[WebAssembly] Split and tidy up target features test #96735
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 splits `target-features.ll` into two tests: `target-features-attrs.ll` and `target-features-cpus.ll`. Now `target-features-attrs.ll` contains tests with bitcode function attributes and `-mattr=` options. The current `target-features.ll` file's FileCheck lines are confusing, mainly because it is unclear how `CHECK` and `ATTRS` lines are meant to be different. Turns out, before llvm@67ec874, `-mattr=` options used to override any existing bitcode function attributes, but after the commit that's not the case anymore. So the original test had a line that tested `i32.atomic.rmw.cmpxchg` was not generated when `-mattr=+simd128` was given (because the existing `+atomics` in the function attributes is overriden). That commit deleted that line and changed some `ATTRS` lines into `CHECK`, which was confusing. This PR simplifies that part and does not test the absence of any instructions, and the effect of `-mattr=` option is only tested with the target features section. And `target-features-cpus.ll` only tests the sets of features enabled by `-mcpu=` lines. It is better to have this as a separate file because once you have bitcode function attributes they end up in the target features section too, making the testing of only the `-mcpu=` options difficult.
|
@llvm/pr-subscribers-backend-webassembly Author: Heejin Ahn (aheejin) ChangesThis splits Now And Full diff: https://github.com/llvm/llvm-project/pull/96735.diff 2 Files Affected:
diff --git a/llvm/test/CodeGen/WebAssembly/target-features-attrs.ll b/llvm/test/CodeGen/WebAssembly/target-features-attrs.ll
index 953a4ac9b41de..25dee51ac8c38 100644
--- a/llvm/test/CodeGen/WebAssembly/target-features-attrs.ll
+++ b/llvm/test/CodeGen/WebAssembly/target-features-attrs.ll
@@ -1,6 +1,5 @@
-; RUN: llc < %s | FileCheck %s --check-prefixes CHECK,ATTRS
-; RUN: llc < %s -mcpu=mvp -mattr=+simd128 | FileCheck %s --check-prefixes CHECK,SIMD128
-; RUN: llc < %s -mcpu=bleeding-edge | FileCheck %s --check-prefixes CHECK,BLEEDING-EDGE
+; RUN: llc -mcpu=mvp < %s | FileCheck %s
+; RUN: llc -mcpu=mvp -mattr=+simd128 < %s | FileCheck %s --check-prefixes SIMD128
; Test that codegen emits target features from the command line or
; function attributes correctly and that features are enabled for the
@@ -33,84 +32,53 @@ attributes #2 = { "target-features"="+reference-types" }
; CHECK-LABEL: fn_atomics:
; Expanded atomicrmw min
-; ATTRS: loop
+; CHECK: loop
; CHECK: i32.atomic.rmw.cmpxchg
-; ATTRS: end_loop
+; CHECK: end_loop
; nontrapping fptoint
; CHECK: i32.trunc_sat_f32_u
-; ATTRS: i32.store
+; CHECK: i32.store
; `fn_nontrapping_fptoint` should be the same as `fn_atomics`
; CHECK-LABEL: fn_nontrapping_fptoint:
; Expanded atomicrmw min
-; ATTRS: loop
+; CHECK: loop
; CHECK: i32.atomic.rmw.cmpxchg
-; ATTRS: end_loop
+; CHECK: end_loop
; nontrapping fptoint
; CHECK: i32.trunc_sat_f32_u
-; ATTRS: i32.store
+; CHECK: i32.store
+; Features in function attributes:
+; +atomics, +nontrapping-fptoint, +reference-types
; CHECK-LABEL: .custom_section.target_features,"",@
-
-; +atomics, +reference-types, +mutable-globals
-; ATTRS-NEXT: .int8 5
-; ATTRS-NEXT: .int8 43
-; ATTRS-NEXT: .int8 7
-; ATTRS-NEXT: .ascii "atomics"
-; ATTRS-NEXT: .int8 43
-; ATTRS-NEXT: .int8 15
-; ATTRS-NEXT: .ascii "mutable-globals"
-; ATTRS-NEXT: .int8 43
-; ATTRS-NEXT: .int8 19
-; ATTRS-NEXT: .ascii "nontrapping-fptoint"
-; ATTRS-NEXT: .int8 43
-; ATTRS-NEXT: .int8 15
-; ATTRS-NEXT: .ascii "reference-types"
-; ATTRS-NEXT: .int8 43
-; ATTRS-NEXT: .int8 8
-
+; CHECK-NEXT: .int8 3
+; CHECK-NEXT: .int8 43
+; CHECK-NEXT: .int8 7
+; CHECK-NEXT: .ascii "atomics"
+; CHECK-NEXT: .int8 43
+; CHECK-NEXT: .int8 19
+; CHECK-NEXT: .ascii "nontrapping-fptoint"
+; CHECK-NEXT: .int8 43
+; CHECK-NEXT: .int8 15
+; CHECK-NEXT: .ascii "reference-types"
+
+; Features in function attributes + features specified by -mattr= option:
; +atomics, +nontrapping-fptoint, +reference-types, +simd128
-; SIMD128-NEXT: .int8 4
-; SIMD128-NEXT: .int8 43
-; SIMD128-NEXT: .int8 7
-; SIMD128-NEXT: .ascii "atomics"
-; SIMD128-NEXT: .int8 43
-; SIMD128-NEXT: .int8 19
-; SIMD128-NEXT: .ascii "nontrapping-fptoint"
-; SIMD128-NEXT: .int8 43
-; SIMD128-NEXT: .int8 15
-; SIMD128-NEXT: .ascii "reference-types"
-; SIMD128-NEXT: .int8 43
-; SIMD128-NEXT: .int8 7
-; SIMD128-NEXT: .ascii "simd128"
-
-; +atomics, +bulk-memory, +mutable-globals, +nontrapping-fptoint,
-; +reference-types, +sign-ext, +simd128, +tail-call
-; BLEEDING-EDGE-NEXT: .int8 8
-; BLEEDING-EDGE-NEXT: .int8 43
-; BLEEDING-EDGE-NEXT: .int8 7
-; BLEEDING-EDGE-NEXT: .ascii "atomics"
-; BLEEDING-EDGE-NEXT: .int8 43
-; BLEEDING-EDGE-NEXT: .int8 11
-; BLEEDING-EDGE-NEXT: .ascii "bulk-memory"
-; BLEEDING-EDGE-NEXT: .int8 43
-; BLEEDING-EDGE-NEXT: .int8 15
-; BLEEDING-EDGE-NEXT: .ascii "mutable-globals"
-; BLEEDING-EDGE-NEXT: .int8 43
-; BLEEDING-EDGE-NEXT: .int8 19
-; BLEEDING-EDGE-NEXT: .ascii "nontrapping-fptoint"
-; BLEEDING-EDGE-NEXT: .int8 43
-; BLEEDING-EDGE-NEXT: .int8 15
-; BLEEDING-EDGE-NEXT: .ascii "reference-types"
-; BLEEDING-EDGE-NEXT: .int8 43
-; BLEEDING-EDGE-NEXT: .int8 8
-; BLEEDING-EDGE-NEXT: .ascii "sign-ext"
-; BLEEDING-EDGE-NEXT: .int8 43
-; BLEEDING-EDGE-NEXT: .int8 7
-; BLEEDING-EDGE-NEXT: .ascii "simd128"
-; BLEEDING-EDGE-NEXT: .int8 43
-; BLEEDING-EDGE-NEXT: .int8 9
-; BLEEDING-EDGE-NEXT: .ascii "tail-call"
+; SIMD128-LABEL: .custom_section.target_features,"",@
+; SIMD128-NEXT: .int8 4
+; SIMD128-NEXT: .int8 43
+; SIMD128-NEXT: .int8 7
+; SIMD128-NEXT: .ascii "atomics"
+; SIMD128-NEXT: .int8 43
+; SIMD128-NEXT: .int8 19
+; SIMD128-NEXT: .ascii "nontrapping-fptoint"
+; SIMD128-NEXT: .int8 43
+; SIMD128-NEXT: .int8 15
+; SIMD128-NEXT: .ascii "reference-types"
+; SIMD128-NEXT: .int8 43
+; SIMD128-NEXT: .int8 7
+; SIMD128-NEXT: .ascii "simd128"
diff --git a/llvm/test/CodeGen/WebAssembly/target-features-cpus.ll b/llvm/test/CodeGen/WebAssembly/target-features-cpus.ll
new file mode 100644
index 0000000000000..d368271682a96
--- /dev/null
+++ b/llvm/test/CodeGen/WebAssembly/target-features-cpus.ll
@@ -0,0 +1,48 @@
+; RUN: llc < %s -mcpu=mvp | FileCheck %s --check-prefixes MVP
+; RUN: llc < %s -mcpu=generic | FileCheck %s --check-prefixes GENERIC
+; RUN: llc < %s | FileCheck %s --check-prefixes GENERIC
+; RUN: llc < %s -mcpu=bleeding-edge | FileCheck %s --check-prefixes BLEEDING-EDGE
+
+; Test that the target features section contains the correct set of features
+; depending on -mcpu= options.
