From: Nadav Amit <namit@vmware.com>
To: <linux-kernel@vger.kernel.org>, <x86@kernel.org>
Cc: Nadav Amit <namit@vmware.com>,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
Josh Poimboeuf <jpoimboe@redhat.com>
Subject: [PATCH v2 4/9] x86: alternatives: macrofy locks for better inlining
Date: Mon, 4 Jun 2018 04:21:26 -0700 [thread overview]
Message-ID: <20180604112131.59100-5-namit@vmware.com> (raw)
In-Reply-To: <20180604112131.59100-1-namit@vmware.com>
GCC considers the number of statements in inlined assembly blocks,
according to new-lines and semicolons, as an indication to the cost of
the block in time and space. This data is distorted by the kernel code,
which puts information in alternative sections. As a result, the
compiler may perform incorrect inlining and branch optimizations.
The solution is to set an assembly macro and call it from the inlined
assembly block. As a result GCC considers the inline assembly block as
a single instruction.
This patch handles the LOCK prefix, allowing more aggresive inlining.
text data bss dec hex filename
18140140 10225284 2957312 31322736 1ddf270 ./vmlinux before
18146889 10225380 2957312 31329581 1de0d2d ./vmlinux after (+6845)
Static text symbols:
Before: 40286
After: 40218 (-68)
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Nadav Amit <namit@vmware.com>
---
arch/x86/include/asm/alternative-asm.h | 20 ++++++++++++++------
arch/x86/include/asm/alternative.h | 16 ++--------------
arch/x86/kernel/macros.S | 1 +
3 files changed, 17 insertions(+), 20 deletions(-)
diff --git a/arch/x86/include/asm/alternative-asm.h b/arch/x86/include/asm/alternative-asm.h
index 31b627b43a8e..8e4ea39e55d0 100644
--- a/arch/x86/include/asm/alternative-asm.h
+++ b/arch/x86/include/asm/alternative-asm.h
@@ -7,16 +7,24 @@
#include <asm/asm.h>
#ifdef CONFIG_SMP
- .macro LOCK_PREFIX
-672: lock
+.macro LOCK_PREFIX_HERE
.pushsection .smp_locks,"a"
.balign 4
- .long 672b - .
+ .long 671f - . # offset
.popsection
- .endm
+671:
+.endm
+
+.macro LOCK_PREFIX insn:vararg
+ LOCK_PREFIX_HERE
+ lock \insn
+.endm
#else
- .macro LOCK_PREFIX
- .endm
+.macro LOCK_PREFIX_HERE
+.endm
+
+.macro LOCK_PREFIX insn:vararg
+.endm
#endif
/*
diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h
index 4cd6a3b71824..c1a3d7c76151 100644
--- a/arch/x86/include/asm/alternative.h
+++ b/arch/x86/include/asm/alternative.h
@@ -30,20 +30,8 @@
* and size information. That keeps the table sizes small.
*/
-#ifdef CONFIG_SMP
-#define LOCK_PREFIX_HERE \
- ".pushsection .smp_locks,\"a\"\n" \
- ".balign 4\n" \
- ".long 671f - .\n" /* offset */ \
- ".popsection\n" \
- "671:"
-
-#define LOCK_PREFIX LOCK_PREFIX_HERE "\n\tlock; "
-
-#else /* ! CONFIG_SMP */
-#define LOCK_PREFIX_HERE ""
-#define LOCK_PREFIX ""
-#endif
+#define LOCK_PREFIX_HERE "LOCK_PREFIX_HERE\n\t"
+#define LOCK_PREFIX "LOCK_PREFIX "
struct alt_instr {
s32 instr_offset; /* original instruction */
diff --git a/arch/x86/kernel/macros.S b/arch/x86/kernel/macros.S
index f1fe1d570365..852487a9fc56 100644
--- a/arch/x86/kernel/macros.S
+++ b/arch/x86/kernel/macros.S
@@ -8,3 +8,4 @@
#include <linux/compiler.h>
#include <asm/refcount.h>
+#include <asm/alternative-asm.h>
--
2.17.0
next prev parent reply other threads:[~2018-06-04 18:37 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-06-04 11:21 [PATCH v2 0/9] x86: macrofying inline asm for better compilation Nadav Amit
2018-06-04 11:21 ` [PATCH v2 1/9] Makefile: Prepare for using macros for inline asm Nadav Amit
2018-06-04 11:21 ` [PATCH v2 2/9] x86: objtool: use asm macro for better compiler decisions Nadav Amit
2018-06-04 19:04 ` Josh Poimboeuf
2018-06-05 5:41 ` kbuild test robot
2018-06-04 11:21 ` [PATCH v2 3/9] x86: refcount: prevent gcc distortions Nadav Amit
2018-06-04 22:06 ` Kees Cook
2018-06-04 22:20 ` Nadav Amit
2018-06-05 8:26 ` kbuild test robot
2018-06-04 11:21 ` Nadav Amit [this message]
2018-06-05 5:36 ` [PATCH v2 4/9] x86: alternatives: macrofy locks for better inlining kbuild test robot
2018-06-05 14:07 ` kbuild test robot
2018-06-07 3:05 ` [lkp-robot] [x86] 1a39381d70: WARNING:at_kernel/locking/mutex.c:#__mutex_unlock_slowpath kernel test robot
2018-06-04 11:21 ` [PATCH v2 5/9] x86: bug: prevent gcc distortions Nadav Amit
2018-06-05 7:34 ` kbuild test robot
2018-06-04 11:21 ` [PATCH v2 6/9] x86: prevent inline distortion by paravirt ops Nadav Amit
2018-06-04 11:21 ` [PATCH v2 7/9] x86: extable: use macros instead of inline assembly Nadav Amit
2018-06-04 11:21 ` [PATCH v2 8/9] x86: cpufeature: " Nadav Amit
2018-06-04 11:21 ` [PATCH v2 9/9] x86: jump-labels: " Nadav Amit
2018-06-04 19:05 ` [PATCH v2 0/9] x86: macrofying inline asm for better compilation Josh Poimboeuf
2018-06-04 19:56 ` Nadav Amit
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=20180604112131.59100-5-namit@vmware.com \
--to=namit@vmware.com \
--cc=hpa@zytor.com \
--cc=jpoimboe@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
--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).