summaryrefslogtreecommitdiffstats
path: root/llvm/test/CodeGen/AMDGPU/local-memory.ll
diff options
context:
space:
mode:
authorMatt Arsenault <Matthew.Arsenault@amd.com>2015-11-30 21:16:03 +0000
committerMatt Arsenault <Matthew.Arsenault@amd.com>2015-11-30 21:16:03 +0000
commit26f8f3db39f8fd856cfdce0689c5e5b84536eb9a (patch)
treea7241314f78b6bf81416571bf7b298cd29b6a884 /llvm/test/CodeGen/AMDGPU/local-memory.ll
parentac234b604da563cbe53364116cf84abe4abbdb04 (diff)
downloadbcm5719-llvm-26f8f3db39f8fd856cfdce0689c5e5b84536eb9a.tar.gz
bcm5719-llvm-26f8f3db39f8fd856cfdce0689c5e5b84536eb9a.zip
AMDGPU: Rework how private buffer passed for HSA
If we know we have stack objects, we reserve the registers that the private buffer resource and wave offset are passed and use them directly. If not, reserve the last 5 SGPRs just in case we need to spill. After register allocation, try to pick the next available registers instead of the last SGPRs, and then insert copies from the inputs to the reserved registers in the progloue. This also only selectively enables all of the input registers which are really required instead of always enabling them. llvm-svn: 254331
Diffstat (limited to 'llvm/test/CodeGen/AMDGPU/local-memory.ll')
-rw-r--r--llvm/test/CodeGen/AMDGPU/local-memory.ll4
1 files changed, 2 insertions, 2 deletions
diff --git a/llvm/test/CodeGen/AMDGPU/local-memory.ll b/llvm/test/CodeGen/AMDGPU/local-memory.ll
index 9494ed75bd0..9ffb59e7092 100644
--- a/llvm/test/CodeGen/AMDGPU/local-memory.ll
+++ b/llvm/test/CodeGen/AMDGPU/local-memory.ll
@@ -9,9 +9,9 @@
; EG: .long 166120
; EG-NEXT: .long 128
; SI: .long 47180
-; SI-NEXT: .long 71560
+; SI-NEXT: .long 65668
; CI: .long 47180
-; CI-NEXT: .long 38792
+; CI-NEXT: .long 32900
; FUNC-LABEL: {{^}}local_memory:
OpenPOWER on IntegriCloud