Page MenuHomePhabricator

nilanjana_basu (NILANJANA BASU)
User

Projects

User does not belong to any projects.

User Details

User Since
Jun 20 2019, 5:36 PM (189 w, 2 d)

Recent Activity

Tue, Jan 31

nilanjana_basu published D142594: [AArch64] Eliminating the use of integer unit in moving from a Neon scalar result of a uaddlv to a Neon vector for review.

The code generated for inserting the scalar result of a 'uaddlv' intrinsic function to a vector exhibits a redundant move to the integer register first, & then moves it back to the destination vector Neon register. This can be done directly. In my understanding, this issue occurs because the selected instruction for vector insertion expects an i32/i64 integer register to hold the value and the selected instruction for the 'uaddlv' function is giving an output of i32/i64 at the instruction selection time, whereas in practice the result of 'uaddlv' goes to a Neon register. Therefore I added patterns to match the combined use of these two instructions (uaddlv & vector_insert), to generate the direct move from the source to destination Neon register.

Tue, Jan 31, 6:37 PM · Restricted Project, Restricted Project
nilanjana_basu requested review of D143038: [AArch64] Added tests for inserting scalar result of uaddlv neon instrinsic function into a vector.
Tue, Jan 31, 5:32 PM · Restricted Project, Restricted Project

Dec 15 2022

nilanjana_basu added a comment to D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Addressed the final comments in a separate commit.

Dec 15 2022, 12:13 PM · Restricted Project, Restricted Project
nilanjana_basu committed rG795868285db9: [AArch64] Minor changes and sanity checks in relation to https://reviews.llvm. (authored by nilanjana_basu).
[AArch64] Minor changes and sanity checks in relation to https://reviews.llvm.
Dec 15 2022, 12:10 PM · Restricted Project, Restricted Project
nilanjana_basu committed rG02d09ffc1b09: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to… (authored by nilanjana_basu).
[AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to…
Dec 15 2022, 7:21 AM · Restricted Project, Restricted Project
nilanjana_basu committed rGa645ec0d3df3: [AArch64] Extra unit tests for trunc lowering of vectors (authored by nilanjana_basu).
[AArch64] Extra unit tests for trunc lowering of vectors
Dec 15 2022, 7:21 AM · Restricted Project, Restricted Project
nilanjana_basu closed D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.
Dec 15 2022, 7:21 AM · Restricted Project, Restricted Project
nilanjana_basu closed D137293: [AArch64] Extra unit tests for trunc lowering of vectors.
Dec 15 2022, 7:21 AM · Restricted Project, Restricted Project

Dec 9 2022

nilanjana_basu committed rG955c0f13cd70: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl… (authored by nilanjana_basu).
[AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl…
Dec 9 2022, 12:51 AM · Restricted Project, Restricted Project
nilanjana_basu committed rGaf42d80a63e3: [AArch64] Unit test for zext lowering for different types of vectors (authored by nilanjana_basu).
[AArch64] Unit test for zext lowering for different types of vectors
Dec 9 2022, 12:51 AM · Restricted Project, Restricted Project
nilanjana_basu closed D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.
Dec 9 2022, 12:51 AM · Restricted Project, Restricted Project
nilanjana_basu closed D137993: [AArch64] Unit test for zext lowering for different types of vectors.
Dec 9 2022, 12:51 AM · Restricted Project, Restricted Project

Dec 8 2022

nilanjana_basu updated the diff for D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.

Added an assert for an extra check

Dec 8 2022, 11:13 AM · Restricted Project, Restricted Project
nilanjana_basu added inline comments to D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.
Dec 8 2022, 10:40 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.

Trying to fix patching error

Dec 8 2022, 10:33 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.

Re-based on newly added tests

Dec 8 2022, 10:27 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D137993: [AArch64] Unit test for zext lowering for different types of vectors.

Added an extra unit test for 'zext <8xi8> to <8xi33>'. Added GISel path testing.

Dec 8 2022, 10:24 AM · Restricted Project, Restricted Project

Dec 2 2022

nilanjana_basu added a comment to D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.

Removed tbl-conversion cases to destination vector element width above 64, due to observed performance regressions. Will move this to a later patch, once we find a fix.

Dec 2 2022, 11:29 AM · Restricted Project, Restricted Project
nilanjana_basu updated the summary of D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.
Dec 2 2022, 11:25 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.

Ran clang-format

Dec 2 2022, 11:24 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.

Blocked tbl-conversion for destination element size above 64 since only 2 or less destination vector elements can be chosen with each tbl instruction in these cases, making it less beneficial

Dec 2 2022, 11:21 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D137993: [AArch64] Unit test for zext lowering for different types of vectors.

Added two new test cases for destination vectors of arbitrary lengths

Dec 2 2022, 11:04 AM · Restricted Project, Restricted Project

Dec 1 2022

nilanjana_basu committed rT08de51078b0a: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for… (authored by nilanjana_basu).
[MicroBenchmarks,AArch64] Added correctness test & other performance tests for…
Dec 1 2022, 10:09 PM · Restricted Project
nilanjana_basu closed D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations.
Dec 1 2022, 10:09 PM · Restricted Project
nilanjana_basu retitled D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations from [MicroBenchmarks,AArch64] Added correctness test for truncate or zero-extend vector operations to [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations.
Dec 1 2022, 9:49 PM · Restricted Project
nilanjana_basu updated the diff for D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations.

Removed duplicate code by adding function pointers as parameter as advised in the reviews. Added more performance tests using ZExt/Trunc operations in combination with addition operation.

Dec 1 2022, 12:21 PM · Restricted Project

Nov 29 2022

nilanjana_basu abandoned D138896: [AArch64] Unit tests for multiple back-to-back zext lowering to tbl for vectors.
Nov 29 2022, 11:14 AM · Restricted Project, Restricted Project
nilanjana_basu added inline comments to D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.
Nov 29 2022, 11:13 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.

Fixed rebasing error of duplicated tests

Nov 29 2022, 11:01 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.

Trying to fix patching error again

Nov 29 2022, 8:56 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.

Trying to fix patching error because of rebasing

Nov 29 2022, 8:54 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.

Rebased on latest updated zext unit tests

Nov 29 2022, 7:44 AM · Restricted Project, Restricted Project
nilanjana_basu updated the summary of D137993: [AArch64] Unit test for zext lowering for different types of vectors.
Nov 29 2022, 7:41 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D137993: [AArch64] Unit test for zext lowering for different types of vectors.

Added tests for multiple back-to-back zext instructions for vectors & rebased on recent commit

Nov 29 2022, 7:13 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D138896: [AArch64] Unit tests for multiple back-to-back zext lowering to tbl for vectors.

Updating labels after rebasing

Nov 29 2022, 2:47 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D138896: [AArch64] Unit tests for multiple back-to-back zext lowering to tbl for vectors.

Trying to patch due to review dependency

Nov 29 2022, 2:17 AM · Restricted Project, Restricted Project
nilanjana_basu added inline comments to D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations.
Nov 29 2022, 12:58 AM · Restricted Project
nilanjana_basu updated the diff for D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations.

Ran clang-format

Nov 29 2022, 12:46 AM · Restricted Project
nilanjana_basu updated the diff for D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations.

Variable name changes

Nov 29 2022, 12:39 AM · Restricted Project
nilanjana_basu updated the diff for D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations.

Addressed reviewer comments

Nov 29 2022, 12:36 AM · Restricted Project

Nov 28 2022

nilanjana_basu updated the diff for D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.

Rebasing & merging on a recent commit

Nov 28 2022, 10:48 PM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D138896: [AArch64] Unit tests for multiple back-to-back zext lowering to tbl for vectors.

