Re: [Buildroot] [PATCH] package/procps-ng: fix pidfd_open checking

[Thomas Petazzoni via buildroot <buildroot@buildroot.org>]

Hello Scott,

Sorry for the slow feedback. Not sure what the status of this is and
whether my answer is still useful/relevant.

On Fri, 1 Nov 2024 09:31:08 +0800
Scott Fan <fancp2007@gmail.com> wrote:

> > So I'm not sure my previous review was clear enough: what is wrong is
> > the procps-ng patch itself, it shouldn't AC_MSG_ERROR(), but instead it
> > should automatically disable pidwait, as if the user had passed
> > --disable-pidwait.
> 
> I understand what you mean, but I may not have expressed my thoughts clearly.
> 
> Before this patch was applied, I got a compile error (not configure
> error) when the package was built with linux-headers-4.19.322,
> Then I tried to figure out what was going wrong and I eventually found
> this patch from upstream.
> 
> After this patch was applied, it threw a configure error (not compile
> error), which is what I expected.
> 
> As the patch said:
> The previous build setup would check for pidfd_open using
> AC_CHECK_FUNC and would be incorrectly reported as true.
> 
> Therefore, I think the key to this patch is that it successfully fixes
> the pidfd_open checking issue.
> 
> If pidfd_open() and __NR_pidfd_open were not present, it will report
> an AC_MSG_ERROR message,
> and gives us the --disable-pidwait configure option, so you are
> explicit in not wanting and knowing you wont get pidwait.
> 
> Whereas you mean, instead of AC_MSG_ERROR(), it should automatically
> disable pidwait, as if the user had passed --disable-pidwait.
> I'm not sure if this is acceptable to the upstream maintainers, who
> want you to be clear about the purpose of the option you passed.
> Even if it is approved , it would require a new patch to improve this behavior.
> 
> 
> To solve my compilation problem (with linux-headers-4.19.322), in the
> current case, I had to pass the --disable-pidwait option.
> If at some point in the future, when the patch you expect is
> successfully applied, I will remove it.

What you can do then is:

ifeq ($(BR2_TOOLCHAIN_HEADERS_AT_LEAST_4_19),)
PROCPS_NG_CONF_OPTS += disable-pidwait
endif

This will ensure to disable pidwait support when headers are < 4.19.

Best regards,

Thomas
-- 
Thomas Petazzoni, co-owner and CEO, Bootlin
Embedded Linux and Kernel engineering and training
https://bootlin.com
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot