2019-05-19 15:51:43 +02:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-16 11:06:36 -04:00
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2020-11-13 00:03:32 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-03-01 07:13:07 -08:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
objtool: Make instruction::alts a single-linked list
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
unsigned int len; /* 64 4 */
enum insn_type type; /* 68 4 */
long unsigned int immediate; /* 72 8 */
u16 dead_end:1; /* 80: 0 2 */
u16 ignore:1; /* 80: 1 2 */
u16 ignore_alts:1; /* 80: 2 2 */
u16 hint:1; /* 80: 3 2 */
u16 save:1; /* 80: 4 2 */
u16 restore:1; /* 80: 5 2 */
u16 retpoline_safe:1; /* 80: 6 2 */
u16 noendbr:1; /* 80: 7 2 */
u16 entry:1; /* 80: 8 2 */
/* XXX 7 bits hole, try to pack */
s8 instr; /* 82 1 */
u8 visited; /* 83 1 */
/* XXX 4 bytes hole, try to pack */
struct alt_group * alt_group; /* 88 8 */
struct symbol * call_dest; /* 96 8 */
struct instruction * jump_dest; /* 104 8 */
struct instruction * first_jump_src; /* 112 8 */
struct reloc * jump_table; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
struct reloc * reloc; /* 128 8 */
- struct list_head alts; /* 136 16 */
- struct symbol * sym; /* 152 8 */
- struct stack_op * stack_ops; /* 160 8 */
- struct cfi_state * cfi; /* 168 8 */
+ struct alternative * alts; /* 136 8 */
+ struct symbol * sym; /* 144 8 */
+ struct stack_op * stack_ops; /* 152 8 */
+ struct cfi_state * cfi; /* 160 8 */
- /* size: 176, cachelines: 3, members: 29 */
- /* sum members: 170, holes: 1, sum holes: 4 */
+ /* size: 168, cachelines: 3, members: 29 */
+ /* sum members: 162, holes: 1, sum holes: 4 */
/* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */
- /* last cacheline: 48 bytes */
+ /* last cacheline: 40 bytes */
};
pre: 5:58.50 real, 229.64 user, 128.65 sys, 26221520 mem
post: 5:48.86 real, 220.30 user, 128.34 sys, 24834672 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.430556498@infradead.org
2023-02-08 18:17:59 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2019-03-01 11:19:03 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-06-05 09:12:21 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2017-07-11 10:33:42 -05:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-16 15:47:27 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2020-03-16 15:47:27 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2018-05-09 22:39:15 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2018-05-09 22:39:15 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2018-05-09 22:39:15 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-04-28 16:45:16 -05:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2020-04-28 16:45:16 -05:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2020-04-28 16:45:16 -05:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2020-04-28 16:45:16 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-10 18:27:24 +01:00
|
|
|
|
2018-05-09 22:39:15 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-10 18:24:59 +01:00
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2020-03-10 18:24:59 +01:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-02-24 10:29:14 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2021-02-24 10:29:14 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2021-02-24 10:29:14 -06:00
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:52 -05:00
|
|
|
|
|
|
|
|
|
2021-01-21 15:29:22 -06:00
|
|
|
|
2022-07-11 11:49:50 +02:00
|
|
|
|
2021-01-21 15:29:22 -06:00
|
|
|
|
2022-07-11 11:49:50 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:52 -05:00
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:52 -05:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:50 -05:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-18 14:27:54 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-04-18 14:27:54 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-04-18 14:27:54 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2019-07-17 20:36:51 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 16:49:37 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:50 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-04-12 16:49:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-05-09 22:39:15 -05:00
|
|
|
|
2019-07-17 20:36:50 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2018-05-09 22:39:15 -05:00
|
|
|
|
powerpc updates for 6.2
- Add powerpc qspinlock implementation optimised for large system scalability and
paravirt. See the merge message for more details.
- Enable objtool to be built on powerpc to generate mcount locations.
- Use a temporary mm for code patching with the Radix MMU, so the writable mapping is
restricted to the patching CPU.
- Add an option to build the 64-bit big-endian kernel with the ELFv2 ABI.
- Sanitise user registers on interrupt entry on 64-bit Book3S.
- Many other small features and fixes.
Thanks to: Aboorva Devarajan, Angel Iglesias, Benjamin Gray, Bjorn Helgaas, Bo Liu, Chen
Lifu, Christoph Hellwig, Christophe JAILLET, Christophe Leroy, Christopher M. Riedl, Colin
Ian King, Deming Wang, Disha Goel, Dmitry Torokhov, Finn Thain, Geert Uytterhoeven,
Gustavo A. R. Silva, Haowen Bai, Joel Stanley, Jordan Niethe, Julia Lawall, Kajol Jain,
Laurent Dufour, Li zeming, Miaoqian Lin, Michael Jeanson, Nathan Lynch, Naveen N. Rao,
Nayna Jain, Nicholas Miehlbradt, Nicholas Piggin, Pali Rohár, Randy Dunlap, Rohan McLure,
Russell Currey, Sathvika Vasireddy, Shaomin Deng, Stephen Kitt, Stephen Rothwell, Thomas
Weißschuh, Tiezhu Yang, Uwe Kleine-König, Xie Shaowen, Xiu Jianfeng, XueBing Chen, Yang
Yingliang, Zhang Jiaming, ruanjinjie, Jessica Yu, Wolfram Sang.
-----BEGIN PGP SIGNATURE-----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=7k3p
-----END PGP SIGNATURE-----
Merge tag 'powerpc-6.2-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc updates from Michael Ellerman:
- Add powerpc qspinlock implementation optimised for large system
scalability and paravirt. See the merge message for more details
- Enable objtool to be built on powerpc to generate mcount locations
- Use a temporary mm for code patching with the Radix MMU, so the
writable mapping is restricted to the patching CPU
- Add an option to build the 64-bit big-endian kernel with the ELFv2
ABI
- Sanitise user registers on interrupt entry on 64-bit Book3S
- Many other small features and fixes
Thanks to Aboorva Devarajan, Angel Iglesias, Benjamin Gray, Bjorn
Helgaas, Bo Liu, Chen Lifu, Christoph Hellwig, Christophe JAILLET,
Christophe Leroy, Christopher M. Riedl, Colin Ian King, Deming Wang,
Disha Goel, Dmitry Torokhov, Finn Thain, Geert Uytterhoeven, Gustavo A.
R. Silva, Haowen Bai, Joel Stanley, Jordan Niethe, Julia Lawall, Kajol
Jain, Laurent Dufour, Li zeming, Miaoqian Lin, Michael Jeanson, Nathan
Lynch, Naveen N. Rao, Nayna Jain, Nicholas Miehlbradt, Nicholas Piggin,
Pali Rohár, Randy Dunlap, Rohan McLure, Russell Currey, Sathvika
Vasireddy, Shaomin Deng, Stephen Kitt, Stephen Rothwell, Thomas
Weißschuh, Tiezhu Yang, Uwe Kleine-König, Xie Shaowen, Xiu Jianfeng,
XueBing Chen, Yang Yingliang, Zhang Jiaming, ruanjinjie, Jessica Yu,
and Wolfram Sang.
* tag 'powerpc-6.2-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux: (181 commits)
powerpc/code-patching: Fix oops with DEBUG_VM enabled
powerpc/qspinlock: Fix 32-bit build
powerpc/prom: Fix 32-bit build
powerpc/rtas: mandate RTAS syscall filtering
powerpc/rtas: define pr_fmt and convert printk call sites
powerpc/rtas: clean up includes
powerpc/rtas: clean up rtas_error_log_max initialization
powerpc/pseries/eeh: use correct API for error log size
powerpc/rtas: avoid scheduling in rtas_os_term()
powerpc/rtas: avoid device tree lookups in rtas_os_term()
powerpc/rtasd: use correct OF API for event scan rate
powerpc/rtas: document rtas_call()
powerpc/pseries: unregister VPA when hot unplugging a CPU
powerpc/pseries: reset the RCU watchdogs after a LPM
powerpc: Take in account addition CPU node when building kexec FDT
powerpc: export the CPU node count
powerpc/cpuidle: Set CPUIDLE_FLAG_POLLING for snooze state
powerpc/dts/fsl: Fix pca954x i2c-mux node names
cxl: Remove unnecessary cxl_pci_window_alignment()
selftests/powerpc: Fix resource leaks
...
2022-12-19 07:13:33 -06:00
|
|
|
|
2019-07-17 20:36:50 -05:00
|
|
|
|
2018-05-09 22:39:15 -05:00
|
|
|
|
2020-03-10 18:27:24 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:50 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:50 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-10 18:27:24 +01:00
|
|
|
|
2019-07-17 20:36:52 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:50 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2019-07-17 20:36:52 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2019-07-17 20:36:52 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:50 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:50 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-08-29 12:51:03 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2017-08-29 12:51:03 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:43 -07:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-23 18:26:03 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
|
|
|
|
|
2020-03-23 18:26:03 +01:00
|
|
|
|
2022-04-18 09:50:43 -07:00
|
|
|
|
2020-03-23 18:26:03 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-07-11 10:33:42 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 17:18:17 +01:00
|
|
|
|
2022-06-14 23:15:47 +02:00
|
|
|
|
2023-01-12 20:43:31 +01:00
|
|
|
|
2023-07-11 11:19:51 +02:00
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
2022-09-15 13:11:08 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-08-22 18:50:37 -04:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
2021-08-22 18:50:37 -04:00
|
|
|
|
2023-02-08 18:17:57 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2022-03-08 16:30:49 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-16 15:47:27 +01:00
|
|
|
|
2020-03-12 09:26:29 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-12-20 20:13:23 +10:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-08-14 13:44:29 +02:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:55 +01:00
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2020-03-12 09:26:29 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
2023-05-30 10:21:06 -07:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:06 -07:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-04-21 15:08:43 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-08-22 18:50:37 -04:00
|
|
|
|
2020-04-21 15:08:43 -07:00
|
|
|
|
2021-08-22 18:50:37 -04:00
|
|
|
|
2020-04-21 15:08:43 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
objtool: Assume unannotated UD2 instructions are dead ends
Arnd reported some false positive warnings with GCC 7:
drivers/hid/wacom_wac.o: warning: objtool: wacom_bpt3_touch()+0x2a5: stack state mismatch: cfa1=7+8 cfa2=6+16
drivers/iio/adc/vf610_adc.o: warning: objtool: vf610_adc_calculate_rates() falls through to next function vf610_adc_sample_set()
drivers/pwm/pwm-hibvt.o: warning: objtool: hibvt_pwm_get_state() falls through to next function hibvt_pwm_remove()
drivers/pwm/pwm-mediatek.o: warning: objtool: mtk_pwm_config() falls through to next function mtk_pwm_enable()
drivers/spi/spi-bcm2835.o: warning: objtool: .text: unexpected end of section
drivers/spi/spi-bcm2835aux.o: warning: objtool: .text: unexpected end of section
drivers/watchdog/digicolor_wdt.o: warning: objtool: dc_wdt_get_timeleft() falls through to next function dc_wdt_restart()
When GCC 7 detects a potential divide-by-zero condition, it sometimes
inserts a UD2 instruction for the case where the divisor is zero,
instead of letting the hardware trap on the divide instruction.
Objtool doesn't consider UD2 to be fatal unless it's annotated with
unreachable(). So it considers the GCC-generated UD2 to be non-fatal,
and it tries to follow the control flow past the UD2 and gets
confused.
Previously, objtool *did* assume UD2 was always a dead end. That
changed with the following commit:
d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
The motivation behind that change was that Peter was planning on using
UD2 for __WARN(), which is *not* a dead end. However, it turns out
that some emulators rely on UD2 being fatal, so he ended up using
'ud0' instead:
9a93848fe787 ("x86/debug: Implement __WARN() using UD0")
For GCC 4.5+, it should be safe to go back to the previous assumption
that UD2 is fatal, even when it's not annotated with unreachable().
But for pre-4.5 versions of GCC, the unreachable() macro isn't
supported, so such cases of UD2 need to be explicitly annotated as
reachable.
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
Link: http://lkml.kernel.org/r/e57fa9dfede25f79487da8126ee9cdf7b856db65.1501188854.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-07-27 15:56:53 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
objtool: Assume unannotated UD2 instructions are dead ends
Arnd reported some false positive warnings with GCC 7:
drivers/hid/wacom_wac.o: warning: objtool: wacom_bpt3_touch()+0x2a5: stack state mismatch: cfa1=7+8 cfa2=6+16
drivers/iio/adc/vf610_adc.o: warning: objtool: vf610_adc_calculate_rates() falls through to next function vf610_adc_sample_set()
drivers/pwm/pwm-hibvt.o: warning: objtool: hibvt_pwm_get_state() falls through to next function hibvt_pwm_remove()
drivers/pwm/pwm-mediatek.o: warning: objtool: mtk_pwm_config() falls through to next function mtk_pwm_enable()
drivers/spi/spi-bcm2835.o: warning: objtool: .text: unexpected end of section
drivers/spi/spi-bcm2835aux.o: warning: objtool: .text: unexpected end of section
drivers/watchdog/digicolor_wdt.o: warning: objtool: dc_wdt_get_timeleft() falls through to next function dc_wdt_restart()
When GCC 7 detects a potential divide-by-zero condition, it sometimes
inserts a UD2 instruction for the case where the divisor is zero,
instead of letting the hardware trap on the divide instruction.
Objtool doesn't consider UD2 to be fatal unless it's annotated with
unreachable(). So it considers the GCC-generated UD2 to be non-fatal,
and it tries to follow the control flow past the UD2 and gets
confused.
Previously, objtool *did* assume UD2 was always a dead end. That
changed with the following commit:
d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
The motivation behind that change was that Peter was planning on using
UD2 for __WARN(), which is *not* a dead end. However, it turns out
that some emulators rely on UD2 being fatal, so he ended up using
'ud0' instead:
9a93848fe787 ("x86/debug: Implement __WARN() using UD0")
For GCC 4.5+, it should be safe to go back to the previous assumption
that UD2 is fatal, even when it's not annotated with unreachable().
But for pre-4.5 versions of GCC, the unreachable() macro isn't
supported, so such cases of UD2 need to be explicitly annotated as
reachable.
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
Link: http://lkml.kernel.org/r/e57fa9dfede25f79487da8126ee9cdf7b856db65.1501188854.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-07-27 15:56:53 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
objtool: Assume unannotated UD2 instructions are dead ends
Arnd reported some false positive warnings with GCC 7:
drivers/hid/wacom_wac.o: warning: objtool: wacom_bpt3_touch()+0x2a5: stack state mismatch: cfa1=7+8 cfa2=6+16
drivers/iio/adc/vf610_adc.o: warning: objtool: vf610_adc_calculate_rates() falls through to next function vf610_adc_sample_set()
drivers/pwm/pwm-hibvt.o: warning: objtool: hibvt_pwm_get_state() falls through to next function hibvt_pwm_remove()
drivers/pwm/pwm-mediatek.o: warning: objtool: mtk_pwm_config() falls through to next function mtk_pwm_enable()
drivers/spi/spi-bcm2835.o: warning: objtool: .text: unexpected end of section
drivers/spi/spi-bcm2835aux.o: warning: objtool: .text: unexpected end of section
drivers/watchdog/digicolor_wdt.o: warning: objtool: dc_wdt_get_timeleft() falls through to next function dc_wdt_restart()
When GCC 7 detects a potential divide-by-zero condition, it sometimes
inserts a UD2 instruction for the case where the divisor is zero,
instead of letting the hardware trap on the divide instruction.
Objtool doesn't consider UD2 to be fatal unless it's annotated with
unreachable(). So it considers the GCC-generated UD2 to be non-fatal,
and it tries to follow the control flow past the UD2 and gets
confused.
Previously, objtool *did* assume UD2 was always a dead end. That
changed with the following commit:
d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
The motivation behind that change was that Peter was planning on using
UD2 for __WARN(), which is *not* a dead end. However, it turns out
that some emulators rely on UD2 being fatal, so he ended up using
'ud0' instead:
9a93848fe787 ("x86/debug: Implement __WARN() using UD0")
For GCC 4.5+, it should be safe to go back to the previous assumption
that UD2 is fatal, even when it's not annotated with unreachable().
But for pre-4.5 versions of GCC, the unreachable() macro isn't
supported, so such cases of UD2 need to be explicitly annotated as
reachable.
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
Link: http://lkml.kernel.org/r/e57fa9dfede25f79487da8126ee9cdf7b856db65.1501188854.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-07-27 15:56:53 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-05-30 10:21:02 -07:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2020-04-21 15:08:43 -07:00
|
|
|
|
2022-05-16 11:06:36 -04:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-16 11:06:36 -04:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Assume unannotated UD2 instructions are dead ends
Arnd reported some false positive warnings with GCC 7:
drivers/hid/wacom_wac.o: warning: objtool: wacom_bpt3_touch()+0x2a5: stack state mismatch: cfa1=7+8 cfa2=6+16
drivers/iio/adc/vf610_adc.o: warning: objtool: vf610_adc_calculate_rates() falls through to next function vf610_adc_sample_set()
drivers/pwm/pwm-hibvt.o: warning: objtool: hibvt_pwm_get_state() falls through to next function hibvt_pwm_remove()
drivers/pwm/pwm-mediatek.o: warning: objtool: mtk_pwm_config() falls through to next function mtk_pwm_enable()
drivers/spi/spi-bcm2835.o: warning: objtool: .text: unexpected end of section
drivers/spi/spi-bcm2835aux.o: warning: objtool: .text: unexpected end of section
drivers/watchdog/digicolor_wdt.o: warning: objtool: dc_wdt_get_timeleft() falls through to next function dc_wdt_restart()
When GCC 7 detects a potential divide-by-zero condition, it sometimes
inserts a UD2 instruction for the case where the divisor is zero,
instead of letting the hardware trap on the divide instruction.
Objtool doesn't consider UD2 to be fatal unless it's annotated with
unreachable(). So it considers the GCC-generated UD2 to be non-fatal,
and it tries to follow the control flow past the UD2 and gets
confused.
Previously, objtool *did* assume UD2 was always a dead end. That
changed with the following commit:
d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
The motivation behind that change was that Peter was planning on using
UD2 for __WARN(), which is *not* a dead end. However, it turns out
that some emulators rely on UD2 being fatal, so he ended up using
'ud0' instead:
9a93848fe787 ("x86/debug: Implement __WARN() using UD0")
For GCC 4.5+, it should be safe to go back to the previous assumption
that UD2 is fatal, even when it's not annotated with unreachable().
But for pre-4.5 versions of GCC, the unreachable() macro isn't
supported, so such cases of UD2 need to be explicitly annotated as
reachable.
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
Link: http://lkml.kernel.org/r/e57fa9dfede25f79487da8126ee9cdf7b856db65.1501188854.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-07-27 15:56:53 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
objtool: Assume unannotated UD2 instructions are dead ends
Arnd reported some false positive warnings with GCC 7:
drivers/hid/wacom_wac.o: warning: objtool: wacom_bpt3_touch()+0x2a5: stack state mismatch: cfa1=7+8 cfa2=6+16
drivers/iio/adc/vf610_adc.o: warning: objtool: vf610_adc_calculate_rates() falls through to next function vf610_adc_sample_set()
drivers/pwm/pwm-hibvt.o: warning: objtool: hibvt_pwm_get_state() falls through to next function hibvt_pwm_remove()
drivers/pwm/pwm-mediatek.o: warning: objtool: mtk_pwm_config() falls through to next function mtk_pwm_enable()
drivers/spi/spi-bcm2835.o: warning: objtool: .text: unexpected end of section
drivers/spi/spi-bcm2835aux.o: warning: objtool: .text: unexpected end of section
drivers/watchdog/digicolor_wdt.o: warning: objtool: dc_wdt_get_timeleft() falls through to next function dc_wdt_restart()
When GCC 7 detects a potential divide-by-zero condition, it sometimes
inserts a UD2 instruction for the case where the divisor is zero,
instead of letting the hardware trap on the divide instruction.
Objtool doesn't consider UD2 to be fatal unless it's annotated with
unreachable(). So it considers the GCC-generated UD2 to be non-fatal,
and it tries to follow the control flow past the UD2 and gets
confused.
Previously, objtool *did* assume UD2 was always a dead end. That
changed with the following commit:
d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
The motivation behind that change was that Peter was planning on using
UD2 for __WARN(), which is *not* a dead end. However, it turns out
that some emulators rely on UD2 being fatal, so he ended up using
'ud0' instead:
9a93848fe787 ("x86/debug: Implement __WARN() using UD0")
For GCC 4.5+, it should be safe to go back to the previous assumption
that UD2 is fatal, even when it's not annotated with unreachable().
But for pre-4.5 versions of GCC, the unreachable() macro isn't
supported, so such cases of UD2 need to be explicitly annotated as
reachable.
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
Link: http://lkml.kernel.org/r/e57fa9dfede25f79487da8126ee9cdf7b856db65.1501188854.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-07-27 15:56:53 -05:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:02 -07:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
objtool: Assume unannotated UD2 instructions are dead ends
Arnd reported some false positive warnings with GCC 7:
drivers/hid/wacom_wac.o: warning: objtool: wacom_bpt3_touch()+0x2a5: stack state mismatch: cfa1=7+8 cfa2=6+16
drivers/iio/adc/vf610_adc.o: warning: objtool: vf610_adc_calculate_rates() falls through to next function vf610_adc_sample_set()
drivers/pwm/pwm-hibvt.o: warning: objtool: hibvt_pwm_get_state() falls through to next function hibvt_pwm_remove()
drivers/pwm/pwm-mediatek.o: warning: objtool: mtk_pwm_config() falls through to next function mtk_pwm_enable()
drivers/spi/spi-bcm2835.o: warning: objtool: .text: unexpected end of section
drivers/spi/spi-bcm2835aux.o: warning: objtool: .text: unexpected end of section
drivers/watchdog/digicolor_wdt.o: warning: objtool: dc_wdt_get_timeleft() falls through to next function dc_wdt_restart()
When GCC 7 detects a potential divide-by-zero condition, it sometimes
inserts a UD2 instruction for the case where the divisor is zero,
instead of letting the hardware trap on the divide instruction.
Objtool doesn't consider UD2 to be fatal unless it's annotated with
unreachable(). So it considers the GCC-generated UD2 to be non-fatal,
and it tries to follow the control flow past the UD2 and gets
confused.
Previously, objtool *did* assume UD2 was always a dead end. That
changed with the following commit:
d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
The motivation behind that change was that Peter was planning on using
UD2 for __WARN(), which is *not* a dead end. However, it turns out
that some emulators rely on UD2 being fatal, so he ended up using
'ud0' instead:
9a93848fe787 ("x86/debug: Implement __WARN() using UD0")
For GCC 4.5+, it should be safe to go back to the previous assumption
that UD2 is fatal, even when it's not annotated with unreachable().
But for pre-4.5 versions of GCC, the unreachable() macro isn't
supported, so such cases of UD2 need to be explicitly annotated as
reachable.
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
Link: http://lkml.kernel.org/r/e57fa9dfede25f79487da8126ee9cdf7b856db65.1501188854.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-07-27 15:56:53 -05:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Assume unannotated UD2 instructions are dead ends
Arnd reported some false positive warnings with GCC 7:
drivers/hid/wacom_wac.o: warning: objtool: wacom_bpt3_touch()+0x2a5: stack state mismatch: cfa1=7+8 cfa2=6+16
drivers/iio/adc/vf610_adc.o: warning: objtool: vf610_adc_calculate_rates() falls through to next function vf610_adc_sample_set()
drivers/pwm/pwm-hibvt.o: warning: objtool: hibvt_pwm_get_state() falls through to next function hibvt_pwm_remove()
drivers/pwm/pwm-mediatek.o: warning: objtool: mtk_pwm_config() falls through to next function mtk_pwm_enable()
drivers/spi/spi-bcm2835.o: warning: objtool: .text: unexpected end of section
drivers/spi/spi-bcm2835aux.o: warning: objtool: .text: unexpected end of section
drivers/watchdog/digicolor_wdt.o: warning: objtool: dc_wdt_get_timeleft() falls through to next function dc_wdt_restart()
When GCC 7 detects a potential divide-by-zero condition, it sometimes
inserts a UD2 instruction for the case where the divisor is zero,
instead of letting the hardware trap on the divide instruction.
Objtool doesn't consider UD2 to be fatal unless it's annotated with
unreachable(). So it considers the GCC-generated UD2 to be non-fatal,
and it tries to follow the control flow past the UD2 and gets
confused.
Previously, objtool *did* assume UD2 was always a dead end. That
changed with the following commit:
d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
The motivation behind that change was that Peter was planning on using
UD2 for __WARN(), which is *not* a dead end. However, it turns out
that some emulators rely on UD2 being fatal, so he ended up using
'ud0' instead:
9a93848fe787 ("x86/debug: Implement __WARN() using UD0")
For GCC 4.5+, it should be safe to go back to the previous assumption
that UD2 is fatal, even when it's not annotated with unreachable().
But for pre-4.5 versions of GCC, the unreachable() macro isn't
supported, so such cases of UD2 need to be explicitly annotated as
reachable.
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
Link: http://lkml.kernel.org/r/e57fa9dfede25f79487da8126ee9cdf7b856db65.1501188854.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-07-27 15:56:53 -05:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2020-04-21 15:08:43 -07:00
|
|
|
|
2022-05-16 11:06:36 -04:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
objtool: Assume unannotated UD2 instructions are dead ends
Arnd reported some false positive warnings with GCC 7:
drivers/hid/wacom_wac.o: warning: objtool: wacom_bpt3_touch()+0x2a5: stack state mismatch: cfa1=7+8 cfa2=6+16
drivers/iio/adc/vf610_adc.o: warning: objtool: vf610_adc_calculate_rates() falls through to next function vf610_adc_sample_set()
drivers/pwm/pwm-hibvt.o: warning: objtool: hibvt_pwm_get_state() falls through to next function hibvt_pwm_remove()
drivers/pwm/pwm-mediatek.o: warning: objtool: mtk_pwm_config() falls through to next function mtk_pwm_enable()
drivers/spi/spi-bcm2835.o: warning: objtool: .text: unexpected end of section
drivers/spi/spi-bcm2835aux.o: warning: objtool: .text: unexpected end of section
drivers/watchdog/digicolor_wdt.o: warning: objtool: dc_wdt_get_timeleft() falls through to next function dc_wdt_restart()
When GCC 7 detects a potential divide-by-zero condition, it sometimes
inserts a UD2 instruction for the case where the divisor is zero,
instead of letting the hardware trap on the divide instruction.
Objtool doesn't consider UD2 to be fatal unless it's annotated with
unreachable(). So it considers the GCC-generated UD2 to be non-fatal,
and it tries to follow the control flow past the UD2 and gets
confused.
Previously, objtool *did* assume UD2 was always a dead end. That
changed with the following commit:
d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
The motivation behind that change was that Peter was planning on using
UD2 for __WARN(), which is *not* a dead end. However, it turns out
that some emulators rely on UD2 being fatal, so he ended up using
'ud0' instead:
9a93848fe787 ("x86/debug: Implement __WARN() using UD0")
For GCC 4.5+, it should be safe to go back to the previous assumption
that UD2 is fatal, even when it's not annotated with unreachable().
But for pre-4.5 versions of GCC, the unreachable() macro isn't
supported, so such cases of UD2 need to be explicitly annotated as
reachable.
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
Link: http://lkml.kernel.org/r/e57fa9dfede25f79487da8126ee9cdf7b856db65.1501188854.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-07-27 15:56:53 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-05-16 11:06:36 -04:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
objtool: Assume unannotated UD2 instructions are dead ends
Arnd reported some false positive warnings with GCC 7:
drivers/hid/wacom_wac.o: warning: objtool: wacom_bpt3_touch()+0x2a5: stack state mismatch: cfa1=7+8 cfa2=6+16
drivers/iio/adc/vf610_adc.o: warning: objtool: vf610_adc_calculate_rates() falls through to next function vf610_adc_sample_set()
drivers/pwm/pwm-hibvt.o: warning: objtool: hibvt_pwm_get_state() falls through to next function hibvt_pwm_remove()
drivers/pwm/pwm-mediatek.o: warning: objtool: mtk_pwm_config() falls through to next function mtk_pwm_enable()
drivers/spi/spi-bcm2835.o: warning: objtool: .text: unexpected end of section
drivers/spi/spi-bcm2835aux.o: warning: objtool: .text: unexpected end of section
drivers/watchdog/digicolor_wdt.o: warning: objtool: dc_wdt_get_timeleft() falls through to next function dc_wdt_restart()
When GCC 7 detects a potential divide-by-zero condition, it sometimes
inserts a UD2 instruction for the case where the divisor is zero,
instead of letting the hardware trap on the divide instruction.
Objtool doesn't consider UD2 to be fatal unless it's annotated with
unreachable(). So it considers the GCC-generated UD2 to be non-fatal,
and it tries to follow the control flow past the UD2 and gets
confused.
Previously, objtool *did* assume UD2 was always a dead end. That
changed with the following commit:
d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
The motivation behind that change was that Peter was planning on using
UD2 for __WARN(), which is *not* a dead end. However, it turns out
that some emulators rely on UD2 being fatal, so he ended up using
'ud0' instead:
9a93848fe787 ("x86/debug: Implement __WARN() using UD0")
For GCC 4.5+, it should be safe to go back to the previous assumption
that UD2 is fatal, even when it's not annotated with unreachable().
But for pre-4.5 versions of GCC, the unreachable() macro isn't
supported, so such cases of UD2 need to be explicitly annotated as
reachable.
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: d1091c7fa3d5 ("objtool: Improve detection of BUG() and other dead ends")
Link: http://lkml.kernel.org/r/e57fa9dfede25f79487da8126ee9cdf7b856db65.1501188854.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-07-27 15:56:53 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-03-26 16:12:12 +01:00
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
2023-05-30 10:20:54 -07:00
|
|
|
|
|
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
2021-03-26 16:12:12 +01:00
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-12-14 16:04:20 -06:00
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-12-05 12:06:42 +04:00
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2021-01-27 17:18:37 -06:00
|
|
|
|
2022-12-05 12:06:42 +04:00
|
|
|
|
2021-01-27 17:18:37 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:36 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:36 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:36 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-14 23:15:38 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-14 23:15:38 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-14 23:15:38 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:55 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2022-03-08 16:30:55 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:55 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-01-18 11:52:15 +01:00
|
|
|
|
2022-03-08 16:30:55 +01:00
|
|
|
|
|
|
|
|
|
2023-01-18 11:52:15 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:55 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-27 11:28:13 +02:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
2022-10-27 11:28:13 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
|
|
|
|
|
2022-10-27 11:28:13 +02:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
|
|
|
|
|
2022-10-27 11:28:13 +02:00
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
2022-10-27 11:28:13 +02:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
2022-10-27 11:28:13 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
|
|
|
|
|
2022-10-27 11:28:13 +02:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
|
|
|
|
|
2022-10-27 11:28:13 +02:00
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
2022-10-27 11:28:13 +02:00
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
2022-10-27 11:28:13 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-06 15:14:09 -07:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:56 -07:00
|
|
|
|
2020-08-06 15:14:09 -07:00
|
|
|
|
2022-11-14 23:27:48 +05:30
|
|
|
|
2020-08-06 15:14:09 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:35 +02:00
|
|
|
|
2020-08-06 15:14:09 -07:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
2020-08-06 15:14:09 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:56 -07:00
|
|
|
|
2022-11-14 23:27:48 +05:30
|
|
|
|
2020-08-06 15:14:09 -07:00
|
|
|
|
2021-10-26 14:01:35 +02:00
|
|
|
|
2020-08-06 15:14:09 -07:00
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
2020-08-06 15:14:09 -07:00
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-06 15:14:09 -07:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:12 -07:00
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
2020-08-06 15:14:09 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-15 13:11:09 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
2022-09-15 13:11:09 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:59 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-15 13:11:09 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
2019-02-27 14:04:13 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-05-30 10:21:02 -07:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2019-02-27 14:04:13 +01:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2019-02-27 14:04:13 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2020-02-17 21:41:54 -06:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2019-02-27 14:04:13 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2019-02-27 14:04:13 +01:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
2019-02-27 14:04:13 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2019-02-27 14:04:13 +01:00
|
|
|
|
2020-03-10 18:27:24 +01:00
|
|
|
|
2019-02-27 14:04:13 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-02-24 12:05:05 -08:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Permit __kasan_check_{read,write} under UACCESS
Building linux-next with JUMP_LABEL=n and KASAN=y, I got this objtool
warning:
arch/x86/lib/copy_mc.o: warning: objtool: copy_mc_to_user()+0x22: call to
__kasan_check_read() with UACCESS enabled
What happens here is that copy_mc_to_user() branches on a static key in a
UACCESS region:
__uaccess_begin();
if (static_branch_unlikely(©_mc_fragile_key))
ret = copy_mc_fragile(to, from, len);
ret = copy_mc_generic(to, from, len);
__uaccess_end();
and the !CONFIG_JUMP_LABEL version of static_branch_unlikely() uses
static_key_enabled(), which uses static_key_count(), which uses
atomic_read(), which calls instrument_atomic_read(), which uses
kasan_check_read(), which is __kasan_check_read().
Let's permit these KASAN helpers in UACCESS regions - static keys should
probably work under UACCESS, I think.
PeterZ adds:
It's not a matter of permitting, it's a matter of being safe and
correct. In this case it is, because it's a thin wrapper around
check_memory_region() which was already marked safe.
check_memory_region() is correct because the only thing it ends up
calling is kasa_report() and that is also marked safe because that is
annotated with user_access_save/restore() before it does anything else.
On top of that, all of KASAN is noinstr, so nothing in here will end up
in tracing and/or call schedule() before the user_access_save().
Signed-off-by: Jann Horn <jannh@google.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-09-29 00:49:16 +02:00
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-14 19:02:57 +01:00
|
|
|
|
2020-03-25 17:41:57 +01:00
|
|
|
|
2021-11-30 12:44:30 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-11-14 19:02:57 +01:00
|
|
|
|
|
|
|
|
|
2020-03-25 17:41:57 +01:00
|
|
|
|
2020-04-24 17:47:30 +02:00
|
|
|
|
|
|
|
|
|
2019-11-14 19:02:57 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-07-24 09:00:02 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-12 11:45:41 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-07-03 15:40:30 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-02-15 14:00:58 +01:00
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
2020-04-29 14:09:04 -05:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-02-16 13:07:49 -05:00
|
|
|
|
2022-09-15 17:04:05 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-10-21 15:11:49 +02:00
|
|
|
|
2023-01-12 20:43:56 +01:00
|
|
|
|
2023-04-12 10:29:35 -07:00
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
x86, powerpc: Rename memcpy_mcsafe() to copy_mc_to_{user, kernel}()
In reaction to a proposal to introduce a memcpy_mcsafe_fast()
implementation Linus points out that memcpy_mcsafe() is poorly named
relative to communicating the scope of the interface. Specifically what
addresses are valid to pass as source, destination, and what faults /
exceptions are handled.
Of particular concern is that even though x86 might be able to handle
the semantics of copy_mc_to_user() with its common copy_user_generic()
implementation other archs likely need / want an explicit path for this
case:
On Fri, May 1, 2020 at 11:28 AM Linus Torvalds <torvalds@linux-foundation.org> wrote:
>
> On Thu, Apr 30, 2020 at 6:21 PM Dan Williams <dan.j.williams@intel.com> wrote:
> >
> > However now I see that copy_user_generic() works for the wrong reason.
> > It works because the exception on the source address due to poison
> > looks no different than a write fault on the user address to the
> > caller, it's still just a short copy. So it makes copy_to_user() work
> > for the wrong reason relative to the name.
>
> Right.
>
> And it won't work that way on other architectures. On x86, we have a
> generic function that can take faults on either side, and we use it
> for both cases (and for the "in_user" case too), but that's an
> artifact of the architecture oddity.
>
> In fact, it's probably wrong even on x86 - because it can hide bugs -
> but writing those things is painful enough that everybody prefers
> having just one function.
Replace a single top-level memcpy_mcsafe() with either
copy_mc_to_user(), or copy_mc_to_kernel().
Introduce an x86 copy_mc_fragile() name as the rename for the
low-level x86 implementation formerly named memcpy_mcsafe(). It is used
as the slow / careful backend that is supplanted by a fast
copy_mc_generic() in a follow-on patch.
