summaryrefslogtreecommitdiffstats
path: root/src/amd
diff options
context:
space:
mode:
authorDave Airlie <[email protected]>2017-07-06 23:09:07 +0100
committerDave Airlie <[email protected]>2017-07-09 22:21:43 +0100
commit7b5f2e00706899e26aaa78e595e21522ccba9c5c (patch)
treecc85762c24daabe523f90dc69188967f1041fc84 /src/amd
parentedf2acbeb16112fe002d3f7208018273d413f25c (diff)
radv/ac: drop setting xnack
Since radv uses compute rings and we can't know when we are setting up the shaders what ring they are to be used on, we should just use the default xnack setting. This may be suboptimal in some places, but if we hit a problem, we likely should try and address this between llvm and mesa. Reviewed-by: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
Diffstat (limited to 'src/amd')
-rw-r--r--src/amd/common/ac_llvm_util.c3
1 files changed, 1 insertions, 2 deletions
diff --git a/src/amd/common/ac_llvm_util.c b/src/amd/common/ac_llvm_util.c
index d14057fa768..ea752e031ef 100644
--- a/src/amd/common/ac_llvm_util.c
+++ b/src/amd/common/ac_llvm_util.c
@@ -126,8 +126,7 @@ LLVMTargetMachineRef ac_create_target_machine(enum radeon_family family, enum ac
LLVMTargetRef target = ac_get_llvm_target(triple);
snprintf(features, sizeof(features),
- "+DumpCode,+vgpr-spilling,-fp32-denormals%s%s",
- family >= CHIP_VEGA10 ? ",+xnack" : ",-xnack",
+ "+DumpCode,+vgpr-spilling,-fp32-denormals%s",
tm_options & AC_TM_SISCHED ? ",+si-scheduler" : "");
LLVMTargetMachineRef tm = LLVMCreateTargetMachine(