Chromium Code Reviews
|
|
Chromium Code Reviews
Issue
1812233003:
SkNx refresh (Closed)
DescriptionSkNx refresh
- rearrange a bit
- fewer macros
- hooks for all operators
- add left and right scalar operator overrides
- add +=, &=, <<=, etc.
- add SkNx_split() and SkNx_join()
- simplify the many rsqrt() and invert() options to just what we actually use
This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code.
It's unclear to me if split/join want to be external, static methods, or non-static methods (SkNx_join(), Sk4f::Join(), x.join()). Time will tell?
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1812233003
CQ_EXTRA_TRYBOTS=client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Release-Trybot;client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot
Committed: https://skia.googlesource.com/skia/+/f8f90e4a85638faa18e7b4133cfe4d1ff5b1b23e
Patch Set 1 #Patch Set 2 : move around #Patch Set 3 : dull #Patch Set 4 : fix? #Patch Set 5 : const U& #Patch Set 6 : limits #Patch Set 7 : msvc u serious? #Patch Set 8 : more neon sk4i #Patch Set 9 : more Sk4i NEON math #Patch Set 10 : use split and join #Patch Set 11 : invert and rsqrt cleanup #Patch Set 12 : explicit makes these fine #Patch Set 13 : clearer #Patch Set 14 : nosimd fixes #Patch Set 15 : typo #
Messages
Total messages: 51 (32 generated)
Description was changed from ========== SkNx refresh - rearrange a bit - add left and right scalar operator overrides - add +=, &=, <<=, etc. BUG=skia: ========== to ========== SkNx refresh - rearrange a bit - add left and right scalar operator overrides - add +=, &=, <<=, etc. BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot ==========
Description was changed from ========== SkNx refresh - rearrange a bit - add left and right scalar operator overrides - add +=, &=, <<=, etc. BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot ========== to ========== SkNx refresh - rearrange a bit - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split and SkNx_join BUG=skia: ==========
Description was changed from ========== SkNx refresh - rearrange a bit - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split and SkNx_join BUG=skia: ========== to ========== SkNx refresh - rearrange a bit - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split and SkNx_join - simplify many rsqrt and invert options to just rsqrt(), invert() BUG=skia: ==========
Description was changed from ========== SkNx refresh - rearrange a bit - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split and SkNx_join - simplify many rsqrt and invert options to just rsqrt(), invert() BUG=skia: ========== to ========== SkNx refresh - rearrange a bit - fewer macros - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split and SkNx_join - simplify the many rsqrt() and invert() options to just what we actually use BUG=skia: ==========
Description was changed from ========== SkNx refresh - rearrange a bit - fewer macros - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split and SkNx_join - simplify the many rsqrt() and invert() options to just what we actually use BUG=skia: ========== to ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split and SkNx_join - simplify the many rsqrt() and invert() options to just what we actually use BUG=skia: ==========
The CQ bit was checked by mtklein@chromium.org to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1812233003/20001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1812233003/20001
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: Try jobs failed on following builders: Build-Ubuntu-GCC-Mips-Debug-Android-Trybot on client.skia.compile (JOB_FAILED, http://build.chromium.org/p/client.skia.compile/builders/Build-Ubuntu-GCC-Mip...) Build-Ubuntu-GCC-x86_64-Release-Trybot on client.skia.compile (JOB_FAILED, http://build.chromium.org/p/client.skia.compile/builders/Build-Ubuntu-GCC-x86...)
Description was changed from ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split and SkNx_join - simplify the many rsqrt() and invert() options to just what we actually use BUG=skia: ========== to ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split and SkNx_join - simplify the many rsqrt() and invert() options to just what we actually use BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot ==========
The CQ bit was checked by mtklein@chromium.org to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1812233003/80001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1812233003/80001
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: Try jobs failed on following builders: Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Debug-Trybot on client.skia (JOB_FAILED, http://build.chromium.org/p/client.skia/builders/Test-Ubuntu-GCC-GCE-CPU-AVX2...) Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-Shared-Trybot on client.skia (JOB_FAILED, http://build.chromium.org/p/client.skia/builders/Test-Ubuntu-GCC-GCE-CPU-AVX2...)
The CQ bit was checked by mtklein@chromium.org to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1812233003/100001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1812233003/100001
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: Try jobs failed on following builders: Build-Ubuntu-GCC-Arm7-Debug-Android-Trybot on client.skia.compile (JOB_FAILED, http://build.chromium.org/p/client.skia.compile/builders/Build-Ubuntu-GCC-Arm...)
Description was changed from ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split and SkNx_join - simplify the many rsqrt() and invert() options to just what we actually use BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot ========== to ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split and SkNx_join - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot ==========
The CQ bit was checked by mtklein@google.com to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1812233003/140001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1812233003/140001
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: Try jobs failed on following builders: Build-Ubuntu-GCC-Arm7-Debug-Android-Trybot on client.skia.compile (JOB_FAILED, http://build.chromium.org/p/client.skia.compile/builders/Build-Ubuntu-GCC-Arm...)
The CQ bit was checked by mtklein@google.com to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1812233003/160001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1812233003/160001
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: This issue passed the CQ dry run.
Description was changed from ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split and SkNx_join - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot ========== to ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split() and SkNx_join() - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot ==========
The CQ bit was checked by mtklein@google.com to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1812233003/180001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1812233003/180001
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: This issue passed the CQ dry run.
Description was changed from ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split() and SkNx_join() - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot ========== to ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split() and SkNx_join() - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot;client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Release-Trybot ==========
Description was changed from ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split() and SkNx_join() - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot;client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Release-Trybot ========== to ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split() and SkNx_join() - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. It's unclear to me if split/join want to be external, static methods, or non-static methods (SkNx_join(), Sk4f::Join(), x.join()). Time will tell? BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot;client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Release-Trybot ==========
Description was changed from ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split() and SkNx_join() - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. It's unclear to me if split/join want to be external, static methods, or non-static methods (SkNx_join(), Sk4f::Join(), x.join()). Time will tell? BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot;client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Release-Trybot ========== to ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split() and SkNx_join() - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. The new <N,T,U> methods that allow constants on the left or right have lead to some odd ambiguities for operator* in Sk4px. I've fixed them up by replacing it with a call to mulWiden(), which is equivalent. Not too worried about Sk4px... I hope to remove it one day in favor of writing things directly with SkNx. It's unclear to me if split/join want to be external, static methods, or non-static methods (SkNx_join(), Sk4f::Join(), x.join()). Time will tell? BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot;client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Release-Trybot ==========
Description was changed from ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split() and SkNx_join() - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. The new <N,T,U> methods that allow constants on the left or right have lead to some odd ambiguities for operator* in Sk4px. I've fixed them up by replacing it with a call to mulWiden(), which is equivalent. Not too worried about Sk4px... I hope to remove it one day in favor of writing things directly with SkNx. It's unclear to me if split/join want to be external, static methods, or non-static methods (SkNx_join(), Sk4f::Join(), x.join()). Time will tell? BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot;client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Release-Trybot ========== to ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split() and SkNx_join() - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. The new <N,T,U> methods that allow constants on the left or right have lead to some odd ambiguities for operator* in Sk4px. I've fixed them up by replacing it with a call to mulWiden(), which is equivalent. Not too worried about Sk4px... I hope to remove it one day in favor of writing things directly with SkNx. It's unclear to me if split/join want to be external, static methods, or non-static methods (SkNx_join(), Sk4f::Join(), x.join()). Time will tell? BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Release-Trybot;client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot ==========
Description was changed from ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split() and SkNx_join() - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. The new <N,T,U> methods that allow constants on the left or right have lead to some odd ambiguities for operator* in Sk4px. I've fixed them up by replacing it with a call to mulWiden(), which is equivalent. Not too worried about Sk4px... I hope to remove it one day in favor of writing things directly with SkNx. It's unclear to me if split/join want to be external, static methods, or non-static methods (SkNx_join(), Sk4f::Join(), x.join()). Time will tell? BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Release-Trybot;client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot ========== to ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split() and SkNx_join() - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. It's unclear to me if split/join want to be external, static methods, or non-static methods (SkNx_join(), Sk4f::Join(), x.join()). Time will tell? BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Release-Trybot;client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot ==========
The CQ bit was checked by mtklein@google.com to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1812233003/240001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1812233003/240001
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: Try jobs failed on following builders: Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot on client.skia (JOB_FAILED, http://build.chromium.org/p/client.skia/builders/Test-Ubuntu-GCC-GCE-CPU-AVX2...) Build-Ubuntu-GCC-Arm7-Debug-Android-Trybot on client.skia.compile (JOB_FAILED, http://build.chromium.org/p/client.skia.compile/builders/Build-Ubuntu-GCC-Arm...) Build-Ubuntu-GCC-Mips-Debug-Android-Trybot on client.skia.compile (JOB_FAILED, http://build.chromium.org/p/client.skia.compile/builders/Build-Ubuntu-GCC-Mip...)
The CQ bit was checked by mtklein@google.com to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1812233003/280001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1812233003/280001
mtklein@chromium.org changed reviewers: + herb@google.com - mtklein@google.com
lgtm
The CQ bit was unchecked by mtklein@google.com
The CQ bit was checked by mtklein@google.com
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/1812233003/280001 View timeline at https://chromium-cq-status.appspot.com/patch-timeline/1812233003/280001
Message was sent while issue was closed.
Description was changed from ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split() and SkNx_join() - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. It's unclear to me if split/join want to be external, static methods, or non-static methods (SkNx_join(), Sk4f::Join(), x.join()). Time will tell? BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Release-Trybot;client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot ========== to ========== SkNx refresh - rearrange a bit - fewer macros - hooks for all operators - add left and right scalar operator overrides - add +=, &=, <<=, etc. - add SkNx_split() and SkNx_join() - simplify the many rsqrt() and invert() options to just what we actually use This refactoring pointed out that our float <-> int NEON conversions are not specialized, so I've implemented them. It seems nice that this is an error rather than silently falling back to serial code. It's unclear to me if split/join want to be external, static methods, or non-static methods (SkNx_join(), Sk4f::Join(), x.join()). Time will tell? BUG=skia: GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is... CQ_EXTRA_TRYBOTS=client.skia.android:Test-Android-GCC-Nexus5-CPU-NEON-Arm7-Release-Trybot;client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-SKNX_NO_SIMD-Trybot Committed: https://skia.googlesource.com/skia/+/f8f90e4a85638faa18e7b4133cfe4d1ff5b1b23e ==========
Message was sent while issue was closed.
Committed patchset #15 (id:280001) as https://skia.googlesource.com/skia/+/f8f90e4a85638faa18e7b4133cfe4d1ff5b1b23e |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
