On Wed, 11 Oct 2023 20:58:23 GMT, Srinivas Vamsi Parasa
wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX
On Thu, 12 Oct 2023 03:44:28 GMT, Danny Thomas wrote:
> At least on Saphire Rapids the [emulation suggested
> here](https://github.com/natmaurice/x86-simd-sort/commit/41d03b2d8f3b62a2ee6a3a97a8da7f193a407026)
> only imposes a 6% penalty for `intSort`, while also mitigating the
> performance is
On Wed, 11 Oct 2023 22:40:20 GMT, Sandhya Viswanathan
wrote:
>> I see now that this is an unrelated change. In that case please avoid
>> changing whitespace in unrelated files for this PR.
>
> @erikj79 This space was inadvertently added as part of
> (https://github.com/openjdk/jdk/pull/14227)
On Wed, 11 Oct 2023 20:58:23 GMT, Srinivas Vamsi Parasa
wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX
On Thu, 12 Oct 2023 04:41:37 GMT, Vladimir Kozlov wrote:
> My tier1-3,xcomp testing for v04 passed. I am integrating these changes. Lets
> continue discussion about changes for AMD in
> https://bugs.openjdk.org/browse/JDK-8317976.
Thank you, Vladimir!
-
PR Comment: https://git.op
On Wed, 11 Oct 2023 20:58:23 GMT, Srinivas Vamsi Parasa
wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX
On Wed, 11 Oct 2023 20:58:23 GMT, Srinivas Vamsi Parasa
wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX
On Wed, 11 Oct 2023 23:40:55 GMT, Sandhya Viswanathan
wrote:
> > It makes sense to let `-XX:ControlIntrinsic=` overrule
> > `VM_Version::is_intel()` check and enable the intrinsics when `AVX512DQ` is
> > supported.
>
> May be it could be done as part of
> https://bugs.openjdk.org/browse/JDK-
On Wed, 11 Oct 2023 23:14:26 GMT, Vladimir Ivanov wrote:
> Proposed patch has one disadvantage: there's no way to override ergonomics
> decisions on AMD CPUs and forcibly enable the intrinsic without rebuilding
> the JVM.
>
> For many other intrinsics there are flags which enable finer grained
On Wed, 11 Oct 2023 20:58:23 GMT, Srinivas Vamsi Parasa
wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX
On Wed, 11 Oct 2023 22:25:14 GMT, Erik Joelsson wrote:
>> Hi Erik (@erikj79),
>> BUILD_LIBFALLBACKLINKER is from different PR (#13079). If I understand
>> correctly, for LIB_SIMD_SORT, are you suggesting that we don't pad the lines
>> with spaces to align features into columns and instead just
On Wed, 11 Oct 2023 21:04:25 GMT, Srinivas Vamsi Parasa
wrote:
>> make/modules/java.base/Lib.gmk line 230:
>>
>>> 228: CFLAGS := $(CFLAGS_JDKLIB) $(LIBFFI_CFLAGS), \
>>> 229: LDFLAGS := $(LDFLAGS_JDKLIB) \
>>> 230: $(call SET_SHARED_LIBRARY_ORIGIN), \
>>
>> If you
On Wed, 11 Oct 2023 20:58:23 GMT, Srinivas Vamsi Parasa
wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX
On Wed, 11 Oct 2023 18:31:44 GMT, Sandhya Viswanathan
wrote:
>> Also @forceinline in these changes only works for case when new intrinsics
>> are not used.
>> I would suggest to adapt/update JMH benchmark to cover all cases and see
>> effect @forceinline without intrinsics.
>> That will tell u
On Wed, 11 Oct 2023 20:48:06 GMT, Erik Joelsson wrote:
>> Srinivas Vamsi Parasa has updated the pull request incrementally with one
>> additional commit since the last revision:
>>
>> Add @ForceInline annotation to insertion and mixedInsertion sort
>
> make/modules/java.base/Lib.gmk line 230:
> The goal of this PR is to address the follow-up comments to the SIMD
> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
> Arrays.sort() methods.
> The proposed changes are:
>
> 1) Restriction of the AVX512 sort acceleration to only Intel CPUs. A
> performance regression (d
On Wed, 11 Oct 2023 20:31:05 GMT, Srinivas Vamsi Parasa
wrote:
>> Hi @vamsi-parasa,
>>
>> Both methods mixedInsertionSort and insertionSort are covered by intrinsics.
>> But insertionSort is run on leftmnost (one) part only and on small ( <
>> MAX_INSERTION_SORT_SIZE = 44) arrays.
>> Do we act
On Wed, 11 Oct 2023 20:10:12 GMT, iaroslavski wrote:
> > > > > Also @forceinline in these changes only works for case when new
> > > > > intrinsics are not used. I would suggest to adapt/update JMH
> > > > > benchmark to cover all cases and see effect @forceinline without
> > > > > intrinsics.
On Wed, 11 Oct 2023 17:28:12 GMT, Srinivas Vamsi Parasa
wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX
On Wed, 11 Oct 2023 07:10:57 GMT, iaroslavski wrote:
> To have clear picture could you please run benchmarking to compare both
> cases: current implementation and implementation with Java insertionSort only?
>
> see changes `sort(int.class, a, Unsafe.ARRAY_INT_BASE_OFFSET, low, high,
> DualPiv
On Wed, 11 Oct 2023 19:56:47 GMT, Srinivas Vamsi Parasa
wrote:
> > > > Also @forceinline in these changes only works for case when new
> > > > intrinsics are not used. I would suggest to adapt/update JMH benchmark
> > > > to cover all cases and see effect @forceinline without intrinsics. That
On Wed, 11 Oct 2023 17:28:12 GMT, Srinivas Vamsi Parasa
wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX
On Wed, 11 Oct 2023 19:06:24 GMT, Vladimir Kozlov wrote:
>> @vnkozlov Please advice if we can integrate this PR or if you would like to
>> run some tests first.
>
> Okay. I will start testing for current changes. @sviswa7 please file RFE for
> Zen 4. If we get patch for it we do followup change
On Wed, 11 Oct 2023 17:22:56 GMT, Srinivas Vamsi Parasa
wrote:
> > Also @forceinline in these changes only works for case when new intrinsics
> > are not used. I would suggest to adapt/update JMH benchmark to cover all
> > cases and see effect @forceinline without intrinsics. That will tell us
On Wed, 11 Oct 2023 18:31:44 GMT, Sandhya Viswanathan
wrote:
>> Also @forceinline in these changes only works for case when new intrinsics
>> are not used.
>> I would suggest to adapt/update JMH benchmark to cover all cases and see
>> effect @forceinline without intrinsics.
>> That will tell u
On Wed, 11 Oct 2023 17:28:12 GMT, Srinivas Vamsi Parasa
wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX
On Tue, 10 Oct 2023 18:40:30 GMT, R1chterScale wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX512 sort a
On Tue, 10 Oct 2023 22:29:55 GMT, Vladimir Kozlov wrote:
>> Srinivas Vamsi Parasa has updated the pull request incrementally with one
>> additional commit since the last revision:
>>
>> fix whitespace in build script
>
> Also @forceinline in these changes only works for case when new intrinsi
On Wed, 11 Oct 2023 09:25:15 GMT, Andrew Haley wrote:
> > Forgive me, I might be missing something very obvious, but is there any
> > particular reason to entirely disable the SIMD accelerated sort on Zen 4
> > rather than having an alternate code path for Zen 4 where it has the
> > `compresss
On Wed, 11 Oct 2023 17:28:12 GMT, Srinivas Vamsi Parasa
wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX
On Wed, 11 Oct 2023 06:59:47 GMT, iaroslavski wrote:
> Also @forceinline in these changes only works for case when new intrinsics
> are not used. I would suggest to adapt/update JMH benchmark to cover all
> cases and see effect @forceinline without intrinsics. That will tell us which
> @forcei
> The goal of this PR is to address the follow-up comments to the SIMD
> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
> Arrays.sort() methods.
> The proposed changes are:
>
> 1) Restriction of the AVX512 sort acceleration to only Intel CPUs. A
> performance regression (d
On Tue, 10 Oct 2023 20:21:56 GMT, Srinivas Vamsi Parasa
wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX
On Tue, 10 Oct 2023 18:40:30 GMT, R1chterScale wrote:
> Forgive me, I might be missing something very obvious, but is there any
> particular reason to entirely disable the SIMD accelerated sort on Zen 4
> rather than having an alternate code path for Zen 4 where it has the
> `compressstoreu` i
On Tue, 10 Oct 2023 20:04:48 GMT, Srinivas Vamsi Parasa
wrote:
>> In #14227, you inadvertently added an extra space at line 230 in
>> make/modules/java.base/Lib.gmk
>> (https://github.com/openjdk/jdk/pull/14227/files#diff-c2e113e4b2661697750fd5e6dcc0908fa98563ccfb3801c8b0e3a70174041b81).
