From: "H. Peter Anvin" <hpa@zytor.com>
To: Borislav Petkov <bp@suse.de>, Brian Gerst <brgerst@gmail.com>
Cc: the arch/x86 maintainers <x86@kernel.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Ingo Molnar <mingo@kernel.org>,
Denys Vlasenko <dvlasenk@redhat.com>,
Andy Lutomirski <luto@amacapital.net>,
Linus Torvalds <torvalds@linux-foundation.org>
Subject: Re: [PATCH] x86: static_cpu_has_safe: discard dynamic check after init
Date: Mon, 18 Jan 2016 17:33:03 -0800 [thread overview]
Message-ID: <DD7599E0-BB19-4BEC-B598-B0B29ACF5A36@zytor.com> (raw)
In-Reply-To: <20160119011026.GA12911@pd.tnic>
On January 18, 2016 5:10:26 PM PST, Borislav Petkov <bp@suse.de> wrote:
>On Mon, Jan 18, 2016 at 07:51:07PM +0100, Borislav Petkov wrote:
>> [45] .static_cpu_has PROGBITS ffffffff97aa655b 16ea655b
>> 00000000000002df 0000000000000000 AX 0 0 1
>
>Ok, staring at this section was wrong. I went and looked at
>the .s file and now it clicked: gcc inlines those calls to
>__static_cpu_has_safe of the dynamic jump target:
>
> .loc 4 538 0
> movl $125, %edi #,
> call __static_cpu_has_safe #
>.LBE885:
>.LBE886:
>.LBE914:
> .loc 1 240 0
> testb %al, %al # D.30157
> je .L150 #,
> jmp .L151 #
>.L152:
>.LBB915:
>.LBB909:
>.LBB905:
>.LBB893:
>.LBB892:
> .loc 4 538 0
> movl $154, %edi #,
> call __static_cpu_has_safe #
>.LBE892:
>.LBE893:
>.LBE905:
> .loc 7 431 0
> testb %al, %al # D.30157
> jne .L154 #,
>
>which turn into:
>
> 751: bf 7d 00 00 00 mov $0x7d,%edi
> 756: e8 00 00 00 00 callq 75b <fpu__copy+0xab>
> 75b: 84 c0 test %al,%al
> 75d: 74 a3 je 702 <fpu__copy+0x52>
> 75f: eb 90 jmp 6f1 <fpu__copy+0x41>
> 761: bf 9a 00 00 00 mov $0x9a,%edi
> 766: e8 00 00 00 00 callq 76b <fpu__copy+0xbb>
> 76b: 84 c0 test %al,%al
>
>That's like 28 bytes in this particular case, which is 14 per call
>site,
>on average.
>
>So grepping through my tailored vmlinux, it has 35 entries with
>X86_FEATURE_ALWAYS in .altinstructions X 14 = 490 bytes.
>
>So yeah, we probably should do this, allyesconfig should give more
>savings.
>
>One thing I'd probably do differently is not call the throwaway section
>.static_cpu_has but something like .altinstr_temporary or so and put it
>after the replacement insns:
>
>.altinstr_replacement : AT(ADDR(.altinstr_replacement) - LOAD_OFFSET) {
> *(.altinstr_replacement)
> *(.altinstr_temporary)
> }
>
>so that we know those instructions belong to the alternatives
>mechanism.
>They'll get discared too, of course.
>
>I could just as well be talking a lot of crap, it is waay too late
>here.
Why the f do we call a subroutine for what amounts to a single bt or test instruction?
--
Sent from my Android device with K-9 Mail. Please excuse brevity and formatting.
