From: Roland Scheidegger <sroland@vmware.com>
Date: Fri, 9 Aug 2013 15:29:52 +0000 (+0200)
Subject: util: (trivial) fix asm input/output list for fxsave
X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=836098f6b2dd275a2f557780d094cde13471f49d;p=mesa.git

util: (trivial) fix asm input/output list for fxsave

Otherwise gcc might do very unsafe optimizations, spotted by Uros Bizjak.
Hopefully this time it's finally right?
---

diff --git a/src/gallium/auxiliary/util/u_cpu_detect.c b/src/gallium/auxiliary/util/u_cpu_detect.c
index 3b36993dcc1..87ad780951e 100644
--- a/src/gallium/auxiliary/util/u_cpu_detect.c
+++ b/src/gallium/auxiliary/util/u_cpu_detect.c
@@ -244,7 +244,7 @@ static INLINE boolean sse2_has_daz(void)
 
    fxarea.mxcsr_mask = 0;
 #if (defined(PIPE_CC_GCC) || defined(PIPE_CC_SUNPRO))
-   __asm __volatile ("fxsave %0" :: "m" (fxarea));
+   __asm __volatile ("fxsave %0" : "+m" (fxarea));
 #elif (defined(PIPE_CC_MSVC) || defined(PIPE_CC_ICL))
    _fxsave(&fxarea);
 #endif