[LoongArch] Add processor models la64v1.0 and la64v1.1 #110211
Closed
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
Add support for the `la64v1.0` and `la64v1.1` processor models, enabling Rust to select features based on the version of the LoongArch instruction set. These values will align with those used in the C/C++ toolchain. [^1] [^1]: https://github.com/loongson/la-toolchain-conventions?tab=readme-ov-file#list
|
@llvm/pr-subscribers-backend-loongarch Author: hev (heiher) ChangesAdd support for the Full diff: https://github.com/llvm/llvm-project/pull/110211.diff 2 Files Affected:
diff --git a/llvm/lib/Target/LoongArch/LoongArch.td b/llvm/lib/Target/LoongArch/LoongArch.td
index ddb27dc6404fa8..c1bd94862b4e5e 100644
--- a/llvm/lib/Target/LoongArch/LoongArch.td
+++ b/llvm/lib/Target/LoongArch/LoongArch.td
@@ -128,30 +128,41 @@ include "LoongArchInstrInfo.td"
// LoongArch processors supported.
//===----------------------------------------------------------------------===//
-def : ProcessorModel<"generic-la32", NoSchedModel, [Feature32Bit]>;
-def : ProcessorModel<"generic-la64", NoSchedModel, [Feature64Bit, FeatureUAL]>;
+def ProcessorFeatures {
+ list<SubtargetFeature> LA32G = [Feature32Bit];
+ list<SubtargetFeature> LA64G = [Feature64Bit,
+ FeatureUAL];
+ list<SubtargetFeature> LA64GD = [Feature64Bit,
+ FeatureUAL,
+ FeatureBasicD];
+ list<SubtargetFeature> LA64V1_0 = [Feature64Bit,
+ FeatureUAL,
+ FeatureExtLASX,
+ FeatureExtLVZ,
+ FeatureExtLBT];
+ list<SubtargetFeature> LA64V1_1 = [Feature64Bit,
+ FeatureUAL,
+ FeatureExtLASX,
+ FeatureExtLVZ,
+ FeatureExtLBT,
+ FeatureFrecipe];
+}
+
+def : ProcessorModel<"generic-la32", NoSchedModel, ProcessorFeatures.LA32G>;
+def : ProcessorModel<"generic-la64", NoSchedModel, ProcessorFeatures.LA64G>;
// Generic 64-bit processor with double-precision floating-point support.
-def : ProcessorModel<"loongarch64", NoSchedModel, [Feature64Bit,
- FeatureUAL,
- FeatureBasicD]>;
+def : ProcessorModel<"loongarch64", NoSchedModel, ProcessorFeatures.LA64GD>;
// Support generic for compatibility with other targets. The triple will be used
// to change to the appropriate la32/la64 version.
def : ProcessorModel<"generic", NoSchedModel, []>;
-def : ProcessorModel<"la464", NoSchedModel, [Feature64Bit,
- FeatureUAL,
- FeatureExtLASX,
- FeatureExtLVZ,
- FeatureExtLBT]>;
-
-def : ProcessorModel<"la664", NoSchedModel, [Feature64Bit,
- FeatureUAL,
- FeatureExtLASX,
- FeatureExtLVZ,
- FeatureExtLBT,
- FeatureFrecipe]>;
+def : ProcessorModel<"la464", NoSchedModel, ProcessorFeatures.LA64V1_0>;
+def : ProcessorModel<"la64v1.0", NoSchedModel, ProcessorFeatures.LA64V1_0>;
+
+def : ProcessorModel<"la664", NoSchedModel, ProcessorFeatures.LA64V1_1>;
+def : ProcessorModel<"la64v1.1", NoSchedModel, ProcessorFeatures.LA64V1_1>;
//===----------------------------------------------------------------------===//
// Define the LoongArch target.
