mirror of
https://github.com/RPCS3/llvm.git
synced 2024-12-28 15:33:16 +00:00
b118a073d7
- Rewrite/merge pseudo-atomic instruction emitters to address the
following issue:
* Reduce one unnecessary load in spin-loop
previously the spin-loop looks like
thisMBB:
newMBB:
ld t1 = [bitinstr.addr]
op t2 = t1, [bitinstr.val]
not t3 = t2 (if Invert)
mov EAX = t1
lcs dest = [bitinstr.addr], t3 [EAX is implicit]
bz newMBB
fallthrough -->nextMBB
the 'ld' at the beginning of newMBB should be lift out of the loop
as lcs (or CMPXCHG on x86) will load the current memory value into
EAX. This loop is refined as:
thisMBB:
EAX = LOAD [MI.addr]
mainMBB:
t1 = OP [MI.val], EAX
LCMPXCHG [MI.addr], t1, [EAX is implicitly used & defined]
JNE mainMBB
sinkMBB:
* Remove immopc as, so far, all pseudo-atomic instructions has
all-register form only, there is no immedidate operand.
* Remove unnecessary attributes/modifiers in pseudo-atomic instruction
td
* Fix issues in PR13458
- Add comprehensive tests on atomic ops on various data types.
NOTE: Some of them are turned off due to missing functionality.
- Revise tests due to the new spin-loop generated.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@164281 91177308-0d34-0410-b5e6-96231b3b80d8
15 lines
767 B
LLVM
15 lines
767 B
LLVM
; RUN: llc < %s
|
|
|
|
target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
|
|
target triple = "x86_64-apple-darwin11.4.2"
|
|
|
|
%v8_uniform_Stats.0.2.4.10 = type { i64, i64, i32, i32, i32, i32, i32, i32, i32, i32, i64, i64, i64, i32, i32, i32, i32, i32, i32, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i64, i32, i64, [7 x i32], [7 x i64] }
|
|
|
|
@globalStats = external global %v8_uniform_Stats.0.2.4.10
|
|
|
|
define void @MergeStats() nounwind {
|
|
allocas:
|
|
%r.i.i720 = atomicrmw max i64* getelementptr inbounds (%v8_uniform_Stats.0.2.4.10* @globalStats, i64 0, i32 30), i64 0 seq_cst
|
|
ret void
|
|
}
|