Since bpf2bpf tailcall support is enabled for 64-bit powerpc with
kernel commit 2ed2d8f6fb ("powerpc64/bpf: Support tailcalls with
subprogs"), 'tailcalls/tailcall_bpf2bpf_hierarchy_fexit' BPF selftest
is triggering "corrupted stack end detected inside scheduler" with the
config option CONFIG_SCHED_STACK_END_CHECK enabled. While reviewing
the stack layout for BPF trampoline, observed that the dummy frame is
trying to protect the redzone of BPF program. This is because tail
call info and NVRs save area are in redzone at the time of tailcall
as the current BPF program stack frame is teared down before the
tailcall. But saving this redzone in the dummy frame of trampoline
is unnecessary because of the follow reasons:
1) Firstly, trampoline can be attached to BPF entry/main program
or subprog. But prologue part of the BPF entry/main program,
where the trampoline attachpoint is, is skipped during tailcall.
So, protecting the redzone does not arise when the trampoline is
not even triggered in this scenario.
2) In case of subprog, the caller's stackframe is already setup
and the subprog's stackframe is yet to be setup. So, nothing
on the redzone to be protected.
Also, using dummy frame in BPF trampoline, wastes critically scarce
kernel stack space, especially in tailcall sequence, for marginal
benefit in stack unwinding. So, drop setting up the dummy frame.
Instead, save return address in bpf trampoline frame and use it as
appropriate. Pruning this unnecessary stack usage mitigates the
likelihood of stack overflow in scenarios where bpf2bpf tailcalls
and fexit programs are mixed.
Reported-by: Saket Kumar Bhaskar <skb99@linux.ibm.com>
Fixes: 2ed2d8f6fb ("powerpc64/bpf: Support tailcalls with subprogs")
Tested-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Link: https://patch.msgid.link/20260303181031.390073-5-hbathini@linux.ibm.com
Ideally, the offset used to load the tail call info field and to find
the pass by reference address for tail call field should be the same.
But while setting up the tail call info in the trampoline, this was
not followed. This can be misleading and can lead to unpredictable
results if and when bpf_has_stack_frame() ends up returning true
for trampoline frame. Since commit 15513beeb6 ("powerpc64/bpf:
Moving tail_call_cnt to bottom of frame") and commit 2ed2d8f6fb
("powerpc64/bpf: Support tailcalls with subprogs") ensured tail call
field is at the bottom of the stack frame for BPF programs as well as
BPF trampoline, avoid relying on bpf_jit_stack_tailcallinfo_offset()
and bpf_has_stack_frame() for trampoline frame and always calculate
tail call field offset with reference to older frame.
Fixes: 2ed2d8f6fb ("powerpc64/bpf: Support tailcalls with subprogs")
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Link: https://patch.msgid.link/20260303181031.390073-4-hbathini@linux.ibm.com
bpf_get_func_ip() helper function returns the address of the traced
function. It relies on the IP address stored at ctx - 16 by the bpf
trampoline. On 64-bit powerpc, this address is recovered from LR
accounting for OOL trampoline. But the address stored here was off
by 4-bytes. Ensure the address is the actual start of the traced
function.
Reported-by: Abhishek Dubey <adubey@linux.ibm.com>
Fixes: d243b62b7b ("powerpc64/bpf: Add support for bpf trampolines")
Cc: stable@vger.kernel.org
Tested-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Link: https://patch.msgid.link/20260303181031.390073-3-hbathini@linux.ibm.com
Do not increment tailcall count, if tailcall did not succeed due to
missing BPF program.
Fixes: ce0761419f ("powerpc/bpf: Implement support for tail calls")
Cc: stable@vger.kernel.org
Tested-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Link: https://patch.msgid.link/20260303181031.390073-2-hbathini@linux.ibm.com
Support for -fpatchable-function-entry on ppc64le was added in Clang
with [1]. However, when no prefix NOPs are specified - as is the case
with CONFIG_PPC_FTRACE_OUT_OF_LINE - the first NOP is emitted at LEP,
but Clang records the Global Entry Point (GEP) unlike GCC which does
record the Local Entry Point (LEP). Issue [2] has been raised to align
Clang's behavior with GCC. As a temporary workaround to ensure ftrace
initialization works as expected with Clang, derive the LEP using
ppc_function_entry() for kernel symbols and by looking for the below
module GEP sequence for module addresses, until [2] is resolved:
ld r2, -8(r12)
add r2, r2, r12
[1] https://github.com/llvm/llvm-project/pull/151569
[2] https://github.com/llvm/llvm-project/issues/163706
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Tested-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Link: https://patch.msgid.link/20260127084926.34497-4-hbathini@linux.ibm.com
The total number of out-of-line (OOL) stubs required for function
tracing is determined using the following command:
$(OBJDUMP) -r -j __patchable_function_entries vmlinux.o
While this works correctly with GNU objdump, llvm-objdump does not
list the expected relocation records for this section. Fix this by
using the -d option and counting R_PPC64_ADDR64 relocation entries.
This works as desired with both objdump and llvm-objdump.
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Tested-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Link: https://patch.msgid.link/20260127084926.34497-3-hbathini@linux.ibm.com
ARCH_USING_PATCHABLE_FUNCTION_ENTRY depends on toolchain support for
-fpatchable-function-entry option. The current script that checks
for this support only handles GCC. Rename the script and extend it
to detect support for -fpatchable-function-entry with Clang as well,
allowing clean cross-compilation with Clang toolchains.
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Tested-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Link: https://patch.msgid.link/20260127084926.34497-2-hbathini@linux.ibm.com
Eduard Zingerman says:
====================
bpf: Fix precision backtracking bug with linked registers
Emil Tsalapatis reported a verifier bug hit by the scx_lavd sched_ext
scheduler. The essential part of the verifier log looks as follows:
436: ...
// checkpoint hit for 438: (1d) if r7 == r8 goto ...
frame 3: propagating r2,r7,r8
frame 2: propagating r6
mark_precise: frame3: last_idx ...
mark_precise: frame3: regs=r2,r7,r8 stack= before 436: ...
mark_precise: frame3: regs=r2,r7 stack= before 435: ...
mark_precise: frame3: regs=r2,r7 stack= before 434: (85) call bpf_trace_vprintk#177
verifier bug: backtracking call unexpected regs 84
The log complains that registers r2 and r7 are tracked as precise
while processing the bpf_trace_vprintk() call in precision backtracking.
This can't be right, as r2 is reset by the call and there is nothing
to backtrack it to. The precision propagation is triggered when
a checkpoint is hit at instruction 438, r2 is dead at that instruction.
This happens because of the following sequence of events:
- Instruction 438 is first reached with registers r2 and r7 having
the same id via a path that does not call bpf_trace_vprintk():
- Checkpoint is created at 438.
- The jump at 438 is predicted, hence r7 and registers linked to it
(r2) are propagated as precise, marking r2 and r7 precise in the
checkpoint.
- Instruction 438 is reached a second time with r2 undefined and via
a path that calls bpf_trace_vprintk():
- Checkpoint is hit.
- propagate_precision() picks registers r2 and r7 and propagates
precision marks for those up to the helper call.
The root cause is the fact that states_equal() and
propagate_precision() assume that the precision flag can't be set for a
dead register (as computed by compute_live_registers()).
However, this is not the case when linked registers are at play.
Fix this by accounting for live register flags in
collect_linked_regs().
---
====================
Link: https://patch.msgid.link/20260306-linked-regs-and-propagate-precision-v1-0-18e859be570d@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Add a test for the scenario described in the previous commit:
an iterator loop with two paths where one ties r2/r7 via
shared scalar id and skips a call, while the other goes
through the call. Precision marks from the linked registers
get spuriously propagated to the call path via
propagate_precision(), hitting "backtracking call unexpected
regs" in backtrack_insn().
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20260306-linked-regs-and-propagate-precision-v1-2-18e859be570d@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Fix an inconsistency between func_states_equal() and
collect_linked_regs():
- regsafe() uses check_ids() to verify that cached and current states
have identical register id mapping.
- func_states_equal() calls regsafe() only for registers computed as
live by compute_live_registers().
- clean_live_states() is supposed to remove dead registers from cached
states, but it can skip states belonging to an iterator-based loop.
- collect_linked_regs() collects all registers sharing the same id,
ignoring the marks computed by compute_live_registers().
Linked registers are stored in the state's jump history.
- backtrack_insn() marks all linked registers for an instruction
as precise whenever one of the linked registers is precise.
The above might lead to a scenario:
- There is an instruction I with register rY known to be dead at I.
- Instruction I is reached via two paths: first A, then B.
- On path A:
- There is an id link between registers rX and rY.
- Checkpoint C is created at I.
- Linked register set {rX, rY} is saved to the jump history.
- rX is marked as precise at I, causing both rX and rY
to be marked precise at C.
- On path B:
- There is no id link between registers rX and rY,
otherwise register states are sub-states of those in C.
