Christopher Ferris 5f45d583b0 Create optimized __strcpy_chk/__strcat_chk.
This change pulls the memcpy code out into a new file so that the
__strcpy_chk and __strcat_chk can use it with an include.

The new versions of the two chk functions uses assembly versions
of strlen and memcpy to implement this check. This allows near
parity with the assembly versions of strcpy/strcat. It also means that
as memcpy implementations get faster, so do the chk functions.

Other included changes:
- Change all of the assembly labels to local labels. The other labels
  confuse gdb and mess up backtracing.
- Add .cfi_startproc and .cfi_endproc directives so that gdb is not
  confused when falling through from one function to another.
- Change all functions to use cfi directives since they are more powerful.
- Move the memcpy_chk fail code outside of the memcpy function definition
  so that backtraces work properly.
- Preserve lr before the calls to __fortify_chk_fail so that the backtrace
  actually works.

Testing:

- Ran the bionic unit tests. Verified all error messages in logs are set
  correctly.
- Ran libc_test, replacing strcpy with __strcpy_chk and replacing
  strcat with __strcat_chk.
- Ran the debugger on nexus10, nexus4, and old nexus7. Verified that the
  backtrace is correct for all fortify check failures. Also verify that
  when falling through from __memcpy_chk to memcpy that the backtrace is
  still correct. Also verified the same for __memset_chk and bzero.
  Verified the two different paths in the cortex-a9 memset routine that
  save variables to the stack still show the backtrace properly.

Bug: 9293744

(cherry-picked from 2be91915dcecc956d14ff281db0c7d216ca98af2)

Change-Id: Ia407b74d3287d0b6af0139a90b6eb3bfaebf2155
2013-08-15 11:13:39 -07:00

199 lines
5.7 KiB
ArmAsm

/*
* Copyright (C) 2013 The Android Open Source Project
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* * Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
* FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
* BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
* OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
* AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
* OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
* OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
#include <machine/cpu-features.h>
#include <machine/asm.h>
#include "libc_events.h"
/*
* Optimized memset() for ARM.
*
* memset() returns its first argument.
*/
.fpu neon
.syntax unified
ENTRY(__memset_chk)
.cfi_startproc
cmp r2, r3
bls .L_done
// Preserve lr for backtrace.
push {lr}
.cfi_def_cfa_offset 4
.cfi_rel_offset lr, 0
ldr r0, error_message
ldr r1, error_code
1:
add r0, pc
bl __fortify_chk_fail
error_code:
.word BIONIC_EVENT_MEMSET_BUFFER_OVERFLOW
error_message:
.word error_string-(1b+8)
.cfi_endproc
END(__memset_chk)
ENTRY(bzero)
.cfi_startproc
mov r2, r1
mov r1, #0
.L_done:
// Fall through to memset...
.cfi_endproc
END(bzero)
ENTRY(memset)
.cfi_startproc
stmfd sp!, {r0}
.cfi_def_cfa_offset 4
.cfi_rel_offset r0, 0
// The new algorithm is slower for copies < 16 so use the old
// neon code in that case.
cmp r2, #16
blo .L_set_less_than_16_unknown_align
// Use strd which requires an even and odd register so move the
// values so that:
// r0 and r1 contain the memset value
// r2 is the number of bytes to set
// r3 is the destination pointer
mov r3, r0
// Copy the byte value in every byte of r1.
mov r1, r1, lsl #24
orr r1, r1, r1, lsr #8
orr r1, r1, r1, lsr #16
.L_check_alignment:
// Align destination to a double word to avoid the strd crossing
// a cache line boundary.
ands ip, r3, #7
bne .L_do_double_word_align
.L_double_word_aligned:
mov r0, r1
subs r2, #64
blo .L_set_less_than_64
1: // Main loop sets 64 bytes at a time.
.irp offset, #0, #8, #16, #24, #32, #40, #48, #56
strd r0, r1, [r3, \offset]
.endr
add r3, #64
subs r2, #64
bge 1b
.L_set_less_than_64:
// Restore r2 to the count of bytes left to set.
add r2, #64
lsls ip, r2, #27
bcc .L_set_less_than_32
// Set 32 bytes.
.irp offset, #0, #8, #16, #24
strd r0, r1, [r3, \offset]
.endr
add r3, #32
.L_set_less_than_32:
bpl .L_set_less_than_16
// Set 16 bytes.
.irp offset, #0, #8
strd r0, r1, [r3, \offset]
.endr
add r3, #16
.L_set_less_than_16:
// Less than 16 bytes to set.
lsls ip, r2, #29
bcc .L_set_less_than_8
// Set 8 bytes.
strd r0, r1, [r3], #8
.L_set_less_than_8:
bpl .L_set_less_than_4
// Set 4 bytes
str r1, [r3], #4
.L_set_less_than_4:
lsls ip, r2, #31
it ne
strbne r1, [r3], #1
itt cs
strbcs r1, [r3], #1
strbcs r1, [r3]
ldmfd sp!, {r0}
bx lr
.L_do_double_word_align:
rsb ip, ip, #8
sub r2, r2, ip
movs r0, ip, lsl #31
it mi
strbmi r1, [r3], #1
itt cs
strbcs r1, [r3], #1
strbcs r1, [r3], #1
// Dst is at least word aligned by this point.
cmp ip, #4
blo .L_double_word_aligned
str r1, [r3], #4
b .L_double_word_aligned
.L_set_less_than_16_unknown_align:
// Set up to 15 bytes.
vdup.8 d0, r1
movs ip, r2, lsl #29
bcc 1f
vst1.8 {d0}, [r0]!
1: bge 2f
vst1.32 {d0[0]}, [r0]!
2: movs ip, r2, lsl #31
it mi
strbmi r1, [r0], #1
itt cs
strbcs r1, [r0], #1
strbcs r1, [r0], #1
ldmfd sp!, {r0}
bx lr
.cfi_endproc
END(memset)
.data
error_string:
.string "memset buffer overflow"