public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Masami Hiramatsu <mhiramat@kernel.org>
To: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Anders Roxell <anders.roxell@linaro.org>,
	Leo Yan <leo.yan@linaro.org>,
	Sergey Senozhatsky <sergey.senozhatsky@gmail.com>,
	linux-kernel@vger.kernel.org
Subject: [PATCH] tools/perf: build issues with Clang on arm64 and CFLAGS
Date: Fri,  7 May 2021 17:54:25 +0900	[thread overview]
Message-ID: <162037766527.94840.4323848545957010011.stgit@devnote2> (raw)

Hi Arnaldo,

Recently, I tested the perf build by clang on arm64 and found an issue
in arch/arm64/util/kvm-stat.c, related to missing-field-initializers warning.

----
 tools/perf $ make CC=clang LLVM=1 
...
arch/arm64/util/kvm-stat.c:74:9: error: missing field 'ops' initializer [-Werror,-Wmissing-field-initializers]
        { NULL },
               ^
1 error generated.
----

That error itself can be fixed easily by just adding field initializers
[1/1] Note that I didn't add Fixes tag, because I'm not sure clang build
of perf is officially supported or not, and this seems correct C code
to initialize a data structure with zero.(*)

And while investigating the error, I found another issue in the Makefile.config.

It seems to make CFLAGS from CORE_CFLAGS, INC_FLAGS, EXTRA_CFLAGS, EXTRA_WARNINGS
in the following order;

CFLAGS = $EXTRA_CFLAGS $EXTRA_WARNINGS $CORE_CFLAGS $INC_FLAGS

But since CORE_CFLAGS includes -Wall and -Wextra, the other -Wno-XXXX in
EXTRA_CFLAGS and EXTRA_WARNINGS are overriden and ignored.
So, I think it is better to define it as

CFLAGS = $CORE_CFLAGS $INC_FLAGS $EXTRA_CFLAGS $EXTRA_WARNINGS

But I also saw some configs tweaks CFLAGS directly. I think they should
modify EXTRA_CFLAGS.

My question is that this order is intentional or not. I might
miss something on it.

(*) BTW, there seems a discussion on the clang warning behavior,
 because gcc doesn't warn it anymore
 (https://gcc.gnu.org/bugzilla/show_bug.cgi?id=36750).
 It might be better to add -Wno-missing-field-initializers in case
 of CC=clang by default.


Thank you,

---

Masami Hiramatsu (1):
      tools/perf: Fix a build error on arm64 with clang


 tools/perf/arch/arm64/util/kvm-stat.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--
Masami Hiramatsu (Linaro) <mhiramat@kernel.org>

             reply	other threads:[~2021-05-07  8:54 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-07  8:54 Masami Hiramatsu [this message]
2021-05-07  8:54 ` [PATCH] tools/perf: Fix a build error on arm64 with clang Masami Hiramatsu
2021-05-09 12:50   ` Arnaldo Carvalho de Melo
2021-05-10 12:14 ` [PATCH] tools/perf: build issues with Clang on arm64 and CFLAGS Arnaldo Carvalho de Melo
2021-05-11 10:36   ` Masami Hiramatsu

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=162037766527.94840.4323848545957010011.stgit@devnote2 \
    --to=mhiramat@kernel.org \
    --cc=acme@kernel.org \
    --cc=anders.roxell@linaro.org \
    --cc=leo.yan@linaro.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sergey.senozhatsky@gmail.com \
    /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