Trying to fix patching error due to local parent commits

Nov 28 2022, 10:42 PM · Restricted Project, Restricted Project
nilanjana_basu requested review of D138896: [AArch64] Unit tests for multiple back-to-back zext lowering to tbl for vectors.
Nov 28 2022, 10:39 PM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.

Removed cases where TBL lowering will not be beneficial

Nov 28 2022, 9:18 AM · Restricted Project, Restricted Project

Nov 25 2022

nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Trying to fix rebasing error

Nov 25 2022, 3:55 PM · Restricted Project, Restricted Project
nilanjana_basu updated the summary of D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.
Nov 25 2022, 3:54 PM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Rebasing on parent patch for tests

Nov 25 2022, 3:07 PM · Restricted Project, Restricted Project

Nov 23 2022

nilanjana_basu added a comment to D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Removed (8|16)xi16 to (8|16)xi8 conversion because it wasn't showing benefits in instruction count, & additionally adding more instructions to the header. Updated comments.

Nov 23 2022, 2:45 AM · Restricted Project, Restricted Project
nilanjana_basu retitled D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions from [AArch64] Extending lowering of 'trunc <(8|16) x (i16|i64)> %x to <(8|16) x i8>' to use tbl instructions to [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.
Nov 23 2022, 2:38 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Updated a comment

Nov 23 2022, 2:37 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Removed case for 'trunc <(8|16)xi16> %x to <(8|16)xi8>' since it was adding more instructions to loop header, while not improving loop instruction count

Nov 23 2022, 2:20 AM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Rebasing on commit of test cases prior to application of this patch

Nov 23 2022, 1:31 AM · Restricted Project, Restricted Project

Nov 22 2022

nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Updated comments as mentioned in the reviews. Rebased on tests for this change prior to applying this patch.

Nov 22 2022, 5:00 PM · Restricted Project, Restricted Project

Nov 21 2022

nilanjana_basu published D137993: [AArch64] Unit test for zext lowering for different types of vectors for review.
Nov 21 2022, 4:16 PM · Restricted Project, Restricted Project
nilanjana_basu added a reviewer for D137293: [AArch64] Extra unit tests for trunc lowering of vectors: fhahn.
Nov 21 2022, 4:15 PM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Added comments

Nov 21 2022, 4:10 PM · Restricted Project, Restricted Project
nilanjana_basu added inline comments to D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.
Nov 21 2022, 1:53 PM · Restricted Project, Restricted Project
nilanjana_basu updated the summary of D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.
Nov 21 2022, 1:40 PM · Restricted Project, Restricted Project
nilanjana_basu updated the summary of D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.
Nov 21 2022, 1:40 PM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions.

Minor update to comment

Nov 21 2022, 1:37 PM · Restricted Project, Restricted Project

Nov 19 2022

nilanjana_basu added a comment to D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations.

Yeah, it is probably fine now, but testing with a single value also seems to make the test less interesting. You could keep the random initialization and add a version of truncOrZextVecInLoopWithVW8 that disables vectorization to generate comparison data for testing.

Nov 19 2022, 1:03 AM · Restricted Project
nilanjana_basu updated the diff for D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations.

Reverted to using random inputs & changed correctness test to compare against same operations with no vectorization

Nov 19 2022, 12:53 AM · Restricted Project

Nov 15 2022

nilanjana_basu added a comment to D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations.

Why randomized?

! In D138059#3929074, @nilanjana_basu wrote:

Removed randomization in input & combined correctness tests with performance ones. Explicitly added vectorization width for 16 elements since the related patches target this width.

I think the main reason for initializing with random data is to make the benchmarks more robust so the optimizer won't be able to (partly) optimize out our benchmark code?

Nov 15 2022, 4:33 PM · Restricted Project
nilanjana_basu added a comment to D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations.

Why randomized?

Nov 15 2022, 3:54 PM · Restricted Project
nilanjana_basu updated the diff for D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations.

