[PATCH v2 2/8] newlib: libc: strcmp M-profile PACBTI-enablement
Victor Do Nascimento <[email protected]>
| Newsgroups | gmane.comp.lib.newlib |
|---|---|
| Message-ID | <[email protected]> |
Add function prologue/epilogue to conditionally add BTI landing pads
and/or PAC code generation & authentication instructions depending on
compilation flags.
This patch enables PACBTI for all relevant variants of strcmp:
* Newlib for armv8.1-m.main+pacbti
* Newlib for armv8.1-m.main+pacbti+mve
* Newlib-nano
---
newlib/libc/machine/arm/strcmp-arm-tiny.S | 7 +++-
newlib/libc/machine/arm/strcmp-armv7.S | 45 +++++++++++++++++------
newlib/libc/machine/arm/strcmp-armv7m.S | 36 +++++++++++++++---
3 files changed, 71 insertions(+), 17 deletions(-)
diff --git a/newlib/libc/machine/arm/strcmp-arm-tiny.S b/newlib/libc/machine/arm/strcmp-arm-tiny.S
index 607a41daf..8085eb4df 100644
--- a/newlib/libc/machine/arm/strcmp-arm-tiny.S
+++ b/newlib/libc/machine/arm/strcmp-arm-tiny.S
@@ -29,10 +29,14 @@
/* Tiny version of strcmp in ARM state. Used only when optimizing
for size. Also supports Thumb-2. */
+#include "pacbti.h"
+
.syntax unified
def_fn strcmp
+ .fnstart
.cfi_sections .debug_frame
.cfi_startproc
+ pacbti_prologue
1:
ldrb r2, [r0], #1
ldrb r3, [r1], #1
@@ -42,6 +46,7 @@ def_fn strcmp
beq 1b
2:
subs r0, r2, r3
- bx lr
+ pacbti_epilogue
.cfi_endproc
+ .fnend
.size strcmp, . - strcmp
diff --git a/newlib/libc/machine/arm/strcmp-armv7.S b/newlib/libc/machine/arm/strcmp-armv7.S
index 2f93bfb73..584c89f7e 100644
--- a/newlib/libc/machine/arm/strcmp-armv7.S
+++ b/newlib/libc/machine/arm/strcmp-armv7.S
@@ -45,6 +45,8 @@
.thumb
.syntax unified
+#include "pacbti.h"
+
/* Parameters and result. */
#define src1 r0
#define src2 r1
@@ -91,8 +93,9 @@
ldrd r4, r5, [sp], #16
.cfi_restore 4
.cfi_restore 5
+ .cfi_adjust_cfa_offset -16
sub result, result, r1, lsr #24
- bx lr
+ pacbti_epilogue
#else
/* To use the big-endian trick we'd have to reverse all three words.
that's slower than this approach. */
@@ -112,22 +115,28 @@
ldrd r4, r5, [sp], #16
.cfi_restore 4
.cfi_restore 5
+ .cfi_adjust_cfa_offset -16
sub result, result, r1
- bx lr
+ pacbti_epilogue
#endif
.endm
+
.text
.p2align 5
+ .fnstart
+ .cfi_sections .debug_frame
+ .cfi_startproc
.Lstrcmp_start_addr:
#ifndef STRCMP_NO_PRECHECK
.Lfastpath_exit:
sub r0, r2, r3
- bx lr
+ pacbti_epilogue
nop
#endif
def_fn strcmp
+ pacbti_prologue
#ifndef STRCMP_NO_PRECHECK
ldrb r2, [src1]
ldrb r3, [src2]
@@ -136,16 +145,26 @@ def_fn strcmp
cmpcs r2, r3
bne .Lfastpath_exit
#endif
- .cfi_sections .debug_frame
- .cfi_startproc
strd r4, r5, [sp, #-16]!
