5.15.40-rc1 review

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 arch/x86/lib/atomic64_386_32.S |   84 ++++++++++++++++++++++-------------------
 1 file changed, 46 insertions(+), 38 deletions(-)

arch/x86/lib/atomic64_386_32.S +46 ~~-38~~

@@ -9,81 +9,83 @@
 #include <asm/alternative.h>
 
 /* if you want SMP support, implement these with real spinlocks */
-.macro LOCK reg
+.macro IRQ_SAVE reg
 	pushfl
 	cli
 .endm
 
-.macro UNLOCK reg
+.macro IRQ_RESTORE reg
 	popfl
 .endm
 
-#define BEGIN(op) \
+#define BEGIN_IRQ_SAVE(op) \
 .macro endp; \
 SYM_FUNC_END(atomic64_##op##_386); \
 .purgem endp; \
 .endm; \
 SYM_FUNC_START(atomic64_##op##_386); \
-	LOCK v;
+	IRQ_SAVE v;
 
 #define ENDP endp
 
-#define RET \
-	UNLOCK v; \
+#define RET_IRQ_RESTORE \
+	IRQ_RESTORE v; \
 	ret
 
-#define RET_ENDP \
-	RET; \
-	ENDP
-
 #define v %ecx
-BEGIN(read)
+BEGIN_IRQ_SAVE(read)
 	movl  (v), %eax
 	movl 4(v), %edx
-RET_ENDP
+	RET_IRQ_RESTORE
+ENDP
 #undef v
 
 #define v %esi
-BEGIN(set)
+BEGIN_IRQ_SAVE(set)
 	movl %ebx,  (v)
 	movl %ecx, 4(v)
-RET_ENDP
+	RET_IRQ_RESTORE
+ENDP
 #undef v
 
 #define v  %esi
-BEGIN(xchg)
+BEGIN_IRQ_SAVE(xchg)
 	movl  (v), %eax
 	movl 4(v), %edx
 	movl %ebx,  (v)
 	movl %ecx, 4(v)
-RET_ENDP
+	RET_IRQ_RESTORE
+ENDP
 #undef v
 
 #define v %ecx
-BEGIN(add)
+BEGIN_IRQ_SAVE(add)
 	addl %eax,  (v)
 	adcl %edx, 4(v)
-RET_ENDP
+	RET_IRQ_RESTORE
+ENDP
 #undef v
 
 #define v %ecx
-BEGIN(add_return)
+BEGIN_IRQ_SAVE(add_return)
 	addl  (v), %eax
 	adcl 4(v), %edx
 	movl %eax,  (v)
 	movl %edx, 4(v)
-RET_ENDP
+	RET_IRQ_RESTORE
+ENDP
 #undef v
 
 #define v %ecx
-BEGIN(sub)
+BEGIN_IRQ_SAVE(sub)
 	subl %eax,  (v)
 	sbbl %edx, 4(v)
-RET_ENDP
+	RET_IRQ_RESTORE
+ENDP
 #undef v
 
 #define v %ecx
-BEGIN(sub_return)
+BEGIN_IRQ_SAVE(sub_return)
 	negl %edx
 	negl %eax
 	sbbl $0, %edx

@@ -91,47 +93,52 @@ BEGIN(sub_return)
 	adcl 4(v), %edx
 	movl %eax,  (v)
 	movl %edx, 4(v)
-RET_ENDP
+	RET_IRQ_RESTORE
+ENDP
 #undef v
 
 #define v %esi
-BEGIN(inc)
+BEGIN_IRQ_SAVE(inc)
 	addl $1,  (v)
 	adcl $0, 4(v)
-RET_ENDP
+	RET_IRQ_RESTORE
+ENDP
 #undef v
 
 #define v %esi
-BEGIN(inc_return)
+BEGIN_IRQ_SAVE(inc_return)
 	movl  (v), %eax
 	movl 4(v), %edx
 	addl $1, %eax
 	adcl $0, %edx
 	movl %eax,  (v)
 	movl %edx, 4(v)
-RET_ENDP
+	RET_IRQ_RESTORE
+ENDP
 #undef v
 
 #define v %esi
-BEGIN(dec)
+BEGIN_IRQ_SAVE(dec)
 	subl $1,  (v)
 	sbbl $0, 4(v)
-RET_ENDP
+	RET_IRQ_RESTORE
+ENDP
 #undef v
 
 #define v %esi
-BEGIN(dec_return)
+BEGIN_IRQ_SAVE(dec_return)
 	movl  (v), %eax
 	movl 4(v), %edx
 	subl $1, %eax
 	sbbl $0, %edx
 	movl %eax,  (v)
 	movl %edx, 4(v)
-RET_ENDP
+	RET_IRQ_RESTORE
+ENDP
 #undef v
 
 #define v %esi
-BEGIN(add_unless)
+BEGIN_IRQ_SAVE(add_unless)
 	addl %eax, %ecx
 	adcl %edx, %edi
 	addl  (v), %eax

@@ -143,7 +150,7 @@ BEGIN(add_unless)
 	movl %edx, 4(v)
 	movl $1, %eax
 2:
-	RET
+	RET_IRQ_RESTORE
 3:
 	cmpl %edx, %edi
 	jne 1b

@@ -153,7 +160,7 @@ ENDP
 #undef v
 
 #define v %esi
-BEGIN(inc_not_zero)
+BEGIN_IRQ_SAVE(inc_not_zero)
 	movl  (v), %eax
 	movl 4(v), %edx
 	testl %eax, %eax

@@ -165,7 +172,7 @@ BEGIN(inc_not_zero)
 	movl %edx, 4(v)
 	movl $1, %eax
 2:
-	RET
+	RET_IRQ_RESTORE
 3:
 	testl %edx, %edx
 	jne 1b

@@ -174,7 +181,7 @@ ENDP
 #undef v
 
 #define v %esi
-BEGIN(dec_if_positive)
+BEGIN_IRQ_SAVE(dec_if_positive)
 	movl  (v), %eax
 	movl 4(v), %edx
 	subl $1, %eax

@@ -183,5 +190,6 @@ BEGIN(dec_if_positive)
 	movl %eax,  (v)
 	movl %edx, 4(v)
 1:
-RET_ENDP
+	RET_IRQ_RESTORE
+ENDP
 #undef v

Greg Kroah-Hartman

author maintainer

2022-05-13 14:30 UTC #2 lore raw

PATCH 2/21 x86: Prepare asm files for straight-line-speculation

From: Peter Zijlstra <[email protected]>

[ Upstream commit f94909ceb1ed4bfdb2ada72f93236305e6d6951f ]

Replace all ret/retq instructions with RET in preparation of making RET a macro. Since AS is case insensitive it's a big no-op without RET defined.

find arch/x86/ -name \*.S | while read file
do

	sed -i 's/\<ret[q]*\>/RET/' $file

done

Signed-off-by Peter Zijlstra (Intel) <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 arch/x86/boot/compressed/efi_thunk_64.S      |    2 
 arch/x86/boot/compressed/head_64.S           |    8 +--
 arch/x86/boot/compressed/mem_encrypt.S       |    6 +-
 arch/x86/crypto/aegis128-aesni-asm.S         |   48 +++++++++++------------
 arch/x86/crypto/aes_ctrby8_avx-x86_64.S      |    2 
 arch/x86/crypto/aesni-intel_asm.S            |   56 +++++++++++++--------------
 arch/x86/crypto/aesni-intel_avx-x86_64.S     |   40 +++++++++----------
 arch/x86/crypto/blake2s-core.S               |    4 -
 arch/x86/crypto/blowfish-x86_64-asm_64.S     |   12 ++---
 arch/x86/crypto/camellia-aesni-avx-asm_64.S  |   14 +++---
 arch/x86/crypto/camellia-aesni-avx2-asm_64.S |   14 +++---
 arch/x86/crypto/camellia-x86_64-asm_64.S     |   12 ++---
 arch/x86/crypto/cast5-avx-x86_64-asm_64.S    |   12 ++---
 arch/x86/crypto/cast6-avx-x86_64-asm_64.S    |   10 ++--
 arch/x86/crypto/chacha-avx2-x86_64.S         |    6 +-
 arch/x86/crypto/chacha-avx512vl-x86_64.S     |    6 +-
 arch/x86/crypto/chacha-ssse3-x86_64.S        |    8 +--
 arch/x86/crypto/crc32-pclmul_asm.S           |    2 
 arch/x86/crypto/crc32c-pcl-intel-asm_64.S    |    2 
 arch/x86/crypto/crct10dif-pcl-asm_64.S       |    2 
 arch/x86/crypto/des3_ede-asm_64.S            |    4 -
 arch/x86/crypto/ghash-clmulni-intel_asm.S    |    6 +-
 arch/x86/crypto/nh-avx2-x86_64.S             |    2 
 arch/x86/crypto/nh-sse2-x86_64.S             |    2 
 arch/x86/crypto/serpent-avx-x86_64-asm_64.S  |   10 ++--
 arch/x86/crypto/serpent-avx2-asm_64.S        |   10 ++--
 arch/x86/crypto/serpent-sse2-i586-asm_32.S   |    6 +-
 arch/x86/crypto/serpent-sse2-x86_64-asm_64.S |    6 +-
 arch/x86/crypto/sha1_avx2_x86_64_asm.S       |    2 
 arch/x86/crypto/sha1_ni_asm.S                |    2 
 arch/x86/crypto/sha1_ssse3_asm.S             |    2 
 arch/x86/crypto/sha256-avx-asm.S             |    2 
 arch/x86/crypto/sha256-avx2-asm.S            |    2 
 arch/x86/crypto/sha256-ssse3-asm.S           |    2 
 arch/x86/crypto/sha256_ni_asm.S              |    2 
 arch/x86/crypto/sha512-avx-asm.S             |    2 
 arch/x86/crypto/sha512-avx2-asm.S            |    2 
 arch/x86/crypto/sha512-ssse3-asm.S           |    2 
 arch/x86/crypto/sm4-aesni-avx-asm_64.S       |   12 ++---
 arch/x86/crypto/sm4-aesni-avx2-asm_64.S      |    8 +--
 arch/x86/crypto/twofish-avx-x86_64-asm_64.S  |   10 ++--
 arch/x86/crypto/twofish-i586-asm_32.S        |    4 -
 arch/x86/crypto/twofish-x86_64-asm_64-3way.S |    6 +-
 arch/x86/crypto/twofish-x86_64-asm_64.S      |    4 -
 arch/x86/entry/entry_32.S                    |    2 
 arch/x86/entry/entry_64.S                    |   10 ++--
 arch/x86/entry/thunk_32.S                    |    2 
 arch/x86/entry/thunk_64.S                    |    2 
 arch/x86/entry/vdso/vdso32/system_call.S     |    2 
 arch/x86/entry/vdso/vsgx.S                   |    2 
 arch/x86/entry/vsyscall/vsyscall_emu_64.S    |    6 +-
 arch/x86/kernel/acpi/wakeup_32.S             |    6 +-
 arch/x86/kernel/ftrace_32.S                  |    6 +-
 arch/x86/kernel/ftrace_64.S                  |   10 ++--
 arch/x86/kernel/head_32.S                    |    2 
 arch/x86/kernel/irqflags.S                   |    2 
 arch/x86/kernel/relocate_kernel_32.S         |   10 ++--
 arch/x86/kernel/relocate_kernel_64.S         |   10 ++--
 arch/x86/kernel/sev_verify_cbit.S            |    2 
 arch/x86/kernel/verify_cpu.S                 |    4 -
 arch/x86/kvm/svm/vmenter.S                   |    4 -
 arch/x86/kvm/vmx/vmenter.S                   |   14 +++---
 arch/x86/lib/atomic64_386_32.S               |    2 
 arch/x86/lib/atomic64_cx8_32.S               |   16 +++----
 arch/x86/lib/checksum_32.S                   |    8 +--
 arch/x86/lib/clear_page_64.S                 |    6 +-
 arch/x86/lib/cmpxchg16b_emu.S                |    4 -
 arch/x86/lib/cmpxchg8b_emu.S                 |    4 -
 arch/x86/lib/copy_mc_64.S                    |    6 +-
 arch/x86/lib/copy_page_64.S                  |    4 -
 arch/x86/lib/copy_user_64.S                  |   12 ++---
 arch/x86/lib/csum-copy_64.S                  |    2 
 arch/x86/lib/getuser.S                       |   22 +++++-----
 arch/x86/lib/hweight.S                       |    6 +-
 arch/x86/lib/iomap_copy_64.S                 |    2 
 arch/x86/lib/memcpy_64.S                     |   12 ++---
 arch/x86/lib/memmove_64.S                    |    4 -
 arch/x86/lib/memset_64.S                     |    6 +-
 arch/x86/lib/msr-reg.S                       |    4 -
 arch/x86/lib/putuser.S                       |    6 +-
 arch/x86/lib/retpoline.S                     |    2 
 arch/x86/math-emu/div_Xsig.S                 |    2 
 arch/x86/math-emu/div_small.S                |    2 
 arch/x86/math-emu/mul_Xsig.S                 |    6 +-
 arch/x86/math-emu/polynom_Xsig.S             |    2 
 arch/x86/math-emu/reg_norm.S                 |    6 +-
 arch/x86/math-emu/reg_round.S                |    2 
 arch/x86/math-emu/reg_u_add.S                |    2 
 arch/x86/math-emu/reg_u_div.S                |    2 
 arch/x86/math-emu/reg_u_mul.S                |    2 
 arch/x86/math-emu/reg_u_sub.S                |    2 
 arch/x86/math-emu/round_Xsig.S               |    4 -
 arch/x86/math-emu/shr_Xsig.S                 |    8 +--
 arch/x86/math-emu/wm_shrx.S                  |   16 +++----
 arch/x86/mm/mem_encrypt_boot.S               |    4 -
 arch/x86/platform/efi/efi_stub_32.S          |    2 
 arch/x86/platform/efi/efi_stub_64.S          |    2 
 arch/x86/platform/efi/efi_thunk_64.S         |    2 
 arch/x86/platform/olpc/xo1-wakeup.S          |    6 +-
 arch/x86/power/hibernate_asm_32.S            |    4 -
 arch/x86/power/hibernate_asm_64.S            |    4 -
 arch/x86/um/checksum_32.S                    |    4 -
 arch/x86/um/setjmp_32.S                      |    2 
 arch/x86/um/setjmp_64.S                      |    2 
 arch/x86/xen/xen-asm.S                       |   12 ++---
 arch/x86/xen/xen-head.S                      |    2 
 106 files changed, 351 insertions(+), 351 deletions(-)

arch/x86/boot/compressed/efi_thunk_64.S +1 -1

@@ -93,7 +93,7 @@ SYM_FUNC_START(__efi64_thunk)
 
 	pop	%rbx
 	pop	%rbp
-	ret
+	RET
 SYM_FUNC_END(__efi64_thunk)
 
 	.code32

arch/x86/boot/compressed/head_64.S +4 -4

@@ -813,7 +813,7 @@ SYM_FUNC_START(efi32_pe_entry)
 2:	popl	%edi				// restore callee-save registers
 	popl	%ebx
 	leave
-	ret
+	RET
 SYM_FUNC_END(efi32_pe_entry)
 
 	.section ".rodata"

@@ -868,7 +868,7 @@ SYM_FUNC_START(startup32_set_idt_entry)
 
 	pop     %ecx
 	pop     %ebx
-	ret
+	RET
 SYM_FUNC_END(startup32_set_idt_entry)
 #endif

@@ -884,7 +884,7 @@ SYM_FUNC_START(startup32_load_idt)
 	movl	%eax, rva(boot32_idt_desc+2)(%ebp)
 	lidt    rva(boot32_idt_desc)(%ebp)
 #endif
-	ret
+	RET
 SYM_FUNC_END(startup32_load_idt)
 
 /*

@@ -954,7 +954,7 @@ SYM_FUNC_START(startup32_check_sev_cbit)
 	popl	%ebx
 	popl	%eax
 #endif
-	ret
+	RET
 SYM_FUNC_END(startup32_check_sev_cbit)
 
 /*

arch/x86/boot/compressed/mem_encrypt.S +3 -3

@@ -58,7 +58,7 @@ SYM_FUNC_START(get_sev_encryption_bit)
 
 #endif	/* CONFIG_AMD_MEM_ENCRYPT */
 
-	ret
+	RET
 SYM_FUNC_END(get_sev_encryption_bit)
 
 /**

@@ -92,7 +92,7 @@ SYM_CODE_START_LOCAL(sev_es_req_cpuid)
 	/* All good - return success */
 	xorl	%eax, %eax
 1:
-	ret
+	RET
 2:
 	movl	$-1, %eax
 	jmp	1b

@@ -221,7 +221,7 @@ SYM_FUNC_START(set_sev_encryption_mask)
 #endif
 
 	xor	%rax, %rax
-	ret
+	RET
 SYM_FUNC_END(set_sev_encryption_mask)
 
 	.data

arch/x86/crypto/aegis128-aesni-asm.S +24 ~~-24~~

181 lines not on this page · raw message

arch/x86/crypto/aes_ctrby8_avx-x86_64.S +1 -1

@@ -525,7 +525,7 @@ ddq_add_8:
 	/* return updated IV */
 	vpshufb	xbyteswap, xcounter, xcounter
 	vmovdqu	xcounter, (p_iv)
-	ret
+	RET
 .endm
 
 /*

arch/x86/crypto/aesni-intel_asm.S +28 ~~-28~~

224 lines not on this page · raw message

arch/x86/crypto/aesni-intel_avx-x86_64.S +20 ~~-20~~

@@ -1767,7 +1767,7 @@ SYM_FUNC_START(aesni_gcm_init_avx_gen2)
         FUNC_SAVE
         INIT GHASH_MUL_AVX, PRECOMPUTE_AVX
         FUNC_RESTORE
-        ret
+        RET
 SYM_FUNC_END(aesni_gcm_init_avx_gen2)
 
 ###############################################################################

@@ -1788,15 +1788,15 @@ SYM_FUNC_START(aesni_gcm_enc_update_avx_
         # must be 192
         GCM_ENC_DEC INITIAL_BLOCKS_AVX, GHASH_8_ENCRYPT_8_PARALLEL_AVX, GHASH_LAST_8_AVX, GHASH_MUL_AVX, ENC, 11
         FUNC_RESTORE
-        ret
+        RET
 key_128_enc_update:
         GCM_ENC_DEC INITIAL_BLOCKS_AVX, GHASH_8_ENCRYPT_8_PARALLEL_AVX, GHASH_LAST_8_AVX, GHASH_MUL_AVX, ENC, 9
         FUNC_RESTORE
-        ret
+        RET
 key_256_enc_update:
         GCM_ENC_DEC INITIAL_BLOCKS_AVX, GHASH_8_ENCRYPT_8_PARALLEL_AVX, GHASH_LAST_8_AVX, GHASH_MUL_AVX, ENC, 13
         FUNC_RESTORE
-        ret
+        RET
 SYM_FUNC_END(aesni_gcm_enc_update_avx_gen2)
 
 ###############################################################################

@@ -1817,15 +1817,15 @@ SYM_FUNC_START(aesni_gcm_dec_update_avx_
         # must be 192
         GCM_ENC_DEC INITIAL_BLOCKS_AVX, GHASH_8_ENCRYPT_8_PARALLEL_AVX, GHASH_LAST_8_AVX, GHASH_MUL_AVX, DEC, 11
         FUNC_RESTORE
-        ret
+        RET
 key_128_dec_update:
         GCM_ENC_DEC INITIAL_BLOCKS_AVX, GHASH_8_ENCRYPT_8_PARALLEL_AVX, GHASH_LAST_8_AVX, GHASH_MUL_AVX, DEC, 9
         FUNC_RESTORE
-        ret
+        RET
 key_256_dec_update:
         GCM_ENC_DEC INITIAL_BLOCKS_AVX, GHASH_8_ENCRYPT_8_PARALLEL_AVX, GHASH_LAST_8_AVX, GHASH_MUL_AVX, DEC, 13
         FUNC_RESTORE
-        ret
+        RET
 SYM_FUNC_END(aesni_gcm_dec_update_avx_gen2)
 
 ###############################################################################

@@ -1846,15 +1846,15 @@ SYM_FUNC_START(aesni_gcm_finalize_avx_ge
         # must be 192
         GCM_COMPLETE GHASH_MUL_AVX, 11, arg3, arg4
         FUNC_RESTORE
-        ret
+        RET
 key_128_finalize:
         GCM_COMPLETE GHASH_MUL_AVX, 9, arg3, arg4
         FUNC_RESTORE
-        ret
+        RET
 key_256_finalize:
         GCM_COMPLETE GHASH_MUL_AVX, 13, arg3, arg4
         FUNC_RESTORE
-        ret
+        RET
 SYM_FUNC_END(aesni_gcm_finalize_avx_gen2)
 
 ###############################################################################

@@ -2735,7 +2735,7 @@ SYM_FUNC_START(aesni_gcm_init_avx_gen4)
         FUNC_SAVE
         INIT GHASH_MUL_AVX2, PRECOMPUTE_AVX2
         FUNC_RESTORE
-        ret
+        RET
 SYM_FUNC_END(aesni_gcm_init_avx_gen4)
 
 ###############################################################################

@@ -2756,15 +2756,15 @@ SYM_FUNC_START(aesni_gcm_enc_update_avx_
         # must be 192
         GCM_ENC_DEC INITIAL_BLOCKS_AVX2, GHASH_8_ENCRYPT_8_PARALLEL_AVX2, GHASH_LAST_8_AVX2, GHASH_MUL_AVX2, ENC, 11
         FUNC_RESTORE
-	ret
+	RET
 key_128_enc_update4:
         GCM_ENC_DEC INITIAL_BLOCKS_AVX2, GHASH_8_ENCRYPT_8_PARALLEL_AVX2, GHASH_LAST_8_AVX2, GHASH_MUL_AVX2, ENC, 9
         FUNC_RESTORE
-	ret
+	RET
 key_256_enc_update4:
         GCM_ENC_DEC INITIAL_BLOCKS_AVX2, GHASH_8_ENCRYPT_8_PARALLEL_AVX2, GHASH_LAST_8_AVX2, GHASH_MUL_AVX2, ENC, 13
         FUNC_RESTORE
-	ret
+	RET
 SYM_FUNC_END(aesni_gcm_enc_update_avx_gen4)
 
