[flang] Add -f[no-]unroll-loops flag #122906
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This patch adds support for the -funroll-loops and -fno-unroll-loops flags with similar behaviour to clang. funroll-loops is enabled at -O2 onwards as in clang.
llvmbot
added
clang
|
@llvm/pr-subscribers-flang-driver @llvm/pr-subscribers-clang-driver Author: David Truby (DavidTruby) ChangesThis patch adds support for the -funroll-loops and -fno-unroll-loops flags with similar behaviour to clang. funroll-loops is enabled at -O2 onwards as is the current default. Full diff: https://github.com/llvm/llvm-project/pull/122906.diff 6 Files Affected:
diff --git a/clang/include/clang/Driver/Options.td b/clang/include/clang/Driver/Options.td
index 2721c1b5d8dc55..4bab2ae4d8dd5c 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -4157,9 +4157,9 @@ def ftrap_function_EQ : Joined<["-"], "ftrap-function=">, Group<f_Group>,
HelpText<"Issue call to specified function rather than a trap instruction">,
MarshallingInfoString<CodeGenOpts<"TrapFuncName">>;
def funroll_loops : Flag<["-"], "funroll-loops">, Group<f_Group>,
- HelpText<"Turn on loop unroller">, Visibility<[ClangOption, CC1Option]>;
+ HelpText<"Turn on loop unroller">, Visibility<[ClangOption, CC1Option, FlangOption, FC1Option]>;
def fno_unroll_loops : Flag<["-"], "fno-unroll-loops">, Group<f_Group>,
- HelpText<"Turn off loop unroller">, Visibility<[ClangOption, CC1Option]>;
+ HelpText<"Turn off loop unroller">, Visibility<[ClangOption, CC1Option, FlangOption, FC1Option]>;
def ffinite_loops: Flag<["-"], "ffinite-loops">, Group<f_Group>,
HelpText<"Assume all non-trivial loops are finite.">, Visibility<[ClangOption, CC1Option]>;
def fno_finite_loops: Flag<["-"], "fno-finite-loops">, Group<f_Group>,
diff --git a/clang/lib/Driver/ToolChains/Flang.cpp b/clang/lib/Driver/ToolChains/Flang.cpp
index a7d0cc99f27d2d..282a4e267b3dfc 100644
--- a/clang/lib/Driver/ToolChains/Flang.cpp
+++ b/clang/lib/Driver/ToolChains/Flang.cpp
@@ -150,12 +150,17 @@ void Flang::addCodegenOptions(const ArgList &Args,
if (shouldLoopVersion(Args))
CmdArgs.push_back("-fversion-loops-for-stride");
+ Args.addAllArgs(CmdArgs, {options::OPT_flang_experimental_hlfir,
+ options::OPT_flang_deprecated_no_hlfir,
+ options::OPT_fno_ppc_native_vec_elem_order,
+ options::OPT_fppc_native_vec_elem_order});
Args.addAllArgs(CmdArgs,
{options::OPT_flang_experimental_hlfir,
options::OPT_flang_deprecated_no_hlfir,
options::OPT_fno_ppc_native_vec_elem_order,
options::OPT_fppc_native_vec_elem_order,
- options::OPT_ftime_report, options::OPT_ftime_report_EQ});
+ options::OPT_ftime_report, options::OPT_ftime_report_EQ,
+ options::OPT_funroll_loops, options::OPT_fno_unroll_loops});
}
void Flang::addPicOptions(const ArgList &Args, ArgStringList &CmdArgs) const {
diff --git a/flang/include/flang/Frontend/CodeGenOptions.def b/flang/include/flang/Frontend/CodeGenOptions.def
index 9d03ec88a56b8a..deb8d1aede518b 100644
--- a/flang/include/flang/Frontend/CodeGenOptions.def
+++ b/flang/include/flang/Frontend/CodeGenOptions.def
@@ -32,6 +32,7 @@ CODEGENOPT(PrepareForThinLTO , 1, 0) ///< Set when -flto=thin is enabled on the
///< compile step.
CODEGENOPT(StackArrays, 1, 0) ///< -fstack-arrays (enable the stack-arrays pass)
CODEGENOPT(LoopVersioning, 1, 0) ///< Enable loop versioning.
