maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/PowerPC/atomics.ll

138 lines
3.9 KiB
LLVM
Raw Normal View History

[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
; RUN: llc < %s -mtriple=powerpc-apple-darwin -march=ppc32 -verify-machineinstrs | FileCheck %s --check-prefix=CHECK --check-prefix=PPC32
; FIXME: -verify-machineinstrs currently fail on ppc64 (mismatched register/instruction).
; This is already checked for in Atomics-64.ll
; RUN: llc < %s -mtriple=powerpc-apple-darwin -march=ppc64 | FileCheck %s --check-prefix=CHECK --check-prefix=PPC64
; FIXME: we don't currently check for the operations themselves with CHECK-NEXT,
; because they are implemented in a very messy way with lwarx/stwcx.
; It should be fixed soon in another patch.
; We first check loads, for all sizes from i8 to i64.
; We also vary orderings to check for barriers.
define i8 @load_i8_unordered(i8* %mem) {
; CHECK-LABEL: load_i8_unordered
[Power] Improve the expansion of atomic loads/stores Summary: Atomic loads and store of up to the native size (32 bits, or 64 for PPC64) can be lowered to a simple load or store instruction (as the synchronization is already handled by AtomicExpand, and the atomicity is guaranteed thanks to the alignment requirements of atomic accesses). This is exactly what this patch does. Previously, these were implemented by complex load-linked/store-conditional loops.. an obvious performance problem. For example, this patch turns ``` define void @store_i8_unordered(i8* %mem) { store atomic i8 42, i8* %mem unordered, align 1 ret void } ``` from ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: rlwinm r2, r3, 3, 27, 28 li r4, 42 xori r5, r2, 24 rlwinm r2, r3, 0, 0, 29 li r3, 255 slw r4, r4, r5 slw r3, r3, r5 and r4, r4, r3 LBB4_1: ; =>This Inner Loop Header: Depth=1 lwarx r5, 0, r2 andc r5, r5, r3 or r5, r4, r5 stwcx. r5, 0, r2 bne cr0, LBB4_1 ; BB#2: blr ``` into ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: li r2, 42 stb r2, 0(r3) blr ``` which looks like a pretty clear win to me. Test Plan: fixed the tests + new test for indexed accesses + make check-all Reviewers: jfb, wschmidt, hfinkel Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5587 llvm-svn: 218922
2014-10-03 06:27:07 +08:00
; CHECK: lbz
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
; CHECK-NOT: sync
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%val = load atomic i8, i8* %mem unordered, align 1
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
ret i8 %val
}
define i16 @load_i16_monotonic(i16* %mem) {
; CHECK-LABEL: load_i16_monotonic
[Power] Improve the expansion of atomic loads/stores Summary: Atomic loads and store of up to the native size (32 bits, or 64 for PPC64) can be lowered to a simple load or store instruction (as the synchronization is already handled by AtomicExpand, and the atomicity is guaranteed thanks to the alignment requirements of atomic accesses). This is exactly what this patch does. Previously, these were implemented by complex load-linked/store-conditional loops.. an obvious performance problem. For example, this patch turns ``` define void @store_i8_unordered(i8* %mem) { store atomic i8 42, i8* %mem unordered, align 1 ret void } ``` from ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: rlwinm r2, r3, 3, 27, 28 li r4, 42 xori r5, r2, 24 rlwinm r2, r3, 0, 0, 29 li r3, 255 slw r4, r4, r5 slw r3, r3, r5 and r4, r4, r3 LBB4_1: ; =>This Inner Loop Header: Depth=1 lwarx r5, 0, r2 andc r5, r5, r3 or r5, r4, r5 stwcx. r5, 0, r2 bne cr0, LBB4_1 ; BB#2: blr ``` into ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: li r2, 42 stb r2, 0(r3) blr ``` which looks like a pretty clear win to me. Test Plan: fixed the tests + new test for indexed accesses + make check-all Reviewers: jfb, wschmidt, hfinkel Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5587 llvm-svn: 218922