Removed randomization in input & combined correctness tests with performance ones. Explicitly added vectorization width for 16 elements since the related patches target this width.

Nov 15 2022, 3:50 PM · Restricted Project
nilanjana_basu added reviewers for D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations: fhahn, t.p.northover, paquette.
Nov 15 2022, 1:14 PM · Restricted Project
nilanjana_basu requested review of D138059: [MicroBenchmarks,AArch64] Added correctness test & other performance tests for truncate or zero-extend vector operations.
Nov 15 2022, 1:13 PM · Restricted Project

Nov 8 2022

nilanjana_basu published D136722: [AArch64] Extending lowering of 'zext <Y x i8> %x to <Y x i8X>' to use tbl instructions for review.
Nov 8 2022, 11:01 AM · Restricted Project, Restricted Project

Nov 3 2022

nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Ran clang-format

Nov 3 2022, 7:46 PM · Restricted Project, Restricted Project
nilanjana_basu added inline comments to D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.
Nov 3 2022, 7:39 PM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Addressed comments by t.p.northover - refactored code to remove redundancy

Nov 3 2022, 7:17 PM · Restricted Project, Restricted Project
nilanjana_basu abandoned D137382: [AArch64] Extending trunc lowering from (8|16)x(i64|i16) to (8|16)xi8 using tbl instructions.

Patch already exists. This was posted by mistake.

Nov 3 2022, 7:15 PM · Restricted Project, Restricted Project
nilanjana_basu requested review of D137382: [AArch64] Extending trunc lowering from (8|16)x(i64|i16) to (8|16)xi8 using tbl instructions.
Nov 3 2022, 7:12 PM · Restricted Project, Restricted Project

Nov 2 2022

nilanjana_basu requested review of D137293: [AArch64] Extra unit tests for trunc lowering of vectors.
Nov 2 2022, 2:59 PM · Restricted Project, Restricted Project
nilanjana_basu committed rT3b44b6bdd3e8: [MicroBenchmarks] Add benchmarks to check runtime of truncate or zero-extend… (authored by nilanjana_basu).
[MicroBenchmarks] Add benchmarks to check runtime of truncate or zero-extend…
Nov 2 2022, 2:05 PM · Restricted Project
nilanjana_basu closed D136274: Microbenchmark to test runtime of truncate or zero-extend vector operations in AArch64.
Nov 2 2022, 2:05 PM · Restricted Project
nilanjana_basu updated the diff for D136274: Microbenchmark to test runtime of truncate or zero-extend vector operations in AArch64.

Minor fix to comments

Nov 2 2022, 10:54 AM · Restricted Project

Nov 1 2022

nilanjana_basu updated the diff for D136274: Microbenchmark to test runtime of truncate or zero-extend vector operations in AArch64.

Removed the addition operation to keep only the truncate or zero-extend operation for a more focused performance comparison

Nov 1 2022, 6:42 PM · Restricted Project
nilanjana_basu abandoned D137221: [MicroBenchmarks] Add benchmarks to check runtime of truncate or zero-extend vector operations in AArch64.
Nov 1 2022, 6:37 PM · Restricted Project
nilanjana_basu requested review of D137221: [MicroBenchmarks] Add benchmarks to check runtime of truncate or zero-extend vector operations in AArch64.
Nov 1 2022, 6:31 PM · Restricted Project

Oct 31 2022

nilanjana_basu added a comment to D136274: Microbenchmark to test runtime of truncate or zero-extend vector operations in AArch64.

All the comments have been addressed in the latest patch.

Oct 31 2022, 1:19 PM · Restricted Project
nilanjana_basu updated the summary of D136274: Microbenchmark to test runtime of truncate or zero-extend vector operations in AArch64.
Oct 31 2022, 1:15 PM · Restricted Project
nilanjana_basu updated the summary of D136274: Microbenchmark to test runtime of truncate or zero-extend vector operations in AArch64.
Oct 31 2022, 1:15 PM · Restricted Project
nilanjana_basu updated the diff for D136274: Microbenchmark to test runtime of truncate or zero-extend vector operations in AArch64.