+CODEGENOPT(UnrollLoops, 1, 0) ///< Enable loop unrolling
CODEGENOPT(AliasAnalysis, 1, 0) ///< Enable alias analysis pass
CODEGENOPT(Underscoring, 1, 1)
diff --git a/flang/lib/Frontend/CompilerInvocation.cpp b/flang/lib/Frontend/CompilerInvocation.cpp
index 5e7127313c1335..15b1e1e0a24881 100644
--- a/flang/lib/Frontend/CompilerInvocation.cpp
+++ b/flang/lib/Frontend/CompilerInvocation.cpp
@@ -246,6 +246,10 @@ static void parseCodeGenArgs(Fortran::frontend::CodeGenOptions &opts,
clang::driver::options::OPT_fno_loop_versioning, false))
opts.LoopVersioning = 1;
+ opts.UnrollLoops = args.hasFlag(clang::driver::options::OPT_funroll_loops,
+ clang::driver::options::OPT_fno_unroll_loops,
+ (opts.OptimizationLevel > 1));
+
opts.AliasAnalysis = opts.OptimizationLevel > 0;
// -mframe-pointer=none/non-leaf/all option.
diff --git a/flang/lib/Frontend/FrontendActions.cpp b/flang/lib/Frontend/FrontendActions.cpp
index 52a18d59c7cda5..b0545a7ac2f99a 100644
--- a/flang/lib/Frontend/FrontendActions.cpp
+++ b/flang/lib/Frontend/FrontendActions.cpp
@@ -1028,6 +1028,8 @@ void CodeGenAction::runOptimizationPipeline(llvm::raw_pwrite_stream &os) {
si.registerCallbacks(pic, &mam);
if (ci.isTimingEnabled())
si.getTimePasses().setOutStream(ci.getTimingStreamLLVM());
+ pto.LoopUnrolling = opts.UnrollLoops;
+ pto.LoopInterleaving = opts.UnrollLoops;
llvm::PassBuilder pb(targetMachine, pto, pgoOpt, &pic);
// Attempt to load pass plugins and register their callbacks with PB.
diff --git a/flang/test/HLFIR/unroll-loops.fir b/flang/test/HLFIR/unroll-loops.fir
new file mode 100644
index 00000000000000..f645132262f8d6
--- /dev/null
+++ b/flang/test/HLFIR/unroll-loops.fir
@@ -0,0 +1,43 @@
+// RUN: %flang_fc1 -emit-llvm -O1 -funroll-loops -mllvm -force-vector-width=1 -o- %s | FileCheck %s --check-prefixes=CHECK,UNROLL
+// RUN: %flang_fc1 -emit-llvm -O2 -mllvm -force-vector-width=1 -o- %s | FileCheck %s --check-prefixes=CHECK,UNROLL
+// RUN: %flang_fc1 -emit-llvm -O1 -fno-unroll-loops -mllvm -force-vector-width=1 -o- %s | FileCheck %s --check-prefixes=CHECK,NO-UNROLL
+// RUN: %flang_fc1 -emit-llvm -O1 -mllvm -force-vector-width=1 -o- %s | FileCheck %s --check-prefixes=CHECK,NO-UNROLL
+
+// CHECK-LABEL: @unroll
+// CHECK-SAME: (ptr nocapture writeonly %[[ARG0:.*]])
+func.func @unroll(%arg0: !fir.ref<!fir.array<1000xf64>> {fir.bindc_name = "a"}) {
+ // CHECK: %[[GEPIV:.*]] = getelementptr i8, ptr %0, i64 -8
+ %scope = fir.dummy_scope : !fir.dscope
+ %c1000 = arith.constant 1000 : index
+ %shape = fir.shape %c1000 : (index) -> !fir.shape<1>
+ %a:2 = hlfir.declare %arg0(%shape) dummy_scope %scope {uniq_name = "unrollEa"} : (!fir.ref<!fir.array<1000xf64>>, !fir.shape<1>, !fir.dscope) -> (!fir.ref<!fir.array<1000xf64>>, !fir.ref<!fir.array<1000xf64>>)
+ %c1 = arith.constant 1 : index
+ fir.do_loop %arg1 = %c1 to %c1000 step %c1 {
+ // CHECK: [[BLK:.*]]:
+
+ // NO-UNROLL-NEXT: %[[PHI:.*]] = phi i64 [ 1, %{{.*}} ], [ %[[NIV:.*]], %[[BLK]] ]
