[RISC-V] Bit counting intrinsics #114337
Merged
[RISC-V] Bit counting intrinsics #114337
Conversation
This file contains 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
dotnet-issue-labeler
bot
added
the
area-CodeGen-coreclr
dotnet-policy-service
bot
added
the
community-contribution
|
Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch |
|
No regressions. Diffs are based on 170,565 contexts (22,709 MinOpts, 147,856 FullOpts). MISSED contexts: 1,020 (0.59%) Overall (-14,440 bytes)
MinOpts (-180 bytes)
FullOpts (-14,260 bytes)
Example diffslinux.riscv64.Checked.mch-36 (-50.00%) : 12110.dasm - System.Numerics.Tensors.TensorPrimitives+LeadingZeroCountOperator`1[int]:Invoke(int):int (FullOpts)@@ -5,12 +5,10 @@
; fp based frame
; partially interruptible
; No matching PGO data
-; 0 inlinees with PGO data; 1 single block inlinees; 1 inlinees without PGO data
; Final local variable assignments
;
-; V00 arg0 [V00,T00] ( 4, 3.50) int -> a0 single-def
+; V00 arg0 [V00,T00] ( 3, 3 ) int -> a0 single-def
;# V01 OutArgs [V01 ] ( 1, 1 ) struct ( 0) [sp+0x00] do-not-enreg[XS] addr-exposed "OutgoingArgSpace" <Empty>
-; V02 tmp1 [V02,T01] ( 3, 2 ) int -> a0 "Inline return value spill temp"
;
; Lcl frame size = 0
Frame info. #outsz=0; #framesz=16; lcl=0
@@ -22,28 +20,15 @@ G_M40987_IG01: ; bbWeight=1, gcrefRegs=0000 {}, byrefRegs=0000 {}, byref,
mv fp, sp
;; size=16 bbWeight=1 PerfScore 9.00
G_M40987_IG02: ; bbWeight=1, gcrefRegs=0000 {}, byrefRegs=0000 {}, byref
- sext.w t6, a0
- beqz t6, G_M40987_IG04
- ;; size=8 bbWeight=1 PerfScore 4.00
-G_M40987_IG03: ; bbWeight=0.50, gcrefRegs=0000 {}, byrefRegs=0000 {}, byref
- lui a1, 0xD1FFAB1E
- addiw a1, a1, 0xD1FFAB1E
- slli a1, a1, 7
- ld a1, 0xD1FFAB1E(a1)
- jalr a1 // System.Numerics.BitOperations:Log2SoftwareFallback(uint):int
- xori a0, a0, 31
- j G_M40987_IG05
- ;; size=28 bbWeight=0.50 PerfScore 4.25
-G_M40987_IG04: ; bbWeight=0.50, gcrefRegs=0000 {}, byrefRegs=0000 {}, byref
- addi a0, zero, 0xD1FFAB1E
- ;; size=4 bbWeight=0.50 PerfScore 0.25
-G_M40987_IG05: ; bbWeight=1, gcrefRegs=0000 {}, byrefRegs=0000 {}, byref, epilog, nogc
+ clzw a0, a0
+ ;; size=4 bbWeight=1 PerfScore 0.50
+G_M40987_IG03: ; bbWeight=1, epilog, nogc, extend
ld ra, 8(sp)
ld fp, 0(sp)
addi sp, sp, 16
ret ;; size=16 bbWeight=1 PerfScore 7.50
-; Total bytes of code 72, prolog size 16, PerfScore 25.00, instruction count 18, allocated bytes for code 72 (MethodHash=a4435fe4) for method System.Numerics.Tensors.TensorPrimitives+LeadingZeroCountOperator`1[int]:Invoke(int):int (FullOpts)
+; Total bytes of code 36, prolog size 16, PerfScore 17.00, instruction count 9, allocated bytes for code 36 (MethodHash=a4435fe4) for method System.Numerics.Tensors.TensorPrimitives+LeadingZeroCountOperator`1[int]:Invoke(int):int (FullOpts)
; ============================================================
Unwind Info:
@@ -54,7 +39,7 @@ Unwind Info:
E bit : 0
X bit : 0
Vers : 0
- Function Length : 18 (0x00012) Actual length = 72 (0x000048)
+ Function Length : 9 (0x00009) Actual length = 36 (0x000024)
---- Epilog scopes ----
---- Scope 0
Epilog Start Offset : 3523193630 (0xd1ffab1e) Actual offset = 3523193630 (0xd1ffab1e) Offset from main function begin = 3523193630 (0xd1ffab1e)-36 (-45.00%) : 32994.dasm - System.Number+BigInteger:CountSignificantBits(uint):uint (FullOpts)@@ -5,12 +5,10 @@
; fp based frame
; partially interruptible
; No matching PGO data
-; 0 inlinees with PGO data; 0 single block inlinees; 1 inlinees without PGO data
; Final local variable assignments
;
-; V00 arg0 [V00,T00] ( 4, 3.50) int -> a0 single-def
+; V00 arg0 [V00,T00] ( 3, 3 ) int -> a0 single-def
;# V01 OutArgs [V01 ] ( 1, 1 ) struct ( 0) [sp+0x00] do-not-enreg[XS] addr-exposed "OutgoingArgSpace" <Empty>
