linux-kernel - Re: [PATCH] Revert "kbuild: use -Oz instead of -Os when using clang"

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <0275C6DD-7F77-4D51-9C35-B25ECF61EB0F@zytor.com>
Date:   Mon, 18 Mar 2019 15:07:47 -0700
From:   hpa@...or.com
To:     Matthias Kaehlcke <mka@...omium.org>,
        Nick Desaulniers <ndesaulniers@...gle.com>
CC:     Masahiro Yamada <yamada.masahiro@...ionext.com>,
        Michal Marek <michal.lkml@...kovi.net>,
        Linux Kbuild mailing list <linux-kbuild@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Behan Webster <behanw@...verseincode.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Manoj Gupta <manojgupta@...omium.org>,
        Tiancong Wang <tcwang@...omium.org>,
        Stephen Hines <srhines@...gle.com>,
        clang-built-linux@...glegroups.com
Subject: Re: [PATCH] Revert "kbuild: use -Oz instead of -Os when using clang"

On March 18, 2019 2:56:05 PM PDT, Matthias Kaehlcke <mka@...omium.org> wrote:
>On Mon, Mar 18, 2019 at 02:47:13PM -0700, 'Nick Desaulniers' via Clang
>Built Linux wrote:
>> On Mon, Mar 18, 2019 at 2:10 PM Matthias Kaehlcke <mka@...omium.org>
>wrote:
>> >
>> > The clang option -Oz enables *aggressive* optimization for size,
>> > which doesn't necessarily result in smaller images, but can have
>> > negative impact on performance. Switch back to the less aggressive
>> > -Os.
>> >
>> > This reverts commit 6748cb3c299de1ffbe56733647b01dbcc398c419.
>> 
>> Does scripts/checkpatch.pl complain about this format (full sha, no
>> title)?
>
>Nope, for reverts checkpatch is happy with it, I suppose because by
>default the subject of the reverted patch is part of the revert's
>subject.
>
>> Not sure that necessitates a v2, but if so would be good to
>> mention that Clang generates excessive calls into libgcc/compiler-rt
>> at -Oz.  Thanks for the patch and the discussion/sanity check.
>
>I can send a v2 if 'needed'.
>
>> Reviewed-by: Nick Desaulniers <ndesaulniers@...gle.com>
>
>Thanks!
>
>Matthias
>
>> >
>> > Suggested-by: Peter Zijlstra <peterz@...radead.org>
>> > Signed-off-by: Matthias Kaehlcke <mka@...omium.org>
>> > ---
>> >  Makefile | 2 +-
>> >  1 file changed, 1 insertion(+), 1 deletion(-)
>> >
>> > diff --git a/Makefile b/Makefile
>> > index 9ef547fc7ffe..191f3ce3cb5e 100644
>> > --- a/Makefile
>> > +++ b/Makefile
>> > @@ -667,7 +667,7 @@ KBUILD_CFLAGS       += $(call
>cc-disable-warning, format-overflow)
>> >  KBUILD_CFLAGS  += $(call cc-disable-warning, int-in-bool-context)
>> >
>> >  ifdef CONFIG_CC_OPTIMIZE_FOR_SIZE
>> > -KBUILD_CFLAGS  += $(call cc-option,-Oz,-Os)
>> > +KBUILD_CFLAGS  += -Os
>> >  else
>> >  KBUILD_CFLAGS   += -O2
>> >  endif
>> 
>> 
>> 

Would be nice, still, to have -Ok.
-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.