Add estimate of how much vram we need to reserve for RAS
when caculating the total available vram.
v2: apply the change to MP0 v13_0_2 and v13_0_14
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
And also make sure the value of msg[1].len should be in the range of u16.
Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Fix LDS size interpretation: 512 bytes (>= gfx12) vs 256 (< gfx12).
Ensure STATE_PRIV.BARRIER_COMPLETE cannot change after reading or
before writing. Other waves in the threadgroup may cause this field
to assert if they complete the barrier.
Do not overwrite EXCP_FLAG_PRIV.{SAVE_CONTEXT,HOST_TRAP} when
restoring this register. Both of these fields can assert while the
wavefront is running the trap handler.
Signed-off-by: Jay Cornwall <jay.cornwall@amd.com>
Reviewed-by: Lancelot Six <lancelot.six@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We have messages when the VCN fails to initialize and
there is no need to report on success.
Also PSP loading FWs is the default for production.
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Sonny Jiang <sonjiang@amd.com>
Signed-off-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This patch changes the implementation of AMDGPU_PTE_MTYPE_GFX12,
clear the bits before setting the new one.
This fixed the potential issue that GFX12 setting memory to NC.
v2: Clear mtype field before setting the new one (Alex)
v3: Fix typo (Felix)
Suggested-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: longlyao <Longlong.Yao@amd.com>
Signed-off-by: Shane Xiao <shane.xiao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Now that all libbpf/bpftool code switched to btf_field_iter, remove
btf_type_visit_type_ids() and btf_type_visit_str_offs() callback-based
helpers as not needed anymore.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20240605001629.4061937-6-andrii@kernel.org
Switch bpftool's code which is using libbpf-internal
btf_type_visit_type_ids() helper to new btf_field_iter functionality.
This makes bpftool code simpler, but also unblocks removing libbpf's
btf_type_visit_type_ids() helper completely.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Reviewed-by: Quentin Monnet <qmo@kernel.org>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20240605001629.4061937-5-andrii@kernel.org
Use new BTF field iterator logic to replace all the callback-based
visitor calls. There is still a .BTF.ext callback-based visitor APIs
that should be converted, which will happens as a follow up.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20240605001629.4061937-4-andrii@kernel.org
Switch all BPF linker code dealing with iterating BTF type ID and string
offset fields to new btf_field_iter facilities.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20240605001629.4061937-3-andrii@kernel.org
Implement iterator-based type ID and string offset BTF field iterator.
This is used extensively in BTF-handling code and BPF linker code for
various sanity checks, rewriting IDs/offsets, etc. Currently this is
implemented as visitor pattern calling custom callbacks, which makes the
logic (especially in simple cases) unnecessarily obscure and harder to
follow.
Having equivalent functionality using iterator pattern makes for simpler
to understand and maintain code. As we add more code for BTF processing
logic in libbpf, it's best to switch to iterator pattern before adding
more callback-based code.
The idea for iterator-based implementation is to record offsets of
necessary fields within fixed btf_type parts (which should be iterated
just once), and, for kinds that have multiple members (based on vlen
field), record where in each member necessary fields are located.
Generic iteration code then just keeps track of last offset that was
returned and handles N members correctly. Return type is just u32
pointer, where NULL is returned when all relevant fields were already
iterated.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20240605001629.4061937-2-andrii@kernel.org
Use preempt_model_preemptible() to detect a preemptible kernel when
deciding whether or not to reschedule in order to drop a contended
spinlock or rwlock. Because PREEMPT_DYNAMIC selects PREEMPTION, kernels
built with PREEMPT_DYNAMIC=y will yield contended locks even if the live
preemption model is "none" or "voluntary". In short, make kernels with
dynamically selected models behave the same as kernels with statically
selected models.
Somewhat counter-intuitively, NOT yielding a lock can provide better
latency for the relevant tasks/processes. E.g. KVM x86's mmu_lock, a
rwlock, is often contended between an invalidation event (takes mmu_lock
for write) and a vCPU servicing a guest page fault (takes mmu_lock for
read). For _some_ setups, letting the invalidation task complete even
if there is mmu_lock contention provides lower latency for *all* tasks,
i.e. the invalidation completes sooner *and* the vCPU services the guest
page fault sooner.
But even KVM's mmu_lock behavior isn't uniform, e.g. the "best" behavior
can vary depending on the host VMM, the guest workload, the number of
vCPUs, the number of pCPUs in the host, why there is lock contention, etc.
In other words, simply deleting the CONFIG_PREEMPTION guard (or doing the
opposite and removing contention yielding entirely) needs to come with a
big pile of data proving that changing the status quo is a net positive.
Opportunistically document this side effect of preempt=full, as yielding
contended spinlocks can have significant, user-visible impact.
Fixes: c597bfddc9 ("sched: Provide Kconfig support for default dynamic preempt mode")
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ankur Arora <ankur.a.arora@oracle.com>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Link: https://lore.kernel.org/kvm/ef81ff36-64bb-4cfe-ae9b-e3acf47bff24@proxmox.com
Move the declarations and inlined implementations of the preempt_model_*()
helpers to preempt.h so that they can be referenced in spinlock.h without
creating a potential circular dependency between spinlock.h and sched.h.
No functional change intended.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ankur Arora <ankur.a.arora@oracle.com>
Link: https://lkml.kernel.org/r/20240528003521.979836-2-ankur.a.arora@oracle.com
in bch2_move_data_btree, we might start with the trans unlocked from a
previous loop iteration - we need a trans_begin() before iter_init().
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
This fixes an issue where setting a device to durability=0 after it's
been used makes it impossible to remove.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Ingo reported that he was seeing these when hitting Control+C during a
perf tools build:
Makefile.perf:1149: *** Missing bpftool input for generating vmlinux.h. Stop.
The failure happens when you don't have vmlinux.h or vmlinux with BTF.
ifeq ($(VMLINUX_H),)
ifeq ($(VMLINUX_BTF),)
$(error Missing bpftool input for generating vmlinux.h)
endif
endif
VMLINUX_BTF can be empty if you didn't build a kernel or it doesn't have
a BTF section and the current kernel also has no BTF. This is totally
ok.
But VMLINUX_H should be set to the minimal version in the source tree
(unless you overwrite it manually) when you don't pass GEN_VMLINUX_H=1
(which requires VMLINUX_BTF should not be empty). The problem is that
it's defined in Makefile.config which is not included for `make clean`.
Reported-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Ingo Molnar <mingo@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Link: http://lore.kernel.org/lkml/CAM9d7ch5HTr+k+_GpbMrX0HUo5BZ11byh1xq0Two7B7RQACuNw@mail.gmail.com
Link: http://lore.kernel.org/lkml/ZjssGrj+abyC6mYP@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
This reverts commit 7d1405c71d.
This causes segfaults in some cases, as reported by Milian:
```
sudo /usr/bin/perf record -z --call-graph dwarf -e cycles -e
raw_syscalls:sys_enter ls
...
[ perf record: Woken up 3 times to write data ]
malloc(): invalid next size (unsorted)
Aborted
```
Backtrace with GDB + debuginfod:
```
malloc(): invalid next size (unsorted)
Thread 1 "perf" received signal SIGABRT, Aborted.
__pthread_kill_implementation (threadid=<optimized out>, signo=signo@entry=6,
no_tid=no_tid@entry=0) at pthread_kill.c:44
Downloading source file /usr/src/debug/glibc/glibc/nptl/pthread_kill.c
44 return INTERNAL_SYSCALL_ERROR_P (ret) ? INTERNAL_SYSCALL_ERRNO
(ret) : 0;
(gdb) bt
#0 __pthread_kill_implementation (threadid=<optimized out>,
signo=signo@entry=6, no_tid=no_tid@entry=0) at pthread_kill.c:44
#1 0x00007ffff6ea8eb3 in __pthread_kill_internal (threadid=<optimized out>,
signo=6) at pthread_kill.c:78
#2 0x00007ffff6e50a30 in __GI_raise (sig=sig@entry=6) at ../sysdeps/posix/
raise.c:26
#3 0x00007ffff6e384c3 in __GI_abort () at abort.c:79
#4 0x00007ffff6e39354 in __libc_message_impl (fmt=fmt@entry=0x7ffff6fc22ea
"%s\n") at ../sysdeps/posix/libc_fatal.c:132
#5 0x00007ffff6eb3085 in malloc_printerr (str=str@entry=0x7ffff6fc5850
"malloc(): invalid next size (unsorted)") at malloc.c:5772
#6 0x00007ffff6eb657c in _int_malloc (av=av@entry=0x7ffff6ff6ac0
<main_arena>, bytes=bytes@entry=368) at malloc.c:4081
#7 0x00007ffff6eb877e in __libc_calloc (n=<optimized out>,
elem_size=<optimized out>) at malloc.c:3754
#8 0x000055555569bdb6 in perf_session.do_write_header ()
#9 0x00005555555a373a in __cmd_record.constprop.0 ()
#10 0x00005555555a6846 in cmd_record ()
#11 0x000055555564db7f in run_builtin ()
#12 0x000055555558ed77 in main ()
```
Valgrind memcheck:
```
==45136== Invalid write of size 8
==45136== at 0x2B38A5: perf_event__synthesize_id_sample (in /usr/bin/perf)
==45136== by 0x157069: __cmd_record.constprop.0 (in /usr/bin/perf)
==45136== by 0x15A845: cmd_record (in /usr/bin/perf)
==45136== by 0x201B7E: run_builtin (in /usr/bin/perf)
