mirror of
https://gitdl.cn/https://github.com/chakralinux/lib32.git
synced 2025-02-16 12:20:25 +08:00
lib32-llvm 5.0.1
This commit is contained in:
AlmAck
parent
d12f3b8689
commit
f0bb2513be
@ -1,140 +0,0 @@
|
||||
From 25e2616626caafb896517e18cd8aa724fba2b200 Mon Sep 17 00:00:00 2001
|
||||
From: Tom Stellard <thomas.stellard@amd.com>
|
||||
Date: Tue, 29 Nov 2016 03:41:28 +0000
|
||||
Subject: [PATCH] Merging r280589:
|
||||
|
||||
------------------------------------------------------------------------
|
||||
r280589 | nhaehnle | 2016-09-03 05:26:32 -0700 (Sat, 03 Sep 2016) | 19 lines
|
||||
|
||||
AMDGPU: Fix an interaction between WQM and polygon stippling
|
||||
|
||||
Summary:
|
||||
This fixes a rare bug in polygon stippling with non-monolithic pixel shaders.
|
||||
|
||||
The underlying problem is as follows: the prolog part contains the polygon
|
||||
stippling sequence, i.e. a kill. The main part then enables WQM based on the
|
||||
_reduced_ exec mask, effectively undoing most of the polygon stippling.
|
||||
|
||||
Since we cannot know whether polygon stippling will be used, the main part
|
||||
of a non-monolithic shader must always return to exact mode to fix this
|
||||
problem.
|
||||
|
||||
Reviewers: arsenm, tstellarAMD, mareko
|
||||
|
||||
Subscribers: arsenm, llvm-commits, kzhuravl
|
||||
|
||||
Differential Revision: https://reviews.llvm.org/D23131
|
||||
|
||||
------------------------------------------------------------------------
|
||||
|
||||
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_39@288105 91177308-0d34-0410-b5e6-96231b3b80d8
|
||||
---
|
||||
lib/Target/AMDGPU/SIInstructions.td | 1 +
|
||||
lib/Target/AMDGPU/SIWholeQuadMode.cpp | 7 -----
|
||||
test/CodeGen/AMDGPU/wqm.ll | 49 ++++++++++++++++++++++++++++++++---
|
||||
3 files changed, 46 insertions(+), 11 deletions(-)
|
||||
|
||||
diff --git a/lib/Target/AMDGPU/SIInstructions.td b/lib/Target/AMDGPU/SIInstructions.td
|
||||
index 18b7d5d..dde5f2f 100644
|
||||
--- a/lib/Target/AMDGPU/SIInstructions.td
|
||||
+++ b/lib/Target/AMDGPU/SIInstructions.td
|
||||
@@ -2029,6 +2029,7 @@ def SI_RETURN : PseudoInstSI <
|
||||
let hasSideEffects = 1;
|
||||
let SALU = 1;
|
||||
let hasNoSchedulingInfo = 1;
|
||||
+ let DisableWQM = 1;
|
||||
}
|
||||
|
||||
let Uses = [EXEC], Defs = [EXEC, VCC, M0],
|
||||
diff --git a/lib/Target/AMDGPU/SIWholeQuadMode.cpp b/lib/Target/AMDGPU/SIWholeQuadMode.cpp
|
||||
index b200c15..1534d58 100644
|
||||
--- a/lib/Target/AMDGPU/SIWholeQuadMode.cpp
|
||||
+++ b/lib/Target/AMDGPU/SIWholeQuadMode.cpp
|
||||
@@ -219,13 +219,6 @@ char SIWholeQuadMode::scanInstructions(MachineFunction &MF,
|
||||
markInstruction(MI, Flags, Worklist);
|
||||
GlobalFlags |= Flags;
|
||||
}
|
||||
-
|
||||
- if (WQMOutputs && MBB.succ_empty()) {
|
||||
- // This is a prolog shader. Make sure we go back to exact mode at the end.
|
||||
- Blocks[&MBB].OutNeeds = StateExact;
|
||||
- Worklist.push_back(&MBB);
|
||||
- GlobalFlags |= StateExact;
|
||||
- }
|
||||
}
|
||||
|
||||
return GlobalFlags;
|
||||
diff --git a/test/CodeGen/AMDGPU/wqm.ll b/test/CodeGen/AMDGPU/wqm.ll
|
||||
index 809a7ba..41e4264 100644
|
||||
--- a/test/CodeGen/AMDGPU/wqm.ll
|
||||
+++ b/test/CodeGen/AMDGPU/wqm.ll
|
||||
@@ -17,17 +17,18 @@ main_body:
|
||||
;CHECK-LABEL: {{^}}test2:
|
||||
;CHECK-NEXT: ; %main_body
|
||||
;CHECK-NEXT: s_wqm_b64 exec, exec
|
||||
-;CHECK: image_sample
|
||||
;CHECK-NOT: exec
|
||||
-;CHECK: _load_dword v0,
|
||||
-define amdgpu_ps float @test2(<8 x i32> inreg %rsrc, <4 x i32> inreg %sampler, float addrspace(1)* inreg %ptr, <4 x i32> %c) {
|
||||
+define amdgpu_ps void @test2(<8 x i32> inreg %rsrc, <4 x i32> inreg %sampler, float addrspace(1)* inreg %ptr, <4 x i32> %c) {
|
||||
main_body:
|
||||
%c.1 = call <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32> %c, <8 x i32> %rsrc, <4 x i32> %sampler, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
|
||||
%c.2 = bitcast <4 x float> %c.1 to <4 x i32>
|
||||
%c.3 = extractelement <4 x i32> %c.2, i32 0
|
||||
%gep = getelementptr float, float addrspace(1)* %ptr, i32 %c.3
|
||||
%data = load float, float addrspace(1)* %gep
|
||||
- ret float %data
|
||||
+
|
||||
+ call void @llvm.SI.export(i32 15, i32 1, i32 1, i32 0, i32 1, float %data, float undef, float undef, float undef)
|
||||
+
|
||||
+ ret void
|
||||
}
|
||||
|
||||
; ... but disabled for stores (and, in this simple case, not re-enabled).
|
||||
@@ -414,6 +415,46 @@ entry:
|
||||
ret void
|
||||
}
|
||||
|
||||
+; Must return to exact at the end of a non-void returning shader,
|
||||
+; otherwise the EXEC mask exported by the epilog will be wrong. This is true
|
||||
+; even if the shader has no kills, because a kill could have happened in a
|
||||
+; previous shader fragment.
|
||||
+;
|
||||
+; CHECK-LABEL: {{^}}test_nonvoid_return:
|
||||
+; CHECK: s_mov_b64 [[LIVE:s\[[0-9]+:[0-9]+\]]], exec
|
||||
+; CHECK: s_wqm_b64 exec, exec
|
||||
+;
|
||||
+; CHECK: s_and_b64 exec, exec, [[LIVE]]
|
||||
+; CHECK-NOT: exec
|
||||
+define amdgpu_ps <4 x float> @test_nonvoid_return() nounwind {
|
||||
+ %tex = call <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
|
||||
+ %tex.i = bitcast <4 x float> %tex to <4 x i32>
|
||||
+ %dtex = call <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32> %tex.i, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
|
||||
+ ret <4 x float> %dtex
|
||||
+}
|
||||
+
|
||||
+; CHECK-LABEL: {{^}}test_nonvoid_return_unreachable:
|
||||
+; CHECK: s_mov_b64 [[LIVE:s\[[0-9]+:[0-9]+\]]], exec
|
||||
+; CHECK: s_wqm_b64 exec, exec
|
||||
+;
|
||||
+; CHECK: s_and_b64 exec, exec, [[LIVE]]
|
||||
+; CHECK-NOT: exec
|
||||
+define amdgpu_ps <4 x float> @test_nonvoid_return_unreachable(i32 inreg %c) nounwind {
|
||||
+entry:
|
||||
+ %tex = call <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
|
||||
+ %tex.i = bitcast <4 x float> %tex to <4 x i32>
|
||||
+ %dtex = call <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32> %tex.i, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
|
||||
+
|
||||
+ %cc = icmp sgt i32 %c, 0
|
||||
+ br i1 %cc, label %if, label %else
|
||||
+
|
||||
+if:
|
||||
+ store volatile <4 x float> %dtex, <4 x float>* undef
|
||||
+ unreachable
|
||||
+
|
||||
+else:
|
||||
+ ret <4 x float> %dtex
|
||||
+}
|
||||
|
||||
declare void @llvm.amdgcn.image.store.v4i32(<4 x float>, <4 x i32>, <8 x i32>, i32, i1, i1, i1, i1) #1
|
||||
declare void @llvm.amdgcn.buffer.store.f32(float, <4 x i32>, i32, i32, i1, i1) #1
