Skip to content
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.

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

Distributed SaveLoad implementation for semi-auto strategy #59659

Merged
merged 45 commits into from
Dec 7, 2023
Merged
Changes from 1 commit
Commits
Show all changes
45 commits
Select commit Hold shift + click to select a range
fc3b3c0
exclude xpu
pangengzheng Jun 25, 2023
e291552
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Jul 24, 2023
7a13c0b
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Aug 25, 2023
d81f305
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Sep 5, 2023
cd6e4fb
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Sep 5, 2023
9d27f27
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Sep 6, 2023
5037694
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Sep 13, 2023
ef695ee
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Sep 13, 2023
23aa6ff
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Sep 25, 2023
f7615b7
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Oct 16, 2023
6605dff
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Oct 19, 2023
767835d
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Oct 30, 2023
f756bc6
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Nov 8, 2023
2ffd709
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Nov 21, 2023
738f5d5
demo of running dygraph distributed save load
pangengzheng Nov 23, 2023
f3d4bb2
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Nov 24, 2023
7134583
support save cross mesh state_dict
pangengzheng Nov 24, 2023
9e2094a
polish
pangengzheng Nov 24, 2023
786a318
fix compute overlap bug
pangengzheng Nov 28, 2023
ef4f374
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Nov 28, 2023
058d5fe
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Nov 29, 2023
8f64e81
test save load in dp_mp unittest
pangengzheng Nov 29, 2023
250b1b7
fix get local file bug and test
pangengzheng Dec 1, 2023
bd9348f
delete useless files, and rename var
pangengzheng Dec 1, 2023
ecee68b
polish
pangengzheng Dec 1, 2023
a8491b9
format codes
pangengzheng Dec 1, 2023
867726d
merge develop
pangengzheng Dec 1, 2023
2bf30c5
test use_dist
pangengzheng Dec 4, 2023
b46042c
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Dec 4, 2023
160552c
fix test
pangengzheng Dec 4, 2023
c5394c5
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Dec 4, 2023
baf2b74
info to debug
pangengzheng Dec 4, 2023
968d611
fix test
pangengzheng Dec 5, 2023
170fd81
fix
pangengzheng Dec 5, 2023
e0d0690
fix coverage ci
pangengzheng Dec 5, 2023
18298b9
fix docstring codes
pangengzheng Dec 5, 2023
13b1d07
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Dec 5, 2023
1dcd0a7
rename and codestyle
pangengzheng Dec 6, 2023
00df8ba
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into…
pangengzheng Dec 6, 2023
c728400
get rid of use_dist argument
pangengzheng Dec 6, 2023
a3125c0
fix copyright
pangengzheng Dec 6, 2023
0543d1f
polish doc
pangengzheng Dec 6, 2023
e4c72cd
polish
pangengzheng Dec 6, 2023
0561180
polish
pangengzheng Dec 6, 2023
4df7f76
use tmp file path
pangengzheng Dec 6, 2023
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
Prev Previous commit
Next Next commit
fix coverage ci
pangengzheng committed Dec 5, 2023
commit e0d0690810713973261238592105b798c53075b6
6 changes: 3 additions & 3 deletions paddle/scripts/paddle_build.sh
Original file line number Diff line number Diff line change
@@ -814,7 +814,7 @@ set -x
fi
if [ -a "$PADDLE_ROOT/added_ut" ];then
added_uts=^$(awk BEGIN{RS=EOF}'{gsub(/\n/,"$|^");print}' $PADDLE_ROOT/added_ut)$
ctest -R "(${added_uts})" -LE "RUN_TYPE=DIST|RUN_TYPE=EXCLUSIVE" --output-on-failure --repeat-until-fail 3 --timeout 15;added_ut_error=$?