>>
>>
On Tue, 10 Oct 2023 22:29:55 GMT, Vladimir Kozlov wrote:
>> Srinivas Vamsi Parasa has updated the pull request incrementally with one
>> additional commit since the last revision:
>>
>> fix whitespace in build script
>
> Also @forceinline in these changes only works for case when new intrinsi
On Tue, 10 Oct 2023 22:29:55 GMT, Vladimir Kozlov wrote:
> Also @forceinline in these changes only works for case when new intrinsics
> are not used. I would suggest to adapt/update JMH benchmark to cover all
> cases and see effect @forceinline without intrinsics. That will tell us which
> @fo
On Tue, 10 Oct 2023 20:21:56 GMT, Srinivas Vamsi Parasa
wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX
On Tue, 10 Oct 2023 20:14:51 GMT, iaroslavski wrote:
> Is ok that partitionDualPivot, partitionSinglePivot and mixedInsertionSort,
> insertionSort are annotated differently?
Good question. Someone familiar with this Java code should answer.
Note, **@forceinline** annotation is used by C2 JIT
On Tue, 10 Oct 2023 20:04:48 GMT, Srinivas Vamsi Parasa
wrote:
>> In #14227, you inadvertently added an extra space at line 230 in
>> make/modules/java.base/Lib.gmk
>> (https://github.com/openjdk/jdk/pull/14227/files#diff-c2e113e4b2661697750fd5e6dcc0908fa98563ccfb3801c8b0e3a70174041b81).
>>
>>
> The goal of this PR is to address the follow-up comments to the SIMD
> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
> Arrays.sort() methods.
> The proposed changes are:
>
> 1) Restriction of the AVX512 sort acceleration to only Intel CPUs. A
> performance regression (d
On Tue, 10 Oct 2023 19:48:50 GMT, Magnus Ihse Bursie wrote:
> In #14227, you inadvertently added an extra space at line 230 in
> make/modules/java.base/Lib.gmk
Hi Magnus (@magicus), please see the extra space fixed in the latest commit.
Thanks,
Vamsi
-
PR Comment: https://git.op
On Tue, 10 Oct 2023 19:03:42 GMT, Srinivas Vamsi Parasa
wrote:
>> The goal of this PR is to address the follow-up comments to the SIMD
>> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
>> Arrays.sort() methods.
>> The proposed changes are:
>>
>> 1) Restriction of the AVX
> The goal of this PR is to address the follow-up comments to the SIMD
> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
> Arrays.sort() methods.
> The proposed changes are:
>
> 1) Restriction of the AVX512 sort acceleration to only Intel CPUs. A
> performance regression (d
On Tue, 10 Oct 2023 18:01:59 GMT, Vladimir Kozlov wrote:
>> Srinivas Vamsi Parasa has updated the pull request incrementally with one
>> additional commit since the last revision:
>>
>> pragma workround for GCC12 bug
>
> What is change for "Addressing the build failure due to a bug in GCC 12"
On Tue, 10 Oct 2023 17:55:43 GMT, Vladimir Kozlov wrote:
>> Srinivas Vamsi Parasa has updated the pull request incrementally with one
>> additional commit since the last revision:
>>
>> pragma workround for GCC12 bug
>
> src/hotspot/cpu/x86/stubGenerator_x86_64.cpp line 4176:
>
>> 4174:
>>
On Tue, 10 Oct 2023 16:44:03 GMT, Srinivas Vamsi Parasa
wrote:
> The goal of this PR is to address the follow-up comments to the SIMD
> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
> Arrays.sort() methods.
> The proposed changes are:
>
> 1) Restriction of the AVX512 so
On Tue, 10 Oct 2023 18:01:59 GMT, Vladimir Kozlov wrote:
> What is change for "Addressing the build failure due to a bug in GCC 12"?
Hello Vladimir,
The change for addressing the build failure will be pushed shortly.
Thanks,
Vamsi
-
PR Comment: https://git.openjdk.org/jdk/pull/1
On Tue, 10 Oct 2023 16:44:03 GMT, Srinivas Vamsi Parasa
wrote:
> The goal of this PR is to address the follow-up comments to the SIMD
> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
> Arrays.sort() methods.
> The proposed changes are:
>
> 1) Restriction of the AVX512 so
On Tue, 10 Oct 2023 16:44:03 GMT, Srinivas Vamsi Parasa
wrote:
> The goal of this PR is to address the follow-up comments to the SIMD
> accelerated sort PR (#14227) which implemented AVX512 intrinsics for
> Arrays.sort() methods.
> The proposed changes are:
>
> 1) Restriction of the AVX512 so
50 matches
Mail list logo