|
||||
@ -2,7 +2,7 @@
|
||||
# adjusted for Chakra
|
||||
|
||||
pkgname=('lib32-llvm' 'lib32-llvm-libs' 'lib32-clang')
|
||||
pkgver=4.0.1
|
||||
pkgver=5.0.1
|
||||
pkgrel=1
|
||||
arch=('x86_64')
|
||||
url="http://llvm.org/"
|
||||
@ -13,11 +13,11 @@ options=('staticlibs')
|
||||
source=(http://llvm.org/releases/$pkgver/llvm-$pkgver.src.tar.xz{,.sig}
|
||||
http://llvm.org/releases/$pkgver/cfe-$pkgver.src.tar.xz{,.sig}
|
||||
http://llvm.org/releases/$pkgver/compiler-rt-$pkgver.src.tar.xz{,.sig})
|
||||
sha256sums=('da783db1f82d516791179fe103c71706046561f7972b18f0049242dee6712b51'
|
||||
sha256sums=('5fa7489fc0225b11821cab0362f5813a05f2bcf2533e8a4ea9c9c860168807b0'
|
||||
'SKIP'
|
||||
'61738a735852c23c3bdbe52d035488cdb2083013f384d67c1ba36fabebd8769b'
|
||||
'135f6c9b0cd2da1aff2250e065946258eb699777888df39ca5a5b4fe5e23d0ff'
|
||||
'SKIP'
|
||||
'a3c87794334887b93b7a766c507244a7cdcce1d48b2e9249fc9a94f2c3beb440'
|
||||
'4edd1417f457a9b3f0eb88082530490edf3cf6a7335cdce8ecbc5d3e16a895da'
|
||||
'SKIP')
|
||||
validpgpkeys=('11E521D646982372EB577A1F8F0871F202119294'
|
||||
'B6C8F98282B944E3B0D5C2530FC3042E345AD05D')
|
||||
@ -88,7 +88,7 @@ package_lib32-llvm() {
|
||||
# Get rid of example Hello transformation
|
||||
rm "$pkgdir"/usr/lib32/*LLVMHello.*
|
||||
|
||||
rm -rf "$pkgdir"/usr/{bin,include,share/{doc,man,llvm}}
|
||||
rm -rf "$pkgdir"/usr/{bin,include,share/{doc,man,llvm,opt-viewer}}
|
||||
|
||||
# Needed for multilib (https://bugs.archlinux.org/task/29951)
|
||||
# Header stub is taken from Fedora
|
||||
|
||||
@ -1,103 +0,0 @@
|
||||
Index: lib/msan/msan_interceptors.cc
|
||||
===================================================================
|
||||
--- lib/msan/msan_interceptors.cc (revision 282231)
|
||||
+++ lib/msan/msan_interceptors.cc (revision 282232)
|
||||
@@ -64,6 +64,23 @@
|
||||
return in_interceptor_scope;
|
||||
}
|
||||
|
||||
+static uptr allocated_for_dlsym;
|
||||
+static const uptr kDlsymAllocPoolSize = 1024;
|
||||
+static uptr alloc_memory_for_dlsym[kDlsymAllocPoolSize];
|
||||
+
|
||||
+static bool IsInDlsymAllocPool(const void *ptr) {
|
||||
+ uptr off = (uptr)ptr - (uptr)alloc_memory_for_dlsym;
|
||||
+ return off < sizeof(alloc_memory_for_dlsym);
|
||||
+}
|
||||
+
|
||||
+static void *AllocateFromLocalPool(uptr size_in_bytes) {
|
||||
+ uptr size_in_words = RoundUpTo(size_in_bytes, kWordSize) / kWordSize;
|
||||
+ void *mem = (void *)&alloc_memory_for_dlsym[allocated_for_dlsym];
|
||||
+ allocated_for_dlsym += size_in_words;
|
||||
+ CHECK_LT(allocated_for_dlsym, kDlsymAllocPoolSize);
|
||||
+ return mem;
|
||||
+}
|
||||
+
|
||||
#define ENSURE_MSAN_INITED() do { \
|
||||
CHECK(!msan_init_is_running); \
|
||||
if (!msan_inited) { \
|
||||
@@ -227,7 +244,7 @@
|
||||
|
||||
INTERCEPTOR(void, free, void *ptr) {
|
||||
GET_MALLOC_STACK_TRACE;
|
||||
- if (!ptr) return;
|
||||
+ if (!ptr || UNLIKELY(IsInDlsymAllocPool(ptr))) return;
|
||||
MsanDeallocate(&stack, ptr);
|
||||
}
|
||||
|
||||
@@ -234,7 +251,7 @@
|
||||
#if !SANITIZER_FREEBSD
|
||||
INTERCEPTOR(void, cfree, void *ptr) {
|
||||
GET_MALLOC_STACK_TRACE;
|
||||
- if (!ptr) return;
|
||||
+ if (!ptr || UNLIKELY(IsInDlsymAllocPool(ptr))) return;
|
||||
MsanDeallocate(&stack, ptr);
|
||||
}
|
||||
#define MSAN_MAYBE_INTERCEPT_CFREE INTERCEPT_FUNCTION(cfree)
|
||||
@@ -907,27 +924,29 @@
|
||||
|
||||
INTERCEPTOR(void *, calloc, SIZE_T nmemb, SIZE_T size) {
|
||||
GET_MALLOC_STACK_TRACE;
|
||||
- if (UNLIKELY(!msan_inited)) {
|
||||
+ if (UNLIKELY(!msan_inited))
|
||||
// Hack: dlsym calls calloc before REAL(calloc) is retrieved from dlsym.
