aboutsummaryrefslogtreecommitdiff
path: root/arch/x86/lib
diff options
context:
space:
mode:
authorDave Airlie2022-05-03 16:08:48 +1000
committerDave Airlie2022-05-03 16:08:48 +1000
commite954d2c94d007afe487044ecfa48f2518643df0e (patch)
tree7903ee4cf7d9387e5bb5285f86a2a82b2eacab44 /arch/x86/lib
parent15e2b419a83777ef652bf5f0ea4385ef7711f710 (diff)
parent672c0c5173427e6b3e2a9bbb7be51ceeec78093a (diff)
Backmerge tag 'v5.18-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux into drm-next
Linux 5.18-rc5 There was a build fix for arm I wanted in drm-next, so backmerge rather then cherry-pick. Signed-off-by: Dave Airlie <airlied@redhat.com>
Diffstat (limited to 'arch/x86/lib')
-rw-r--r--arch/x86/lib/copy_user_64.S87
-rw-r--r--arch/x86/lib/putuser.S4
-rw-r--r--arch/x86/lib/retpoline.S2
-rw-r--r--arch/x86/lib/usercopy_64.c2
4 files changed, 58 insertions, 37 deletions
diff --git a/arch/x86/lib/copy_user_64.S b/arch/x86/lib/copy_user_64.S
index 8ca5ecf16dc4..9dec1b38a98f 100644
--- a/arch/x86/lib/copy_user_64.S
+++ b/arch/x86/lib/copy_user_64.S
@@ -53,12 +53,12 @@
SYM_FUNC_START(copy_user_generic_unrolled)
ASM_STAC
cmpl $8,%edx
- jb 20f /* less then 8 bytes, go to byte copy loop */
+ jb .Lcopy_user_short_string_bytes
ALIGN_DESTINATION
movl %edx,%ecx
andl $63,%edx
shrl $6,%ecx
- jz .L_copy_short_string
+ jz copy_user_short_string
1: movq (%rsi),%r8
2: movq 1*8(%rsi),%r9
3: movq 2*8(%rsi),%r10
@@ -79,37 +79,11 @@ SYM_FUNC_START(copy_user_generic_unrolled)
leaq 64(%rdi),%rdi
decl %ecx
jnz 1b
-.L_copy_short_string:
- movl %edx,%ecx
- andl $7,%edx
- shrl $3,%ecx
- jz 20f
-18: movq (%rsi),%r8
-19: movq %r8,(%rdi)
- leaq 8(%rsi),%rsi
- leaq 8(%rdi),%rdi
- decl %ecx
- jnz 18b
-20: andl %edx,%edx
- jz 23f
- movl %edx,%ecx
-21: movb (%rsi),%al
-22: movb %al,(%rdi)
- incq %rsi
- incq %rdi
- decl %ecx
- jnz 21b
-23: xor %eax,%eax
- ASM_CLAC
- RET
+ jmp copy_user_short_string
30: shll $6,%ecx
addl %ecx,%edx
- jmp 60f
-40: leal (%rdx,%rcx,8),%edx
- jmp 60f
-50: movl %ecx,%edx
-60: jmp .Lcopy_user_handle_tail /* ecx is zerorest also */
+ jmp .Lcopy_user_handle_tail
_ASM_EXTABLE_CPY(1b, 30b)
_ASM_EXTABLE_CPY(2b, 30b)
@@ -127,10 +101,6 @@ SYM_FUNC_START(copy_user_generic_unrolled)
_ASM_EXTABLE_CPY(14b, 30b)
_ASM_EXTABLE_CPY(15b, 30b)
_ASM_EXTABLE_CPY(16b, 30b)
- _ASM_EXTABLE_CPY(18b, 40b)
- _ASM_EXTABLE_CPY(19b, 40b)
- _ASM_EXTABLE_CPY(21b, 50b)
- _ASM_EXTABLE_CPY(22b, 50b)
SYM_FUNC_END(copy_user_generic_unrolled)
EXPORT_SYMBOL(copy_user_generic_unrolled)
@@ -191,7 +161,7 @@ EXPORT_SYMBOL(copy_user_generic_string)
SYM_FUNC_START(copy_user_enhanced_fast_string)
ASM_STAC
/* CPUs without FSRM should avoid rep movsb for short copies */
- ALTERNATIVE "cmpl $64, %edx; jb .L_copy_short_string", "", X86_FEATURE_FSRM
+ ALTERNATIVE "cmpl $64, %edx; jb copy_user_short_string", "", X86_FEATURE_FSRM
movl %edx,%ecx
1: rep movsb
xorl %eax,%eax
@@ -244,6 +214,53 @@ SYM_CODE_START_LOCAL(.Lcopy_user_handle_tail)
SYM_CODE_END(.Lcopy_user_handle_tail)
/*
+ * Finish memcpy of less than 64 bytes. #AC should already be set.
+ *
+ * Input:
+ * rdi destination
+ * rsi source
+ * rdx count (< 64)
+ *
+ * Output:
+ * eax uncopied bytes or 0 if successful.
