[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <tencent_492BE2E81E62BEC199106096C025954AFD06@qq.com>
Date: Tue, 22 Aug 2023 08:38:58 +0800
From: Rong Tao <rtoax@...mail.com>
To: olsajiri@...il.com
Cc: andrii@...nel.org, ast@...nel.org, bpf@...r.kernel.org,
daniel@...earbox.net, haoluo@...gle.com, john.fastabend@...il.com,
kpsingh@...nel.org, linux-kernel@...r.kernel.org,
linux-kselftest@...r.kernel.org, martin.lau@...ux.dev,
mykolal@...com, rongtao@...tc.cn, rtoax@...mail.com,
sdf@...gle.com, shuah@...nel.org, song@...nel.org,
yonghong.song@...ux.dev
Subject: Re: [PATCH bpf-next v6] selftests/bpf: trace_helpers.c: optimize kallsyms cache
> I think you need to fix this on samples/bpf side
>
> I tried to play with the samples/bpf/ includes, but couldn't find a way to
> make this work.. selftests base includes on tools/include, while samples
> have $(objtree)/usr/include as first include and AFAICS the __must_check is
> defined under __KERNEL__ ifdef
>
> I guess the reason samples use $(objtree)/usr/include is to get some struct
> definitions which are not in tools/include, but looks like some samples objects
> already use vmlinux.h include, so that could be the way forward to fix that
I tried the method you suggested, and some unexpected problems occurred. Maybe,
we can apply v5 [0] first, and then solve this problem?
Best wishes,
Rong Tao
[0] v5: https://lore.kernel.org/lkml/tencent_0E9E1A1C0981678D5E7EA9E4BDBA8EE2200A@qq.com/
Powered by blists - more mailing lists