One side-effect of this reorganization is that separating copy_mc_64.S
to its own file means that perf no longer needs to track dependencies
for its memcpy_64.S benchmarks.
[ bp: Massage a bit. ]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Cc: <stable@vger.kernel.org>
Link: http://lore.kernel.org/r/CAHk-=wjSqtXAqfUJxFtWNwmguFASTgB0dz1dT3V-78Quiezqbg@mail.gmail.com
Link: https://lkml.kernel.org/r/160195561680.2163339.11574962055305783722.stgit@dwillia2-desk3.amr.corp.intel.com
2020-10-05 20:40:16 -07:00
|
|
|
|
|
|
|
|
|
2020-10-05 20:40:25 -07:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
2023-04-16 14:06:58 -07:00
|
|
|
|
x86: improve on the non-rep 'copy_user' function
The old 'copy_user_generic_unrolled' function was oddly implemented for
largely historical reasons: it had been largely based on the uncached
copy case, which has some other concerns.
For example, the __copy_user_nocache() function uses 'movnti' for the
destination stores, and those want the destination to be aligned. In
contrast, the regular copy function doesn't really care, and trying to
align things only complicates matters.
Also, like the clear_user function, the copy function had some odd
handling of the repeat counts, complicating the exception handling for
no really good reason. So as with clear_user, just write it to keep all
the byte counts in the %rcx register, exactly like the 'rep movs'
functionality that this replaces.
Unlike a real 'rep movs', we do allow for this to trash a few temporary
registers to not have to unnecessarily save/restore registers on the
stack.
And like the clearing case, rename this to what it now clearly is:
'rep_movs_alternative', and make it one coherent function, so that it
shows up as such in profiles (instead of the odd split between
"copy_user_generic_unrolled" and "copy_user_short_string", the latter of
which was not about strings at all, and which was shared with the
uncached case).
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2023-04-16 18:23:06 -07:00
|
|
|
|
2023-04-15 13:39:15 -07:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:48 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-11 21:46:24 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-03-18 14:33:07 +01:00
|
|
|
|
2018-01-11 21:46:24 +00:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2018-01-11 21:46:24 +00:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
2018-01-11 21:46:24 +00:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:02 -07:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2018-01-11 21:46:24 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2018-01-11 21:46:24 +00:00
|
|
|
|
2019-03-18 14:33:07 +01:00
|
|
|
|
2018-01-11 21:46:24 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-08-14 13:44:29 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-03-26 16:12:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-08-14 13:44:29 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-14 23:15:38 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-08-14 13:44:29 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-03-26 16:12:13 +01:00
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::reloc
Instead of caching the reloc for each instruction, only keep a
negative cache of not having a reloc (by far the most common case).
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
+ u16 no_reloc:1; /* 78: 5 2 */
- /* XXX 3 bits hole, try to pack */
+ /* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
struct symbol * call_dest; /* 88 8 */
struct instruction * jump_dest; /* 96 8 */
struct instruction * first_jump_src; /* 104 8 */
struct reloc * jump_table; /* 112 8 */
- struct reloc * reloc; /* 120 8 */
+ struct alternative * alts; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct alternative * alts; /* 128 8 */
- struct symbol * sym; /* 136 8 */
- struct stack_op * stack_ops; /* 144 8 */
- struct cfi_state * cfi; /* 152 8 */
+ struct symbol * sym; /* 128 8 */
+ struct stack_op * stack_ops; /* 136 8 */
+ struct cfi_state * cfi; /* 144 8 */
- /* size: 160, cachelines: 3, members: 29 */
- /* sum members: 158 */
- /* sum bitfield members: 13 bits, bit holes: 1, sum bit holes: 3 bits */
- /* last cacheline: 32 bytes */
+ /* size: 152, cachelines: 3, members: 29 */
+ /* sum members: 150 */
+ /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
+ /* last cacheline: 24 bytes */
};
pre: 5:48.89 real, 220.96 user, 127.55 sys, 24834672 mem
post: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.572145269@infradead.org
2023-02-08 18:18:01 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-03-26 16:12:13 +01:00
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::reloc
Instead of caching the reloc for each instruction, only keep a
negative cache of not having a reloc (by far the most common case).
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
+ u16 no_reloc:1; /* 78: 5 2 */
- /* XXX 3 bits hole, try to pack */
+ /* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
struct symbol * call_dest; /* 88 8 */
struct instruction * jump_dest; /* 96 8 */
struct instruction * first_jump_src; /* 104 8 */
struct reloc * jump_table; /* 112 8 */
- struct reloc * reloc; /* 120 8 */
+ struct alternative * alts; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct alternative * alts; /* 128 8 */
- struct symbol * sym; /* 136 8 */
- struct stack_op * stack_ops; /* 144 8 */
- struct cfi_state * cfi; /* 152 8 */
+ struct symbol * sym; /* 128 8 */
+ struct stack_op * stack_ops; /* 136 8 */
+ struct cfi_state * cfi; /* 144 8 */
- /* size: 160, cachelines: 3, members: 29 */
- /* sum members: 158 */
- /* sum bitfield members: 13 bits, bit holes: 1, sum bit holes: 3 bits */
- /* last cacheline: 32 bytes */
+ /* size: 152, cachelines: 3, members: 29 */
+ /* sum members: 150 */
+ /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
+ /* last cacheline: 24 bytes */
};
pre: 5:48.89 real, 220.96 user, 127.55 sys, 24834672 mem
post: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.572145269@infradead.org
2023-02-08 18:18:01 +01:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
objtool: Remove instruction::reloc
Instead of caching the reloc for each instruction, only keep a
negative cache of not having a reloc (by far the most common case).
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
+ u16 no_reloc:1; /* 78: 5 2 */
- /* XXX 3 bits hole, try to pack */
+ /* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
struct symbol * call_dest; /* 88 8 */
struct instruction * jump_dest; /* 96 8 */
struct instruction * first_jump_src; /* 104 8 */
struct reloc * jump_table; /* 112 8 */
- struct reloc * reloc; /* 120 8 */
+ struct alternative * alts; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct alternative * alts; /* 128 8 */
- struct symbol * sym; /* 136 8 */
- struct stack_op * stack_ops; /* 144 8 */
- struct cfi_state * cfi; /* 152 8 */
+ struct symbol * sym; /* 128 8 */
+ struct stack_op * stack_ops; /* 136 8 */
+ struct cfi_state * cfi; /* 144 8 */
- /* size: 160, cachelines: 3, members: 29 */
- /* sum members: 158 */
- /* sum bitfield members: 13 bits, bit holes: 1, sum bit holes: 3 bits */
- /* last cacheline: 32 bytes */
+ /* size: 152, cachelines: 3, members: 29 */
+ /* sum members: 150 */
+ /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
+ /* last cacheline: 24 bytes */
};
pre: 5:48.89 real, 220.96 user, 127.55 sys, 24834672 mem
post: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.572145269@infradead.org
2023-02-08 18:18:01 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-03-26 16:12:13 +01:00
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::reloc
Instead of caching the reloc for each instruction, only keep a
negative cache of not having a reloc (by far the most common case).
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
+ u16 no_reloc:1; /* 78: 5 2 */
- /* XXX 3 bits hole, try to pack */
+ /* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
struct symbol * call_dest; /* 88 8 */
struct instruction * jump_dest; /* 96 8 */
struct instruction * first_jump_src; /* 104 8 */
struct reloc * jump_table; /* 112 8 */
- struct reloc * reloc; /* 120 8 */
+ struct alternative * alts; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct alternative * alts; /* 128 8 */
- struct symbol * sym; /* 136 8 */
- struct stack_op * stack_ops; /* 144 8 */
- struct cfi_state * cfi; /* 152 8 */
+ struct symbol * sym; /* 128 8 */
+ struct stack_op * stack_ops; /* 136 8 */
+ struct cfi_state * cfi; /* 144 8 */
- /* size: 160, cachelines: 3, members: 29 */
- /* sum members: 158 */
- /* sum bitfield members: 13 bits, bit holes: 1, sum bit holes: 3 bits */
- /* last cacheline: 32 bytes */
+ /* size: 152, cachelines: 3, members: 29 */
+ /* sum members: 150 */
+ /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
+ /* last cacheline: 24 bytes */
};
pre: 5:48.89 real, 220.96 user, 127.55 sys, 24834672 mem
post: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.572145269@infradead.org
2023-02-08 18:18:01 +01:00
|
|
|
|
2021-03-26 16:12:13 +01:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
|
|
|
|
|
objtool: Make instruction::stack_ops a single-linked list
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
unsigned int len; /* 64 4 */
enum insn_type type; /* 68 4 */
long unsigned int immediate; /* 72 8 */
u16 dead_end:1; /* 80: 0 2 */
u16 ignore:1; /* 80: 1 2 */
u16 ignore_alts:1; /* 80: 2 2 */
u16 hint:1; /* 80: 3 2 */
u16 save:1; /* 80: 4 2 */
u16 restore:1; /* 80: 5 2 */
u16 retpoline_safe:1; /* 80: 6 2 */
u16 noendbr:1; /* 80: 7 2 */
u16 entry:1; /* 80: 8 2 */
/* XXX 7 bits hole, try to pack */
s8 instr; /* 82 1 */
u8 visited; /* 83 1 */
/* XXX 4 bytes hole, try to pack */
struct alt_group * alt_group; /* 88 8 */
struct symbol * call_dest; /* 96 8 */
struct instruction * jump_dest; /* 104 8 */
struct instruction * first_jump_src; /* 112 8 */
struct reloc * jump_table; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
struct reloc * reloc; /* 128 8 */
struct list_head alts; /* 136 16 */
struct symbol * sym; /* 152 8 */
- struct list_head stack_ops; /* 160 16 */
- struct cfi_state * cfi; /* 176 8 */
+ struct stack_op * stack_ops; /* 160 8 */
+ struct cfi_state * cfi; /* 168 8 */
- /* size: 184, cachelines: 3, members: 29 */
- /* sum members: 178, holes: 1, sum holes: 4 */
+ /* size: 176, cachelines: 3, members: 29 */
+ /* sum members: 170, holes: 1, sum holes: 4 */
/* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */
- /* last cacheline: 56 bytes */
+ /* last cacheline: 48 bytes */
};
pre: 5:58.22 real, 226.69 user, 131.22 sys, 26221520 mem
post: 5:58.50 real, 229.64 user, 128.65 sys, 26221520 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.362196959@infradead.org
2023-02-08 18:17:58 +01:00
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
objtool: Make instruction::stack_ops a single-linked list
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
unsigned int len; /* 64 4 */
enum insn_type type; /* 68 4 */
long unsigned int immediate; /* 72 8 */
u16 dead_end:1; /* 80: 0 2 */
u16 ignore:1; /* 80: 1 2 */
u16 ignore_alts:1; /* 80: 2 2 */
u16 hint:1; /* 80: 3 2 */
u16 save:1; /* 80: 4 2 */
u16 restore:1; /* 80: 5 2 */
u16 retpoline_safe:1; /* 80: 6 2 */
u16 noendbr:1; /* 80: 7 2 */
u16 entry:1; /* 80: 8 2 */
/* XXX 7 bits hole, try to pack */
s8 instr; /* 82 1 */
u8 visited; /* 83 1 */
/* XXX 4 bytes hole, try to pack */
struct alt_group * alt_group; /* 88 8 */
struct symbol * call_dest; /* 96 8 */
struct instruction * jump_dest; /* 104 8 */
struct instruction * first_jump_src; /* 112 8 */
struct reloc * jump_table; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
struct reloc * reloc; /* 128 8 */
struct list_head alts; /* 136 16 */
struct symbol * sym; /* 152 8 */
- struct list_head stack_ops; /* 160 16 */
- struct cfi_state * cfi; /* 176 8 */
+ struct stack_op * stack_ops; /* 160 8 */
+ struct cfi_state * cfi; /* 168 8 */
- /* size: 184, cachelines: 3, members: 29 */
- /* sum members: 178, holes: 1, sum holes: 4 */
+ /* size: 176, cachelines: 3, members: 29 */
+ /* sum members: 170, holes: 1, sum holes: 4 */
/* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */
- /* last cacheline: 56 bytes */
+ /* last cacheline: 48 bytes */
};
pre: 5:58.22 real, 226.69 user, 131.22 sys, 26221520 mem
post: 5:58.50 real, 229.64 user, 128.65 sys, 26221520 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.362196959@infradead.org
2023-02-08 18:17:58 +01:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
|
|
|
|
|
objtool: Make instruction::stack_ops a single-linked list
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
unsigned int len; /* 64 4 */
enum insn_type type; /* 68 4 */
long unsigned int immediate; /* 72 8 */
u16 dead_end:1; /* 80: 0 2 */
u16 ignore:1; /* 80: 1 2 */
u16 ignore_alts:1; /* 80: 2 2 */
u16 hint:1; /* 80: 3 2 */
u16 save:1; /* 80: 4 2 */
u16 restore:1; /* 80: 5 2 */
u16 retpoline_safe:1; /* 80: 6 2 */
u16 noendbr:1; /* 80: 7 2 */
u16 entry:1; /* 80: 8 2 */
/* XXX 7 bits hole, try to pack */
s8 instr; /* 82 1 */
u8 visited; /* 83 1 */
/* XXX 4 bytes hole, try to pack */
struct alt_group * alt_group; /* 88 8 */
struct symbol * call_dest; /* 96 8 */
struct instruction * jump_dest; /* 104 8 */
struct instruction * first_jump_src; /* 112 8 */
struct reloc * jump_table; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
struct reloc * reloc; /* 128 8 */
struct list_head alts; /* 136 16 */
struct symbol * sym; /* 152 8 */
- struct list_head stack_ops; /* 160 16 */
- struct cfi_state * cfi; /* 176 8 */
+ struct stack_op * stack_ops; /* 160 8 */
+ struct cfi_state * cfi; /* 168 8 */
- /* size: 184, cachelines: 3, members: 29 */
- /* sum members: 178, holes: 1, sum holes: 4 */
+ /* size: 176, cachelines: 3, members: 29 */
+ /* sum members: 170, holes: 1, sum holes: 4 */
/* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */
- /* last cacheline: 56 bytes */
+ /* last cacheline: 48 bytes */
};
pre: 5:58.22 real, 226.69 user, 131.22 sys, 26221520 mem
post: 5:58.50 real, 229.64 user, 128.65 sys, 26221520 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.362196959@infradead.org
2023-02-08 18:17:58 +01:00
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:34 +02:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
2021-10-26 14:01:34 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:34 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:36 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
2021-11-30 12:44:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
2022-04-18 09:50:40 -07:00
|
|
|
|
2023-05-30 10:21:12 -07:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-23 23:35:01 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:34 +02:00
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-11-14 23:27:49 +05:30
|
|
|
|
2023-05-30 10:21:12 -07:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
2022-11-14 23:27:49 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
2022-11-14 23:27:49 +05:30
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
2021-10-26 14:01:35 +02:00
|
|
|
|
2021-10-26 14:01:34 +02:00
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
2022-03-08 16:30:49 +01:00
|
|
|
|
2022-09-15 13:11:09 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:49 +01:00
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:34 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2021-10-26 14:01:34 +02:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:34 +02:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:36 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2022-06-14 23:15:48 +02:00
|
|
|
|
2022-06-14 23:15:38 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-14 23:15:48 +02:00
|
|
|
|
|
|
|
|
|
2022-06-14 23:15:38 +02:00
|
|
|
|
|
|
|
|
|
2022-07-11 11:49:50 +02:00
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2022-07-11 11:49:50 +02:00
|
|
|
|
2022-03-22 12:33:31 +01:00
|
|
|
|
|
|
|
|
|
2022-07-11 11:49:50 +02:00
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2022-03-22 12:33:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-07-11 11:49:50 +02:00
|
|
|
|
2022-03-22 12:33:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
|
|
|
|
|
2022-07-11 11:49:50 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-11 16:10:30 -07:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-11 16:10:31 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-02-10 12:32:39 -06:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2021-03-26 16:12:13 +01:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2020-03-27 15:28:45 +00:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2021-10-26 14:01:33 +02:00
|
|
|
|
2021-10-26 14:01:36 +02:00
|
|
|
|
2018-01-11 21:46:23 +00:00
|
|
|
|
2022-06-14 23:15:38 +02:00
|
|
|
|
2022-06-14 23:15:48 +02:00
|
|
|
|
2022-06-14 23:15:38 +02:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2022-04-11 16:10:30 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2021-01-21 15:29:22 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2021-01-21 15:29:22 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2022-04-11 16:10:30 -07:00
|
|
|
|
|
|
|
|
|
2022-06-14 23:15:48 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-08-14 13:44:32 +02:00
|
|
|
|
2022-06-14 23:15:48 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-08-14 13:44:29 +02:00
|
|
|
|
2022-06-14 23:15:48 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2018-06-01 07:23:51 -05:00
|
|
|
|
2023-10-11 19:47:37 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-06-01 07:23:51 -05:00
|
|
|
|
2019-03-06 12:58:15 +01:00
|
|
|
|
2018-06-01 07:23:51 -05:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
|
|
|
|
|
2019-03-06 12:58:15 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:53 -05:00
|
|
|
|
2019-03-06 12:58:15 +01:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-03-06 12:58:15 +01:00
|
|
|
|
2018-06-01 07:23:51 -05:00
|
|
|
|
2022-04-11 16:10:30 -07:00
|
|
|
|
2022-07-11 11:49:50 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-11 16:10:30 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-15 08:53:18 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-03-26 16:12:13 +01:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2020-03-27 15:28:45 +00:00
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-29 22:00:39 -06:00
|
|
|
|
2020-02-17 21:41:54 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2018-01-29 22:00:39 -06:00
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-02-17 21:41:54 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2021-03-26 16:12:03 +01:00
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:33 +02:00
|
|
|
|
2021-10-26 14:01:36 +02:00
|
|
|
|
2021-03-26 16:12:03 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2021-06-24 11:41:02 +02:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-02-08 18:18:03 +01:00
|
|
|
|
2020-12-18 14:19:32 -06:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2023-02-08 18:18:03 +01:00
|
|
|
|
2020-12-18 14:19:32 -06:00
|
|
|
|
2023-02-08 18:18:03 +01:00
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
2023-02-08 18:18:03 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-02-08 18:18:03 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-02-08 18:18:03 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2023-02-08 18:18:03 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2023-02-08 18:18:03 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-29 22:00:40 -06:00
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-29 22:00:40 -06:00
|
|
|
|
|
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2018-01-29 22:00:40 -06:00
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-04 16:30:23 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-29 22:00:39 -06:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2020-12-18 14:19:32 -06:00
|
|
|
|
2018-01-29 22:00:39 -06:00
|
|
|
|
2020-02-10 12:32:40 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-03-26 16:12:13 +01:00
|
|
|
|
2022-09-15 13:11:07 +02:00
|
|
|
|
2020-09-04 16:30:23 +01:00
|
|
|
|
2020-02-10 12:32:40 -06:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-02-10 12:32:40 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-02-10 12:32:39 -06:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-27 15:28:45 +00:00
|
|
|
|
2022-04-11 16:10:31 -07:00
|
|
|
|
2023-02-08 18:18:03 +01:00
|
|
|
|
2022-04-11 16:10:31 -07:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-04-11 16:10:31 -07:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
2020-12-18 14:19:32 -06:00
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-13 16:15:50 +02:00
|
|
|
|
|
|
|
|
|
2021-05-06 21:34:04 +02:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:39 -07:00
|
|
|
|
2021-05-06 21:34:03 +02:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:12 -07:00
|
|
|
|
|
|
|
|
|
2021-05-06 21:34:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-13 16:15:50 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-05-06 21:34:03 +02:00
|
|
|
|
|
|
|
|
|
2021-05-06 21:34:04 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-27 15:28:41 +00:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-03-27 15:28:41 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-11 21:46:24 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2019-03-01 11:19:03 +01:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
objtool: Make instruction::alts a single-linked list
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
unsigned int len; /* 64 4 */
enum insn_type type; /* 68 4 */
long unsigned int immediate; /* 72 8 */
u16 dead_end:1; /* 80: 0 2 */
u16 ignore:1; /* 80: 1 2 */
u16 ignore_alts:1; /* 80: 2 2 */
u16 hint:1; /* 80: 3 2 */
u16 save:1; /* 80: 4 2 */
u16 restore:1; /* 80: 5 2 */
u16 retpoline_safe:1; /* 80: 6 2 */
u16 noendbr:1; /* 80: 7 2 */
u16 entry:1; /* 80: 8 2 */
/* XXX 7 bits hole, try to pack */
s8 instr; /* 82 1 */
u8 visited; /* 83 1 */
/* XXX 4 bytes hole, try to pack */
struct alt_group * alt_group; /* 88 8 */
struct symbol * call_dest; /* 96 8 */
struct instruction * jump_dest; /* 104 8 */
struct instruction * first_jump_src; /* 112 8 */
struct reloc * jump_table; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
struct reloc * reloc; /* 128 8 */
- struct list_head alts; /* 136 16 */
- struct symbol * sym; /* 152 8 */
- struct stack_op * stack_ops; /* 160 8 */
- struct cfi_state * cfi; /* 168 8 */
+ struct alternative * alts; /* 136 8 */
+ struct symbol * sym; /* 144 8 */
+ struct stack_op * stack_ops; /* 152 8 */
+ struct cfi_state * cfi; /* 160 8 */
- /* size: 176, cachelines: 3, members: 29 */
- /* sum members: 170, holes: 1, sum holes: 4 */
+ /* size: 168, cachelines: 3, members: 29 */
+ /* sum members: 162, holes: 1, sum holes: 4 */
/* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */
- /* last cacheline: 48 bytes */
+ /* last cacheline: 40 bytes */
};
pre: 5:58.50 real, 229.64 user, 128.65 sys, 26221520 mem
post: 5:48.86 real, 220.30 user, 128.34 sys, 24834672 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.430556498@infradead.org
2023-02-08 18:17:59 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2021-05-06 21:34:04 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:53 -05:00
|
|
|
|
2023-05-30 10:21:09 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2023-05-30 10:21:09 -07:00
|
|
|
|
|
|
|
|
|
2018-05-10 17:48:49 -05:00
|
|
|
|
2023-05-30 10:21:09 -07:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2019-07-17 20:36:53 -05:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2019-07-17 20:36:53 -05:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:02 -07:00
|
|
|
|
objtool: Support repeated uses of the same C jump table
This fixes objtool for both a GCC issue and a Clang issue:
1) GCC issue:
kernel/bpf/core.o: warning: objtool: ___bpf_prog_run()+0x8d5: sibling call from callable instruction with modified stack frame
With CONFIG_RETPOLINE=n, GCC is doing the following optimization in
___bpf_prog_run().
Before:
select_insn:
jmp *jumptable(,%rax,8)
...
ALU64_ADD_X:
...
jmp select_insn
ALU_ADD_X:
...
jmp select_insn
After:
select_insn:
jmp *jumptable(, %rax, 8)
...
ALU64_ADD_X:
...
jmp *jumptable(, %rax, 8)
ALU_ADD_X:
...
jmp *jumptable(, %rax, 8)
This confuses objtool. It has never seen multiple indirect jump
sites which use the same jump table.
For GCC switch tables, the only way of detecting the size of a table
is by continuing to scan for more tables. The size of the previous
table can only be determined after another switch table is found, or
when the scan reaches the end of the function.
That logic was reused for C jump tables, and was based on the
assumption that each jump table only has a single jump site. The
above optimization breaks that assumption.
2) Clang issue:
drivers/usb/misc/sisusbvga/sisusb.o: warning: objtool: sisusb_write_mem_bulk()+0x588: can't find switch jump table
With clang 9, code can be generated where a function contains two
indirect jump instructions which use the same switch table.
The fix is the same for both issues: split the jump table parsing into
two passes.
In the first pass, locate the heads of all switch tables for the
function and mark their locations.
In the second pass, parse the switch tables and add them.
Fixes: e55a73251da3 ("bpf: Fix ORC unwinding in non-JIT BPF code")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Nick Desaulniers <ndesaulniers@google.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/e995befaada9d4d8b2cf788ff3f566ba900d2b4d.1563413318.git.jpoimboe@redhat.com
Co-developed-by: Josh Poimboeuf <jpoimboe@redhat.com>
2019-07-17 20:36:54 -05:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:09 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:53 -05:00
|
|
|
|
2023-05-30 10:21:06 -07:00
|
|
|
|
2018-05-10 17:48:49 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2018-05-10 17:48:49 -05:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2019-07-17 20:36:53 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:53 -05:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2018-05-09 22:39:15 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:53 -05:00
|
|
|
|
objtool: Make instruction::alts a single-linked list
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
unsigned int len; /* 64 4 */
enum insn_type type; /* 68 4 */
long unsigned int immediate; /* 72 8 */
u16 dead_end:1; /* 80: 0 2 */
u16 ignore:1; /* 80: 1 2 */
u16 ignore_alts:1; /* 80: 2 2 */
u16 hint:1; /* 80: 3 2 */
u16 save:1; /* 80: 4 2 */
u16 restore:1; /* 80: 5 2 */
u16 retpoline_safe:1; /* 80: 6 2 */
u16 noendbr:1; /* 80: 7 2 */
u16 entry:1; /* 80: 8 2 */
/* XXX 7 bits hole, try to pack */
s8 instr; /* 82 1 */
u8 visited; /* 83 1 */
/* XXX 4 bytes hole, try to pack */
struct alt_group * alt_group; /* 88 8 */
struct symbol * call_dest; /* 96 8 */
struct instruction * jump_dest; /* 104 8 */
struct instruction * first_jump_src; /* 112 8 */
struct reloc * jump_table; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
struct reloc * reloc; /* 128 8 */
- struct list_head alts; /* 136 16 */
- struct symbol * sym; /* 152 8 */
- struct stack_op * stack_ops; /* 160 8 */
- struct cfi_state * cfi; /* 168 8 */
+ struct alternative * alts; /* 136 8 */
+ struct symbol * sym; /* 144 8 */
+ struct stack_op * stack_ops; /* 152 8 */
+ struct cfi_state * cfi; /* 160 8 */
- /* size: 176, cachelines: 3, members: 29 */
- /* sum members: 170, holes: 1, sum holes: 4 */
+ /* size: 168, cachelines: 3, members: 29 */
+ /* sum members: 162, holes: 1, sum holes: 4 */
/* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */
- /* last cacheline: 48 bytes */
+ /* last cacheline: 40 bytes */
};
pre: 5:58.50 real, 229.64 user, 128.65 sys, 26221520 mem
post: 5:48.86 real, 220.30 user, 128.34 sys, 24834672 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.430556498@infradead.org
2023-02-08 18:17:59 +01:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:06 -07:00
|
|
|
|
2018-05-10 17:48:49 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2018-05-10 17:48:49 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-04 16:30:24 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-04 16:30:24 +01:00
|
|
|
|
2020-02-17 21:41:53 -06:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2018-02-08 14:02:32 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-05-18 15:10:34 -05:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2020-04-28 16:45:16 -05:00
|
|
|
|
2018-02-08 14:02:32 +01:00
|
|
|
|
2018-05-18 15:10:34 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-04 16:30:24 +01:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2019-07-17 20:36:53 -05:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2020-02-17 21:41:53 -06:00
|
|
|
|
2018-05-18 15:10:34 -05:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Support repeated uses of the same C jump table
This fixes objtool for both a GCC issue and a Clang issue:
1) GCC issue:
kernel/bpf/core.o: warning: objtool: ___bpf_prog_run()+0x8d5: sibling call from callable instruction with modified stack frame
With CONFIG_RETPOLINE=n, GCC is doing the following optimization in
___bpf_prog_run().
Before:
select_insn:
jmp *jumptable(,%rax,8)
...
ALU64_ADD_X:
...
jmp select_insn
ALU_ADD_X:
...
jmp select_insn
After:
select_insn:
jmp *jumptable(, %rax, 8)
...
ALU64_ADD_X:
...
jmp *jumptable(, %rax, 8)
ALU_ADD_X:
...
jmp *jumptable(, %rax, 8)
This confuses objtool. It has never seen multiple indirect jump
sites which use the same jump table.
For GCC switch tables, the only way of detecting the size of a table
is by continuing to scan for more tables. The size of the previous
table can only be determined after another switch table is found, or
when the scan reaches the end of the function.
That logic was reused for C jump tables, and was based on the
assumption that each jump table only has a single jump site. The
above optimization breaks that assumption.
2) Clang issue:
drivers/usb/misc/sisusbvga/sisusb.o: warning: objtool: sisusb_write_mem_bulk()+0x588: can't find switch jump table
With clang 9, code can be generated where a function contains two
indirect jump instructions which use the same switch table.
The fix is the same for both issues: split the jump table parsing into
two passes.
In the first pass, locate the heads of all switch tables for the
function and mark their locations.
In the second pass, parse the switch tables and add them.
Fixes: e55a73251da3 ("bpf: Fix ORC unwinding in non-JIT BPF code")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Nick Desaulniers <ndesaulniers@google.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/e995befaada9d4d8b2cf788ff3f566ba900d2b4d.1563413318.git.jpoimboe@redhat.com
Co-developed-by: Josh Poimboeuf <jpoimboe@redhat.com>
2019-07-17 20:36:54 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
objtool: Support repeated uses of the same C jump table
This fixes objtool for both a GCC issue and a Clang issue:
1) GCC issue:
kernel/bpf/core.o: warning: objtool: ___bpf_prog_run()+0x8d5: sibling call from callable instruction with modified stack frame
With CONFIG_RETPOLINE=n, GCC is doing the following optimization in
___bpf_prog_run().
Before:
select_insn:
jmp *jumptable(,%rax,8)
...
ALU64_ADD_X:
...
jmp select_insn
ALU_ADD_X:
...
jmp select_insn
After:
select_insn:
jmp *jumptable(, %rax, 8)
...
ALU64_ADD_X:
...
jmp *jumptable(, %rax, 8)
ALU_ADD_X:
...
jmp *jumptable(, %rax, 8)
This confuses objtool. It has never seen multiple indirect jump
sites which use the same jump table.
For GCC switch tables, the only way of detecting the size of a table
is by continuing to scan for more tables. The size of the previous
table can only be determined after another switch table is found, or
when the scan reaches the end of the function.
That logic was reused for C jump tables, and was based on the
assumption that each jump table only has a single jump site. The
above optimization breaks that assumption.
2) Clang issue:
drivers/usb/misc/sisusbvga/sisusb.o: warning: objtool: sisusb_write_mem_bulk()+0x588: can't find switch jump table
With clang 9, code can be generated where a function contains two
indirect jump instructions which use the same switch table.
The fix is the same for both issues: split the jump table parsing into
two passes.
In the first pass, locate the heads of all switch tables for the
function and mark their locations.
In the second pass, parse the switch tables and add them.
Fixes: e55a73251da3 ("bpf: Fix ORC unwinding in non-JIT BPF code")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Nick Desaulniers <ndesaulniers@google.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/e995befaada9d4d8b2cf788ff3f566ba900d2b4d.1563413318.git.jpoimboe@redhat.com
Co-developed-by: Josh Poimboeuf <jpoimboe@redhat.com>
2019-07-17 20:36:54 -05:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2020-03-10 18:27:24 +01:00
|
|
|
|
2018-02-08 14:02:32 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:53 -05:00
|
|
|
|
2018-02-08 14:02:32 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2023-05-30 10:21:09 -07:00
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
objtool: Support repeated uses of the same C jump table
This fixes objtool for both a GCC issue and a Clang issue:
1) GCC issue:
kernel/bpf/core.o: warning: objtool: ___bpf_prog_run()+0x8d5: sibling call from callable instruction with modified stack frame
With CONFIG_RETPOLINE=n, GCC is doing the following optimization in
___bpf_prog_run().
Before:
select_insn:
jmp *jumptable(,%rax,8)
...
ALU64_ADD_X:
...
jmp select_insn
ALU_ADD_X:
...
jmp select_insn
After:
select_insn:
jmp *jumptable(, %rax, 8)
...
ALU64_ADD_X:
...
jmp *jumptable(, %rax, 8)
ALU_ADD_X:
...
jmp *jumptable(, %rax, 8)
This confuses objtool. It has never seen multiple indirect jump
sites which use the same jump table.
For GCC switch tables, the only way of detecting the size of a table
is by continuing to scan for more tables. The size of the previous
table can only be determined after another switch table is found, or
when the scan reaches the end of the function.
That logic was reused for C jump tables, and was based on the
assumption that each jump table only has a single jump site. The
above optimization breaks that assumption.
2) Clang issue:
drivers/usb/misc/sisusbvga/sisusb.o: warning: objtool: sisusb_write_mem_bulk()+0x588: can't find switch jump table
With clang 9, code can be generated where a function contains two
indirect jump instructions which use the same switch table.
The fix is the same for both issues: split the jump table parsing into
two passes.
In the first pass, locate the heads of all switch tables for the
function and mark their locations.
In the second pass, parse the switch tables and add them.
Fixes: e55a73251da3 ("bpf: Fix ORC unwinding in non-JIT BPF code")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Nick Desaulniers <ndesaulniers@google.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/e995befaada9d4d8b2cf788ff3f566ba900d2b4d.1563413318.git.jpoimboe@redhat.com
Co-developed-by: Josh Poimboeuf <jpoimboe@redhat.com>
2019-07-17 20:36:54 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:09 -07:00
|
|
|
|
|
|
|
|
|
objtool: Support repeated uses of the same C jump table
This fixes objtool for both a GCC issue and a Clang issue:
1) GCC issue:
kernel/bpf/core.o: warning: objtool: ___bpf_prog_run()+0x8d5: sibling call from callable instruction with modified stack frame
With CONFIG_RETPOLINE=n, GCC is doing the following optimization in
___bpf_prog_run().
Before:
select_insn:
jmp *jumptable(,%rax,8)
...
ALU64_ADD_X:
...
jmp select_insn
ALU_ADD_X:
...
jmp select_insn
After:
select_insn:
jmp *jumptable(, %rax, 8)
...
ALU64_ADD_X:
...
jmp *jumptable(, %rax, 8)
ALU_ADD_X:
...
jmp *jumptable(, %rax, 8)
This confuses objtool. It has never seen multiple indirect jump
sites which use the same jump table.
For GCC switch tables, the only way of detecting the size of a table
is by continuing to scan for more tables. The size of the previous
table can only be determined after another switch table is found, or
when the scan reaches the end of the function.
That logic was reused for C jump tables, and was based on the
assumption that each jump table only has a single jump site. The
above optimization breaks that assumption.
2) Clang issue:
drivers/usb/misc/sisusbvga/sisusb.o: warning: objtool: sisusb_write_mem_bulk()+0x588: can't find switch jump table
With clang 9, code can be generated where a function contains two
indirect jump instructions which use the same switch table.
The fix is the same for both issues: split the jump table parsing into
two passes.
In the first pass, locate the heads of all switch tables for the
function and mark their locations.
In the second pass, parse the switch tables and add them.
Fixes: e55a73251da3 ("bpf: Fix ORC unwinding in non-JIT BPF code")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Nick Desaulniers <ndesaulniers@google.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/e995befaada9d4d8b2cf788ff3f566ba900d2b4d.1563413318.git.jpoimboe@redhat.com
Co-developed-by: Josh Poimboeuf <jpoimboe@redhat.com>
2019-07-17 20:36:54 -05:00
|
|
|
|
2020-03-10 18:27:24 +01:00
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
objtool: Support repeated uses of the same C jump table
This fixes objtool for both a GCC issue and a Clang issue:
1) GCC issue:
kernel/bpf/core.o: warning: objtool: ___bpf_prog_run()+0x8d5: sibling call from callable instruction with modified stack frame
With CONFIG_RETPOLINE=n, GCC is doing the following optimization in
___bpf_prog_run().
Before:
select_insn:
jmp *jumptable(,%rax,8)
...
ALU64_ADD_X:
...
jmp select_insn
ALU_ADD_X:
...
jmp select_insn
After:
select_insn:
jmp *jumptable(, %rax, 8)
...
ALU64_ADD_X:
...
jmp *jumptable(, %rax, 8)
ALU_ADD_X:
...
jmp *jumptable(, %rax, 8)
This confuses objtool. It has never seen multiple indirect jump
sites which use the same jump table.
For GCC switch tables, the only way of detecting the size of a table
is by continuing to scan for more tables. The size of the previous
table can only be determined after another switch table is found, or
when the scan reaches the end of the function.