+ */
+SYM_CODE_START_LOCAL(copy_user_short_string)
+ movl %edx,%ecx
+ andl $7,%edx
+ shrl $3,%ecx
+ jz .Lcopy_user_short_string_bytes
+18: movq (%rsi),%r8
+19: movq %r8,(%rdi)
+ leaq 8(%rsi),%rsi
+ leaq 8(%rdi),%rdi
+ decl %ecx
+ jnz 18b
+.Lcopy_user_short_string_bytes:
+ andl %edx,%edx
+ jz 23f
+ movl %edx,%ecx
+21: movb (%rsi),%al
+22: movb %al,(%rdi)
+ incq %rsi
+ incq %rdi
+ decl %ecx
+ jnz 21b
+23: xor %eax,%eax
+ ASM_CLAC
+ RET
+
+40: leal (%rdx,%rcx,8),%edx
+ jmp 60f
+50: movl %ecx,%edx /* ecx is zerorest also */
+60: jmp .Lcopy_user_handle_tail
+
+ _ASM_EXTABLE_CPY(18b, 40b)
+ _ASM_EXTABLE_CPY(19b, 40b)
+ _ASM_EXTABLE_CPY(21b, 50b)
+ _ASM_EXTABLE_CPY(22b, 50b)
+SYM_CODE_END(copy_user_short_string)
+
+/*
* copy_user_nocache - Uncached memory copy with exception handling
* This will force destination out of cache for more performance.
*
diff --git a/arch/x86/lib/putuser.S b/arch/x86/lib/putuser.S
index ecb2049c1273..b7dfd60243b7 100644
--- a/arch/x86/lib/putuser.S
+++ b/arch/x86/lib/putuser.S
@@ -48,6 +48,7 @@ SYM_FUNC_START(__put_user_1)
cmp %_ASM_BX,%_ASM_CX
jae .Lbad_put_user
SYM_INNER_LABEL(__put_user_nocheck_1, SYM_L_GLOBAL)
+ ENDBR
ASM_STAC
1: movb %al,(%_ASM_CX)
xor %ecx,%ecx
@@ -62,6 +63,7 @@ SYM_FUNC_START(__put_user_2)
cmp %_ASM_BX,%_ASM_CX
jae .Lbad_put_user
SYM_INNER_LABEL(__put_user_nocheck_2, SYM_L_GLOBAL)
+ ENDBR
ASM_STAC
2: movw %ax,(%_ASM_CX)
xor %ecx,%ecx
@@ -76,6 +78,7 @@ SYM_FUNC_START(__put_user_4)
cmp %_ASM_BX,%_ASM_CX
jae .Lbad_put_user
SYM_INNER_LABEL(__put_user_nocheck_4, SYM_L_GLOBAL)
+ ENDBR
ASM_STAC
3: movl %eax,(%_ASM_CX)
xor %ecx,%ecx
@@ -90,6 +93,7 @@ SYM_FUNC_START(__put_user_8)
cmp %_ASM_BX,%_ASM_CX
jae .Lbad_put_user
SYM_INNER_LABEL(__put_user_nocheck_8, SYM_L_GLOBAL)
+ ENDBR
ASM_STAC
4: mov %_ASM_AX,(%_ASM_CX)
#ifdef CONFIG_X86_32
diff --git a/arch/x86/lib/retpoline.S b/arch/x86/lib/retpoline.S
index 5f87bab4fb8d..b2b2366885a2 100644
--- a/arch/x86/lib/retpoline.S
+++ b/arch/x86/lib/retpoline.S
@@ -31,6 +31,7 @@
.align RETPOLINE_THUNK_SIZE
SYM_INNER_LABEL(__x86_indirect_thunk_\reg, SYM_L_GLOBAL)
UNWIND_HINT_EMPTY
+ ANNOTATE_NOENDBR
ALTERNATIVE_2 __stringify(ANNOTATE_RETPOLINE_SAFE; jmp *%\reg), \
__stringify(RETPOLINE \reg), X86_FEATURE_RETPOLINE, \
@@ -55,7 +56,6 @@ SYM_INNER_LABEL(__x86_indirect_thunk_\reg, SYM_L_GLOBAL)
.align RETPOLINE_THUNK_SIZE
SYM_CODE_START(__x86_indirect_thunk_array)
- ANNOTATE_NOENDBR // apply_retpolines
#define GEN(reg) THUNK reg
#include <asm/GEN-for-each-reg.h>
diff --git a/arch/x86/lib/usercopy_64.c b/arch/x86/lib/usercopy_64.c
index 0402a749f3a0..0ae6cf804197 100644
--- a/arch/x86/lib/usercopy_64.c
+++ b/arch/x86/lib/usercopy_64.c
@@ -119,7 +119,7 @@ void __memcpy_flushcache(void *_dst, const void *_src, size_t size)
/* cache copy and flush to align dest */
if (!IS_ALIGNED(dest, 8)) {
- unsigned len = min_t(unsigned, size, ALIGN(dest, 8) - dest);
+ size_t len = min_t(size_t, size, ALIGN(dest, 8) - dest);
memcpy((void *) dest, (void *) source, len);
clean_cache_range((void *) dest, len);