Linux Perf Users
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Alireza Haghdoost" <haghdoost@uber.com>
Cc: linux-perf-users@vger.kernel.org
Subject: Re: [PATCH v7 3/4] perf script: Add --lazy-load-symbols to load symbol names lazily
Date: Sat, 14 Nov 2026 05:31:00 +0000	[thread overview]
Message-ID: <sashiko-outbox-214179@kernel.org> (raw)
In-Reply-To: <20261113-perf-symbol-memory-send-v7-3-d992eced9f9c@uber.com>

Thank you for your contribution! Sashiko AI review found 1 potential
issue to consider:
- [Low] perf symbols: skipping .dynsym copy in dso__load_sym_internal()
  alters fixup
--

commit bbcf49c4a773356002f2cf428fb0dbd0071729ba
Author: Alireza Haghdoost <haghdoost@uber.com>

perf script: Add --lazy-load-symbols to load symbol names lazily

This commit adds a --lazy-load-symbols option to perf script that defers
reading and demangling userspace ELF symbol names until a symbol lookup
returns them, storing the string table file offset in the meantime.

> diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c
> index 98b6d7c73d235..7b7f06a882f61 100644
> --- a/tools/perf/util/symbol-elf.c
> +++ b/tools/perf/util/symbol-elf.c

[ ... ]

> @@ -1774,14 +2020,36 @@ dso__load_sym_internal(struct dso *dso, struct map *map, struct symsrc *syms_ss,
>  			}
>  		}
>  
> -		demangled = dso__demangle_sym(dso, kmodule, elf_name);
> -		if (demangled != NULL)
> -			elf_name = demangled;
> +		/*
> +		 * Most .dynsym entries are copies of .symtab symbols, which
> +		 * duplicate selection would drop anyway. Skip them here, so
> +		 * that perf does not read both names just to choose one.
> +		 */
> +		if (lazy_names && dynsym && !symbol_conf.allow_aliases) {
> +			f = lazy_dynsym_copy(dso__symbols(curr_dso), elf, &sym, elf_name);
> +			if (f) {
> +				if (GELF_ST_TYPE(sym.st_info) == STT_GNU_IFUNC)
> +					symbol__set_ifunc_alias(f, true);
> +				continue;
> +			}
> +		}

[Severity: Low]
Can skipping the .dynsym copy here change how symbols__fixup_end() and
symbols__fixup_duplicate() resolve symbols when .symtab also has a
zero-size symbol at the same address?

__symbols__insert() places symbols with equal start addresses in insertion
order.  If .symtab has a sized symbol followed by a zero-size symbol at
the same address (such as an assembly label without a .size directive),
and .dynsym has a copy of the sized symbol, eager loading inserts the
.dynsym copy after the zero-size symbol.  When symbols__fixup_end() runs,
the node following the zero-size symbol still has the same start address,
so the zero-size symbol keeps a zero length and symbols__fixup_duplicate()
drops it in favor of the sized symbol.

When this check skips inserting the .dynsym copy, the zero-size symbol is
left as the last node at that start address.  symbols__fixup_end() then
stretches its end to the next higher symbol address (or rounds it up by
4096 if it is the last symbol), giving it a bogus non-zero size so
symbols__fixup_duplicate() no longer discards it for having zero length
and can select it over the real sized symbol.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261113-perf-symbol-memory-send-v7-0-d992eced9f9c@uber.com?part=3

  reply	other threads:[~2026-11-14  5:31 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-14  5:09 [PATCH v7 0/4] perf script: Lazy symbol loading Alireza Haghdoost via B4 Relay
2026-11-14  5:09 ` [PATCH v7 1/4] perf symbols: Split out name-independent duplicate symbol selection Alireza Haghdoost via B4 Relay
2026-11-14  5:12   ` sashiko-bot
2026-11-14  5:09 ` [PATCH v7 2/4] perf symbols: Point to the symbol name instead of embedding it Alireza Haghdoost via B4 Relay
2026-11-14  5:16   ` sashiko-bot
2026-11-14  5:09 ` [PATCH v7 3/4] perf script: Add --lazy-load-symbols to load symbol names lazily Alireza Haghdoost via B4 Relay
2026-11-14  5:31   ` sashiko-bot [this message]
2026-11-14  5:09 ` [PATCH v7 4/4] perf test: Test lazy symbol name loading Alireza Haghdoost via B4 Relay
2026-11-14  5:19   ` 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=sashiko-outbox-214179@kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=haghdoost@uber.com \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=sashiko-reviews@lists.linux.dev \
    /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