Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Fri, 11 Nov 2011 19:52:12 +0100
From: magnum <>
Subject: Re: more targets using sse-intrinsics.S

Actually I had the exact same thought! It can't hurt, please include that.


2011-11-11 19:43, jfoug wrote:
> It appears (at least in the cygwin build), that gcc is called, and in the background, calls the asm, vs having makefile properly call it, and passing along the ASFLAGS.  Thus for cygwin/mingw, which REQUIRE the underscore define, to properly mangle the asm build, having the make interpreter properly put in the ASFLAGS is crucial.
> There may be other environs where this is also crucial.
> Thus, I think we should do this:
> 64 bit  make intrisics:  sse-intrinsics.c -> sse-intrinsics-64.S
> 32 bit  make intrisics:  sse-intrinsics.c -> sse-intrinsics-32.S
> Then, we can properly use sse-intrinsics-64.o or sse-intrinsics-32.o in the OBJS line of each target, and make will properly know how to build that object, and will be able to call the assembler itself, passing all of the proper defines.
> Jim.
>> From: magnum []
>> I see... For the x86-64 builds, we *must* use .S because otherwise the
>> compiler will (or may?) use sse-intrinsics.c to produce
>> sse-intrinsics.o!  But since there is no sse-intrinsics-32.c we can
>> indeed replace all sse-intrinsics-32.S in the targets' object lists to
>> sse-intrinsics-32.o.
>> I just tested this with the linux sse2i target and it works fine. I
>> suppose it's easier to "follow" if you change them too.
>> magnum

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.