 ###############################################################################

@@ -2785,15 +2785,15 @@ SYM_FUNC_START(aesni_gcm_dec_update_avx_
         # must be 192
         GCM_ENC_DEC INITIAL_BLOCKS_AVX2, GHASH_8_ENCRYPT_8_PARALLEL_AVX2, GHASH_LAST_8_AVX2, GHASH_MUL_AVX2, DEC, 11
         FUNC_RESTORE
-        ret
+        RET
 key_128_dec_update4:
         GCM_ENC_DEC INITIAL_BLOCKS_AVX2, GHASH_8_ENCRYPT_8_PARALLEL_AVX2, GHASH_LAST_8_AVX2, GHASH_MUL_AVX2, DEC, 9
         FUNC_RESTORE
-        ret
+        RET
 key_256_dec_update4:
         GCM_ENC_DEC INITIAL_BLOCKS_AVX2, GHASH_8_ENCRYPT_8_PARALLEL_AVX2, GHASH_LAST_8_AVX2, GHASH_MUL_AVX2, DEC, 13
         FUNC_RESTORE
-        ret
+        RET
 SYM_FUNC_END(aesni_gcm_dec_update_avx_gen4)
 
 ###############################################################################

@@ -2814,13 +2814,13 @@ SYM_FUNC_START(aesni_gcm_finalize_avx_ge
         # must be 192
         GCM_COMPLETE GHASH_MUL_AVX2, 11, arg3, arg4
         FUNC_RESTORE
-        ret
+        RET
 key_128_finalize4:
         GCM_COMPLETE GHASH_MUL_AVX2, 9, arg3, arg4
         FUNC_RESTORE
-        ret
+        RET
 key_256_finalize4:
         GCM_COMPLETE GHASH_MUL_AVX2, 13, arg3, arg4
         FUNC_RESTORE
-        ret
+        RET
 SYM_FUNC_END(aesni_gcm_finalize_avx_gen4)

arch/x86/crypto/blake2s-core.S +2 -2

@@ -171,7 +171,7 @@ SYM_FUNC_START(blake2s_compress_ssse3)
 	movdqu		%xmm1,0x10(%rdi)
 	movdqu		%xmm14,0x20(%rdi)
 .Lendofloop:
-	ret
+	RET
 SYM_FUNC_END(blake2s_compress_ssse3)
 
 #ifdef CONFIG_AS_AVX512

@@ -251,6 +251,6 @@ SYM_FUNC_START(blake2s_compress_avx512)
 	vmovdqu		%xmm1,0x10(%rdi)
 	vmovdqu		%xmm4,0x20(%rdi)
 	vzeroupper
-	retq
+	RET
 SYM_FUNC_END(blake2s_compress_avx512)
 #endif /* CONFIG_AS_AVX512 */

arch/x86/crypto/blowfish-x86_64-asm_64.S +6 -6

@@ -135,10 +135,10 @@ SYM_FUNC_START(__blowfish_enc_blk)
 	jnz .L__enc_xor;
 
 	write_block();
-	ret;
+	RET;
 .L__enc_xor:
 	xor_block();
-	ret;
+	RET;
 SYM_FUNC_END(__blowfish_enc_blk)
 
 SYM_FUNC_START(blowfish_dec_blk)

@@ -170,7 +170,7 @@ SYM_FUNC_START(blowfish_dec_blk)
 
 	movq %r11, %r12;
 
-	ret;
+	RET;
 SYM_FUNC_END(blowfish_dec_blk)
 
 /**********************************************************************

@@ -322,14 +322,14 @@ SYM_FUNC_START(__blowfish_enc_blk_4way)
 
 	popq %rbx;
 	popq %r12;
-	ret;
+	RET;
 
 .L__enc_xor4:
 	xor_block4();
 
 	popq %rbx;
 	popq %r12;
-	ret;
+	RET;
 SYM_FUNC_END(__blowfish_enc_blk_4way)
 
 SYM_FUNC_START(blowfish_dec_blk_4way)

@@ -364,5 +364,5 @@ SYM_FUNC_START(blowfish_dec_blk_4way)
 	popq %rbx;
 	popq %r12;
 
-	ret;
+	RET;
 SYM_FUNC_END(blowfish_dec_blk_4way)

arch/x86/crypto/camellia-aesni-avx-asm_64.S +7 -7

@@ -192,7 +192,7 @@ SYM_FUNC_START_LOCAL(roundsm16_x0_x1_x2_
 	roundsm16(%xmm0, %xmm1, %xmm2, %xmm3, %xmm4, %xmm5, %xmm6, %xmm7,
 		  %xmm8, %xmm9, %xmm10, %xmm11, %xmm12, %xmm13, %xmm14, %xmm15,
 		  %rcx, (%r9));
-	ret;
+	RET;
 SYM_FUNC_END(roundsm16_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd)
 
 .align 8

@@ -200,7 +200,7 @@ SYM_FUNC_START_LOCAL(roundsm16_x4_x5_x6_
 	roundsm16(%xmm4, %xmm5, %xmm6, %xmm7, %xmm0, %xmm1, %xmm2, %xmm3,
 		  %xmm12, %xmm13, %xmm14, %xmm15, %xmm8, %xmm9, %xmm10, %xmm11,
 		  %rax, (%r9));
-	ret;
+	RET;
 SYM_FUNC_END(roundsm16_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
 
 /*

@@ -778,7 +778,7 @@ SYM_FUNC_START_LOCAL(__camellia_enc_blk1
 		    %xmm15, (key_table)(CTX, %r8, 8), (%rax), 1 * 16(%rax));
 
 	FRAME_END
-	ret;
+	RET;
 
 .align 8
 .Lenc_max32:

@@ -865,7 +865,7 @@ SYM_FUNC_START_LOCAL(__camellia_dec_blk1
 		    %xmm15, (key_table)(CTX), (%rax), 1 * 16(%rax));
 
 	FRAME_END
-	ret;
+	RET;
 
 .align 8
 .Ldec_max32:

@@ -906,7 +906,7 @@ SYM_FUNC_START(camellia_ecb_enc_16way)
 		     %xmm8, %rsi);
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(camellia_ecb_enc_16way)
 
 SYM_FUNC_START(camellia_ecb_dec_16way)

@@ -936,7 +936,7 @@ SYM_FUNC_START(camellia_ecb_dec_16way)
 		     %xmm8, %rsi);
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(camellia_ecb_dec_16way)
 
 SYM_FUNC_START(camellia_cbc_dec_16way)

@@ -987,5 +987,5 @@ SYM_FUNC_START(camellia_cbc_dec_16way)
 		     %xmm8, %rsi);
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(camellia_cbc_dec_16way)

arch/x86/crypto/camellia-aesni-avx2-asm_64.S +7 -7

@@ -226,7 +226,7 @@ SYM_FUNC_START_LOCAL(roundsm32_x0_x1_x2_
 	roundsm32(%ymm0, %ymm1, %ymm2, %ymm3, %ymm4, %ymm5, %ymm6, %ymm7,
 		  %ymm8, %ymm9, %ymm10, %ymm11, %ymm12, %ymm13, %ymm14, %ymm15,
 		  %rcx, (%r9));
-	ret;
+	RET;
 SYM_FUNC_END(roundsm32_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd)
 
 .align 8

@@ -234,7 +234,7 @@ SYM_FUNC_START_LOCAL(roundsm32_x4_x5_x6_
 	roundsm32(%ymm4, %ymm5, %ymm6, %ymm7, %ymm0, %ymm1, %ymm2, %ymm3,
 		  %ymm12, %ymm13, %ymm14, %ymm15, %ymm8, %ymm9, %ymm10, %ymm11,
 		  %rax, (%r9));
-	ret;
+	RET;
 SYM_FUNC_END(roundsm32_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
 
 /*

@@ -814,7 +814,7 @@ SYM_FUNC_START_LOCAL(__camellia_enc_blk3
 		    %ymm15, (key_table)(CTX, %r8, 8), (%rax), 1 * 32(%rax));
 
 	FRAME_END
-	ret;
+	RET;
 
 .align 8
 .Lenc_max32:

@@ -901,7 +901,7 @@ SYM_FUNC_START_LOCAL(__camellia_dec_blk3
 		    %ymm15, (key_table)(CTX), (%rax), 1 * 32(%rax));
 
 	FRAME_END
-	ret;
+	RET;
 
 .align 8
 .Ldec_max32:

@@ -946,7 +946,7 @@ SYM_FUNC_START(camellia_ecb_enc_32way)
 	vzeroupper;
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(camellia_ecb_enc_32way)
 
 SYM_FUNC_START(camellia_ecb_dec_32way)

@@ -980,7 +980,7 @@ SYM_FUNC_START(camellia_ecb_dec_32way)
 	vzeroupper;
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(camellia_ecb_dec_32way)
 
 SYM_FUNC_START(camellia_cbc_dec_32way)

@@ -1047,5 +1047,5 @@ SYM_FUNC_START(camellia_cbc_dec_32way)
 
 	addq $(16 * 32), %rsp;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(camellia_cbc_dec_32way)

arch/x86/crypto/camellia-x86_64-asm_64.S +6 -6

@@ -213,13 +213,13 @@ SYM_FUNC_START(__camellia_enc_blk)
 	enc_outunpack(mov, RT1);
 
 	movq RR12, %r12;
-	ret;
+	RET;
 
 .L__enc_xor:
 	enc_outunpack(xor, RT1);
 
 	movq RR12, %r12;
-	ret;
+	RET;
 SYM_FUNC_END(__camellia_enc_blk)
 
 SYM_FUNC_START(camellia_dec_blk)

@@ -257,7 +257,7 @@ SYM_FUNC_START(camellia_dec_blk)
 	dec_outunpack();
 
 	movq RR12, %r12;
-	ret;
+	RET;
 SYM_FUNC_END(camellia_dec_blk)
 
 /**********************************************************************

@@ -448,14 +448,14 @@ SYM_FUNC_START(__camellia_enc_blk_2way)
 
 	movq RR12, %r12;
 	popq %rbx;
-	ret;
+	RET;
 
 .L__enc2_xor:
 	enc_outunpack2(xor, RT2);
 
 	movq RR12, %r12;
 	popq %rbx;
-	ret;
+	RET;
 SYM_FUNC_END(__camellia_enc_blk_2way)
 
 SYM_FUNC_START(camellia_dec_blk_2way)

@@ -495,5 +495,5 @@ SYM_FUNC_START(camellia_dec_blk_2way)
 
 	movq RR12, %r12;
 	movq RXOR, %rbx;
-	ret;
+	RET;
 SYM_FUNC_END(camellia_dec_blk_2way)

arch/x86/crypto/cast5-avx-x86_64-asm_64.S +6 -6

@@ -279,7 +279,7 @@ SYM_FUNC_START_LOCAL(__cast5_enc_blk16)
 	outunpack_blocks(RR3, RL3, RTMP, RX, RKM);
 	outunpack_blocks(RR4, RL4, RTMP, RX, RKM);
 
-	ret;
+	RET;
 SYM_FUNC_END(__cast5_enc_blk16)
 
 .align 16

@@ -352,7 +352,7 @@ SYM_FUNC_START_LOCAL(__cast5_dec_blk16)
 	outunpack_blocks(RR3, RL3, RTMP, RX, RKM);
 	outunpack_blocks(RR4, RL4, RTMP, RX, RKM);
 
-	ret;
+	RET;
 
 .L__skip_dec:
 	vpsrldq $4, RKR, RKR;

@@ -393,7 +393,7 @@ SYM_FUNC_START(cast5_ecb_enc_16way)
 
 	popq %r15;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(cast5_ecb_enc_16way)
 
 SYM_FUNC_START(cast5_ecb_dec_16way)

@@ -431,7 +431,7 @@ SYM_FUNC_START(cast5_ecb_dec_16way)
 
 	popq %r15;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(cast5_ecb_dec_16way)
 
 SYM_FUNC_START(cast5_cbc_dec_16way)

@@ -483,7 +483,7 @@ SYM_FUNC_START(cast5_cbc_dec_16way)
 	popq %r15;
 	popq %r12;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(cast5_cbc_dec_16way)
 
 SYM_FUNC_START(cast5_ctr_16way)

@@ -559,5 +559,5 @@ SYM_FUNC_START(cast5_ctr_16way)
 	popq %r15;
 	popq %r12;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(cast5_ctr_16way)

arch/x86/crypto/cast6-avx-x86_64-asm_64.S +5 -5

@@ -289,7 +289,7 @@ SYM_FUNC_START_LOCAL(__cast6_enc_blk8)
 	outunpack_blocks(RA1, RB1, RC1, RD1, RTMP, RX, RKRF, RKM);
 	outunpack_blocks(RA2, RB2, RC2, RD2, RTMP, RX, RKRF, RKM);
 
-	ret;
+	RET;
 SYM_FUNC_END(__cast6_enc_blk8)
 
 .align 8

@@ -336,7 +336,7 @@ SYM_FUNC_START_LOCAL(__cast6_dec_blk8)
 	outunpack_blocks(RA1, RB1, RC1, RD1, RTMP, RX, RKRF, RKM);
 	outunpack_blocks(RA2, RB2, RC2, RD2, RTMP, RX, RKRF, RKM);
 
-	ret;
+	RET;
 SYM_FUNC_END(__cast6_dec_blk8)
 
 SYM_FUNC_START(cast6_ecb_enc_8way)

@@ -359,7 +359,7 @@ SYM_FUNC_START(cast6_ecb_enc_8way)
 
 	popq %r15;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(cast6_ecb_enc_8way)
 
 SYM_FUNC_START(cast6_ecb_dec_8way)

@@ -382,7 +382,7 @@ SYM_FUNC_START(cast6_ecb_dec_8way)
 
 	popq %r15;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(cast6_ecb_dec_8way)
 
 SYM_FUNC_START(cast6_cbc_dec_8way)

@@ -408,5 +408,5 @@ SYM_FUNC_START(cast6_cbc_dec_8way)
 	popq %r15;
 	popq %r12;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(cast6_cbc_dec_8way)

arch/x86/crypto/chacha-avx2-x86_64.S +3 -3

@@ -193,7 +193,7 @@ SYM_FUNC_START(chacha_2block_xor_avx2)
 
 .Ldone2:
 	vzeroupper
-	ret
+	RET
 
 .Lxorpart2:
 	# xor remaining bytes from partial register into output

@@ -498,7 +498,7 @@ SYM_FUNC_START(chacha_4block_xor_avx2)
 
 .Ldone4:
 	vzeroupper
-	ret
+	RET
 
 .Lxorpart4:
 	# xor remaining bytes from partial register into output

@@ -992,7 +992,7 @@ SYM_FUNC_START(chacha_8block_xor_avx2)
 .Ldone8:
 	vzeroupper
 	lea		-8(%r10),%rsp
-	ret
+	RET
 
 .Lxorpart8:
 	# xor remaining bytes from partial register into output

arch/x86/crypto/chacha-avx512vl-x86_64.S +3 -3

@@ -166,7 +166,7 @@ SYM_FUNC_START(chacha_2block_xor_avx512v
 
 .Ldone2:
 	vzeroupper
-	ret
+	RET
 
 .Lxorpart2:
 	# xor remaining bytes from partial register into output

@@ -432,7 +432,7 @@ SYM_FUNC_START(chacha_4block_xor_avx512v
 
 .Ldone4:
 	vzeroupper
-	ret
+	RET
 
 .Lxorpart4:
 	# xor remaining bytes from partial register into output

@@ -812,7 +812,7 @@ SYM_FUNC_START(chacha_8block_xor_avx512v
 
 .Ldone8:
 	vzeroupper
-	ret
+	RET
 
 .Lxorpart8:
 	# xor remaining bytes from partial register into output

arch/x86/crypto/chacha-ssse3-x86_64.S +4 -4

@@ -108,7 +108,7 @@ SYM_FUNC_START_LOCAL(chacha_permute)
 	sub		$2,%r8d
 	jnz		.Ldoubleround
 
-	ret
+	RET
 SYM_FUNC_END(chacha_permute)
 
 SYM_FUNC_START(chacha_block_xor_ssse3)

@@ -166,7 +166,7 @@ SYM_FUNC_START(chacha_block_xor_ssse3)
 
 .Ldone:
 	FRAME_END
-	ret
+	RET
 
 .Lxorpart:
 	# xor remaining bytes from partial register into output

@@ -217,7 +217,7 @@ SYM_FUNC_START(hchacha_block_ssse3)
 	movdqu		%xmm3,0x10(%rsi)
 
 	FRAME_END
-	ret
+	RET
 SYM_FUNC_END(hchacha_block_ssse3)
 
 SYM_FUNC_START(chacha_4block_xor_ssse3)

@@ -762,7 +762,7 @@ SYM_FUNC_START(chacha_4block_xor_ssse3)
 
 .Ldone4:
 	lea		-8(%r10),%rsp
-	ret
+	RET
 
 .Lxorpart4:
 	# xor remaining bytes from partial register into output

arch/x86/crypto/crc32-pclmul_asm.S +1 -1

@@ -236,5 +236,5 @@ fold_64:
 	pxor    %xmm2, %xmm1
 	pextrd  $0x01, %xmm1, %eax
 
-	ret
+	RET
 SYM_FUNC_END(crc32_pclmul_le_16)

arch/x86/crypto/crc32c-pcl-intel-asm_64.S +1 -1

@@ -306,7 +306,7 @@ do_return:
 	popq    %rsi
 	popq    %rdi
 	popq    %rbx
-        ret
+        RET
 SYM_FUNC_END(crc_pcl)
 
 .section	.rodata, "a", @progbits

arch/x86/crypto/crct10dif-pcl-asm_64.S +1 -1

@@ -257,7 +257,7 @@ SYM_FUNC_START(crc_t10dif_pcl)
 	# Final CRC value (x^16 * M(x)) mod G(x) is in low 16 bits of xmm0.
 
 	pextrw	$0, %xmm0, %eax
-	ret
+	RET
 
 .align 16
 .Lless_than_256_bytes:

arch/x86/crypto/des3_ede-asm_64.S +2 -2

@@ -243,7 +243,7 @@ SYM_FUNC_START(des3_ede_x86_64_crypt_blk
 	popq %r12;
 	popq %rbx;
 
-	ret;
+	RET;
 SYM_FUNC_END(des3_ede_x86_64_crypt_blk)
 
 /***********************************************************************

@@ -528,7 +528,7 @@ SYM_FUNC_START(des3_ede_x86_64_crypt_blk
 	popq %r12;
 	popq %rbx;
 
-	ret;
+	RET;
 SYM_FUNC_END(des3_ede_x86_64_crypt_blk_3way)
 
 .section	.rodata, "a", @progbits

arch/x86/crypto/ghash-clmulni-intel_asm.S +3 -3

@@ -85,7 +85,7 @@ SYM_FUNC_START_LOCAL(__clmul_gf128mul_bl
 	psrlq $1, T2
 	pxor T2, T1
 	pxor T1, DATA
-	ret
+	RET
 SYM_FUNC_END(__clmul_gf128mul_ble)
 
 /* void clmul_ghash_mul(char *dst, const u128 *shash) */

@@ -99,7 +99,7 @@ SYM_FUNC_START(clmul_ghash_mul)
 	pshufb BSWAP, DATA
 	movups DATA, (%rdi)
 	FRAME_END
-	ret
+	RET
 SYM_FUNC_END(clmul_ghash_mul)
 
 /*

@@ -128,5 +128,5 @@ SYM_FUNC_START(clmul_ghash_update)
 	movups DATA, (%rdi)
 .Lupdate_just_ret:
 	FRAME_END
-	ret
+	RET
 SYM_FUNC_END(clmul_ghash_update)

arch/x86/crypto/nh-avx2-x86_64.S +1 -1

@@ -153,5 +153,5 @@ SYM_FUNC_START(nh_avx2)
 	vpaddq		T1, T0, T0
 	vpaddq		T4, T0, T0
 	vmovdqu		T0, (HASH)
-	ret
+	RET
 SYM_FUNC_END(nh_avx2)

arch/x86/crypto/nh-sse2-x86_64.S +1 -1

@@ -119,5 +119,5 @@ SYM_FUNC_START(nh_sse2)
 	paddq		PASS2_SUMS, T1
 	movdqu		T0, 0x00(HASH)
 	movdqu		T1, 0x10(HASH)
-	ret
+	RET
 SYM_FUNC_END(nh_sse2)

arch/x86/crypto/serpent-avx-x86_64-asm_64.S +5 -5

@@ -601,7 +601,7 @@ SYM_FUNC_START_LOCAL(__serpent_enc_blk8_
 	write_blocks(RA1, RB1, RC1, RD1, RK0, RK1, RK2);
 	write_blocks(RA2, RB2, RC2, RD2, RK0, RK1, RK2);
 
-	ret;
+	RET;
 SYM_FUNC_END(__serpent_enc_blk8_avx)
 
 .align 8

@@ -655,7 +655,7 @@ SYM_FUNC_START_LOCAL(__serpent_dec_blk8_
 	write_blocks(RC1, RD1, RB1, RE1, RK0, RK1, RK2);
 	write_blocks(RC2, RD2, RB2, RE2, RK0, RK1, RK2);
 
-	ret;
+	RET;
 SYM_FUNC_END(__serpent_dec_blk8_avx)
 
 SYM_FUNC_START(serpent_ecb_enc_8way_avx)

@@ -673,7 +673,7 @@ SYM_FUNC_START(serpent_ecb_enc_8way_avx)
 	store_8way(%rsi, RA1, RB1, RC1, RD1, RA2, RB2, RC2, RD2);
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(serpent_ecb_enc_8way_avx)
 
 SYM_FUNC_START(serpent_ecb_dec_8way_avx)

@@ -691,7 +691,7 @@ SYM_FUNC_START(serpent_ecb_dec_8way_avx)
 	store_8way(%rsi, RC1, RD1, RB1, RE1, RC2, RD2, RB2, RE2);
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(serpent_ecb_dec_8way_avx)
 
 SYM_FUNC_START(serpent_cbc_dec_8way_avx)

@@ -709,5 +709,5 @@ SYM_FUNC_START(serpent_cbc_dec_8way_avx)
 	store_cbc_8way(%rdx, %rsi, RC1, RD1, RB1, RE1, RC2, RD2, RB2, RE2);
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(serpent_cbc_dec_8way_avx)

arch/x86/crypto/serpent-avx2-asm_64.S +5 -5

@@ -601,7 +601,7 @@ SYM_FUNC_START_LOCAL(__serpent_enc_blk16
 	write_blocks(RA1, RB1, RC1, RD1, RK0, RK1, RK2);
 	write_blocks(RA2, RB2, RC2, RD2, RK0, RK1, RK2);
 
-	ret;
+	RET;
 SYM_FUNC_END(__serpent_enc_blk16)
 
 .align 8

@@ -655,7 +655,7 @@ SYM_FUNC_START_LOCAL(__serpent_dec_blk16
 	write_blocks(RC1, RD1, RB1, RE1, RK0, RK1, RK2);
 	write_blocks(RC2, RD2, RB2, RE2, RK0, RK1, RK2);
 
-	ret;
+	RET;
 SYM_FUNC_END(__serpent_dec_blk16)
 
 SYM_FUNC_START(serpent_ecb_enc_16way)

@@ -677,7 +677,7 @@ SYM_FUNC_START(serpent_ecb_enc_16way)
 	vzeroupper;
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(serpent_ecb_enc_16way)
 
 SYM_FUNC_START(serpent_ecb_dec_16way)

@@ -699,7 +699,7 @@ SYM_FUNC_START(serpent_ecb_dec_16way)
 	vzeroupper;
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(serpent_ecb_dec_16way)
 
 SYM_FUNC_START(serpent_cbc_dec_16way)

@@ -722,5 +722,5 @@ SYM_FUNC_START(serpent_cbc_dec_16way)
 	vzeroupper;
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(serpent_cbc_dec_16way)

arch/x86/crypto/serpent-sse2-i586-asm_32.S +3 -3

@@ -553,12 +553,12 @@ SYM_FUNC_START(__serpent_enc_blk_4way)
 
 	write_blocks(%eax, RA, RB, RC, RD, RT0, RT1, RE);
 
-	ret;
+	RET;
 
 .L__enc_xor4:
 	xor_blocks(%eax, RA, RB, RC, RD, RT0, RT1, RE);
 
-	ret;
+	RET;
 SYM_FUNC_END(__serpent_enc_blk_4way)
 
 SYM_FUNC_START(serpent_dec_blk_4way)

@@ -612,5 +612,5 @@ SYM_FUNC_START(serpent_dec_blk_4way)
 	movl arg_dst(%esp), %eax;
 	write_blocks(%eax, RC, RD, RB, RE, RT0, RT1, RA);
 
-	ret;
+	RET;
 SYM_FUNC_END(serpent_dec_blk_4way)

arch/x86/crypto/serpent-sse2-x86_64-asm_64.S +3 -3

@@ -675,13 +675,13 @@ SYM_FUNC_START(__serpent_enc_blk_8way)
 	write_blocks(%rsi, RA1, RB1, RC1, RD1, RK0, RK1, RK2);
 	write_blocks(%rax, RA2, RB2, RC2, RD2, RK0, RK1, RK2);
 
-	ret;
+	RET;
 
 .L__enc_xor8:
 	xor_blocks(%rsi, RA1, RB1, RC1, RD1, RK0, RK1, RK2);
 	xor_blocks(%rax, RA2, RB2, RC2, RD2, RK0, RK1, RK2);
 
-	ret;
+	RET;
 SYM_FUNC_END(__serpent_enc_blk_8way)
 
 SYM_FUNC_START(serpent_dec_blk_8way)

@@ -735,5 +735,5 @@ SYM_FUNC_START(serpent_dec_blk_8way)
 	write_blocks(%rsi, RC1, RD1, RB1, RE1, RK0, RK1, RK2);
 	write_blocks(%rax, RC2, RD2, RB2, RE2, RK0, RK1, RK2);
 
-	ret;
+	RET;
 SYM_FUNC_END(serpent_dec_blk_8way)

arch/x86/crypto/sha1_avx2_x86_64_asm.S +1 -1

@@ -674,7 +674,7 @@ _loop3:
 	pop	%r12
 	pop	%rbx
 
-	ret
+	RET
 
 	SYM_FUNC_END(\name)
 .endm

arch/x86/crypto/sha1_ni_asm.S +1 -1

@@ -290,7 +290,7 @@ SYM_FUNC_START(sha1_ni_transform)
 	mov		%rbp, %rsp
 	pop		%rbp
 
-	ret
+	RET
 SYM_FUNC_END(sha1_ni_transform)
 
 .section	.rodata.cst16.PSHUFFLE_BYTE_FLIP_MASK, "aM", @progbits, 16

arch/x86/crypto/sha1_ssse3_asm.S +1 -1

@@ -99,7 +99,7 @@
 	pop	%rbp
 	pop	%r12
 	pop	%rbx
-	ret
+	RET
 
 	SYM_FUNC_END(\name)
 .endm

arch/x86/crypto/sha256-avx-asm.S +1 -1

@@ -458,7 +458,7 @@ done_hash:
 	popq    %r13
 	popq	%r12
 	popq    %rbx
-	ret
+	RET
 SYM_FUNC_END(sha256_transform_avx)
 
 .section	.rodata.cst256.K256, "aM", @progbits, 256

arch/x86/crypto/sha256-avx2-asm.S +1 -1

@@ -710,7 +710,7 @@ done_hash:
 	popq	%r13
 	popq	%r12
 	popq	%rbx
-	ret
+	RET
 SYM_FUNC_END(sha256_transform_rorx)
 
 .section	.rodata.cst512.K256, "aM", @progbits, 512

arch/x86/crypto/sha256-ssse3-asm.S +1 -1

@@ -472,7 +472,7 @@ done_hash:
 	popq    %r12
 	popq    %rbx
 
-	ret
+	RET
 SYM_FUNC_END(sha256_transform_ssse3)
 
 .section	.rodata.cst256.K256, "aM", @progbits, 256

arch/x86/crypto/sha256_ni_asm.S +1 -1

@@ -326,7 +326,7 @@ SYM_FUNC_START(sha256_ni_transform)
 
 .Ldone_hash:
 
-	ret
+	RET
 SYM_FUNC_END(sha256_ni_transform)
 
 .section	.rodata.cst256.K256, "aM", @progbits, 256

arch/x86/crypto/sha512-avx-asm.S +1 -1

@@ -361,7 +361,7 @@ updateblock:
 	pop	%rbx
 
 nowork:
-	ret
+	RET
 SYM_FUNC_END(sha512_transform_avx)
 
 ########################################################################

arch/x86/crypto/sha512-avx2-asm.S +1 -1

@@ -679,7 +679,7 @@ done_hash:
 	pop	%r12
 	pop	%rbx
 
-	ret
+	RET
 SYM_FUNC_END(sha512_transform_rorx)
 
 ########################################################################

arch/x86/crypto/sha512-ssse3-asm.S +1 -1

@@ -363,7 +363,7 @@ updateblock:
 	pop	%rbx
 
 nowork:
-	ret
+	RET
 SYM_FUNC_END(sha512_transform_ssse3)
 
 ########################################################################

arch/x86/crypto/sm4-aesni-avx-asm_64.S +6 -6

@@ -246,7 +246,7 @@ SYM_FUNC_START(sm4_aesni_avx_crypt4)
 .Lblk4_store_output_done:
 	vzeroall;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(sm4_aesni_avx_crypt4)
 
 .align 8

@@ -356,7 +356,7 @@ SYM_FUNC_START_LOCAL(__sm4_crypt_blk8)
 	vpshufb RTMP2, RB3, RB3;
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(__sm4_crypt_blk8)
 
 /*

@@ -412,7 +412,7 @@ SYM_FUNC_START(sm4_aesni_avx_crypt8)
 .Lblk8_store_output_done:
 	vzeroall;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(sm4_aesni_avx_crypt8)
 
 /*

@@ -487,7 +487,7 @@ SYM_FUNC_START(sm4_aesni_avx_ctr_enc_blk
 
 	vzeroall;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(sm4_aesni_avx_ctr_enc_blk8)
 
 /*

@@ -537,7 +537,7 @@ SYM_FUNC_START(sm4_aesni_avx_cbc_dec_blk
 
 	vzeroall;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(sm4_aesni_avx_cbc_dec_blk8)
 
 /*

@@ -590,5 +590,5 @@ SYM_FUNC_START(sm4_aesni_avx_cfb_dec_blk
 
 	vzeroall;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(sm4_aesni_avx_cfb_dec_blk8)

arch/x86/crypto/sm4-aesni-avx2-asm_64.S +4 -4

@@ -268,7 +268,7 @@ SYM_FUNC_START_LOCAL(__sm4_crypt_blk16)
 	vpshufb RTMP2, RB3, RB3;
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(__sm4_crypt_blk16)
 
 #define inc_le128(x, minus_one, tmp) \

@@ -387,7 +387,7 @@ SYM_FUNC_START(sm4_aesni_avx2_ctr_enc_bl
 
 	vzeroall;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(sm4_aesni_avx2_ctr_enc_blk16)
 
 /*

@@ -441,7 +441,7 @@ SYM_FUNC_START(sm4_aesni_avx2_cbc_dec_bl
 
 	vzeroall;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(sm4_aesni_avx2_cbc_dec_blk16)
 
 /*

@@ -497,5 +497,5 @@ SYM_FUNC_START(sm4_aesni_avx2_cfb_dec_bl
 
 	vzeroall;
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(sm4_aesni_avx2_cfb_dec_blk16)

arch/x86/crypto/twofish-avx-x86_64-asm_64.S +5 -5

@@ -267,7 +267,7 @@ SYM_FUNC_START_LOCAL(__twofish_enc_blk8)
 	outunpack_blocks(RC1, RD1, RA1, RB1, RK1, RX0, RY0, RK2);
 	outunpack_blocks(RC2, RD2, RA2, RB2, RK1, RX0, RY0, RK2);
 
-	ret;
+	RET;
 SYM_FUNC_END(__twofish_enc_blk8)
 
 .align 8

@@ -307,7 +307,7 @@ SYM_FUNC_START_LOCAL(__twofish_dec_blk8)
 	outunpack_blocks(RA1, RB1, RC1, RD1, RK1, RX0, RY0, RK2);
 	outunpack_blocks(RA2, RB2, RC2, RD2, RK1, RX0, RY0, RK2);
 
-	ret;
+	RET;
 SYM_FUNC_END(__twofish_dec_blk8)
 
 SYM_FUNC_START(twofish_ecb_enc_8way)

@@ -327,7 +327,7 @@ SYM_FUNC_START(twofish_ecb_enc_8way)
 	store_8way(%r11, RC1, RD1, RA1, RB1, RC2, RD2, RA2, RB2);
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(twofish_ecb_enc_8way)
 
 SYM_FUNC_START(twofish_ecb_dec_8way)

@@ -347,7 +347,7 @@ SYM_FUNC_START(twofish_ecb_dec_8way)
 	store_8way(%r11, RA1, RB1, RC1, RD1, RA2, RB2, RC2, RD2);
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(twofish_ecb_dec_8way)
 
 SYM_FUNC_START(twofish_cbc_dec_8way)

@@ -372,5 +372,5 @@ SYM_FUNC_START(twofish_cbc_dec_8way)
 	popq %r12;
 
 	FRAME_END
-	ret;
+	RET;
 SYM_FUNC_END(twofish_cbc_dec_8way)

arch/x86/crypto/twofish-i586-asm_32.S +2 -2

@@ -260,7 +260,7 @@ SYM_FUNC_START(twofish_enc_blk)
 	pop	%ebx
 	pop	%ebp
 	mov	$1,	%eax
-	ret
+	RET
 SYM_FUNC_END(twofish_enc_blk)
 
 SYM_FUNC_START(twofish_dec_blk)

@@ -317,5 +317,5 @@ SYM_FUNC_START(twofish_dec_blk)
 	pop	%ebx
 	pop	%ebp
 	mov	$1,	%eax
-	ret
+	RET
 SYM_FUNC_END(twofish_dec_blk)

arch/x86/crypto/twofish-x86_64-asm_64-3way.S +3 -3

@@ -258,7 +258,7 @@ SYM_FUNC_START(__twofish_enc_blk_3way)
 	popq %rbx;
 	popq %r12;
 	popq %r13;
-	ret;
+	RET;
 
 .L__enc_xor3:
 	outunpack_enc3(xor);

@@ -266,7 +266,7 @@ SYM_FUNC_START(__twofish_enc_blk_3way)
 	popq %rbx;
 	popq %r12;
 	popq %r13;
-	ret;
+	RET;
 SYM_FUNC_END(__twofish_enc_blk_3way)
 
 SYM_FUNC_START(twofish_dec_blk_3way)

@@ -301,5 +301,5 @@ SYM_FUNC_START(twofish_dec_blk_3way)
 	popq %rbx;
 	popq %r12;
 	popq %r13;
-	ret;
+	RET;
 SYM_FUNC_END(twofish_dec_blk_3way)

arch/x86/crypto/twofish-x86_64-asm_64.S +2 -2

@@ -252,7 +252,7 @@ SYM_FUNC_START(twofish_enc_blk)
 
 	popq	R1
 	movl	$1,%eax
-	ret
+	RET
 SYM_FUNC_END(twofish_enc_blk)
 
 SYM_FUNC_START(twofish_dec_blk)

@@ -304,5 +304,5 @@ SYM_FUNC_START(twofish_dec_blk)
 
 	popq	R1
 	movl	$1,%eax
-	ret
+	RET
 SYM_FUNC_END(twofish_dec_blk)

arch/x86/entry/entry_32.S +1 -1

@@ -740,7 +740,7 @@ SYM_FUNC_START(schedule_tail_wrapper)
 	popl	%eax
 
 	FRAME_END
-	ret
+	RET
 SYM_FUNC_END(schedule_tail_wrapper)
 .popsection

arch/x86/entry/entry_64.S +5 -5

@@ -738,7 +738,7 @@ SYM_FUNC_START(asm_load_gs_index)
 2:	ALTERNATIVE "", "mfence", X86_BUG_SWAPGS_FENCE
 	swapgs
 	FRAME_END
-	ret
+	RET
 SYM_FUNC_END(asm_load_gs_index)
 EXPORT_SYMBOL(asm_load_gs_index)

@@ -889,7 +889,7 @@ SYM_CODE_START_LOCAL(paranoid_entry)
 	 * is needed here.
 	 */
 	SAVE_AND_SET_GSBASE scratch_reg=%rax save_reg=%rbx
-	ret
+	RET
 
 .Lparanoid_entry_checkgs:
 	/* EBX = 1 -> kernel GSBASE active, no restore required */

@@ -910,7 +910,7 @@ SYM_CODE_START_LOCAL(paranoid_entry)
 .Lparanoid_kernel_gsbase:
 
 	FENCE_SWAPGS_KERNEL_ENTRY
-	ret
+	RET
 SYM_CODE_END(paranoid_entry)
 
 /*

@@ -989,7 +989,7 @@ SYM_CODE_START_LOCAL(error_entry)
 	movq	%rax, %rsp			/* switch stack */
 	ENCODE_FRAME_POINTER
 	pushq	%r12
-	ret
+	RET
 
 	/*
 	 * There are two places in the kernel that can potentially fault with

@@ -1020,7 +1020,7 @@ SYM_CODE_START_LOCAL(error_entry)
 	 */
 .Lerror_entry_done_lfence:
 	FENCE_SWAPGS_KERNEL_ENTRY
-	ret
+	RET
 
 .Lbstep_iret:
 	/* Fix truncated RIP */

arch/x86/entry/thunk_32.S +1 -1

@@ -24,7 +24,7 @@ SYM_CODE_START_NOALIGN(\name)
 	popl %edx
 	popl %ecx
 	popl %eax
-	ret
+	RET
 	_ASM_NOKPROBE(\name)
 SYM_CODE_END(\name)
 	.endm

arch/x86/entry/thunk_64.S +1 -1

@@ -50,7 +50,7 @@ SYM_CODE_START_LOCAL_NOALIGN(__thunk_res
 	popq %rsi
 	popq %rdi
 	popq %rbp
-	ret
+	RET
 	_ASM_NOKPROBE(__thunk_restore)
 SYM_CODE_END(__thunk_restore)
 #endif

arch/x86/entry/vdso/vdso32/system_call.S +1 -1

@@ -78,7 +78,7 @@ SYM_INNER_LABEL(int80_landing_pad, SYM_L
 	popl	%ecx
 	CFI_RESTORE		ecx
 	CFI_ADJUST_CFA_OFFSET	-4
-	ret
+	RET
 	CFI_ENDPROC
 
 	.size __kernel_vsyscall,.-__kernel_vsyscall

arch/x86/entry/vdso/vsgx.S +1 -1

@@ -81,7 +81,7 @@ SYM_FUNC_START(__vdso_sgx_enter_enclave)
 	pop	%rbx
 	leave
 	.cfi_def_cfa		%rsp, 8
-	ret
+	RET
 
 	/* The out-of-line code runs with the pre-leave stack frame. */
 	.cfi_def_cfa		%rbp, 16

arch/x86/entry/vsyscall/vsyscall_emu_64.S +3 -3

@@ -19,17 +19,17 @@ __vsyscall_page:
 
 	mov $__NR_gettimeofday, %rax
 	syscall
-	ret
+	RET
 
 	.balign 1024, 0xcc
 	mov $__NR_time, %rax
 	syscall
-	ret
+	RET
 
 	.balign 1024, 0xcc
 	mov $__NR_getcpu, %rax
 	syscall
-	ret
+	RET
 
 	.balign 4096, 0xcc

arch/x86/kernel/acpi/wakeup_32.S +3 -3

@@ -60,7 +60,7 @@ save_registers:
 	popl	saved_context_eflags
 
 	movl	$ret_point, saved_eip
-	ret
+	RET
 
 
 restore_registers:

@@ -70,7 +70,7 @@ restore_registers:
 	movl	saved_context_edi, %edi
 	pushl	saved_context_eflags
 	popfl
-	ret
+	RET
 
 SYM_CODE_START(do_suspend_lowlevel)
 	call	save_processor_state

@@ -86,7 +86,7 @@ SYM_CODE_START(do_suspend_lowlevel)
 ret_point:
 	call	restore_registers
 	call	restore_processor_state
-	ret
+	RET
 SYM_CODE_END(do_suspend_lowlevel)
 
 .data

arch/x86/kernel/ftrace_32.S +3 -3

@@ -19,7 +19,7 @@
 #endif
 
 SYM_FUNC_START(__fentry__)
-	ret
+	RET
 SYM_FUNC_END(__fentry__)
 EXPORT_SYMBOL(__fentry__)

@@ -84,7 +84,7 @@ ftrace_graph_call:
 
 /* This is weak to keep gas from relaxing the jumps */
 SYM_INNER_LABEL_ALIGN(ftrace_stub, SYM_L_WEAK)
-	ret
+	RET
 SYM_CODE_END(ftrace_caller)
 
 SYM_CODE_START(ftrace_regs_caller)

@@ -177,7 +177,7 @@ SYM_CODE_START(ftrace_graph_caller)
 	popl	%edx
 	popl	%ecx
 	popl	%eax
-	ret
+	RET
 SYM_CODE_END(ftrace_graph_caller)
 
 .globl return_to_handler

arch/x86/kernel/ftrace_64.S +5 -5

@@ -132,7 +132,7 @@
 #ifdef CONFIG_DYNAMIC_FTRACE
 
 SYM_FUNC_START(__fentry__)
-	retq
+	RET
 SYM_FUNC_END(__fentry__)
 EXPORT_SYMBOL(__fentry__)

@@ -181,11 +181,11 @@ SYM_INNER_LABEL(ftrace_graph_call, SYM_L
 
 /*
  * This is weak to keep gas from relaxing the jumps.
- * It is also used to copy the retq for trampolines.
+ * It is also used to copy the RET for trampolines.
  */
 SYM_INNER_LABEL_ALIGN(ftrace_stub, SYM_L_WEAK)
 	UNWIND_HINT_FUNC
-	retq
+	RET
 SYM_FUNC_END(ftrace_epilogue)
 
 SYM_FUNC_START(ftrace_regs_caller)

@@ -299,7 +299,7 @@ fgraph_trace:
 #endif
 
 SYM_INNER_LABEL(ftrace_stub, SYM_L_GLOBAL)
-	retq
+	RET
 
 trace:
 	/* save_mcount_regs fills in first two parameters */

@@ -331,7 +331,7 @@ SYM_FUNC_START(ftrace_graph_caller)
 
 	restore_mcount_regs
 
-	retq
+	RET
 SYM_FUNC_END(ftrace_graph_caller)
 
 SYM_FUNC_START(return_to_handler)

arch/x86/kernel/head_32.S +1 -1

@@ -340,7 +340,7 @@ SYM_FUNC_END(startup_32_smp)
 __INIT
 setup_once:
 	andl $0,setup_once_ref	/* Once is enough, thanks */
-	ret
+	RET
 
 SYM_FUNC_START(early_idt_handler_array)
 	# 36(%esp) %eflags

arch/x86/kernel/irqflags.S +1 -1

@@ -10,6 +10,6 @@
 SYM_FUNC_START(native_save_fl)
 	pushf
 	pop %_ASM_AX
-	ret
+	RET
 SYM_FUNC_END(native_save_fl)
 EXPORT_SYMBOL(native_save_fl)

arch/x86/kernel/relocate_kernel_32.S +5 -5

@@ -91,7 +91,7 @@ SYM_CODE_START_NOALIGN(relocate_kernel)
 	movl    %edi, %eax
 	addl    $(identity_mapped - relocate_kernel), %eax
 	pushl   %eax
-	ret
+	RET
 SYM_CODE_END(relocate_kernel)
 
 SYM_CODE_START_LOCAL_NOALIGN(identity_mapped)

@@ -159,7 +159,7 @@ SYM_CODE_START_LOCAL_NOALIGN(identity_ma
 	xorl    %edx, %edx
 	xorl    %esi, %esi
 	xorl    %ebp, %ebp
-	ret
+	RET
 1:
 	popl	%edx
 	movl	CP_PA_SWAP_PAGE(%edi), %esp

@@ -190,7 +190,7 @@ SYM_CODE_START_LOCAL_NOALIGN(identity_ma
 	movl	%edi, %eax
 	addl	$(virtual_mapped - relocate_kernel), %eax
 	pushl	%eax
-	ret
+	RET
 SYM_CODE_END(identity_mapped)
 
 SYM_CODE_START_LOCAL_NOALIGN(virtual_mapped)

@@ -208,7 +208,7 @@ SYM_CODE_START_LOCAL_NOALIGN(virtual_map
 	popl	%edi
 	popl	%esi
 	popl	%ebx
-	ret
+	RET
 SYM_CODE_END(virtual_mapped)
 
 	/* Do the copies */

@@ -271,7 +271,7 @@ SYM_CODE_START_LOCAL_NOALIGN(swap_pages)
 	popl	%edi
 	popl	%ebx
 	popl	%ebp
-	ret
+	RET
 SYM_CODE_END(swap_pages)
 
 	.globl kexec_control_code_size

arch/x86/kernel/relocate_kernel_64.S +5 -5

@@ -104,7 +104,7 @@ SYM_CODE_START_NOALIGN(relocate_kernel)
 	/* jump to identity mapped page */
 	addq	$(identity_mapped - relocate_kernel), %r8
 	pushq	%r8
-	ret
+	RET
 SYM_CODE_END(relocate_kernel)
 
 SYM_CODE_START_LOCAL_NOALIGN(identity_mapped)

@@ -191,7 +191,7 @@ SYM_CODE_START_LOCAL_NOALIGN(identity_ma
 	xorl	%r14d, %r14d
 	xorl	%r15d, %r15d
 
-	ret
+	RET
 
 1:
 	popq	%rdx

@@ -210,7 +210,7 @@ SYM_CODE_START_LOCAL_NOALIGN(identity_ma
 	call	swap_pages
 	movq	$virtual_mapped, %rax
 	pushq	%rax
-	ret
+	RET
 SYM_CODE_END(identity_mapped)
 
 SYM_CODE_START_LOCAL_NOALIGN(virtual_mapped)

@@ -231,7 +231,7 @@ SYM_CODE_START_LOCAL_NOALIGN(virtual_map
 	popq	%r12
 	popq	%rbp
 	popq	%rbx
-	ret
+	RET
 SYM_CODE_END(virtual_mapped)
 
 	/* Do the copies */

@@ -288,7 +288,7 @@ SYM_CODE_START_LOCAL_NOALIGN(swap_pages)
 	lea	PAGE_SIZE(%rax), %rsi
 	jmp	0b
 3:
-	ret
+	RET
 SYM_CODE_END(swap_pages)
 
 	.globl kexec_control_code_size

arch/x86/kernel/sev_verify_cbit.S +1 -1

@@ -85,5 +85,5 @@ SYM_FUNC_START(sev_verify_cbit)
 #endif
 	/* Return page-table pointer */
 	movq	%rdi, %rax
-	ret
+	RET
 SYM_FUNC_END(sev_verify_cbit)

arch/x86/kernel/verify_cpu.S +2 -2

@@ -132,9 +132,9 @@ SYM_FUNC_START_LOCAL(verify_cpu)
 .Lverify_cpu_no_longmode:
 	popf				# Restore caller passed flags
 	movl $1,%eax
-	ret
+	RET
 .Lverify_cpu_sse_ok:
 	popf				# Restore caller passed flags
 	xorl %eax, %eax
-	ret
+	RET
 SYM_FUNC_END(verify_cpu)

arch/x86/kvm/svm/vmenter.S +2 -2

@@ -148,7 +148,7 @@ SYM_FUNC_START(__svm_vcpu_run)
 	pop %edi
 #endif
 	pop %_ASM_BP
-	ret
+	RET
 
 3:	cmpb $0, kvm_rebooting
 	jne 2b

@@ -202,7 +202,7 @@ SYM_FUNC_START(__svm_sev_es_vcpu_run)
 	pop %edi
 #endif
 	pop %_ASM_BP
-	ret
+	RET
 
 3:	cmpb $0, kvm_rebooting
 	jne 2b

arch/x86/kvm/vmx/vmenter.S +7 -7

@@ -49,14 +49,14 @@ SYM_FUNC_START_LOCAL(vmx_vmenter)
 	je 2f
 
 1:	vmresume
-	ret
+	RET
 
 2:	vmlaunch
-	ret
+	RET
 
 3:	cmpb $0, kvm_rebooting
 	je 4f
-	ret
+	RET
 4:	ud2
 
 	_ASM_EXTABLE(1b, 3b)

@@ -89,7 +89,7 @@ SYM_FUNC_START(vmx_vmexit)
 	pop %_ASM_AX
 .Lvmexit_skip_rsb:
 #endif
-	ret
+	RET
 SYM_FUNC_END(vmx_vmexit)
 
 /**

@@ -228,7 +228,7 @@ SYM_FUNC_START(__vmx_vcpu_run)
 	pop %edi
 #endif
 	pop %_ASM_BP
-	ret
+	RET
 
 	/* VM-Fail.  Out-of-line to avoid a taken Jcc after VM-Exit. */
 2:	mov $1, %eax

@@ -293,7 +293,7 @@ SYM_FUNC_START(vmread_error_trampoline)
 	pop %_ASM_AX
 	pop %_ASM_BP
 
-	ret
+	RET
 SYM_FUNC_END(vmread_error_trampoline)
 
 SYM_FUNC_START(vmx_do_interrupt_nmi_irqoff)

@@ -326,5 +326,5 @@ SYM_FUNC_START(vmx_do_interrupt_nmi_irqo
 	 */
 	mov %_ASM_BP, %_ASM_SP
 	pop %_ASM_BP
-	ret
+	RET
 SYM_FUNC_END(vmx_do_interrupt_nmi_irqoff)

arch/x86/lib/atomic64_386_32.S +1 -1

@@ -30,7 +30,7 @@ SYM_FUNC_START(atomic64_##op##_386); \
 
 #define RET_IRQ_RESTORE \
 	IRQ_RESTORE v; \
-	ret
+	RET
 
 #define v %ecx
 BEGIN_IRQ_SAVE(read)

arch/x86/lib/atomic64_cx8_32.S +8 -8

@@ -18,7 +18,7 @@
 
 SYM_FUNC_START(atomic64_read_cx8)
 	read64 %ecx
-	ret
+	RET
 SYM_FUNC_END(atomic64_read_cx8)
 
 SYM_FUNC_START(atomic64_set_cx8)

@@ -28,7 +28,7 @@ SYM_FUNC_START(atomic64_set_cx8)
 	cmpxchg8b (%esi)
 	jne 1b
 
-	ret
+	RET
 SYM_FUNC_END(atomic64_set_cx8)
 
 SYM_FUNC_START(atomic64_xchg_cx8)

@@ -37,7 +37,7 @@ SYM_FUNC_START(atomic64_xchg_cx8)
 	cmpxchg8b (%esi)
 	jne 1b
 
-	ret
+	RET
 SYM_FUNC_END(atomic64_xchg_cx8)
 
 .macro addsub_return func ins insc

@@ -68,7 +68,7 @@ SYM_FUNC_START(atomic64_\func\()_return_
 	popl %esi
 	popl %ebx
 	popl %ebp
-	ret
+	RET
 SYM_FUNC_END(atomic64_\func\()_return_cx8)
 .endm

@@ -93,7 +93,7 @@ SYM_FUNC_START(atomic64_\func\()_return_
 	movl %ebx, %eax
 	movl %ecx, %edx
 	popl %ebx
-	ret
+	RET
 SYM_FUNC_END(atomic64_\func\()_return_cx8)
 .endm

@@ -118,7 +118,7 @@ SYM_FUNC_START(atomic64_dec_if_positive_
 	movl %ebx, %eax
 	movl %ecx, %edx
 	popl %ebx
-	ret
+	RET
 SYM_FUNC_END(atomic64_dec_if_positive_cx8)
 
 SYM_FUNC_START(atomic64_add_unless_cx8)

@@ -149,7 +149,7 @@ SYM_FUNC_START(atomic64_add_unless_cx8)
 	addl $8, %esp
 	popl %ebx
 	popl %ebp
-	ret
+	RET
 4:
 	cmpl %edx, 4(%esp)
 	jne 2b

@@ -176,5 +176,5 @@ SYM_FUNC_START(atomic64_inc_not_zero_cx8
 	movl $1, %eax
 3:
 	popl %ebx
-	ret
+	RET
 SYM_FUNC_END(atomic64_inc_not_zero_cx8)

arch/x86/lib/checksum_32.S +4 -4

@@ -127,7 +127,7 @@ SYM_FUNC_START(csum_partial)
 8:
 	popl %ebx
 	popl %esi
-	ret
+	RET
 SYM_FUNC_END(csum_partial)
 
 #else

@@ -245,7 +245,7 @@ SYM_FUNC_START(csum_partial)
 90: 
 	popl %ebx
 	popl %esi
-	ret
+	RET
 SYM_FUNC_END(csum_partial)
 				
 #endif

@@ -371,7 +371,7 @@ EXC(	movb %cl, (%edi)	)
 	popl %esi
 	popl %edi
 	popl %ecx			# equivalent to addl $4,%esp
-	ret	
+	RET
 SYM_FUNC_END(csum_partial_copy_generic)
 
 #else

@@ -447,7 +447,7 @@ EXC(	movb %dl, (%edi)         )
 	popl %esi
 	popl %edi
 	popl %ebx
-	ret
+	RET
 SYM_FUNC_END(csum_partial_copy_generic)
 				
 #undef ROUND

arch/x86/lib/clear_page_64.S +3 -3

@@ -17,7 +17,7 @@ SYM_FUNC_START(clear_page_rep)
 	movl $4096/8,%ecx
 	xorl %eax,%eax
 	rep stosq
-	ret
+	RET
 SYM_FUNC_END(clear_page_rep)
 EXPORT_SYMBOL_GPL(clear_page_rep)

@@ -39,7 +39,7 @@ SYM_FUNC_START(clear_page_orig)
 	leaq	64(%rdi),%rdi
 	jnz	.Lloop
 	nop
-	ret
+	RET
 SYM_FUNC_END(clear_page_orig)
 EXPORT_SYMBOL_GPL(clear_page_orig)

@@ -47,6 +47,6 @@ SYM_FUNC_START(clear_page_erms)
 	movl $4096,%ecx
 	xorl %eax,%eax
 	rep stosb
-	ret
+	RET
 SYM_FUNC_END(clear_page_erms)
 EXPORT_SYMBOL_GPL(clear_page_erms)

arch/x86/lib/cmpxchg16b_emu.S +2 -2

@@ -37,11 +37,11 @@ SYM_FUNC_START(this_cpu_cmpxchg16b_emu)
 
 	popfq
 	mov $1, %al
-	ret
+	RET
 
 .Lnot_same:
 	popfq
 	xor %al,%al
-	ret
+	RET
 
 SYM_FUNC_END(this_cpu_cmpxchg16b_emu)

arch/x86/lib/cmpxchg8b_emu.S +2 -2

@@ -32,7 +32,7 @@ SYM_FUNC_START(cmpxchg8b_emu)
 	movl %ecx, 4(%esi)
 
 	popfl
-	ret
+	RET
 
 .Lnot_same:
 	movl  (%esi), %eax

@@ -40,7 +40,7 @@ SYM_FUNC_START(cmpxchg8b_emu)
 	movl 4(%esi), %edx
 
 	popfl
-	ret
+	RET
 
 SYM_FUNC_END(cmpxchg8b_emu)
 EXPORT_SYMBOL(cmpxchg8b_emu)

arch/x86/lib/copy_mc_64.S +3 -3

@@ -77,7 +77,7 @@ SYM_FUNC_START(copy_mc_fragile)
 .L_done_memcpy_trap:
 	xorl %eax, %eax
 .L_done:
-	ret
+	RET
 SYM_FUNC_END(copy_mc_fragile)
 
 	.section .fixup, "ax"

@@ -132,7 +132,7 @@ SYM_FUNC_START(copy_mc_enhanced_fast_str
 	rep movsb
 	/* Copy successful. Return zero */
 	xorl %eax, %eax
-	ret
+	RET
 SYM_FUNC_END(copy_mc_enhanced_fast_string)
 
 	.section .fixup, "ax"

@@ -145,7 +145,7 @@ SYM_FUNC_END(copy_mc_enhanced_fast_strin
 	 * user-copy routines.
 	 */
 	movq %rcx, %rax
-	ret
+	RET
 
 	.previous

arch/x86/lib/copy_page_64.S +2 -2

@@ -17,7 +17,7 @@ SYM_FUNC_START(copy_page)
 	ALTERNATIVE "jmp copy_page_regs", "", X86_FEATURE_REP_GOOD
 	movl	$4096/8, %ecx
 	rep	movsq
-	ret
+	RET
 SYM_FUNC_END(copy_page)
 EXPORT_SYMBOL(copy_page)

@@ -85,5 +85,5 @@ SYM_FUNC_START_LOCAL(copy_page_regs)
 	movq	(%rsp), %rbx
 	movq	1*8(%rsp), %r12
 	addq	$2*8, %rsp
-	ret
+	RET
 SYM_FUNC_END(copy_page_regs)

arch/x86/lib/copy_user_64.S +6 -6

@@ -105,7 +105,7 @@ SYM_FUNC_START(copy_user_generic_unrolle
 	jnz 21b
 23:	xor %eax,%eax
 	ASM_CLAC
-	ret
+	RET
 
 	.section .fixup,"ax"
 30:	shll $6,%ecx

@@ -173,7 +173,7 @@ SYM_FUNC_START(copy_user_generic_string)
 	movsb
 	xorl %eax,%eax
 	ASM_CLAC
-	ret
+	RET
 
 	.section .fixup,"ax"
 11:	leal (%rdx,%rcx,8),%ecx

@@ -207,7 +207,7 @@ SYM_FUNC_START(copy_user_enhanced_fast_s
 	movsb
 	xorl %eax,%eax
 	ASM_CLAC
-	ret
+	RET
 
 	.section .fixup,"ax"
 12:	movl %ecx,%edx		/* ecx is zerorest also */

@@ -239,7 +239,7 @@ SYM_CODE_START_LOCAL(.Lcopy_user_handle_
 1:	rep movsb
 2:	mov %ecx,%eax
 	ASM_CLAC
-	ret
+	RET
 
 	/*
 	 * Return zero to pretend that this copy succeeded. This

@@ -250,7 +250,7 @@ SYM_CODE_START_LOCAL(.Lcopy_user_handle_
 	 */
 3:	xorl %eax,%eax
 	ASM_CLAC
-	ret
+	RET
 
 	_ASM_EXTABLE_CPY(1b, 2b)
 SYM_CODE_END(.Lcopy_user_handle_tail)

@@ -361,7 +361,7 @@ SYM_FUNC_START(__copy_user_nocache)
 	xorl %eax,%eax
 	ASM_CLAC
 	sfence
-	ret
+	RET
 
 	.section .fixup,"ax"
 .L_fixup_4x8b_copy:

arch/x86/lib/csum-copy_64.S +1 -1

@@ -201,7 +201,7 @@ SYM_FUNC_START(csum_partial_copy_generic
 	movq 3*8(%rsp), %r13
 	movq 4*8(%rsp), %r15
 	addq $5*8, %rsp
-	ret
+	RET
 .Lshort:
 	movl %ecx, %r10d
 	jmp  .L1

arch/x86/lib/getuser.S +11 ~~-11~~

@@ -57,7 +57,7 @@ SYM_FUNC_START(__get_user_1)
 1:	movzbl (%_ASM_AX),%edx
 	xor %eax,%eax
 	ASM_CLAC
-	ret
+	RET
 SYM_FUNC_END(__get_user_1)
 EXPORT_SYMBOL(__get_user_1)

@@ -71,7 +71,7 @@ SYM_FUNC_START(__get_user_2)
 2:	movzwl (%_ASM_AX),%edx
 	xor %eax,%eax
 	ASM_CLAC
-	ret
+	RET
 SYM_FUNC_END(__get_user_2)
 EXPORT_SYMBOL(__get_user_2)

@@ -85,7 +85,7 @@ SYM_FUNC_START(__get_user_4)
 3:	movl (%_ASM_AX),%edx
 	xor %eax,%eax
 	ASM_CLAC
-	ret
+	RET
 SYM_FUNC_END(__get_user_4)
 EXPORT_SYMBOL(__get_user_4)

@@ -100,7 +100,7 @@ SYM_FUNC_START(__get_user_8)
 4:	movq (%_ASM_AX),%rdx
 	xor %eax,%eax
 	ASM_CLAC
-	ret
+	RET
 #else
 	LOAD_TASK_SIZE_MINUS_N(7)
 	cmp %_ASM_DX,%_ASM_AX

@@ -112,7 +112,7 @@ SYM_FUNC_START(__get_user_8)
 5:	movl 4(%_ASM_AX),%ecx
 	xor %eax,%eax
 	ASM_CLAC
-	ret
+	RET
 #endif
 SYM_FUNC_END(__get_user_8)
 EXPORT_SYMBOL(__get_user_8)

@@ -124,7 +124,7 @@ SYM_FUNC_START(__get_user_nocheck_1)
 6:	movzbl (%_ASM_AX),%edx
 	xor %eax,%eax
 	ASM_CLAC
-	ret
+	RET
 SYM_FUNC_END(__get_user_nocheck_1)
 EXPORT_SYMBOL(__get_user_nocheck_1)

@@ -134,7 +134,7 @@ SYM_FUNC_START(__get_user_nocheck_2)
 7:	movzwl (%_ASM_AX),%edx
 	xor %eax,%eax
 	ASM_CLAC
-	ret
+	RET
 SYM_FUNC_END(__get_user_nocheck_2)
 EXPORT_SYMBOL(__get_user_nocheck_2)

@@ -144,7 +144,7 @@ SYM_FUNC_START(__get_user_nocheck_4)
 8:	movl (%_ASM_AX),%edx
 	xor %eax,%eax
 	ASM_CLAC
-	ret
+	RET
 SYM_FUNC_END(__get_user_nocheck_4)
 EXPORT_SYMBOL(__get_user_nocheck_4)

@@ -159,7 +159,7 @@ SYM_FUNC_START(__get_user_nocheck_8)
 #endif
 	xor %eax,%eax
 	ASM_CLAC
-	ret
+	RET
 SYM_FUNC_END(__get_user_nocheck_8)
 EXPORT_SYMBOL(__get_user_nocheck_8)

@@ -169,7 +169,7 @@ SYM_CODE_START_LOCAL(.Lbad_get_user_clac
 bad_get_user:
 	xor %edx,%edx
 	mov $(-EFAULT),%_ASM_AX
-	ret
+	RET
 SYM_CODE_END(.Lbad_get_user_clac)
 
 #ifdef CONFIG_X86_32

@@ -179,7 +179,7 @@ bad_get_user_8:
 	xor %edx,%edx
 	xor %ecx,%ecx
 	mov $(-EFAULT),%_ASM_AX
-	ret
+	RET
 SYM_CODE_END(.Lbad_get_user_8_clac)
 #endif

arch/x86/lib/hweight.S +3 -3

@@ -32,7 +32,7 @@ SYM_FUNC_START(__sw_hweight32)
 	imull $0x01010101, %eax, %eax		# w_tmp *= 0x01010101
 	shrl $24, %eax				# w = w_tmp >> 24
 	__ASM_SIZE(pop,) %__ASM_REG(dx)
-	ret
+	RET
 SYM_FUNC_END(__sw_hweight32)
 EXPORT_SYMBOL(__sw_hweight32)

@@ -65,7 +65,7 @@ SYM_FUNC_START(__sw_hweight64)
 
 	popq    %rdx
 	popq    %rdi
-	ret
+	RET
 #else /* CONFIG_X86_32 */
 	/* We're getting an u64 arg in (%eax,%edx): unsigned long hweight64(__u64 w) */
 	pushl   %ecx

@@ -77,7 +77,7 @@ SYM_FUNC_START(__sw_hweight64)
 	addl    %ecx, %eax                      # result
 
 	popl    %ecx
-	ret
+	RET
 #endif
 SYM_FUNC_END(__sw_hweight64)
 EXPORT_SYMBOL(__sw_hweight64)

arch/x86/lib/iomap_copy_64.S +1 -1

@@ -11,5 +11,5 @@
 SYM_FUNC_START(__iowrite32_copy)
 	movl %edx,%ecx
 	rep movsd
-	ret
+	RET
 SYM_FUNC_END(__iowrite32_copy)

arch/x86/lib/memcpy_64.S +6 -6

@@ -39,7 +39,7 @@ SYM_FUNC_START_WEAK(memcpy)
 	rep movsq
 	movl %edx, %ecx
 	rep movsb
-	ret
+	RET
 SYM_FUNC_END(memcpy)
 SYM_FUNC_END_ALIAS(__memcpy)
 EXPORT_SYMBOL(memcpy)

@@ -53,7 +53,7 @@ SYM_FUNC_START_LOCAL(memcpy_erms)
 	movq %rdi, %rax
 	movq %rdx, %rcx
 	rep movsb
-	ret
+	RET
 SYM_FUNC_END(memcpy_erms)
 
 SYM_FUNC_START_LOCAL(memcpy_orig)

@@ -137,7 +137,7 @@ SYM_FUNC_START_LOCAL(memcpy_orig)
 	movq %r9,	1*8(%rdi)
 	movq %r10,	-2*8(%rdi, %rdx)
 	movq %r11,	-1*8(%rdi, %rdx)
-	retq
+	RET
 	.p2align 4
 .Lless_16bytes:
 	cmpl $8,	%edx

@@ -149,7 +149,7 @@ SYM_FUNC_START_LOCAL(memcpy_orig)
 	movq -1*8(%rsi, %rdx),	%r9
 	movq %r8,	0*8(%rdi)
 	movq %r9,	-1*8(%rdi, %rdx)
-	retq
+	RET
 	.p2align 4
 .Lless_8bytes:
 	cmpl $4,	%edx

@@ -162,7 +162,7 @@ SYM_FUNC_START_LOCAL(memcpy_orig)
 	movl -4(%rsi, %rdx), %r8d
 	movl %ecx, (%rdi)
 	movl %r8d, -4(%rdi, %rdx)
-	retq
+	RET
 	.p2align 4
 .Lless_3bytes:
 	subl $1, %edx

@@ -180,7 +180,7 @@ SYM_FUNC_START_LOCAL(memcpy_orig)
 	movb %cl, (%rdi)
 
 .Lend:
-	retq
+	RET
 SYM_FUNC_END(memcpy_orig)
 
 .popsection

arch/x86/lib/memmove_64.S +2 -2

@@ -40,7 +40,7 @@ SYM_FUNC_START(__memmove)
 	/* FSRM implies ERMS => no length checks, do the copy directly */
 .Lmemmove_begin_forward:
 	ALTERNATIVE "cmp $0x20, %rdx; jb 1f", "", X86_FEATURE_FSRM
-	ALTERNATIVE "", "movq %rdx, %rcx; rep movsb; retq", X86_FEATURE_ERMS
+	ALTERNATIVE "", "movq %rdx, %rcx; rep movsb; RET", X86_FEATURE_ERMS
 
 	/*
 	 * movsq instruction have many startup latency

@@ -205,7 +205,7 @@ SYM_FUNC_START(__memmove)
 	movb (%rsi), %r11b
 	movb %r11b, (%rdi)
 13:
-	retq
+	RET
 SYM_FUNC_END(__memmove)
 SYM_FUNC_END_ALIAS(memmove)
 EXPORT_SYMBOL(__memmove)

arch/x86/lib/memset_64.S +3 -3

@@ -40,7 +40,7 @@ SYM_FUNC_START(__memset)
 	movl %edx,%ecx
 	rep stosb
 	movq %r9,%rax
-	ret
+	RET
 SYM_FUNC_END(__memset)
 SYM_FUNC_END_ALIAS(memset)
 EXPORT_SYMBOL(memset)

@@ -63,7 +63,7 @@ SYM_FUNC_START_LOCAL(memset_erms)
 	movq %rdx,%rcx
 	rep stosb
 	movq %r9,%rax
-	ret
+	RET
 SYM_FUNC_END(memset_erms)
 
 SYM_FUNC_START_LOCAL(memset_orig)

@@ -125,7 +125,7 @@ SYM_FUNC_START_LOCAL(memset_orig)
 
 .Lende:
 	movq	%r10,%rax
-	ret
+	RET
 
 .Lbad_alignment:
 	cmpq $7,%rdx

arch/x86/lib/msr-reg.S +2 -2

@@ -35,7 +35,7 @@ SYM_FUNC_START(\op\()_safe_regs)
 	movl    %edi, 28(%r10)
 	popq %r12
 	popq %rbx
-	ret
+	RET
 3:
 	movl    $-EIO, %r11d
 	jmp     2b

@@ -77,7 +77,7 @@ SYM_FUNC_START(\op\()_safe_regs)
 	popl %esi
 	popl %ebp
 	popl %ebx
-	ret
+	RET
 3:
 	movl    $-EIO, 4(%esp)
 	jmp     2b

arch/x86/lib/putuser.S +3 -3

@@ -52,7 +52,7 @@ SYM_INNER_LABEL(__put_user_nocheck_1, SY
 1:	movb %al,(%_ASM_CX)
 	xor %ecx,%ecx
 	ASM_CLAC
-	ret
+	RET
 SYM_FUNC_END(__put_user_1)
 EXPORT_SYMBOL(__put_user_1)
 EXPORT_SYMBOL(__put_user_nocheck_1)

@@ -66,7 +66,7 @@ SYM_INNER_LABEL(__put_user_nocheck_2, SY
 2:	movw %ax,(%_ASM_CX)
 	xor %ecx,%ecx
 	ASM_CLAC
-	ret
+	RET
 SYM_FUNC_END(__put_user_2)
 EXPORT_SYMBOL(__put_user_2)
 EXPORT_SYMBOL(__put_user_nocheck_2)

@@ -80,7 +80,7 @@ SYM_INNER_LABEL(__put_user_nocheck_4, SY
 3:	movl %eax,(%_ASM_CX)
 	xor %ecx,%ecx
 	ASM_CLAC
-	ret
+	RET
 SYM_FUNC_END(__put_user_4)
 EXPORT_SYMBOL(__put_user_4)
 EXPORT_SYMBOL(__put_user_nocheck_4)

arch/x86/lib/retpoline.S +1 -1

@@ -23,7 +23,7 @@
 .Ldo_rop_\@:
 	mov     %\reg, (%_ASM_SP)
 	UNWIND_HINT_FUNC
-	ret
+	RET
 .endm
 
 .macro THUNK reg

arch/x86/math-emu/div_Xsig.S +1 -1

@@ -341,7 +341,7 @@ L_exit:
 	popl	%esi
 
 	leave
-	ret
+	RET
 
 
 #ifdef PARANOID

arch/x86/math-emu/div_small.S +1 -1

@@ -44,5 +44,5 @@ SYM_FUNC_START(FPU_div_small)
 	popl	%esi
 
 	leave
-	ret
+	RET
 SYM_FUNC_END(FPU_div_small)

arch/x86/math-emu/mul_Xsig.S +3 -3

@@ -62,7 +62,7 @@ SYM_FUNC_START(mul32_Xsig)
 
 	popl %esi
 	leave
-	ret
+	RET
 SYM_FUNC_END(mul32_Xsig)

@@ -115,7 +115,7 @@ SYM_FUNC_START(mul64_Xsig)
 
 	popl %esi
 	leave
-	ret
+	RET
 SYM_FUNC_END(mul64_Xsig)

@@ -175,5 +175,5 @@ SYM_FUNC_START(mul_Xsig_Xsig)
 
 	popl %esi
 	leave
-	ret
+	RET
 SYM_FUNC_END(mul_Xsig_Xsig)

arch/x86/math-emu/polynom_Xsig.S +1 -1

@@ -133,5 +133,5 @@ L_accum_done:
 	popl	%edi
 	popl	%esi
 	leave
-	ret
+	RET
 SYM_FUNC_END(polynomial_Xsig)

arch/x86/math-emu/reg_norm.S +3 -3

@@ -72,7 +72,7 @@ L_exit_valid:
 L_exit:
 	popl	%ebx
 	leave
-	ret
+	RET
 
 
 L_zero:

@@ -138,7 +138,7 @@ L_exit_nuo_valid:
 
 	popl	%ebx
 	leave
-	ret
+	RET
 
 L_exit_nuo_zero:
 	movl	TAG_Zero,%eax

@@ -146,5 +146,5 @@ L_exit_nuo_zero:
 
 	popl	%ebx
 	leave
-	ret
+	RET
 SYM_FUNC_END(FPU_normalize_nuo)

arch/x86/math-emu/reg_round.S +1 -1

@@ -437,7 +437,7 @@ fpu_Arith_exit:
 	popl	%edi
 	popl	%esi
 	leave
-	ret
+	RET
 
 
 /*

arch/x86/math-emu/reg_u_add.S +1 -1

@@ -164,6 +164,6 @@ L_exit:
 	popl	%edi
 	popl	%esi
 	leave
-	ret
+	RET
 #endif /* PARANOID */
 SYM_FUNC_END(FPU_u_add)

arch/x86/math-emu/reg_u_div.S +1 -1

@@ -468,7 +468,7 @@ L_exit:
 	popl	%esi
 
 	leave
-	ret
+	RET
 #endif /* PARANOID */ 
 
 SYM_FUNC_END(FPU_u_div)

arch/x86/math-emu/reg_u_mul.S +1 -1

@@ -144,7 +144,7 @@ L_exit:
 	popl	%edi
 	popl	%esi
 	leave
-	ret
+	RET
 #endif /* PARANOID */ 
 
 SYM_FUNC_END(FPU_u_mul)

arch/x86/math-emu/reg_u_sub.S +1 -1

@@ -270,5 +270,5 @@ L_exit:
 	popl	%edi
 	popl	%esi
 	leave
-	ret
+	RET
 SYM_FUNC_END(FPU_u_sub)

arch/x86/math-emu/round_Xsig.S +2 -2

@@ -78,7 +78,7 @@ L_exit:
 	popl	%esi
 	popl	%ebx
 	leave
-	ret
+	RET
 SYM_FUNC_END(round_Xsig)

@@ -138,5 +138,5 @@ L_n_exit:
 	popl	%esi
 	popl	%ebx
 	leave
-	ret
+	RET
 SYM_FUNC_END(norm_Xsig)

arch/x86/math-emu/shr_Xsig.S +4 -4

@@ -45,7 +45,7 @@ SYM_FUNC_START(shr_Xsig)
 	popl	%ebx
 	popl	%esi
 	leave
-	ret
+	RET
 
 L_more_than_31:
 	cmpl	$64,%ecx

@@ -61,7 +61,7 @@ L_more_than_31:
 	movl	$0,8(%esi)
 	popl	%esi
 	leave
-	ret
+	RET
 
 L_more_than_63:
 	cmpl	$96,%ecx

@@ -76,7 +76,7 @@ L_more_than_63:
 	movl	%edx,8(%esi)
 	popl	%esi
 	leave
-	ret
+	RET
 
 L_more_than_95:
 	xorl	%eax,%eax

@@ -85,5 +85,5 @@ L_more_than_95:
 	movl	%eax,8(%esi)
 	popl	%esi
 	leave
-	ret
+	RET
 SYM_FUNC_END(shr_Xsig)

arch/x86/math-emu/wm_shrx.S +8 -8

@@ -55,7 +55,7 @@ SYM_FUNC_START(FPU_shrx)
 	popl	%ebx
 	popl	%esi
 	leave
-	ret
+	RET
 
 L_more_than_31:
 	cmpl	$64,%ecx

@@ -70,7 +70,7 @@ L_more_than_31:
 	movl	$0,4(%esi)
 	popl	%esi
 	leave
-	ret
+	RET
 
 L_more_than_63:
 	cmpl	$96,%ecx

@@ -84,7 +84,7 @@ L_more_than_63:
 	movl	%edx,4(%esi)
 	popl	%esi
 	leave
-	ret
+	RET
 
 L_more_than_95:
 	xorl	%eax,%eax

@@ -92,7 +92,7 @@ L_more_than_95:
 	movl	%eax,4(%esi)
 	popl	%esi
 	leave
-	ret
+	RET
 SYM_FUNC_END(FPU_shrx)

@@ -146,7 +146,7 @@ SYM_FUNC_START(FPU_shrxs)
 	popl	%ebx
 	popl	%esi
 	leave
-	ret
+	RET
 
 /* Shift by [0..31] bits */
 Ls_less_than_32:

@@ -163,7 +163,7 @@ Ls_less_than_32:
 	popl	%ebx
 	popl	%esi
 	leave
-	ret
+	RET
 
 /* Shift by [64..95] bits */
 Ls_more_than_63:

@@ -189,7 +189,7 @@ Ls_more_than_63:
 	popl	%ebx
 	popl	%esi
 	leave
-	ret
+	RET
 
 Ls_more_than_95:
 /* Shift by [96..inf) bits */

@@ -203,5 +203,5 @@ Ls_more_than_95:
 	popl	%ebx
 	popl	%esi
 	leave
-	ret
+	RET
 SYM_FUNC_END(FPU_shrxs)

arch/x86/mm/mem_encrypt_boot.S +2 -2

@@ -65,7 +65,7 @@ SYM_FUNC_START(sme_encrypt_execute)
 	movq	%rbp, %rsp		/* Restore original stack pointer */
 	pop	%rbp
 
-	ret
+	RET
 SYM_FUNC_END(sme_encrypt_execute)
 
 SYM_FUNC_START(__enc_copy)

@@ -151,6 +151,6 @@ SYM_FUNC_START(__enc_copy)
 	pop	%r12
 	pop	%r15
 
-	ret
+	RET
 .L__enc_copy_end:
 SYM_FUNC_END(__enc_copy)

arch/x86/platform/efi/efi_stub_32.S +1 -1

@@ -56,5 +56,5 @@ SYM_FUNC_START(efi_call_svam)
 
 	movl	16(%esp), %ebx
 	leave
-	ret
+	RET
 SYM_FUNC_END(efi_call_svam)

arch/x86/platform/efi/efi_stub_64.S +1 -1

@@ -23,5 +23,5 @@ SYM_FUNC_START(__efi_call)
 	mov %rsi, %rcx
 	CALL_NOSPEC rdi
 	leave
-	ret
+	RET
 SYM_FUNC_END(__efi_call)

arch/x86/platform/efi/efi_thunk_64.S +1 -1

@@ -63,7 +63,7 @@ SYM_CODE_START(__efi64_thunk)
 1:	movq	24(%rsp), %rsp
 	pop	%rbx
 	pop	%rbp
-	retq
+	RET
 
 	.code32
 2:	pushl	$__KERNEL_CS

arch/x86/platform/olpc/xo1-wakeup.S +3 -3

@@ -77,7 +77,7 @@ save_registers:
 	pushfl
 	popl saved_context_eflags
 
-	ret
+	RET
 
 restore_registers:
 	movl saved_context_ebp, %ebp

@@ -88,7 +88,7 @@ restore_registers:
 	pushl saved_context_eflags
 	popfl
 
-	ret
+	RET
 
 SYM_CODE_START(do_olpc_suspend_lowlevel)
 	call	save_processor_state

@@ -109,7 +109,7 @@ ret_point:
 
 	call	restore_registers
 	call	restore_processor_state
-	ret
+	RET
 SYM_CODE_END(do_olpc_suspend_lowlevel)
 
 .data

arch/x86/power/hibernate_asm_32.S +2 -2

@@ -32,7 +32,7 @@ SYM_FUNC_START(swsusp_arch_suspend)
 	FRAME_BEGIN
 	call swsusp_save
 	FRAME_END
-	ret
+	RET
 SYM_FUNC_END(swsusp_arch_suspend)
 
 SYM_CODE_START(restore_image)

@@ -108,5 +108,5 @@ SYM_FUNC_START(restore_registers)
 	/* tell the hibernation core that we've just restored the memory */
 	movl	%eax, in_suspend
 
-	ret
+	RET
 SYM_FUNC_END(restore_registers)

arch/x86/power/hibernate_asm_64.S +2 -2

@@ -66,7 +66,7 @@ SYM_FUNC_START(restore_registers)
 	/* tell the hibernation core that we've just restored the memory */
 	movq	%rax, in_suspend(%rip)
 
-	ret
+	RET
 SYM_FUNC_END(restore_registers)
 
 SYM_FUNC_START(swsusp_arch_suspend)

@@ -96,7 +96,7 @@ SYM_FUNC_START(swsusp_arch_suspend)
 	FRAME_BEGIN
 	call swsusp_save
 	FRAME_END
-	ret
+	RET
 SYM_FUNC_END(swsusp_arch_suspend)
 
 SYM_FUNC_START(restore_image)

arch/x86/um/checksum_32.S +2 -2

@@ -110,7 +110,7 @@ csum_partial:
 7:	
 	popl %ebx
 	popl %esi
-	ret
+	RET
 
 #else

@@ -208,7 +208,7 @@ csum_partial:
 80: 
 	popl %ebx
 	popl %esi
-	ret
+	RET
 				
 #endif
 	EXPORT_SYMBOL(csum_partial)

arch/x86/um/setjmp_32.S +1 -1

@@ -34,7 +34,7 @@ kernel_setjmp:
 	movl %esi,12(%edx)
 	movl %edi,16(%edx)
 	movl %ecx,20(%edx)		# Return address
-	ret
+	RET
 
 	.size kernel_setjmp,.-kernel_setjmp

arch/x86/um/setjmp_64.S +1 -1

@@ -33,7 +33,7 @@ kernel_setjmp:
 	movq %r14,40(%rdi)
 	movq %r15,48(%rdi)
 	movq %rsi,56(%rdi)		# Return address
-	ret
+	RET
 
 	.size kernel_setjmp,.-kernel_setjmp

arch/x86/xen/xen-asm.S +6 -6

@@ -45,7 +45,7 @@ SYM_FUNC_START(xen_irq_enable_direct)
 	call check_events
 1:
 	FRAME_END
-	ret
+	RET
 SYM_FUNC_END(xen_irq_enable_direct)

@@ -55,7 +55,7 @@ SYM_FUNC_END(xen_irq_enable_direct)
  */
 SYM_FUNC_START(xen_irq_disable_direct)
 	movb $1, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_mask
-	ret
+	RET
 SYM_FUNC_END(xen_irq_disable_direct)
 
 /*

@@ -71,7 +71,7 @@ SYM_FUNC_START(xen_save_fl_direct)
 	testb $0xff, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_mask
 	setz %ah
 	addb %ah, %ah
-	ret
+	RET
 SYM_FUNC_END(xen_save_fl_direct)
 
 /*

@@ -100,7 +100,7 @@ SYM_FUNC_START(check_events)
 	pop %rcx
 	pop %rax
 	FRAME_END
-	ret
+	RET
 SYM_FUNC_END(check_events)
 
 SYM_FUNC_START(xen_read_cr2)

@@ -108,14 +108,14 @@ SYM_FUNC_START(xen_read_cr2)
 	_ASM_MOV PER_CPU_VAR(xen_vcpu), %_ASM_AX
 	_ASM_MOV XEN_vcpu_info_arch_cr2(%_ASM_AX), %_ASM_AX
 	FRAME_END
-	ret
+	RET
 SYM_FUNC_END(xen_read_cr2);
 
 SYM_FUNC_START(xen_read_cr2_direct)
 	FRAME_BEGIN
 	_ASM_MOV PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_arch_cr2, %_ASM_AX
 	FRAME_END
-	ret
+	RET
 SYM_FUNC_END(xen_read_cr2_direct);
 
 .macro xen_pv_trap name

arch/x86/xen/xen-head.S +1 -1

@@ -70,7 +70,7 @@ SYM_CODE_START(hypercall_page)
 	.rept (PAGE_SIZE / 32)
 		UNWIND_HINT_FUNC
 		.skip 31, 0x90
-		ret
+		RET
 	.endr
 
 #define HYPERCALL(n) \

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #3 lore raw

PATCH 3/21 x86: Prepare inline-asm for straight-line-speculation

From: Peter Zijlstra <[email protected]>

[ Upstream commit b17c2baa305cccbd16bafa289fd743cc2db77966 ]

Replace all ret/retq instructions with ASM_RET in preparation of making it more than a single instruction.

Signed-off-by Peter Zijlstra (Intel) <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 arch/x86/include/asm/linkage.h            |    4 ++++
 arch/x86/include/asm/paravirt.h           |    2 +-
 arch/x86/include/asm/qspinlock_paravirt.h |    4 ++--
 arch/x86/kernel/alternative.c             |    2 +-
 arch/x86/kernel/kprobes/core.c            |    2 +-
 arch/x86/kernel/paravirt.c                |    2 +-
 arch/x86/kvm/emulate.c                    |    4 ++--
 arch/x86/lib/error-inject.c               |    3 ++-
 samples/ftrace/ftrace-direct-modify.c     |    4 ++--
 samples/ftrace/ftrace-direct-too.c        |    2 +-
 samples/ftrace/ftrace-direct.c            |    2 +-
 11 files changed, 18 insertions(+), 13 deletions(-)

arch/x86/include/asm/linkage.h +4 -0

@@ -18,6 +18,10 @@
 #define __ALIGN_STR	__stringify(__ALIGN)
 #endif
 
+#else /* __ASSEMBLY__ */
+
+#define ASM_RET	"ret\n\t"
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* _ASM_X86_LINKAGE_H */

arch/x86/include/asm/paravirt.h +1 -1

@@ -665,7 +665,7 @@ bool __raw_callee_save___native_vcpu_is_
 	    "call " #func ";"						\
 	    PV_RESTORE_ALL_CALLER_REGS					\
 	    FRAME_END							\
-	    "ret;"							\
+	    ASM_RET							\
 	    ".size " PV_THUNK_NAME(func) ", .-" PV_THUNK_NAME(func) ";"	\
 	    ".popsection")

arch/x86/include/asm/qspinlock_paravirt.h +2 -2

@@ -48,7 +48,7 @@ asm    (".pushsection .text;"
 	"jne   .slowpath;"
 	"pop   %rdx;"
 	FRAME_END
-	"ret;"
+	ASM_RET
 	".slowpath: "
 	"push   %rsi;"
 	"movzbl %al,%esi;"

@@ -56,7 +56,7 @@ asm    (".pushsection .text;"
 	"pop    %rsi;"
 	"pop    %rdx;"
 	FRAME_END
-	"ret;"
+	ASM_RET
 	".size " PV_UNLOCK ", .-" PV_UNLOCK ";"
 	".popsection");

arch/x86/kernel/alternative.c +1 -1

@@ -537,7 +537,7 @@ asm (
 "	.type		int3_magic, @function\n"
 "int3_magic:\n"
 "	movl	$1, (%" _ASM_ARG1 ")\n"
-"	ret\n"
+	ASM_RET
 "	.size		int3_magic, .-int3_magic\n"
 "	.popsection\n"
 );

arch/x86/kernel/kprobes/core.c +1 -1

@@ -1044,7 +1044,7 @@ asm(
 	RESTORE_REGS_STRING
 	"	popfl\n"
 #endif
-	"	ret\n"
+	ASM_RET
 	".size kretprobe_trampoline, .-kretprobe_trampoline\n"
 );
 NOKPROBE_SYMBOL(kretprobe_trampoline);

arch/x86/kernel/paravirt.c +1 -1

@@ -41,7 +41,7 @@ extern void _paravirt_nop(void);
 asm (".pushsection .entry.text, \"ax\"\n"
      ".global _paravirt_nop\n"
      "_paravirt_nop:\n\t"
-     "ret\n\t"
+     ASM_RET
      ".size _paravirt_nop, . - _paravirt_nop\n\t"
      ".type _paravirt_nop, @function\n\t"
      ".popsection");

arch/x86/kvm/emulate.c +2 -2

@@ -315,7 +315,7 @@ static int fastop(struct x86_emulate_ctx
 	__FOP_FUNC(#name)
 
 #define __FOP_RET(name) \
-	"ret \n\t" \
+	ASM_RET \
 	".size " name ", .-" name "\n\t"
 
 #define FOP_RET(name) \

@@ -435,7 +435,7 @@ static int fastop(struct x86_emulate_ctx
 	__FOP_RET(#op)
 
 asm(".pushsection .fixup, \"ax\"\n"
-    "kvm_fastop_exception: xor %esi, %esi; ret\n"
+    "kvm_fastop_exception: xor %esi, %esi; " ASM_RET
     ".popsection");
 
 FOP_START(setcc)

arch/x86/lib/error-inject.c +2 -1

@@ -1,5 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 
+#include <linux/linkage.h>
 #include <linux/error-injection.h>
 #include <linux/kprobes.h>

@@ -10,7 +11,7 @@ asm(
 	".type just_return_func, @function\n"
 	".globl just_return_func\n"
 	"just_return_func:\n"
-	"	ret\n"
+		ASM_RET
 	".size just_return_func, .-just_return_func\n"
 );

samples/ftrace/ftrace-direct-modify.c +2 -2

@@ -31,7 +31,7 @@ asm (
 "	call my_direct_func1\n"
 "	leave\n"
 "	.size		my_tramp1, .-my_tramp1\n"
-"	ret\n"
+	ASM_RET
 "	.type		my_tramp2, @function\n"
 "	.globl		my_tramp2\n"
 "   my_tramp2:"

@@ -39,7 +39,7 @@ asm (
 "	movq %rsp, %rbp\n"
 "	call my_direct_func2\n"
 "	leave\n"
-"	ret\n"
+	ASM_RET
 "	.size		my_tramp2, .-my_tramp2\n"
 "	.popsection\n"
 );

samples/ftrace/ftrace-direct-too.c +1 -1

@@ -31,7 +31,7 @@ asm (
 "	popq %rsi\n"
 "	popq %rdi\n"
 "	leave\n"
-"	ret\n"
+	ASM_RET
 "	.size		my_tramp, .-my_tramp\n"
 "	.popsection\n"
 );

samples/ftrace/ftrace-direct.c +1 -1

@@ -24,7 +24,7 @@ asm (
 "	call my_direct_func\n"
 "	popq %rdi\n"
 "	leave\n"
-"	ret\n"
+	ASM_RET
 "	.size		my_tramp, .-my_tramp\n"
 "	.popsection\n"
 );

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #4 lore raw

PATCH 4/21 objtool: Add straight-line-speculation validation

From: Peter Zijlstra <[email protected]>

[ Upstream commit 1cc1e4c8aab4213bd4e6353dec2620476a233d6d ]

Teach objtool to validate the straight-line-speculation constraints:

- speculation trap after indirect calls
- speculation trap after RET

Notable: when an instruction is annotated RETPOLINE_SAFE, indicating
speculation isn't a problem, also don't care about sls for that instruction.

Signed-off-by Peter Zijlstra (Intel) <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 tools/objtool/arch/x86/decode.c         |   13 +++++++++----
 tools/objtool/builtin-check.c           |    3 ++-
 tools/objtool/check.c                   |   14 ++++++++++++++
 tools/objtool/include/objtool/arch.h    |    1 +
 tools/objtool/include/objtool/builtin.h |    2 +-
 5 files changed, 27 insertions(+), 6 deletions(-)

tools/objtool/arch/x86/decode.c +9 -4

@@ -529,6 +529,11 @@ int arch_decode_instruction(const struct
 		}
 		break;
 
+	case 0xcc:
+		/* int3 */
+		*type = INSN_TRAP;
+		break;
+
 	case 0xe3:
 		/* jecxz/jrcxz */
 		*type = INSN_JUMP_CONDITIONAL;

@@ -665,10 +670,10 @@ const char *arch_ret_insn(int len)
 {
 	static const char ret[5][5] = {
 		{ BYTE_RET },
-		{ BYTE_RET, BYTES_NOP1 },
-		{ BYTE_RET, BYTES_NOP2 },
-		{ BYTE_RET, BYTES_NOP3 },
-		{ BYTE_RET, BYTES_NOP4 },
+		{ BYTE_RET, 0xcc },
+		{ BYTE_RET, 0xcc, BYTES_NOP1 },
+		{ BYTE_RET, 0xcc, BYTES_NOP2 },
+		{ BYTE_RET, 0xcc, BYTES_NOP3 },
 	};
 
 	if (len < 1 || len > 5) {

tools/objtool/builtin-check.c +2 -1

@@ -20,7 +20,7 @@
 #include <objtool/objtool.h>
 
 bool no_fp, no_unreachable, retpoline, module, backtrace, uaccess, stats,
-     validate_dup, vmlinux, mcount, noinstr, backup;
+     validate_dup, vmlinux, mcount, noinstr, backup, sls;
 
 static const char * const check_usage[] = {
 	"objtool check [<options>] file.o",

@@ -45,6 +45,7 @@ const struct option check_options[] = {
 	OPT_BOOLEAN('l', "vmlinux", &vmlinux, "vmlinux.o validation"),
 	OPT_BOOLEAN('M', "mcount", &mcount, "generate __mcount_loc"),
 	OPT_BOOLEAN('B', "backup", &backup, "create .orig files before modification"),
+	OPT_BOOLEAN('S', "sls", &sls, "validate straight-line-speculation"),
 	OPT_END(),
 };

tools/objtool/check.c +14 -0

@@ -2776,6 +2776,12 @@ static int validate_branch(struct objtoo
 		switch (insn->type) {
 
 		case INSN_RETURN:
+			if (next_insn && next_insn->type == INSN_TRAP) {
+				next_insn->ignore = true;
+			} else if (sls && !insn->retpoline_safe) {
+				WARN_FUNC("missing int3 after ret",
+					  insn->sec, insn->offset);
+			}
 			return validate_return(func, insn, &state);
 
 		case INSN_CALL:

@@ -2819,6 +2825,14 @@ static int validate_branch(struct objtoo
 			break;
 
 		case INSN_JUMP_DYNAMIC:
+			if (next_insn && next_insn->type == INSN_TRAP) {
+				next_insn->ignore = true;
+			} else if (sls && !insn->retpoline_safe) {
+				WARN_FUNC("missing int3 after indirect jump",
+					  insn->sec, insn->offset);
+			}
+
+			/* fallthrough */
 		case INSN_JUMP_DYNAMIC_CONDITIONAL:
 			if (is_sibling_call(insn)) {
 				ret = validate_sibling_call(insn, &state);

tools/objtool/include/objtool/arch.h +1 -0

@@ -26,6 +26,7 @@ enum insn_type {
 	INSN_CLAC,
 	INSN_STD,
 	INSN_CLD,
+	INSN_TRAP,
 	INSN_OTHER,
 };

tools/objtool/include/objtool/builtin.h +1 -1

@@ -9,7 +9,7 @@
 
 extern const struct option check_options[];
 extern bool no_fp, no_unreachable, retpoline, module, backtrace, uaccess, stats,
-            validate_dup, vmlinux, mcount, noinstr, backup;
+            validate_dup, vmlinux, mcount, noinstr, backup, sls;
 
 extern int cmd_parse_options(int argc, const char **argv, const char * const usage[]);

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #5 lore raw

PATCH 5/21 x86/alternative: Relax text_poke_bp() constraint

From: Peter Zijlstra <[email protected]>

[ Upstream commit 26c44b776dba4ac692a0bf5a3836feb8a63fea6b ]

Currently, text_poke_bp() is very strict to only allow patching a single instruction; however with straight-line-speculation it will be required to patch: ret; int3, which is two instructions.

As such, relax the constraints a little to allow int3 padding for all instructions that do not imply the execution of the next instruction,
ie: RET, JMP.d8 and JMP.d32.

While there, rename the text_poke_loc::rel32 field to ::disp.

Note: this fills up the text_poke_loc structure which is now a round
16 bytes big.

[ bp: Put comments ontop instead of on the side. ]

Signed-off-by Peter Zijlstra (Intel) <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 arch/x86/kernel/alternative.c |   49 +++++++++++++++++++++++++++++-------------
 1 file changed, 34 insertions(+), 15 deletions(-)

arch/x86/kernel/alternative.c +34 ~~-15~~

@@ -930,10 +930,13 @@ void text_poke_sync(void)
 }
 
 struct text_poke_loc {
-	s32 rel_addr; /* addr := _stext + rel_addr */
-	s32 rel32;
+	/* addr := _stext + rel_addr */
+	s32 rel_addr;
+	s32 disp;
+	u8 len;
 	u8 opcode;
 	const u8 text[POKE_MAX_OPCODE_SIZE];
+	/* see text_poke_bp_batch() */
 	u8 old;
 };

@@ -948,7 +951,8 @@ static struct bp_patching_desc *bp_desc;
 static __always_inline
 struct bp_patching_desc *try_get_desc(struct bp_patching_desc **descp)
 {
-	struct bp_patching_desc *desc = __READ_ONCE(*descp); /* rcu_dereference */
+	/* rcu_dereference */
+	struct bp_patching_desc *desc = __READ_ONCE(*descp);
 
 	if (!desc || !arch_atomic_inc_not_zero(&desc->refs))
 		return NULL;

@@ -982,7 +986,7 @@ noinstr int poke_int3_handler(struct pt_
 {
 	struct bp_patching_desc *desc;
 	struct text_poke_loc *tp;
-	int len, ret = 0;
+	int ret = 0;
 	void *ip;
 
 	if (user_mode(regs))

@@ -1022,8 +1026,7 @@ noinstr int poke_int3_handler(struct pt_
 			goto out_put;
 	}
 
-	len = text_opcode_size(tp->opcode);
-	ip += len;
+	ip += tp->len;
 
 	switch (tp->opcode) {
 	case INT3_INSN_OPCODE:

@@ -1038,12 +1041,12 @@ noinstr int poke_int3_handler(struct pt_
 		break;
 
 	case CALL_INSN_OPCODE:
-		int3_emulate_call(regs, (long)ip + tp->rel32);
+		int3_emulate_call(regs, (long)ip + tp->disp);
 		break;
 
 	case JMP32_INSN_OPCODE:
 	case JMP8_INSN_OPCODE:
-		int3_emulate_jmp(regs, (long)ip + tp->rel32);
+		int3_emulate_jmp(regs, (long)ip + tp->disp);
 		break;
 
 	default:

@@ -1118,7 +1121,7 @@ static void text_poke_bp_batch(struct te
 	 */
 	for (do_sync = 0, i = 0; i < nr_entries; i++) {
 		u8 old[POKE_MAX_OPCODE_SIZE] = { tp[i].old, };
-		int len = text_opcode_size(tp[i].opcode);
+		int len = tp[i].len;
 
 		if (len - INT3_INSN_SIZE > 0) {
 			memcpy(old + INT3_INSN_SIZE,

@@ -1195,21 +1198,37 @@ static void text_poke_loc_init(struct te
 			       const void *opcode, size_t len, const void *emulate)
 {
 	struct insn insn;
-	int ret;
+	int ret, i;
 
 	memcpy((void *)tp->text, opcode, len);
 	if (!emulate)
 		emulate = opcode;
 
 	ret = insn_decode_kernel(&insn, emulate);
-
 	BUG_ON(ret < 0);
-	BUG_ON(len != insn.length);
 
 	tp->rel_addr = addr - (void *)_stext;
+	tp->len = len;
 	tp->opcode = insn.opcode.bytes[0];
 
 	switch (tp->opcode) {
+	case RET_INSN_OPCODE:
+	case JMP32_INSN_OPCODE:
+	case JMP8_INSN_OPCODE:
+		/*
+		 * Control flow instructions without implied execution of the
+		 * next instruction can be padded with INT3.
+		 */
+		for (i = insn.length; i < len; i++)
+			BUG_ON(tp->text[i] != INT3_INSN_OPCODE);
+		break;
+
+	default:
+		BUG_ON(len != insn.length);
+	};
+
+
+	switch (tp->opcode) {
 	case INT3_INSN_OPCODE:
 	case RET_INSN_OPCODE:
 		break;

@@ -1217,7 +1236,7 @@ static void text_poke_loc_init(struct te
 	case CALL_INSN_OPCODE:
 	case JMP32_INSN_OPCODE:
 	case JMP8_INSN_OPCODE:
-		tp->rel32 = insn.immediate.value;
+		tp->disp = insn.immediate.value;
 		break;
 
 	default: /* assume NOP */

@@ -1225,13 +1244,13 @@ static void text_poke_loc_init(struct te
 		case 2: /* NOP2 -- emulate as JMP8+0 */
 			BUG_ON(memcmp(emulate, x86_nops[len], len));
 			tp->opcode = JMP8_INSN_OPCODE;
-			tp->rel32 = 0;
+			tp->disp = 0;
 			break;
 
 		case 5: /* NOP5 -- emulate as JMP32+0 */
 			BUG_ON(memcmp(emulate, x86_nops[len], len));
 			tp->opcode = JMP32_INSN_OPCODE;
-			tp->rel32 = 0;
+			tp->disp = 0;
 			break;
 
 		default: /* unknown instruction */

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #6 lore raw

PATCH 6/21 kbuild: move objtool_args back to scripts/Makefile.build

From: Masahiro Yamada <[email protected]>

[ Upstream commit 8f0c32c788fffa8e88f995372415864039347c8a ]

Commit b1a1a1a09b46 ("kbuild: lto: postpone objtool") moved objtool_args to Makefile.lib, so the arguments can be used in Makefile.modfinal as well as Makefile.build.

With commit 850ded46c642 ("kbuild: Fix TRIM_UNUSED_KSYMS with LTO_CLANG"), module LTO linking came back to scripts/Makefile.build again.

So, there is no more reason to keep objtool_args in a separate file.

Get it back to the original place, close to the objtool command.

Remove the stale comment too.

Signed-off-by Masahiro Yamada <[email protected]>

Reviewed-by Kees Cook <[email protected]>

Acked-by Josh Poimboeuf <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 scripts/Makefile.build |   10 ++++++++++
 scripts/Makefile.lib   |   11 -----------
 2 files changed, 10 insertions(+), 11 deletions(-)

scripts/Makefile.build +10 -0

@@ -224,6 +224,16 @@ cmd_record_mcount = $(if $(findstring $(
 endif # CONFIG_FTRACE_MCOUNT_USE_RECORDMCOUNT
 
 ifdef CONFIG_STACK_VALIDATION
+
+objtool_args =								\
+	$(if $(CONFIG_UNWINDER_ORC),orc generate,check)			\
+	$(if $(part-of-module), --module)				\
+	$(if $(CONFIG_FRAME_POINTER),, --no-fp)				\
+	$(if $(CONFIG_GCOV_KERNEL)$(CONFIG_LTO_CLANG), --no-unreachable)\
+	$(if $(CONFIG_RETPOLINE), --retpoline)				\
+	$(if $(CONFIG_X86_SMAP), --uaccess)				\
+	$(if $(CONFIG_FTRACE_MCOUNT_USE_OBJTOOL), --mcount)
+
 ifndef CONFIG_LTO_CLANG
 
 __objtool_obj := $(objtree)/tools/objtool/objtool

scripts/Makefile.lib +0 ~~-11~~

@@ -232,17 +232,6 @@ ifeq ($(CONFIG_LTO_CLANG),y)
 mod-prelink-ext := .lto
 endif
 
-# Objtool arguments are also needed for modfinal with LTO, so we define
-# then here to avoid duplication.
-objtool_args =								\
-	$(if $(CONFIG_UNWINDER_ORC),orc generate,check)			\
-	$(if $(part-of-module), --module)				\
-	$(if $(CONFIG_FRAME_POINTER),, --no-fp)				\
-	$(if $(CONFIG_GCOV_KERNEL)$(CONFIG_LTO_CLANG), --no-unreachable)\
-	$(if $(CONFIG_RETPOLINE), --retpoline)				\
-	$(if $(CONFIG_X86_SMAP), --uaccess)				\
-	$(if $(CONFIG_FTRACE_MCOUNT_USE_OBJTOOL), --mcount)
-
 # Useful for describing the dependency of composite objects
 # Usage:
 #   $(call multi_depend, multi_used_targets, suffix_to_remove, suffix_to_add)

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #7 lore raw

PATCH 7/21 x86: Add straight-line-speculation mitigation

https://gcc.gnu.org/g:53a643f8568067d7700a9f2facc8ba39974973d3
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102952
https://bugs.llvm.org/show_bug.cgi?id=52323

From: Peter Zijlstra <[email protected]>

[ Upstream commit e463a09af2f0677b9485a7e8e4e70b396b2ffb6f ]

Make use of an upcoming GCC feature to mitigate
straight-line-speculation for x86:

It's built tested on x86_64-allyesconfig using GCC-12 and GCC-11.

Maintenance overhead of this should be fairly low due to objtool validation.

Size overhead of all these additional int3 instructions comes to:

text data bss dec hex filename
22267751 6933356 2011368 31212475 1dc43bb defconfig-build/vmlinux
22804126 6933356 1470696 31208178 1dc32f2 defconfig-build/vmlinux.sls

Or roughly 2.4% additional text.

Signed-off-by Peter Zijlstra (Intel) <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 arch/x86/Kconfig                   |   12 ++++++++++++
 arch/x86/Makefile                  |    4 ++++
 arch/x86/include/asm/linkage.h     |   10 ++++++++++
 arch/x86/include/asm/static_call.h |    2 +-
 arch/x86/kernel/ftrace.c           |    2 +-
 arch/x86/kernel/static_call.c      |    5 +++--
 arch/x86/lib/memmove_64.S          |    2 +-
 arch/x86/lib/retpoline.S           |    2 +-
 scripts/Makefile.build             |    3 ++-
 scripts/link-vmlinux.sh            |    3 +++
 10 files changed, 38 insertions(+), 7 deletions(-)

arch/x86/Kconfig +12 -0

@@ -468,6 +468,18 @@ config RETPOLINE
 	  branches. Requires a compiler with -mindirect-branch=thunk-extern
 	  support for full protection. The kernel may run slower.
 
+config CC_HAS_SLS
+	def_bool $(cc-option,-mharden-sls=all)
+
+config SLS
+	bool "Mitigate Straight-Line-Speculation"
+	depends on CC_HAS_SLS && X86_64
+	default n
+	help
+	  Compile the kernel with straight-line-speculation options to guard
+	  against straight line speculation. The kernel image might be slightly
+	  larger.
+
 config X86_CPU_RESCTRL
 	bool "x86 CPU resource control support"
 	depends on X86 && (CPU_SUP_INTEL || CPU_SUP_AMD)

arch/x86/Makefile +4 -0

@@ -179,6 +179,10 @@ ifdef CONFIG_RETPOLINE
   endif
 endif
 
+ifdef CONFIG_SLS
+  KBUILD_CFLAGS += -mharden-sls=all
+endif
+
 KBUILD_LDFLAGS += -m elf_$(UTS_MACHINE)
 
 ifdef CONFIG_LTO_CLANG

arch/x86/include/asm/linkage.h +10 -0

@@ -18,9 +18,19 @@
 #define __ALIGN_STR	__stringify(__ALIGN)
 #endif
 
+#ifdef CONFIG_SLS
+#define RET	ret; int3
+#else
+#define RET	ret
+#endif
+
 #else /* __ASSEMBLY__ */
 
+#ifdef CONFIG_SLS
+#define ASM_RET	"ret; int3\n\t"
+#else
 #define ASM_RET	"ret\n\t"
+#endif
 
 #endif /* __ASSEMBLY__ */

arch/x86/include/asm/static_call.h +1 -1

@@ -35,7 +35,7 @@
 	__ARCH_DEFINE_STATIC_CALL_TRAMP(name, ".byte 0xe9; .long " #func " - (. + 4)")
 
 #define ARCH_DEFINE_STATIC_CALL_NULL_TRAMP(name)			\
-	__ARCH_DEFINE_STATIC_CALL_TRAMP(name, "ret; nop; nop; nop; nop")
+	__ARCH_DEFINE_STATIC_CALL_TRAMP(name, "ret; int3; nop; nop; nop")
 
 
 #define ARCH_ADD_TRAMP_KEY(name)					\

arch/x86/kernel/ftrace.c +1 -1

@@ -308,7 +308,7 @@ union ftrace_op_code_union {
 	} __attribute__((packed));
 };
 
-#define RET_SIZE		1
+#define RET_SIZE		1 + IS_ENABLED(CONFIG_SLS)
 
 static unsigned long
 create_trampoline(struct ftrace_ops *ops, unsigned int *tramp_size)

arch/x86/kernel/static_call.c +3 -2

@@ -16,6 +16,8 @@ enum insn_type {
  */
 static const u8 xor5rax[] = { 0x2e, 0x2e, 0x2e, 0x31, 0xc0 };
 
+static const u8 retinsn[] = { RET_INSN_OPCODE, 0xcc, 0xcc, 0xcc, 0xcc };
+
 static void __ref __static_call_transform(void *insn, enum insn_type type, void *func)
 {
 	const void *emulate = NULL;

@@ -41,8 +43,7 @@ static void __ref __static_call_transfor
 		break;
 
 	case RET:
-		code = text_gen_insn(RET_INSN_OPCODE, insn, func);
-		size = RET_INSN_SIZE;
+		code = &retinsn;
 		break;
 	}

arch/x86/lib/memmove_64.S +1 -1

@@ -40,7 +40,7 @@ SYM_FUNC_START(__memmove)
 	/* FSRM implies ERMS => no length checks, do the copy directly */
 .Lmemmove_begin_forward:
 	ALTERNATIVE "cmp $0x20, %rdx; jb 1f", "", X86_FEATURE_FSRM
-	ALTERNATIVE "", "movq %rdx, %rcx; rep movsb; RET", X86_FEATURE_ERMS
+	ALTERNATIVE "", __stringify(movq %rdx, %rcx; rep movsb; RET), X86_FEATURE_ERMS
 
 	/*
 	 * movsq instruction have many startup latency

arch/x86/lib/retpoline.S +1 -1

@@ -34,7 +34,7 @@ SYM_FUNC_START(__x86_indirect_thunk_\reg
 
 	ALTERNATIVE_2 __stringify(ANNOTATE_RETPOLINE_SAFE; jmp *%\reg), \
 		      __stringify(RETPOLINE \reg), X86_FEATURE_RETPOLINE, \
-		      __stringify(lfence; ANNOTATE_RETPOLINE_SAFE; jmp *%\reg), X86_FEATURE_RETPOLINE_LFENCE
+		      __stringify(lfence; ANNOTATE_RETPOLINE_SAFE; jmp *%\reg; int3), X86_FEATURE_RETPOLINE_LFENCE
 
 SYM_FUNC_END(__x86_indirect_thunk_\reg)

scripts/Makefile.build +2 -1

@@ -232,7 +232,8 @@ objtool_args =								\
 	$(if $(CONFIG_GCOV_KERNEL)$(CONFIG_LTO_CLANG), --no-unreachable)\
 	$(if $(CONFIG_RETPOLINE), --retpoline)				\
 	$(if $(CONFIG_X86_SMAP), --uaccess)				\
-	$(if $(CONFIG_FTRACE_MCOUNT_USE_OBJTOOL), --mcount)
+	$(if $(CONFIG_FTRACE_MCOUNT_USE_OBJTOOL), --mcount)		\
+	$(if $(CONFIG_SLS), --sls)
 
 ifndef CONFIG_LTO_CLANG

scripts/link-vmlinux.sh +3 -0

@@ -139,6 +139,9 @@ objtool_link()
 		if [ -n "${CONFIG_X86_SMAP}" ]; then
 			objtoolopt="${objtoolopt} --uaccess"
 		fi
+		if [ -n "${CONFIG_SLS}" ]; then
+			objtoolopt="${objtoolopt} --sls"
+		fi
 		info OBJTOOL ${1}
 		tools/objtool/objtool ${objtoolcmd} ${objtoolopt} ${1}
 	fi

Greg Kroah-Hartman

author maintainer

2022-05-13 14:32 UTC #8 lore raw

PATCH 8/21 tools arch: Update arch/x86/lib/mem{cpy,set}_64.S copies used in perf bench mem memcpy

From: Arnaldo Carvalho de Melo <[email protected]>

[ Upstream commit 35cb8c713a496e8c114eed5e2a5a30b359876df2 ]

To bring in the change made in this cset:

f94909ceb1ed4bfd ("x86: Prepare asm files for straight-line-speculation")

It silences these perf tools build warnings, no change in the tools:

Warning: Kernel ABI header at 'tools/arch/x86/lib/memcpy_64.S' differs from latest version at 'arch/x86/lib/memcpy_64.S' diff -u tools/arch/x86/lib/memcpy_64.S arch/x86/lib/memcpy_64.S
Warning: Kernel ABI header at 'tools/arch/x86/lib/memset_64.S' differs from latest version at 'arch/x86/lib/memset_64.S' diff -u tools/arch/x86/lib/memset_64.S arch/x86/lib/memset_64.S

The code generated was checked before and after using 'objdump -d /tmp/build/perf/bench/mem-memcpy-x86-64-asm.o', no changes.

