| Age | Commit message (Collapse) | Author |
|---|
|
__replace_page()
Ever since commit b756a3b5e7ea ("mm: device exclusive memory access") we
can return with a device-exclusive entry from page_vma_mapped_walk().
__replace_page() is not prepared for that, so teach it about these PFN
swap PTEs. Note that device-private entries are so far not applicable on
that path, because GUP would never have returned such folios (conversion
to device-private happens by page migration, not in-place conversion of
the PTE).
There is a race between GUP and us locking the folio to look it up using
page_vma_mapped_walk(), so this is likely a fix (unless something else
could prevent that race, but it doesn't look like). pte_pfn() on
something that is not a present pte could give use garbage, and we'd
wrongly mess up the mapcount because it was already adjusted by calling
folio_remove_rmap_pte() when making the entry device-exclusive.
Link: https://lkml.kernel.org/r/20250210193801.781278-9-david@redhat.com
Fixes: b756a3b5e7ea ("mm: device exclusive memory access")
Signed-off-by: David Hildenbrand <david@redhat.com>
Tested-by: Alistair Popple <apopple@nvidia.com>
Cc: Alex Shi <alexs@kernel.org>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Dave Airlie <airlied@gmail.com>
Cc: Jann Horn <jannh@google.com>
Cc: Jason Gunthorpe <jgg@nvidia.com>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Karol Herbst <kherbst@redhat.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Lyude <lyude@redhat.com>
Cc: "Masami Hiramatsu (Google)" <mhiramat@kernel.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Pasha Tatashin <pasha.tatashin@soleen.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: SeongJae Park <sj@kernel.org>
Cc: Simona Vetter <simona.vetter@ffwll.ch>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Yanteng Si <si.yanteng@linux.dev>
Cc: Barry Song <v-songbaohua@oppo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Follow the advice in Documentation/filesystems/sysfs.rst:
"- show() should only use sysfs_emit() or sysfs_emit_at() when formatting
the value to be returned to user space."
No change in functionality intended.
[ mingo: Updated the changelog ]
Signed-off-by: XieLudan <xie.ludan@zte.com.cn>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250315141738452lXIH39UJAXlCmcATCzcBv@zte.com.cn
|
|
The 'size' parameter is optional and strscpy() automatically determines
the length of the destination buffer using sizeof() if the argument is
omitted. This makes the explicit sizeof() calls unnecessary.
Furthermore, KSYM_NAME_LEN is equal to sizeof(name) and can also be
removed. Remove them to shorten and simplify the code.
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20250310192336.442994-1-thorsten.blum@linux.dev
|
|
Jann reported a possible issue when trampoline_check_ip returns
address near the bottom of the address space that is allowed to
call into the syscall if uretprobes are not set up:
https://lore.kernel.org/bpf/202502081235.5A6F352985@keescook/T/#m9d416df341b8fbc11737dacbcd29f0054413cbbf
Though the mmap minimum address restrictions will typically prevent
creating mappings there, let's make sure uretprobe syscall checks
for that.
Fixes: ff474a78cef5 ("uprobe: Add uretprobe syscall to speed up return probe")
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Kees Cook <kees@kernel.org>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20250212220433.3624297-1-jolsa@kernel.org
|
|
Make sure that perf_try_init_event() doesn't leave event->pmu nor
event->destroy set on failure.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20250205102449.110145835@infradead.org
|
|
When f_ops->mmap() returns failure, m_ops->close() is *not* called.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20241104135519.248358497@infradead.org
|
|
In prepration for being able to unregister a PMU with existing events,
it becomes important to detach struct perf_cpu_pmu_context lifetimes
from that of struct pmu.
Notably struct perf_cpu_pmu_context embeds a struct perf_event_pmu_context
that can stay referenced until the last event goes.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20241104135518.760214287@infradead.org
|
|
This puts 'all' of perf_mmap() under single event->mmap_mutex.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20241104135519.582252957@infradead.org
|
|
AFAICT there is no actual benefit from the mutex drop on re-try. The
'worst' case scenario is that we instantly re-gain the mutex without
perf_mmap_close() getting it. So might as well make that the normal
case.
Reflow the code to make the ring buffer detach case naturally flow
into the no ring buffer case.
[ mingo: Forward ported it ]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20241104135519.463607258@infradead.org
|
|
Perform CSE and such.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20241104135519.354909594@infradead.org
|
|
Identity-transform:
if (c) {
X1;
} else {
Y;
goto l;
}
X2;
l:
into the simpler:
if (c) {
X1;
X2;
} else {
Y;
}
[ mingo: Forward ported it ]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20241104135519.095904637@infradead.org
|
|
Ensure perf_event_free_bpf_prog() is safe to call a second time;
notably without making any references to event->pmu when there is no
prog left.
Note: perf_event_detach_bpf_prog() might leave a stale event->prog
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20241104135518.978956692@infradead.org
|
|
Replace _free_event()'s use of perf_addr_filters_splice()s use with an
explicit perf_free_addr_filters() with the explicit propery that it is
able to be called a second time without ill effect.
Most notable, referencing event->pmu must be avoided when there are no
filters left (from eg a previous call).
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20241104135518.868460518@infradead.org
|
|
As a preparation for adding yet another indirection.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20241104135518.650051565@infradead.org
|
|
perf_cpu_pmu_context::pmu_disable_count
Because it makes no sense to have two per-cpu allocations per pmu.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20241104135518.518730578@infradead.org
|
|
Using the previous simplifications, transition perf_event_alloc() to
the cleanup way of things -- reducing error path magic.
[ mingo: Ported it to recent kernels. ]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20241104135518.410755241@infradead.org
|
|
Use the <linux/cleanup.h> guard() and scoped_guard() infrastructure
to simplify the control flow.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20241104135518.302444446@infradead.org
|
|
Using the previously introduced perf_pmu_free() and a new IDR helper,
simplify the perf_pmu_register error paths.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20241104135518.198937277@infradead.org
|
|
The error path of perf_pmu_register() is of course very similar to a
subset of perf_pmu_unregister(). Extract this common part in
perf_pmu_free() and simplify things.
[ mingo: Forward ported it ]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20241104135518.090915501@infradead.org
|
|
The error cleanup sequence in perf_event_alloc() is a subset of the
existing _free_event() function (it must of course be).
Split this out into __free_event() and simplify the error path.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20241104135517.967889521@infradead.org
|
|
Currently, __reserve_bp_slot() returns -ENOSPC for unsupported
breakpoint types on the architecture. For example, powerpc
does not support hardware instruction breakpoints. This causes
the perf_skip BPF selftest to fail, as neither ENOENT nor
EOPNOTSUPP is returned by perf_event_open for unsupported
breakpoint types. As a result, the test that should be skipped
for this arch is not correctly identified.
To resolve this, hw_breakpoint_event_init() should exit early by
checking for unsupported breakpoint types using
hw_breakpoint_slots_cached() and return the appropriate error
(-EOPNOTSUPP).
Signed-off-by: Saket Kumar Bhaskar <skb99@linux.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Marco Elver <elver@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Link: https://lore.kernel.org/r/20250303092451.1862862-1-skb99@linux.ibm.com
|
|
fixes
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
There is a fairly obvious race between perf_init_event() doing
idr_find() and perf_pmu_register() doing idr_alloc() with an
incompletely initialized PMU pointer.
Avoid by doing idr_alloc() on a NULL pointer to register the id, and
swizzling the real struct pmu pointer at the end using idr_replace().
Also making sure to not set struct pmu members after publishing
the struct pmu, duh.
[ introduce idr_cmpxchg() in order to better handle the idr_replace()
error case -- if it were to return an unexpected pointer, it will
already have replaced the value and there is no going back. ]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20241104135517.858805880@infradead.org
|
|
Commit a63fbed776c7 ("perf/tracing/cpuhotplug: Fix locking order")
placed pmus_lock inside pmus_srcu, this makes perf_pmu_unregister()
trip lockdep.
Move the locking about such that only pmu_idr and pmus (list) are
modified while holding pmus_lock. This avoids doing synchronize_srcu()
while holding pmus_lock and all is well again.
Fixes: a63fbed776c7 ("perf/tracing/cpuhotplug: Fix locking order")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20241104135517.679556858@infradead.org
|
|
It seems that the attr parameter was never been used in security
checks since it was first introduced by:
commit da97e18458fb ("perf_event: Add support for LSM and SELinux checks")
so remove it.
Signed-off-by: Luo Gengkun <luogengkun@huaweicloud.com>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Paul Moore <paul@paul-moore.com>
|
|
hprobe_expire() is used to atomically switch pending uretprobe instance
(struct return_instance) from being SRCU protected to be refcounted.
This can be done from background timer thread, or synchronously within
current thread when task is forked.
In the former case, return_instance has to be protected through RCU read
lock, and that's what hprobe_expire() used to check with
lockdep_assert(rcu_read_lock_held()).
But in the latter case (hprobe_expire() called from dup_utask()) there
is no RCU lock being held, and it's both unnecessary and incovenient.
Inconvenient due to the intervening memory allocations inside
dup_return_instance()'s loop. Unnecessary because dup_utask() is called
synchronously in current thread, and no uretprobe can run at that point,
so return_instance can't be freed either.
So drop rcu_read_lock_held() condition, and expand corresponding comment
to explain necessary lifetime guarantees. lockdep_assert()-detected
issue is a false positive.
Fixes: dd1a7567784e ("uprobes: SRCU-protect uretprobe lifetime (with timeout)")
Reported-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20250225223214.2970740-1-andrii@kernel.org
|
|
A low attr::freq value cannot be set via IOC_PERIOD on some platforms.
The perf_event_check_period() introduced in:
81ec3f3c4c4d ("perf/x86: Add check_period PMU callback")
was intended to check the period, rather than the frequency.
A low frequency may be mistakenly rejected by limit_period().
Fix it.
Fixes: 81ec3f3c4c4d ("perf/x86: Add check_period PMU callback")
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20250117151913.3043942-2-kan.liang@linux.intel.com
Closes: https://lore.kernel.org/lkml/20250115154949.3147-1-ravi.bangoria@amd.com/
|
|
We triggered the following crash in syzkaller tests:
BUG: Bad page state in process syz.7.38 pfn:1eff3
page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x1eff3
flags: 0x3fffff00004004(referenced|reserved|node=0|zone=1|lastcpupid=0x1fffff)
raw: 003fffff00004004 ffffe6c6c07bfcc8 ffffe6c6c07bfcc8 0000000000000000
raw: 0000000000000000 0000000000000000 00000000fffffffe 0000000000000000
page dumped because: PAGE_FLAGS_CHECK_AT_FREE flag(s) set
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014
Call Trace:
<TASK>
dump_stack_lvl+0x32/0x50
bad_page+0x69/0xf0
free_unref_page_prepare+0x401/0x500
free_unref_page+0x6d/0x1b0
uprobe_write_opcode+0x460/0x8e0
install_breakpoint.part.0+0x51/0x80
register_for_each_vma+0x1d9/0x2b0
__uprobe_register+0x245/0x300
bpf_uprobe_multi_link_attach+0x29b/0x4f0
link_create+0x1e2/0x280
__sys_bpf+0x75f/0xac0
__x64_sys_bpf+0x1a/0x30
do_syscall_64+0x56/0x100
entry_SYSCALL_64_after_hwframe+0x78/0xe2
BUG: Bad rss-counter state mm:00000000452453e0 type:MM_FILEPAGES val:-1
The following syzkaller test case can be used to reproduce:
r2 = creat(&(0x7f0000000000)='./file0\x00', 0x8)
write$nbd(r2, &(0x7f0000000580)=ANY=[], 0x10)
r4 = openat(0xffffffffffffff9c, &(0x7f0000000040)='./file0\x00', 0x42, 0x0)
mmap$IORING_OFF_SQ_RING(&(0x7f0000ffd000/0x3000)=nil, 0x3000, 0x0, 0x12, r4, 0x0)
r5 = userfaultfd(0x80801)
ioctl$UFFDIO_API(r5, 0xc018aa3f, &(0x7f0000000040)={0xaa, 0x20})
r6 = userfaultfd(0x80801)
ioctl$UFFDIO_API(r6, 0xc018aa3f, &(0x7f0000000140))
ioctl$UFFDIO_REGISTER(r6, 0xc020aa00, &(0x7f0000000100)={{&(0x7f0000ffc000/0x4000)=nil, 0x4000}, 0x2})
ioctl$UFFDIO_ZEROPAGE(r5, 0xc020aa04, &(0x7f0000000000)={{&(0x7f0000ffd000/0x1000)=nil, 0x1000}})
r7 = bpf$PROG_LOAD(0x5, &(0x7f0000000140)={0x2, 0x3, &(0x7f0000000200)=ANY=[@ANYBLOB="1800000000120000000000000000000095"], &(0x7f0000000000)='GPL\x00', 0x7, 0x0, 0x0, 0x0, 0x0, '\x00', 0x0, @fallback=0x30, 0xffffffffffffffff, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x10, 0x0, @void, @value}, 0x94)
bpf$BPF_LINK_CREATE_XDP(0x1c, &(0x7f0000000040)={r7, 0x0, 0x30, 0x1e, @val=@uprobe_multi={&(0x7f0000000080)='./file0\x00', &(0x7f0000000100)=[0x2], 0x0, 0x0, 0x1}}, 0x40)
The cause is that zero pfn is set to the PTE without increasing the RSS
count in mfill_atomic_pte_zeropage() and the refcount of zero folio does
not increase accordingly. Then, the operation on the same pfn is performed
in uprobe_write_opcode()->__replace_page() to unconditional decrease the
RSS count and old_folio's refcount.
Therefore, two bugs are introduced:
1. The RSS count is incorrect, when process exit, the check_mm() report
error "Bad rss-count".
2. The reserved folio (zero folio) is freed when folio->refcount is zero,
then free_pages_prepare->free_page_is_bad() report error
"Bad page state".
There is more, the following warning could also theoretically be triggered:
__replace_page()
-> ...
-> folio_remove_rmap_pte()
-> VM_WARN_ON_FOLIO(is_zero_folio(folio), folio)
Considering that uprobe hit on the zero folio is a very rare case, just
reject zero old folio immediately after get_user_page_vma_remote().
[ mingo: Cleaned up the changelog ]
Fixes: 7396fa818d62 ("uprobes/core: Make background page replacement logic account for rss_stat counters")
Fixes: 2b1444983508 ("uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints")
Signed-off-by: Tong Tiangen <tongtiangen@huawei.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Link: https://lore.kernel.org/r/20250224031149.1598949-1-tongtiangen@huawei.com
|
|
Syskaller triggers a warning due to prev_epc->pmu != next_epc->pmu in
perf_event_swap_task_ctx_data(). vmcore shows that two lists have the same
perf_event_pmu_context, but not in the same order.
The problem is that the order of pmu_ctx_list for the parent is impacted by
the time when an event/PMU is added. While the order for a child is
impacted by the event order in the pinned_groups and flexible_groups. So
the order of pmu_ctx_list in the parent and child may be different.
To fix this problem, insert the perf_event_pmu_context to its proper place
after iteration of the pmu_ctx_list.
The follow testcase can trigger above warning:
# perf record -e cycles --call-graph lbr -- taskset -c 3 ./a.out &
# perf stat -e cpu-clock,cs -p xxx // xxx is the pid of a.out
test.c
void main() {
int count = 0;
pid_t pid;
printf("%d running\n", getpid());
sleep(30);
printf("running\n");
pid = fork();
if (pid == -1) {
printf("fork error\n");
return;
}
if (pid == 0) {
while (1) {
count++;
}
} else {
while (1) {
count++;
}
}
}
The testcase first opens an LBR event, so it will allocate task_ctx_data,
and then open tracepoint and software events, so the parent context will
have 3 different perf_event_pmu_contexts. On inheritance, child ctx will
insert the perf_event_pmu_context in another order and the warning will
trigger.
[ mingo: Tidied up the changelog. ]
Fixes: bd2756811766 ("perf: Rewrite core context handling")
Signed-off-by: Luo Gengkun <luogengkun@huaweicloud.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Link: https://lore.kernel.org/r/20250122073356.1824736-1-luogengkun@huaweicloud.com
|
|
The perf_iterate_ctx() function performs RCU list traversal but
currently lacks RCU read lock protection. This causes lockdep warnings
when running perf probe with unshare(1) under CONFIG_PROVE_RCU_LIST=y:
WARNING: suspicious RCU usage
kernel/events/core.c:8168 RCU-list traversed in non-reader section!!
Call Trace:
lockdep_rcu_suspicious
? perf_event_addr_filters_apply
perf_iterate_ctx
perf_event_exec
begin_new_exec
? load_elf_phdrs
load_elf_binary
? lock_acquire
? find_held_lock
? bprm_execve
bprm_execve
do_execveat_common.isra.0
__x64_sys_execve
do_syscall_64
entry_SYSCALL_64_after_hwframe
This protection was previously present but was removed in commit
bd2756811766 ("perf: Rewrite core context handling"). Add back the
necessary rcu_read_lock()/rcu_read_unlock() pair around
perf_iterate_ctx() call in perf_event_exec().
[ mingo: Use scoped_guard() as suggested by Peter ]
Fixes: bd2756811766 ("perf: Rewrite core context handling")
Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20250117-fix_perf_rcu-v1-1-13cb9210fc6a@debian.org
|
|
Move ctl tables to two files:
- perf_event_{paranoid,mlock_kb,max_sample_rate} and
perf_cpu_time_max_percent into kernel/events/core.c
- perf_event_max_{stack,context_per_stack} into
kernel/events/callchain.c
Make static variables and functions that are fully contained in core.c
and callchain.cand remove them from include/linux/perf_event.h.
Additionally six_hundred_forty_kb is moved to callchain.c.
Two new sysctl tables are added ({callchain,events_core}_sysctl_table)
with their respective sysctl registration functions.
This is part of a greater effort to move ctl tables into their
respective subsystems which will reduce the merge conflicts in
kerenel/sysctl.c.
Signed-off-by: Joel Granados <joel.granados@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250218-jag-mv_ctltables-v1-5-cd3698ab8d29@kernel.org
|
|
new patches
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
hrtimer_setup() takes the callback function pointer as argument and
initializes the timer completely.
Replace hrtimer_init() and the open coded initialization of
hrtimer::function with the new setup mechanism.
Patch was created by using Coccinelle.
Signed-off-by: Nam Cao <namcao@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/all/f611e6d3fc6996bbcf0e19fe234f75edebe4332f.1738746821.git.namcao@linutronix.de
|
|
Restricted pointers ("%pK") are not meant to be used through printk().
It can unintentionally expose security sensitive, raw pointer values.
Use regular pointer formatting instead.
For more background, see:
https://lore.kernel.org/lkml/20250113171731-dc10e3c1-da64-4af0-b767-7c7070468023@linutronix.de/
Signed-off-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250217-restricted-pointers-uprobes-v1-1-e8cbe5bb22a7@linutronix.de
|
|
The event may have been updated in the PMU-specific implementation,
e.g., Intel PEBS counters snapshotting. The common code should not
read and overwrite the value.
The PERF_SAMPLE_READ in the data->sample_type can be used to detect
whether the PMU-specific value is available. If yes, avoid the
pmu->read() in the common code. Add a new flag, skip_read, to track the
case.
Factor out a perf_pmu_read() to clean up the code.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20250121152303.3128733-3-kan.liang@linux.intel.com
|
|
This patch introduces a flag to track TIF_SIGPENDING is suppress
temporarily during the uprobe single-step. Upon uprobe singlestep is
handled and the flag is confirmed, it could resume the TIF_SIGPENDING
directly without acquiring the siglock in most case, then reducing
contention and improving overall performance.
I've use the script developed by Andrii in [1] to run benchmark. The CPU
used was Kunpeng916 (Hi1616), 4 NUMA nodes, 64 cores@2.4GHz running the
kernel on next tree + the optimization for get_xol_insn_slot() [2].
before-opt
----------
uprobe-nop ( 1 cpus): 0.907 ± 0.003M/s ( 0.907M/s/cpu)
uprobe-nop ( 2 cpus): 1.676 ± 0.008M/s ( 0.838M/s/cpu)
uprobe-nop ( 4 cpus): 3.210 ± 0.003M/s ( 0.802M/s/cpu)
uprobe-nop ( 8 cpus): 4.457 ± 0.003M/s ( 0.557M/s/cpu)
uprobe-nop (16 cpus): 3.724 ± 0.011M/s ( 0.233M/s/cpu)
uprobe-nop (32 cpus): 2.761 ± 0.003M/s ( 0.086M/s/cpu)
uprobe-nop (64 cpus): 1.293 ± 0.015M/s ( 0.020M/s/cpu)
uprobe-push ( 1 cpus): 0.883 ± 0.001M/s ( 0.883M/s/cpu)
uprobe-push ( 2 cpus): 1.642 ± 0.005M/s ( 0.821M/s/cpu)
uprobe-push ( 4 cpus): 3.086 ± 0.002M/s ( 0.771M/s/cpu)
uprobe-push ( 8 cpus): 3.390 ± 0.003M/s ( 0.424M/s/cpu)
uprobe-push (16 cpus): 2.652 ± 0.005M/s ( 0.166M/s/cpu)
uprobe-push (32 cpus): 2.713 ± 0.005M/s ( 0.085M/s/cpu)
uprobe-push (64 cpus): 1.313 ± 0.009M/s ( 0.021M/s/cpu)
uprobe-ret ( 1 cpus): 1.774 ± 0.000M/s ( 1.774M/s/cpu)
uprobe-ret ( 2 cpus): 3.350 ± 0.001M/s ( 1.675M/s/cpu)
uprobe-ret ( 4 cpus): 6.604 ± 0.000M/s ( 1.651M/s/cpu)
uprobe-ret ( 8 cpus): 6.706 ± 0.005M/s ( 0.838M/s/cpu)
uprobe-ret (16 cpus): 5.231 ± 0.001M/s ( 0.327M/s/cpu)
uprobe-ret (32 cpus): 5.743 ± 0.003M/s ( 0.179M/s/cpu)
uprobe-ret (64 cpus): 4.726 ± 0.016M/s ( 0.074M/s/cpu)
after-opt
---------
uprobe-nop ( 1 cpus): 0.985 ± 0.002M/s ( 0.985M/s/cpu)
uprobe-nop ( 2 cpus): 1.773 ± 0.005M/s ( 0.887M/s/cpu)
uprobe-nop ( 4 cpus): 3.304 ± 0.001M/s ( 0.826M/s/cpu)
uprobe-nop ( 8 cpus): 5.328 ± 0.002M/s ( 0.666M/s/cpu)
uprobe-nop (16 cpus): 6.475 ± 0.002M/s ( 0.405M/s/cpu)
uprobe-nop (32 cpus): 4.831 ± 0.082M/s ( 0.151M/s/cpu)
uprobe-nop (64 cpus): 2.564 ± 0.053M/s ( 0.040M/s/cpu)
uprobe-push ( 1 cpus): 0.964 ± 0.001M/s ( 0.964M/s/cpu)
uprobe-push ( 2 cpus): 1.766 ± 0.002M/s ( 0.883M/s/cpu)
uprobe-push ( 4 cpus): 3.290 ± 0.009M/s ( 0.823M/s/cpu)
uprobe-push ( 8 cpus): 4.670 ± 0.002M/s ( 0.584M/s/cpu)
uprobe-push (16 cpus): 5.197 ± 0.004M/s ( 0.325M/s/cpu)
uprobe-push (32 cpus): 5.068 ± 0.161M/s ( 0.158M/s/cpu)
uprobe-push (64 cpus): 2.605 ± 0.026M/s ( 0.041M/s/cpu)
uprobe-ret ( 1 cpus): 1.833 ± 0.001M/s ( 1.833M/s/cpu)
uprobe-ret ( 2 cpus): 3.384 ± 0.003M/s ( 1.692M/s/cpu)
uprobe-ret ( 4 cpus): 6.677 ± 0.004M/s ( 1.669M/s/cpu)
uprobe-ret ( 8 cpus): 6.854 ± 0.005M/s ( 0.857M/s/cpu)
uprobe-ret (16 cpus): 6.508 ± 0.006M/s ( 0.407M/s/cpu)
uprobe-ret (32 cpus): 5.793 ± 0.009M/s ( 0.181M/s/cpu)
uprobe-ret (64 cpus): 4.743 ± 0.016M/s ( 0.074M/s/cpu)
Above benchmark results demonstrates a obivious improvement in the
scalability of trig-uprobe-nop and trig-uprobe-push, the peak throughput
of which are from 4.5M/s to 6.4M/s and 3.3M/s to 5.1M/s individually.
[1] https://lore.kernel.org/all/20240731214256.3588718-1-andrii@kernel.org
[2] https://lore.kernel.org/all/20240727094405.1362496-1-liaochang1@huawei.com
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Liao Chang <liaochang1@huawei.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20250124093826.2123675-3-liaochang1@huawei.com
|
|
Since clearing a bit in thread_info is an atomic operation, the spinlock
is redundant and can be removed, reducing lock contention is good for
performance.
Signed-off-by: Liao Chang <liaochang1@huawei.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: "Masami Hiramatsu (Google)" <mhiramat@kernel.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Link: https://lore.kernel.org/r/20250124093826.2123675-2-liaochang1@huawei.com
|
|
If a memory allocation fails during dup_mmap(), the maple tree can be left
in an unsafe state for other iterators besides the exit path. All the
locks are dropped before the exit_mmap() call (in mm/mmap.c), but the
incomplete mm_struct can be reached through (at least) the rmap finding
the vmas which have a pointer back to the mm_struct.
Up to this point, there have been no issues with being able to find an
mm_struct that was only partially initialised. Syzbot was able to make
the incomplete mm_struct fail with recent forking changes, so it has been
proven unsafe to use the mm_struct that hasn't been initialised, as
referenced in the link below.
Although 8ac662f5da19f ("fork: avoid inappropriate uprobe access to
invalid mm") fixed the uprobe access, it does not completely remove the
race.
This patch sets the MMF_OOM_SKIP to avoid the iteration of the vmas on the
oom side (even though this is extremely unlikely to be selected as an oom
victim in the race window), and sets MMF_UNSTABLE to avoid other potential
users from using a partially initialised mm_struct.
When registering vmas for uprobe, skip the vmas in an mm that is marked
unstable. Modifying a vma in an unstable mm may cause issues if the mm
isn't fully initialised.
Link: https://lore.kernel.org/all/6756d273.050a0220.2477f.003d.GAE@google.com/
Link: https://lkml.kernel.org/r/20250127170221.1761366-1-Liam.Howlett@oracle.com
Fixes: d24062914837 ("fork: use __mt_dup() to duplicate maple tree in dup_mmap()")
Signed-off-by: Liam R. Howlett <Liam.Howlett@Oracle.com>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Jann Horn <jannh@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Peng Zhang <zhangpeng.00@bytedance.com>
Cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull performance events updates from Ingo Molnar:
"Seqlock optimizations that arose in a perf context and were merged
into the perf tree:
- seqlock: Add raw_seqcount_try_begin (Suren Baghdasaryan)
- mm: Convert mm_lock_seq to a proper seqcount (Suren Baghdasaryan)
- mm: Introduce mmap_lock_speculate_{try_begin|retry} (Suren
Baghdasaryan)
- mm/gup: Use raw_seqcount_try_begin() (Peter Zijlstra)
Core perf enhancements:
- Reduce 'struct page' footprint of perf by mapping pages in advance
(Lorenzo Stoakes)
- Save raw sample data conditionally based on sample type (Yabin Cui)
- Reduce sampling overhead by checking sample_type in
perf_sample_save_callchain() and perf_sample_save_brstack() (Yabin
Cui)
- Export perf_exclude_event() (Namhyung Kim)
Uprobes scalability enhancements: (Andrii Nakryiko)
- Simplify find_active_uprobe_rcu() VMA checks
- Add speculative lockless VMA-to-inode-to-uprobe resolution
- Simplify session consumer tracking
- Decouple return_instance list traversal and freeing
- Ensure return_instance is detached from the list before freeing
- Reuse return_instances between multiple uretprobes within task
- Guard against kmemdup() failing in dup_return_instance()
AMD core PMU driver enhancements:
- Relax privilege filter restriction on AMD IBS (Namhyung Kim)
AMD RAPL energy counters support: (Dhananjay Ugwekar)
- Introduce topology_logical_core_id() (K Prateek Nayak)
- Remove the unused get_rapl_pmu_cpumask() function
- Remove the cpu_to_rapl_pmu() function
- Rename rapl_pmu variables
- Make rapl_model struct global
- Add arguments to the init and cleanup functions
- Modify the generic variable names to *_pkg*
- Remove the global variable rapl_msrs
- Move the cntr_mask to rapl_pmus struct
- Add core energy counter support for AMD CPUs
Intel core PMU driver enhancements:
- Support RDPMC 'metrics clear mode' feature (Kan Liang)
- Clarify adaptive PEBS processing (Kan Liang)
- Factor out functions for PEBS records processing (Kan Liang)
- Simplify the PEBS records processing for adaptive PEBS (Kan Liang)
Intel uncore driver enhancements: (Kan Liang)
- Convert buggy pmu->func_id use to pmu->registered
- Support more units on Granite Rapids"
* tag 'perf-core-2025-01-20' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (33 commits)
perf: map pages in advance
perf/x86/intel/uncore: Support more units on Granite Rapids
perf/x86/intel/uncore: Clean up func_id
perf/x86/intel: Support RDPMC metrics clear mode
uprobes: Guard against kmemdup() failing in dup_return_instance()
perf/x86: Relax privilege filter restriction on AMD IBS
perf/core: Export perf_exclude_event()
uprobes: Reuse return_instances between multiple uretprobes within task
uprobes: Ensure return_instance is detached from the list before freeing
uprobes: Decouple return_instance list traversal and freeing
uprobes: Simplify session consumer tracking
uprobes: add speculative lockless VMA-to-inode-to-uprobe resolution
uprobes: simplify find_active_uprobe_rcu() VMA checks
mm: introduce mmap_lock_speculate_{try_begin|retry}
mm: convert mm_lock_seq to a proper seqcount
mm/gup: Use raw_seqcount_try_begin()
seqlock: add raw_seqcount_try_begin
perf/x86/rapl: Add core energy counter support for AMD CPUs
perf/x86/rapl: Move the cntr_mask to rapl_pmus struct
perf/x86/rapl: Remove the global variable rapl_msrs
...
|
|
We are adjusting struct page to make it smaller, removing unneeded fields
which correctly belong to struct folio.
Two of those fields are page->index and page->mapping. Perf is currently
making use of both of these. This is unnecessary. This patch eliminates
this.
Perf establishes its own internally controlled memory-mapped pages using
vm_ops hooks. The first page in the mapping is the read/write user control
page, and the rest of the mapping consists of read-only pages.
The VMA is backed by kernel memory either from the buddy allocator or
vmalloc depending on configuration. It is intended to be mapped read/write,
but because it has a page_mkwrite() hook, vma_wants_writenotify() indicates
that it should be mapped read-only.
When a write fault occurs, the provided page_mkwrite() hook,
perf_mmap_fault() (doing double duty handing faults as well) uses the
vmf->pgoff field to determine if this is the first page, allowing for the
desired read/write first page, read-only rest mapping.
For this to work the implementation has to carefully work around faulting
logic. When a page is write-faulted, the fault() hook is called first, then
its page_mkwrite() hook is called (to allow for dirty tracking in file
systems).
On fault we set the folio's mapping in perf_mmap_fault(), this is because
when do_page_mkwrite() is subsequently invoked, it treats a missing mapping
as an indicator that the fault should be retried.
We also set the folio's index so, given the folio is being treated as faux
user memory, it correctly references its offset within the VMA.
This explains why the mapping and index fields are used - but it's not
necessary.
We preallocate pages when perf_mmap() is called for the first time via
rb_alloc(), and further allocate auxiliary pages via rb_aux_alloc() as
needed if the mapping requires it.
This allocation is done in the f_ops->mmap() hook provided in perf_mmap(),
and so we can instead simply map all the memory right away here - there's
no point in handling (read) page faults when we don't demand page nor need
to be notified about them (perf does not).
This patch therefore changes this logic to map everything when the mmap()
hook is called, establishing a PFN map. It implements vm_ops->pfn_mkwrite()
to provide the required read/write vs. read-only behaviour, which does not
require the previously implemented workarounds.
While it is not ideal to use a VM_PFNMAP here, doing anything else will
result in the page_mkwrite() hook need to be provided, which requires the
same page->mapping hack this patch seeks to undo.
It will also result in the pages being treated as folios and placed on the
rmap, which really does not make sense for these mappings.
Semantically it makes sense to establish this as some kind of special
mapping, as the pages are managed by perf and are not strictly user pages,
but currently the only means by which we can do so functionally while
maintaining the required R/W and R/O behaviour is a PFN map.
There should be no change to actual functionality as a result of this
change.
Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20250103153151.124163-1-lorenzo.stoakes@oracle.com
|
|
Max Makarov reported kernel panic [1] in perf user callchain code.
The reason for that is the race between uprobe_free_utask and bpf
profiler code doing the perf user stack unwind and is triggered
within uprobe_free_utask function:
- after current->utask is freed and
- before current->utask is set to NULL
general protection fault, probably for non-canonical address 0x9e759c37ee555c76: 0000 [#1] SMP PTI
RIP: 0010:is_uprobe_at_func_entry+0x28/0x80
...
? die_addr+0x36/0x90
? exc_general_protection+0x217/0x420
? asm_exc_general_protection+0x26/0x30
? is_uprobe_at_func_entry+0x28/0x80
perf_callchain_user+0x20a/0x360
get_perf_callchain+0x147/0x1d0
bpf_get_stackid+0x60/0x90
bpf_prog_9aac297fb833e2f5_do_perf_event+0x434/0x53b
? __smp_call_single_queue+0xad/0x120
bpf_overflow_handler+0x75/0x110
...
asm_sysvec_apic_timer_interrupt+0x1a/0x20
RIP: 0010:__kmem_cache_free+0x1cb/0x350
...
? uprobe_free_utask+0x62/0x80
? acct_collect+0x4c/0x220
uprobe_free_utask+0x62/0x80
mm_release+0x12/0xb0
do_exit+0x26b/0xaa0
__x64_sys_exit+0x1b/0x20
do_syscall_64+0x5a/0x80
It can be easily reproduced by running following commands in
separate terminals:
# while :; do bpftrace -e 'uprobe:/bin/ls:_start { printf("hit\n"); }' -c ls; done
# bpftrace -e 'profile:hz:100000 { @[ustack()] = count(); }'
Fixing this by making sure current->utask pointer is set to NULL
before we start to release the utask object.
[1] https://github.com/grafana/pyroscope/issues/3673
Fixes: cfa7f3d2c526 ("perf,x86: avoid missing caller address in stack traces captured in uprobe")
Reported-by: Max Makarov <maxpain@linux.com>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/r/20250109141440.2692173-1-jolsa@kernel.org
|
|
If kmemdup() failed to alloc memory, don't proceed with extra_consumers
copy.
Fixes: e62f2d492728 ("uprobes: Simplify session consumer tracking")
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20241206183436.968068-1-andrii@kernel.org
|
|
While at it, rename the same function in s390 cpum_sf PMU.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Tested-by: Ravi Bangoria <ravi.bangoria@amd.com>
Reviewed-by: Ravi Bangoria <ravi.bangoria@amd.com>
Acked-by: Thomas Richter <tmricht@linux.ibm.com>
Link: https://lore.kernel.org/r/20241203180441.1634709-2-namhyung@kernel.org
|
|
Instead of constantly allocating and freeing very short-lived
struct return_instance, reuse it as much as possible within current
task. For that, store a linked list of reusable return_instances within
current->utask.
The only complication is that ri_timer() might be still processing such
return_instance. And so while the main uretprobe processing logic might
be already done with return_instance and would be OK to immediately
reuse it for the next uretprobe instance, it's not correct to
unconditionally reuse it just like that.
Instead we make sure that ri_timer() can't possibly be processing it by
using seqcount_t, with ri_timer() being "a writer", while
free_ret_instance() being "a reader". If, after we unlink return
instance from utask->return_instances list, we know that ri_timer()
hasn't gotten to processing utask->return_instances yet, then we can be
sure that immediate return_instance reuse is OK, and so we put it
onto utask->ri_pool for future (potentially, almost immediate) reuse.
This change shows improvements both in single CPU performance (by
avoiding relatively expensive kmalloc/free combon) and in terms of
multi-CPU scalability, where you can see that per-CPU throughput doesn't
decline as steeply with increased number of CPUs (which were previously
attributed to kmalloc()/free() through profiling):
BASELINE (latest perf/core)
===========================
uretprobe-nop ( 1 cpus): 1.898 ± 0.002M/s ( 1.898M/s/cpu)
uretprobe-nop ( 2 cpus): 3.574 ± 0.011M/s ( 1.787M/s/cpu)
uretprobe-nop ( 3 cpus): 5.279 ± 0.066M/s ( 1.760M/s/cpu)
uretprobe-nop ( 4 cpus): 6.824 ± 0.047M/s ( 1.706M/s/cpu)
uretprobe-nop ( 5 cpus): 8.339 ± 0.060M/s ( 1.668M/s/cpu)
uretprobe-nop ( 6 cpus): 9.812 ± 0.047M/s ( 1.635M/s/cpu)
uretprobe-nop ( 7 cpus): 11.030 ± 0.048M/s ( 1.576M/s/cpu)
uretprobe-nop ( 8 cpus): 12.453 ± 0.126M/s ( 1.557M/s/cpu)
uretprobe-nop (10 cpus): 14.838 ± 0.044M/s ( 1.484M/s/cpu)
uretprobe-nop (12 cpus): 17.092 ± 0.115M/s ( 1.424M/s/cpu)
uretprobe-nop (14 cpus): 19.576 ± 0.022M/s ( 1.398M/s/cpu)
uretprobe-nop (16 cpus): 22.264 ± 0.015M/s ( 1.391M/s/cpu)
uretprobe-nop (24 cpus): 33.534 ± 0.078M/s ( 1.397M/s/cpu)
uretprobe-nop (32 cpus): 43.262 ± 0.127M/s ( 1.352M/s/cpu)
uretprobe-nop (40 cpus): 53.252 ± 0.080M/s ( 1.331M/s/cpu)
uretprobe-nop (48 cpus): 55.778 ± 0.045M/s ( 1.162M/s/cpu)
uretprobe-nop (56 cpus): 56.850 ± 0.227M/s ( 1.015M/s/cpu)
uretprobe-nop (64 cpus): 62.005 ± 0.077M/s ( 0.969M/s/cpu)
uretprobe-nop (72 cpus): 66.445 ± 0.236M/s ( 0.923M/s/cpu)
uretprobe-nop (80 cpus): 68.353 ± 0.180M/s ( 0.854M/s/cpu)
THIS PATCHSET (on top of latest perf/core)
==========================================
uretprobe-nop ( 1 cpus): 2.253 ± 0.004M/s ( 2.253M/s/cpu)
uretprobe-nop ( 2 cpus): 4.281 ± 0.003M/s ( 2.140M/s/cpu)
uretprobe-nop ( 3 cpus): 6.389 ± 0.027M/s ( 2.130M/s/cpu)
uretprobe-nop ( 4 cpus): 8.328 ± 0.005M/s ( 2.082M/s/cpu)
uretprobe-nop ( 5 cpus): 10.353 ± 0.001M/s ( 2.071M/s/cpu)
uretprobe-nop ( 6 cpus): 12.513 ± 0.010M/s ( 2.086M/s/cpu)
uretprobe-nop ( 7 cpus): 14.525 ± 0.017M/s ( 2.075M/s/cpu)
uretprobe-nop ( 8 cpus): 15.633 ± 0.013M/s ( 1.954M/s/cpu)
uretprobe-nop (10 cpus): 19.532 ± 0.011M/s ( 1.953M/s/cpu)
uretprobe-nop (12 cpus): 21.405 ± 0.009M/s ( 1.784M/s/cpu)
uretprobe-nop (14 cpus): 24.857 ± 0.020M/s ( 1.776M/s/cpu)
uretprobe-nop (16 cpus): 26.466 ± 0.018M/s ( 1.654M/s/cpu)
uretprobe-nop (24 cpus): 40.513 ± 0.222M/s ( 1.688M/s/cpu)
uretprobe-nop (32 cpus): 54.180 ± 0.074M/s ( 1.693M/s/cpu)
uretprobe-nop (40 cpus): 66.100 ± 0.082M/s ( 1.652M/s/cpu)
uretprobe-nop (48 cpus): 70.544 ± 0.068M/s ( 1.470M/s/cpu)
uretprobe-nop (56 cpus): 74.494 ± 0.055M/s ( 1.330M/s/cpu)
uretprobe-nop (64 cpus): 79.317 ± 0.029M/s ( 1.239M/s/cpu)
uretprobe-nop (72 cpus): 84.875 ± 0.020M/s ( 1.179M/s/cpu)
uretprobe-nop (80 cpus): 92.318 ± 0.224M/s ( 1.154M/s/cpu)
For reference, with uprobe-nop we hit the following throughput:
uprobe-nop (80 cpus): 143.485 ± 0.035M/s ( 1.794M/s/cpu)
So now uretprobe stays a bit closer to that performance.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Link: https://lore.kernel.org/r/20241206002417.3295533-5-andrii@kernel.org
|
|
Ensure that by the time we call free_ret_instance() to clean up an
instance of struct return_instance it isn't reachable from
utask->return_instances anymore.
free_ret_instance() is called in a few different situations, all but one
of which already are fine w.r.t. return_instance visibility:
- uprobe_free_utask() guarantees that ri_timer() won't be called
(through timer_delete_sync() call), and so there is no need to
unlink anything, because entire utask is being freed;
- uprobe_handle_trampoline() is already unlinking to-be-freed
return_instance with rcu_assign_pointer() before calling
free_ret_instance().
Only cleanup_return_instances() violates this property, which so far is
not causing problems due to RCU-delayed freeing of return_instance,
which we'll change in the next patch. So make sure we unlink
return_instance before passing it into free_ret_instance(), as otherwise
reuse will be unsafe.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Link: https://lore.kernel.org/r/20241206002417.3295533-4-andrii@kernel.org
|
|
free_ret_instance() has two unrelated responsibilities: actually
cleaning up return_instance's resources and freeing memory, and also
helping with utask->return_instances list traversal by returning the
next alive pointer.
There is no reason why these two aspects have to be mixed together, so
turn free_ret_instance() into void-returning function and make callers
do list traversal on their own.
We'll use this simplification in the next patch that will guarantee that
to-be-freed return_instance isn't reachable from utask->return_instances
list.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Link: https://lore.kernel.org/r/20241206002417.3295533-3-andrii@kernel.org
|
|
In practice, each return_instance will typically contain either zero or
one return_consumer, depending on whether it has any uprobe session
consumer attached or not. It's highly unlikely that more than one uprobe
session consumers will be attached to any given uprobe, so there is no
need to optimize for that case. But the way we currently do memory
allocation and accounting is by pre-allocating the space for 4 session
consumers in contiguous block of memory next to struct return_instance
fixed part. This is unnecessarily wasteful.
This patch changes this to keep struct return_instance fixed-sized with one
pre-allocated return_consumer, while (in a highly unlikely scenario)
allowing for more session consumers in a separate dynamically
allocated and reallocated array.
We also simplify accounting a bit by not maintaining a separate
temporary capacity for consumers array, and, instead, relying on
krealloc() to be a no-op if underlying memory can accommodate a slightly
bigger allocation (but again, it's very uncommon scenario to even have
to do this reallocation).
All this gets rid of ri_size(), simplifies push_consumer() and removes
confusing ri->consumers_cnt re-assignment, while containing this
singular preallocated consumer logic contained within a few simple
preexisting helpers.
Having fixed-sized struct return_instance simplifies and speeds up
return_instance reuse that we ultimately add later in this patch set,
see follow up patches.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Link: https://lore.kernel.org/r/20241206002417.3295533-2-andrii@kernel.org
|
|
Given filp_cachep is marked SLAB_TYPESAFE_BY_RCU (and FMODE_BACKING
files, a special case, now goes through RCU-delated freeing), we can
safely access vma->vm_file->f_inode field locklessly under just
rcu_read_lock() protection, which enables looking up uprobe from
uprobes_tree completely locklessly and speculatively without the need to
acquire mmap_lock for reads. In most cases, anyway, assuming that there
are no parallel mm and/or VMA modifications. The underlying struct
file's memory won't go away from under us (even if struct file can be
reused in the meantime).
We rely on newly added mmap_lock_speculate_{try_begin,retry}() helpers to
validate that mm_struct stays intact for entire duration of this
speculation. If not, we fall back to mmap_lock-protected lookup.
The speculative logic is written in such a way that it will safely
handle any garbage values that might be read from vma or file structs.
Benchmarking results speak for themselves.
BEFORE (latest tip/perf/core)
=============================
uprobe-nop ( 1 cpus): 3.384 ± 0.004M/s ( 3.384M/s/cpu)
uprobe-nop ( 2 cpus): 5.456 ± 0.005M/s ( 2.728M/s/cpu)
uprobe-nop ( 3 cpus): 7.863 ± 0.015M/s ( 2.621M/s/cpu)
uprobe-nop ( 4 cpus): 9.442 ± 0.008M/s ( 2.360M/s/cpu)
uprobe-nop ( 5 cpus): 11.036 ± 0.013M/s ( 2.207M/s/cpu)
uprobe-nop ( 6 cpus): 10.884 ± 0.019M/s ( 1.814M/s/cpu)
uprobe-nop ( 7 cpus): 7.897 ± 0.145M/s ( 1.128M/s/cpu)
uprobe-nop ( 8 cpus): 10.021 ± 0.128M/s ( 1.253M/s/cpu)
uprobe-nop (10 cpus): 9.932 ± 0.170M/s ( 0.993M/s/cpu)
uprobe-nop (12 cpus): 8.369 ± 0.056M/s ( 0.697M/s/cpu)
uprobe-nop (14 cpus): 8.678 ± 0.017M/s ( 0.620M/s/cpu)
uprobe-nop (16 cpus): 7.392 ± 0.003M/s ( 0.462M/s/cpu)
uprobe-nop (24 cpus): 5.326 ± 0.178M/s ( 0.222M/s/cpu)
uprobe-nop (32 cpus): 5.426 ± 0.059M/s ( 0.170M/s/cpu)
uprobe-nop (40 cpus): 5.262 ± 0.070M/s ( 0.132M/s/cpu)
uprobe-nop (48 cpus): 6.121 ± 0.010M/s ( 0.128M/s/cpu)
uprobe-nop (56 cpus): 6.252 ± 0.035M/s ( 0.112M/s/cpu)
uprobe-nop (64 cpus): 7.644 ± 0.023M/s ( 0.119M/s/cpu)
uprobe-nop (72 cpus): 7.781 ± 0.001M/s ( 0.108M/s/cpu)
uprobe-nop (80 cpus): 8.992 ± 0.048M/s ( 0.112M/s/cpu)
AFTER
=====
uprobe-nop ( 1 cpus): 3.534 ± 0.033M/s ( 3.534M/s/cpu)
uprobe-nop ( 2 cpus): 6.701 ± 0.007M/s ( 3.351M/s/cpu)
uprobe-nop ( 3 cpus): 10.031 ± 0.007M/s ( 3.344M/s/cpu)
uprobe-nop ( 4 cpus): 13.003 ± 0.012M/s ( 3.251M/s/cpu)
uprobe-nop ( 5 cpus): 16.274 ± 0.006M/s ( 3.255M/s/cpu)
uprobe-nop ( 6 cpus): 19.563 ± 0.024M/s ( 3.261M/s/cpu)
uprobe-nop ( 7 cpus): 22.696 ± 0.054M/s ( 3.242M/s/cpu)
uprobe-nop ( 8 cpus): 24.534 ± 0.010M/s ( 3.067M/s/cpu)
uprobe-nop (10 cpus): 30.475 ± 0.117M/s ( 3.047M/s/cpu)
uprobe-nop (12 cpus): 33.371 ± 0.017M/s ( 2.781M/s/cpu)
uprobe-nop (14 cpus): 38.864 ± 0.004M/s ( 2.776M/s/cpu)
uprobe-nop (16 cpus): 41.476 ± 0.020M/s ( 2.592M/s/cpu)
uprobe-nop (24 cpus): 64.696 ± 0.021M/s ( 2.696M/s/cpu)
uprobe-nop (32 cpus): 85.054 ± 0.027M/s ( 2.658M/s/cpu)
uprobe-nop (40 cpus): 101.979 ± 0.032M/s ( 2.549M/s/cpu)
uprobe-nop (48 cpus): 110.518 ± 0.056M/s ( 2.302M/s/cpu)
uprobe-nop (56 cpus): 117.737 ± 0.020M/s ( 2.102M/s/cpu)
uprobe-nop (64 cpus): 124.613 ± 0.079M/s ( 1.947M/s/cpu)
uprobe-nop (72 cpus): 133.239 ± 0.032M/s ( 1.851M/s/cpu)
uprobe-nop (80 cpus): 142.037 ± 0.138M/s ( 1.775M/s/cpu)
Previously total throughput was maxing out at 11mln/s, and gradually
declining past 8 cores. With this change, it now keeps growing with each
added CPU, reaching 142mln/s at 80 CPUs (this was measured on a 80-core
Intel(R) Xeon(R) Gold 6138 CPU @ 2.00GHz).
Suggested-by: Matthew Wilcox <willy@infradead.org>
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Link: https://lkml.kernel.org/r/20241122035922.3321100-3-andrii@kernel.org
|
|
At the point where find_active_uprobe_rcu() is used we know that VMA in
question has triggered software breakpoint, so we don't need to validate
vma->vm_flags. Keep only vma->vm_file NULL check.
Suggested-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Link: https://lkml.kernel.org/r/20241122035922.3321100-2-andrii@kernel.org
|
|
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