2014-10-03 06:27:07 +08:00
; CHECK: lhz
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
; CHECK-NOT: sync
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%val = load atomic i16, i16* %mem monotonic, align 2
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
ret i16 %val
}
define i32 @load_i32_acquire(i32* %mem) {
; CHECK-LABEL: load_i32_acquire
[Power] Improve the expansion of atomic loads/stores Summary: Atomic loads and store of up to the native size (32 bits, or 64 for PPC64) can be lowered to a simple load or store instruction (as the synchronization is already handled by AtomicExpand, and the atomicity is guaranteed thanks to the alignment requirements of atomic accesses). This is exactly what this patch does. Previously, these were implemented by complex load-linked/store-conditional loops.. an obvious performance problem. For example, this patch turns ``` define void @store_i8_unordered(i8* %mem) { store atomic i8 42, i8* %mem unordered, align 1 ret void } ``` from ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: rlwinm r2, r3, 3, 27, 28 li r4, 42 xori r5, r2, 24 rlwinm r2, r3, 0, 0, 29 li r3, 255 slw r4, r4, r5 slw r3, r3, r5 and r4, r4, r3 LBB4_1: ; =>This Inner Loop Header: Depth=1 lwarx r5, 0, r2 andc r5, r5, r3 or r5, r4, r5 stwcx. r5, 0, r2 bne cr0, LBB4_1 ; BB#2: blr ``` into ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: li r2, 42 stb r2, 0(r3) blr ``` which looks like a pretty clear win to me. Test Plan: fixed the tests + new test for indexed accesses + make check-all Reviewers: jfb, wschmidt, hfinkel Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5587 llvm-svn: 218922
2014-10-03 06:27:07 +08:00
; CHECK: lwz
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%val = load atomic i32, i32* %mem acquire, align 4
[PowerPC] Enable printing instructions using aliases TableGen had been nicely generating code to print a number of instructions using shorter aliases (and PowerPC has plenty of short mnemonics), but we were not calling it. For some of the aliases we support in the parser, TableGen can't infer the "inverse" alias relationship, so there is still more to do. Thus, after some hours of updating test cases... llvm-svn: 235616
2015-04-24 02:30:38 +08:00
; CHECK: lwsync
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
ret i32 %val
}
define i64 @load_i64_seq_cst(i64* %mem) {
; CHECK-LABEL: load_i64_seq_cst
[PowerPC] Use sync inst alias when printing So long as the choice between printing msync and sync is not ambiguous, we can print 'sync 0' and just 'sync'. llvm-svn: 235663
2015-04-24 07:05:08 +08:00
; CHECK: sync
[Power] Improve the expansion of atomic loads/stores Summary: Atomic loads and store of up to the native size (32 bits, or 64 for PPC64) can be lowered to a simple load or store instruction (as the synchronization is already handled by AtomicExpand, and the atomicity is guaranteed thanks to the alignment requirements of atomic accesses). This is exactly what this patch does. Previously, these were implemented by complex load-linked/store-conditional loops.. an obvious performance problem. For example, this patch turns ``` define void @store_i8_unordered(i8* %mem) { store atomic i8 42, i8* %mem unordered, align 1 ret void } ``` from ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: rlwinm r2, r3, 3, 27, 28 li r4, 42 xori r5, r2, 24 rlwinm r2, r3, 0, 0, 29 li r3, 255 slw r4, r4, r5 slw r3, r3, r5 and r4, r4, r3 LBB4_1: ; =>This Inner Loop Header: Depth=1 lwarx r5, 0, r2 andc r5, r5, r3 or r5, r4, r5 stwcx. r5, 0, r2 bne cr0, LBB4_1 ; BB#2: blr ``` into ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: li r2, 42 stb r2, 0(r3) blr ``` which looks like a pretty clear win to me. Test Plan: fixed the tests + new test for indexed accesses + make check-all Reviewers: jfb, wschmidt, hfinkel Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5587 llvm-svn: 218922
2014-10-03 06:27:07 +08:00
; PPC32: __sync_
; PPC64-NOT: __sync_
; PPC64: ld
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%val = load atomic i64, i64* %mem seq_cst, align 8
[PowerPC] Enable printing instructions using aliases TableGen had been nicely generating code to print a number of instructions using shorter aliases (and PowerPC has plenty of short mnemonics), but we were not calling it. For some of the aliases we support in the parser, TableGen can't infer the "inverse" alias relationship, so there is still more to do. Thus, after some hours of updating test cases... llvm-svn: 235616
2015-04-24 02:30:38 +08:00
; CHECK: lwsync
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
ret i64 %val
}
; Stores
define void @store_i8_unordered(i8* %mem) {
; CHECK-LABEL: store_i8_unordered
; CHECK-NOT: sync
[Power] Improve the expansion of atomic loads/stores Summary: Atomic loads and store of up to the native size (32 bits, or 64 for PPC64) can be lowered to a simple load or store instruction (as the synchronization is already handled by AtomicExpand, and the atomicity is guaranteed thanks to the alignment requirements of atomic accesses). This is exactly what this patch does. Previously, these were implemented by complex load-linked/store-conditional loops.. an obvious performance problem. For example, this patch turns ``` define void @store_i8_unordered(i8* %mem) { store atomic i8 42, i8* %mem unordered, align 1 ret void } ``` from ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: rlwinm r2, r3, 3, 27, 28 li r4, 42 xori r5, r2, 24 rlwinm r2, r3, 0, 0, 29 li r3, 255 slw r4, r4, r5 slw r3, r3, r5 and r4, r4, r3 LBB4_1: ; =>This Inner Loop Header: Depth=1 lwarx r5, 0, r2 andc r5, r5, r3 or r5, r4, r5 stwcx. r5, 0, r2 bne cr0, LBB4_1 ; BB#2: blr ``` into ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: li r2, 42 stb r2, 0(r3) blr ``` which looks like a pretty clear win to me. Test Plan: fixed the tests + new test for indexed accesses + make check-all Reviewers: jfb, wschmidt, hfinkel Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5587 llvm-svn: 218922
2014-10-03 06:27:07 +08:00
; CHECK: stb
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
store atomic i8 42, i8* %mem unordered, align 1
ret void
}
define void @store_i16_monotonic(i16* %mem) {
; CHECK-LABEL: store_i16_monotonic
; CHECK-NOT: sync
[Power] Improve the expansion of atomic loads/stores Summary: Atomic loads and store of up to the native size (32 bits, or 64 for PPC64) can be lowered to a simple load or store instruction (as the synchronization is already handled by AtomicExpand, and the atomicity is guaranteed thanks to the alignment requirements of atomic accesses). This is exactly what this patch does. Previously, these were implemented by complex load-linked/store-conditional loops.. an obvious performance problem. For example, this patch turns ``` define void @store_i8_unordered(i8* %mem) { store atomic i8 42, i8* %mem unordered, align 1 ret void } ``` from ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: rlwinm r2, r3, 3, 27, 28 li r4, 42 xori r5, r2, 24 rlwinm r2, r3, 0, 0, 29 li r3, 255 slw r4, r4, r5 slw r3, r3, r5 and r4, r4, r3 LBB4_1: ; =>This Inner Loop Header: Depth=1 lwarx r5, 0, r2 andc r5, r5, r3 or r5, r4, r5 stwcx. r5, 0, r2 bne cr0, LBB4_1 ; BB#2: blr ``` into ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: li r2, 42 stb r2, 0(r3) blr ``` which looks like a pretty clear win to me. Test Plan: fixed the tests + new test for indexed accesses + make check-all Reviewers: jfb, wschmidt, hfinkel Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5587 llvm-svn: 218922