Cc Borislav Petkov <[email protected]>

Cc Peter Zijlstra <[email protected]>

Signed-off-by Arnaldo Carvalho de Melo <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 tools/arch/x86/lib/memcpy_64.S |   12 ++++++------
 tools/arch/x86/lib/memset_64.S |    6 +++---
 2 files changed, 9 insertions(+), 9 deletions(-)

tools/arch/x86/lib/memcpy_64.S +6 -6

@@ -39,7 +39,7 @@ SYM_FUNC_START_WEAK(memcpy)
 	rep movsq
 	movl %edx, %ecx
 	rep movsb
-	ret
+	RET
 SYM_FUNC_END(memcpy)
 SYM_FUNC_END_ALIAS(__memcpy)
 EXPORT_SYMBOL(memcpy)

@@ -53,7 +53,7 @@ SYM_FUNC_START_LOCAL(memcpy_erms)
 	movq %rdi, %rax
 	movq %rdx, %rcx
 	rep movsb
-	ret
+	RET
 SYM_FUNC_END(memcpy_erms)
 
 SYM_FUNC_START_LOCAL(memcpy_orig)

@@ -137,7 +137,7 @@ SYM_FUNC_START_LOCAL(memcpy_orig)
 	movq %r9,	1*8(%rdi)
 	movq %r10,	-2*8(%rdi, %rdx)
 	movq %r11,	-1*8(%rdi, %rdx)
-	retq
+	RET
 	.p2align 4
 .Lless_16bytes:
 	cmpl $8,	%edx

@@ -149,7 +149,7 @@ SYM_FUNC_START_LOCAL(memcpy_orig)
 	movq -1*8(%rsi, %rdx),	%r9
 	movq %r8,	0*8(%rdi)
 	movq %r9,	-1*8(%rdi, %rdx)
-	retq
+	RET
 	.p2align 4
 .Lless_8bytes:
 	cmpl $4,	%edx

@@ -162,7 +162,7 @@ SYM_FUNC_START_LOCAL(memcpy_orig)
 	movl -4(%rsi, %rdx), %r8d
 	movl %ecx, (%rdi)
 	movl %r8d, -4(%rdi, %rdx)
-	retq
+	RET
 	.p2align 4
 .Lless_3bytes:
 	subl $1, %edx

@@ -180,7 +180,7 @@ SYM_FUNC_START_LOCAL(memcpy_orig)
 	movb %cl, (%rdi)
 
 .Lend:
-	retq
+	RET
 SYM_FUNC_END(memcpy_orig)
 
 .popsection

tools/arch/x86/lib/memset_64.S +3 -3

@@ -40,7 +40,7 @@ SYM_FUNC_START(__memset)
 	movl %edx,%ecx
 	rep stosb
 	movq %r9,%rax
-	ret
+	RET
 SYM_FUNC_END(__memset)
 SYM_FUNC_END_ALIAS(memset)
 EXPORT_SYMBOL(memset)

@@ -63,7 +63,7 @@ SYM_FUNC_START_LOCAL(memset_erms)
 	movq %rdx,%rcx
 	rep stosb
 	movq %r9,%rax
-	ret
+	RET
 SYM_FUNC_END(memset_erms)
 
 SYM_FUNC_START_LOCAL(memset_orig)

@@ -125,7 +125,7 @@ SYM_FUNC_START_LOCAL(memset_orig)
 
 .Lende:
 	movq	%r10,%rax
-	ret
+	RET
 
 .Lbad_alignment:
 	cmpq $7,%rdx

Greg Kroah-Hartman

author maintainer

2022-05-13 14:32 UTC #9 lore raw

PATCH 9/21 kvm/emulate: Fix SETcc emulation function offsets with SLS

From: Borislav Petkov <[email protected]>

[ Upstream commit fe83f5eae432ccc8e90082d6ed506d5233547473 ]

The commit in Fixes started adding INT3 after RETs as a mitigation against straight-line speculation.

The fastop SETcc implementation in kvm's insn emulator uses macro magic to generate all possible SETcc functions and to jump to them when emulating the respective instruction.

However, it hardcodes the size and alignment of those functions to 4: a three-byte SETcc insn and a single-byte RET. BUT, with SLS, there's an INT3 that gets slapped after the RET, which brings the whole scheme out of alignment:

  15:   0f 90 c0                seto   %al
  18:   c3                      ret
  19:   cc                      int3
  1a:   0f 1f 00                nopl   (%rax)
  1d:   0f 91 c0                setno  %al
  20:   c3                      ret
  21:   cc                      int3
  22:   0f 1f 00                nopl   (%rax)
  25:   0f 92 c0                setb   %al
  28:   c3                      ret
  29:   cc                      int3

and this explodes like this:

int3: 0000 [#1] PREEMPT SMP PTI

  CPU: 0 PID: 2435 Comm: qemu-system-x86 Not tainted 5.17.0-rc8-sls #1
  Hardware name: Dell Inc. Precision WorkStation T3400  /0TP412, BIOS A14 04/30/2012
  RIP: 0010:setc+0x5/0x8 [kvm]
  Code: 00 00 0f 1f 00 0f b6 05 43 24 06 00 c3 cc 0f 1f 80 00 00 00 00 0f 90 c0 c3 cc 0f \
	  1f 00 0f 91 c0 c3 cc 0f 1f 00 0f 92 c0 c3 cc <0f> 1f 00 0f 93 c0 c3 cc 0f 1f 00 \
	  0f 94 c0 c3 cc 0f 1f 00 0f 95 c0
  Call Trace:
   <TASK>
   ? x86_emulate_insn [kvm]
   ? x86_emulate_instruction [kvm]
   ? vmx_handle_exit [kvm_intel]
   ? kvm_arch_vcpu_ioctl_run [kvm]
   ? kvm_vcpu_ioctl [kvm]
   ? __x64_sys_ioctl
   ? do_syscall_64
   ? entry_SYSCALL_64_after_hwframe
   </TASK>

Raise the alignment value when SLS is enabled and use a macro for that instead of hard-coding naked numbers.

Fixes e463a09af2f0 ("x86: Add straight-line-speculation mitigation")

Reported-by Jamie Heilman <[email protected]>

Acked-by Peter Zijlstra (Intel) <[email protected]>

Tested-by Jamie Heilman <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

[Add a comment and a bit of safety checking, since this is going to be changed
again for IBT support. - Paolo]

Signed-off-by Paolo Bonzini <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 arch/x86/kvm/emulate.c |   19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

arch/x86/kvm/emulate.c +17 -2

@@ -427,8 +427,23 @@ static int fastop(struct x86_emulate_ctx
 	FOP_END
 
 /* Special case for SETcc - 1 instruction per cc */
+
+/*
+ * Depending on .config the SETcc functions look like:
+ *
+ * SETcc %al   [3 bytes]
+ * RET         [1 byte]
+ * INT3        [1 byte; CONFIG_SLS]
+ *
+ * Which gives possible sizes 4 or 5.  When rounded up to the
+ * next power-of-two alignment they become 4 or 8.
+ */
+#define SETCC_LENGTH	(4 + IS_ENABLED(CONFIG_SLS))
+#define SETCC_ALIGN	(4 << IS_ENABLED(CONFIG_SLS))
+static_assert(SETCC_LENGTH <= SETCC_ALIGN);
+
 #define FOP_SETCC(op) \
-	".align 4 \n\t" \
+	".align " __stringify(SETCC_ALIGN) " \n\t" \
 	".type " #op ", @function \n\t" \
 	#op ": \n\t" \
 	#op " %al \n\t" \

@@ -1053,7 +1068,7 @@ static int em_bsr_c(struct x86_emulate_c
 static __always_inline u8 test_cc(unsigned int condition, unsigned long flags)
 {
 	u8 rc;
-	void (*fop)(void) = (void *)em_setcc + 4 * (condition & 0xf);
+	void (*fop)(void) = (void *)em_setcc + SETCC_ALIGN * (condition & 0xf);
 
 	flags = (flags & EFLAGS_MASK) | X86_EFLAGS_IF;
 	asm("push %[flags]; popf; " CALL_NOSPEC

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #10 lore raw

PATCH 10/21 crypto: x86/poly1305 - Fixup SLS

From: Peter Zijlstra <[email protected]>

[ Upstream commit 7ed7aa4de9421229be6d331ed52d5cd09c99f409 ]

Due to being a perl generated asm file, it got missed by the mass convertion script.

arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_init_x86_64()+0x3a: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_x86_64()+0xf2: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_emit_x86_64()+0x37: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: __poly1305_block()+0x6d: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: __poly1305_init_avx()+0x1e8: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx()+0x18a: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx()+0xaf8: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_emit_avx()+0x99: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx2()+0x18a: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx2()+0x776: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx512()+0x18a: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx512()+0x796: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx512()+0x10bd: missing int3 after ret

Fixes f94909ceb1ed ("x86: Prepare asm files for straight-line-speculation")

Signed-off-by Peter Zijlstra (Intel) <[email protected]>

Signed-off-by Herbert Xu <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 arch/x86/crypto/poly1305-x86_64-cryptogams.pl |   38 +++++++++++++-------------
 1 file changed, 19 insertions(+), 19 deletions(-)

arch/x86/crypto/poly1305-x86_64-cryptogams.pl +19 ~~-19~~

@@ -297,7 +297,7 @@ ___
 $code.=<<___;
 	mov	\$1,%eax
 .Lno_key:
-	ret
+	RET
 ___
 &end_function("poly1305_init_x86_64");

@@ -373,7 +373,7 @@ $code.=<<___;
 .cfi_adjust_cfa_offset	-48
 .Lno_data:
 .Lblocks_epilogue:
-	ret
+	RET
 .cfi_endproc
 ___
 &end_function("poly1305_blocks_x86_64");

@@ -399,7 +399,7 @@ $code.=<<___;
 	mov	%rax,0($mac)	# write result
 	mov	%rcx,8($mac)
 
-	ret
+	RET
 ___
 &end_function("poly1305_emit_x86_64");
 if ($avx) {

@@ -429,7 +429,7 @@ ___
 	&poly1305_iteration();
 $code.=<<___;
 	pop $ctx
-	ret
+	RET
 .size	__poly1305_block,.-__poly1305_block
 
 .type	__poly1305_init_avx,\@abi-omnipotent

@@ -594,7 +594,7 @@ __poly1305_init_avx:
 
 	lea	-48-64($ctx),$ctx	# size [de-]optimization
 	pop %rbp
-	ret
+	RET
 .size	__poly1305_init_avx,.-__poly1305_init_avx
 ___

@@ -747,7 +747,7 @@ $code.=<<___;
 .cfi_restore	%rbp
 .Lno_data_avx:
 .Lblocks_avx_epilogue:
-	ret
+	RET
 .cfi_endproc
 
 .align	32

@@ -1452,7 +1452,7 @@ $code.=<<___	if (!$win64);
 ___
 $code.=<<___;
 	vzeroupper
-	ret
+	RET
 .cfi_endproc
 ___
 &end_function("poly1305_blocks_avx");

@@ -1508,7 +1508,7 @@ $code.=<<___;
 	mov	%rax,0($mac)	# write result
 	mov	%rcx,8($mac)
 
-	ret
+	RET
 ___
 &end_function("poly1305_emit_avx");

@@ -1675,7 +1675,7 @@ $code.=<<___;
 .cfi_restore 	%rbp
 .Lno_data_avx2$suffix:
 .Lblocks_avx2_epilogue$suffix:
-	ret
+	RET
 .cfi_endproc
 
 .align	32

@@ -2201,7 +2201,7 @@ $code.=<<___	if (!$win64);
 ___
 $code.=<<___;
 	vzeroupper
-	ret
+	RET
 .cfi_endproc
 ___
 if($avx > 2 && $avx512) {

@@ -2792,7 +2792,7 @@ $code.=<<___	if (!$win64);
 .cfi_def_cfa_register	%rsp
 ___
 $code.=<<___;
-	ret
+	RET
 .cfi_endproc
 ___

@@ -2893,7 +2893,7 @@ $code.=<<___	if ($flavour =~ /elf32/);
 ___
 $code.=<<___;
 	mov	\$1,%eax
-	ret
+	RET
 .size	poly1305_init_base2_44,.-poly1305_init_base2_44
 ___
 {

@@ -3010,7 +3010,7 @@ poly1305_blocks_vpmadd52:
 	jnz		.Lblocks_vpmadd52_4x
 
 .Lno_data_vpmadd52:
-	ret
+	RET
 .size	poly1305_blocks_vpmadd52,.-poly1305_blocks_vpmadd52
 ___
 }

@@ -3451,7 +3451,7 @@ poly1305_blocks_vpmadd52_4x:
 	vzeroall
 
 .Lno_data_vpmadd52_4x:
-	ret
+	RET
 .size	poly1305_blocks_vpmadd52_4x,.-poly1305_blocks_vpmadd52_4x
 ___
 }

@@ -3824,7 +3824,7 @@ $code.=<<___;
 	vzeroall
 
 .Lno_data_vpmadd52_8x:
-	ret
+	RET
 .size	poly1305_blocks_vpmadd52_8x,.-poly1305_blocks_vpmadd52_8x
 ___
 }

@@ -3861,7 +3861,7 @@ poly1305_emit_base2_44:
 	mov	%rax,0($mac)	# write result
 	mov	%rcx,8($mac)
 
-	ret
+	RET
 .size	poly1305_emit_base2_44,.-poly1305_emit_base2_44
 ___
 }	}	}

@@ -3916,7 +3916,7 @@ xor128_encrypt_n_pad:
 
 .Ldone_enc:
 	mov	$otp,%rax
-	ret
+	RET
 .size	xor128_encrypt_n_pad,.-xor128_encrypt_n_pad
 
 .globl	xor128_decrypt_n_pad

@@ -3967,7 +3967,7 @@ xor128_decrypt_n_pad:
 
 .Ldone_dec:
 	mov	$otp,%rax
-	ret
+	RET
 .size	xor128_decrypt_n_pad,.-xor128_decrypt_n_pad
 ___
 }

@@ -4109,7 +4109,7 @@ avx_handler:
 	pop	%rbx
 	pop	%rdi
 	pop	%rsi
-	ret
+	RET
 .size	avx_handler,.-avx_handler
 
 .section	.pdata

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #11 lore raw

PATCH 11/21 objtool: Fix SLS validation for kcov tail-call replacement

From: Peter Zijlstra <[email protected]>

[ Upstream commit 7a53f408902d913cd541b4f8ad7dbcd4961f5b82 ]

Since not all compilers have a function attribute to disable KCOV instrumentation, objtool can rewrite KCOV instrumentation in noinstr functions as per commit:

f56dae88a81f ("objtool: Handle __sanitize_cov*() tail calls")

However, this has subtle interaction with the SLS validation from commit:

1cc1e4c8aab4 ("objtool: Add straight-line-speculation validation")

In that when a tail-call instrucion is replaced with a RET an additional INT3 instruction is also written, but is not represented in the decoded instruction stream.

This then leads to false positive missing INT3 objtool warnings in noinstr code.

Instead of adding additional struct instruction objects, mark the RET instruction with retpoline_safe to suppress the warning (since we know there really is an INT3).

Fixes 1cc1e4c8aab4 ("objtool: Add straight-line-speculation validation")

Signed-off-by Peter Zijlstra (Intel) <[email protected]>

Link https://lkml.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 tools/objtool/check.c |   10 ++++++++++
 1 file changed, 10 insertions(+)

tools/objtool/check.c +10 -0

@@ -871,6 +871,16 @@ static void add_call_dest(struct objtool
 			               : arch_nop_insn(insn->len));
 
 		insn->type = sibling ? INSN_RETURN : INSN_NOP;
+
+		if (sibling) {
+			/*
+			 * We've replaced the tail-call JMP insn by two new
+			 * insn: RET; INT3, except we only have a single struct
+			 * insn here. Mark it retpoline_safe to avoid the SLS
+			 * warning, instead of adding another insn.
+			 */
+			insn->retpoline_safe = true;
+		}
 	}
 
 	if (mcount && !strcmp(insn->call_dest->name, "__fentry__")) {

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #12 lore raw

PATCH 12/21 Bluetooth: Fix the creation of hdev->name

From: Itay Iellin <[email protected]>

commit 103a2f3255a95991252f8f13375c3a96a75011cd upstream.

Set a size limit of 8 bytes of the written buffer to "hdev->name" including the terminating null byte, as the size of "hdev->name" is 8 bytes. If an id value which is greater than 9999 is allocated, then the "snprintf(hdev->name, sizeof(hdev->name), "hci%d", id)" function call would lead to a truncation of the id value in decimal notation.

Set an explicit maximum id parameter in the id allocation function call. The id allocation function defines the maximum allocated id value as the maximum id parameter value minus one. Therefore, HCI_MAX_ID is defined as 10000.

Signed-off-by Itay Iellin <[email protected]>

Signed-off-by Luiz Augusto von Dentz <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 include/net/bluetooth/hci_core.h |    3 +++
 net/bluetooth/hci_core.c         |    6 +++---
 2 files changed, 6 insertions(+), 3 deletions(-)

include/net/bluetooth/hci_core.h +3 -0

@@ -35,6 +35,9 @@
 /* HCI priority */
 #define HCI_PRIO_MAX	7
 
+/* HCI maximum id value */
+#define HCI_MAX_ID 10000
+
 /* HCI Core structures */
 struct inquiry_data {
 	bdaddr_t	bdaddr;

net/bluetooth/hci_core.c +3 -3

@@ -3907,10 +3907,10 @@ int hci_register_dev(struct hci_dev *hde
 	 */
 	switch (hdev->dev_type) {
 	case HCI_PRIMARY:
-		id = ida_simple_get(&hci_index_ida, 0, 0, GFP_KERNEL);
+		id = ida_simple_get(&hci_index_ida, 0, HCI_MAX_ID, GFP_KERNEL);
 		break;
 	case HCI_AMP:
-		id = ida_simple_get(&hci_index_ida, 1, 0, GFP_KERNEL);
+		id = ida_simple_get(&hci_index_ida, 1, HCI_MAX_ID, GFP_KERNEL);
 		break;
 	default:
 		return -EINVAL;

@@ -3919,7 +3919,7 @@ int hci_register_dev(struct hci_dev *hde
 	if (id < 0)
 		return id;
 
-	sprintf(hdev->name, "hci%d", id);
+	snprintf(hdev->name, sizeof(hdev->name), "hci%d", id);
 	hdev->id = id;
 
 	BT_DBG("%p name %s bus %d", hdev, hdev->name, hdev->bus);

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #13 lore raw

PATCH 13/21 rfkill: uapi: fix RFKILL_IOCTL_MAX_SIZE ioctl request definition

From: Gleb Fotengauer-Malinovskiy <[email protected]>

commit a36e07dfe6ee71e209383ea9288cd8d1617e14f9 upstream.

The definition of RFKILL_IOCTL_MAX_SIZE introduced by commit 54f586a91532 ("rfkill: make new event layout opt-in") is unusable since it is based on RFKILL_IOC_EXT_SIZE which has not been defined. Fix that by replacing the undefined constant with the constant which is intended to be used in this definition.

Fixes 54f586a91532 ("rfkill: make new event layout opt-in")

Cc [email protected] # 5.11+

Signed-off-by Gleb Fotengauer-Malinovskiy <[email protected]>

Signed-off-by Dmitry V. Levin <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

[add commit message provided later by Dmitry]

Signed-off-by Johannes Berg <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 include/uapi/linux/rfkill.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

include/uapi/linux/rfkill.h +1 -1

@@ -184,7 +184,7 @@ struct rfkill_event_ext {
 #define RFKILL_IOC_NOINPUT	1
 #define RFKILL_IOCTL_NOINPUT	_IO(RFKILL_IOC_MAGIC, RFKILL_IOC_NOINPUT)
 #define RFKILL_IOC_MAX_SIZE	2
-#define RFKILL_IOCTL_MAX_SIZE	_IOW(RFKILL_IOC_MAGIC, RFKILL_IOC_EXT_SIZE, __u32)
+#define RFKILL_IOCTL_MAX_SIZE	_IOW(RFKILL_IOC_MAGIC, RFKILL_IOC_MAX_SIZE, __u32)
 
 /* and that's all userspace gets */

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #14 lore raw

PATCH 14/21 udf: Avoid using stale lengthOfImpUse

From: Jan Kara <[email protected]>

commit c1ad35dd0548ce947d97aaf92f7f2f9a202951cf upstream.

udf_write_fi() uses lengthOfImpUse of the entry it is writing to. However this field has not yet been initialized so it either contains completely bogus value or value from last directory entry at that place. In either case this is wrong and can lead to filesystem corruption or kernel crashes.

Reported-by butt3rflyh4ck <[email protected]>

CC [email protected]

Fixes 979a6e28dd96 ("udf: Get rid of 0-length arrays in struct fileIdentDesc")

Signed-off-by Jan Kara <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 fs/udf/namei.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

fs/udf/namei.c +4 -4

@@ -75,11 +75,11 @@ int udf_write_fi(struct inode *inode, st
 
 	if (fileident) {
 		if (adinicb || (offset + lfi < 0)) {
-			memcpy(udf_get_fi_ident(sfi), fileident, lfi);
+			memcpy(sfi->impUse + liu, fileident, lfi);
 		} else if (offset >= 0) {
 			memcpy(fibh->ebh->b_data + offset, fileident, lfi);
 		} else {
-			memcpy(udf_get_fi_ident(sfi), fileident, -offset);
+			memcpy(sfi->impUse + liu, fileident, -offset);
 			memcpy(fibh->ebh->b_data, fileident - offset,
 				lfi + offset);
 		}

@@ -88,11 +88,11 @@ int udf_write_fi(struct inode *inode, st
 	offset += lfi;
 
 	if (adinicb || (offset + padlen < 0)) {
-		memset(udf_get_fi_ident(sfi) + lfi, 0x00, padlen);
+		memset(sfi->impUse + liu + lfi, 0x00, padlen);
 	} else if (offset >= 0) {
 		memset(fibh->ebh->b_data + offset, 0x00, padlen);
 	} else {
-		memset(udf_get_fi_ident(sfi) + lfi, 0x00, -offset);
+		memset(sfi->impUse + liu + lfi, 0x00, -offset);
 		memset(fibh->ebh->b_data, 0x00, padlen + offset);
 	}

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #15 lore raw

PATCH 15/21 mm: fix missing cache flush for all tail pages of compound page

From: Muchun Song <[email protected]>

commit 2771739a7162782c0aa6424b2e3dd874e884a15d upstream.

