Linux Perf Users
 help / color / mirror / Atom feed
From: Namhyung Kim <namhyung@kernel.org>
To: sashiko-reviews@lists.linux.dev
Cc: linux-perf-users@vger.kernel.org
Subject: Re: [PATCH v1 6/7] perf symbol: Set binary type for JIT map DSOs
Date: Mon, 5 Oct 2026 23:05:12 -0700	[thread overview]
Message-ID: <asSPmK_QgxgE6SON@z2> (raw)
In-Reply-To: <20261005064616.F07241F00893@smtp.kernel.org>

On Mon, Oct 05, 2026 at 06:46:16AM +0000, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [Medium] perf: annotate regression on JIT map files due to unhandled binary type
> --
> 
> commit 79781ef5615895fe2ec492b108f82094b6a0246b
> Author: Namhyung Kim <namhyung@kernel.org>
> 
> perf symbol: Set binary type for JIT map DSOs
> 
> This commit explicitly sets the binary type of JIT map DSOs to JAVA_JIT. This
> ensures they are identified correctly and bypass attempts to load dso_data, 
> as they only contain symbol names and lack binary instructions.
> 
> > diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
> > --- a/tools/perf/util/symbol.c
> > +++ b/tools/perf/util/symbol.c
> > @@ -1859,6 +1859,7 @@ int dso__load(struct dso *dso, struct map *map)
> >  
> >  	if (perfmap) {
> >  		ret = dso__load_perf_map(map_path, dso);
> > +		dso__set_binary_type(dso, DSO_BINARY_TYPE__JAVA_JIT);
> 
> [Severity: Medium]
> Does setting the binary type to DSO_BINARY_TYPE__JAVA_JIT here bypass the
> historical early exit guard in symbol__disassemble()?
> 
> Looking at symbol__disassemble() in tools/perf/util/disasm.c, it guards
> against unsupported files with:
> 
>     if (dso__binary_type(dso) == DSO_BINARY_TYPE__NOT_FOUND) {
>         return SYMBOL_ANNOTATE_ERRNO__COULDNT_DETERMINE_FILE_TYPE;
>     }
> 
> With the type now explicitly set to DSO_BINARY_TYPE__JAVA_JIT, this check
> fails and allows perf annotate to incorrectly attempt disassembling the text
> map file. 
> 
> The fallback binary disassemblers inevitably fail to parse the text file as
> ELF and return -1. This propagates back to symbol__strerror_disassemble(),
> which handles the -1 return by printing an unhelpful message to the user
> instead of gracefully exiting:
> 
>     default:
>         scnprintf(buf, buflen, "Internal error: Invalid %d error code\n", errnum);
>         break;
> 
> Should symbol__disassemble() be updated to check for DSO_BINARY_TYPE__JAVA_JIT
> so that it can exit gracefully when users try to annotate a JIT symbol profile?

Ok, but I think the dso__disassemble_filename() should be updated to set
binary type correctly and just use dso__get_filename().  Will do in v2.

Thanks,
Namhyung

  reply	other threads:[~2026-10-06  6:05 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-05  6:34 [PATCH v1 0/7] perf symbol: Properly set DSO binary/symtab types Namhyung Kim
2026-10-05  6:34 ` [PATCH v1 1/7] perf tools: Remove redundant dso data init Namhyung Kim
2026-10-05  6:39   ` sashiko-bot
2026-10-05  6:34 ` [PATCH v1 2/7] perf tools: Try linked debug files for DSO debug info Namhyung Kim
2026-10-05  6:43   ` sashiko-bot
2026-10-05  6:34 ` [PATCH v1 3/7] perf symbol: Update symtab type of vmlinux from build-id cache Namhyung Kim
2026-10-05  6:40   ` sashiko-bot
2026-10-05  6:34 ` [PATCH v1 4/7] perf symbol: Set binary/symtab type for split kallsyms Namhyung Kim
2026-10-05  6:40   ` sashiko-bot
2026-10-05  6:34 ` [PATCH v1 5/7] perf symbol: Set dso symtab type for libbfd Namhyung Kim
2026-10-05  6:43   ` sashiko-bot
2026-10-05  6:34 ` [PATCH v1 6/7] perf symbol: Set binary type for JIT map DSOs Namhyung Kim
2026-10-05  6:46   ` sashiko-bot
2026-10-06  6:05     ` Namhyung Kim [this message]
2026-10-05  6:34 ` [PATCH v1 7/7] perf symbol: Do not set binary type from symtab type Namhyung Kim
2026-10-05  6:45   ` sashiko-bot
2026-10-06  6:30     ` Namhyung Kim

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=asSPmK_QgxgE6SON@z2 \
    --to=namhyung@kernel.org \
    --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