[FFmpeg-devel] [PATCH] Fix compilation with NASM.
Reimar Döffinger
Reimar.Doeffinger at gmx.de
Sun Apr 15 19:31:53 CEST 2012
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger at gmx.de>
---
libavutil/x86/x86inc.asm | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/libavutil/x86/x86inc.asm b/libavutil/x86/x86inc.asm
index 7e389d0..1303902 100644
--- a/libavutil/x86/x86inc.asm
+++ b/libavutil/x86/x86inc.asm
@@ -97,7 +97,10 @@
%endif
; Always use long nops (reduces 0x90 spam in disassembly on x86_32)
+; Not supported by NASM (except via smartalign package)
+%ifdef __YASM_VER__
CPU amdnop
+%endif
; Macros to eliminate most code duplication between x86_32 and x86_64:
; Currently this works only for leaf functions which load all their arguments
@@ -142,10 +145,10 @@ CPU amdnop
%define r%1mp %2
%elif ARCH_X86_64 ; memory
%define r%1m [rsp + stack_offset + %6]
- %define r%1mp qword r %+ %1m
+ %define r%1mp qword r %+ %1 %+ m
%else
%define r%1m [esp + stack_offset + %6]
- %define r%1mp dword r %+ %1m
+ %define r%1mp dword r %+ %1 %+ m
%endif
%define r%1 %2
%endmacro
--
1.7.10
More information about the ffmpeg-devel
mailing list