Skip to content

Add NVTE_BACKWARD_MODE=default|unquant|dequant #2644

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

Open
zianglih wants to merge 45 commits into
base: main
Choose a base branch
from
Open

Add NVTE_BACKWARD_MODE=default|unquant|dequant #2644

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
45 commits
Select commit Hold shift + click to select a range
74d982d
Add NVTE_KEEP_BACKWARD_UNQUANTIZED
zianglih Feb 3, 2026
f04ae52
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Feb 3, 2026
28fadbc
Disable ub and clean up
zianglih Feb 3, 2026
8d7cbbb
Drop fuser changes
zianglih Feb 3, 2026
7eda433
Replace use_quantized_bwd with use_fp8_bwd
zianglih Feb 3, 2026
dbc60c5
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Feb 3, 2026
40d42d4
Ignore keep_backward_unquantized if delayed scaling
zianglih Feb 3, 2026
f87e17b
Refactor ignoring NVTE_KEEP_BACKWARD_UNQUANTIZED when delayed scaling…
zianglih Feb 3, 2026
455a905
Add back missing ctx.debug
zianglih Feb 3, 2026
41415ff
Refactor changes under fused
zianglih Feb 3, 2026
65d44ff
Clean up
zianglih Feb 3, 2026
e3a651c
Refactor high-precision overwrite if keep_backward_unquantized
zianglih Feb 3, 2026
fba242b
Clean up
zianglih Feb 3, 2026
d25fc47
Drop redundant fp8_recipe_bwd
zianglih Feb 4, 2026
4df62fa
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Feb 4, 2026
06e7060
Drop redundant ub changes
zianglih Feb 4, 2026
e26d318
Drop more redundant ub changes
zianglih Feb 4, 2026
986f173
Drop redundant delayed scaling changes
zianglih Feb 4, 2026
5019d3b
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Feb 4, 2026
70ef66c
Drop unneeded backwards_needs_fc1_input
zianglih Feb 4, 2026
88c58fa
Drop and disallow LayerNormMLP implementation
zianglih Feb 4, 2026
a097f3e
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Feb 4, 2026
8414358
Move interface changes to recipe
zianglih Feb 5, 2026
eecfcf8
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Feb 5, 2026
bfb840c
Move ub overrides to fwd
zianglih Feb 5, 2026
23f14ea
Remove duplication
zianglih Feb 5, 2026
ae25dee
Simplify use_fp8_bwd logic in bwd
zianglih Feb 5, 2026
491dd44
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Feb 5, 2026
764ee6f
Set grad quantizers to none if keep bwd unquantized
zianglih Feb 5, 2026
bbaa6c5
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Feb 5, 2026
5e638e0
Drop delayed scaling change
zianglih Feb 6, 2026
75079a4
Simplify env var logic
zianglih Feb 9, 2026
0874804
Move validation check to recipe
zianglih Feb 9, 2026
1347578
Simplify effective_enabled
zianglih Feb 9, 2026
2e81568
Fix inverted assertion logic
zianglih Feb 9, 2026
cf04de6
Simplify changes under ops
zianglih Feb 9, 2026
edcb1f6
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Feb 9, 2026
891fc7b
Simplify ctx.keep_backward_unquantized
zianglih Feb 9, 2026
5d20f77
Fix missing attribute
zianglih Feb 9, 2026
bbd22c7
Add unit tests
zianglih Feb 10, 2026
7d74aa3
Fix bias errors in unit test
zianglih Feb 10, 2026
5bc3a57
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Feb 10, 2026
255589e
Add more shapes to unit test
zianglih Feb 10, 2026
c915bc3
Refator interface to `NVTE_BACKWARD_MODE=default|unquant|dequant`
zianglih Feb 24, 2026
0dee809
Fix override and clean up
zianglih Feb 25, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions qa/L0_pytorch_unittest/test.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,6 +40,7 @@ python3 -m pytest --tb=auto --junitxml=$XML_LOG_DIR/pytest_test_gqa.xml $TE_PATH
python3 -m pytest --tb=auto --junitxml=$XML_LOG_DIR/pytest_test_fused_optimizer.xml $TE_PATH/tests/pytorch/test_fused_optimizer.py || test_fail "test_fused_optimizer.py"
python3 -m pytest --tb=auto --junitxml=$XML_LOG_DIR/pytest_test_multi_tensor.xml $TE_PATH/tests/pytorch/test_multi_tensor.py || test_fail "test_multi_tensor.py"
python3 -m pytest --tb=auto --junitxml=$XML_LOG_DIR/pytest_test_fusible_ops.xml $TE_PATH/tests/pytorch/test_fusible_ops.py || test_fail "test_fusible_ops.py"
python3 -m pytest --tb=auto --junitxml=$XML_LOG_DIR/pytest_test_backward_mode.xml $TE_PATH/tests/pytorch/test_backward_mode.py || test_fail "test_backward_mode.py"
python3 -m pytest --tb=auto --junitxml=$XML_LOG_DIR/pytest_test_permutation.xml $TE_PATH/tests/pytorch/test_permutation.py || test_fail "test_permutation.py"
python3 -m pytest --tb=auto --junitxml=$XML_LOG_DIR/pytest_test_parallel_cross_entropy.xml $TE_PATH/tests/pytorch/test_parallel_cross_entropy.py || test_fail "test_parallel_cross_entropy.py"
python3 -m pytest --tb=auto --junitxml=$XML_LOG_DIR/pytest_test_cpu_offloading.xml $TE_PATH/tests/pytorch/test_cpu_offloading.py || test_fail "test_cpu_offloading.py"
Expand Down
Loading