diff options
| author | Max Filippov <jcmvbkbc@gmail.com> | 2018-12-05 12:48:19 -0800 |
|---|---|---|
| committer | Max Filippov <jcmvbkbc@gmail.com> | 2018-12-05 12:53:07 -0800 |
| commit | 037602705109ec2ab96340bea93ad87daa3ac046 (patch) | |
| tree | 616c10e2e60b8a6bfbe388366ea1c1c3f170dafb /arch/xtensa/include | |
| parent | f37598be4e3896359e87c824be57ddddc280cc3f (diff) | |
| download | linux-037602705109ec2ab96340bea93ad87daa3ac046.tar.gz | |
xtensa: don't use l32r opcode directly
xtensa assembler is capable of representing register loads with either
movi + addmi, l32r or const16, depending on the core configuration.
Don't use '.literal' and 'l32r' directly in the code, use 'movi' and let
the assembler relax them.
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
Diffstat (limited to 'arch/xtensa/include')
| -rw-r--r-- | arch/xtensa/include/asm/futex.h | 8 | ||||
| -rw-r--r-- | arch/xtensa/include/asm/uaccess.h | 10 |
2 files changed, 8 insertions, 10 deletions
diff --git a/arch/xtensa/include/asm/futex.h b/arch/xtensa/include/asm/futex.h index 5bfbc1c401d4..fd0eef6b8e7c 100644 --- a/arch/xtensa/include/asm/futex.h +++ b/arch/xtensa/include/asm/futex.h @@ -32,8 +32,8 @@ "3:\n" \ " .section .fixup,\"ax\"\n" \ " .align 4\n" \ - "4: .long 3b\n" \ - "5: l32r %0, 4b\n" \ + " .literal_position\n" \ + "5: movi %0, 3b\n" \ " movi %1, %3\n" \ " jx %0\n" \ " .previous\n" \ @@ -108,8 +108,8 @@ futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, "2:\n" " .section .fixup,\"ax\"\n" " .align 4\n" - "3: .long 2b\n" - "4: l32r %1, 3b\n" + " .literal_position\n" + "4: movi %1, 2b\n" " movi %0, %7\n" " jx %1\n" " .previous\n" diff --git a/arch/xtensa/include/asm/uaccess.h b/arch/xtensa/include/asm/uaccess.h index f1158b4c629c..d11ef2939652 100644 --- a/arch/xtensa/include/asm/uaccess.h +++ b/arch/xtensa/include/asm/uaccess.h @@ -159,10 +159,9 @@ __asm__ __volatile__( \ "2: \n" \ " .section .fixup,\"ax\" \n" \ " .align 4 \n" \ - "4: \n" \ - " .long 2b \n" \ + " .literal_position \n" \ "5: \n" \ - " l32r %1, 4b \n" \ + " movi %1, 2b \n" \ " movi %0, %4 \n" \ " jx %1 \n" \ " .previous \n" \ @@ -217,10 +216,9 @@ __asm__ __volatile__( \ "2: \n" \ " .section .fixup,\"ax\" \n" \ " .align 4 \n" \ - "4: \n" \ - " .long 2b \n" \ + " .literal_position \n" \ "5: \n" \ - " l32r %1, 4b \n" \ + " movi %1, 2b \n" \ " movi %2, 0 \n" \ " movi %0, %4 \n" \ " jx %1 \n" \ |