<div dir="ltr">We need a test case for this patch at the least.<div><br></div><div>I also wonder if we shouldn't support both names to avoid breaking any code that has been written to Clang and never tested with GCC? Supporting both names in this particular context seems less onerous than usual.</div>
</div><div class="gmail_extra"><br><br><div class="gmail_quote">On Thu, May 22, 2014 at 9:22 PM, Matt Turner <span dir="ltr"><<a href="mailto:mattst88@gmail.com" target="_blank">mattst88@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">These have been incorrect since they were added to clang, and wouldn't<br>
have been useful to programs without additional workarounds.<br>
---<br>
I added CPU feature detection via cpuid.h to Mesa, and discovered<br>
that clang's macro names for SSE4.* were broken.<br>
<br>
Please Cc me on replies, and please commit, I don't have access.<br>
<br>
lib/Headers/cpuid.h | 4 ++--<br>
1 file changed, 2 insertions(+), 2 deletions(-)<br>
<br>
diff --git a/lib/Headers/cpuid.h b/lib/Headers/cpuid.h<br>
index f9254e9..ae78e19 100644<br>
--- a/lib/Headers/cpuid.h<br>
+++ b/lib/Headers/cpuid.h<br>
@@ -43,8 +43,8 @@<br>
#define bit_PDCM 0x00008000<br>
#define bit_PCID 0x00020000<br>
#define bit_DCA 0x00040000<br>
-#define bit_SSE41 0x00080000<br>
-#define bit_SSE42 0x00100000<br>
+#define bit_SSE4_1 0x00080000<br>
+#define bit_SSE4_2 0x00100000<br>
#define bit_x2APIC 0x00200000<br>
#define bit_MOVBE 0x00400000<br>
#define bit_POPCNT 0x00800000<br>
<span class="HOEnZb"><font color="#888888">--<br>
1.8.3.2<br>
<br>
_______________________________________________<br>
cfe-dev mailing list<br>
<a href="mailto:cfe-dev@cs.uiuc.edu">cfe-dev@cs.uiuc.edu</a><br>
<a href="http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev" target="_blank">http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev</a><br>
</font></span></blockquote></div><br></div>