On Wed, Oct 26, 2016 at 09:17:36AM +0300, Martin Storsjö wrote:
> On Wed, 26 Oct 2016, Diego Biurrun wrote:
> 
> >We currently use a mix of full path and -I assembler command line flags.
> >---
> >
> >This is a simplification related to the pending assembly language
> >build refactoring work.
> >
> >Makefile                            | 4 ++--
> >libavresample/x86/audio_convert.asm | 2 +-
> >libavresample/x86/audio_mix.asm     | 2 +-
> >libavutil/x86/cpuid.asm             | 2 +-
> >libavutil/x86/emms.asm              | 2 +-
> >libavutil/x86/float_dsp.asm         | 2 +-
> >libavutil/x86/lls.asm               | 2 +-
> >7 files changed, 8 insertions(+), 8 deletions(-)
> >
> >diff --git a/Makefile b/Makefile
> >index e860f2c..4237b31 100644
> >--- a/Makefile
> >+++ b/Makefile
> >@@ -57,8 +57,8 @@ COMPILE_HOSTC = $(call COMPILE,HOSTCC)
> >     $(COMPILE_HOSTC)
> >
> >%.o: %.asm
> >-    $(DEPYASM) $(YASMFLAGS) -I $(<D)/ -M -o $@ $< > $(@:.o=.d)
> >-    $(YASM) $(YASMFLAGS) -I $(<D)/ -o $@ $<
> >+    $(DEPYASM) $(YASMFLAGS) -M -o $@ $< > $(@:.o=.d)
> >+    $(YASM) $(YASMFLAGS) -o $@ $<
> >     -$(STRIP) $(STRIPFLAGS) $@
> >
> >%.i: %.c
> >diff --git a/libavresample/x86/audio_convert.asm 
> >b/libavresample/x86/audio_convert.asm
> >index 28f05a2..addfce0 100644
> >--- a/libavresample/x86/audio_convert.asm
> >+++ b/libavresample/x86/audio_convert.asm
> >@@ -21,7 +21,7 @@
> >;******************************************************************************
> >
> >%include "libavutil/x86/x86util.asm"
> >-%include "util.asm"
> >+%include "libavresample/x86/util.asm"
> 
> I'm not sure I would consider this an improvement; in general I would expect
> to be able to include files from the current directory without spelling out
> the full path. Although in practice this mostly seems to be used by the
> libavresample util.asm...

It's used by libavresample's util.asm and a few files in the libavutil/x86
subdirectory. I certainly don't feel strongly about this. Simpler rules
helped a bit with the refactoring, but it is not an insurmountable
obstacle, so ...

Diego
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to