New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Prepare transpose_embedding_input for VLE #1634

Closed

sryap wants to merge 1 commit into pytorch:main from sryap:export-D43256880

Contributor

sryap commented Mar 9, 2023

Summary:
Prepare transpose_embedding_input for variable length TBE (VLE).

Update the frontend API to accept VLE args
Change linearize_index_kernel to generate new info for pooled
TBE backward. Bag ID (b) and table ID (t) are stored in a 32-bit
info variable.
- We use lower info_B_num_bits bits to store b (b < max_B).
  Supported max_B = 2^info_B_num_bits
- We use upper 32 - info_B_num_bits bits to store t (t <
  T). Supported T = 2^(32 - info_B_num_bits)
- Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VLE cases too for
  easy code maintenace.
Update all backward kernels (including Triton's) to process the new
info

Differential Revision: D43256880

facebook-github-bot added the cla signed label

netlify bot commented Mar 9, 2023 •

edited

✅ Deploy Preview for pytorch-fbgemm-docs canceled.

Name	Link
🔨 Latest commit	`ceecab6`
🔍 Latest deploy log	https://app.netlify.com/sites/pytorch-fbgemm-docs/deploys/64594860d68d9900076aaaf9

facebook-github-bot added the fb-exported label

Contributor

facebook-github-bot commented Mar 9, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

1 similar comment

Contributor

facebook-github-bot commented Mar 14, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VLE (pytorch#1634)

192958e

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable length TBE (VLE).

- Update the frontend API to accept VLE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VLE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Differential Revision: D43256880

fbshipit-source-id: c058763b5b02a7e60dab0d5287a03f013c29dd31

sryap force-pushed the export-D43256880 branch from 8e73be7 to 192958e Compare

March 14, 2023 02:13

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VLE (pytorch#1634)

1c66acc

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable length TBE (VLE).

- Update the frontend API to accept VLE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VLE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Differential Revision: D43256880

fbshipit-source-id: ac0d62f07401273f6ff1bad68a9b63797436a147

sryap force-pushed the export-D43256880 branch from 192958e to 1c66acc Compare

March 14, 2023 02:16

Contributor

facebook-github-bot commented Mar 14, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VLE (pytorch#1634)

b3ac6b3

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable length TBE (VLE).

- Update the frontend API to accept VLE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VLE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: 2fea1ea8509308ba95018eb2b340702568ccea5b

Contributor

facebook-github-bot commented Mar 21, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

sryap force-pushed the export-D43256880 branch from 1c66acc to b3ac6b3 Compare

March 21, 2023 05:11

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VLE (pytorch#1634)

d6ab723

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable length TBE (VLE).

- Update the frontend API to accept VLE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VLE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: b439643214b05b79bb4cbb9ddca7cc07dfd9e3ed

Contributor

facebook-github-bot commented Mar 21, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

sryap force-pushed the export-D43256880 branch from b3ac6b3 to d6ab723 Compare

March 21, 2023 05:15

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VLE (pytorch#1634)

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable length TBE (VLE).

- Update the frontend API to accept VLE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VLE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: 7028199aff979b21b8b2ed159df73bf4c206ef2e

sryap force-pushed the export-D43256880 branch from d6ab723 to 2723637 Compare

March 27, 2023 00:19

Contributor

facebook-github-bot commented Mar 27, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

1 similar comment

Contributor

facebook-github-bot commented Mar 27, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VLE (pytorch#1634)

97ba24e

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable length TBE (VLE).

- Update the frontend API to accept VLE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VLE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: 85f0cc67600cb0c671d729afbe7d4dd882f14762

sryap force-pushed the export-D43256880 branch from 2723637 to 97ba24e Compare

March 27, 2023 00:23

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VLE (pytorch#1634)

9f61d68

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable length TBE (VLE).

- Update the frontend API to accept VLE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VLE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: aba1f1228b7f49c86b541341ef0bd1cc479e46ed

sryap force-pushed the export-D43256880 branch from 97ba24e to 9f61d68 Compare

March 27, 2023 18:40

Contributor

facebook-github-bot commented Mar 27, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VLE (pytorch#1634)

2fbfe6c

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable length TBE (VLE).

- Update the frontend API to accept VLE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VLE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: 27875c98eef34fe7f7f38337e236f4a73398ad7e

sryap force-pushed the export-D43256880 branch from 9f61d68 to 2fbfe6c Compare

March 27, 2023 18:45

Contributor

facebook-github-bot commented Mar 27, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

1 similar comment

Contributor

facebook-github-bot commented Mar 29, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VBE (pytorch#1634)

9f8b84e

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable batch size TBE (VBE).

- Update the frontend API to accept VBE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VBE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: 68eb9cd9f11b087c08a665149a24b8afda6a5270

Contributor

facebook-github-bot commented May 1, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

sryap force-pushed the export-D43256880 branch from 1249f5a to dd69d17 Compare

May 1, 2023 20:37

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VBE (pytorch#1634)

605bc6e

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable batch size TBE (VBE).

- Update the frontend API to accept VBE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VBE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: 91b3cc0211d51154965cb6223fbbde5620ab0661

sryap force-pushed the export-D43256880 branch from dd69d17 to 605bc6e Compare

May 2, 2023 18:11

Contributor

facebook-github-bot commented May 2, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VBE (pytorch#1634)

814c94b

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable batch size TBE (VBE).

- Update the frontend API to accept VBE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VBE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: c445248a3cdb9a507a20cf4462a1566344f1469a

sryap force-pushed the export-D43256880 branch from 605bc6e to 814c94b Compare

May 2, 2023 18:20

Contributor

facebook-github-bot commented May 2, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VBE (pytorch#1634)

fc3236e

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable batch size TBE (VBE).

- Update the frontend API to accept VBE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VBE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: c71e65e43aa0ba195f25e2ed2fa1d45ce1ef74e1

sryap force-pushed the export-D43256880 branch from 814c94b to fc3236e Compare

May 3, 2023 23:36

Contributor

facebook-github-bot commented May 3, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VBE (pytorch#1634)

13854bf

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable batch size TBE (VBE).

- Update the frontend API to accept VBE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VBE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: 48285ebaf48aa72173c651d37ab115f2da6f3290

sryap force-pushed the export-D43256880 branch from fc3236e to 13854bf Compare

May 3, 2023 23:41

Contributor

facebook-github-bot commented May 3, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

1 similar comment

Contributor

facebook-github-bot commented May 5, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VBE (pytorch#1634)

24e9f69

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable batch size TBE (VBE).

- Update the frontend API to accept VBE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VBE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: 4c19e649d5021627386dbccd0727e5de6967a684

sryap force-pushed the export-D43256880 branch from 13854bf to 24e9f69 Compare

May 5, 2023 17:32

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VBE (pytorch#1634)

00938c0

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable batch size TBE (VBE).

- Update the frontend API to accept VBE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VBE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: 3b15b534a18cc8a1953d3c0e8e65bbea364d5b99

sryap force-pushed the export-D43256880 branch from 24e9f69 to 00938c0 Compare

May 5, 2023 17:37

Contributor

facebook-github-bot commented May 5, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

sryap added a commit to sryap/FBGEMM that referenced this pull request


          Prepare transpose_embedding_input for VBE (pytorch#1634)

60d1d83

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable batch size TBE (VBE).

- Update the frontend API to accept VBE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VBE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: 8d29a3181dcc5f864c064c955c1a37fe1580c744

sryap force-pushed the export-D43256880 branch from 00938c0 to 60d1d83 Compare

May 8, 2023 19:00

Contributor

facebook-github-bot commented May 8, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880


          Prepare transpose_embedding_input for VBE (pytorch#1634)

ceecab6

Summary:
Pull Request resolved: pytorch#1634

Prepare `transpose_embedding_input` for variable batch size TBE (VBE).

- Update the frontend API to accept VBE args
- Change `linearize_index_kernel` to generate new `info` for pooled
  TBE backward. Bag ID (`b`) and table ID (`t`) are stored in a 32-bit
  `info` variable.
  - We use lower `info_B_num_bits` bits to store `b` (`b` < `max_B`).
    Supported `max_B` = `2^info_B_num_bits`
  - We use upper `32 - info_B_num_bits` bits to store `t` (`t` <
    `T`).  Supported `T` = `2^(32 - info_B_num_bits)`
  - Although this change is mainly for avoiding binary search in the
  backward pass, the change is applied to the non-VBE cases too for
  easy code maintenace.
- Update all backward kernels (including Triton's) to process the new
  `info`

Reviewed By: jianyuh

Differential Revision: D43256880

fbshipit-source-id: 4ec4ef7e6c922e5b6c13ed735d9264851e79419c

sryap force-pushed the export-D43256880 branch from 60d1d83 to ceecab6 Compare

May 8, 2023 19:07

Contributor

facebook-github-bot commented May 8, 2023

This pull request was exported from Phabricator. Differential Revision: D43256880

facebook-github-bot closed this in

f4c83b4

facebook-github-bot added the Merged label

Contributor

facebook-github-bot commented May 9, 2023

This pull request has been merged in f4c83b4.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment