[RISCV] Add a new subtarget feature for throttled FP64 vector performance #162399
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
mshockwave
requested review from
lukel97,
mikhailramalho,
preames,
topperc and
wangpc-pp
|
@llvm/pr-subscribers-backend-risc-v Author: Min-Yih Hsu (mshockwave) Changessifive-x390 and sifive-x280 both share the SiFIve7 scheduling model, yet the former has a limited FP64 vector performance. Right now we account for it by instantiating two separate scheduling models (throttled v.s. non-throttled) from the base SiFive7 model. However, this approach (which is also used in other performance features like fast vrgather in SiFive7) does not scale if we add more of these performance features in the future -- the number of scheduling models will simply become unmanageable. The new solution I've been working on is to let a single scheduling model be configured by subtarget features on performance features like these, such that we no longer need to create those derived models. This patch creates the subtarget feature that'll ultimately replace the Full diff: https://github.com/llvm/llvm-project/pull/162399.diff 2 Files Affected:
diff --git a/llvm/lib/Target/RISCV/RISCVFeatures.td b/llvm/lib/Target/RISCV/RISCVFeatures.td
index 27cf057112869..0d3df0e188505 100644
--- a/llvm/lib/Target/RISCV/RISCVFeatures.td
+++ b/llvm/lib/Target/RISCV/RISCVFeatures.td
@@ -1823,6 +1823,10 @@ def TuneConditionalCompressedMoveFusion
def HasConditionalMoveFusion : Predicate<"Subtarget->hasConditionalMoveFusion()">;
def NoConditionalMoveFusion : Predicate<"!Subtarget->hasConditionalMoveFusion()">;
+def TuneHasThrottledVecFP64
+ : SubtargetFeature<"throttled-vec-fp64", "HasThrottledVectorFP64", "true",
+ "Certain vector FP64 operations have limited performance">;
+
def TuneMIPSP8700
: SubtargetFeature<"mips-p8700", "RISCVProcFamily", "MIPSP8700",
"MIPS p8700 processor">;
diff --git a/llvm/test/CodeGen/RISCV/features-info.ll b/llvm/test/CodeGen/RISCV/features-info.ll
index 1a7a72d3e072b..40a976e871988 100644
--- a/llvm/test/CodeGen/RISCV/features-info.ll
+++ b/llvm/test/CodeGen/RISCV/features-info.ll
@@ -179,6 +179,7 @@
; CHECK-NEXT: svpbmt - 'Svpbmt' (Page-Based Memory Types).
; CHECK-NEXT: svvptc - 'Svvptc' (Obviating Memory-Management Instructions after Marking PTEs Valid).
; CHECK-NEXT: tagged-globals - Use an instruction sequence for taking the address of a global that allows a memory tag in the upper address bits.
+; CHECK-NEXT: throttled-vec-fp64 - Certain vector FP64 operations have limited performance.
; CHECK-NEXT: unaligned-scalar-mem - Has reasonably performant unaligned scalar loads and stores.
; CHECK-NEXT: unaligned-vector-mem - Has reasonably performant unaligned vector loads and stores.
; CHECK-NEXT: use-postra-scheduler - Schedule again after register allocation.
|
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/65/builds/23808 Here is the relevant piece of the build log for the reference |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
sifive-x390 and sifive-x280 both share the SiFIve7 scheduling model, yet the former has a limited FP64 vector performance. Right now we account for it by instantiating two separate scheduling models (throttled v.s. non-throttled) from the base SiFive7 model. However, this approach (which is also used in other performance features like fast vrgather in SiFive7) does not scale if we add more of these performance features in the future -- the number of scheduling models will simply become unmanageable.
The new solution I've been working on is to let a single scheduling model be configured by subtarget features on performance features like these, such that we no longer need to create those derived models. This patch creates the subtarget feature that'll ultimately replace the
isFP64Throttledknob in SiFive7 scheduling model mentioned earlier. There will be a follow-up patch to integrate this into the scheduling model.