next prev parent reply other threads:[~2016-01-19 1:33 UTC|newest]
Thread overview: 66+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-01-16 19:22 [PATCH] x86: static_cpu_has_safe: discard dynamic check after init Brian Gerst
2016-01-16 19:36 ` Borislav Petkov
2016-01-16 19:58 ` Brian Gerst
2016-01-17 10:33 ` Borislav Petkov
2016-01-18 16:52 ` Brian Gerst
2016-01-18 17:49 ` Andy Lutomirski
2016-01-18 18:14 ` Borislav Petkov
2016-01-18 18:29 ` Andy Lutomirski
2016-01-18 18:39 ` Borislav Petkov
2016-01-18 19:45 ` H. Peter Anvin
2016-01-18 23:05 ` Borislav Petkov
2016-01-18 23:13 ` H. Peter Anvin
2016-01-18 23:25 ` Borislav Petkov
2016-01-19 13:57 ` Borislav Petkov
2016-01-19 16:23 ` Borislav Petkov
2016-01-19 23:10 ` Borislav Petkov
2016-01-19 23:26 ` Andy Lutomirski
2016-01-19 23:49 ` Boris Petkov
2016-01-20 4:03 ` H. Peter Anvin
2016-01-20 10:33 ` Borislav Petkov
2016-01-20 10:41 ` H. Peter Anvin
2016-01-21 22:14 ` Borislav Petkov
2016-01-21 22:22 ` H. Peter Anvin
2016-01-21 22:56 ` Borislav Petkov
2016-01-21 23:36 ` H. Peter Anvin
2016-01-21 23:37 ` H. Peter Anvin
2016-01-22 10:32 ` Borislav Petkov
2016-01-18 18:51 ` Borislav Petkov
2016-01-19 1:10 ` Borislav Petkov
2016-01-19 1:33 ` H. Peter Anvin [this message]
2016-01-19 9:22 ` Borislav Petkov
2016-01-20 4:02 ` H. Peter Anvin
2016-01-20 4:39 ` Brian Gerst
2016-01-20 4:42 ` H. Peter Anvin
2016-01-20 10:50 ` Borislav Petkov
2016-01-20 10:55 ` H. Peter Anvin
2016-01-20 11:05 ` Borislav Petkov
2016-01-20 14:48 ` H. Peter Anvin
2016-01-20 15:01 ` Borislav Petkov
2016-01-20 15:09 ` H. Peter Anvin
2016-01-20 16:04 ` Borislav Petkov
2016-01-20 16:16 ` H. Peter Anvin
2016-01-23 6:50 [PATCH] x86/head_64.S: do not use temporary register to check alignment Alexander Kuleshov
2016-01-26 9:31 ` Borislav Petkov
2016-01-26 21:12 [PATCH 00/10] tip-queue 2016-01-26, rest Borislav Petkov
2016-01-26 21:12 ` [PATCH 01/10] x86/asm: Add condition codes clobber to memory barrier macros Borislav Petkov
2016-01-26 21:12 ` [PATCH 02/10] x86/asm: Drop a comment left over from X86_OOSTORE Borislav Petkov
2016-01-26 21:12 ` [PATCH 03/10] x86/asm: Tweak the comment about wmb() use for IO Borislav Petkov
2016-01-26 21:12 ` [PATCH 04/10] x86/cpufeature: Carve out X86_FEATURE_* Borislav Petkov
2016-01-30 13:18 ` [tip:x86/asm] " tip-bot for Borislav Petkov
2016-01-26 21:12 ` [PATCH 05/10] x86/cpufeature: Replace the old static_cpu_has() with safe variant Borislav Petkov
2016-01-30 13:19 ` [tip:x86/asm] " tip-bot for Borislav Petkov
2016-01-26 21:12 ` [PATCH 06/10] x86/cpufeature: Get rid of the non-asm goto variant Borislav Petkov
2016-01-27 3:36 ` Brian Gerst
2016-01-27 8:41 ` Borislav Petkov
2016-01-27 8:43 ` [PATCH -v1.1 " Borislav Petkov
2016-01-30 13:19 ` [tip:x86/asm] " tip-bot for Borislav Petkov
2016-01-27 8:45 ` [PATCH -v1.1 8/10] x86/alternatives: Discard dynamic check after init Borislav Petkov
2016-01-30 13:20 ` [tip:x86/asm] " tip-bot for Brian Gerst
2016-01-26 21:12 ` [PATCH 07/10] x86/alternatives: Add an auxilary section Borislav Petkov
2016-01-30 13:19 ` [tip:x86/asm] " tip-bot for Borislav Petkov
2016-01-26 21:12 ` [PATCH 08/10] x86/alternatives: Discard dynamic check after init Borislav Petkov
2016-01-26 21:12 ` [PATCH 09/10] x86/vdso: Use static_cpu_has() Borislav Petkov
2016-01-30 13:20 ` [tip:x86/asm] " tip-bot for Borislav Petkov
2016-01-26 21:12 ` [PATCH 10/10] x86/head_64: Simplify kernel load address alignment check Borislav Petkov
2016-01-30 13:20 ` [tip:x86/boot] x86/boot: " tip-bot for Alexander Kuleshov
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=DD7599E0-BB19-4BEC-B598-B0B29ACF5A36@zytor.com \
--to=hpa@zytor.com \
--cc=bp@suse.de \
--cc=brgerst@gmail.com \
--cc=dvlasenk@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@amacapital.net \
--cc=mingo@kernel.org \
--cc=torvalds@linux-foundation.org \
--cc=x86@kernel.org \
/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;
as well as URLs for NNTP newsgroup(s).