From: Tanushree Shah <tshah@linux.ibm.com>
To: Matt Turner <mattst88@gmail.com>,
Madhavan Srinivasan <maddy@linux.ibm.com>,
Michael Ellerman <mpe@ellerman.id.au>,
Nicholas Piggin <npiggin@gmail.com>,
"Christophe Leroy (CS GROUP)" <chleroy@kernel.org>,
"Ritesh Harjani (IBM)" <ritesh.list@gmail.com>,
Shrikanth Hegde <sshegde@linux.ibm.com>,
Paul Mackerras <paulus@ozlabs.org>
Cc: linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3 7/8] powerpc/perf: Use SIAR for kernel sample addresses on 32-bit
Date: Thu, 19 Nov 2026 12:58:05 +0530 [thread overview]
Message-ID: <cf0df071-7e06-4da4-bc08-d0781c5aebcb@linux.ibm.com> (raw)
In-Reply-To: <20261118161706.2454031-1-mattst88@gmail.com>
On 11/18/26 9:47 PM, Matt Turner wrote:
> On 32-bit, perf_read_regs() sets regs->result to 0, so regs_use_siar()
> is always false and every sample is attributed to the interrupted PC.
>
> The MPC7450 family's SIAR holds the address of the last instruction to
> complete before the performance monitor exception is generated
> (MPC7450UM Rev. 5, section 11.3.7). The exception can be generated
> with MSR[EE] = 0 and is taken once EE is set again (section 11.2), so
> when interrupts are disabled the interrupted PC is wherever they are
> re-enabled, while SIAR still points into the disabled region.
>
> On a 7448, sampling cycles system-wide with --intr-regs=nip during dd,
> perf bench sched pipe and find, 46% of the kernel samples have SIAR and
> the interrupted PC in different functions. 13270 samples have their PC
> on the instruction after the mtmsr that restores interrupts in
> ata_scsi_queuecmd(), and 11661 of those have SIAR in udelay().
>
> With interrupts enabled SIAR helps less. Sampling a 32-instruction
> loop with one store and one branch with -c 10007, 99.8% of the samples
> land at:
>
> event interrupted PC SIAR
> r122 (branches) br + 13 br + 9
> r014 (stores) stw + 21 stw + 17
>
> Because the exception is held off, it can also be taken on the other
> side of a user/kernel transition from the instruction in SIAR, e.g.
> just after the rfi at the end of a system call. Recording a loop that
> alternates user work with getppid() with perf record -g -c 1000000,
> 3022 of 16953 samples had SIAR in a different mode from the
> interrupted PC. The callchain is walked from regs, so taking its first
> entry from SIAR there starts a user callchain with a kernel address.
>
> SIAR also holds a real address while translation is off, and a real
> address looks like a user address. The hash_page() fast path on
> CONFIG_PPC_BOOK3S_604 runs with translation off and EE clear and
> returns straight to user space, so an exception generated there is
> taken at a user PC with SIAR holding a kernel real address. Reading
> one word from each of 2048 pages after an mprotect() round trip has
> flushed their hash PTEs, with -c 10007, 1674 of 2659 user samples land
> on the real addresses of DataAccess, hash_page() and
> fast_hash_page_return when SIAR is used whenever its address matches
> the mode of the interrupted context.
>
> A user SIAR cannot be told apart from a kernel real address, and the
> table above shows user samples gain little from SIAR. Use SIAR for PMU
> exceptions only when the interrupted context is the kernel and SIAR is
> a kernel address, and the interrupted PC otherwise. With that, neither
> test has user samples at kernel addresses or callchains with mixed
> modes, and 99.9% of the kernel samples in the mprotect() test still
> take their address from SIAR.
>
> Suggested-by: Tanushree Shah <tshah@linux.ibm.com>
> Signed-off-by: Matt Turner <mattst88@gmail.com>
Hi Matt, v3 looks good for this patch, and the hash_page() explanation
in the message and comment is clear.
Reviewed-by: Tanushree Shah <tshah@linux.ibm.com>
Thanks
Tanushree Shah
> ---
> Changes in v3:
> - Use SIAR only for kernel samples, and only when it is a kernel
> address. A real-mode SIAR from the hash_page() fast path looks like a
> user address and was used for user samples taken after its rfi
> (Tanushree Shah).
> - Retitled from "Use SIAR for sample addresses on 32-bit".
> - Link to v2: https://patch.msgid.link/20261117-ppc32-perf-v2-7-2072e9b90009@gmail.com
>
> arch/powerpc/perf/core-book3s.c | 21 ++++++++++++++++++++-
> 1 file changed, 20 insertions(+), 1 deletion(-)
>
> diff --git a/arch/powerpc/perf/core-book3s.c b/arch/powerpc/perf/core-book3s.c
> index 10ded865583c..aab18f4dda94 100644
> --- a/arch/powerpc/perf/core-book3s.c
> +++ b/arch/powerpc/perf/core-book3s.c
> @@ -112,9 +112,28 @@ static inline u32 perf_get_misc_flags(struct pt_regs *regs)
> {
> return 0;
> }
> +
> +/*
> + * The MPC7450 family's SIAR holds the address of the last instruction to
> + * complete before the performance monitor exception was generated, so for
> + * a PMU exception it is closer to the event than the interrupted PC. The
> + * exception is held off while MSR[EE] is clear, so with interrupts
> + * disabled the interrupted PC is wherever they are re-enabled.
> + *
> + * Because it is held off, the exception can also be taken on the other
> + * side of a user/kernel transition from the instruction in SIAR, e.g.
> + * just after the rfi at the end of a system call. SIAR also holds a real
> + * address for code running with translation off, and a real address
> + * looks like a user address. The hash_page() fast path runs with
> + * translation off and returns straight to user space, so a user SIAR
> + * cannot be trusted. Use SIAR only for kernel samples, and only when it
> + * is a kernel address.
> + */
> static inline void perf_read_regs(struct pt_regs *regs)
> {
> - regs->result = 0;
> + unsigned long siar = mfspr(SPRN_SIAR);
> +
> + regs->result = siar && !user_mode(regs) && is_kernel_addr(siar);
> }
>
> static inline int siar_valid(struct pt_regs *regs)
next prev parent reply other threads:[~2026-11-19 7:28 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-17 17:43 [PATCH v2 0/8] powerpc/perf: mpc7450: Event fixes, SIAR sampling and sysfs event names Matt Turner
2026-11-17 17:43 ` [PATCH v2 1/8] powerpc/perf: mpc7450: Event 0x30d does not use the threshold Matt Turner
2026-11-17 17:43 ` [PATCH v2 2/8] powerpc/perf: mpc7450: Event 0x634 counts L2 prefetches, not L1I Matt Turner
2026-11-17 17:43 ` [PATCH v2 3/8] powerpc/perf: mpc7450: Remove the snoop retry alternatives Matt Turner
2026-11-17 17:43 ` [PATCH v2 4/8] powerpc/perf: mpc7450: Fill in the L1 data cache events Matt Turner
2026-11-17 17:43 ` [PATCH v2 5/8] powerpc/perf: mpc7450: Add the PMC6 L3 events to the alternatives table Matt Turner
2026-11-17 17:43 ` [PATCH v2 6/8] powerpc/perf: mpc7450: Export the event format in sysfs Matt Turner
2026-11-17 17:43 ` [PATCH v2 7/8] powerpc/perf: Use SIAR for sample addresses on 32-bit Matt Turner
[not found] ` <53a68daf-e694-4c88-b99e-fb8a4c17c1a7@linux.ibm.com>
2026-11-18 16:01 ` Matt Turner
2026-11-18 16:17 ` [PATCH v3 7/8] powerpc/perf: Use SIAR for kernel " Matt Turner
2026-11-19 7:28 ` Tanushree Shah [this message]
2026-11-17 17:43 ` [PATCH v2 8/8] powerpc/perf: mpc7450: Export event names in sysfs Matt Turner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=cf0df071-7e06-4da4-bc08-d0781c5aebcb@linux.ibm.com \
--to=tshah@linux.ibm.com \
--cc=chleroy@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=maddy@linux.ibm.com \
--cc=mattst88@gmail.com \
--cc=mpe@ellerman.id.au \
--cc=npiggin@gmail.com \
--cc=paulus@ozlabs.org \
--cc=ritesh.list@gmail.com \
--cc=sshegde@linux.ibm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®