- Because rY is dead at I, check_ids() returns true.
- Current state is considered equal to checkpoint C,
propagate_precision() propagates spurious precision
mark for register rY along the path B.
- Depending on a program, this might hit verifier_bug()
in the backtrack_insn(), e.g. if rY ∈ [r1..r5]
and backtrack_insn() spots a function call.
The reproducer program is in the next patch.
This was hit by sched_ext scx_lavd scheduler code.
Changes in tests:
- verifier_scalar_ids.c selftests need modification to preserve
some registers as live for __msg() checks.
- exceptions_assert.c adjusted to match changes in the verifier log,
R0 is dead after conditional instruction and thus does not get
range.
- precise.c adjusted to match changes in the verifier log, register r9
is dead after comparison and it's range is not important for test.
Reported-by: Emil Tsalapatis <emil@etsalapatis.com>
Fixes: 0fb3cf6110 ("bpf: use register liveness information for func_states_equal")
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20260306-linked-regs-and-propagate-precision-v1-1-18e859be570d@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Return sysfs_emit() directly and drop 'ret' in cap_rem_show().
In cap_rem_store(), use 'ret' when calling set_param_u() instead of
assigning it to 'val' first, and remove 'val'.
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Acked-by: Giovanni Cabiddu <giovanni.cabiddu@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
According to archive.org the site threw HTTP errors 404 since early 2024.
The last snapshot in the archive having actual content was from late 2023.
The page behind the new URL has more or less the same content as the
archived page from 2023, so it probably was just moved without setting
up a redirect.
Signed-off-by: Alexander Dahl <ada@thorsis.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add support for CTR mode of operation for AES algorithm in the AES
Engine of the DTHEv2 hardware cryptographic engine.
Signed-off-by: T Pratham <t-pratham@ti.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
During the CPU offline process, the dying CPU is cleared from the
cpu_online_mask in takedown_cpu(). After this step, various CPUHP_*_DEAD
callbacks are executed to perform cleanup jobs for the dead CPU, so this
cpu online check in padata_cpu_dead() is unnecessary.
Similarly, when executing padata_cpu_online() during the
CPUHP_AP_ONLINE_DYN phase, the CPU has already been set in the
cpu_online_mask, the action even occurs earlier than the
CPUHP_AP_ONLINE_IDLE stage.
Remove this unnecessary cpu online check in __padata_add_cpu() and
__padata_remove_cpu().
Signed-off-by: Chuyi Zhou <zhouchuyi@bytedance.com>
Acked-by: Daniel Jordan <daniel.m.jordan@oracle.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Fix all kernel-doc warnings in ecc.h:
- use correct kernel-doc format
- add some Returns: sections
- fix spelling and parameter names
Fixes these warnings:
Warning: include/crypto/internal/ecc.h:82 function parameter 'nbytes' not
described in 'ecc_digits_from_bytes'
Warning: include/crypto/internal/ecc.h:82 function parameter 'out' not
described in 'ecc_digits_from_bytes'
Warning: include/crypto/internal/ecc.h:95 No description found for return
value of 'ecc_is_key_valid'
Warning: include/crypto/internal/ecc.h:110 No description found for return
value of 'ecc_gen_privkey'
Warning: include/crypto/internal/ecc.h:124 No description found for return
value of 'ecc_make_pub_key'
Warning: include/crypto/internal/ecc.h:143 No description found for return
value of 'crypto_ecdh_shared_secret'
Warning: include/crypto/internal/ecc.h:182 No description found for return
value of 'vli_is_zero'
Warning: include/crypto/internal/ecc.h:194 No description found for return
value of 'vli_cmp'
Warning: include/crypto/internal/ecc.h:209 function parameter 'right' not
described in 'vli_sub'
Warning: include/crypto/internal/ecc.h:271 expecting prototype for
ecc_aloc_point(). Prototype was for ecc_alloc_point() instead
Warning: include/crypto/internal/ecc.h:287 function parameter 'point' not
described in 'ecc_point_is_zero'
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Use correct function parameter names and add Returns: sections to
eliminate all kernel-doc warnings in des.h:
Warning: include/crypto/des.h:41 function parameter 'keylen' not
described in 'des_expand_key'
Warning: include/crypto/des.h:41 No description found for return value
of 'des_expand_key'
Warning: include/crypto/des.h:54 function parameter 'keylen' not
described in 'des3_ede_expand_key'
Warning: include/crypto/des.h:54 No description found for return value
of 'des3_ede_expand_key'
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Correct kernel-doc:
- add the @extra function parameter
- add "_extra" to the mismatched function name
- spell the "cmpl" parameter correctly
to avoid these warnings:
Warning: include/crypto/acompress.h:251 function parameter 'extra' not
described in 'acomp_request_alloc_extra'
Warning: include/crypto/acompress.h:251 expecting prototype for
acomp_request_alloc(). Prototype was for acomp_request_alloc_extra()
instead
Warning: include/crypto/acompress.h:327 function parameter 'cmpl' not
described in 'acomp_request_set_callback'
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The else-if and else branches in rht_key_get_hash() both compute a hash
using either params.hashfn or jhash, differing only in the source of
key_len (params.key_len vs ht->p.key_len). Merge the two branches into
one by using the ternary `params.key_len ?: ht->p.key_len` to select
the key length, removing the duplicated logic.
This also improves the performance of the else branch which previously
always used jhash and never fell through to jhash2. This branch is going
to be used by BPF resizable hashmap, which wraps rhashtable:
https://lore.kernel.org/bpf/20260205-rhash-v1-0-30dd6d63c462@meta.com/
Signed-off-by: Mykyta Yatsenko <yatsenko@meta.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Replace snprintf("%s", ...) with the faster and more direct strscpy().
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
I'm leaving Intel soon. Remove myself as maintainer of Keem Bay
architecture and related crypto drivers.
The INTEL KEEM BAY OCS AES/SM4 CRYPTO DRIVER has no replacement
maintainer available, so mark it as Orphan.
Signed-off-by: Daniele Alessandrelli <daniele.alessandrelli@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Document the Inline Crypto Engine (ICE) on the Eliza platform.
Signed-off-by: Abel Vesa <abel.vesa@oss.qualcomm.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
CONFIG_CRYPTO_DEV_VMX has been moved to arch/powerpc - delete the
disabled build directive.
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Call atmel_ecc_i2c_client_free() to release the I2C client reserved by
atmel_ecc_i2c_client_alloc() when crypto_alloc_kpp() fails. Otherwise
->tfm_count will be out of sync.
Fixes: 11105693fa ("crypto: atmel-ecc - introduce Microchip / Atmel ECC driver")
Cc: stable@vger.kernel.org
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Return early if atmel_i2c_send_receive() fails to avoid checking
potentially uninitialized data in 'cmd.data'.
Cc: stable@vger.kernel.org
Fixes: e05ce444e9 ("crypto: atmel-sha204a - add reading from otp zone")
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
- Split out .modinfo section from ELF_DETAILS macro, as that macro may
be used in other areas that expect to discard .modinfo, breaking
certain image layouts
- Adjust genksyms parser to handle optional attributes in certain
declarations, necessary after commit 07919126ec ("netfilter:
annotate NAT helper hook pointers with __rcu")
- Include resolve_btfids in external module build created by
scripts/package/install-extmod-build when it may be run on
external modules
- Avoid removing objtool binary with 'make clean', as it is required for
external module builds
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQR74yXHMTGczQHYypIdayaRccAalgUCaat33gAKCRAdayaRccAa
lizMAQCxm0P5WsJf3ydYR+5ZzzM7wreNtpMVMXsCbwOKBGY3VwEAyvB7om1a00Ex
Z6WFa9P4VKW+L4PWMnWoyxcnvl/CdgM=
=mvIb
-----END PGP SIGNATURE-----
Merge tag 'kbuild-fixes-7.0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/kbuild/linux
Pull Kbuild fixes from Nathan Chancellor:
- Split out .modinfo section from ELF_DETAILS macro, as that macro may
be used in other areas that expect to discard .modinfo, breaking
certain image layouts
- Adjust genksyms parser to handle optional attributes in certain
declarations, necessary after commit 07919126ec ("netfilter:
annotate NAT helper hook pointers with __rcu")
- Include resolve_btfids in external module build created by
scripts/package/install-extmod-build when it may be run on external
modules
- Avoid removing objtool binary with 'make clean', as it is required
for external module builds
* tag 'kbuild-fixes-7.0-2' of git://git.kernel.org/pub/scm/linux/kernel/git/kbuild/linux:
kbuild: Leave objtool binary around with 'make clean'
kbuild: install-extmod-build: Package resolve_btfids if necessary
genksyms: Fix parsing a declarator with a preceding attribute
kbuild: Split .modinfo out from ELF_DETAILS
- Fix kexec/hibernation hang due to bogus read-only mappings.
- Fix sparse warnings in our cmpxchg() implementation.
- Prevent runtime-const being used in modules, just like x86.
- Fix broken elision of access flag modifications for contiguous entries
on systems without support for hardware updates.
- Fix a broken SVE selftest that was testing the wrong instruction.
-----BEGIN PGP SIGNATURE-----
iQFEBAABCgAuFiEEPxTL6PPUbjXGY88ct6xw3ITBYzQFAmmrH5wQHHdpbGxAa2Vy
bmVsLm9yZwAKCRC3rHDchMFjNLiWB/40+A3Q3gz9VB3obupFeC/s688TjGMwLbIO
m03Qu/ulGwBZaPRPZxsxnr8pFZKjSple5NJiHv5kQ/wR4Cfc4zwF2zOSdRvAI/c3
qPT2YL0CcVt0OgbWd2VCjiThTuFREewdCqRWbmkDaPYd27k0KWY14gHHpriRw7XM
QY0OOz8wrWi3lg2Wyvub9wWLkyjKtFlrkwZaACD5D90k/CwKVgncC1z4vh41hQxk
qjxdygNJt2sV+31+F7QMoY/rbyVnUkdSvWSwe9z2Bs9mwebaoGgx4c1l47Wq+oQD
NiVgHOZnPQkDgd2MWkUiCwzAr6C3B0aF2BCu+NTgILkbX7PyG792
=knFu
-----END PGP SIGNATURE-----
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fixes from Will Deacon:
"The main changes are a fix to the way in which we manage the access
flag setting for mappings using the contiguous bit and a fix for a
hang on the kexec/hibernation path.
Summary:
- Fix kexec/hibernation hang due to bogus read-only mappings
- Fix sparse warnings in our cmpxchg() implementation
- Prevent runtime-const being used in modules, just like x86
- Fix broken elision of access flag modifications for contiguous
entries on systems without support for hardware updates
- Fix a broken SVE selftest that was testing the wrong instruction"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
selftest/arm64: Fix sve2p1_sigill() to hwcap test
arm64: contpte: fix set_access_flags() no-op check for SMMU/ATS faults
arm64: make runtime const not usable by modules
arm64: mm: Add PTE_DIRTY back to PAGE_KERNEL* to fix kexec/hibernation
arm64: Silence sparse warnings caused by the type casting in (cmp)xchg
The pmtu.sh kselftest configures OVS using ovs-dpctl.py and falls back
to ovs-vsctl only when ovs-dpctl.py fails. However, ovs-dpctl.py exits
with a success status when the installed pyroute2 package version is
lower than 0.6, even though the OVS datapath is not configured.
As a result, pmtu.sh assumes that the setup was successful and
continues running the test, which later fails due to the missing
OVS configuration.
Fix the exit code handling in ovs-dpctl.py so that pmtu.sh can detect
that the setup did not complete successfully and fall back to
ovs-vsctl.
Signed-off-by: Aleksei Oladko <aleksey.oladko@virtuozzo.com>
Reviewed-by: Aaron Conole <aconole@redhat.com>
Link: https://patch.msgid.link/20260306000127.519064-3-aleksey.oladko@virtuozzo.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Some of the sizing logic through tracer_alloc_buffers() uses int
internally, causing unexpected behavior if the user passes a value that
does not fit in an int (on my x86 machine, the result is uselessly tiny
buffers).
Fix by plumbing the parameter's real type (unsigned long) through to the
ring buffer allocation functions, which already use unsigned long.
It has always been possible to create larger ring buffers via the sysfs
interface: this only affects the cmdline parameter.
Cc: stable@vger.kernel.org
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://patch.msgid.link/bff42a4288aada08bdf74da3f5b67a2c28b761f8.1772852067.git.calvin@wbinvd.org
Fixes: 73c5162aa3 ("tracing: keep ring buffer to minimum size till used")
Signed-off-by: Calvin Owens <calvin@wbinvd.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Driver core holds a reference to the USB interface and its parent USB
device while the interface is bound to a driver and there is no need to
take additional references unless the structures are needed after
disconnect.
Drop the redundant device reference to reduce cargo culting, make it
easier to spot drivers where an extra reference is needed, and reduce
the risk of memory leaks when drivers fail to release it.
Signed-off-by: Johan Hovold <johan@kernel.org>
Link: https://patch.msgid.link/20260305105006.16415-1-johan@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Koichiro Den says:
====================
net: ntb_netdev: Add Multi-queue support
ntb_netdev currently hard-codes a single NTB transport queue pair, which
means the datapath effectively runs as a single-queue netdev regardless
of available CPUs / parallel flows.