2014-10-03 06:27:07 +08:00
; CHECK: sth
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
store atomic i16 42, i16* %mem monotonic, align 2
ret void
}
define void @store_i32_release(i32* %mem) {
; CHECK-LABEL: store_i32_release
[PowerPC] Enable printing instructions using aliases TableGen had been nicely generating code to print a number of instructions using shorter aliases (and PowerPC has plenty of short mnemonics), but we were not calling it. For some of the aliases we support in the parser, TableGen can't infer the "inverse" alias relationship, so there is still more to do. Thus, after some hours of updating test cases... llvm-svn: 235616
2015-04-24 02:30:38 +08:00
; CHECK: lwsync
[Power] Improve the expansion of atomic loads/stores Summary: Atomic loads and store of up to the native size (32 bits, or 64 for PPC64) can be lowered to a simple load or store instruction (as the synchronization is already handled by AtomicExpand, and the atomicity is guaranteed thanks to the alignment requirements of atomic accesses). This is exactly what this patch does. Previously, these were implemented by complex load-linked/store-conditional loops.. an obvious performance problem. For example, this patch turns ``` define void @store_i8_unordered(i8* %mem) { store atomic i8 42, i8* %mem unordered, align 1 ret void } ``` from ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: rlwinm r2, r3, 3, 27, 28 li r4, 42 xori r5, r2, 24 rlwinm r2, r3, 0, 0, 29 li r3, 255 slw r4, r4, r5 slw r3, r3, r5 and r4, r4, r3 LBB4_1: ; =>This Inner Loop Header: Depth=1 lwarx r5, 0, r2 andc r5, r5, r3 or r5, r4, r5 stwcx. r5, 0, r2 bne cr0, LBB4_1 ; BB#2: blr ``` into ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: li r2, 42 stb r2, 0(r3) blr ``` which looks like a pretty clear win to me. Test Plan: fixed the tests + new test for indexed accesses + make check-all Reviewers: jfb, wschmidt, hfinkel Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5587 llvm-svn: 218922
2014-10-03 06:27:07 +08:00
; CHECK: stw
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
store atomic i32 42, i32* %mem release, align 4
ret void
}
define void @store_i64_seq_cst(i64* %mem) {
; CHECK-LABEL: store_i64_seq_cst
[PowerPC] Use sync inst alias when printing So long as the choice between printing msync and sync is not ambiguous, we can print 'sync 0' and just 'sync'. llvm-svn: 235663
2015-04-24 07:05:08 +08:00
; CHECK: sync
[Power] Improve the expansion of atomic loads/stores Summary: Atomic loads and store of up to the native size (32 bits, or 64 for PPC64) can be lowered to a simple load or store instruction (as the synchronization is already handled by AtomicExpand, and the atomicity is guaranteed thanks to the alignment requirements of atomic accesses). This is exactly what this patch does. Previously, these were implemented by complex load-linked/store-conditional loops.. an obvious performance problem. For example, this patch turns ``` define void @store_i8_unordered(i8* %mem) { store atomic i8 42, i8* %mem unordered, align 1 ret void } ``` from ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: rlwinm r2, r3, 3, 27, 28 li r4, 42 xori r5, r2, 24 rlwinm r2, r3, 0, 0, 29 li r3, 255 slw r4, r4, r5 slw r3, r3, r5 and r4, r4, r3 LBB4_1: ; =>This Inner Loop Header: Depth=1 lwarx r5, 0, r2 andc r5, r5, r3 or r5, r4, r5 stwcx. r5, 0, r2 bne cr0, LBB4_1 ; BB#2: blr ``` into ``` _store_i8_unordered: ; @store_i8_unordered ; BB#0: li r2, 42 stb r2, 0(r3) blr ``` which looks like a pretty clear win to me. Test Plan: fixed the tests + new test for indexed accesses + make check-all Reviewers: jfb, wschmidt, hfinkel Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5587 llvm-svn: 218922
2014-10-03 06:27:07 +08:00
; PPC32: __sync_