+
+target triple = "wasm32-unknown-unknown"
+
+; mvp: should not contain the target features section
+; MVP-NOT: .custom_section.target_features,"",@
+
+; generic: +mutable-globals, +sign-ext
+; GENERIC-LABEL: .custom_section.target_features,"",@
+; GENERIC-NEXT: .int8 2
+; GENERIC-NEXT: .int8 43
+; GENERIC-NEXT: .int8 15
+; GENERIC-NEXT: .ascii "mutable-globals"
+; GENERIC-NEXT: .int8 43
+; GENERIC-NEXT: .int8 8
+; GENERIC-NEXT: .ascii "sign-ext"
+
+; bleeding-edge: +atomics, +bulk-memory, +mutable-globals, +nontrapping-fptoint,
+; +sign-ext, +simd128, +tail-call
+; BLEEDING-EDGE-LABEL: .section .custom_section.target_features,"",@
+; BLEEDING-EDGE-NEXT: .int8 7
+; BLEEDING-EDGE-NEXT: .int8 43
+; BLEEDING-EDGE-NEXT: .int8 7
+; BLEEDING-EDGE-NEXT: .ascii "atomics"
+; BLEEDING-EDGE-NEXT: .int8 43
+; BLEEDING-EDGE-NEXT: .int8 11
+; BLEEDING-EDGE-NEXT: .ascii "bulk-memory"
+; BLEEDING-EDGE-NEXT: .int8 43
+; BLEEDING-EDGE-NEXT: .int8 15
+; BLEEDING-EDGE-NEXT: .ascii "mutable-globals"
+; BLEEDING-EDGE-NEXT: .int8 43
+; BLEEDING-EDGE-NEXT: .int8 19
+; BLEEDING-EDGE-NEXT: .ascii "nontrapping-fptoint"
+; BLEEDING-EDGE-NEXT: .int8 43
+; BLEEDING-EDGE-NEXT: .int8 8
+; BLEEDING-EDGE-NEXT: .ascii "sign-ext"
+; BLEEDING-EDGE-NEXT: .int8 43
+; BLEEDING-EDGE-NEXT: .int8 7
+; BLEEDING-EDGE-NEXT: .ascii "simd128"
+; BLEEDING-EDGE-NEXT: .int8 43
+; BLEEDING-EDGE-NEXT: .int8 9
+; BLEEDING-EDGE-NEXT: .ascii "tail-call"
|
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/190/builds/749 Here is the relevant piece of the build log for the reference: |
|
@llvm-ci The failure doesn't seem to be related to this PR. |
It is not, the build bot was running out of disk space, and I've since taken care of that issue. Sorry for the noise. |
lravenclaw
pushed a commit
to lravenclaw/llvm-project
that referenced
this pull request
Jul 3, 2024
This splits `target-features.ll` into two tests: `target-features-attrs.ll` and `target-features-cpus.ll`. Now `target-features-attrs.ll` contains tests with bitcode function attributes and `-mattr=` options. The current `target-features.ll` file's FileCheck lines are confusing, mainly because it is unclear how `CHECK` and `ATTRS` lines are meant to be different. Turns out, before llvm@67ec874, `-mattr=` options used to override any existing bitcode function attributes, but after the commit that's not the case anymore. So the original test had a line that tested `i32.atomic.rmw.cmpxchg` was not generated when `-mattr=+simd128` was given (because the existing `+atomics` in the function attributes is overriden). That commit deleted that line and changed some `ATTRS` lines into `CHECK`, which was confusing. This PR simplifies that part and does not test the absence of any instructions, and the effect of `-mattr=` option is only tested with the target features section. And `target-features-cpus.ll` only tests the sets of features enabled by `-mcpu=` lines. It is better to have this as a separate file because once you have bitcode function attributes they end up in the target features section too, making the testing of only the `-mcpu=` options difficult.
AlexisPerry
pushed a commit
to llvm-project-tlp/llvm-project
that referenced
this pull request
Jul 9, 2024
This splits `target-features.ll` into two tests: `target-features-attrs.ll` and `target-features-cpus.ll`. Now `target-features-attrs.ll` contains tests with bitcode function attributes and `-mattr=` options. The current `target-features.ll` file's FileCheck lines are confusing, mainly because it is unclear how `CHECK` and `ATTRS` lines are meant to be different. Turns out, before llvm@67ec874, `-mattr=` options used to override any existing bitcode function attributes, but after the commit that's not the case anymore. So the original test had a line that tested `i32.atomic.rmw.cmpxchg` was not generated when `-mattr=+simd128` was given (because the existing `+atomics` in the function attributes is overriden). That commit deleted that line and changed some `ATTRS` lines into `CHECK`, which was confusing. This PR simplifies that part and does not test the absence of any instructions, and the effect of `-mattr=` option is only tested with the target features section. And `target-features-cpus.ll` only tests the sets of features enabled by `-mcpu=` lines. It is better to have this as a separate file because once you have bitcode function attributes they end up in the target features section too, making the testing of only the `-mcpu=` options difficult.
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 splits
target-features.llinto two tests:target-features-attrs.llandtarget-features-cpus.ll.Now
target-features-attrs.llcontains tests with bitcode function attributes and-mattr=options. The currenttarget-features.llfile's FileCheck lines are confusing, mainly because it is unclear howCHECKandATTRSlines are meant to be different. Turns out, before 67ec874,-mattr=options used to override any existing bitcode function attributes, but after the commit that's not the case anymore. So the original test had a line that testedi32.atomic.rmw.cmpxchgwas not generated when-mattr=+simd128was given (because the existing+atomicsin the function attributes is overriden). That commit deleted that line and changed someATTRSlines intoCHECK, which was confusing. This PR simplifies that part and does not test the absence of any instructions, and the effect of-mattr=option is only tested with the target features section.And
target-features-cpus.llonly tests the sets of features enabled by-mcpu=lines. It is better to have this as a separate file because once you have bitcode function attributes they end up in the target features section too, making the testing of only the-mcpu=options difficult.