That logic was reused for C jump tables, and was based on the
assumption that each jump table only has a single jump site. The
above optimization breaks that assumption.
2) Clang issue:
drivers/usb/misc/sisusbvga/sisusb.o: warning: objtool: sisusb_write_mem_bulk()+0x588: can't find switch jump table
With clang 9, code can be generated where a function contains two
indirect jump instructions which use the same switch table.
The fix is the same for both issues: split the jump table parsing into
two passes.
In the first pass, locate the heads of all switch tables for the
function and mark their locations.
In the second pass, parse the switch tables and add them.
Fixes: e55a73251da3 ("bpf: Fix ORC unwinding in non-JIT BPF code")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Nick Desaulniers <ndesaulniers@google.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/e995befaada9d4d8b2cf788ff3f566ba900d2b4d.1563413318.git.jpoimboe@redhat.com
Co-developed-by: Josh Poimboeuf <jpoimboe@redhat.com>
2019-07-17 20:36:54 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-05-30 10:21:09 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:09 -07:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:09 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:53 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-09-07 08:12:01 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-21 15:29:24 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:12 -08:00
|
|
|
|
2021-01-21 15:29:24 -06:00
|
|
|
|
|
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-08-22 18:50:37 -04:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-08-22 18:50:37 -04:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-21 15:29:24 -06:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
2023-06-05 09:12:21 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-24 12:52:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
|
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-21 15:29:24 -06:00
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-11-14 23:27:47 +05:30
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
2023-02-16 12:34:41 -08:00
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
|
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:52 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2022-03-08 16:30:52 +01:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:52 +01:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:02 -07:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:52 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-16 10:24:06 +01:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2018-01-16 10:24:06 +01:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2018-01-16 10:24:06 +01:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
2018-01-16 10:24:06 +01:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:02 -07:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2018-01-16 10:24:06 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2018-01-16 10:24:06 +01:00
|
|
|
|
2018-03-06 17:58:15 -06:00
|
|
|
|
2018-01-16 10:24:06 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-14 23:15:59 +02:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2018-01-16 10:24:06 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:02 -07:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:02 -07:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:02 -07:00
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-04-14 12:36:12 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2020-04-14 12:36:12 +02:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
2020-04-14 12:36:12 +02:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:02 -07:00
|
|
|
|
2020-04-14 12:36:12 +02:00
|
|
|
|
|
|
|
|
|
objtool: Rename rela to reloc
Before supporting additional relocation types rename the relevant
types and functions from "rela" to "reloc". This work be done with
the following regex:
sed -e 's/struct rela/struct reloc/g' \
-e 's/\([_\*]\)rela\(s\{0,1\}\)/\1reloc\2/g' \
-e 's/tmprela\(s\{0,1\}\)/tmpreloc\1/g' \
-e 's/relasec/relocsec/g' \
-e 's/rela_list/reloc_list/g' \
-e 's/rela_hash/reloc_hash/g' \
-e 's/add_rela/add_reloc/g' \
-e 's/rela->/reloc->/g' \
-e '/rela[,\.]/{ s/\([^\.>]\)rela\([\.,]\)/\1reloc\2/g ; }' \
-e 's/rela =/reloc =/g' \
-e 's/relas =/relocs =/g' \
-e 's/relas\[/relocs[/g' \
-e 's/relaname =/relocname =/g' \
-e 's/= rela\;/= reloc\;/g' \
-e 's/= relas\;/= relocs\;/g' \
-e 's/= relaname\;/= relocname\;/g' \
-e 's/, rela)/, reloc)/g' \
-e 's/\([ @]\)rela\([ "]\)/\1reloc\2/g' \
-e 's/ rela$/ reloc/g' \
-e 's/, relaname/, relocname/g' \
-e 's/sec->rela/sec->reloc/g' \
-e 's/(\(!\{0,1\}\)rela/(\1reloc/g' \
-i \
arch.h \
arch/x86/decode.c \
check.c \
check.h \
elf.c \
elf.h \
orc_gen.c \
special.c
Notable exceptions which complicate the regex include gelf_*
library calls and standard/expected section names which still use
"rela" because they encode the type of relocation expected. Also, keep
"rela" in the struct because it encodes a specific type of relocation
we currently expect.
It will eventually turn into a member of an anonymous union when a
susequent patch adds implicit addend, or "rel", relocation support.
Signed-off-by: Matt Helsley <mhelsley@vmware.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-05-29 14:01:13 -07:00
|
|
|
|
2020-04-14 12:36:12 +02:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2020-04-14 12:36:12 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2020-04-14 12:36:12 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-04-14 12:36:12 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-08-18 09:45:53 +08:00
|
|
|
|
2020-04-14 12:36:12 +02:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-04-14 12:36:12 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-11-30 12:44:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:33 +02:00
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:33 +02:00
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:33 +02:00
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:33 +02:00
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
|
|
|
|
|
2022-06-14 23:15:38 +02:00
|
|
|
|
2023-08-14 13:44:29 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:33 +02:00
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
|
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-09-07 08:12:01 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-06-27 20:50:46 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-09-07 08:12:01 -05:00
|
|
|
|
|
|
|
|
|
2020-04-12 22:44:05 +08:00
|
|
|
|
|
|
|
|
|
2018-09-07 08:12:01 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-09-07 08:12:01 -05:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2019-03-18 14:33:07 +01:00
|
|
|
|
2018-01-11 21:46:24 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:52 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-03-26 16:12:12 +01:00
|
|
|
|
2022-04-11 16:10:31 -07:00
|
|
|
|
|
|
|
|
|
2021-03-26 16:12:12 +01:00
|
|
|
|
2022-11-14 23:27:50 +05:30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2022-04-11 16:10:31 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-03-26 16:12:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-04-14 12:36:12 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-29 22:00:39 -06:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:49 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:53 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-16 10:24:06 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-08-16 13:59:21 +02:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-08-16 13:59:21 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-04-01 16:38:19 +02:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2020-04-01 16:38:19 +02:00
|
|
|
|
|
|
|
|
|
2021-01-21 15:29:24 -06:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2021-01-21 15:29:24 -06:00
|
|
|
|
2020-04-01 16:38:19 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2020-04-01 16:38:19 +02:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-10-14 08:38:00 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2020-10-14 08:38:00 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2020-03-27 15:28:47 +00:00
|
|
|
|
2017-07-11 10:33:42 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-07-11 10:33:42 -05:00
|
|
|
|
2020-04-25 05:03:00 -05:00
|
|
|
|
2017-07-11 10:33:42 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
2017-07-11 10:33:42 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
2017-07-11 10:33:42 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
objtool: Track DRAP separately from callee-saved registers
When GCC realigns a function's stack, it sometimes uses %r13 as the DRAP
register, like:
push %r13
lea 0x10(%rsp), %r13
and $0xfffffffffffffff0, %rsp
pushq -0x8(%r13)
push %rbp
mov %rsp, %rbp
push %r13
...
mov -0x8(%rbp),%r13
leaveq
lea -0x10(%r13), %rsp
pop %r13
retq
Since %r13 was pushed onto the stack twice, its two stack locations need
to be stored separately. The first push of %r13 is its original value,
and the second push of %r13 is the caller's stack frame address.
Since %r13 is a callee-saved register, we need to track the stack
location of its original value separately from the DRAP register.
This fixes the following false positive warning:
lib/ubsan.o: warning: objtool: val_to_string.constprop.7()+0x97: leave instruction with modified stack frame
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: baa41469a7b9 ("objtool: Implement stack validation 2.0")
Link: http://lkml.kernel.org/r/3da23a6d4c5b3c1e21fc2ccc21a73941b97ff20a.1502401017.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-08-10 16:37:26 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-02-11 13:03:28 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2023-06-05 09:12:21 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-04 16:30:27 +01:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-07-11 10:33:42 -05:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-09-20 16:24:32 -05:00
|
|
|
|
|
|
|
|
|
2020-10-14 08:38:00 +01:00
|
|
|
|
2017-09-20 16:24:32 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-09-20 16:24:32 -05:00
|
|
|
|
2017-08-29 12:51:03 -05:00
|
|
|
|
2017-09-20 16:24:32 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-08-29 12:51:03 -05:00
|
|
|
|
2017-09-20 16:24:32 -05:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2017-09-20 16:24:32 -05:00
|
|
|
|
2017-08-29 12:51:03 -05:00
|
|
|
|
2017-09-20 16:24:32 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2017-08-29 12:51:03 -05:00
|
|
|
|
|
|
|
|
|
2018-03-22 13:00:37 -05:00
|
|
|
|
2021-02-09 21:41:13 +01:00
|
|
|
|
2018-03-22 13:00:37 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2018-03-22 13:00:37 -05:00
|
|
|
|
|
|
|
|
|
2017-08-29 12:51:03 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-08-29 12:51:03 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2017-08-29 12:51:03 -05:00
|
|
|
|
2021-02-03 12:02:17 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-08-29 12:51:03 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2021-02-18 17:14:10 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-08-29 12:51:03 -05:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-08-29 12:51:03 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-08-29 12:51:03 -05:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-02-11 13:03:28 +01:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
2021-02-03 12:02:17 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
objtool: Track DRAP separately from callee-saved registers
When GCC realigns a function's stack, it sometimes uses %r13 as the DRAP
register, like:
push %r13
lea 0x10(%rsp), %r13
and $0xfffffffffffffff0, %rsp
pushq -0x8(%r13)
push %rbp
mov %rsp, %rbp
push %r13
...
mov -0x8(%rbp),%r13
leaveq
lea -0x10(%r13), %rsp
pop %r13
retq
Since %r13 was pushed onto the stack twice, its two stack locations need
to be stored separately. The first push of %r13 is its original value,
and the second push of %r13 is the caller's stack frame address.
Since %r13 is a callee-saved register, we need to track the stack
location of its original value separately from the DRAP register.
This fixes the following false positive warning:
lib/ubsan.o: warning: objtool: val_to_string.constprop.7()+0x97: leave instruction with modified stack frame
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: baa41469a7b9 ("objtool: Implement stack validation 2.0")
Link: http://lkml.kernel.org/r/3da23a6d4c5b3c1e21fc2ccc21a73941b97ff20a.1502401017.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-08-10 16:37:26 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
objtool: Track DRAP separately from callee-saved registers
When GCC realigns a function's stack, it sometimes uses %r13 as the DRAP
register, like:
push %r13
lea 0x10(%rsp), %r13
and $0xfffffffffffffff0, %rsp
pushq -0x8(%r13)
push %rbp
mov %rsp, %rbp
push %r13
...
mov -0x8(%rbp),%r13
leaveq
lea -0x10(%r13), %rsp
pop %r13
retq
Since %r13 was pushed onto the stack twice, its two stack locations need
to be stored separately. The first push of %r13 is its original value,
and the second push of %r13 is the caller's stack frame address.
Since %r13 is a callee-saved register, we need to track the stack
location of its original value separately from the DRAP register.
This fixes the following false positive warning:
lib/ubsan.o: warning: objtool: val_to_string.constprop.7()+0x97: leave instruction with modified stack frame
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: baa41469a7b9 ("objtool: Implement stack validation 2.0")
Link: http://lkml.kernel.org/r/3da23a6d4c5b3c1e21fc2ccc21a73941b97ff20a.1502401017.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-08-10 16:37:26 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2021-02-09 21:41:13 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
objtool: Track DRAP separately from callee-saved registers
When GCC realigns a function's stack, it sometimes uses %r13 as the DRAP
register, like:
push %r13
lea 0x10(%rsp), %r13
and $0xfffffffffffffff0, %rsp
pushq -0x8(%r13)
push %rbp
mov %rsp, %rbp
push %r13
...
mov -0x8(%rbp),%r13
leaveq
lea -0x10(%r13), %rsp
pop %r13
retq
Since %r13 was pushed onto the stack twice, its two stack locations need
to be stored separately. The first push of %r13 is its original value,
and the second push of %r13 is the caller's stack frame address.
Since %r13 is a callee-saved register, we need to track the stack
location of its original value separately from the DRAP register.
This fixes the following false positive warning:
lib/ubsan.o: warning: objtool: val_to_string.constprop.7()+0x97: leave instruction with modified stack frame
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: baa41469a7b9 ("objtool: Implement stack validation 2.0")
Link: http://lkml.kernel.org/r/3da23a6d4c5b3c1e21fc2ccc21a73941b97ff20a.1502401017.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-08-10 16:37:26 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-10-14 08:38:02 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
objtool: Track DRAP separately from callee-saved registers
When GCC realigns a function's stack, it sometimes uses %r13 as the DRAP
register, like:
push %r13
lea 0x10(%rsp), %r13
and $0xfffffffffffffff0, %rsp
pushq -0x8(%r13)
push %rbp
mov %rsp, %rbp
push %r13
...
mov -0x8(%rbp),%r13
leaveq
lea -0x10(%r13), %rsp
pop %r13
retq
Since %r13 was pushed onto the stack twice, its two stack locations need
to be stored separately. The first push of %r13 is its original value,
and the second push of %r13 is the caller's stack frame address.
Since %r13 is a callee-saved register, we need to track the stack
location of its original value separately from the DRAP register.
This fixes the following false positive warning:
lib/ubsan.o: warning: objtool: val_to_string.constprop.7()+0x97: leave instruction with modified stack frame
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: baa41469a7b9 ("objtool: Implement stack validation 2.0")
Link: http://lkml.kernel.org/r/3da23a6d4c5b3c1e21fc2ccc21a73941b97ff20a.1502401017.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-08-10 16:37:26 -05:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
objtool: Track DRAP separately from callee-saved registers
When GCC realigns a function's stack, it sometimes uses %r13 as the DRAP
register, like:
push %r13
lea 0x10(%rsp), %r13
and $0xfffffffffffffff0, %rsp
pushq -0x8(%r13)
push %rbp
mov %rsp, %rbp
push %r13
...
mov -0x8(%rbp),%r13
leaveq
lea -0x10(%r13), %rsp
pop %r13
retq
Since %r13 was pushed onto the stack twice, its two stack locations need
to be stored separately. The first push of %r13 is its original value,
and the second push of %r13 is the caller's stack frame address.
Since %r13 is a callee-saved register, we need to track the stack
location of its original value separately from the DRAP register.
This fixes the following false positive warning:
lib/ubsan.o: warning: objtool: val_to_string.constprop.7()+0x97: leave instruction with modified stack frame
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: baa41469a7b9 ("objtool: Implement stack validation 2.0")
Link: http://lkml.kernel.org/r/3da23a6d4c5b3c1e21fc2ccc21a73941b97ff20a.1502401017.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-08-10 16:37:26 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
objtool: Track DRAP separately from callee-saved registers
When GCC realigns a function's stack, it sometimes uses %r13 as the DRAP
register, like:
push %r13
lea 0x10(%rsp), %r13
and $0xfffffffffffffff0, %rsp
pushq -0x8(%r13)
push %rbp
mov %rsp, %rbp
push %r13
...
mov -0x8(%rbp),%r13
leaveq
lea -0x10(%r13), %rsp
pop %r13
retq
Since %r13 was pushed onto the stack twice, its two stack locations need
to be stored separately. The first push of %r13 is its original value,
and the second push of %r13 is the caller's stack frame address.
Since %r13 is a callee-saved register, we need to track the stack
location of its original value separately from the DRAP register.
This fixes the following false positive warning:
lib/ubsan.o: warning: objtool: val_to_string.constprop.7()+0x97: leave instruction with modified stack frame
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: baa41469a7b9 ("objtool: Implement stack validation 2.0")
Link: http://lkml.kernel.org/r/3da23a6d4c5b3c1e21fc2ccc21a73941b97ff20a.1502401017.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-08-10 16:37:26 -05:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2020-10-14 08:38:02 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
objtool: Track DRAP separately from callee-saved registers
When GCC realigns a function's stack, it sometimes uses %r13 as the DRAP
register, like:
push %r13
lea 0x10(%rsp), %r13
and $0xfffffffffffffff0, %rsp
pushq -0x8(%r13)
push %rbp
mov %rsp, %rbp
push %r13
...
mov -0x8(%rbp),%r13
leaveq
lea -0x10(%r13), %rsp
pop %r13
retq
Since %r13 was pushed onto the stack twice, its two stack locations need
to be stored separately. The first push of %r13 is its original value,
and the second push of %r13 is the caller's stack frame address.
Since %r13 is a callee-saved register, we need to track the stack
location of its original value separately from the DRAP register.
This fixes the following false positive warning:
lib/ubsan.o: warning: objtool: val_to_string.constprop.7()+0x97: leave instruction with modified stack frame
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: baa41469a7b9 ("objtool: Implement stack validation 2.0")
Link: http://lkml.kernel.org/r/3da23a6d4c5b3c1e21fc2ccc21a73941b97ff20a.1502401017.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-08-10 16:37:26 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2020-09-15 08:53:16 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Track DRAP separately from callee-saved registers
When GCC realigns a function's stack, it sometimes uses %r13 as the DRAP
register, like:
push %r13
lea 0x10(%rsp), %r13
and $0xfffffffffffffff0, %rsp
pushq -0x8(%r13)
push %rbp
mov %rsp, %rbp
push %r13
...
mov -0x8(%rbp),%r13
leaveq
lea -0x10(%r13), %rsp
pop %r13
retq
Since %r13 was pushed onto the stack twice, its two stack locations need
to be stored separately. The first push of %r13 is its original value,
and the second push of %r13 is the caller's stack frame address.
Since %r13 is a callee-saved register, we need to track the stack
location of its original value separately from the DRAP register.