; PPC64-NOT: __sync_
; PPC64: std
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
store atomic i64 42, i64* %mem seq_cst, align 8
ret void
}
; Atomic CmpXchg
define i8 @cas_strong_i8_sc_sc(i8* %mem) {
; CHECK-LABEL: cas_strong_i8_sc_sc
[PowerPC] Use sync inst alias when printing So long as the choice between printing msync and sync is not ambiguous, we can print 'sync 0' and just 'sync'. llvm-svn: 235663
2015-04-24 07:05:08 +08:00
; CHECK: sync
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
%val = cmpxchg i8* %mem, i8 0, i8 1 seq_cst seq_cst
[PowerPC] Enable printing instructions using aliases TableGen had been nicely generating code to print a number of instructions using shorter aliases (and PowerPC has plenty of short mnemonics), but we were not calling it. For some of the aliases we support in the parser, TableGen can't infer the "inverse" alias relationship, so there is still more to do. Thus, after some hours of updating test cases... llvm-svn: 235616
2015-04-24 02:30:38 +08:00
; CHECK: lwsync
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
%loaded = extractvalue { i8, i1} %val, 0
ret i8 %loaded
}
define i16 @cas_weak_i16_acquire_acquire(i16* %mem) {
; CHECK-LABEL: cas_weak_i16_acquire_acquire
;CHECK-NOT: sync
%val = cmpxchg weak i16* %mem, i16 0, i16 1 acquire acquire
[PowerPC] Enable printing instructions using aliases TableGen had been nicely generating code to print a number of instructions using shorter aliases (and PowerPC has plenty of short mnemonics), but we were not calling it. For some of the aliases we support in the parser, TableGen can't infer the "inverse" alias relationship, so there is still more to do. Thus, after some hours of updating test cases... llvm-svn: 235616
2015-04-24 02:30:38 +08:00
; CHECK: lwsync
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
%loaded = extractvalue { i16, i1} %val, 0
ret i16 %loaded
}
define i32 @cas_strong_i32_acqrel_acquire(i32* %mem) {
; CHECK-LABEL: cas_strong_i32_acqrel_acquire
[PowerPC] Enable printing instructions using aliases TableGen had been nicely generating code to print a number of instructions using shorter aliases (and PowerPC has plenty of short mnemonics), but we were not calling it. For some of the aliases we support in the parser, TableGen can't infer the "inverse" alias relationship, so there is still more to do. Thus, after some hours of updating test cases... llvm-svn: 235616
2015-04-24 02:30:38 +08:00
; CHECK: lwsync
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
%val = cmpxchg i32* %mem, i32 0, i32 1 acq_rel acquire
[PowerPC] Enable printing instructions using aliases TableGen had been nicely generating code to print a number of instructions using shorter aliases (and PowerPC has plenty of short mnemonics), but we were not calling it. For some of the aliases we support in the parser, TableGen can't infer the "inverse" alias relationship, so there is still more to do. Thus, after some hours of updating test cases... llvm-svn: 235616
2015-04-24 02:30:38 +08:00
; CHECK: lwsync
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
%loaded = extractvalue { i32, i1} %val, 0
ret i32 %loaded
}
define i64 @cas_weak_i64_release_monotonic(i64* %mem) {
; CHECK-LABEL: cas_weak_i64_release_monotonic
[PowerPC] Enable printing instructions using aliases TableGen had been nicely generating code to print a number of instructions using shorter aliases (and PowerPC has plenty of short mnemonics), but we were not calling it. For some of the aliases we support in the parser, TableGen can't infer the "inverse" alias relationship, so there is still more to do. Thus, after some hours of updating test cases... llvm-svn: 235616
2015-04-24 02:30:38 +08:00
; CHECK: lwsync
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