-; V02 tmp1 [V02,T01] ( 3, 2 ) int -> a0 "Inline return value spill temp"
;
; Lcl frame size = 0
Frame info. #outsz=0; #framesz=16; lcl=0
@@ -22,32 +20,17 @@ G_M36112_IG01: ; bbWeight=1, gcrefRegs=0000 {}, byrefRegs=0000 {}, byref,
mv fp, sp
;; size=16 bbWeight=1 PerfScore 9.00
G_M36112_IG02: ; bbWeight=1, gcrefRegs=0000 {}, byrefRegs=0000 {}, byref
- sext.w t6, a0
- beqz t6, G_M36112_IG04
- ;; size=8 bbWeight=1 PerfScore 4.00
-G_M36112_IG03: ; bbWeight=0.50, gcrefRegs=0000 {}, byrefRegs=0000 {}, byref
- lui a1, 0xD1FFAB1E
- addiw a1, a1, 0xD1FFAB1E
- slli a1, a1, 7
- ld a1, 0xD1FFAB1E(a1)
- jalr a1 // System.Numerics.BitOperations:Log2SoftwareFallback(uint):int
- xori a0, a0, 31
- j G_M36112_IG05
- ;; size=28 bbWeight=0.50 PerfScore 4.25
-G_M36112_IG04: ; bbWeight=0.50, gcrefRegs=0000 {}, byrefRegs=0000 {}, byref
- addi a0, zero, 0xD1FFAB1E
- ;; size=4 bbWeight=0.50 PerfScore 0.25
-G_M36112_IG05: ; bbWeight=1, gcrefRegs=0000 {}, byrefRegs=0000 {}, byref
+ clzw a0, a0
subw a0, zero, a0
addiw a0, a0, 0xD1FFAB1E
- ;; size=8 bbWeight=1 PerfScore 1.00
-G_M36112_IG06: ; bbWeight=1, epilog, nogc, extend
+ ;; size=12 bbWeight=1 PerfScore 1.50
+G_M36112_IG03: ; bbWeight=1, epilog, nogc, extend
ld ra, 8(sp)
ld fp, 0(sp)
addi sp, sp, 16
ret ;; size=16 bbWeight=1 PerfScore 7.50
-; Total bytes of code 80, prolog size 16, PerfScore 26.00, instruction count 20, allocated bytes for code 80 (MethodHash=581172ef) for method System.Number+BigInteger:CountSignificantBits(uint):uint (FullOpts)
+; Total bytes of code 44, prolog size 16, PerfScore 18.00, instruction count 11, allocated bytes for code 44 (MethodHash=581172ef) for method System.Number+BigInteger:CountSignificantBits(uint):uint (FullOpts)
; ============================================================
Unwind Info:
@@ -58,7 +41,7 @@ Unwind Info:
E bit : 0
X bit : 0
Vers : 0
- Function Length : 20 (0x00014) Actual length = 80 (0x000050)
+ Function Length : 11 (0x0000b) Actual length = 44 (0x00002c)
---- Epilog scopes ----
---- Scope 0
Epilog Start Offset : 3523193630 (0xd1ffab1e) Actual offset = 3523193630 (0xd1ffab1e) Offset from main function begin = 3523193630 (0xd1ffab1e)-24 (-40.00%) : 12401.dasm - System.Numerics.Tensors.TensorPrimitives+PopCountOperator`1[long]:Invoke(long):long (FullOpts)@@ -5,7 +5,6 @@
; fp based frame
; partially interruptible
; No matching PGO data
-; 0 inlinees with PGO data; 1 single block inlinees; 1 inlinees without PGO data
; Final local variable assignments
;
; V00 arg0 [V00,T00] ( 3, 3 ) long -> a0 single-def
@@ -21,21 +20,15 @@ G_M15459_IG01: ; bbWeight=1, gcrefRegs=0000 {}, byrefRegs=0000 {}, byref,
mv fp, sp
;; size=16 bbWeight=1 PerfScore 9.00
G_M15459_IG02: ; bbWeight=1, gcrefRegs=0000 {}, byrefRegs=0000 {}, byref
- lui a1, 0xD1FFAB1E
- addiw a1, a1, 0xD1FFAB1E
- slli a1, a1, 31
- srli a1, a1, 26
- ld a1, 0xD1FFAB1E(a1)
- jalr a1 // System.Numerics.BitOperations:<PopCount>g__SoftwareFallback|23_0(ulong):int
- slliw a0, a0, 0
- ;; size=28 bbWeight=1 PerfScore 7.50
+ cpop a0, a0
+ ;; size=4 bbWeight=1 PerfScore 0.50
G_M15459_IG03: ; bbWeight=1, epilog, nogc, extend
ld ra, 8(sp)
ld fp, 0(sp)
addi sp, sp, 16
ret ;; size=16 bbWeight=1 PerfScore 7.50
-; Total bytes of code 60, prolog size 16, PerfScore 24.00, instruction count 15, allocated bytes for code 60 (MethodHash=1bb0c39c) for method System.Numerics.Tensors.TensorPrimitives+PopCountOperator`1[long]:Invoke(long):long (FullOpts)
+; Total bytes of code 36, prolog size 16, PerfScore 17.00, instruction count 9, allocated bytes for code 36 (MethodHash=1bb0c39c) for method System.Numerics.Tensors.TensorPrimitives+PopCountOperator`1[long]:Invoke(long):long (FullOpts)
; ============================================================
Unwind Info:
@@ -46,7 +39,7 @@ Unwind Info:
E bit : 0
X bit : 0
