Re: [PATCH v12 03/19] perf env, dso, thread: Add _endian variants for e_machine helpers

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] Calling `maps__machine()` with a NULL `maps` pointer causes a segmentation fault.
--

commit 53562e7c68bd480e23dc08ba72db35bfbe234752
Author: Ian Rogers <irogers@google.com>
Subject: perf env, dso, thread: Add _endian variants for e_machine helpers

This commit adds new _endian variants for ELF machine helpers to support cross
architecture analysis and bi-endianness. These additions allow querying the
absolute endianness of a DSO or thread, which is required for external tools.

> diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c
> --- a/tools/perf/util/thread.c
> +++ b/tools/perf/util/thread.c
[ ... ]
> @@ -483,6 +500,7 @@ uint16_t thread__e_machine(struct thread *thread, struct machine *machine, uint3
>  		struct maps *maps = thread__maps(thread);
>  
>  		machine = maps__machine(maps);
> +		args.machine = machine;
>  	}
>  	tid = thread__tid(thread);

[Severity: High]
This is a pre-existing issue, but could this dereference a NULL pointer if
the thread lacks attached maps?

If thread__e_machine_endian is called concurrently while a new thread is
being initialized (for instance, via a concurrent iteration like
machines__for_each_thread), it seems the thread might be visible in the
machine's thread hashmap before thread__init_maps sets up the maps.

If thread__maps returns NULL here, will maps__machine attempt to
dereference the NULL pointer and trigger a segmentation fault? Should we add
a check to ensure maps is valid before looking up the machine?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260602152516.2831152-1-irogers@google.com?part=3