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 v4 13/23] perf annotate-arm64: Track return type after call instructions
Date: Sat, 08 Aug 2026 13:05:02 +0000	[thread overview]
Message-ID: <20260808130503.AA8CA1F00A3A@smtp.kernel.org> (raw)
In-Reply-To: <20260808122400.2961238-14-wutengda@huaweicloud.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Missing invalidation of DWARF lifetime tracking when a caller-saved return register is updated with a function's return type.
--

--- Patch [13]: [PATCH v4 13/23] perf annotate-arm64: Track return type after call instructions ---
commit 001dcc403f6841490dc601cf7b406aa4f0f30321
Author: Tengda Wu <wutengda@huaweicloud.com>

perf annotate-arm64: Track return type after call instructions

This commit enables data type profiling to track function return values in
arm64 annotations by looking up the DWARF return type of called functions.

> diff --git a/tools/perf/util/annotate-arch/annotate-arm64.c b/tools/perf/util/annotate-arch/annotate-arm64.c
> index 44daa0176e512..acff14ca01e00 100644
> --- a/tools/perf/util/annotate-arch/annotate-arm64.c
> +++ b/tools/perf/util/annotate-arch/annotate-arm64.c

[ ... ]

> @@ -395,6 +397,20 @@ static void update_insn_state_arm64(struct type_state *state,
>  				continue;
>  			invalidate_reg_state(reg);
>  		}
> +
> +		/* Update register with the return type (if any) */
> +		if (call_name && die_find_func_rettype(cu_die, call_name, &type_die)) {
> +			tsr = &state->regs[state->ret_reg];
> +			tsr->copied_from = -1;
> +			tsr->type = type_die;
> +			tsr->kind = TSR_KIND_TYPE;
> +			tsr->offset = 0;
> +			tsr->ok = true;

[Severity: Medium]
Does this code need to clear the existing dynamic lifetime tracking fields
(such as lifetime_active and lifetime_end) before updating the return
register?

If the return register (e.g., x0) previously held a variable with an active
DWARF lifetime that spanned across this call, failing to call
invalidate_reg_state() or manually clear those fields could leave stale
lifetime flags in the register state.

If a subsequent function call occurs before that original lifetime ends, and
it does not return a known type, the return register might skip invalidation
due to the stale lifetime flag. This could erroneously preserve the first
call's return type and corrupt the data type profiling statistics.

> +
> +			pr_debug_dtp("call [%x] return -> reg%d",
> +				     insn_offset, state->ret_reg);
> +			pr_debug_type_name(&type_die, tsr->kind);
> +		}
>  		return;
>  	}

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260808122400.2961238-1-wutengda@huaweicloud.com?part=13

  reply	other threads:[~2026-08-08 13:05 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-08 12:23 [PATCH v4 00/23] perf arm64: Support data type profiling Tengda Wu
2026-08-08 12:23 ` [PATCH v4 01/23] perf capstone: Fix arm64 jump/adrp disassembly mismatch with objdump Tengda Wu
2026-08-10 13:08   ` Shuai Xue
2026-08-08 12:23 ` [PATCH v4 02/23] perf llvm: Fix arm64 adrp instruction " Tengda Wu
2026-08-08 13:03   ` sashiko-bot
2026-08-08 12:23 ` [PATCH v4 03/23] perf annotate-arm64: Generalize arm64_mov__parse to support more instructions Tengda Wu
2026-08-08 13:05   ` sashiko-bot
2026-08-08 12:23 ` [PATCH v4 04/23] perf annotate-arm64: Handle load and store instructions Tengda Wu
2026-08-08 13:07   ` sashiko-bot
2026-08-08 12:23 ` [PATCH v4 05/23] perf dwarf-regs: Adapt get_dwarf_regnum() for arm64 Tengda Wu
2026-08-08 13:12   ` sashiko-bot
2026-08-08 12:23 ` [PATCH v4 06/23] perf annotate: Adapt arch__dwarf_regnum() " Tengda Wu
2026-08-08 13:07   ` sashiko-bot
2026-08-08 12:23 ` [PATCH v4 07/23] perf annotate: Introduce extract_op_location callback for arch-specific parsing Tengda Wu
2026-08-08 13:11   ` sashiko-bot
2026-08-08 12:23 ` [PATCH v4 08/23] perf annotate-arm64: Implement extract_op_location() callback Tengda Wu
2026-08-08 12:23 ` [PATCH v4 09/23] perf annotate: Deduplicate overlapping ARM SPE events for data type profiling Tengda Wu
2026-08-10  6:57   ` Adrian Hunter
2026-08-08 12:23 ` [PATCH v4 10/23] perf arm-spe: Set default synthesized event period to 1 Tengda Wu
2026-08-08 12:23 ` [PATCH v4 11/23] perf annotate-data: Extract invalidate_reg_state() as a common helper Tengda Wu
2026-08-08 12:23 ` [PATCH v4 12/23] perf annotate-arm64: Enable instruction tracking support Tengda Wu
2026-08-08 13:22   ` sashiko-bot
2026-08-08 12:23 ` [PATCH v4 13/23] perf annotate-arm64: Track return type after call instructions Tengda Wu
2026-08-08 13:05   ` sashiko-bot [this message]
2026-08-08 12:23 ` [PATCH v4 14/23] perf annotate-arm64: Support load instruction tracking Tengda Wu
2026-08-08 13:08   ` sashiko-bot
2026-08-08 12:23 ` [PATCH v4 15/23] perf annotate-arm64: Support store " Tengda Wu
2026-08-08 13:11   ` sashiko-bot
2026-08-08 12:23 ` [PATCH v4 16/23] perf annotate-data: Expand type_state_reg imm_value to u64 Tengda Wu
2026-08-08 13:17   ` sashiko-bot
2026-08-08 12:23 ` [PATCH v4 17/23] perf annotate-data: Track imm_value for stack variables Tengda Wu
2026-08-08 12:23 ` [PATCH v4 18/23] perf annotate-arm64: Support stack variable tracking Tengda Wu
2026-08-08 13:25   ` sashiko-bot
2026-08-08 12:23 ` [PATCH v4 19/23] perf annotate-arm64: Support 'mov' instruction tracking Tengda Wu
2026-08-08 13:20   ` sashiko-bot
2026-08-08 12:23 ` [PATCH v4 20/23] perf annotate-arm64: Support 'add' " Tengda Wu
2026-08-08 13:14   ` sashiko-bot
2026-08-08 12:23 ` [PATCH v4 21/23] perf annotate-arm64: Support 'adrp' instruction to track global variables Tengda Wu
2026-08-08 12:23 ` [PATCH v4 22/23] perf annotate-arm64: Support per-cpu variable access tracking Tengda Wu
2026-08-08 13:18   ` sashiko-bot
2026-08-08 12:24 ` [PATCH v4 23/23] perf annotate-arm64: Support 'mrs' instruction to track 'current' pointer Tengda Wu
2026-08-08 13:20   ` sashiko-bot

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=20260808130503.AA8CA1F00A3A@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