This fixes the following false positive warning:
lib/ubsan.o: warning: objtool: val_to_string.constprop.7()+0x97: leave instruction with modified stack frame
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: baa41469a7b9 ("objtool: Implement stack validation 2.0")
Link: http://lkml.kernel.org/r/3da23a6d4c5b3c1e21fc2ccc21a73941b97ff20a.1502401017.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2017-08-10 16:37:26 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2020-09-15 08:53:16 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2020-10-14 08:38:02 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-02-03 12:02:17 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-27 15:28:47 +00:00
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
2020-03-27 15:28:47 +00:00
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
2020-03-27 15:28:47 +00:00
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
2020-03-27 15:28:47 +00:00
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
2023-02-08 18:18:03 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2023-02-08 18:18:03 +01:00
|
|
|
|
2020-05-08 12:34:33 +02:00
|
|
|
|
|
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-02-11 13:03:28 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Make instruction::stack_ops a single-linked list
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
unsigned int len; /* 64 4 */
enum insn_type type; /* 68 4 */
long unsigned int immediate; /* 72 8 */
u16 dead_end:1; /* 80: 0 2 */
u16 ignore:1; /* 80: 1 2 */
u16 ignore_alts:1; /* 80: 2 2 */
u16 hint:1; /* 80: 3 2 */
u16 save:1; /* 80: 4 2 */
u16 restore:1; /* 80: 5 2 */
u16 retpoline_safe:1; /* 80: 6 2 */
u16 noendbr:1; /* 80: 7 2 */
u16 entry:1; /* 80: 8 2 */
/* XXX 7 bits hole, try to pack */
s8 instr; /* 82 1 */
u8 visited; /* 83 1 */
/* XXX 4 bytes hole, try to pack */
struct alt_group * alt_group; /* 88 8 */
struct symbol * call_dest; /* 96 8 */
struct instruction * jump_dest; /* 104 8 */
struct instruction * first_jump_src; /* 112 8 */
struct reloc * jump_table; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
struct reloc * reloc; /* 128 8 */
struct list_head alts; /* 136 16 */
struct symbol * sym; /* 152 8 */
- struct list_head stack_ops; /* 160 16 */
- struct cfi_state * cfi; /* 176 8 */
+ struct stack_op * stack_ops; /* 160 8 */
+ struct cfi_state * cfi; /* 168 8 */
- /* size: 184, cachelines: 3, members: 29 */
- /* sum members: 178, holes: 1, sum holes: 4 */
+ /* size: 176, cachelines: 3, members: 29 */
+ /* sum members: 170, holes: 1, sum holes: 4 */
/* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */
- /* last cacheline: 56 bytes */
+ /* last cacheline: 48 bytes */
};
pre: 5:58.22 real, 226.69 user, 131.22 sys, 26221520 mem
post: 5:58.50 real, 229.64 user, 128.65 sys, 26221520 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.362196959@infradead.org
2023-02-08 18:17:58 +01:00
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
2021-02-11 13:03:28 +01:00
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
2020-05-08 12:34:33 +02:00
|
|
|
|
2021-03-08 15:46:04 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-27 15:28:47 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-03-27 15:28:47 +00:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-07-11 10:33:42 -05:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:48 -05:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:52 -05:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
objtool: Fix truncated string warning
On GCC 12, the build fails due to a possible truncated string:
check.c: In function 'validate_call':
check.c:2865:58: error: '%d' directive output may be truncated writing between 1 and 10 bytes into a region of size 9 [-Werror=format-truncation=]
2865 | snprintf(pvname, sizeof(pvname), "pv_ops[%d]", idx);
| ^~
In theory it's a valid bug:
static char pvname[16];
int idx;
...
idx = (rel->addend / sizeof(void *));
snprintf(pvname, sizeof(pvname), "pv_ops[%d]", idx);
There are only 7 chars for %d while it could take up to 9, so the
printed "pv_ops[%d]" string could get truncated.
In reality the bug should never happen, because pv_ops only has ~80
entries, so 7 chars for the integer is more than enough. Still, it's
worth fixing. Bump the buffer size by 2 bytes to silence the warning.
[ jpoimboe: changed size to 19; massaged changelog ]
Fixes: db2b0c5d7b6f ("objtool: Support pv_opsindirect calls for noinstr")
Reported-by: Adam Borowski <kilobyte@angband.pl>
Reported-by: Martin Liška <mliska@suse.cz>
Signed-off-by: Sergei Trofimovich <slyich@gmail.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Link: https://lore.kernel.org/r/20220120233748.2062559-1-slyich@gmail.com
2022-01-20 23:37:48 +00:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-06-03 20:09:06 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-06-03 20:09:06 +02:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
2020-06-03 20:09:06 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-01-12 20:43:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-06-03 20:09:06 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-25 11:10:55 +01:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2019-02-25 11:10:55 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-03-06 12:58:15 +01:00
|
|
|
|
2022-07-11 11:49:50 +02:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2019-03-06 12:58:15 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
2019-03-06 12:58:15 +01:00
|
|
|
|
|
|
|
|
|
2020-03-10 18:07:44 +01:00
|
|
|
|
|
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-10 18:07:44 +01:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-03-10 18:07:44 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-03-10 18:07:44 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-03-10 18:07:44 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-04-01 16:38:19 +02:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-03-10 18:07:44 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2020-03-10 18:07:44 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
2020-04-28 19:37:01 +02:00
|
|
|
|
2020-12-18 14:19:32 -06:00
|
|
|
|
2020-04-28 19:37:01 +02:00
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-04-28 19:37:01 +02:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-26 10:11:01 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2019-07-17 20:36:47 -05:00
|
|
|
|
2020-04-02 10:15:51 +02:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2019-07-24 17:47:26 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2022-09-08 14:54:59 -07:00
|
|
|
|
2022-10-28 15:50:42 +02:00
|
|
|
|
|
|
|
|
|
2022-09-08 14:54:59 -07:00
|
|
|
|
|
|
|
|
|
2017-08-11 12:24:15 -05:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2017-08-11 12:24:15 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2017-07-07 09:19:42 -05:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2017-08-10 16:37:25 -05:00
|
|
|
|
2017-07-07 09:19:42 -05:00
|
|
|
|
|
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2019-07-24 17:47:26 -05:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
2022-06-24 12:52:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-06-24 12:52:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-06-24 12:52:40 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2019-07-24 17:47:26 -05:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
objtool: Support stack layout changes in alternatives
The ORC unwinder showed a warning [1] which revealed the stack layout
didn't match what was expected. The problem was that paravirt patching
had replaced "CALL *pv_ops.irq.save_fl" with "PUSHF;POP". That changed
the stack layout between the PUSHF and the POP, so unwinding from an
interrupt which occurred between those two instructions would fail.
Part of the agreed upon solution was to rework the custom paravirt
patching code to use alternatives instead, since objtool already knows
how to read alternatives (and converging runtime patching infrastructure
is always a good thing anyway). But the main problem still remains,
which is that runtime patching can change the stack layout.
Making stack layout changes in alternatives was disallowed with commit
7117f16bf460 ("objtool: Fix ORC vs alternatives"), but now that paravirt
is going to be doing it, it needs to be supported.
One way to do so would be to modify the ORC table when the code gets
patched. But ORC is simple -- a good thing! -- and it's best to leave
it alone.
Instead, support stack layout changes by "flattening" all possible stack
states (CFI) from parallel alternative code streams into a single set of
linear states. The only necessary limitation is that CFI conflicts are
disallowed at all possible instruction boundaries.
For example, this scenario is allowed:
Alt1 Alt2 Alt3
0x00 CALL *pv_ops.save_fl CALL xen_save_fl PUSHF
0x01 POP %RAX
0x02 NOP
...
0x05 NOP
...
0x07 <insn>
The unwind information for offset-0x00 is identical for all 3
alternatives. Similarly offset-0x05 and higher also are identical (and
the same as 0x00). However offset-0x01 has deviating CFI, but that is
only relevant for Alt3, neither of the other alternative instruction
streams will ever hit that offset.
This scenario is NOT allowed:
Alt1 Alt2
0x00 CALL *pv_ops.save_fl PUSHF
0x01 NOP6
...
0x07 NOP POP %RAX
The problem here is that offset-0x7, which is an instruction boundary in
both possible instruction patch streams, has two conflicting stack
layouts.
[ The above examples were stolen from Peter Zijlstra. ]
The new flattened CFI array is used both for the detection of conflicts
(like the second example above) and the generation of linear ORC
entries.
BTW, another benefit of these changes is that, thanks to some related
cleanups (new fake nops and alt_group struct) objtool can finally be rid
of fake jumps, which were a constant source of headaches.
[1] https://lkml.kernel.org/r/20201111170536.arx2zbn4ngvjoov7@treble
Cc: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
2020-12-18 14:26:21 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Make instruction::alts a single-linked list
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
unsigned int len; /* 64 4 */
enum insn_type type; /* 68 4 */
long unsigned int immediate; /* 72 8 */
u16 dead_end:1; /* 80: 0 2 */
u16 ignore:1; /* 80: 1 2 */
u16 ignore_alts:1; /* 80: 2 2 */
u16 hint:1; /* 80: 3 2 */
u16 save:1; /* 80: 4 2 */
u16 restore:1; /* 80: 5 2 */
u16 retpoline_safe:1; /* 80: 6 2 */
u16 noendbr:1; /* 80: 7 2 */
u16 entry:1; /* 80: 8 2 */
/* XXX 7 bits hole, try to pack */
s8 instr; /* 82 1 */
u8 visited; /* 83 1 */
/* XXX 4 bytes hole, try to pack */
struct alt_group * alt_group; /* 88 8 */
struct symbol * call_dest; /* 96 8 */
struct instruction * jump_dest; /* 104 8 */
struct instruction * first_jump_src; /* 112 8 */
struct reloc * jump_table; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
struct reloc * reloc; /* 128 8 */
- struct list_head alts; /* 136 16 */
- struct symbol * sym; /* 152 8 */
- struct stack_op * stack_ops; /* 160 8 */
- struct cfi_state * cfi; /* 168 8 */
+ struct alternative * alts; /* 136 8 */
+ struct symbol * sym; /* 144 8 */
+ struct stack_op * stack_ops; /* 152 8 */
+ struct cfi_state * cfi; /* 160 8 */
- /* size: 176, cachelines: 3, members: 29 */
- /* sum members: 170, holes: 1, sum holes: 4 */
+ /* size: 168, cachelines: 3, members: 29 */
+ /* sum members: 162, holes: 1, sum holes: 4 */
/* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */
- /* last cacheline: 48 bytes */
+ /* last cacheline: 40 bytes */
};
pre: 5:58.50 real, 229.64 user, 128.65 sys, 26221520 mem
post: 5:48.86 real, 220.30 user, 128.34 sys, 24834672 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.430556498@infradead.org
2023-02-08 18:17:59 +01:00
|
|
|
|
2019-03-01 11:19:03 +01:00
|
|
|
|
|
|
|
|
|
objtool: Make instruction::alts a single-linked list
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
unsigned int len; /* 64 4 */
enum insn_type type; /* 68 4 */
long unsigned int immediate; /* 72 8 */
u16 dead_end:1; /* 80: 0 2 */
u16 ignore:1; /* 80: 1 2 */
u16 ignore_alts:1; /* 80: 2 2 */
u16 hint:1; /* 80: 3 2 */
u16 save:1; /* 80: 4 2 */
u16 restore:1; /* 80: 5 2 */
u16 retpoline_safe:1; /* 80: 6 2 */
u16 noendbr:1; /* 80: 7 2 */
u16 entry:1; /* 80: 8 2 */
/* XXX 7 bits hole, try to pack */
s8 instr; /* 82 1 */
u8 visited; /* 83 1 */
/* XXX 4 bytes hole, try to pack */
struct alt_group * alt_group; /* 88 8 */
struct symbol * call_dest; /* 96 8 */
struct instruction * jump_dest; /* 104 8 */
struct instruction * first_jump_src; /* 112 8 */
struct reloc * jump_table; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
struct reloc * reloc; /* 128 8 */
- struct list_head alts; /* 136 16 */
- struct symbol * sym; /* 152 8 */
- struct stack_op * stack_ops; /* 160 8 */
- struct cfi_state * cfi; /* 168 8 */
+ struct alternative * alts; /* 136 8 */
+ struct symbol * sym; /* 144 8 */
+ struct stack_op * stack_ops; /* 152 8 */
+ struct cfi_state * cfi; /* 160 8 */
- /* size: 176, cachelines: 3, members: 29 */
- /* sum members: 170, holes: 1, sum holes: 4 */
+ /* size: 168, cachelines: 3, members: 29 */
+ /* sum members: 162, holes: 1, sum holes: 4 */
/* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */
- /* last cacheline: 48 bytes */
+ /* last cacheline: 40 bytes */
};
pre: 5:58.50 real, 229.64 user, 128.65 sys, 26221520 mem
post: 5:48.86 real, 220.30 user, 128.34 sys, 24834672 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.430556498@infradead.org
2023-02-08 18:17:59 +01:00
|
|
|
|
2019-03-01 11:19:03 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:47 -05:00
|
|
|
|
objtool: Add --backtrace support
For when you want to know the path that reached your fail state:
$ ./objtool check --no-fp --backtrace arch/x86/lib/usercopy_64.o
arch/x86/lib/usercopy_64.o: warning: objtool: .altinstr_replacement+0x3: UACCESS disable without MEMOPs: __clear_user()
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x3a: (alt)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x2e: (branch)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x18: (branch)
arch/x86/lib/usercopy_64.o: warning: objtool: .altinstr_replacement+0xffffffffffffffff: (branch)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x5: (alt)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x0: <=== (func)
0000000000000000 <__clear_user>:
0: e8 00 00 00 00 callq 5 <__clear_user+0x5>
1: R_X86_64_PLT32 __fentry__-0x4
5: 90 nop
6: 90 nop
7: 90 nop
8: 48 89 f0 mov %rsi,%rax
b: 48 c1 ee 03 shr $0x3,%rsi
f: 83 e0 07 and $0x7,%eax
12: 48 89 f1 mov %rsi,%rcx
15: 48 85 c9 test %rcx,%rcx
18: 74 0f je 29 <__clear_user+0x29>
1a: 48 c7 07 00 00 00 00 movq $0x0,(%rdi)
21: 48 83 c7 08 add $0x8,%rdi
25: ff c9 dec %ecx
27: 75 f1 jne 1a <__clear_user+0x1a>
29: 48 89 c1 mov %rax,%rcx
2c: 85 c9 test %ecx,%ecx
2e: 74 0a je 3a <__clear_user+0x3a>
30: c6 07 00 movb $0x0,(%rdi)
33: 48 ff c7 inc %rdi
36: ff c9 dec %ecx
38: 75 f6 jne 30 <__clear_user+0x30>
3a: 90 nop
3b: 90 nop
3c: 90 nop
3d: 48 89 c8 mov %rcx,%rax
40: c3 retq
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2019-03-01 11:15:49 +01:00
|
|
|
|
2023-04-18 14:27:49 -07:00
|
|
|
|
objtool: Add --backtrace support
For when you want to know the path that reached your fail state:
$ ./objtool check --no-fp --backtrace arch/x86/lib/usercopy_64.o
arch/x86/lib/usercopy_64.o: warning: objtool: .altinstr_replacement+0x3: UACCESS disable without MEMOPs: __clear_user()
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x3a: (alt)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x2e: (branch)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x18: (branch)
arch/x86/lib/usercopy_64.o: warning: objtool: .altinstr_replacement+0xffffffffffffffff: (branch)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x5: (alt)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x0: <=== (func)
0000000000000000 <__clear_user>:
0: e8 00 00 00 00 callq 5 <__clear_user+0x5>
1: R_X86_64_PLT32 __fentry__-0x4
5: 90 nop
6: 90 nop
7: 90 nop
8: 48 89 f0 mov %rsi,%rax
b: 48 c1 ee 03 shr $0x3,%rsi
f: 83 e0 07 and $0x7,%eax
12: 48 89 f1 mov %rsi,%rcx
15: 48 85 c9 test %rcx,%rcx
18: 74 0f je 29 <__clear_user+0x29>
1a: 48 c7 07 00 00 00 00 movq $0x0,(%rdi)
21: 48 83 c7 08 add $0x8,%rdi
25: ff c9 dec %ecx
27: 75 f1 jne 1a <__clear_user+0x1a>
29: 48 89 c1 mov %rax,%rcx
2c: 85 c9 test %ecx,%ecx
2e: 74 0a je 3a <__clear_user+0x3a>
30: c6 07 00 movb $0x0,(%rdi)
33: 48 ff c7 inc %rdi
36: ff c9 dec %ecx
38: 75 f6 jne 30 <__clear_user+0x30>
3a: 90 nop
3b: 90 nop
3c: 90 nop
3d: 48 89 c8 mov %rcx,%rax
40: c3 retq
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2019-03-01 11:15:49 +01:00
|
|
|
|
|
|
|
|
|
2018-01-29 22:00:39 -06:00
|
|
|
|
2019-03-01 11:19:03 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2021-02-11 13:03:28 +01:00
|
|
|
|
2020-04-24 16:16:41 +02:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-10 18:07:44 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-08-16 13:59:21 +02:00
|
|
|
|
2019-07-17 20:36:51 -05:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:51 -05:00
|
|
|
|
2022-03-08 16:30:49 +01:00
|
|
|
|
2019-07-17 20:36:51 -05:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-21 15:29:22 -06:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2019-03-06 12:58:15 +01:00
|
|
|
|
2017-07-07 09:19:42 -05:00
|
|
|
|
2019-07-17 20:36:52 -05:00
|
|
|
|
2019-07-17 20:36:47 -05:00
|
|
|
|
|
|
|
|
|
objtool: Add --backtrace support
For when you want to know the path that reached your fail state:
$ ./objtool check --no-fp --backtrace arch/x86/lib/usercopy_64.o
arch/x86/lib/usercopy_64.o: warning: objtool: .altinstr_replacement+0x3: UACCESS disable without MEMOPs: __clear_user()
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x3a: (alt)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x2e: (branch)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x18: (branch)
arch/x86/lib/usercopy_64.o: warning: objtool: .altinstr_replacement+0xffffffffffffffff: (branch)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x5: (alt)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x0: <=== (func)
0000000000000000 <__clear_user>:
0: e8 00 00 00 00 callq 5 <__clear_user+0x5>