==45136== by 0x142D76: main (in /usr/bin/perf)
==45136== Address 0x6a866a8 is 0 bytes after a block of size 40 alloc'd
==45136== at 0x4849BF3: calloc (vg_replace_malloc.c:1675)
==45136== by 0x3574AB: zalloc (in /usr/bin/perf)
==45136== by 0x1570E0: __cmd_record.constprop.0 (in /usr/bin/perf)
==45136== by 0x15A845: cmd_record (in /usr/bin/perf)
==45136== by 0x201B7E: run_builtin (in /usr/bin/perf)
==45136== by 0x142D76: main (in /usr/bin/perf)
==45136==
==45136== Syscall param write(buf) points to unaddressable byte(s)
==45136== at 0x575953D: __libc_write (write.c:26)
==45136== by 0x575953D: write (write.c:24)
==45136== by 0x35761F: ion (in /usr/bin/perf)
==45136== by 0x357778: writen (in /usr/bin/perf)
==45136== by 0x1548F7: record__write (in /usr/bin/perf)
==45136== by 0x15708A: __cmd_record.constprop.0 (in /usr/bin/perf)
==45136== by 0x15A845: cmd_record (in /usr/bin/perf)
==45136== by 0x201B7E: run_builtin (in /usr/bin/perf)
==45136== by 0x142D76: main (in /usr/bin/perf)
==45136== Address 0x6a866a8 is 0 bytes after a block of size 40 alloc'd
==45136== at 0x4849BF3: calloc (vg_replace_malloc.c:1675)
==45136== by 0x3574AB: zalloc (in /usr/bin/perf)
==45136== by 0x1570E0: __cmd_record.constprop.0 (in /usr/bin/perf)
==45136== by 0x15A845: cmd_record (in /usr/bin/perf)
==45136== by 0x201B7E: run_builtin (in /usr/bin/perf)
==45136== by 0x142D76: main (in /usr/bin/perf)
==45136==
-----
Closes: https://lore.kernel.org/linux-perf-users/23879991.0LEYPuXRzz@milian-workstation/
Reported-by: Milian Wolff <milian.wolff@kdab.com>
Tested-by: Milian Wolff <milian.wolff@kdab.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: stable@kernel.org # 6.8+
Link: https://lore.kernel.org/lkml/Zl9ksOlHJHnKM70p@x1
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
We observed that the overhead on trigger_load_balance(), now renamed
sched_balance_trigger(), has risen with a system's core counts.
For an OLTP workload running 6.8 kernel on a 2 socket x86 systems
having 96 cores/socket, we saw that 0.7% cpu cycles are spent in
trigger_load_balance(). On older systems with fewer cores/socket, this
function's overhead was less than 0.1%.
The cause of this overhead was that there are multiple cpus calling
kick_ilb(flags), updating the balancing work needed to a common idle
load balancer cpu. The ilb_cpu's flags field got updated unconditionally
with atomic_fetch_or(). The atomic read and writes to ilb_cpu's flags
causes much cache bouncing and cpu cycles overhead. This is seen in the
annotated profile below.
kick_ilb():
if (ilb_cpu < 0)
test %r14d,%r14d
↑ js 6c
flags = atomic_fetch_or(flags, nohz_flags(ilb_cpu));
mov $0x2d600,%rdi
movslq %r14d,%r8
mov %rdi,%rdx
add -0x7dd0c3e0(,%r8,8),%rdx
arch_atomic_read():
0.01 mov 0x64(%rdx),%esi
35.58 add $0x64,%rdx
arch_atomic_fetch_or():
static __always_inline int arch_atomic_fetch_or(int i, atomic_t *v)
{
int val = arch_atomic_read(v);
do { } while (!arch_atomic_try_cmpxchg(v, &val, val | i));
0.03 157: mov %r12d,%ecx
arch_atomic_try_cmpxchg():
return arch_try_cmpxchg(&v->counter, old, new);
0.00 mov %esi,%eax
arch_atomic_fetch_or():
do { } while (!arch_atomic_try_cmpxchg(v, &val, val | i));
or %esi,%ecx
arch_atomic_try_cmpxchg():
return arch_try_cmpxchg(&v->counter, old, new);
0.01 lock cmpxchg %ecx,(%rdx)
42.96 ↓ jne 2d2
kick_ilb():
With instrumentation, we found that 81% of the updates do not result in
any change in the ilb_cpu's flags. That is, multiple cpus are asking
the ilb_cpu to do the same things over and over again, before the ilb_cpu
has a chance to run NOHZ load balance.
Skip updates to ilb_cpu's flags if no new work needs to be done.
Such updates do not change ilb_cpu's NOHZ flags. This requires an extra
atomic read but it is less expensive than frequent unnecessary atomic
updates that generate cache bounces.
We saw that on the OLTP workload, cpu cycles from trigger_load_balance()
(or sched_balance_trigger()) got reduced from 0.7% to 0.2%.
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Chen Yu <yu.c.chen@intel.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20240531205452.65781-1-tim.c.chen@linux.intel.com
The call of rcu_idle_enter() from within cpuidle_idle_call() was
removed in commit 1098582a0f ("sched,idle,rcu: Push rcu_idle deeper
into the idle path") which makes the comment out of place.
