Optimize __memset_chk, __memcpy_chk. DO NOT MERGE.
This change creates assembler versions of __memcpy_chk/__memset_chk
that is implemented in the memcpy/memset assembler code. This change
avoids an extra call to memcpy/memset, instead allowing a simple fall
through to occur from the chk code into the body of the real
implementation.
Testing:
- Ran the libc_test on __memcpy_chk/__memset_chk on all nexus devices.
- Wrote a small test executable that has three calls to __memcpy_chk and
three calls to __memset_chk. First call dest_len is length + 1. Second
call dest_len is length. Third call dest_len is length - 1.
Verified that the first two calls pass, and the third fails. Examined
the logcat output on all nexus devices to verify that the fortify
error message was sent properly.
- I benchmarked the new __memcpy_chk and __memset_chk on all systems. For
__memcpy_chk and large copies, the savings is relatively small (about 1%).
For small copies, the savings is large on cortex-a15/krait devices
(between 5% to 30%).
For cortex-a9 and small copies, the speed up is present, but relatively
small (about 3% to 5%).
For __memset_chk and large copies, the savings is also small (about 1%).
However, all processors show larger speed-ups on small copies (about 30% to
100%).
Bug: 9293744
Merge from internal master.
(cherry-picked from 7c860db0747f6276a6e43984d43f8fa5181ea936)
Change-Id: I916ad305e4001269460ca6ebd38aaa0be8ac7f52
diff --git a/libc/arch-arm/cortex-a15/bionic/memcpy.S b/libc/arch-arm/cortex-a15/bionic/memcpy.S
index d297064..2394024 100644
--- a/libc/arch-arm/cortex-a15/bionic/memcpy.S
+++ b/libc/arch-arm/cortex-a15/bionic/memcpy.S
@@ -59,6 +59,7 @@
#include <machine/cpu-features.h>
#include <machine/asm.h>
+#include "libc_events.h"
.text
.syntax unified
@@ -66,6 +67,13 @@
#define CACHE_LINE_SIZE 64
+ENTRY(__memcpy_chk)
+ cmp r2, r3
+ bgt fortify_check_failed
+
+ // Fall through to memcpy...
+END(__memcpy_chk)
+
ENTRY(memcpy)
// Assumes that n >= 0, and dst, src are valid pointers.
// For any sizes less than 832 use the neon code that doesn't
@@ -321,4 +329,21 @@
// Src is guaranteed to be at least word aligned by this point.
b word_aligned
+
+
+ // Only reached when the __memcpy_chk check fails.
+fortify_check_failed:
+ ldr r0, error_message
+ ldr r1, error_code
+1:
+ add r0, pc
+ bl __fortify_chk_fail
+error_code:
+ .word BIONIC_EVENT_MEMCPY_BUFFER_OVERFLOW
+error_message:
+ .word error_string-(1b+8)
END(memcpy)
+
+ .data
+error_string:
+ .string "memcpy buffer overflow"