+ // NO-UNROLL-NEXT: %[[IV_D:.*]] = uitofp nneg i64 %[[PHI]] to double
+ // NO-UNROLL-NEXT: %[[GEP:.*]] = getelementptr double, ptr %[[GEPIV]], i64 %[[PHI]]
+ // NO-UNROLL-NEXT: store double %[[IV_D]], ptr %[[GEP]]
+ // NO-UNROLL-NEXT: %[[NIV:.*]] = add nuw nsw i64 %{{.*}}, 1
+ // NO-UNROLL-NEXT: %[[EXIT:.*]] = icmp eq i64 %[[NIV]], 1001
+ // NO-UNROLL-NEXT: br i1 %[[EXIT]], label %{{.*}}, label %[[BLK]]
+
+ // UNROLL-NEXT: %[[PHI:.*]] = phi i64 [ 0, %{{.*}} ], [ %[[NIV:.*]], %[[BLK]] ]
+ // UNROLL-NEXT: %[[IV0:.*]] = or disjoint i64 %[[PHI]], 1
+ // UNROLL-NEXT: %[[IV1:.*]] = add i64 %[[PHI]], 2
+ // UNROLL-NEXT: %[[IV0_D:.*]] = uitofp nneg i64 %[[IV0]] to double
+ // UNROLL-NEXT: %[[IV1_D:.*]] = uitofp nneg i64 %[[IV1]] to double
+ // UNROLL-NEXT: %[[GEP0:.*]] = getelementptr double, ptr %[[ARG0]], i64 %[[PHI]]
+ // UNROLL-NEXT: %[[GEP1:.*]] = getelementptr double, ptr %[[GEPIV]], i64 %[[IV1]]
+ // UNROLL-NEXT: store double %[[IV0_D]], ptr %[[GEP0]]
+ // UNROLL-NEXT: store double %[[IV1_D]], ptr %[[GEP1]]
+ // UNROLL-NEXT: %[[NIV:.*]] = add nuw i64 %[[PHI]], 2
+ // UNROLL-NEXT: %[[EXIT:.*]] = icmp eq i64 %[[NIV]], 1000
+ // UNROLL-NEXT: br i1 %[[EXIT]], label %{{.*}}, label %[[BLK]]
+ %iv = fir.convert %arg1 : (index) -> f64
+ %ai = hlfir.designate %a#0 (%arg1) : (!fir.ref<!fir.array<1000xf64>>, index) -> !fir.ref<f64>
+ hlfir.assign %iv to %ai : f64, !fir.ref<f64>
+ }
+ return
+}
|
|
@llvm/pr-subscribers-clang Author: David Truby (DavidTruby) ChangesThis patch adds support for the -funroll-loops and -fno-unroll-loops flags with similar behaviour to clang. funroll-loops is enabled at -O2 onwards as is the current default. Full diff: https://github.com/llvm/llvm-project/pull/122906.diff 6 Files Affected:
diff --git a/clang/include/clang/Driver/Options.td b/clang/include/clang/Driver/Options.td
index 2721c1b5d8dc55..4bab2ae4d8dd5c 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -4157,9 +4157,9 @@ def ftrap_function_EQ : Joined<["-"], "ftrap-function=">, Group<f_Group>,
HelpText<"Issue call to specified function rather than a trap instruction">,
MarshallingInfoString<CodeGenOpts<"TrapFuncName">>;
def funroll_loops : Flag<["-"], "funroll-loops">, Group<f_Group>,
- HelpText<"Turn on loop unroller">, Visibility<[ClangOption, CC1Option]>;
+ HelpText<"Turn on loop unroller">, Visibility<[ClangOption, CC1Option, FlangOption, FC1Option]>;
def fno_unroll_loops : Flag<["-"], "fno-unroll-loops">, Group<f_Group>,
- HelpText<"Turn off loop unroller">, Visibility<[ClangOption, CC1Option]>;
+ HelpText<"Turn off loop unroller">, Visibility<[ClangOption, CC1Option, FlangOption, FC1Option]>;
def ffinite_loops: Flag<["-"], "ffinite-loops">, Group<f_Group>,
HelpText<"Assume all non-trivial loops are finite.">, Visibility<[ClangOption, CC1Option]>;
def fno_finite_loops: Flag<["-"], "fno-finite-loops">, Group<f_Group>,
diff --git a/clang/lib/Driver/ToolChains/Flang.cpp b/clang/lib/Driver/ToolChains/Flang.cpp
