[PATCH i-g-t] lib/igt_aux.c: since procps-ng 4.0.5, PIDS_VAL() takes 3 arguments, not 4

[Cavitt, Jonathan]

-----Original Message-----
From: igt-dev <igt-dev-bounces at lists.freedesktop.org> On Behalf Of Thomas Petazzoni
Sent: Friday, January 3, 2025 2:26 PM
To: igt-dev at lists.freedesktop.org
Cc: Thomas Petazzoni <thomas.petazzoni at bootlin.com>
Subject: [PATCH i-g-t] lib/igt_aux.c: since procps-ng 4.0.5, PIDS_VAL() takes 3 arguments, not 4
> 
> Unfortunately, the API of procps-ng has changed between procps-ng
> 4.0.4 and procps-ng 4.0.5, with the PIDS_VAL() macro changing in
> upstream commit
> https://gitlab.com/procps-ng/procps/-/commit/967fdcfb06e20aad0f31073474cb94545c9bdea5.
> 
> This API breakage has been reported upstream as of
> https://gitlab.com/procps-ng/procps/-/issues/366, but in the mean
> time, the build of igt-gpu-tools breaks with:
> 
> ../lib/igt_aux.c:1358:78: error: macro "PIDS_VAL" passed 4 arguments, but takes just 3
> 
> We fix this by detecting the version of the libproc2 library, and
> adjusting accordingly how PIDS_VAL() gets used.
> 
> Signed-off-by: Thomas Petazzoni <thomas.petazzoni at bootlin.com>
> ---
>  lib/igt_aux.c | 7 +++++++
>  meson.build   | 3 +++
>  2 files changed, 10 insertions(+)
> 
> diff --git a/lib/igt_aux.c b/lib/igt_aux.c
> index 3407cc4f2..f5bf48da6 100644
> --- a/lib/igt_aux.c
> +++ b/lib/igt_aux.c
> @@ -1358,10 +1358,17 @@ static bool get_process_ids(struct igt_process *prcs)
>  	prcs->comm = NULL;
>  	prcs->stack = procps_pids_get(prcs->info, PIDS_FETCH_TASKS_ONLY);
>  	if (prcs->stack) {
> +#if defined(HAVE_LIBPROC2_POST_4_0_5_API)

"HAVE_LIBPROC2_POST_4_0_5_API" works as a name for this new tag,
though I wonder if "HAVE_LIBPROC2_NO_INFO" would also work?

I don't see any other granular version checks in the meson build file (at
least, I don't see any that aren't strict version requirements for certain
dependencies), so AFAICT this type of tag is fairly novel.  So whatever
name we end up choosing may end up inadvertently becoming a 
standard naming convention for future tags like this one.

... Well, assuming I haven't just missed where the standard is already 
set, which is fairly likely all things considered.

But again, the current name works, so I won't block on this decision.

Reviewed-by: Jonathan Cavitt <jonathan.cavitt at intel.com>
-Jonathan Cavitt

> +		prcs->tid = PIDS_VAL(EU_PID, s_int, prcs->stack);
> +		prcs->euid = PIDS_VAL(EU_EUID, s_int, prcs->stack);
> +		prcs->egid = PIDS_VAL(EU_EGID, s_int, prcs->stack);
> +		prcs->comm = PIDS_VAL(EU_CMD, str, prcs->stack);
> +#else
>  		prcs->tid = PIDS_VAL(EU_PID, s_int, prcs->stack, prcs->info);
>  		prcs->euid = PIDS_VAL(EU_EUID, s_int, prcs->stack, prcs->info);
>  		prcs->egid = PIDS_VAL(EU_EGID, s_int, prcs->stack, prcs->info);
>  		prcs->comm = PIDS_VAL(EU_CMD, str, prcs->stack, prcs->info);
> +#endif
>  	}
>  #endif
>  	return prcs->tid != 0;
> diff --git a/meson.build b/meson.build
> index 8b2a2a64a..f3e645c4d 100644
> --- a/meson.build
> +++ b/meson.build
> @@ -139,6 +139,9 @@ if libprocps.found()
>    config.set('HAVE_LIBPROCPS', 1)
>  elif libproc2.found()
>    config.set('HAVE_LIBPROC2', 1)
> +  if libproc2.version().version_compare('>= 4.0.5')
> +    config.set('HAVE_LIBPROC2_POST_4_0_5_API', 1)
> +  endif
>  else
>    error('Either libprocps or libproc2 is required')
>  endif
> -- 
> 2.47.1
> 
>