From: Steven Rostedt <rostedt@goodmis.org>
To: linux-kernel@vger.kernel.org
Cc: "H. Peter Anvin" <hpa@zytor.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
Ingo Molnar <mingo@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Thomas Gleixner <tglx@linutronix.de>,
Peter Zijlstra <peterz@infradead.org>,
Jason Baron <jbaron@akamai.com>
Subject: [RFC][PATCH 2/2] x86/jump labels: Use etiher 5 byte or 2 byte jumps
Date: Wed, 07 Aug 2013 13:36:08 -0400 [thread overview]
Message-ID: <20130807174121.001163270@goodmis.org> (raw)
In-Reply-To: 20130807173606.872055511@goodmis.org
[-- Attachment #1: 0002-x86-jump-labels-Use-etiher-5-byte-or-2-byte-jumps.patch --]
[-- Type: text/plain, Size: 6709 bytes --]
From: Steven Rostedt <srostedt@redhat.com>
Have the jump labels add a "jmp" in the assembly instead
of a default nop. This will cause the assembler to put in
either a 2 byte or 5 byte jmp depending on where the target
lable is.
Then at compile time, the update_jump_label code will replace
the jmps with either 2 or 5 byte nops.
On boot up, the code can be examined to see if the jump label
uses either a 2 or 5 byte nop and replace it.
By allowing the jump labels to be 2 bytes, it speeds up the
nops, not only 2 byte nops are faster than 5 byte nops, but also
because it saves on cache foot print.
text data bss dec hex filename
13403667 3666856 2998272 20068795 13239bb ../nobackup/mxtest/vmlinux-old
13398536 3666856 2998272 20063664 13225b0 ../nobackup/mxtest/vmlinux-new
Converting the current v3.2 trace points saved 5,131 bytes.
As more places use jump labels, this will have a bigger savings.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
arch/x86/Kconfig | 1 +
arch/x86/include/asm/jump_label.h | 7 ++-
arch/x86/kernel/jump_label.c | 88 +++++++++++++++++++++++++++----------
3 files changed, 73 insertions(+), 23 deletions(-)
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index b32ebf9..5dbbee2 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -81,6 +81,7 @@ config X86
select HAVE_USER_RETURN_NOTIFIER
select ARCH_BINFMT_ELF_RANDOMIZE_PIE
select HAVE_ARCH_JUMP_LABEL
+ select HAVE_BUILD_TIME_JUMP_LABEL
select HAVE_TEXT_POKE_SMP
select HAVE_GENERIC_HARDIRQS
select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE
diff --git a/arch/x86/include/asm/jump_label.h b/arch/x86/include/asm/jump_label.h
index 64507f3..615adb4 100644
--- a/arch/x86/include/asm/jump_label.h
+++ b/arch/x86/include/asm/jump_label.h
@@ -10,6 +10,11 @@
#define JUMP_LABEL_NOP_SIZE 5
+/*
+ * The STATIC_KEY_INIT_NOP must match the nops used in
+ * scripts/update_jump_label.c. Otherwise the boot time checks
+ * will fail and trigger a BUG() on boot up.
+ */
#ifdef CONFIG_X86_64
# define STATIC_KEY_INIT_NOP P6_NOP5_ATOMIC
#else
@@ -19,7 +24,7 @@
static __always_inline bool arch_static_branch(struct static_key *key)
{
asm goto("1:"
- ".byte " __stringify(STATIC_KEY_INIT_NOP) "\n\t"
+ "jmp %l[l_yes]\n"
".pushsection __jump_table, \"aw\" \n\t"
_ASM_ALIGN "\n\t"
_ASM_PTR "1b, %l[l_yes], %c0 \n\t"
diff --git a/arch/x86/kernel/jump_label.c b/arch/x86/kernel/jump_label.c
index 912a528..1a943aa 100644
--- a/arch/x86/kernel/jump_label.c
+++ b/arch/x86/kernel/jump_label.c
@@ -16,12 +16,20 @@
#ifdef HAVE_JUMP_LABEL
+/* These are the nops added at compile time */
+static const unsigned char nop_short[] = { P6_NOP2 };
+static const unsigned char default_nop[] = { STATIC_KEY_INIT_NOP };
+
union jump_code_union {
char code[JUMP_LABEL_NOP_SIZE];
struct {
char jump;
int offset;
- } __attribute__((packed));
+ } __packed;
+ struct {
+ char jump_short;
+ char offset_short;
+ } __packed;
};
static void bug_at(unsigned char *ip, int line)
@@ -42,19 +50,29 @@ static void __jump_label_transform(struct jump_entry *entry,
int init)
{
union jump_code_union code;
+ unsigned char nop;
+ unsigned char op;
+ unsigned size;
+ void *ip = (void *)entry->code;
const unsigned char *ideal_nop = ideal_nops[NOP_ATOMIC5];
- if (type == JUMP_LABEL_ENABLE) {
- /*
- * We are enabling this jump label. If it is not a nop
- * then something must have gone wrong.