- .cfi_def_cfa_offset 16
+ .save {r4, r5}
+ .cfi_adjust_cfa_offset 16
+#if __HAVE_PAC_LEAF
+ .cfi_offset 4, -20
+ .cfi_offset 5, -16
+#else
.cfi_offset 4, -16
.cfi_offset 5, -12
+#endif /* __HAVE_PAC_LEAF */
orr tmp1, src1, src2
strd r6, r7, [sp, #8]
+ .save {r6, r7}
+#if __HAVE_PAC_LEAF
+ .cfi_offset 6, -12
+ .cfi_offset 7, -8
+#else
.cfi_offset 6, -8
.cfi_offset 7, -4
+#endif /* __HAVE_PAC_LEAF */
mvn const_m1, #0
lsl r2, tmp1, #29
cbz r2, .Lloop_aligned8
@@ -270,7 +289,6 @@ def_fn strcmp
ldr data1, [src1], #4
beq .Laligned_m2
bcs .Laligned_m1
-
#ifdef STRCMP_NO_PRECHECK
ldrb data2, [src2, #1]
uxtb tmp1, data1, ror #BYTE1_OFFSET
@@ -314,7 +332,8 @@ def_fn strcmp
mov result, tmp1
ldr r4, [sp], #16
.cfi_restore 4
- bx lr
+ .cfi_adjust_cfa_offset -16
+ pacbti_epilogue
#ifndef STRCMP_NO_PRECHECK
.Laligned_m1:
@@ -364,8 +383,9 @@ def_fn strcmp
/* R6/7 Not used in this sequence. */
.cfi_restore 6
.cfi_restore 7
+ .cfi_adjust_cfa_offset -16
neg result, result
- bx lr
+ pacbti_epilogue
6:
.cfi_restore_state
@@ -441,7 +461,8 @@ def_fn strcmp
/* R6/7 not used in this sequence. */
.cfi_restore 6
.cfi_restore 7
- bx lr
+ .cfi_adjust_cfa_offset -16
+ pacbti_epilogue
.Lstrcmp_tail:
.cfi_restore_state
@@ -463,7 +484,9 @@ def_fn strcmp
/* R6/7 not used in this sequence. */
.cfi_restore 6
.cfi_restore 7
+ .cfi_adjust_cfa_offset -16
sub result, result, data2, lsr #24
- bx lr
+ pacbti_epilogue
.cfi_endproc
+ .fnend
.size strcmp, . - .Lstrcmp_start_addr
diff --git a/newlib/libc/machine/arm/strcmp-armv7m.S b/newlib/libc/machine/arm/strcmp-armv7m.S
index cdb4912df..fe1519f4d 100644
--- a/newlib/libc/machine/arm/strcmp-armv7m.S
+++ b/newlib/libc/machine/arm/strcmp-armv7m.S
@@ -29,6 +29,8 @@
/* Very similar to the generic code, but uses Thumb2 as implemented
in ARMv7-M. */
+#include "pacbti.h"
+
/* Parameters and result. */
#define src1 r0
#define src2 r1
@@ -44,8 +46,10 @@
.thumb
.syntax unified
def_fn strcmp
+ .fnstart
.cfi_sections .debug_frame
.cfi_startproc
+ pacbti_prologue
eor tmp1, src1, src2
tst tmp1, #3
/* Strings not at same byte offset from a word boundary. */
@@ -106,7 +110,7 @@ def_fn strcmp
lsrs result, result, #24
subs result, result, data2
#endif
- bx lr
+ pacbti_epilogue
#if 0
@@ -214,12 +218,18 @@ def_fn strcmp
cmpcs data1, data2
beq .Lstrcmp_unaligned
sub result, data1, data2
- bx lr
+ pacbti_epilogue
2:
stmfd sp!, {r5}
- .cfi_def_cfa_offset 4
+ .save {r5}
+ .cfi_adjust_cfa_offset 4
+#if __HAVE_PAC_LEAF
+ .cfi_offset 5, -8 /* Account for ip register already on stack. */
+#else
.cfi_offset 5, -4
+#endif /* __HAVE_PAC_LEAF */
+
ldr data1, [src1], #4
and tmp2, src2, #3
@@ -353,10 +363,17 @@ def_fn strcmp
.Lstrcmp_done_equal:
mov result, #0
.cfi_remember_state
+#if __HAVE_PAC_LEAF
+ pop {r5, ip}
+ .cfi_restore 5
+ .cfi_restore 143
+ aut ip, lr, sp
+#else
ldmfd sp!, {r5}
.cfi_restore 5
+#endif /* __HAVE_PAC_LEAF */
.cfi_def_cfa_offset 0
- bx lr
+ bx lr
.Lstrcmp_tail:
.cfi_restore_state
@@ -370,9 +387,18 @@ def_fn strcmp
S2LOEQ data2, data2, #8
beq .Lstrcmp_tail
sub result, r2, result
+#if __HAVE_PAC_LEAF
+ pop {r5, ip}
+ .cfi_restore 5
+ .cfi_restore 143
+ .cfi_def_cfa_offset 0
+ aut ip, lr, sp
+#else
ldmfd sp!, {r5}
.cfi_restore 5
.cfi_def_cfa_offset 0
- bx lr
+#endif /* __HAVE_PAC_LEAF */
+ bx lr
.cfi_endproc
+ .fnend
.size strcmp, . - strcmp
--
2.36.1