index a7d0cc99f27d2d..282a4e267b3dfc 100644
--- a/clang/lib/Driver/ToolChains/Flang.cpp
+++ b/clang/lib/Driver/ToolChains/Flang.cpp
@@ -150,12 +150,17 @@ void Flang::addCodegenOptions(const ArgList &Args,
if (shouldLoopVersion(Args))
CmdArgs.push_back("-fversion-loops-for-stride");
+ Args.addAllArgs(CmdArgs, {options::OPT_flang_experimental_hlfir,
+ options::OPT_flang_deprecated_no_hlfir,
+ options::OPT_fno_ppc_native_vec_elem_order,
+ options::OPT_fppc_native_vec_elem_order});
Args.addAllArgs(CmdArgs,
{options::OPT_flang_experimental_hlfir,
options::OPT_flang_deprecated_no_hlfir,
options::OPT_fno_ppc_native_vec_elem_order,
options::OPT_fppc_native_vec_elem_order,
- options::OPT_ftime_report, options::OPT_ftime_report_EQ});
+ options::OPT_ftime_report, options::OPT_ftime_report_EQ,
+ options::OPT_funroll_loops, options::OPT_fno_unroll_loops});
}
void Flang::addPicOptions(const ArgList &Args, ArgStringList &CmdArgs) const {
diff --git a/flang/include/flang/Frontend/CodeGenOptions.def b/flang/include/flang/Frontend/CodeGenOptions.def
index 9d03ec88a56b8a..deb8d1aede518b 100644
--- a/flang/include/flang/Frontend/CodeGenOptions.def
+++ b/flang/include/flang/Frontend/CodeGenOptions.def
@@ -32,6 +32,7 @@ CODEGENOPT(PrepareForThinLTO , 1, 0) ///< Set when -flto=thin is enabled on the
///< compile step.
CODEGENOPT(StackArrays, 1, 0) ///< -fstack-arrays (enable the stack-arrays pass)
CODEGENOPT(LoopVersioning, 1, 0) ///< Enable loop versioning.
+CODEGENOPT(UnrollLoops, 1, 0) ///< Enable loop unrolling
CODEGENOPT(AliasAnalysis, 1, 0) ///< Enable alias analysis pass
CODEGENOPT(Underscoring, 1, 1)
diff --git a/flang/lib/Frontend/CompilerInvocation.cpp b/flang/lib/Frontend/CompilerInvocation.cpp
index 5e7127313c1335..15b1e1e0a24881 100644
--- a/flang/lib/Frontend/CompilerInvocation.cpp
+++ b/flang/lib/Frontend/CompilerInvocation.cpp
@@ -246,6 +246,10 @@ static void parseCodeGenArgs(Fortran::frontend::CodeGenOptions &opts,
clang::driver::options::OPT_fno_loop_versioning, false))
opts.LoopVersioning = 1;
+ opts.UnrollLoops = args.hasFlag(clang::driver::options::OPT_funroll_loops,
+ clang::driver::options::OPT_fno_unroll_loops,
+ (opts.OptimizationLevel > 1));
+
opts.AliasAnalysis = opts.OptimizationLevel > 0;
// -mframe-pointer=none/non-leaf/all option.
diff --git a/flang/lib/Frontend/FrontendActions.cpp b/flang/lib/Frontend/FrontendActions.cpp
index 52a18d59c7cda5..b0545a7ac2f99a 100644
--- a/flang/lib/Frontend/FrontendActions.cpp
+++ b/flang/lib/Frontend/FrontendActions.cpp
@@ -1028,6 +1028,8 @@ void CodeGenAction::runOptimizationPipeline(llvm::raw_pwrite_stream &os) {
si.registerCallbacks(pic, &mam);
if (ci.isTimingEnabled())
si.getTimePasses().setOutStream(ci.getTimingStreamLLVM());
+ pto.LoopUnrolling = opts.UnrollLoops;
+ pto.LoopInterleaving = opts.UnrollLoops;
llvm::PassBuilder pb(targetMachine, pto, pgoOpt, &pic);
// Attempt to load pass plugins and register their callbacks with PB.