Removed two test cases whose related patches are not yet available.

Oct 31 2022, 1:07 PM · Restricted Project

Oct 28 2022

nilanjana_basu updated the summary of D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.
Oct 28 2022, 4:57 PM · Restricted Project, Restricted Project
nilanjana_basu added a reviewer for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions: paquette.
Oct 28 2022, 4:41 PM · Restricted Project, Restricted Project

Oct 25 2022

nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Ran clang-format since it was failing in the build report at https://buildkite.com/llvm-project/diff-checks/builds/133184

Oct 25 2022, 5:42 PM · Restricted Project, Restricted Project
nilanjana_basu set the repository for D136274: Microbenchmark to test runtime of truncate or zero-extend vector operations in AArch64 to rT test-suite.
Oct 25 2022, 5:12 PM · Restricted Project

Oct 24 2022

nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

The automated build tests failed for the previous patch because it was based on a previous commit for a unit test that isn't submitted yet. This patch fixes it by squashing the previous commit, removing the dependency & showing the final update.

Oct 24 2022, 11:03 AM · Restricted Project, Restricted Project

Oct 22 2022

nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Extended the trunc lowering for other types like 16xi64, 16xi16, 8xi16

Oct 22 2022, 1:19 PM · Restricted Project, Restricted Project

Oct 21 2022

nilanjana_basu updated the diff for D136274: Microbenchmark to test runtime of truncate or zero-extend vector operations in AArch64.

Extended it to be generic enough for both truncate & zero-extend vector operations

Oct 21 2022, 6:36 PM · Restricted Project
nilanjana_basu updated the diff for D136274: Microbenchmark to test runtime of truncate or zero-extend vector operations in AArch64.

Fixed a mistake where the same test was being ran twice

Oct 21 2022, 5:57 PM · Restricted Project

Oct 20 2022

nilanjana_basu updated the diff for D136274: Microbenchmark to test runtime of truncate or zero-extend vector operations in AArch64.

Made a separate file for testing vector operations for truncate or zero extend. Added tests for truncate of different types of data types, with different vectorization width settings.

Oct 20 2022, 7:24 PM · Restricted Project

Oct 19 2022

nilanjana_basu updated the diff for D136274: Microbenchmark to test runtime of truncate or zero-extend vector operations in AArch64.

Removed redundant code

Oct 19 2022, 11:01 AM · Restricted Project
nilanjana_basu requested review of D136274: Microbenchmark to test runtime of truncate or zero-extend vector operations in AArch64.
Oct 19 2022, 10:59 AM · Restricted Project

Oct 4 2022

nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Removed an unused variable warning

Oct 4 2022, 6:09 PM · Restricted Project, Restricted Project
nilanjana_basu added reviewers for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions: fhahn, t.p.northover.
Oct 4 2022, 6:05 PM · Restricted Project, Restricted Project
nilanjana_basu updated the diff for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions.

Ran git-clang-format & made minor change to reduce LoC.

Oct 4 2022, 6:03 PM · Restricted Project, Restricted Project
nilanjana_basu retitled D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions from Extending trunc lowering of 'trunc <8 x i64> %x to <8 x i8>' to use tbl.4 instruction to [AArch64] Extending trunc lowering of 'trunc <8 x i64> %x to <8 x i8>' to use tbl.4 instruction.
Oct 4 2022, 6:01 PM · Restricted Project, Restricted Project
nilanjana_basu set the repository for D135229: [AArch64] Extending lowering of 'trunc <(8|16) x i64> %x to <(8|16) x i8>' to use tbl instructions to rG LLVM Github Monorepo.
Oct 4 2022, 5:47 PM · Restricted Project, Restricted Project