%val = cmpxchg weak i64* %mem, i64 0, i64 1 release monotonic
; CHECK-NOT: [sync ]
%loaded = extractvalue { i64, i1} %val, 0
ret i64 %loaded
}
; AtomicRMW
define i8 @add_i8_monotonic(i8* %mem, i8 %operand) {
; CHECK-LABEL: add_i8_monotonic
; CHECK-NOT: sync
%val = atomicrmw add i8* %mem, i8 %operand monotonic
ret i8 %val
}
define i16 @xor_i16_seq_cst(i16* %mem, i16 %operand) {
; CHECK-LABEL: xor_i16_seq_cst
[PowerPC] Use sync inst alias when printing So long as the choice between printing msync and sync is not ambiguous, we can print 'sync 0' and just 'sync'. llvm-svn: 235663
2015-04-24 07:05:08 +08:00
; CHECK: sync
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
%val = atomicrmw xor i16* %mem, i16 %operand seq_cst
[PowerPC] Enable printing instructions using aliases TableGen had been nicely generating code to print a number of instructions using shorter aliases (and PowerPC has plenty of short mnemonics), but we were not calling it. For some of the aliases we support in the parser, TableGen can't infer the "inverse" alias relationship, so there is still more to do. Thus, after some hours of updating test cases... llvm-svn: 235616
2015-04-24 02:30:38 +08:00
; CHECK: lwsync
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
ret i16 %val
}
define i32 @xchg_i32_acq_rel(i32* %mem, i32 %operand) {
; CHECK-LABEL: xchg_i32_acq_rel
[PowerPC] Enable printing instructions using aliases TableGen had been nicely generating code to print a number of instructions using shorter aliases (and PowerPC has plenty of short mnemonics), but we were not calling it. For some of the aliases we support in the parser, TableGen can't infer the "inverse" alias relationship, so there is still more to do. Thus, after some hours of updating test cases... llvm-svn: 235616
2015-04-24 02:30:38 +08:00
; CHECK: lwsync
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
%val = atomicrmw xchg i32* %mem, i32 %operand acq_rel
[PowerPC] Enable printing instructions using aliases TableGen had been nicely generating code to print a number of instructions using shorter aliases (and PowerPC has plenty of short mnemonics), but we were not calling it. For some of the aliases we support in the parser, TableGen can't infer the "inverse" alias relationship, so there is still more to do. Thus, after some hours of updating test cases... llvm-svn: 235616
2015-04-24 02:30:38 +08:00
; CHECK: lwsync
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
ret i32 %val
}
define i64 @and_i64_release(i64* %mem, i64 %operand) {
; CHECK-LABEL: and_i64_release
[PowerPC] Enable printing instructions using aliases TableGen had been nicely generating code to print a number of instructions using shorter aliases (and PowerPC has plenty of short mnemonics), but we were not calling it. For some of the aliases we support in the parser, TableGen can't infer the "inverse" alias relationship, so there is still more to do. Thus, after some hours of updating test cases... llvm-svn: 235616
2015-04-24 02:30:38 +08:00
; CHECK: lwsync
[Power] Use AtomicExpandPass for fence insertion, and use lwsync where appropriate Summary: This patch makes use of AtomicExpandPass in Power for inserting fences around atomic as part of an effort to remove fence insertion from SelectionDAGBuilder. As a big bonus, it lets us use sync 1 (lightweight sync, often used by the mnemonic lwsync) instead of sync 0 (heavyweight sync) in many cases. I also added a test, as there was no test for the barriers emitted by the Power backend for atomic loads and stores. Test Plan: new test + make check-all Reviewers: jfb Subscribers: llvm-commits Differential Revision: http://reviews.llvm.org/D5180 llvm-svn: 218331
2014-09-24 04:46:49 +08:00
%val = atomicrmw and i64* %mem, i64 %operand release
; CHECK-NOT: [sync ]
ret i64 %val
}