Linux Perf Users
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Tengda Wu" <wutengda@huaweicloud.com>
Cc: linux-perf-users@vger.kernel.org
Subject: Re: [PATCH v6 25/26] perf annotate-arm64: Support per-cpu variable access tracking
Date: Wed, 16 Sep 2026 01:53:47 +0000	[thread overview]
Message-ID: <20260916015348.2114B1F000FF@smtp.kernel.org> (raw)
In-Reply-To: <735c23b4a55db2c67be4920792922550ff4f6409.1789521520.git.wutengda@huaweicloud.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] perf: retry mechanism fails for per-cpu variable loads
--

commit 0a8000b94df79930832d6fd4535fbe42f682b339
Author: Tengda Wu <wutengda@huaweicloud.com>

perf annotate-arm64: Support per-cpu variable access tracking

This commit extends update_insn_state_arm64() to handle per-cpu variable
addressing. It detects pairs resolving to the '__per_cpu_offset' symbol
and marks the destination register as TSR_KIND_PERCPU_BASE to propagate
type information during subsequent instructions.

> diff --git a/tools/perf/util/annotate-arch/annotate-arm64.c b/tools/perf/util/annotate-arch/annotate-arm64.c
> index b25a3c70675c1..f4b7b0eb44f42 100644
> --- a/tools/perf/util/annotate-arch/annotate-arm64.c
> +++ b/tools/perf/util/annotate-arch/annotate-arm64.c
[ ... ]
> @@ -666,6 +677,12 @@ static int propagate_load_reg_state(struct type_state *state,
>  	}
>  
>  load_non_regoff:
> +	/*
> +	 * If src_tsr->kind is TSR_KIND_PERCPU_BASE, treat it as an array.
> +	 * No special handling is needed here; just fall through to retry
> +	 * and resolve the type from the second register.
> +	 */
> +
>  	/* Try another register if any */
>  	if (src->multi_regs && src->reg1 != src->reg2 && sreg != src->reg2 &&
>  	    !(src->extend_type || src->shift_type)) {

[Severity: Medium]
Will this retry logic actually succeed for per-cpu variable loads?

Since both registers in the per-cpu load are non-constant
(TSR_KIND_PERCPU_BASE and TSR_KIND_GLOBAL_ADDR), arch_get_reg_offset()
returns -1 for both the initial attempt and the retry pass because it
requires the 'other' register in a multi-register operand to be
TSR_KIND_CONST.

Does this break the intended fallback logic here in
propagate_load_reg_state(), resulting in an unhandled -1 return and
silent type propagation failure?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/cover.1789521520.git.wutengda@huaweicloud.com?part=25

  reply	other threads:[~2026-09-16  1:53 UTC|newest]

Thread overview: 70+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-16  1:29 [PATCH v6 00/26] perf arm64: Support data type profiling Tengda Wu
2026-09-16  1:29 ` [PATCH v6 01/26] perf capstone: Symbolize address operands to match objdump on arm64 Tengda Wu
2026-09-16  1:37   ` sashiko-bot
2026-09-19  7:03   ` Namhyung Kim
2026-09-16  1:29 ` [PATCH v6 02/26] perf llvm: Fix arm64 adrp instruction disassembly mismatch with objdump Tengda Wu
2026-09-16  1:43   ` sashiko-bot
2026-09-16  2:51     ` Tengda Wu
2026-09-16  1:29 ` [PATCH v6 03/26] perf annotate-arm64: Generalize arm64_mov__parse to support more instructions Tengda Wu
2026-09-16  1:39   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 04/26] perf annotate-arm64: Handle load and store instructions Tengda Wu
2026-09-16  1:43   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 05/26] perf annotate: Normalize arch__dwarf_regnum() error return values Tengda Wu
2026-09-16  1:39   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 06/26] perf annotate: Introduce extract_op_location callback for arch-specific parsing Tengda Wu
2026-09-16  1:44   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 07/26] perf dwarf-regs: Adapt get_dwarf_regnum() for arm64 Tengda Wu
2026-09-16  1:43   ` sashiko-bot
2026-09-16  2:05     ` Tengda Wu
2026-09-16  1:29 ` [PATCH v6 08/26] perf annotate: Adapt arch__dwarf_regnum() " Tengda Wu
2026-09-16  1:43   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 09/26] perf annotate-arm64: Implement extract_op_location() callback Tengda Wu
2026-09-16  1:38   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 10/26] perf annotate: Default to --itrace=i1i for data type profiling Tengda Wu
2026-09-16  1:40   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 11/26] perf arm-spe: Set default synthesized event period to 1 Tengda Wu
2026-09-16  1:38   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 12/26] perf annotate-data: Extract invalidate_reg_state() as a common helper Tengda Wu
2026-09-16  1:37   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 13/26] perf annotate-arm64: Enable instruction tracking support Tengda Wu
2026-09-16  1:43   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 14/26] perf annotate-data: Add arch_get_reg_offset helper Tengda Wu
2026-09-16  1:39   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 15/26] perf annotate-arm64: Track return type after call instructions Tengda Wu
2026-09-16  1:37   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 16/26] perf annotate-arm64: Support load instruction tracking Tengda Wu
2026-09-16  1:51   ` sashiko-bot
2026-09-16 11:31     ` Tengda Wu
2026-09-16  1:29 ` [PATCH v6 17/26] perf annotate-arm64: Support store " Tengda Wu
2026-09-16  1:41   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 18/26] perf annotate-data: Expand type_state_reg imm_value to u64 Tengda Wu
2026-09-16  1:44   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 19/26] perf annotate-data: Track imm_value for stack variables Tengda Wu
2026-09-16  1:44   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 20/26] perf annotate-x86: Delete stale stack state on store of untracked register Tengda Wu
2026-09-16  1:43   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 21/26] perf annotate-arm64: Support stack variable tracking Tengda Wu
2026-09-16  1:49   ` sashiko-bot
2026-09-16 11:55     ` Tengda Wu
2026-09-20  1:43       ` Tengda Wu
2026-09-16  1:29 ` [PATCH v6 22/26] perf annotate-arm64: Support 'mov' instruction tracking Tengda Wu
2026-09-16  1:45   ` sashiko-bot
2026-09-16  1:29 ` [PATCH v6 23/26] perf annotate-arm64: Support 'add' " Tengda Wu
2026-09-16  1:44   ` sashiko-bot
2026-09-16  1:30 ` [PATCH v6 24/26] perf annotate-arm64: Support 'adrp' instruction to track global variables Tengda Wu
2026-09-16  1:48   ` sashiko-bot
2026-09-16  1:30 ` [PATCH v6 25/26] perf annotate-arm64: Support per-cpu variable access tracking Tengda Wu
2026-09-16  1:53   ` sashiko-bot [this message]
2026-09-17  2:16     ` Tengda Wu
2026-09-20  1:45       ` Tengda Wu
2026-09-20  1:52   ` Tengda Wu
2026-09-16  1:30 ` [PATCH v6 26/26] perf annotate-arm64: Support 'mrs' instruction to track 'current' pointer Tengda Wu
2026-09-16  1:52   ` sashiko-bot
2026-09-16  4:17 ` [PATCH v6 00/26] perf arm64: Support data type profiling Ian Rogers
2026-09-16 11:36   ` Tengda Wu
2026-09-19  7:10 ` Namhyung Kim
2026-09-19 11:22   ` Arnaldo Melo
2026-09-19 17:41     ` Namhyung Kim
2026-09-20  2:02   ` Tengda Wu
2026-09-21  6:47     ` Arnaldo Carvalho de Melo
2026-09-21  7:08       ` Tengda Wu

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=20260916015348.2114B1F000FF@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=sashiko-reviews@lists.linux.dev \
    --cc=wutengda@huaweicloud.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