- */
- if (unlikely(memcmp((void *)entry->code, ideal_nop, 5) != 0))
- bug_at((void *)entry->code, __LINE__);
+ /* Use probe_kernel_read()? */
+ op = *(unsigned char *)ip;
+ nop = ideal_nops[NOP_ATOMIC5][0];
- code.jump = 0xe9;
- code.offset = entry->target -
- (entry->code + JUMP_LABEL_NOP_SIZE);
+ if (type == JUMP_LABEL_ENABLE) {
+ if (memcmp(ip, nop_short, 2) == 0) {
+ size = 2;
+ code.jump_short = 0xeb;
+ code.offset = entry->target - (entry->code + 2);
+ /* Check for overflow ? */
+ } else if ((!init && memcmp(ip, ideal_nop, 5) == 0) ||
+ (init && memcmp(ip, default_nop, 5) == 0)) {
+ size = JUMP_LABEL_NOP_SIZE;
+ code.jump = 0xe9;
+ code.offset = entry->target - (entry->code + size);
+ } else
+ bug_at(ip, __LINE__);
} else {
/*
* We are disabling this jump label. If it is not what
@@ -63,20 +81,48 @@ static void __jump_label_transform(struct jump_entry *entry,
* are converting the default nop to the ideal nop.
*/
if (init) {
- const unsigned char default_nop[] = { STATIC_KEY_INIT_NOP };
- if (unlikely(memcmp((void *)entry->code, default_nop, 5) != 0))
- bug_at((void *)entry->code, __LINE__);
- } else {
+ /* Ignore short nops, we do not change them */
+ if (memcmp(ip, nop_short, 2) == 0)
+ return;
+
+ /* We are initializing from the default nop */
+ if (unlikely(memcmp(ip, default_nop, 5) != 0))
+ bug_at(ip, __LINE__);
+
+ /* Set to the ideal nop */
+ size = JUMP_LABEL_NOP_SIZE;
+ memcpy(&code, ideal_nops[NOP_ATOMIC5], size);
+
+ } else if (op == 0xe9) {
+ /* Replace a 5 byte jmp */
+
+ /* Make sure this is what we expected it to be */
code.jump = 0xe9;
code.offset = entry->target -
(entry->code + JUMP_LABEL_NOP_SIZE);
- if (unlikely(memcmp((void *)entry->code, &code, 5) != 0))
- bug_at((void *)entry->code, __LINE__);
- }
- memcpy(&code, ideal_nops[NOP_ATOMIC5], JUMP_LABEL_NOP_SIZE);
+
+ if (unlikely(memcmp(ip, &code, 5) != 0))
+ bug_at(ip, __LINE__);
+
+ size = JUMP_LABEL_NOP_SIZE;
+ memcpy(&code, ideal_nops[NOP_ATOMIC5], size);
+ } else if (op == 0xeb) {
+ /* Replace a 2 byte jmp */
+
+ /* Had better be a 2 byte jmp */
+ code.jump_short = 0xeb;
+ code.offset = entry->target - (entry->code + 2);
+ if (unlikely(memcmp(ip, &code, 2) != 0))
+ bug_at(ip, __LINE__);
+
+ size = 2;
+ memcpy(&code, nop_short, size);
+ } else
+ /* The code was not what we expected! */
+ bug_at(ip, __LINE__);
}
- (*poker)((void *)entry->code, &code, JUMP_LABEL_NOP_SIZE);
+ (*poker)(ip, &code, size);
}
void arch_jump_label_transform(struct jump_entry *entry,
@@ -106,7 +152,6 @@ __init_or_module void arch_jump_label_transform_static(struct jump_entry *entry,
* If it is not, then we need to update the nop to the ideal nop.
*/
if (jlstate == JL_STATE_START) {
- const unsigned char default_nop[] = { STATIC_KEY_INIT_NOP };
const unsigned char *ideal_nop = ideal_nops[NOP_ATOMIC5];
if (memcmp(ideal_nop, default_nop, 5) != 0)
@@ -117,5 +162,4 @@ __init_or_module void arch_jump_label_transform_static(struct jump_entry *entry,
if (jlstate == JL_STATE_UPDATE)
__jump_label_transform(entry, type, text_poke_early, 1);
}
-
#endif
--
1.7.10.4
next prev parent reply other threads:[~2013-08-07 17:41 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-08-07 17:36 [RFC][PATCH 0/2] x86/jump labels: Add the 5 byte or 2 byte jumps Steven Rostedt
2013-08-07 17:36 ` [RFC][PATCH 1/2] jump labels: Add infrastructure to update jump labels at compile time Steven Rostedt
2013-08-07 17:36 ` Steven Rostedt [this message]
2013-08-07 17:57 ` [RFC][PATCH 2/2] x86/jump labels: Use etiher 5 byte or 2 byte jumps Steven Rostedt
2013-08-07 17:54 ` [RFC][PATCH 3/2] x86/jump labels: Count and display the short jumps used Steven Rostedt
2013-08-07 19:22 ` Linus Torvalds
2013-08-07 19:27 ` H. Peter Anvin
2013-08-07 19:49 ` Linus Torvalds
2013-08-07 20:30 ` Steven Rostedt
2013-08-07 20:19 ` Jason Baron
2013-08-07 20:33 ` Steven Rostedt
2013-08-07 20:47 ` Linus Torvalds
2013-08-07 21:37 ` Jason Baron
2013-08-07 21:56 ` Linus Torvalds
2013-08-10 0:14 ` Andy Lutomirski
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=20130807174121.001163270@goodmis.org \
--to=rostedt@goodmis.org \
--cc=akpm@linux-foundation.org \
--cc=hpa@zytor.com \
--cc=jbaron@akamai.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.