The longer-term motivation here is throughput scale-out: allow
ntb_netdev to grow beyond the single-QP bottleneck and make it possible
to spread TX/RX work across multiple queue pairs as link speeds and core
counts keep increasing.
Multi-queue also unlocks the standard networking knobs on top of it. In
particular, once the device exposes multiple TX queues, qdisc/tc can
steer flows/traffic classes into different queues (via
skb->queue_mapping), enabling per-flow/per-class scheduling and QoS in a
familiar way.
Usage
=====
1. Ensure the NTB device you want to use has multiple Memory Windows.
2. modprobe ntb_transport on both sides, if it's not built-in.
3. modprobe ntb_netdev on both sides, if it's not built-in.
4. Use ethtool -L to configure the desired number of queues.
The default number of real (combined) queues is 1.
e.g. ethtool -L eth0 combined 2 # to increase
ethtool -L eth0 combined 1 # to reduce back to 1
Note:
* If the NTB device has only a single Memory Window, ethtool -L eth0
combined N (N > 1) fails with:
"netlink error: No space left on device".
* ethtool -L can be executed while the net_device is up.
Compatibility
=============
The default remains a single queue, so behavior is unchanged unless
the user explicitly increases the number of queues.
Kernel base
===========
ntb-next (latest as of 2026-03-06):
commit 7b3302c687ca ("ntb_hw_amd: Fix incorrect debug message in link
disable path")
Testing / Results
=================
Environment / command line:
- 2x R-Car S4 Spider boards
"Kernel base" (see above) + this series
TCP:
[RC] $ sudo iperf3 -s
[EP] $ sudo iperf3 -Z -c ${SERVER_IP} -l 65480 -w 512M -P 4
UDP:
[RC] $ sudo iperf3 -s
[EP] $ sudo iperf3 -ub0 -c ${SERVER_IP} -l 65480 -w 512M -P 4
Without this series:
TCP / UDP : 589 Mbps / 580 Mbps
With this series (default single queue):
TCP / UDP : 583 Mbps / 583 Mbps
With this series + `ethtool -L eth0 combined 2`:
TCP / UDP : 576 Mbps / 584 Mbps
With this series + `ethtool -L eth0 combined 2` + [1], where flows are
properly distributed across queues:
TCP / UDP : 1.13 Gbps / 1.16 Gbps (re-measured with v3)
The 575~590 Mbps variation is run-to-run variance i.e. no measurable
regression or improvement is observed with a single queue. The key
point is scaling from ~600 Mbps to ~1.20 Gbps once flows are
distributed across multiple queues.
Note: On R-Car S4 Spider, only BAR2 is usable for ntb_transport MW.
For testing, BAR2 was expanded from 1 MiB to 2 MiB and split into two
Memory Windows. A follow-up series is planned to add split BAR support
for vNTB. On platforms where multiple BARs can be used for the
datapath, this series should allow >=2 queues without additional
changes.
[1] [PATCH v2 00/10] NTB: epf: Enable per-doorbell bit handling while keeping legacy offset
https://lore.kernel.org/linux-pci/20260227084955.3184017-1-den@valinux.co.jp/
(subject was accidentally incorrect in the original posting)
====================
Link: https://patch.msgid.link/20260305155639.1885517-1-den@valinux.co.jp
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Support dynamic queue pair addition/removal via ethtool channels.
Use the combined channel count to control the number of netdev TX/RX
queues, each corresponding to a ntb_transport queue pair.
When the number of queues is reduced, tear down and free the removed
ntb_transport queue pairs (not just deactivate them) so other
ntb_transport clients can reuse the freed resources.
When the number of queues is increased, create additional queue pairs up
to NTB_NETDEV_MAX_QUEUES (=64). The effective limit is determined by the
underlying ntb_transport implementation and NTB hardware resources (the
number of MWs), so set_channels may return -ENOSPC if no more QPs can be
allocated.
Keep the default at one queue pair to preserve the previous behavior.
Signed-off-by: Koichiro Den <den@valinux.co.jp>
Link: https://patch.msgid.link/20260305155639.1885517-5-den@valinux.co.jp
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Implementing .set_channels will otherwise duplicate the same multi-queue
operations at multiple call sites. Factor out the following helpers:
- ntb_netdev_update_carrier(): carrier is switched on when at least
one QP link is up
- ntb_netdev_queue_rx_drain(): drain and free all queued RX packets
for one QP
- ntb_netdev_queue_rx_fill(): prefill RX ring for one QP
No functional change.