Vers : 0
- Function Length : 15 (0x0000f) Actual length = 60 (0x00003c)
+ Function Length : 9 (0x00009) Actual length = 36 (0x000024)
---- Epilog scopes ----
---- Scope 0
Epilog Start Offset : 3523193630 (0xd1ffab1e) Actual offset = 3523193630 (0xd1ffab1e) Offset from main function begin = 3523193630 (0xd1ffab1e)+0 (0.00%) : 171568.dasm - ILGEN_0x2193e4b5:Method_0xeffa27b5(ubyte,long,float,float,byte,double,uint,ubyte,short):int (FullOpts)No diffs found? +0 (0.00%) : 171536.dasm - Generated442:StructConstrainedInterfaceCallsTest() (FullOpts)No diffs found? +0 (0.00%) : 171472.dasm - ValueNumberingCheckedCastsOfConstants:g__ConfirmUInt64OneDecrementUnderUInt64MaxValueCastToUInt32Overflows|97_24() (FullOpts)No diffs found? DetailsSize improvements/regressions per collection
PerfScore improvements/regressions per collection
Context information
jit-analyze output |
RISC-V Release-CLR-VF2: 9531 / 9551 (99.79%)Release-CLR-VF2.md, Release-CLR-VF2.xml, testclr_output.tar.gz Build information and commandsGIT: RISC-V Release-CLR-QEMU: 9531 / 9551 (99.79%)Release-CLR-QEMU.md, Release-CLR-QEMU.xml, testclr_output.tar.gz Build information and commandsGIT: RISC-V Release-FX-VF2: 628462 / 660362 (95.17%)Build information and commandsGIT: RISC-V Release-FX-QEMU: 643107 / 667247 (96.38%)Release-FX-QEMU.md, Release-FX-QEMU.xml, testfx_output.tar.gz Build information and commandsGIT: |
<testsuite name="readytorun.determinism.crossgen2determinism" time="1.970">
<testcase name="crossgen2determinism.sh" classname="readytorun.determinism.crossgen2determinism.crossgen2determinism" result="FAILED" time="1.970">
<failure message="Unknown exit code 101." type="exitcode_101">Could not find file '/var/lib/go-agent/pipelines/Release-CLR-VF2/coreclr.Release/readytorun/determinism/crossgen2determinism/crossgen2smoke1.ildll'.
File name: '/var/lib/go-agent/pipelines/Release-CLR-VF2/coreclr.Release/readytorun/determinism/crossgen2determinism/crossgen2smoke1.ildll'
at Interop.ThrowExceptionForIoErrno(ErrorInfo errorInfo, String path, Boolean isDirError)
at Microsoft.Win32.SafeHandles.SafeFileHandle.Open(String path, OpenFlags flags, Int32 mode, Boolean failForSymlink, Boolean& wasSymlink, Func`4 createOpenException)
at Microsoft.Win32.SafeHandles.SafeFileHandle.Open(String fullPath, FileMode mode, FileAccess access, FileShare share, FileOptions options, Int64 preallocationSize, UnixFileMode openPermissions, Int64& fileLength, UnixFileMode& filePermissions, Boolean failForSymlink, Boolean& wasSymlink, Func`4 createOpenException)
at System.IO.File.OpenHandle(String path, FileMode mode, FileAccess access, FileShare share, FileOptions options, Int64 preallocationSize)
at System.IO.File.ReadAllBytes(String path)
at Program.TestEntryPoint() in /godata/pipelines/Release-build/runtime/src/tests/readytorun/determinism/Program.cs:line 13
at __GeneratedMainWrapper.Main() in /godata/pipelines/Release-build/runtime/artifacts/tests/coreclr/obj/linux.riscv64.Release/Managed/readytorun/determinism/crossgen2determinism/XUnitWrapperGenerator/XUnitWrapperGenerator.XUnitWrapperGenerator/SimpleRunner.g.cs:line 7do we know if that path is wrong on the test runner or the file wasn't generated by the build? File should be generated at build time and placed at pwd when the test runs: |
The file is built, it does pass locally. Looks like it's not collected or at wrong path for the test run. @sirntar will be working on it. |
|
@am11 The file is generated at build time, but isn't sent to CI tests runners. I am investigating this at the moment. It's probably some packing issue... |
BruceForstall
approved these changes
Apr 8, 2025
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.
(Leading|Trailing)ZeroCount, PopCount.
Part of #84834, cc @dotnet/samsung