|
||||
- const SIZE_T kCallocPoolSize = 1024;
|
||||
- static uptr calloc_memory_for_dlsym[kCallocPoolSize];
|
||||
- static SIZE_T allocated;
|
||||
- SIZE_T size_in_words = ((nmemb * size) + kWordSize - 1) / kWordSize;
|
||||
- void *mem = (void*)&calloc_memory_for_dlsym[allocated];
|
||||
- allocated += size_in_words;
|
||||
- CHECK(allocated < kCallocPoolSize);
|
||||
- return mem;
|
||||
- }
|
||||
+ return AllocateFromLocalPool(nmemb * size);
|
||||
return MsanCalloc(&stack, nmemb, size);
|
||||
}
|
||||
|
||||
INTERCEPTOR(void *, realloc, void *ptr, SIZE_T size) {
|
||||
GET_MALLOC_STACK_TRACE;
|
||||
+ if (UNLIKELY(IsInDlsymAllocPool(ptr))) {
|
||||
+ uptr offset = (uptr)ptr - (uptr)alloc_memory_for_dlsym;
|
||||
+ uptr copy_size = Min(size, kDlsymAllocPoolSize - offset);
|
||||
+ void *new_ptr = AllocateFromLocalPool(size);
|
||||
+ internal_memcpy(new_ptr, ptr, copy_size);
|
||||
+ return new_ptr;
|
||||
+ }
|
||||
return MsanReallocate(&stack, ptr, size, sizeof(u64), false);
|
||||
}
|
||||
|
||||
INTERCEPTOR(void *, malloc, SIZE_T size) {
|
||||
GET_MALLOC_STACK_TRACE;
|
||||
+ if (UNLIKELY(!msan_inited))
|
||||
+ // Hack: dlsym calls malloc before REAL(malloc) is retrieved from dlsym.
|
||||
+ return AllocateFromLocalPool(size);
|
||||
return MsanReallocate(&stack, nullptr, size, sizeof(u64), false);
|
||||
}
|
||||
|
||||
Index: lib/asan/asan_malloc_linux.cc
|
||||
===================================================================
|
||||
--- lib/asan/asan_malloc_linux.cc (revision 282231)
|
||||
+++ lib/asan/asan_malloc_linux.cc (revision 282232)
|
||||
@@ -78,7 +78,11 @@
|
||||
if (UNLIKELY(IsInDlsymAllocPool(ptr))) {
|
||||
uptr offset = (uptr)ptr - (uptr)alloc_memory_for_dlsym;
|
||||
uptr copy_size = Min(size, kDlsymAllocPoolSize - offset);
|
||||
- void *new_ptr = asan_malloc(size, &stack);
|
||||
+ void *new_ptr;
|
||||
+ if (UNLIKELY(!asan_inited))
|
||||
+ new_ptr = AllocateFromLocalPool(size);
|
||||
+ else
|
||||
+ new_ptr = asan_malloc(size, &stack);
|
||||
internal_memcpy(new_ptr, ptr, copy_size);
|
||||
return new_ptr;
|
||||
}
|
||||
Loading…
Reference in New Issue
Block a user