[llvm] AMDGPU: Don't pad .text with s_code_end if it would otherwise be empty (PR #147980)
Tim Renouf via llvm-commits
llvm-commits at lists.llvm.org
Fri Jul 11 06:20:39 PDT 2025
https://github.com/trenouf updated https://github.com/llvm/llvm-project/pull/147980
>From 9f4e8e170bb2978fd5bbb6dbac0900c51ea09af7 Mon Sep 17 00:00:00 2001
From: Tim Renouf <tim.renouf at amd.com>
Date: Thu, 20 Feb 2025 09:47:55 +0000
Subject: [PATCH 1/3] AMDGPU: Don't pad .text with s_code_end if it would
otherwise be empty
We don't want that padding in a module that only contains data, not
code.
---
llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp b/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp
index 749b9efc81378..3765c1d2b106c 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp
@@ -486,10 +486,12 @@ bool AMDGPUAsmPrinter::doFinalization(Module &M) {
// Pad with s_code_end to help tools and guard against instruction prefetch
// causing stale data in caches. Arguably this should be done by the linker,
// which is why this isn't done for Mesa.
+ // Don't do it if there are no functions.
const MCSubtargetInfo &STI = *getGlobalSTI();
if ((AMDGPU::isGFX10Plus(STI) || AMDGPU::isGFX90A(STI)) &&
(STI.getTargetTriple().getOS() == Triple::AMDHSA ||
- STI.getTargetTriple().getOS() == Triple::AMDPAL)) {
+ STI.getTargetTriple().getOS() == Triple::AMDPAL) &&
+ !M.empty()) {
OutStreamer->switchSection(getObjFileLowering().getTextSection());
getTargetStreamer()->EmitCodeEnd(STI);
}
>From dab2b13ecf7809b4dc05fae9b267afdd25d222f3 Mon Sep 17 00:00:00 2001
From: Tim Renouf <tim.renouf at amd.com>
Date: Fri, 11 Jul 2025 10:49:46 +0100
Subject: [PATCH 2/3] Use hasInstructions(), and fix hasInstructions() to work
on the asm streamer
---
llvm/lib/MC/MCAsmStreamer.cpp | 2 ++
llvm/lib/MC/MCObjectStreamer.cpp | 6 ++----
llvm/lib/MC/MCStreamer.cpp | 3 +++
llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp | 12 +++++++-----
4 files changed, 14 insertions(+), 9 deletions(-)
diff --git a/llvm/lib/MC/MCAsmStreamer.cpp b/llvm/lib/MC/MCAsmStreamer.cpp
index aae02652564d3..0231a71b4bd23 100644
--- a/llvm/lib/MC/MCAsmStreamer.cpp
+++ b/llvm/lib/MC/MCAsmStreamer.cpp
@@ -2427,6 +2427,8 @@ void MCAsmStreamer::AddEncodingComment(const MCInst &Inst,
void MCAsmStreamer::emitInstruction(const MCInst &Inst,
const MCSubtargetInfo &STI) {
+ MCStreamer::emitInstruction(Inst, STI);
+
if (MAI->isAIX() && CurFrag)
// Now that a machine instruction has been assembled into this section, make
// a line entry for any .loc directive that has been seen.
diff --git a/llvm/lib/MC/MCObjectStreamer.cpp b/llvm/lib/MC/MCObjectStreamer.cpp
index 5cc9bed2669d4..5d8822b452f59 100644
--- a/llvm/lib/MC/MCObjectStreamer.cpp
+++ b/llvm/lib/MC/MCObjectStreamer.cpp
@@ -342,12 +342,10 @@ void MCObjectStreamer::emitInstructionImpl(const MCInst &Inst,
const MCSubtargetInfo &STI) {
MCStreamer::emitInstruction(Inst, STI);
- MCSection *Sec = getCurrentSectionOnly();
- Sec->setHasInstructions(true);
-
// Now that a machine instruction has been assembled into this section, make
// a line entry for any .loc directive that has been seen.
- MCDwarfLineEntry::make(this, getCurrentSectionOnly());
+ MCSection *Sec = getCurrentSectionOnly();
+ MCDwarfLineEntry::make(this, Sec);
// If this instruction doesn't need relaxation, just emit it as data.
MCAssembler &Assembler = getAssembler();
diff --git a/llvm/lib/MC/MCStreamer.cpp b/llvm/lib/MC/MCStreamer.cpp
index 6cd6b4abdd327..8603b60d91915 100644
--- a/llvm/lib/MC/MCStreamer.cpp
+++ b/llvm/lib/MC/MCStreamer.cpp
@@ -1194,6 +1194,9 @@ void MCStreamer::visitUsedExpr(const MCExpr &Expr) {
}
void MCStreamer::emitInstruction(const MCInst &Inst, const MCSubtargetInfo &) {
+ MCSection *Sec = getCurrentSectionOnly();
+ Sec->setHasInstructions(true);
+
// Scan for values.
for (unsigned i = Inst.getNumOperands(); i--;)
if (Inst.getOperand(i).isExpr())
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp b/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp
index 3765c1d2b106c..65dade76150f2 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp
@@ -486,14 +486,16 @@ bool AMDGPUAsmPrinter::doFinalization(Module &M) {
// Pad with s_code_end to help tools and guard against instruction prefetch
// causing stale data in caches. Arguably this should be done by the linker,
// which is why this isn't done for Mesa.
- // Don't do it if there are no functions.
+ // Don't do it if there is no code.
const MCSubtargetInfo &STI = *getGlobalSTI();
if ((AMDGPU::isGFX10Plus(STI) || AMDGPU::isGFX90A(STI)) &&
(STI.getTargetTriple().getOS() == Triple::AMDHSA ||
- STI.getTargetTriple().getOS() == Triple::AMDPAL) &&
- !M.empty()) {
- OutStreamer->switchSection(getObjFileLowering().getTextSection());
- getTargetStreamer()->EmitCodeEnd(STI);
+ STI.getTargetTriple().getOS() == Triple::AMDPAL)) {
+ MCSection *TextSect = getObjFileLowering().getTextSection();
+ if (TextSect->hasInstructions()) {
+ OutStreamer->switchSection(TextSect);
+ getTargetStreamer()->EmitCodeEnd(STI);
+ }
}
// Assign expressions which can only be resolved when all other functions are
>From 7ab441218bfb4913c3409740c0168c19f4db17bb Mon Sep 17 00:00:00 2001
From: Tim Renouf <tim.renouf at amd.com>
Date: Fri, 11 Jul 2025 09:20:53 +0100
Subject: [PATCH 3/3] Add test
---
llvm/test/CodeGen/AMDGPU/empty-text.ll | 9 +++++++++
1 file changed, 9 insertions(+)
create mode 100644 llvm/test/CodeGen/AMDGPU/empty-text.ll
diff --git a/llvm/test/CodeGen/AMDGPU/empty-text.ll b/llvm/test/CodeGen/AMDGPU/empty-text.ll
new file mode 100644
index 0000000000000..c6a035cd03b3e
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/empty-text.ll
@@ -0,0 +1,9 @@
+; Test that there is no s_code_end padding if .text is othertwise empty.
+
+; RUN: llc -mtriple=amdgcn--amdpal -mcpu=gfx1200 < %s | FileCheck %s --check-prefixes=GCN
+
+ at globalVar = global i32 37
+
+declare amdgpu_ps void @funcDecl()
+
+; GCN-NOT: .fill
More information about the llvm-commits
mailing list