ctest -R "(${added_uts})" -LE "RUN_TYPE=DIST|RUN_TYPE=EXCLUSIVE|RUN_TYPE=HYBRID" --output-on-failure --repeat-until-fail 3 --timeout 15;added_ut_error=$?
ctest -R "(${added_uts})" -L "RUN_TYPE=DIST|RUN_TYPE=EXCLUSIVE" --output-on-failure --repeat-until-fail 3 --timeout 15;added_ut_error_1=$?
if [ "$added_ut_error" != 0 ] && [ "$added_ut_error_1" != 0 ];then
echo "========================================"
@@ -1545,7 +1545,7 @@ set -x
fi
if [ -a "$PADDLE_ROOT/added_ut" ];then
added_uts=^$(awk BEGIN{RS=EOF}'{gsub(/\n/,"$|^");print}' $PADDLE_ROOT/added_ut)$
env CUDA_VISIBLE_DEVICES=0 ctest -R "(${added_uts})" -LE "RUN_TYPE=DIST|RUN_TYPE=EXCLUSIVE" --output-on-failure --repeat-until-fail 3 --timeout 15;added_ut_error=$?
env CUDA_VISIBLE_DEVICES=0 ctest -R "(${added_uts})" -LE "RUN_TYPE=DIST|RUN_TYPE=EXCLUSIVE|RUN_TYPE=HYBRID" --output-on-failure --repeat-until-fail 3 --timeout 15;added_ut_error=$?
ctest -R "(${added_uts})" -L "RUN_TYPE=DIST|RUN_TYPE=EXCLUSIVE" --output-on-failure --repeat-until-fail 3 --timeout 15;added_ut_error_1=$?
if [ "$added_ut_error" != 0 ] && [ "$added_ut_error_1" != 0 ];then
echo "========================================"
@@ -2544,7 +2544,7 @@ set -x
fi
if [ -a "$PADDLE_ROOT/added_ut" ];then
added_uts=^$(awk BEGIN{RS=EOF}'{gsub(/\n/,"$|^");print}' $PADDLE_ROOT/added_ut)$
env CUDA_VISIBLE_DEVICES=0 ctest -R "(${added_uts})" -LE "RUN_TYPE=DIST|RUN_TYPE=EXCLUSIVE" --output-on-failure --repeat-until-fail 3 --timeout 15;added_ut_error=$?
env CUDA_VISIBLE_DEVICES=0 ctest -R "(${added_uts})" -LE "RUN_TYPE=DIST|RUN_TYPE=EXCLUSIVE|RUN_TYPE=HYBRID" --output-on-failure --repeat-until-fail 3 --timeout 15;added_ut_error=$?
ctest -R "(${added_uts})" -L "RUN_TYPE=DIST|RUN_TYPE=EXCLUSIVE" --output-on-failure --repeat-until-fail 3 --timeout 15;added_ut_error_1=$?
if [ "$added_ut_error" != 0 ] && [ "$added_ut_error_1" != 0 ];then
echo "========================================"
4 changes: 2 additions & 2 deletions tools/check_file_diff_approvals.sh
Original file line number Diff line number Diff line change
@@ -553,13 +553,13 @@ RUNTYPE_FILE_CHANGED=`git diff --name-only --diff-filter=AM upstream/$BRANCH|gre
if [ "${RUNTYPE_FILE_CHANGED}" != "" ] && [ "${GIT_PR_ID}" != "" ]; then
for CMAKELISTS_FILE in ${RUNTYPE_FILE_CHANGED};
do
RUNTYPE_ADD=`git diff -U0 upstream/$BRANCH ${PADDLE_ROOT}/${CMAKELISTS_FILE} |grep "^+" |grep -E "SERIAL|RUN_TYPE=EXCLUSIVE|RUN_TYPE=DIST|RUN_TYPE=NIGHTLY|RUN_TYPE=EXCLUSIVE:NIGHTLY|RUN_TYPE=DIST:NIGHTLY|PROPERTIES[[:space:]]+TIMEOUT" || true`
RUNTYPE_ADD=`git diff -U0 upstream/$BRANCH ${PADDLE_ROOT}/${CMAKELISTS_FILE} |grep "^+" |grep -E "SERIAL|RUN_TYPE=EXCLUSIVE|RUN_TYPE=DIST|RUN_TYPE=HYBRID|RUN_TYPE=NIGHTLY|RUN_TYPE=EXCLUSIVE:NIGHTLY|RUN_TYPE=DIST:NIGHTLY|PROPERTIES[[:space:]]+TIMEOUT" || true`
if [[ ${RUNTYPE_ADD} != "" ]];then
RUNTYPE_ADD_LINES="${RUNTYPE_ADD_LINES}\n${CMAKELISTS_FILE}\n${RUNTYPE_ADD}\n"
fi
done
if [[ ${RUNTYPE_ADD_LINES} != "" ]];then
echo_line="You must have one QA (XieYunshen(Recommend) or chalsliu) approval for setting parameter RUN_TYPE as EXCLUSIVE, DIST, NIGHTLY, EXCLUSIVE:NIGHTLY or DISTNIGHTLY, or setting parameter SERIAL, or setting TIMEOUT properties.\nThe corresponding lines are as follows:\n${RUNTYPE_ADD_LINES}\nFor more information, please refer to:https://github.com/PaddlePaddle/Paddle/wiki/PaddlePaddle-Unit-test-specification"
echo_line="You must have one QA (XieYunshen(Recommend) or chalsliu) approval for setting parameter RUN_TYPE as EXCLUSIVE, DIST, HYBRID, NIGHTLY, EXCLUSIVE:NIGHTLY or DISTNIGHTLY, or setting parameter SERIAL, or setting TIMEOUT properties.\nThe corresponding lines are as follows:\n${RUNTYPE_ADD_LINES}\nFor more information, please refer to:https://github.com/PaddlePaddle/Paddle/wiki/PaddlePaddle-Unit-test-specification"
check_approval 1 XieYunshen chalsliu
fi
fi