Signed-off-by: Christian Loehle <christian.loehle@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/5b936388-47df-4050-9229-6617a6c2bba5@arm.com
For ${atomic}_sub_and_test() the @i parameter is the value to subtract,
not add. Fix the typo in the kerneldoc template and generate the headers
with this update.
Fixes: ad8110706f ("locking/atomic: scripts: generate kerneldoc comments")
Suggested-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/20240515133844.3502360-1-cmllamas@google.com
In our production environment, we found many hung tasks which are
blocked for more than 18 hours. Their call traces are like this:
[346278.191038] __schedule+0x2d8/0x890
[346278.191046] schedule+0x4e/0xb0
[346278.191049] perf_event_free_task+0x220/0x270
[346278.191056] ? init_wait_var_entry+0x50/0x50
[346278.191060] copy_process+0x663/0x18d0
[346278.191068] kernel_clone+0x9d/0x3d0
[346278.191072] __do_sys_clone+0x5d/0x80
[346278.191076] __x64_sys_clone+0x25/0x30
[346278.191079] do_syscall_64+0x5c/0xc0
[346278.191083] ? syscall_exit_to_user_mode+0x27/0x50
[346278.191086] ? do_syscall_64+0x69/0xc0
[346278.191088] ? irqentry_exit_to_user_mode+0x9/0x20
[346278.191092] ? irqentry_exit+0x19/0x30
[346278.191095] ? exc_page_fault+0x89/0x160
[346278.191097] ? asm_exc_page_fault+0x8/0x30
[346278.191102] entry_SYSCALL_64_after_hwframe+0x44/0xae
The task was waiting for the refcount become to 1, but from the vmcore,
we found the refcount has already been 1. It seems that the task didn't
get woken up by perf_event_release_kernel() and got stuck forever. The
below scenario may cause the problem.
Thread A Thread B
... ...
perf_event_free_task perf_event_release_kernel
...
acquire event->child_mutex
...
get_ctx
... release event->child_mutex
acquire ctx->mutex
...
perf_free_event (acquire/release event->child_mutex)
...
release ctx->mutex
wait_var_event
acquire ctx->mutex
acquire event->child_mutex
# move existing events to free_list
release event->child_mutex
release ctx->mutex
put_ctx
... ...
In this case, all events of the ctx have been freed, so we couldn't
find the ctx in free_list and Thread A will miss the wakeup. It's thus
necessary to add a wakeup after dropping the reference.
Fixes: 1cf8dfe8a6 ("perf/core: Fix race between close() and fork()")
Signed-off-by: Haifeng Xu <haifeng.xu@shopee.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/20240513103948.33570-1-haifeng.xu@shopee.com
The WCN7850 package contains discrete modules for WLAN and Bluetooth. They
are powered by the Power Management Unit (PMU) that takes inputs from the
host and provides LDO outputs. Extend the bindings for QCA6390 to also
document this model.
Acked-by: Mark Brown <broonie@kernel.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Link: https://msgid.link/r/20240605121450.23697-3-brgl@bgdev.pl
Signed-off-by: Mark Brown <broonie@kernel.org>
The QCA6390 package contains discrete modules for WLAN and Bluetooth. They
are powered by the Power Management Unit (PMU) that takes inputs from the
host and provides LDO outputs. This document describes this module.
Acked-by: Mark Brown <broonie@kernel.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Link: https://msgid.link/r/20240605121450.23697-2-brgl@bgdev.pl
Signed-off-by: Mark Brown <broonie@kernel.org>
Keep kvm_apicv_inhibit enum naming consistent with the current pattern by
renaming the reason/enumerator defined as APICV_INHIBIT_REASON_DISABLE to
APICV_INHIBIT_REASON_DISABLED.
No functional change intended.
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Alejandro Jimenez <alejandro.j.jimenez@oracle.com>
Link: https://lore.kernel.org/r/20240506225321.3440701-3-alejandro.j.jimenez@oracle.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Use the tracing infrastructure helper __print_flags() for printing flag
bitfields, to enhance the trace output by displaying a string describing
each of the inhibit reasons set.
The kvm_apicv_inhibit_changed tracepoint currently shows the raw bitmap
value, requiring the user to consult the source file where the inhibit
reasons are defined to decode the trace output.
Signed-off-by: Alejandro Jimenez <alejandro.j.jimenez@oracle.com>
Reviewed-by: Vasant Hegde <vasant.hegde@amd.com>
Link: https://lore.kernel.org/r/20240506225321.3440701-2-alejandro.j.jimenez@oracle.com
Co-developed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Add KVM_CAP_X86_APIC_BUS_CYCLES_NS capability to configure the APIC
bus clock frequency for APIC timer emulation.
Allow KVM_ENABLE_CAPABILITY(KVM_CAP_X86_APIC_BUS_CYCLES_NS) to set the
frequency in nanoseconds. When using this capability, the user space
VMM should configure CPUID leaf 0x15 to advertise the frequency.
Vishal reported that the TDX guest kernel expects a 25MHz APIC bus
frequency but ends up getting interrupts at a significantly higher rate.
The TDX architecture hard-codes the core crystal clock frequency to
25MHz and mandates exposing it via CPUID leaf 0x15. The TDX architecture
does not allow the VMM to override the value.
In addition, per Intel SDM:
"The APIC timer frequency will be the processor’s bus clock or core
crystal clock frequency (when TSC/core crystal clock ratio is
enumerated in CPUID leaf 0x15) divided by the value specified in
the divide configuration register."
The resulting 25MHz APIC bus frequency conflicts with the KVM hardcoded
APIC bus frequency of 1GHz.
The KVM doesn't enumerate CPUID leaf 0x15 to the guest unless the user
space VMM sets it using KVM_SET_CPUID. If the CPUID leaf 0x15 is
enumerated, the guest kernel uses it as the APIC bus frequency. If not,
the guest kernel measures the frequency based on other known timers like
the ACPI timer or the legacy PIT. As reported by Vishal the TDX guest
kernel expects a 25MHz timer frequency but gets timer interrupt more
frequently due to the 1GHz frequency used by KVM.
To ensure that the guest doesn't have a conflicting view of the APIC bus
frequency, allow the userspace to tell KVM to use the same frequency that
TDX mandates instead of the default 1Ghz.
Reported-by: Vishal Annapurve <vannapurve@google.com>
Closes: https://lore.kernel.org/lkml/20231006011255.4163884-1-vannapurve@google.com
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Reviewed-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Co-developed-by: Reinette Chatre <reinette.chatre@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Reviewed-by: Yuan Yao <yuan.yao@intel.com>
Link: https://lore.kernel.org/r/6748a4c12269e756f0c48680da8ccc5367c31ce7.1714081726.git.reinette.chatre@intel.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Introduce the VM variable "nanoseconds per APIC bus cycle" in
preparation to make the APIC bus frequency configurable.
The TDX architecture hard-codes the core crystal clock frequency to
25MHz and mandates exposing it via CPUID leaf 0x15. The TDX architecture
does not allow the VMM to override the value.
In addition, per Intel SDM:
"The APIC timer frequency will be the processor’s bus clock or core
crystal clock frequency (when TSC/core crystal clock ratio is
enumerated in CPUID leaf 0x15) divided by the value specified in
the divide configuration register."
The resulting 25MHz APIC bus frequency conflicts with the KVM hardcoded
APIC bus frequency of 1GHz.
Introduce the VM variable "nanoseconds per APIC bus cycle" to prepare
for allowing userspace to tell KVM to use the frequency that TDX mandates
instead of the default 1Ghz. Doing so ensures that the guest doesn't have
a conflicting view of the APIC bus frequency.
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
[reinette: rework changelog]
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Link: https://lore.kernel.org/r/ae75ce37c6c38bb4efd10a0a41932984c40b24ac.1714081726.git.reinette.chatre@intel.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Remove APIC_BUS_FREQUENCY and calculate it based on nanoseconds per APIC
bus cycle. APIC_BUS_FREQUENCY is used only for HV_X64_MSR_APIC_FREQUENCY.
The MSR is not frequently read, calculate it every time.
There are two constants related to the APIC bus frequency:
APIC_BUS_FREQUENCY and APIC_BUS_CYCLE_NS.
Only one value is required because one can be calculated from the other:
APIC_BUS_CYCLES_NS = 1000 * 1000 * 1000 / APIC_BUS_FREQUENCY.
Remove APIC_BUS_FREQUENCY and instead calculate it when needed.
This prepares for support of configurable APIC bus frequency by
requiring to change only a single variable.
Suggested-by: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>
Reviewed-by: Maxim Levitsky <maximlevitsky@gmail.com>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Reviewed-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
[reinette: rework changelog]
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Link: https://lore.kernel.org/r/76a659d0898e87ebd73ee7c922f984a87a6ab370.1714081726.git.reinette.chatre@intel.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Use {READ,WRITE}_ONCE() to access kvm->last_boosted_vcpu to ensure the
loads and stores are atomic. In the extremely unlikely scenario the
compiler tears the stores, it's theoretically possible for KVM to attempt
to get a vCPU using an out-of-bounds index, e.g. if the write is split
into multiple 8-bit stores, and is paired with a 32-bit load on a VM with
257 vCPUs:
CPU0 CPU1
last_boosted_vcpu = 0xff;
(last_boosted_vcpu = 0x100)
last_boosted_vcpu[15:8] = 0x01;
i = (last_boosted_vcpu = 0x1ff)
last_boosted_vcpu[7:0] = 0x00;
vcpu = kvm->vcpu_array[0x1ff];
As detected by KCSAN:
BUG: KCSAN: data-race in kvm_vcpu_on_spin [kvm] / kvm_vcpu_on_spin [kvm]
write to 0xffffc90025a92344 of 4 bytes by task 4340 on cpu 16:
kvm_vcpu_on_spin (arch/x86/kvm/../../../virt/kvm/kvm_main.c:4112) kvm
handle_pause (arch/x86/kvm/vmx/vmx.c:5929) kvm_intel
vmx_handle_exit (arch/x86/kvm/vmx/vmx.c:?
arch/x86/kvm/vmx/vmx.c:6606) kvm_intel
vcpu_run (arch/x86/kvm/x86.c:11107 arch/x86/kvm/x86.c:11211) kvm
kvm_arch_vcpu_ioctl_run (arch/x86/kvm/x86.c:?) kvm
kvm_vcpu_ioctl (arch/x86/kvm/../../../virt/kvm/kvm_main.c:?) kvm
__se_sys_ioctl (fs/ioctl.c:52 fs/ioctl.c:904 fs/ioctl.c:890)
__x64_sys_ioctl (fs/ioctl.c:890)
x64_sys_call (arch/x86/entry/syscall_64.c:33)
do_syscall_64 (arch/x86/entry/common.c:?)
entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)
read to 0xffffc90025a92344 of 4 bytes by task 4342 on cpu 4:
kvm_vcpu_on_spin (arch/x86/kvm/../../../virt/kvm/kvm_main.c:4069) kvm
handle_pause (arch/x86/kvm/vmx/vmx.c:5929) kvm_intel
vmx_handle_exit (arch/x86/kvm/vmx/vmx.c:?
arch/x86/kvm/vmx/vmx.c:6606) kvm_intel
vcpu_run (arch/x86/kvm/x86.c:11107 arch/x86/kvm/x86.c:11211) kvm
kvm_arch_vcpu_ioctl_run (arch/x86/kvm/x86.c:?) kvm
kvm_vcpu_ioctl (arch/x86/kvm/../../../virt/kvm/kvm_main.c:?) kvm
__se_sys_ioctl (fs/ioctl.c:52 fs/ioctl.c:904 fs/ioctl.c:890)
__x64_sys_ioctl (fs/ioctl.c:890)
x64_sys_call (arch/x86/entry/syscall_64.c:33)
do_syscall_64 (arch/x86/entry/common.c:?)
entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)
value changed: 0x00000012 -> 0x00000000
Fixes: 217ece6129 ("KVM: use yield_to instead of sleep in kvm_vcpu_on_spin")
Cc: stable@vger.kernel.org
Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://lore.kernel.org/r/20240510092353.2261824-1-leitao@debian.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Use the max mappable GPA via GuestPhysBits advertised by KVM to calculate
max_gfn. Currently some selftests (e.g. access_tracking_perf_test,
dirty_log_test...) add RAM regions close to max_gfn, so guest may access
GPA beyond its mappable range and cause infinite loop.
Adjust max_gfn in vm_compute_max_gfn() since x86 selftests already
overrides vm_compute_max_gfn() specifically to deal with goofy edge cases.
Reported-by: Yi Lai <yi1.lai@intel.com>
Signed-off-by: Tao Su <tao1.su@linux.intel.com>
Tested-by: Yi Lai <yi1.lai@intel.com>
Reviewed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Link: https://lore.kernel.org/r/20240513014003.104593-1-tao1.su@linux.intel.com
[sean: tweak name, add comment and sanity check]
Signed-off-by: Sean Christopherson <seanjc@google.com>
Currrentl a 32 bit 1u value is being shifted more than 32 bits causing
overflow and incorrect checking of bits 32-63. Fix this by using the
BIT_ULL macro for shifting bits.
Detected by cppcheck:
sev_init2_tests.c:108:34: error: Shifting 32-bit value by 63 bits is
undefined behaviour [shiftTooManyBits]
Fixes: dfc083a181 ("selftests: kvm: add tests for KVM_SEV_INIT2")
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Link: https://lore.kernel.org/r/20240523154102.2236133-1-colin.i.king@gmail.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Tariq Toukan says:
====================
mlx5 core fixes 20240603
This small patchset provides two bug fixes from the team to the mlx5 core driver.
Series generated against:
commit 33700a0c9b ("net/tcp: Don't consider TCP_CLOSE in TCP_AO_ESTABLISHED")
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
In case pci channel becomes offline the driver should not wait for PCI
reads during health dump and recovery flow. The driver has timeout for
each of these loops trying to read PCI, so it would fail anyway.
However, in case of recovery waiting till timeout may cause the pci
error_detected() callback fail to meet pci_dpc_recovered() wait timeout.
Fixes: b3bd076f75 ("net/mlx5: Report devlink health on FW fatal issues")
Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Shay Drori <shayd@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The mainline MTK ethernet driver suffers long time from rarly but
annoying tx queue timeouts. We think that this is caused by fixed
dma sizes hardcoded for all SoCs.
We suspect this problem arises from a low level of free TX DMADs,
the TX Ring alomost full.
The transmit timeout is caused by the Tx queue not waking up. The
Tx queue stops when the free counter is less than ring->thres, and
it will wake up once the free counter is greater than ring->thres.
If the CPU is too late to wake up the Tx queues, it may cause a
transmit timeout.
Therefore, we increased the TX and RX DMADs to improve this error
situation.
Use the dma-size implementation from SDK in a per SoC manner. In
difference to SDK we have no RSS feature yet, so all RX/TX sizes
should be raised from 512 to 2048 byte except fqdma on mt7988 to
avoid the tx timeout issue.
Fixes: 656e705243 ("net-next: mediatek: add support for MT7623 ethernet")
Suggested-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: Frank Wunderlich <frank-w@public-files.de>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
*-objs suffix is reserved rather for (user-space) host programs while
usually *-y suffix is used for kernel drivers (although *-objs works
for that purpose for now).
Let's correct the old usages of *-objs in Makefiles.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20240508151611.1444352-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Will Deacon <will@kernel.org>
Static checker is complaining about the ASID possibly set uninitialized.
This only happens in case of error and this value would be ignored anyway.
A simple fix would be just to initialize the local variable to zero,
this path will only be reached on the first attach to a domain where
the CD is already initialized to zero.
This avoids having to bloat the function with an error path.
Closes: https://lore.kernel.org/linux-iommu/849e3d77-0a3c-43c4-878d-a0e061c8cd61@moroto.mountain/T/#u
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Mostafa Saleh <smostafa@google.com>
Fixes: 04905c17f6 ("iommu/arm-smmu-v3: Build the whole CD in arm_smmu_make_s1_cd()")
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20240604185218.2602058-1-smostafa@google.com
Signed-off-by: Will Deacon <will@kernel.org>
Kevin Yang says:
====================
tcp: add sysctl_tcp_rto_min_us
Adding a sysctl knob to allow user to specify a default
rto_min at socket init time.
After this patch series, the rto_min will has multiple sources:
route option has the highest precedence, followed by the
TCP_BPF_RTO_MIN socket option, followed by this new
tcp_rto_min_us sysctl.
v3:
fix typo, simplify min/max_t to min/max
v2:
fit line width to 80 column.
v2: https://lore.kernel.org/netdev/20240530153436.2202800-1-yyd@google.com/
v1: https://lore.kernel.org/netdev/20240528171320.1332292-1-yyd@google.com/
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Adding a sysctl knob to allow user to specify a default
rto_min at socket init time, other than using the hard
coded 200ms default rto_min.
Note that the rto_min route option has the highest precedence
for configuring this setting, followed by the TCP_BPF_RTO_MIN
socket option, followed by the tcp_rto_min_us sysctl.
Signed-off-by: Kevin Yang <yyd@google.com>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Yuchung Cheng <ycheng@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rto_min now has multiple sources, ordered by preprecedence high to
low: ip route option rto_min, icsk->icsk_rto_min.
When derive delack_max from rto_min, we should not only use ip
route option, but should use tcp_rto_min helper to get the correct
rto_min.
Signed-off-by: Kevin Yang <yyd@google.com>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Yuchung Cheng <ycheng@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
In some configurations __const_iowrite32_copy() does not get inlined
and gcc runs into the BUILD_BUG():
In file included from <command-line>:
In function '__const_memcpy_toio_aligned32',
inlined from '__const_iowrite32_copy' at arch/arm64/include/asm/io.h:203:3,
inlined from '__const_iowrite32_copy' at arch/arm64/include/asm/io.h:199:20:
include/linux/compiler_types.h:487:45: error: call to '__compiletime_assert_538' declared with attribute error: BUILD_BUG failed
487 | _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
| ^
include/linux/compiler_types.h:468:25: note: in definition of macro '__compiletime_assert'
468 | prefix ## suffix(); \
| ^~~~~~
include/linux/compiler_types.h:487:9: note: in expansion of macro '_compiletime_assert'
487 | _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
| ^~~~~~~~~~~~~~~~~~~
include/linux/build_bug.h:39:37: note: in expansion of macro 'compiletime_assert'
39 | #define BUILD_BUG_ON_MSG(cond, msg) compiletime_assert(!(cond), msg)
| ^~~~~~~~~~~~~~~~~~
include/linux/build_bug.h:59:21: note: in expansion of macro 'BUILD_BUG_ON_MSG'
59 | #define BUILD_BUG() BUILD_BUG_ON_MSG(1, "BUILD_BUG failed")
| ^~~~~~~~~~~~~~~~
arch/arm64/include/asm/io.h:193:17: note: in expansion of macro 'BUILD_BUG'
193 | BUILD_BUG();
| ^~~~~~~~~
Move the check for constant arguments into the inline function to ensure
it is still constant if the compiler decides against inlining it, and
mark them as __always_inline to override the logic that sometimes leads
to the compiler not producing the simplified output.
Note that either the __always_inline annotation or the check for a
constant value are sufficient here, but combining the two looks cleaner
as it also avoids the macro. With clang-8 and older, the macro was still
needed, but all versions of gcc and clang can reliably perform constant
folding here.
Fixes: ead79118da ("arm64/io: Provide a WC friendly __iowriteXX_copy()")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20240604210006.668912-1-arnd@kernel.org
Signed-off-by: Will Deacon <will@kernel.org>
Since "Headphone Switch" kcontrol name has already been used by da7219,
rename the control name from "Headphone" to "Headphones" to prevent the
colision. Also, this change makes kcontrol name align with the one in
mt8186-mt6366-da7219-max98357.c.
Fixes: 9c7388baa2 ("ASoC: mediatek: mt8183-da7219-max98357: Map missing jack kcontrols")
Change-Id: I9ae69a4673cd04786b247cc514fdd20f878ef009
Signed-off-by: Hsin-Te Yuan <yuanhsinte@chromium.org>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://msgid.link/r/20240531-da7219-v1-1-ac3343f3ae6a@chromium.org
Signed-off-by: Mark Brown <broonie@kernel.org>
When TRCM mode is enabled, I2S RX and TX clocks are synchronized through
selected clock source. Without this fix BCLK and LRCK might get parented
to an uninitialized MCLK and the DAI will receive data at wrong pace.
However, unlike in original i2s-tdm driver, there is no need to manually
synchronize mclk_rx and mclk_tx, as only one gets used anyway.
Tested on a board with RK3568 SoC and Silergy SY24145S codec with enabled and
disabled TRCM mode.
Fixes: 9e2ab4b18e ("ASoC: rockchip: i2s-tdm: Fix inaccurate sampling rates")
Signed-off-by: Alibek Omarov <a1ba.omarov@gmail.com>
Reviewed-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Link: https://msgid.link/r/20240604184752.697313-1-a1ba.omarov@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
There's a random number of structure pre-declaration lines in
kernel/sched/sched.h, some of which are unnecessary duplicates.
Move them to the head & order them a bit for readability.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: linux-kernel@vger.kernel.org
SEV-ES and thus SNP guest mandates LBR Virtualization to be _always_ ON.
Although commit b7e4be0a22 ("KVM: SEV-ES: Delegate LBR virtualization
to the processor") did the correct change for SEV-ES guests, it missed
the SNP. Fix it.
Reported-by: Srikanth Aithal <sraithal@amd.com>
Fixes: b7e4be0a22 ("KVM: SEV-ES: Delegate LBR virtualization to the processor")
Signed-off-by: Ravi Bangoria <ravi.bangoria@amd.com>
Message-ID: <20240605114810.1304-1-ravi.bangoria@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Jaroslav reports Dell's OMSA Systems Management Data Engine
expects NLM_DONE in a separate recvmsg(), both for rtnl_dump_ifinfo()
and inet_dump_ifaddr(). We already added a similar fix previously in
commit 460b0d33cf ("inet: bring NLM_DONE out to a separate recv() again")
Instead of modifying all the dump handlers, and making them look
different than modern for_each_netdev_dump()-based dump handlers -
put the workaround in rtnetlink code. This will also help us move
the custom rtnl-locking from af_netlink in the future (in net-next).
Note that this change is not touching rtnl_dump_all(). rtnl_dump_all()
is different kettle of fish and a potential problem. We now mix families
in a single recvmsg(), but NLM_DONE is not coalesced.
Tested:
./cli.py --dbg-small-recv 4096 --spec netlink/specs/rt_addr.yaml \
--dump getaddr --json '{"ifa-family": 2}'
./cli.py --dbg-small-recv 4096 --spec netlink/specs/rt_route.yaml \
--dump getroute --json '{"rtm-family": 2}'
./cli.py --dbg-small-recv 4096 --spec netlink/specs/rt_link.yaml \
--dump getlink
Fixes: 3e41af9076 ("rtnetlink: use xarray iterator to implement rtnl_dump_ifinfo()")
Fixes: cdb2f80f1c ("inet: use xa_array iterator to implement inet_dump_ifaddr()")
Reported-by: Jaroslav Pulchart <jaroslav.pulchart@gooddata.com>
Link: https://lore.kernel.org/all/CAK8fFZ7MKoFSEzMBDAOjoUt+vTZRRQgLDNXEOfdCCXSoXXKE0g@mail.gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Like previous patch does in TCP, we need to adhere to RFC 1213:
"tcpCurrEstab OBJECT-TYPE
...
The number of TCP connections for which the current state
is either ESTABLISHED or CLOSE- WAIT."
So let's consider CLOSE-WAIT sockets.
The logic of counting
When we increment the counter?
a) Only if we change the state to ESTABLISHED.
When we decrement the counter?
a) if the socket leaves ESTABLISHED and will never go into CLOSE-WAIT,
say, on the client side, changing from ESTABLISHED to FIN-WAIT-1.
b) if the socket leaves CLOSE-WAIT, say, on the server side, changing
from CLOSE-WAIT to LAST-ACK.
Fixes: d9cd27b8cd ("mptcp: add CurrEstab MIB counter support")
Signed-off-by: Jason Xing <kernelxing@tencent.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
