mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: x86@kernel.org, tony.luck@intel.com, pjt@google.com
Cc: linux-kernel@vger.kernel.org, peterz@infradead.org,
	r.marek@assembler.cz, jpoimboe@redhat.com, jikos@kernel.org
Subject: [RFC][PATCH 2/2] x86/retpoline: Compress retpolines
Date: Thu, 18 Feb 2021 17:59:40 +0100	[thread overview]
Message-ID: <20210218172151.942060087@infradead.org> (raw)
In-Reply-To: <20210218165938.213678824@infradead.org>

By using int3 as a speculation fence instead of lfence, we can shrink
the longest alternative to just 15 bytes:

  0:   e8 05 00 00 00          callq  a <.altinstr_replacement+0xa>
  5:   f3 90                   pause  
  7:   cc                      int3   
  8:   eb fb                   jmp    5 <.altinstr_replacement+0x5>
  a:   48 89 04 24             mov    %rax,(%rsp)
  e:   c3                      retq   

This means we can change the alignment from 32 to 16 bytes and get 4
retpolines per cacheline, $I win.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
 arch/x86/lib/retpoline.S |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

--- a/arch/x86/lib/retpoline.S
+++ b/arch/x86/lib/retpoline.S
@@ -16,7 +16,7 @@
 .Lspec_trap_\@:
 	UNWIND_HINT_EMPTY
 	pause
-	lfence
+	int3
 	jmp .Lspec_trap_\@
 .Ldo_rop_\@:
 	mov     %\reg, (%_ASM_SP)
@@ -27,7 +27,7 @@
 .macro THUNK reg
 	.section .text.__x86.indirect_thunk
 
-	.align 32
+	.align 16
 SYM_FUNC_START(__x86_indirect_thunk_\reg)
 
 	ALTERNATIVE_2 __stringify(ANNOTATE_RETPOLINE_SAFE; jmp *%\reg), \



  parent reply	other threads:[~2021-02-18 19:05 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-02-18 16:59 [RFC][PATCH 0/2] x86/retpoline: Retpoline on a diet Peter Zijlstra
2021-02-18 16:59 ` [RFC][PATCH 1/2] x86/retpoline: Simplify retpolines Peter Zijlstra
2021-02-22 11:36   ` Peter Zijlstra
2021-02-18 16:59 ` Peter Zijlstra [this message]
2021-02-19  7:14   ` [RFC][PATCH 2/2] x86/retpoline: Compress retpolines Borislav Petkov
2021-02-22 11:27     ` Peter Zijlstra
2021-02-18 18:46 ` [RFC PATCH] x86/retpolines: Prevent speculation after RET Borislav Petkov
2021-02-18 19:02   ` Peter Zijlstra
2021-02-18 19:11     ` Borislav Petkov
2021-02-19  8:15       ` Peter Zijlstra
2021-02-19 12:08         ` Andrew Cooper
2021-02-19  9:28     ` David Laight

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=20210218172151.942060087@infradead.org \
    --to=peterz@infradead.org \
    --cc=jikos@kernel.org \
    --cc=jpoimboe@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pjt@google.com \
    --cc=r.marek@assembler.cz \
    --cc=tony.luck@intel.com \
    --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

Powered by JetHome