diff --git a/flang/test/HLFIR/unroll-loops.fir b/flang/test/HLFIR/unroll-loops.fir
new file mode 100644
index 00000000000000..f645132262f8d6
--- /dev/null
+++ b/flang/test/HLFIR/unroll-loops.fir
@@ -0,0 +1,43 @@
+// RUN: %flang_fc1 -emit-llvm -O1 -funroll-loops -mllvm -force-vector-width=1 -o- %s | FileCheck %s --check-prefixes=CHECK,UNROLL
+// RUN: %flang_fc1 -emit-llvm -O2 -mllvm -force-vector-width=1 -o- %s | FileCheck %s --check-prefixes=CHECK,UNROLL
+// RUN: %flang_fc1 -emit-llvm -O1 -fno-unroll-loops -mllvm -force-vector-width=1 -o- %s | FileCheck %s --check-prefixes=CHECK,NO-UNROLL
+// RUN: %flang_fc1 -emit-llvm -O1 -mllvm -force-vector-width=1 -o- %s | FileCheck %s --check-prefixes=CHECK,NO-UNROLL
+
+// CHECK-LABEL: @unroll
+// CHECK-SAME: (ptr nocapture writeonly %[[ARG0:.*]])
+func.func @unroll(%arg0: !fir.ref<!fir.array<1000xf64>> {fir.bindc_name = "a"}) {
+ // CHECK: %[[GEPIV:.*]] = getelementptr i8, ptr %0, i64 -8
+ %scope = fir.dummy_scope : !fir.dscope
+ %c1000 = arith.constant 1000 : index
+ %shape = fir.shape %c1000 : (index) -> !fir.shape<1>
+ %a:2 = hlfir.declare %arg0(%shape) dummy_scope %scope {uniq_name = "unrollEa"} : (!fir.ref<!fir.array<1000xf64>>, !fir.shape<1>, !fir.dscope) -> (!fir.ref<!fir.array<1000xf64>>, !fir.ref<!fir.array<1000xf64>>)
+ %c1 = arith.constant 1 : index
+ fir.do_loop %arg1 = %c1 to %c1000 step %c1 {
+ // CHECK: [[BLK:.*]]:
+
+ // NO-UNROLL-NEXT: %[[PHI:.*]] = phi i64 [ 1, %{{.*}} ], [ %[[NIV:.*]], %[[BLK]] ]
+ // NO-UNROLL-NEXT: %[[IV_D:.*]] = uitofp nneg i64 %[[PHI]] to double
+ // NO-UNROLL-NEXT: %[[GEP:.*]] = getelementptr double, ptr %[[GEPIV]], i64 %[[PHI]]
+ // NO-UNROLL-NEXT: store double %[[IV_D]], ptr %[[GEP]]
+ // NO-UNROLL-NEXT: %[[NIV:.*]] = add nuw nsw i64 %{{.*}}, 1
+ // NO-UNROLL-NEXT: %[[EXIT:.*]] = icmp eq i64 %[[NIV]], 1001
+ // NO-UNROLL-NEXT: br i1 %[[EXIT]], label %{{.*}}, label %[[BLK]]
+
+ // UNROLL-NEXT: %[[PHI:.*]] = phi i64 [ 0, %{{.*}} ], [ %[[NIV:.*]], %[[BLK]] ]
+ // UNROLL-NEXT: %[[IV0:.*]] = or disjoint i64 %[[PHI]], 1
+ // UNROLL-NEXT: %[[IV1:.*]] = add i64 %[[PHI]], 2
+ // UNROLL-NEXT: %[[IV0_D:.*]] = uitofp nneg i64 %[[IV0]] to double
+ // UNROLL-NEXT: %[[IV1_D:.*]] = uitofp nneg i64 %[[IV1]] to double
+ // UNROLL-NEXT: %[[GEP0:.*]] = getelementptr double, ptr %[[ARG0]], i64 %[[PHI]]
+ // UNROLL-NEXT: %[[GEP1:.*]] = getelementptr double, ptr %[[GEPIV]], i64 %[[IV1]]
+ // UNROLL-NEXT: store double %[[IV0_D]], ptr %[[GEP0]]
+ // UNROLL-NEXT: store double %[[IV1_D]], ptr %[[GEP1]]
+ // UNROLL-NEXT: %[[NIV:.*]] = add nuw i64 %[[PHI]], 2
+ // UNROLL-NEXT: %[[EXIT:.*]] = icmp eq i64 %[[NIV]], 1000
+ // UNROLL-NEXT: br i1 %[[EXIT]], label %{{.*}}, label %[[BLK]]
+ %iv = fir.convert %arg1 : (index) -> f64
+ %ai = hlfir.designate %a#0 (%arg1) : (!fir.ref<!fir.array<1000xf64>>, index) -> !fir.ref<f64>
+ hlfir.assign %iv to %ai : f64, !fir.ref<f64>
+ }
+ return
+}
|
DavidTruby
requested review from
Meinersbur,
kiranchandramohan and
tarunprabhu
flang/test/HLFIR/unroll-loops.fir
Outdated
| @@ -0,0 +1,43 @@ | |||
| // RUN: %flang_fc1 -emit-llvm -O1 -funroll-loops -mllvm -force-vector-width=1 -o- %s | FileCheck %s --check-prefixes=CHECK,UNROLL | |||
There was a problem hiding this comment.
