From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.remote-ip=109.224.244.22 ARC-Seal: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; cv=none; t=1793343576; b=PlwI+czPrfJ7/zmXfpo59OSrP7eeVk73IWOudpf7KAdkH1PFXmZ0Hgxb2OZpovbmfJFc 1OHdnumf3oqvfWkyzNTnJV7vbpaaFjBk7lG7lsMHiOAL90H6sBMge+8l8TogpmqiJuvIR oljEcsBqhDRJWW+HsEPyU1SBP4F9bYFObVeoBUWtE93d5lU3o27AaNLrZa2dVbNXwe8fM goT51nuixcbYHBtXzgnTYPlbV9VOQap46vOsnAhSytI8G3BnXNgLnwOhy4jczVQMig6BR O57QlCOKn5FpPt32amopR+IYSkw6sYFhoF0+bEumkl4EyWJOlRlJSbuKhMNmE3e1Szg== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1793343576; h=DMARC-Filter:DKIM-Signature:Date:To:From:Cc:Subject:Message-ID: Feedback-ID:X-Pm-Message-ID:MIME-Version:Content-Type: Content-Transfer-Encoding; bh=Cs9D0HY32Cso7YnxVSoyyLLlor/aF98StXHYbAJIbtM=; b=yOXSvAwW5PtmNiJqArF4yYthw2/6BYj31v0T4SPSiqvm68dC2x4hrZc9IjFHFcT7e/ch ar0s5Rj+wpHNsC68Ri1qsyHGzcChl3GndodPXEEgakk5wxT+sYY7IWylAt8fF1XlN35p7 f1lF6cMd7k0RDP/Lqaf3wRUyAHmxDKxFA5CKeB4CcrobNgqPOYiMU9FfffuDbmVPnmki3 hZpsBXmbx7yrSe2OydWLxcamEUAp1+UX8kzij05Z/ZqTiolw1KJyyhMoAUpr+0Ye3xyCG wUsvVDhR0mWfi/1lLndN5e6QNi8Nnhot4dxLuUmbAtvVQAocLsBBj22Cka7rdziihZA== ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass header.from=1g4.org; spf=pass smtp.mailfrom=1g4.org; dkim=pass header.d=1g4.org header.i=@1g4.org header.a=rsa-sha256 header.s=protonmail header.b="rQ/lO6+M"; arc=none smtp.remote-ip=109.224.244.22 Received: from mail-24422.protonmail.ch (mail-24422.protonmail.ch [109.224.244.22]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (prime256v1) server-digest SHA256) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 85CFB606A31C; Fri, 30 Oct 2026 06:59:35 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.2 smtp.subspace.kernel.org 85CFB606A31C Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=quarantine dis=none) header.from=1g4.org Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=1g4.org Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key, unprotected) header.d=1g4.org header.i=@1g4.org header.a=rsa-sha256 header.s=protonmail header.b=rQ/lO6+M DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1g4.org; s=protonmail; t=1793343573; x=1793602773; bh=Cs9D0HY32Cso7YnxVSoyyLLlor/aF98StXHYbAJIbtM=; h=Date:To:From:Cc:Subject:Message-ID:Feedback-ID:From:To:Cc:Date: Subject:Reply-To:Feedback-ID:Message-ID:BIMI-Selector; b=rQ/lO6+MM6O0c9Lri28+sr03loz1eKU8Moh/sR8lKA2yMySoq1XuhynlwKOUVJPIq 9+rEIgCWiQGQY+j57N8lvhdif2jGUOZDMQM9NWaOnjdady6levdVLKFmSzzbMl3/Jj 3Uxm+nXIUoQadHtmXNQT982D6CPLQozBUqtdhZt7aClt1OMaFng2AnextvBcWewKPl p/dHZf8rq55/+h93GJwO3NNH1tPay13VmjjuKwB/iGwMbMxvaMW5J/Yl8YeYndu4ml 7lPB0ECAGEyDgERnN8/zkjQHMJ78i9dqNxschgO/JJxSQPY6DfxaRCiL8XDr/gkH4m ImYNbJwlh5qVw== Date: Fri, 30 Oct 2026 06:59:28 +0000 To: acme@kernel.org, namhyung@kernel.org From: Paul Cc: peterz@infradead.org, mingo@redhat.com, mark.rutland@arm.com, alexander.shishkin@linux.intel.com, jolsa@kernel.org, irogers@google.com, adrian.hunter@intel.com, james.clark@linaro.org, alexey.budankov@linux.intel.com, linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, Paul Moses Subject: [PATCH] perf record: Skip affinity for per-thread mmap buffers Message-ID: <20261030065857.2157899-1-p@1g4.org> Feedback-ID: 8253658:user:proton X-Pm-Message-ID: 30567b2a5a2c3d284c4b55dfa40f2fb37edd2741 X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Post: List-Help: List-Subscribe: List-Unsubscribe: List-Owner: Precedence: list MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable From: Paul Moses Per-thread mmap buffers use CPU -1 to mean any CPU. Combining --per-thread with --affinity=3Dcpu passes this value to __set_bit() when constructing th= e reader affinity mask, causing an out-of-bounds bitmap access and SIGSEGV. Program received signal SIGSEGV, Segmentation fault. 0x0000555555f3488e in mmap.mmap () #0 0x0000555555f3488e in mmap.mmap () #1 0x0000555555e2f7bf in mmap_per_evsel (evlist=3Devlist@entry=3D0x7fffedc= c1150, ops=3Dops@entry=3D0x7fffffffaf50, idx=3Didx@entry=3D0, mp=3Dmp@entry= =3D0x7fffffffaf70, cpu_idx=3Dcpu_idx@entry=3D0, thread=3Dthread@entry=3D0, = _output=3D0x7fffffffaf08, _output_overwrite=3D0x7fffffffaf04, nr_mmaps=3D0x= 7fffffffaf0c) at evlist.c:598 #2 0x0000555555e2ef96 in mmap_per_thread (evlist=3D0x7fffedcc1150, ops=3D0= x7fffffffaf50, mp=3D0x7fffffffaf70) at evlist.c:650 #3 perf_evlist__mmap_ops (evlist=3D0x7fffedcc1150, ops=3D0x7fffffffaf50, m= p=3D0x7fffffffaf70) at evlist.c:748 #4 0x0000555555f2150d in evlist.mmap_ex () #5 0x0000555555db0b0a in record.open () #6 0x0000555555dadea7 in __cmd_record () #7 0x0000555555daca89 in cmd_record () #8 0x0000555555e27d53 in run_builtin () #9 0x0000555555e277cf in main () rip 0x555555f3488e 0x555555f3488e rax 0x7fffef9e9bd0 140737213537232 rbx 0x7fffedd249a0 140737183369632 rcx 0x3ffffffffffffff 288230376151711743 rdx 0x8000000000000000 -9223372036854775808 rsi 0x0 0 rdi 0x7fffef9e9bd0 140737213537232 0x555555f34876 :=09movswq 0x10(%rbx),%rcx 0x555555f3487b :=09mov 0xb8(%rbx),%rax 0x555555f34882 :=09mov $0x1,%edx 0x555555f34887 :=09shl %cl,%rdx 0x555555f3488a :=09shr $0x6,%rcx =3D> 0x555555f3488e :=09or %rdx,(%rax,%rcx,8) 0x555555f34892 :=09lea 0x17711b(%rip),%r14 # 0= x5555560ab9b4 0x555555f34899 :=09cmpl $0x2,(%r14) 0x555555f3489d :=09jne 0x555555f348f7 0x555555f3489f :=09mov 0xb8(%rbx),%rdi 0x555555f348a6 :=09mov 0xc0(%rbx),%esi 0x555555f348ac :=09lea -0x430(%rbp),%r12 Only construct and apply reader affinity masks for buffers with a nonnegative CPU. Skip the reader-side bitmap operations as well, since per-thread buffers no longer allocate an affinity mask. Fixes: 8384a2600c7d ("perf record: Adapt affinity to machines with #CPUs > = 1K") Signed-off-by: Paul Moses --- tools/perf/builtin-record.c | 2 +- tools/perf/util/mmap.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index a57987851cf0..d7c8f490cb62 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -1565,7 +1565,7 @@ static struct perf_event_header finished_init_event = =3D { =20 static void record__adjust_affinity(struct record *rec, struct mmap *map) { -=09if (rec->opts.affinity !=3D PERF_AFFINITY_SYS && +=09if (map->core.cpu.cpu >=3D 0 && rec->opts.affinity !=3D PERF_AFFINITY_S= YS && =09 !bitmap_equal(thread->mask->affinity.bits, map->affinity_mask.bits, =09=09=09 thread->mask->affinity.nbits)) { =09=09bitmap_zero(thread->mask->affinity.bits, thread->mask->affinity.nbit= s); diff --git a/tools/perf/util/mmap.c b/tools/perf/util/mmap.c index 358e70c4f3ed..be3dffdaede8 100644 --- a/tools/perf/util/mmap.c +++ b/tools/perf/util/mmap.c @@ -292,8 +292,8 @@ int mmap__mmap(struct mmap *map, struct mmap_params *mp= , int fd, struct perf_cpu =09=09return -1; =09} =20 -=09if (mp->affinity !=3D PERF_AFFINITY_SYS && -=09=09perf_mmap__setup_affinity_mask(map, mp)) { +=09if (cpu.cpu >=3D 0 && mp->affinity !=3D PERF_AFFINITY_SYS && +=09 perf_mmap__setup_affinity_mask(map, mp)) { =09=09pr_debug2("failed to alloc mmap affinity mask, error %d\n", =09=09=09 errno); =09=09return -1; --=20 2.56.GIT