linux-kernel - Re: [RFC] [PATCH] arch: x86: change GCC optimisation target from atom to bonnell

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] [day] [month] [year] [list]

Message-ID: <CALjTZvbzc+FP-OrZQZQceSOvxygDFFF5XuKEREFWLR_rpiwBnA@mail.gmail.com>
Date:   Tue, 11 Oct 2016 09:40:09 +0100
From:   Rui Salvaterra <rsalvaterra@...il.com>
To:     hpa@...or.com
Cc:     tglx@...utronix.de, mingo@...hat.com, x86@...nel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [RFC] [PATCH] arch: x86: change GCC optimisation target from atom
 to bonnell

2016-10-11 1:11 GMT+01:00  <hpa@...or.com>:
> On October 10, 2016 2:45:38 PM PDT, Rui Salvaterra <rsalvaterra@...il.com> wrote:
>>Hi, Thomas, Ingo, Peter,
>>
>>(Sending as RFC, since I don't know if this patch is acceptable.)
>>
>>The GCC team has deprecated atom as a march/mtune target since almost
>>three
>>years ago, according to this mailing list thread [1], in favour of
>>specific
>>microarchitecture names (bonnell, silvermont). This patch changes the
>>Atom
>>optimisation target in the kernel to bonnell, as it was originally
>>intended.
>>
>>Tested on an x86-64 Atom 330 machine. No functional changes.
>>
>>[1] https://gcc.gnu.org/ml/gcc-patches/2013-12/msg01805.html
>>
>>Signed-off-by: Rui Salvaterra <rsalvaterra@...il.com>
>>---
>> arch/x86/Makefile        | 4 ++--
>> arch/x86/Makefile_32.cpu | 4 ++--
>> 2 files changed, 4 insertions(+), 4 deletions(-)
>>
>>diff --git a/arch/x86/Makefile b/arch/x86/Makefile
>>index 2d44933..b7d615f 100644
>>--- a/arch/x86/Makefile
>>+++ b/arch/x86/Makefile
>>@@ -109,8 +109,8 @@ else
>>
>>         cflags-$(CONFIG_MCORE2) += \
>>        $(call cc-option,-march=core2,$(call cc-option,-mtune=generic))
>>-      cflags-$(CONFIG_MATOM) += $(call cc-option,-march=atom) \
>>-              $(call cc-option,-mtune=atom,$(call cc-option,-mtune=generic))
>>+      cflags-$(CONFIG_MATOM) += $(call cc-option,-march=bonnell) \
>>+              $(call cc-option,-mtune=bonnell,$(call cc-option,-mtune=generic))
>>       cflags-$(CONFIG_GENERIC_CPU) += $(call cc-option,-mtune=generic)
>>         KBUILD_CFLAGS += $(cflags-y)
>>
>>diff --git a/arch/x86/Makefile_32.cpu b/arch/x86/Makefile_32.cpu
>>index 6647ed4..d66e5e3 100644
>>--- a/arch/x86/Makefile_32.cpu
>>+++ b/arch/x86/Makefile_32.cpu
>>@@ -32,8 +32,8 @@ cflags-$(CONFIG_MCYRIXIII)   += $(call
>>cc-option,-march=c3,-march=i486) $(align)-f
>> cflags-$(CONFIG_MVIAC3_2)     += $(call cc-option,-march=c3-2,-march=i686)
>> cflags-$(CONFIG_MVIAC7)               += -march=i686
>> cflags-$(CONFIG_MCORE2)               += -march=i686 $(call tune,core2)
>>-cflags-$(CONFIG_MATOM)                += $(call cc-option,-march=atom,$(call
>>cc-option,-march=core2,-march=i686)) \
>>-      $(call cc-option,-mtune=atom,$(call cc-option,-mtune=generic))
>>+cflags-$(CONFIG_MATOM)                += $(call cc-option,-march=bonnell,$(call
>>cc-option,-march=core2,-march=i686)) \
>>+      $(call cc-option,-mtune=bonnell,$(call cc-option,-mtune=generic))
>>
>> # AMD Elan support
>> cflags-$(CONFIG_MELAN)                += -march=i486
>
> This just breaks backwards compatibility with older gcc.  If you want to add a comment that's fine, though.
>
> Nacked-by: hpa
> --
> Sent from my Android device with K-9 Mail. Please excuse my brevity.


Hi, Hans,

I suspected that (hence the RFC), but I couldn't find the minimum supported
gcc version. I was just worried the bonnell microarch could get neglected in
the long term, when it needs all the (compiler) help it can get, being in-order.

Thanks,

Rui