The test should probably be in Integration directory and possibly a source to LLVM IR test.
There was a problem hiding this comment.
A source to (at least) HLFIR test will also check that the -f(no-)unroll loops option has not disappeared from the frontend, and is being passed correctly to fc1.
There was a problem hiding this comment.
I guess we should possibly have all 3? A source to HLFIR, this test, and source->llvmir?
There was a problem hiding this comment.
Actually, source->HLFIR isn't useful when the driver test already exists. I'll add a source->llvm integration test.
Comment on lines
153
to
156
| Args.addAllArgs(CmdArgs, {options::OPT_flang_experimental_hlfir, | ||
| options::OPT_flang_deprecated_no_hlfir, | ||
| options::OPT_fno_ppc_native_vec_elem_order, | ||
| options::OPT_fppc_native_vec_elem_order}); |
There was a problem hiding this comment.
Why is this added separately from below?
| Args.addAllArgs(CmdArgs, | ||
| {options::OPT_flang_experimental_hlfir, | ||
| options::OPT_flang_deprecated_no_hlfir, | ||
| options::OPT_fno_ppc_native_vec_elem_order, | ||
| options::OPT_fppc_native_vec_elem_order, | ||
| options::OPT_ftime_report, options::OPT_ftime_report_EQ}); | ||
| options::OPT_ftime_report, options::OPT_ftime_report_EQ, | ||
| options::OPT_funroll_loops, options::OPT_fno_unroll_loops}); |
There was a problem hiding this comment.
We can add a forwarding test from the driver to the frontend driver.
Comment on lines
16
to
20
| ! NO-UNROLL-NEXT: %[[IV_D:.*]] = uitofp nneg <2 x i64> %[[VIND]] to <2 x double> | ||
| ! NO-UNROLL-NEXT: %[[GEP:.*]] = getelementptr double, ptr %[[ARG0]], i64 %[[IND]] | ||
| ! NO-UNROLL-NEXT: store <2 x double> %[[IV_D]], ptr %[[GEP]] | ||
| ! NO-UNROLL-NEXT: %[[NIV:.*]] = add nuw i64 %{{.*}}, 2 | ||
| ! NO-UNROLL-NEXT: %[[NVIND]] = add <2 x i64> %[[VIND]], splat (i64 2) |
There was a problem hiding this comment.
Unroll should ideally check for the branch back to the body. And nounroll should probably check that there is no such branch?
There was a problem hiding this comment.
-funroll-loops doesn't mean that the loop will be fully unrolled, just that some unrolling can occur. So there's a branch back to the body of the loop in both cases.
There was a problem hiding this comment.
For small loops, it will fully unroll if the body is small.
The issue with checking for two iterations is that this could be due to interleaving during vectorisation as well.
There was a problem hiding this comment.
I actually did try to only have (e.g.) 2 iterations of the loop; for some reason the full loop is still emitted, at least at -O1. What I ended up with here was essentially copied from the clang -funroll-loops tests but converted to fortran.
| ! CHECK-LABEL: @unroll | ||
| ! CHECK-SAME: (ptr nocapture writeonly %[[ARG0:.*]]) | ||
| subroutine unroll(a) | ||
| real(kind=8), intent(out) :: a(1000) |
There was a problem hiding this comment.
Can this be an integer array to avoid the uitofp ?
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/157/builds/17336 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/201/builds/1461 Here is the relevant piece of the build log for the reference |
|
Hi @DavidTruby, have are you investigating this failure on |
kamaub
pushed a commit
that referenced
this pull request
Jan 20, 2025
mustartt
added a commit
to mustartt/llvm-project
that referenced
this pull request
Jan 21, 2025
This reverts commit 0195ec4.
|
I guess the tests in this should have had --target= for the targets I checked them on (x86_64 and aarch64). I can add those and remove the xfail? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This patch adds support for the -funroll-loops and -fno-unroll-loops flags with similar behaviour to clang. funroll-loops is enabled at -O2 onwards as is the current default.