Signed-off-by: Koichiro Den <den@valinux.co.jp>
Link: https://patch.msgid.link/20260305155639.1885517-4-den@valinux.co.jp
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When ntb_netdev is extended to multiple ntb_transport queue pairs, the
netdev carrier can be up as long as at least one QP link is up. In that
setup, a given QP may be link-down while the carrier remains on.
Make the link event handler start/stop the corresponding netdev TX
subqueue and drive carrier state based on whether any QP link is up.
Also guard subqueue wake/start points in the TX completion and timer
paths so a subqueue is not restarted while its QP link is down.
Stop all queues in ndo_open() and let the link event handler wake each
subqueue once ntb_transport link negotiation succeeds.
Signed-off-by: Koichiro Den <den@valinux.co.jp>
Link: https://patch.msgid.link/20260305155639.1885517-3-den@valinux.co.jp
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Prepare ntb_netdev for multi-queue operation by moving queue-pair state
out of struct ntb_netdev.
Introduce struct ntb_netdev_queue to carry the ntb_transport_qp pointer,
the per-QP TX timer and queue id. Pass this object as the callback
context and convert the RX/TX handlers and link event path accordingly.
The probe path allocates a fixed upper bound for netdev queues while
instantiating only a single ntb_transport queue pair, preserving the
previous behavior. Also store client_dev for future queue pair
creation/removal via the ntb_transport API.
Signed-off-by: Koichiro Den <den@valinux.co.jp>
Link: https://patch.msgid.link/20260305155639.1885517-2-den@valinux.co.jp
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Even if we get a dma_mapping_error() while mapping an RX buffer, we
should still update rx_ring->head to ensure that the buffers we were
able to allocate and map are used. Fix this by breaking out to the
existing code after the loop, analogous to the existing handling for skb
allocation failure.
Fixes: bfec6d7f20 ("net: spacemit: Add K1 Ethernet MAC")
Signed-off-by: Vivian Wang <wangruikang@iscas.ac.cn>
Link: https://patch.msgid.link/20260305-k1-ethernet-more-fixes-v2-1-e4e434d65055@iscas.ac.cn
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Johan Hovold says:
====================
nfc: drop redundant USB device references
Driver core holds a reference to the USB interface and its parent USB
device while the interface is bound to a driver and there is no need to
take additional references unless the structures are needed after
disconnect.
Drop redundant device references to reduce cargo culting, make it easier
to spot drivers where an extra reference is needed, and reduce the risk
of memory leaks when drivers fail to release them.
====================
Link: https://patch.msgid.link/20260305111019.18030-1-johan@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Driver core holds a reference to the USB interface and its parent USB
device while the interface is bound to a driver and there is no need to
take additional references unless the structures are needed after
disconnect.
Drop the redundant device reference to reduce cargo culting, make it
easier to spot drivers where an extra reference is needed, and reduce
the risk of memory leaks when drivers fail to release it.
Signed-off-by: Johan Hovold <johan@kernel.org>
Link: https://patch.msgid.link/20260305111019.18030-3-johan@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Driver core holds a reference to the USB interface and its parent USB
device while the interface is bound to a driver and there is no need to
take additional references unless the structures are needed after
disconnect.
Drop the redundant device reference to reduce cargo culting, make it
easier to spot drivers where an extra reference is needed, and reduce
the risk of memory leaks when drivers fail to release it.
Signed-off-by: Johan Hovold <johan@kernel.org>
Link: https://patch.msgid.link/20260305111019.18030-2-johan@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Eduard Zingerman says:
====================
bpf: Fix u32/s32 bounds when ranges cross min/max boundary
Cover the following cases in range refinement logic for 32-bit ranges:
- s32 range crosses U32_MAX/0 boundary, positive part of the s32 range
overlaps with u32 range.
- s32 range crosses U32_MAX/0 boundary, negative part of the s32 range
overlaps with u32 range.
These cases are already handled for 64-bit range refinement.
Without the fix the test in patch 2 is rejected by the verifier.
The test was reduced from sched-ext program.
Changelog:
- v2 -> v3:
- Reverted da653de268 (Paul)
- Removed !BPF_F_TEST_REG_INVARIANTS flag from
crossing_32_bit_signed_boundary_2() (Paul)
- v1 -> v2:
- Extended commit message and comments (Emil)
- Targeting 'bpf' tree instead of bpf-next (Alexei)
v1: https://lore.kernel.org/bpf/9a23fbacdc6d33ec8fcb3f6988395b5129f75369.camel@gmail.com/T
v2: https://lore.kernel.org/bpf/20260305-bpf-32-bit-range-overflow-v2-0-7169206a3041@gmail.com/
---
====================
Link: https://patch.msgid.link/20260306-bpf-32-bit-range-overflow-v3-0-f7f67e060a6b@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Two test cases for signed/unsigned 32-bit bounds refinement
when s32 range crosses the sign boundary:
- s32 range [S32_MIN..1] overlapping with u32 range [3..U32_MAX],
s32 range tail before sign boundary overlaps with u32 range.
- s32 range [-3..5] overlapping with u32 range [0..S32_MIN+3],
s32 range head after the sign boundary overlaps with u32 range.
This covers both branches added in the __reg32_deduce_bounds().
Also, crossing_32_bit_signed_boundary_2() no longer triggers invariant
violations.
Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com>
Reviewed-by: Paul Chaignon <paul.chaignon@gmail.com>
Acked-by: Shung-Hsi Yu <shung-hsi.yu@suse.com>
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20260306-bpf-32-bit-range-overflow-v3-2-f7f67e060a6b@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Same as in __reg64_deduce_bounds(), refine s32/u32 ranges
in __reg32_deduce_bounds() in the following situations:
- s32 range crosses U32_MAX/0 boundary, positive part of the s32 range
overlaps with u32 range:
0 U32_MAX
| [xxxxxxxxxxxxxx u32 range xxxxxxxxxxxxxx] |
|----------------------------|----------------------------|
|xxxxx s32 range xxxxxxxxx] [xxxxxxx|
0 S32_MAX S32_MIN -1
- s32 range crosses U32_MAX/0 boundary, negative part of the s32 range
overlaps with u32 range:
0 U32_MAX
| [xxxxxxxxxxxxxx u32 range xxxxxxxxxxxxxx] |
|----------------------------|----------------------------|
|xxxxxxxxx] [xxxxxxxxxxxx s32 range |
0 S32_MAX S32_MIN -1
- No refinement if ranges overlap in two intervals.
This helps for e.g. consider the following program:
call %[bpf_get_prandom_u32];
w0 &= 0xffffffff;
if w0 < 0x3 goto 1f; // on fall-through u32 range [3..U32_MAX]
if w0 s> 0x1 goto 1f; // on fall-through s32 range [S32_MIN..1]
if w0 s< 0x0 goto 1f; // range can be narrowed to [S32_MIN..-1]
r10 = 0;
1: ...;
The reg_bounds.c selftest is updated to incorporate identical logic,
refinement based on non-overflowing range halves:
((x ∩ [0, smax]) ∩ (y ∩ [0, smax])) ∪
((x ∩ [smin,-1]) ∩ (y ∩ [smin,-1]))
Reported-by: Andrea Righi <arighi@nvidia.com>
Reported-by: Emil Tsalapatis <emil@etsalapatis.com>
Closes: https://lore.kernel.org/bpf/aakqucg4vcujVwif@gpd4/T/
Reviewed-by: Emil Tsalapatis <emil@etsalapatis.com>
Acked-by: Shung-Hsi Yu <shung-hsi.yu@suse.com>
Signed-off-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/r/20260306-bpf-32-bit-range-overflow-v3-1-f7f67e060a6b@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
rxrpc_kernel_lookup_peer() can also return error pointers in addition to
NULL, so just checking for NULL is not sufficient.
Fix this by:
(1) Changing rxrpc_kernel_lookup_peer() to return -ENOMEM rather than NULL
on allocation failure.
(2) Making the callers in afs use IS_ERR() and PTR_ERR() to pass on the
error code returned.
Fixes: 72904d7b9b ("rxrpc, afs: Allow afs to pin rxrpc_peer objects")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Co-developed-by: David Howells <dhowells@redhat.com>
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: Simon Horman <horms@kernel.org>
cc: linux-afs@lists.infradead.org
Link: https://patch.msgid.link/368272.1772713861@warthog.procyon.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Vladimir Oltean says:
====================
Further SJA1105 phylink link replay fixups
While I was playing around with the subsystem knowledge in Chris Mason's
review-prompts to see what LLMs would have needed to catch the bug
behind commit bfd264fbbb ("net: dsa: sja1105: protect link replay
helpers against NULL phylink instance"), it flagged another issue
instead, which IMO is valid. This is being fixed in patch 2/2.
Patch 1/2 is preparatory reordering for that.
I haven't noticed any physical issues, it only has to do with the
soundness of the new call path introduced in January in commit
0b2edc531e ("net: dsa: sja1105: let phylink help with the replay of
link callbacks").
====================
Link: https://patch.msgid.link/20260304220900.3865120-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
