Hi, regarding the problem I mentionted about the new amd64 optimized functions in imlib2, I think I found the problem, has something to do with the fact that memory was not aligned in some (SSE2 128 bit) MOV operations - ie, I just changed a couple of MOVDQA to MOVDQU in file amd64_blend.S, treating memory as unaligned; Now if this has some other side effects (speed?) I don't know, but for me it worked now...
Cheers, Tiago Gehring
diff -u -r e17/libs/imlib2/src/lib/amd64_blend.S ../e17_ori/e17/libs/imlib2/src/lib/amd64_blend.S --- e17/libs/imlib2/src/lib/amd64_blend.S 2005-08-07 07:07:23.000000000 +0000 +++ ../e17_ori/e17/libs/imlib2/src/lib/amd64_blend.S 2005-08-22 07:10:00.000000000 +0000 @@ -168,8 +168,8 @@ ENTER pxor %xmm4, %xmm4 - movdqa c1(%rip), %xmm5 - movdqa m00XXXXXX(%rip), %xmm6 + movdqu c1(%rip), %xmm5 + movdqu m00XXXXXX(%rip), %xmm6 /* Move right to left across each line, */ /* processing in two pixel chunks */ @@ -565,9 +565,9 @@ ENTER pxor %xmm4, %xmm4 - movdqa c1(%rip), %xmm5 + movdqu c1(%rip), %xmm5 xorq %rax, %rax - movdqa mX000X000X000X000(%rip), %xmm6 + movdqu mX000X000X000X000(%rip), %xmm6 movq [EMAIL PROTECTED](%rip), %r13 /* Move right to left across each line, */ @@ -994,8 +994,8 @@ PR_(imlib_amd64_copy_rgba_to_rgb): ENTER - movdqa m0XXX0XXX0XXX0XXX(%rip), %xmm5 - movdqa mX000X000X000X000(%rip), %xmm6 + movdqu m0XXX0XXX0XXX0XXX(%rip), %xmm5 + movdqu mX000X000X000X000(%rip), %xmm6 leaq (%rsi, %r8, 4), %rsi leaq (%rdi, %r8, 4), %rdi