The D-cache maintenance inside move_to_new_page() only consider one page, there is still D-cache maintenance issue for tail pages of compound page (e.g. THP or HugeTLB).

THP migration is only enabled on x86_64, ARM64 and powerpc, while powerpc and arm64 need to maintain the consistency between I-Cache and D-Cache, which depends on flush_dcache_page() to maintain the consistency between I-Cache and D-Cache.

But there is no issues on arm64 and powerpc since they already considers the compound page cache flushing in their icache flush function. HugeTLB migration is enabled on arm, arm64, mips, parisc, powerpc, riscv, s390 and sh, while arm has handled the compound page cache flush in flush_dcache_page(), but most others do not.

In theory, the issue exists on many architectures. Fix this by not using flush_dcache_folio() since it is not backportable.

Link https://lkml.kernel.org/r/[email protected] open here

Fixes 290408d4a250 ("hugetlb: hugepage migration core")

Signed-off-by Muchun Song <[email protected]>

Reviewed-by Zi Yan <[email protected]>

Cc Axel Rasmussen <[email protected]>

Cc David Rientjes <[email protected]>

Cc Fam Zheng <[email protected]>

Cc Kirill A. Shutemov <[email protected]>

Cc Lars Persson <[email protected]>

Cc Mike Kravetz <[email protected]>

Cc Peter Xu <[email protected]>

Cc Xiongchun Duan <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 mm/migrate.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

mm/migrate.c +5 -2

@@ -948,9 +948,12 @@ static int move_to_new_page(struct page
 		if (!PageMappingFlags(page))
 			page->mapping = NULL;
 
-		if (likely(!is_zone_device_page(newpage)))
-			flush_dcache_page(newpage);
+		if (likely(!is_zone_device_page(newpage))) {
+			int i, nr = compound_nr(newpage);
 
+			for (i = 0; i < nr; i++)
+				flush_dcache_page(newpage + i);
+		}
 	}
 out:
 	return rc;

Greg Kroah-Hartman

author maintainer

2022-05-13 14:32 UTC #16 lore raw

PATCH 16/21 mm: hugetlb: fix missing cache flush in copy_huge_page_from_user()

From: Muchun Song <[email protected]>

commit e763243cc6cb1fcc720ec58cfd6e7c35ae90a479 upstream.

userfaultfd calls copy_huge_page_from_user() which does not do any cache flushing for the target page. Then the target page will be mapped to the user space with a different address (user address), which might have an alias issue with the kernel address used to copy the data from the user to.

Fix this issue by flushing dcache in copy_huge_page_from_user().

Link https://lkml.kernel.org/r/[email protected] open here

Fixes fa4d75c1de13 ("userfaultfd: hugetlbfs: add copy_huge_page_from_user for hugetlb userfaultfd support")

Signed-off-by Muchun Song <[email protected]>

Reviewed-by Mike Kravetz <[email protected]>

Cc Axel Rasmussen <[email protected]>

Cc David Rientjes <[email protected]>

Cc Fam Zheng <[email protected]>

Cc Kirill A. Shutemov <[email protected]>

Cc Lars Persson <[email protected]>

Cc Peter Xu <[email protected]>

Cc Xiongchun Duan <[email protected]>

Cc Zi Yan <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 mm/memory.c |    2 ++
 1 file changed, 2 insertions(+)

mm/memory.c +2 -0

@@ -5467,6 +5467,8 @@ long copy_huge_page_from_user(struct pag
 		if (rc)
 			break;
 
+		flush_dcache_page(subpage);
+
 		cond_resched();
 	}
 	return ret_val;

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #17 lore raw

PATCH 17/21 mm: shmem: fix missing cache flush in shmem_mfill_atomic_pte()

From: Muchun Song <[email protected]>

commit 19b482c29b6f3805f1d8e93015847b89e2f7f3b1 upstream.

userfaultfd calls shmem_mfill_atomic_pte() which does not do any cache flushing for the target page. Then the target page will be mapped to the user space with a different address (user address), which might have an alias issue with the kernel address used to copy the data from the user to. Insert flush_dcache_page() in non-zero-page case. And replace clear_highpage() with clear_user_highpage() which already considers the cache maintenance.

Link https://lkml.kernel.org/r/[email protected] open here

Fixes 8d1039634206 ("userfaultfd: shmem: add shmem_mfill_zeropage_pte for userfaultfd support")

Fixes 4c27fe4c4c84 ("userfaultfd: shmem: add shmem_mcopy_atomic_pte for userfaultfd support")

Signed-off-by Muchun Song <[email protected]>

Reviewed-by Mike Kravetz <[email protected]>

Cc Axel Rasmussen <[email protected]>

Cc David Rientjes <[email protected]>

Cc Fam Zheng <[email protected]>

Cc Kirill A. Shutemov <[email protected]>

Cc Lars Persson <[email protected]>

Cc Peter Xu <[email protected]>

Cc Xiongchun Duan <[email protected]>

Cc Zi Yan <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 mm/shmem.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

mm/shmem.c +3 -1

@@ -2394,8 +2394,10 @@ int shmem_mfill_atomic_pte(struct mm_str
 				/* don't free the page */
 				goto out_unacct_blocks;
 			}
+
+			flush_dcache_page(page);
 		} else {		/* ZEROPAGE */
-			clear_highpage(page);
+			clear_user_highpage(page, dst_addr);
 		}
 	} else {
 		page = *pagep;

Greg Kroah-Hartman

author maintainer

2022-05-13 14:32 UTC #18 lore raw

PATCH 18/21 mm: userfaultfd: fix missing cache flush in mcopy_atomic_pte() and __mcopy_atomic()

From: Muchun Song <[email protected]>

commit 7c25a0b89a487878b0691e6524fb5a8827322194 upstream.

userfaultfd calls mcopy_atomic_pte() and __mcopy_atomic() which do not do any cache flushing for the target page. Then the target page will be mapped to the user space with a different address (user address), which might have an alias issue with the kernel address used to copy the data from the user to. Fix this by insert flush_dcache_page() after copy_from_user() succeeds.

Link https://lkml.kernel.org/r/[email protected] open here

Fixes b6ebaedb4cb1 ("userfaultfd: avoid mmap_sem read recursion in mcopy_atomic")

Fixes c1a4de99fada ("userfaultfd: mcopy_atomic|mfill_zeropage: UFFDIO_COPY|UFFDIO_ZEROPAGE preparation")

Signed-off-by Muchun Song <[email protected]>

Cc Axel Rasmussen <[email protected]>

Cc David Rientjes <[email protected]>

Cc Fam Zheng <[email protected]>

Cc Kirill A. Shutemov <[email protected]>

Cc Lars Persson <[email protected]>

Cc Mike Kravetz <[email protected]>

Cc Peter Xu <[email protected]>

Cc Xiongchun Duan <[email protected]>

Cc Zi Yan <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 mm/userfaultfd.c |    3 +++
 1 file changed, 3 insertions(+)

mm/userfaultfd.c +3 -0

@@ -151,6 +151,8 @@ static int mcopy_atomic_pte(struct mm_st
 			/* don't free the page */
 			goto out;
 		}
+
+		flush_dcache_page(page);
 	} else {
 		page = *pagep;
 		*pagep = NULL;

@@ -621,6 +623,7 @@ retry:
 				err = -EFAULT;
 				goto out;
 			}
+			flush_dcache_page(page);
 			goto retry;
 		} else
 			BUG_ON(page);

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #19 lore raw

PATCH 19/21 mm/hwpoison: fix error page recovered but reported "not recovered"

From: Naoya Horiguchi <[email protected]>

commit 046545a661af2beec21de7b90ca0e35f05088a81 upstream.

When an uncorrected memory error is consumed there is a race between the CMCI from the memory controller reporting an uncorrected error with a UCNA signature, and the core reporting and SRAR signature machine check when the data is about to be consumed.

If the CMCI wins that race, the page is marked poisoned when uc_decode_notifier() calls memory_failure() and the machine check processing code finds the page already poisoned. It calls kill_accessing_process() to make sure a SIGBUS is sent. But returns the wrong error code.

Console log looks like this:

mce: Uncorrected hardware memory error in user-access at 3710b3400
Memory failure: 0x3710b3: recovery action for dirty LRU page: Recovered
Memory failure: 0x3710b3: already hardware poisoned
Memory failure: 0x3710b3: Sending SIGBUS to einj_mem_uc:361438 due to hardware memory corruption
mce: Memory error not recovered

kill_accessing_process() is supposed to return -EHWPOISON to notify that SIGBUS is already set to the process and kill_me_maybe() doesn't have to send it again. But current code simply fails to do this, so fix it to make sure to work as intended. This change avoids the noise message "Memory error not recovered" and skips duplicate SIGBUSs.

[[email protected]: reword some parts of commit message]

Link https://lkml.kernel.org/r/[email protected] open here

Fixes a3f5d80ea401 ("mm,hwpoison: send SIGBUS with error virutal address")

Signed-off-by Naoya Horiguchi <[email protected]>

Reported-by Youquan Song <[email protected]>

Cc Tony Luck <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 mm/memory-failure.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

mm/memory-failure.c +3 -1

@@ -705,8 +705,10 @@ static int kill_accessing_process(struct
 			      (void *)&priv);
 	if (ret == 1 && priv.tk.addr)
 		kill_proc(&priv.tk, pfn, flags);
+	else
+		ret = 0;
 	mmap_read_unlock(p->mm);
-	return ret ? -EFAULT : -EHWPOISON;
+	return ret > 0 ? -EHWPOISON : -EFAULT;
 }
 
 static const char *action_name[] = {

Greg Kroah-Hartman

author maintainer

2022-05-13 14:32 UTC #20 lore raw

PATCH 20/21 mm/mlock: fix potential imbalanced rlimit ucounts adjustment

From: Miaohe Lin <[email protected]>

commit 5c2a956c3eea173b2bc89f632507c0eeaebf6c4a upstream.

user_shm_lock forgets to set allowed to 0 when get_ucounts fails. So the later user_shm_unlock might do the extra dec_rlimit_ucounts. Fix this by resetting allowed to 0.

Link https://lkml.kernel.org/r/[email protected] open here

Fixes d7c9e99aee48 ("Reimplement RLIMIT_MEMLOCK on top of ucounts")

Signed-off-by Miaohe Lin <[email protected]>

Reviewed-by Andrew Morton <[email protected]>

Acked-by Hugh Dickins <[email protected]>

Cc Herbert van den Bergh <[email protected]>

Cc Chris Mason <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 mm/mlock.c |    1 +
 1 file changed, 1 insertion(+)

mm/mlock.c +1 -0

@@ -837,6 +837,7 @@ int user_shm_lock(size_t size, struct uc
 	}
 	if (!get_ucounts(ucounts)) {
 		dec_rlimit_ucounts(ucounts, UCOUNT_RLIMIT_MEMLOCK, locked);
+		allowed = 0;
 		goto out;
 	}
 	allowed = 1;

Greg Kroah-Hartman

author maintainer

2022-05-13 14:31 UTC #21 lore raw

PATCH 21/21 mm: fix invalid page pointer returned with FOLL_PIN gups

From: Peter Xu <[email protected]>

commit 7196040e19ad634293acd3eff7083149d7669031 upstream.

Patch series "mm/gup: some cleanups", v5.

This patch (of 5):

Alex reported invalid page pointer returned with pin_user_pages_remote() from vfio after upstream commit 4b6c33b32296 ("vfio/type1: Prepare for batched pinning with struct vfio_batch").

It turns out that it's not the fault of the vfio commit; however after vfio switches to a full page buffer to store the page pointers it starts to expose the problem easier.

The problem is for VM_PFNMAP vmas we should normally fail with an -EFAULT then vfio will carry on to handle the MMIO regions. However when the bug triggered, follow_page_mask() returned -EEXIST for such a page, which will jump over the current page, leaving that entry in **pages untouched. However the caller is not aware of it, hence the caller will reference the page as usual even if the pointer data can be anything.

We had that -EEXIST logic since commit 1027e4436b6a ("mm: make GUP handle pfn mapping unless FOLL_GET is requested") which seems very reasonable. It could be that when we reworked GUP with FOLL_PIN we could have overlooked that special path in commit 3faa52c03f44 ("mm/gup:
track FOLL_PIN pages"), even if that commit rightfully touched up follow_devmap_pud() on checking FOLL_PIN when it needs to return an -EEXIST.

Attaching the Fixes to the FOLL_PIN rework commit, as it happened later than 1027e4436b6a.

[[email protected]: added some tags, removed a reference to an out of tree module.]

Link https://lkml.kernel.org/r/[email protected] open here

Fixes 3faa52c03f44 ("mm/gup: track FOLL_PIN pages")

Signed-off-by Peter Xu <[email protected]>

Signed-off-by John Hubbard <[email protected]>

Reviewed-by Claudio Imbrenda <[email protected]>

Reported-by Alex Williamson <[email protected]>

Debugged-by Alex Williamson <[email protected]>

Tested-by Alex Williamson <[email protected]>

Reviewed-by Christoph Hellwig <[email protected]>

Reviewed-by Jan Kara <[email protected]>

Cc Andrea Arcangeli <[email protected]>

Cc Kirill A. Shutemov <[email protected]>

Cc Jason Gunthorpe <[email protected]>

Cc David Hildenbrand <[email protected]>

Cc Lukas Bulwahn <[email protected]>

Cc Matthew Wilcox (Oracle) <[email protected]>

Cc Jason Gunthorpe <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 mm/gup.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

mm/gup.c +1 -1

@@ -465,7 +465,7 @@ static int follow_pfn_pte(struct vm_area
 		pte_t *pte, unsigned int flags)
 {
 	/* No page to get reference */
-	if (flags & FOLL_GET)
+	if (flags & (FOLL_GET | FOLL_PIN))
 		return -EFAULT;
 
 	if (flags & FOLL_TOUCH) {

Jon Hunter

maintainer Tested-by on the v1 series

2022-05-13 16:40 UTC #22 lore raw

Quoted 9 lines · Greg Kroah-Hartman

Quoted 8 lines

All tests passing for Tegra ...

Test results for stable-v5.15:

    10 builds:	10 pass, 0 fail
    28 boots:	28 pass, 0 fail
    114 tests:	114 pass, 0 fail

Linux version: 5.15.40-rc1-g13b089c28632
Boards tested: tegra124-jetson-tk1, tegra186-p2771-0000,

                tegra194-p2972-0000, tegra194-p3509-0000+p3668-0000,
                tegra20-ventana, tegra210-p2371-2180,
                tegra210-p3450-0000, tegra30-cardhu-a04

Tested-by Jon Hunter <[email protected]>

Jon

Shuah Khan

Tested-by on the v1 series

2022-05-13 20:38 UTC #23 lore raw

Quoted 9 lines · Greg Kroah-Hartman

Quoted 9 lines

Compiled and booted on my test system. No dmesg regressions.

Tested-by Shuah Khan <[email protected]>

thanks,
-- Shuah

Florian Fainelli

Tested-by on the v1 series

2022-05-14 03:19 UTC #24 lore raw

Quoted 9 lines · Greg Kroah-Hartman

Quoted 8 lines

On ARCH_BRCMSTB using 32-bit and 64-bit ARM kernels:

Tested-by Florian Fainelli <[email protected]>

Signature · 1 line

Florian

Fox Chen

Tested-by on the v1 series

2022-05-14 05:35 UTC #25 lore raw

Quoted 9 lines · Greg Kroah-Hartman

Quoted 9 lines

5.15.40-rc1 Successfully Compiled and booted on my Raspberry PI 4b (8g) (bcm2711)

Tested-by Fox Chen <[email protected]>

Ron Economos

Tested-by on the v1 series

2022-05-14 11:38 UTC #26 lore raw

Quoted 9 lines · Greg Kroah-Hartman

Quoted 8 lines

Built and booted successfully on RISC-V RV64 (HiFive Unmatched).

Tested-by Ron Economos <[email protected]>

Naresh Kamboju

Tested-by Linux Kernel Functional Testing on the v1 series

2022-05-14 11:41 UTC #27 lore raw

Quoted 10 lines · Greg Kroah-Hartman

Quoted 8 lines

Results from Linaro’s test farm.
No regressions on arm64, arm, x86_64, and i386.

Tested-by Linux Kernel Functional Testing <[email protected]>

## Build
* kernel: 5.15.40-rc1
* git: https://gitlab.com/Linaro/lkft/mirrors/stable/linux-stable-rc
* git branch: linux-5.15.y
* git commit: 13b089c28632ad5a051bdbb1951ee189f96b01fd
* git describe: v5.15.39-22-g13b089c28632
* test details:
https://qa-reports.linaro.org/lkft/linux-stable-rc-linux-5.15.y/build/v5.15.39-22-g13b089c28632

## Test Regressions (compared to v5.15.38-136-g60041d098524) No test regressions found.

## Metric Regressions (compared to v5.15.38-136-g60041d098524) No metric regressions found.

## Test Fixes (compared to v5.15.38-136-g60041d098524) No test fixes found.

## Metric Fixes (compared to v5.15.38-136-g60041d098524) No metric fixes found.

## Test result summary
total: 110223, pass: 92720, fail: 1126, skip: 15100, xfail: 1277

## Build Summary
* arc: 10 total, 10 passed, 0 failed
* arm: 296 total, 293 passed, 3 failed
* arm64: 47 total, 47 passed, 0 failed
* dragonboard-410c: 1 total, 1 passed, 0 failed
* hi6220-hikey: 1 total, 1 passed, 0 failed
* i386: 45 total, 41 passed, 4 failed
* juno-r2: 1 total, 1 passed, 0 failed
* mips: 41 total, 38 passed, 3 failed
* parisc: 14 total, 14 passed, 0 failed
* powerpc: 59 total, 56 passed, 3 failed
* riscv: 27 total, 27 passed, 0 failed
* s390: 26 total, 23 passed, 3 failed
* sh: 26 total, 24 passed, 2 failed
* sparc: 14 total, 14 passed, 0 failed
* x15: 1 total, 1 passed, 0 failed
* x86: 1 total, 1 passed, 0 failed
* x86_64: 47 total, 47 passed, 0 failed

## Test suites summary
* fwts
* igt-gpu-tools
* kselftest-
* kselftest-android
* kselftest-arm64
* kselftest-bpf
* kselftest-breakpoints
* kselftest-capabilities
* kselftest-cgroup
* kselftest-clone3
* kselftest-core
* kselftest-cpu-hotplug
* kselftest-cpufreq
* kselftest-drivers
* kselftest-efivarfs
* kselftest-filesystems
* kselftest-firmware
* kselftest-fpu
* kselftest-futex
* kselftest-gpio
* kselftest-intel_pstate
* kselftest-ipc
* kselftest-ir
* kselftest-kcmp
* kselftest-kexec
* kselftest-kvm
* kselftest-lib
* kselftest-livepatch
* kselftest-membarrier
* kselftest-memfd
* kselftest-memory-hotplug
* kselftest-mincore
* kselftest-mount
* kselftest-mqueue
* kselftest-net
* kselftest-netfilter
* kselftest-nsfs
* kselftest-openat2
* kselftest-pid_namespace
* kselftest-pidfd
* kselftest-proc
* kselftest-pstore
* kselftest-ptrace
* kselftest-rseq
* kselftest-rtc
* kselftest-seccomp
* kselftest-sigaltstack
* kselftest-size
* kselftest-splice
* kselftest-static_keys
* kselftest-sync
* kselftest-sysctl
* kselftest-tc-testing
* kselftest-timens
* kselftest-timers
* kselftest-tmpfs
* kselftest-tpm2
* kselftest-user
* kselftest-vm
* kselftest-x86
* kselftest-zram
* kunit
* kvm-unit-tests
* libgpiod
* libhugetlbfs
* linux-log-parser
* ltp-cap_bounds-tests
* ltp-commands-tests
* ltp-containers-tests
* ltp-controllers-tests
* ltp-cpuhotplug-tests
* ltp-crypto-tests
* ltp-cve-tests
* ltp-dio-tests
* ltp-fcntl-locktests-tests
* ltp-filecaps-tests
* ltp-fs-tests
* ltp-fs_bind-tests
* ltp-fs_perms_simple-tests
* ltp-fsx-tests
* ltp-hugetlb-tests
* ltp-io-tests
* ltp-ipc-tests
* ltp-math-tests
* ltp-mm-tests
* ltp-nptl-tests
* ltp-open-posix-tests
* ltp-pty-tests
* ltp-sched-tests
* ltp-securebits-tests
* ltp-syscalls-tests
* ltp-tracing-tests
* network-basic-tests
* packetdrill
* perf
* rcutorture
* ssuite
* v4l2-compliance
* vdso

--
Linaro LKFT
https://lkft.linaro.org

Sudip Mukherjee

Tested-by on the v1 series

2022-05-14 14:26 UTC #28 lore raw

Hi Greg,

Quoted 7 lines · Greg Kroah-Hartman

Build test:
mips (gcc version 11.2.1 20220408): 62 configs -> no failure arm (gcc version 11.2.1 20220408): 100 configs -> no new failure arm64 (gcc version 11.2.1 20220408): 3 configs -> no failure x86_64 (gcc version 11.2.1 20220408): 4 configs -> no failure

Boot test:
x86_64: Booted on my test laptop. No regression.
x86_64: Booted on qemu. No regression. [1]
arm64: Booted on rpi4b (4GB model). No regression. [2]
mips: Booted on ci20 board. No regression. [3]

[1]. https://openqa.qa.codethink.co.uk/tests/1141
[2]. https://openqa.qa.codethink.co.uk/tests/1145
[3]. https://openqa.qa.codethink.co.uk/tests/1142

Tested-by Sudip Mukherjee <[email protected]>

--
Regards
Sudip

Guenter Roeck

Tested-by on the v1 series

2022-05-14 14:57 UTC #29 lore raw