diff --git a/llvm/test/CodeGen/LoongArch/cpus.ll b/llvm/test/CodeGen/LoongArch/cpus.ll
index be3e05e394ff9a..aa5a81411a2d71 100644
--- a/llvm/test/CodeGen/LoongArch/cpus.ll
+++ b/llvm/test/CodeGen/LoongArch/cpus.ll
@@ -4,6 +4,8 @@
; RUN: llc < %s --mtriple=loongarch64 -mattr=+d --mcpu=loongarch64 2>&1 | FileCheck %s
; RUN: llc < %s --mtriple=loongarch64 -mattr=+d --mcpu=la464 2>&1 | FileCheck %s
; RUN: llc < %s --mtriple=loongarch64 -mattr=+d --mcpu=la664 2>&1 | FileCheck %s
+; RUN: llc < %s --mtriple=loongarch64 -mattr=+d --mcpu=la64v1.0 2>&1 | FileCheck %s
+; RUN: llc < %s --mtriple=loongarch64 -mattr=+d --mcpu=la64v1.1 2>&1 | FileCheck %s
; RUN: llc < %s --mtriple=loongarch64 -mattr=+d 2>&1 | FileCheck %s
; CHECK-NOT: {{.*}} is not a recognized processor for this target
@@ -23,3 +25,11 @@ define void @tune_cpu_la464() "tune-cpu"="la464" {
define void @tune_cpu_la664() "tune-cpu"="la664" {
ret void
}
+
+define void @tune_cpu_la64v1_0() "tune-cpu"="la64v1.0" {
+ ret void
+}
+
+define void @tune_cpu_la64v1_1() "tune-cpu"="la64v1.1" {
+ ret void
+}
Footnotes |
xen0n
reviewed
Oct 2, 2024
There was a problem hiding this comment.
While the changes here are good, I think it may be a good idea to also make the generic target to support FP and LSX (but not LASX), as per the recommended general computing baseline by Loongson. This is tangential to the issue at hand though, and we can do it in a separate PR if you agree. (I'm surprised to realize that generic does not support FP but the equally generic-sounding loongarch64 does -- this is very confusing and I always struggle to remember these details, even to this day.)
|
CLOSE: Internal reviewers believe that this change will make the The changes to |
Ami-zhang
added a commit
to Ami-zhang/llvm-project
that referenced
this pull request
Oct 23, 2024
This commit makes the `generic` target to support FP and LSX, as discussed in llvm#110211. Thereby, it allows 128-bit vector to be enabled by default in the loongarch64 backend.
|
BTW, what are the plans for adding |
|
Ami-zhang
added a commit
to Ami-zhang/llvm-project
that referenced
this pull request
Oct 24, 2024
This commit makes the `generic` target to support FP and LSX, as discussed in llvm#110211. Thereby, it allows 128-bit vector to be enabled by default in the loongarch64 backend.
Ami-zhang
added a commit
to Ami-zhang/llvm-project
that referenced
this pull request
Oct 25, 2024
This commit makes the `generic` target to support FP and LSX, as discussed in llvm#110211. Thereby, it allows 128-bit vector to be enabled by default in the loongarch64 backend.
Ami-zhang
added a commit
that referenced
this pull request
Oct 31, 2024
This commit makes the `generic` target to support FP and LSX, as discussed in #110211. Thereby, it allows 128-bit vector to be enabled by default in the loongarch64 backend.
smallp-o-p
pushed a commit
to smallp-o-p/llvm-project
that referenced
this pull request
Nov 3, 2024
) This commit makes the `generic` target to support FP and LSX, as discussed in llvm#110211. Thereby, it allows 128-bit vector to be enabled by default in the loongarch64 backend.
NoumanAmir657
pushed a commit
to NoumanAmir657/llvm-project
that referenced
this pull request
Nov 4, 2024
) This commit makes the `generic` target to support FP and LSX, as discussed in llvm#110211. Thereby, it allows 128-bit vector to be enabled by default in the loongarch64 backend.
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.
Add support for the
la64v1.0andla64v1.1processor models, enabling Rust to select features based on the version of the LoongArch instruction set. These values will align with those used in the C/C++ toolchain. 1Footnotes
https://github.com/loongson/la-toolchain-conventions?tab=readme-ov-file#list ↩