1: R_X86_64_PLT32 __fentry__-0x4
5: 90 nop
6: 90 nop
7: 90 nop
8: 48 89 f0 mov %rsi,%rax
b: 48 c1 ee 03 shr $0x3,%rsi
f: 83 e0 07 and $0x7,%eax
12: 48 89 f1 mov %rsi,%rcx
15: 48 85 c9 test %rcx,%rcx
18: 74 0f je 29 <__clear_user+0x29>
1a: 48 c7 07 00 00 00 00 movq $0x0,(%rdi)
21: 48 83 c7 08 add $0x8,%rdi
25: ff c9 dec %ecx
27: 75 f1 jne 1a <__clear_user+0x1a>
29: 48 89 c1 mov %rax,%rcx
2c: 85 c9 test %ecx,%ecx
2e: 74 0a je 3a <__clear_user+0x3a>
30: c6 07 00 movb $0x0,(%rdi)
33: 48 ff c7 inc %rdi
36: ff c9 dec %ecx
38: 75 f6 jne 30 <__clear_user+0x30>
3a: 90 nop
3b: 90 nop
3c: 90 nop
3d: 48 89 c8 mov %rcx,%rax
40: c3 retq
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2019-03-01 11:15:49 +01:00
|
|
|
|
2023-04-18 14:27:49 -07:00
|
|
|
|
objtool: Add --backtrace support
For when you want to know the path that reached your fail state:
$ ./objtool check --no-fp --backtrace arch/x86/lib/usercopy_64.o
arch/x86/lib/usercopy_64.o: warning: objtool: .altinstr_replacement+0x3: UACCESS disable without MEMOPs: __clear_user()
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x3a: (alt)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x2e: (branch)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x18: (branch)
arch/x86/lib/usercopy_64.o: warning: objtool: .altinstr_replacement+0xffffffffffffffff: (branch)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x5: (alt)
arch/x86/lib/usercopy_64.o: warning: objtool: __clear_user()+0x0: <=== (func)
0000000000000000 <__clear_user>:
0: e8 00 00 00 00 callq 5 <__clear_user+0x5>
1: R_X86_64_PLT32 __fentry__-0x4
5: 90 nop
6: 90 nop
7: 90 nop
8: 48 89 f0 mov %rsi,%rax
b: 48 c1 ee 03 shr $0x3,%rsi
f: 83 e0 07 and $0x7,%eax
12: 48 89 f1 mov %rsi,%rcx
15: 48 85 c9 test %rcx,%rcx
18: 74 0f je 29 <__clear_user+0x29>
1a: 48 c7 07 00 00 00 00 movq $0x0,(%rdi)
21: 48 83 c7 08 add $0x8,%rdi
25: ff c9 dec %ecx
27: 75 f1 jne 1a <__clear_user+0x1a>
29: 48 89 c1 mov %rax,%rcx
2c: 85 c9 test %ecx,%ecx
2e: 74 0a je 3a <__clear_user+0x3a>
30: c6 07 00 movb $0x0,(%rdi)
33: 48 ff c7 inc %rdi
36: ff c9 dec %ecx
38: 75 f6 jne 30 <__clear_user+0x30>
3a: 90 nop
3b: 90 nop
3c: 90 nop
3d: 48 89 c8 mov %rcx,%rax
40: c3 retq
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2019-03-01 11:15:49 +01:00
|
|
|
|
|
|
|
|
|
2017-07-07 09:19:42 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:57 -05:00
|
|
|
|
2021-01-21 15:29:22 -06:00
|
|
|
|
2021-06-24 11:41:23 +02:00
|
|
|
|
2019-03-06 12:58:15 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:57 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:47 -05:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2019-02-25 12:50:09 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2019-02-25 11:10:55 +01:00
|
|
|
|
2021-01-21 15:29:17 -06:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2021-01-21 15:29:17 -06:00
|
|
|
|
|
|
|
|
|
2019-02-25 11:10:55 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-01-21 15:29:17 -06:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2021-01-21 15:29:17 -06:00
|
|
|
|
|
|
|
|
|
2019-02-25 11:10:55 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-09-18 21:43:30 -05:00
|
|
|
|
2020-03-25 14:04:45 +01:00
|
|
|
|
2017-09-18 21:43:30 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-09-18 21:43:30 -05:00
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
2017-09-18 21:43:30 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-18 14:27:49 -07:00
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-23 18:26:03 +01:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:43 -07:00
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
2020-03-23 18:26:03 +01:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
2020-03-23 18:26:03 +01:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
|
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
2023-05-30 15:56:49 +08:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
objtool: Make instruction::alts a single-linked list
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
unsigned int len; /* 64 4 */
enum insn_type type; /* 68 4 */
long unsigned int immediate; /* 72 8 */
u16 dead_end:1; /* 80: 0 2 */
u16 ignore:1; /* 80: 1 2 */
u16 ignore_alts:1; /* 80: 2 2 */
u16 hint:1; /* 80: 3 2 */
u16 save:1; /* 80: 4 2 */
u16 restore:1; /* 80: 5 2 */
u16 retpoline_safe:1; /* 80: 6 2 */
u16 noendbr:1; /* 80: 7 2 */
u16 entry:1; /* 80: 8 2 */
/* XXX 7 bits hole, try to pack */
s8 instr; /* 82 1 */
u8 visited; /* 83 1 */
/* XXX 4 bytes hole, try to pack */
struct alt_group * alt_group; /* 88 8 */
struct symbol * call_dest; /* 96 8 */
struct instruction * jump_dest; /* 104 8 */
struct instruction * first_jump_src; /* 112 8 */
struct reloc * jump_table; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
struct reloc * reloc; /* 128 8 */
- struct list_head alts; /* 136 16 */
- struct symbol * sym; /* 152 8 */
- struct stack_op * stack_ops; /* 160 8 */
- struct cfi_state * cfi; /* 168 8 */
+ struct alternative * alts; /* 136 8 */
+ struct symbol * sym; /* 144 8 */
+ struct stack_op * stack_ops; /* 152 8 */
+ struct cfi_state * cfi; /* 160 8 */
- /* size: 176, cachelines: 3, members: 29 */
- /* sum members: 170, holes: 1, sum holes: 4 */
+ /* size: 168, cachelines: 3, members: 29 */
+ /* sum members: 162, holes: 1, sum holes: 4 */
/* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */
- /* last cacheline: 48 bytes */
+ /* last cacheline: 40 bytes */
};
pre: 5:58.50 real, 229.64 user, 128.65 sys, 26221520 mem
post: 5:48.86 real, 220.30 user, 128.34 sys, 24834672 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.430556498@infradead.org
2023-02-08 18:17:59 +01:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Make instruction::alts a single-linked list
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
unsigned int len; /* 64 4 */
enum insn_type type; /* 68 4 */
long unsigned int immediate; /* 72 8 */
u16 dead_end:1; /* 80: 0 2 */
u16 ignore:1; /* 80: 1 2 */
u16 ignore_alts:1; /* 80: 2 2 */
u16 hint:1; /* 80: 3 2 */
u16 save:1; /* 80: 4 2 */
u16 restore:1; /* 80: 5 2 */
u16 retpoline_safe:1; /* 80: 6 2 */
u16 noendbr:1; /* 80: 7 2 */
u16 entry:1; /* 80: 8 2 */
/* XXX 7 bits hole, try to pack */
s8 instr; /* 82 1 */
u8 visited; /* 83 1 */
/* XXX 4 bytes hole, try to pack */
struct alt_group * alt_group; /* 88 8 */
struct symbol * call_dest; /* 96 8 */
struct instruction * jump_dest; /* 104 8 */
struct instruction * first_jump_src; /* 112 8 */
struct reloc * jump_table; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
struct reloc * reloc; /* 128 8 */
- struct list_head alts; /* 136 16 */
- struct symbol * sym; /* 152 8 */
- struct stack_op * stack_ops; /* 160 8 */
- struct cfi_state * cfi; /* 168 8 */
+ struct alternative * alts; /* 136 8 */
+ struct symbol * sym; /* 144 8 */
+ struct stack_op * stack_ops; /* 152 8 */
+ struct cfi_state * cfi; /* 160 8 */
- /* size: 176, cachelines: 3, members: 29 */
- /* sum members: 170, holes: 1, sum holes: 4 */
+ /* size: 168, cachelines: 3, members: 29 */
+ /* sum members: 162, holes: 1, sum holes: 4 */
/* sum bitfield members: 9 bits, bit holes: 1, sum bit holes: 7 bits */
- /* last cacheline: 48 bytes */
+ /* last cacheline: 40 bytes */
};
pre: 5:58.50 real, 229.64 user, 128.65 sys, 26221520 mem
post: 5:48.86 real, 220.30 user, 128.34 sys, 24834672 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.430556498@infradead.org
2023-02-08 18:17:59 +01:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
2023-04-18 14:27:49 -07:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
2023-04-18 14:27:49 -07:00
|
|
|
|
|
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
|
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
2023-04-18 14:27:49 -07:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 15:56:49 +08:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
|
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-01-16 10:24:06 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-14 23:15:59 +02:00
|
|
|
|
|
|
|
|
|
2018-01-16 10:24:06 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-15 13:11:08 +02:00
|
|
|
|
2018-01-31 10:18:28 +01:00
|
|
|
|
|
|
|
|
|
2022-06-14 23:15:59 +02:00
|
|
|
|
2022-06-27 22:21:17 +00:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-06-27 22:21:17 +00:00
|
|
|
|
|
|
|
|
|
2022-06-14 23:15:59 +02:00
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-06-14 23:15:59 +02:00
|
|
|
|
|
|
|
|
|
2018-01-16 10:24:06 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Union instruction::{call_dest,jump_table}
The instruction call_dest and jump_table members can never be used at
the same time, their usage depends on type.
struct instruction {
struct list_head list; /* 0 16 */
struct hlist_node hash; /* 16 16 */
struct list_head call_node; /* 32 16 */
struct section * sec; /* 48 8 */
long unsigned int offset; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int immediate; /* 64 8 */
unsigned int len; /* 72 4 */
u8 type; /* 76 1 */
/* Bitfield combined with previous fields */
u16 dead_end:1; /* 76: 8 2 */
u16 ignore:1; /* 76: 9 2 */
u16 ignore_alts:1; /* 76:10 2 */
u16 hint:1; /* 76:11 2 */
u16 save:1; /* 76:12 2 */
u16 restore:1; /* 76:13 2 */
u16 retpoline_safe:1; /* 76:14 2 */
u16 noendbr:1; /* 76:15 2 */
u16 entry:1; /* 78: 0 2 */
u16 visited:4; /* 78: 1 2 */
u16 no_reloc:1; /* 78: 5 2 */
/* XXX 2 bits hole, try to pack */
/* Bitfield combined with next fields */
s8 instr; /* 79 1 */
struct alt_group * alt_group; /* 80 8 */
- struct symbol * call_dest; /* 88 8 */
- struct instruction * jump_dest; /* 96 8 */
- struct instruction * first_jump_src; /* 104 8 */
- struct reloc * jump_table; /* 112 8 */
- struct alternative * alts; /* 120 8 */
+ struct instruction * jump_dest; /* 88 8 */
+ struct instruction * first_jump_src; /* 96 8 */
+ union {
+ struct symbol * _call_dest; /* 104 8 */
+ struct reloc * _jump_table; /* 104 8 */
+ }; /* 104 8 */
+ struct alternative * alts; /* 112 8 */
+ struct symbol * sym; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
- struct symbol * sym; /* 128 8 */
- struct stack_op * stack_ops; /* 136 8 */
- struct cfi_state * cfi; /* 144 8 */
+ struct stack_op * stack_ops; /* 128 8 */
+ struct cfi_state * cfi; /* 136 8 */
- /* size: 152, cachelines: 3, members: 29 */
- /* sum members: 150 */
+ /* size: 144, cachelines: 3, members: 28 */
+ /* sum members: 142 */
/* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 24 bytes */
+ /* last cacheline: 16 bytes */
};
pre: 5:39.35 real, 215.58 user, 123.69 sys, 23448736 mem
post: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.640914454@infradead.org
2023-02-08 18:18:02 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-19 09:41:18 +03:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2020-09-19 09:41:18 +03:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2022-03-08 16:30:14 +01:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-11-10 11:01:25 +01:00
|
|
|
|
2017-07-27 15:56:54 -05:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2021-11-10 11:01:25 +01:00
|
|
|
|
2017-07-27 15:56:54 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:46 +01:00
|
|
|
|
2022-04-18 09:50:43 -07:00
|
|
|
|
2022-03-08 16:30:46 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2022-03-08 16:30:46 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:46 +01:00
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2022-03-08 16:30:46 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2020-04-01 13:23:25 -05:00
|
|
|
|
|
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2021-10-30 09:47:58 +02:00
|
|
|
|
|
|
|
|
|
2020-04-01 13:23:25 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-09-19 09:41:18 +03:00
|
|
|
|
|
|
|
|
|
2020-04-01 13:23:25 -05:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2023-04-12 12:03:22 -07:00
|
|
|
|
2020-04-01 13:23:25 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2018-02-08 17:09:25 -06:00
|
|
|
|
|
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2018-02-08 17:09:25 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2018-02-08 17:09:25 -06:00
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 13:26:13 -07:00
|
|
|
|
2022-10-28 15:50:42 +02:00
|
|
|
|
2023-04-12 13:26:13 -07:00
|
|
|
|
2023-04-12 13:26:15 -07:00
|
|
|
|
2022-10-28 15:50:42 +02:00
|
|
|
|
2023-04-12 13:26:13 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-28 15:50:42 +02:00
|
|
|
|
2023-04-12 13:26:13 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-28 15:50:42 +02:00
|
|
|
|
|
|
|
|
|
2023-04-12 13:26:13 -07:00
|
|
|
|
2022-10-28 15:50:42 +02:00
|
|
|
|
|
|
|
|
|
2023-04-12 13:26:13 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-28 15:50:42 +02:00
|
|
|
|
|
|
|
|
|
2023-04-12 13:26:13 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 13:26:15 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-28 15:50:42 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 13:26:13 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 15:56:49 +08:00
|
|
|
|
2023-04-12 13:26:13 -07:00
|
|
|
|
|
|
|
|
|
2020-03-23 21:17:50 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2020-03-23 21:17:50 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-22 22:03:50 +02:00
|
|
|
|
2023-04-18 14:27:49 -07:00
|
|
|
|
|
|
|
|
|
2020-03-23 21:17:50 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2020-03-23 21:17:50 +01:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-04-12 12:03:19 -07:00
|
|
|
|
2020-03-23 20:57:13 +01:00
|
|
|
|
|
|
|
|
|
2019-07-17 20:36:48 -05:00
|
|
|
|
2022-04-18 09:50:43 -07:00
|
|
|
|
2021-01-21 15:29:24 -06:00
|
|
|
|
2020-03-27 15:28:40 +00:00
|
|
|
|
2020-03-23 21:17:50 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:43 -07:00
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
2020-03-23 18:26:03 +01:00
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
2020-03-25 17:18:17 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
2020-03-25 17:18:17 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-23 18:26:03 +01:00
|
|
|
|
2023-01-12 20:43:31 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-23 18:26:03 +01:00
|
|
|
|
2020-03-10 18:57:41 +01:00
|
|
|
|
|
|
|
|
|
2020-03-23 20:57:13 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-23 21:11:14 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-03-23 20:57:13 +01:00
|
|
|
|
2020-03-23 21:11:14 +01:00
|
|
|
|
2020-03-23 20:57:13 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-15 13:11:13 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:06 -07:00
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
2023-05-30 10:21:07 -07:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2023-09-11 16:56:13 -07:00
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2022-09-15 13:11:13 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
|
|
|
|
|
2022-09-15 13:11:13 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:08 -07:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:06 -07:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
|
|
|
|
|
2023-05-30 10:20:55 -07:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-08-18 14:39:27 -07:00
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-08 14:54:59 -07:00
|
|
|
|
Networking changes for 6.1.
Core
----
- Introduce and use a single page frag cache for allocating small skb
heads, clawing back the 10-20% performance regression in UDP flood
test from previous fixes.
- Run packets which already went thru HW coalescing thru SW GRO.
This significantly improves TCP segment coalescing and simplifies
deployments as different workloads benefit from HW or SW GRO.
- Shrink the size of the base zero-copy send structure.
- Move TCP init under a new slow / sleepable version of DO_ONCE().
BPF
---
- Add BPF-specific, any-context-safe memory allocator.
- Add helpers/kfuncs for PKCS#7 signature verification from BPF
programs.
- Define a new map type and related helpers for user space -> kernel
communication over a ring buffer (BPF_MAP_TYPE_USER_RINGBUF).
- Allow targeting BPF iterators to loop through resources of one
task/thread.
- Add ability to call selected destructive functions.
Expose crash_kexec() to allow BPF to trigger a kernel dump.
Use CAP_SYS_BOOT check on the loading process to judge permissions.
- Enable BPF to collect custom hierarchical cgroup stats efficiently
by integrating with the rstat framework.
- Support struct arguments for trampoline based programs.
Only structs with size <= 16B and x86 are supported.
- Invoke cgroup/connect{4,6} programs for unprivileged ICMP ping
sockets (instead of just TCP and UDP sockets).
- Add a helper for accessing CLOCK_TAI for time sensitive network
related programs.
- Support accessing network tunnel metadata's flags.
- Make TCP SYN ACK RTO tunable by BPF programs with TCP Fast Open.
- Add support for writing to Netfilter's nf_conn:mark.
Protocols
---------
- WiFi: more Extremely High Throughput (EHT) and Multi-Link
Operation (MLO) work (802.11be, WiFi 7).
- vsock: improve support for SO_RCVLOWAT.
- SMC: support SO_REUSEPORT.
- Netlink: define and document how to use netlink in a "modern" way.
Support reporting missing attributes via extended ACK.
- IPSec: support collect metadata mode for xfrm interfaces.
- TCPv6: send consistent autoflowlabel in SYN_RECV state
and RST packets.
- TCP: introduce optional per-netns connection hash table to allow
better isolation between namespaces (opt-in, at the cost of memory
and cache pressure).
- MPTCP: support TCP_FASTOPEN_CONNECT.
- Add NEXT-C-SID support in Segment Routing (SRv6) End behavior.
- Adjust IP_UNICAST_IF sockopt behavior for connected UDP sockets.
- Open vSwitch:
- Allow specifying ifindex of new interfaces.
- Allow conntrack and metering in non-initial user namespace.
- TLS: support the Korean ARIA-GCM crypto algorithm.
- Remove DECnet support.
Driver API
----------
- Allow selecting the conduit interface used by each port
in DSA switches, at runtime.
- Ethernet Power Sourcing Equipment and Power Device support.
- Add tc-taprio support for queueMaxSDU parameter, i.e. setting
per traffic class max frame size for time-based packet schedules.
- Support PHY rate matching - adapting between differing host-side
and link-side speeds.
- Introduce QUSGMII PHY mode and 1000BASE-KX interface mode.
- Validate OF (device tree) nodes for DSA shared ports; make
phylink-related properties mandatory on DSA and CPU ports.
Enforcing more uniformity should allow transitioning to phylink.
- Require that flash component name used during update matches one
of the components for which version is reported by info_get().
- Remove "weight" argument from driver-facing NAPI API as much
as possible. It's one of those magic knobs which seemed like
a good idea at the time but is too indirect to use in practice.
- Support offload of TLS connections with 256 bit keys.
New hardware / drivers
----------------------
- Ethernet:
- Microchip KSZ9896 6-port Gigabit Ethernet Switch
- Renesas Ethernet AVB (EtherAVB-IF) Gen4 SoCs
- Analog Devices ADIN1110 and ADIN2111 industrial single pair
Ethernet (10BASE-T1L) MAC+PHY.
- Rockchip RV1126 Gigabit Ethernet (a version of stmmac IP).
- Ethernet SFPs / modules:
- RollBall / Hilink / Turris 10G copper SFPs
- HALNy GPON module
- WiFi:
- CYW43439 SDIO chipset (brcmfmac)
- CYW89459 PCIe chipset (brcmfmac)
- BCM4378 on Apple platforms (brcmfmac)
Drivers
-------
- CAN:
- gs_usb: HW timestamp support
- Ethernet PHYs:
- lan8814: cable diagnostics
- Ethernet NICs:
- Intel (100G):
- implement control of FCS/CRC stripping
- port splitting via devlink
- L2TPv3 filtering offload
- nVidia/Mellanox:
- tunnel offload for sub-functions
- MACSec offload, w/ Extended packet number and replay
window offload
- significantly restructure, and optimize the AF_XDP support,
align the behavior with other vendors
- Huawei:
- configuring DSCP map for traffic class selection
- querying standard FEC statistics
- querying SerDes lane number via ethtool
- Marvell/Cavium:
- egress priority flow control
- MACSec offload
- AMD/SolarFlare:
- PTP over IPv6 and raw Ethernet
- small / embedded:
- ax88772: convert to phylink (to support SFP cages)
- altera: tse: convert to phylink
- ftgmac100: support fixed link
- enetc: standard Ethtool counters
- macb: ZynqMP SGMII dynamic configuration support
- tsnep: support multi-queue and use page pool
- lan743x: Rx IP & TCP checksum offload
- igc: add xdp frags support to ndo_xdp_xmit
- Ethernet high-speed switches:
- Marvell (prestera):
- support SPAN port features (traffic mirroring)
- nexthop object offloading
- Microchip (sparx5):
- multicast forwarding offload
- QoS queuing offload (tc-mqprio, tc-tbf, tc-ets)
- Ethernet embedded switches:
- Marvell (mv88e6xxx):
- support RGMII cmode
- NXP (felix):
- standardized ethtool counters
- Microchip (lan966x):
- QoS queuing offload (tc-mqprio, tc-tbf, tc-cbs, tc-ets)
- traffic policing and mirroring
- link aggregation / bonding offload
- QUSGMII PHY mode support
- Qualcomm 802.11ax WiFi (ath11k):
- cold boot calibration support on WCN6750
- support to connect to a non-transmit MBSSID AP profile
- enable remain-on-channel support on WCN6750
- Wake-on-WLAN support for WCN6750
- support to provide transmit power from firmware via nl80211
- support to get power save duration for each client
- spectral scan support for 160 MHz
- MediaTek WiFi (mt76):
- WiFi-to-Ethernet bridging offload for MT7986 chips
- RealTek WiFi (rtw89):
- P2P support
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
-----BEGIN PGP SIGNATURE-----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=Gsio
-----END PGP SIGNATURE-----
Merge tag 'net-next-6.1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next
Pull networking updates from Jakub Kicinski:
"Core:
- Introduce and use a single page frag cache for allocating small skb
heads, clawing back the 10-20% performance regression in UDP flood
test from previous fixes.
- Run packets which already went thru HW coalescing thru SW GRO. This
significantly improves TCP segment coalescing and simplifies
deployments as different workloads benefit from HW or SW GRO.
- Shrink the size of the base zero-copy send structure.
- Move TCP init under a new slow / sleepable version of DO_ONCE().
BPF:
- Add BPF-specific, any-context-safe memory allocator.
- Add helpers/kfuncs for PKCS#7 signature verification from BPF
programs.
- Define a new map type and related helpers for user space -> kernel
communication over a ring buffer (BPF_MAP_TYPE_USER_RINGBUF).
- Allow targeting BPF iterators to loop through resources of one
task/thread.
- Add ability to call selected destructive functions. Expose
crash_kexec() to allow BPF to trigger a kernel dump. Use
CAP_SYS_BOOT check on the loading process to judge permissions.
- Enable BPF to collect custom hierarchical cgroup stats efficiently
by integrating with the rstat framework.
- Support struct arguments for trampoline based programs. Only
structs with size <= 16B and x86 are supported.
- Invoke cgroup/connect{4,6} programs for unprivileged ICMP ping
sockets (instead of just TCP and UDP sockets).
- Add a helper for accessing CLOCK_TAI for time sensitive network
related programs.
- Support accessing network tunnel metadata's flags.
- Make TCP SYN ACK RTO tunable by BPF programs with TCP Fast Open.
- Add support for writing to Netfilter's nf_conn:mark.
Protocols:
- WiFi: more Extremely High Throughput (EHT) and Multi-Link Operation
(MLO) work (802.11be, WiFi 7).
- vsock: improve support for SO_RCVLOWAT.
- SMC: support SO_REUSEPORT.
- Netlink: define and document how to use netlink in a "modern" way.
Support reporting missing attributes via extended ACK.
- IPSec: support collect metadata mode for xfrm interfaces.
- TCPv6: send consistent autoflowlabel in SYN_RECV state and RST
packets.
- TCP: introduce optional per-netns connection hash table to allow
better isolation between namespaces (opt-in, at the cost of memory
and cache pressure).
- MPTCP: support TCP_FASTOPEN_CONNECT.
- Add NEXT-C-SID support in Segment Routing (SRv6) End behavior.
- Adjust IP_UNICAST_IF sockopt behavior for connected UDP sockets.
- Open vSwitch:
- Allow specifying ifindex of new interfaces.
- Allow conntrack and metering in non-initial user namespace.
- TLS: support the Korean ARIA-GCM crypto algorithm.
- Remove DECnet support.
Driver API:
- Allow selecting the conduit interface used by each port in DSA
switches, at runtime.
- Ethernet Power Sourcing Equipment and Power Device support.
- Add tc-taprio support for queueMaxSDU parameter, i.e. setting per
traffic class max frame size for time-based packet schedules.
- Support PHY rate matching - adapting between differing host-side
and link-side speeds.
- Introduce QUSGMII PHY mode and 1000BASE-KX interface mode.
- Validate OF (device tree) nodes for DSA shared ports; make
phylink-related properties mandatory on DSA and CPU ports.
Enforcing more uniformity should allow transitioning to phylink.
- Require that flash component name used during update matches one of
the components for which version is reported by info_get().
- Remove "weight" argument from driver-facing NAPI API as much as
possible. It's one of those magic knobs which seemed like a good
idea at the time but is too indirect to use in practice.
- Support offload of TLS connections with 256 bit keys.
New hardware / drivers:
- Ethernet:
- Microchip KSZ9896 6-port Gigabit Ethernet Switch
- Renesas Ethernet AVB (EtherAVB-IF) Gen4 SoCs
- Analog Devices ADIN1110 and ADIN2111 industrial single pair
Ethernet (10BASE-T1L) MAC+PHY.
- Rockchip RV1126 Gigabit Ethernet (a version of stmmac IP).
- Ethernet SFPs / modules:
- RollBall / Hilink / Turris 10G copper SFPs
- HALNy GPON module
- WiFi:
- CYW43439 SDIO chipset (brcmfmac)
- CYW89459 PCIe chipset (brcmfmac)
- BCM4378 on Apple platforms (brcmfmac)
Drivers:
- CAN:
- gs_usb: HW timestamp support
- Ethernet PHYs:
- lan8814: cable diagnostics
- Ethernet NICs:
- Intel (100G):
- implement control of FCS/CRC stripping
- port splitting via devlink
- L2TPv3 filtering offload
- nVidia/Mellanox:
- tunnel offload for sub-functions
- MACSec offload, w/ Extended packet number and replay window
offload
- significantly restructure, and optimize the AF_XDP support,
align the behavior with other vendors
- Huawei:
- configuring DSCP map for traffic class selection
- querying standard FEC statistics
- querying SerDes lane number via ethtool
- Marvell/Cavium:
- egress priority flow control
- MACSec offload
- AMD/SolarFlare:
- PTP over IPv6 and raw Ethernet
- small / embedded:
- ax88772: convert to phylink (to support SFP cages)
- altera: tse: convert to phylink
- ftgmac100: support fixed link
- enetc: standard Ethtool counters
- macb: ZynqMP SGMII dynamic configuration support
- tsnep: support multi-queue and use page pool
- lan743x: Rx IP & TCP checksum offload
- igc: add xdp frags support to ndo_xdp_xmit
- Ethernet high-speed switches:
- Marvell (prestera):
- support SPAN port features (traffic mirroring)
- nexthop object offloading
- Microchip (sparx5):
- multicast forwarding offload
- QoS queuing offload (tc-mqprio, tc-tbf, tc-ets)
- Ethernet embedded switches:
- Marvell (mv88e6xxx):
- support RGMII cmode
- NXP (felix):
- standardized ethtool counters
- Microchip (lan966x):
- QoS queuing offload (tc-mqprio, tc-tbf, tc-cbs, tc-ets)
- traffic policing and mirroring
- link aggregation / bonding offload
- QUSGMII PHY mode support
- Qualcomm 802.11ax WiFi (ath11k):
- cold boot calibration support on WCN6750
- support to connect to a non-transmit MBSSID AP profile
- enable remain-on-channel support on WCN6750
- Wake-on-WLAN support for WCN6750
- support to provide transmit power from firmware via nl80211
- support to get power save duration for each client
- spectral scan support for 160 MHz
- MediaTek WiFi (mt76):
- WiFi-to-Ethernet bridging offload for MT7986 chips
- RealTek WiFi (rtw89):
- P2P support"
* tag 'net-next-6.1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1864 commits)
eth: pse: add missing static inlines
once: rename _SLOW to _SLEEPABLE
net: pse-pd: add regulator based PSE driver
dt-bindings: net: pse-dt: add bindings for regulator based PoDL PSE controller
ethtool: add interface to interact with Ethernet Power Equipment
net: mdiobus: search for PSE nodes by parsing PHY nodes.
net: mdiobus: fwnode_mdiobus_register_phy() rework error handling
net: add framework to support Ethernet PSE and PDs devices
dt-bindings: net: phy: add PoDL PSE property
net: marvell: prestera: Propagate nh state from hw to kernel
net: marvell: prestera: Add neighbour cache accounting
net: marvell: prestera: add stub handler neighbour events
net: marvell: prestera: Add heplers to interact with fib_notifier_info
net: marvell: prestera: Add length macros for prestera_ip_addr
net: marvell: prestera: add delayed wq and flush wq on deinit
net: marvell: prestera: Add strict cleanup of fib arbiter
net: marvell: prestera: Add cleanup of allocated fib_nodes
net: marvell: prestera: Add router nexthops ABI
eth: octeon: fix build after netif_napi_add() changes
net/mlx5: E-Switch, Return EBUSY if can't get mode lock
...
2022-10-04 13:38:03 -07:00
|
|
|
|
2022-09-03 15:11:53 +02:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
2023-05-30 10:21:02 -07:00
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:34 -07:00
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:35 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-04-18 09:50:35 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2022-04-18 09:50:35 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-18 14:27:51 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2023-04-18 14:27:50 -07:00
|
|
|
|
|
|
|
|
|
2023-04-18 14:27:47 -07:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2020-09-19 09:41:18 +03:00
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2023-04-18 14:27:50 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-18 14:27:51 -07:00
|
|
|
|
2023-04-18 14:27:50 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 12:03:17 -07:00
|
|
|
|
2023-04-18 14:27:47 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2023-04-18 14:27:47 -07:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2023-04-18 14:27:48 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:13 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-25 13:47:40 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-06-05 09:12:21 -07:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
2020-08-25 13:47:39 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2020-08-25 13:47:39 +01:00
|
|
|
|
2018-01-16 10:24:06 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:35 -07:00
|
|
|
|
2022-04-18 09:50:33 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-07-11 10:33:43 -05:00
|
|
|
|
2022-04-18 09:50:33 -07:00
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:33 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:43 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-03-08 16:30:54 +01:00
|
|
|
|
|
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-03-01 07:13:11 -08:00
|
|
|
|
2022-06-14 23:16:03 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:33 -07:00
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:35 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:07 -05:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:38 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2020-08-18 15:57:45 +02:00
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2021-10-26 14:01:36 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-10-27 11:28:13 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-06-27 22:21:17 +00:00
|
|
|
|
2022-06-14 23:15:38 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-09-15 13:11:09 +02:00
|
|
|
|
2022-09-15 13:11:10 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2021-10-26 14:01:36 +02:00
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2020-08-06 15:14:09 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-04-12 13:26:13 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2022-03-08 16:30:55 +01:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
objtool: Remove instruction::list
Replace the instruction::list by allocating instructions in arrays of
256 entries and stringing them together by (amortized) find_insn().
This shrinks instruction by 16 bytes and brings it down to 128.
struct instruction {
- struct list_head list; /* 0 16 */
- struct hlist_node hash; /* 16 16 */
- struct list_head call_node; /* 32 16 */
- struct section * sec; /* 48 8 */
- long unsigned int offset; /* 56 8 */
- /* --- cacheline 1 boundary (64 bytes) --- */
- long unsigned int immediate; /* 64 8 */
- unsigned int len; /* 72 4 */
- u8 type; /* 76 1 */
-
- /* Bitfield combined with previous fields */
+ struct hlist_node hash; /* 0 16 */
+ struct list_head call_node; /* 16 16 */
+ struct section * sec; /* 32 8 */
+ long unsigned int offset; /* 40 8 */
+ long unsigned int immediate; /* 48 8 */
+ u8 len; /* 56 1 */
+ u8 prev_len; /* 57 1 */
+ u8 type; /* 58 1 */
+ s8 instr; /* 59 1 */
+ u32 idx:8; /* 60: 0 4 */
+ u32 dead_end:1; /* 60: 8 4 */
+ u32 ignore:1; /* 60: 9 4 */
+ u32 ignore_alts:1; /* 60:10 4 */
+ u32 hint:1; /* 60:11 4 */
+ u32 save:1; /* 60:12 4 */
+ u32 restore:1; /* 60:13 4 */
+ u32 retpoline_safe:1; /* 60:14 4 */
+ u32 noendbr:1; /* 60:15 4 */
+ u32 entry:1; /* 60:16 4 */
+ u32 visited:4; /* 60:17 4 */
+ u32 no_reloc:1; /* 60:21 4 */
- u16 dead_end:1; /* 76: 8 2 */
- u16 ignore:1; /* 76: 9 2 */
- u16 ignore_alts:1; /* 76:10 2 */
- u16 hint:1; /* 76:11 2 */
- u16 save:1; /* 76:12 2 */
- u16 restore:1; /* 76:13 2 */
- u16 retpoline_safe:1; /* 76:14 2 */
- u16 noendbr:1; /* 76:15 2 */
- u16 entry:1; /* 78: 0 2 */
- u16 visited:4; /* 78: 1 2 */
- u16 no_reloc:1; /* 78: 5 2 */
+ /* XXX 10 bits hole, try to pack */
- /* XXX 2 bits hole, try to pack */
- /* Bitfield combined with next fields */
-
- s8 instr; /* 79 1 */
- struct alt_group * alt_group; /* 80 8 */
- struct instruction * jump_dest; /* 88 8 */
- struct instruction * first_jump_src; /* 96 8 */
+ /* --- cacheline 1 boundary (64 bytes) --- */
+ struct alt_group * alt_group; /* 64 8 */
+ struct instruction * jump_dest; /* 72 8 */
+ struct instruction * first_jump_src; /* 80 8 */
union {
- struct symbol * _call_dest; /* 104 8 */
- struct reloc * _jump_table; /* 104 8 */
- }; /* 104 8 */
- struct alternative * alts; /* 112 8 */
- struct symbol * sym; /* 120 8 */
- /* --- cacheline 2 boundary (128 bytes) --- */
- struct stack_op * stack_ops; /* 128 8 */
- struct cfi_state * cfi; /* 136 8 */
+ struct symbol * _call_dest; /* 88 8 */
+ struct reloc * _jump_table; /* 88 8 */
+ }; /* 88 8 */
+ struct alternative * alts; /* 96 8 */
+ struct symbol * sym; /* 104 8 */
+ struct stack_op * stack_ops; /* 112 8 */
+ struct cfi_state * cfi; /* 120 8 */
- /* size: 144, cachelines: 3, members: 28 */
- /* sum members: 142 */
- /* sum bitfield members: 14 bits, bit holes: 1, sum bit holes: 2 bits */
- /* last cacheline: 16 bytes */
+ /* size: 128, cachelines: 2, members: 29 */
+ /* sum members: 124 */
+ /* sum bitfield members: 22 bits, bit holes: 1, sum bit holes: 10 bits */
};
pre: 5:38.18 real, 213.25 user, 124.90 sys, 23449040 mem
post: 5:03.34 real, 210.75 user, 88.80 sys, 20241232 mem
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Josh Poimboeuf <jpoimboe@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build only
Tested-by: Thomas Weißschuh <linux@weissschuh.net> # compile and run
Link: https://lore.kernel.org/r/20230208172245.851307606@infradead.org
2023-02-08 18:18:05 +01:00
|
|
|
|
2022-04-18 09:50:27 -07:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2023-05-30 10:21:13 -07:00
|
|
|
|
|
|
|
|
|
2023-04-18 14:27:48 -07:00
|
|
|
|
|
|
|
|
|
2022-04-18 09:50:27 -07:00
|
|
|
|
2022-04-18 09:50:26 -07:00
|
|
|
|
2021-06-24 11:41:01 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
2021-01-14 16:32:42 -06:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2017-06-28 10:11:05 -05:00
|
|
|
|
|
|
|
|
|
