DeepSpeed
[zero_to_fp32] adapt to 4-bytes alignment in z2
#1372
Merged

[zero_to_fp32] adapt to 4-bytes alignment in z2 #1372

stas00
stas00 [zero_to_fp32] adapt to 4-bytes alignment in z2
d8077876
stas00 stas00 requested a review from awan-10 awan-10 4 years ago
stas00 stas00 requested a review from cli99 cli99 4 years ago
stas00 stas00 requested a review from conglongli conglongli 4 years ago
stas00 stas00 requested a review from eltonzheng eltonzheng 4 years ago
stas00 stas00 requested a review from jeffra jeffra 4 years ago
stas00 stas00 requested a review from minjiaz minjiaz 4 years ago
stas00 stas00 requested a review from niumanar niumanar 4 years ago
stas00 stas00 requested a review from RezaYazdaniAminabadi RezaYazdaniAminabadi 4 years ago
stas00 stas00 requested a review from samyam samyam 4 years ago
stas00 stas00 requested a review from ShadenSmith ShadenSmith 4 years ago
stas00 stas00 requested a review from tjruwase tjruwase 4 years ago
stas00 align both sides
a0080589
stas00 cleanup
a3bc8d98
stas00 adjust the existing test to reproduce the bug
08b41a8d
stas00 test only on one gpu
a38e3ae8
stas00 handle the edge case of param with 1 element
9bee134f
tjruwase Merge branch 'master' into zero_to_fp32_z2_align4
aa049500
tjruwase Merge branch 'master' into zero_to_fp32_z2_align4
3dc0083c
stas00 remove the buffer
592a7d79
stas00 Merge branch 'zero_to_fp32_z2_align4' of github.com:stas00/DeepSpeed …
7ca1c7f8
tjruwase
tjruwase approved these changes on 2021-09-16
jeffra
jeffra approved these changes on 2021-09-16
jeffra jeffra merged 30537e71 into master 4 years ago
stas00 stas00 deleted the zero_to_fp32_z2_align4 branch 4 years ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone