PR #7348 Fix perf issue in Conv CUDA kernel

Fix perf issue in Conv CUDA kernel #7348

SherlockNoMad merged 6 commits into master from bahuang/conv_fix

Fix perf issue in Conv CUDA kernel

caf652cb

SherlockNoMad added core runtime

SherlockNoMad requested a review from

duli2012 4 years ago

SherlockNoMad requested a review from

weixingzhang 4 years ago

SherlockNoMad requested a review from

RandySheriffH 4 years ago

SherlockNoMad requested a review 4 years ago

hariharans29 commented on 2021-04-15

Read avaiable memory from device

9235444d

SherlockNoMad commented on 2021-04-15

assuming 10% fragmentation

ebc1b735

Merge branch 'master' into bahuang/conv_fix

fc34fc71

fix build

3eb1cf1f

Merge branch 'master' into bahuang/conv_fix

f30aef6e

duli2012 commented on 2021-04-16

duli2012 approved these changes on 2021-04-19

SherlockNoMad merged ce7ff27b into master 4 years ago

SherlockNoMad deleted the bahuang/conv_fix branch 4 years ago

Reviewers

duli2012

hariharans29

vadimkantorov

weixingzhang

RandySheriffH

Assignees

No one assigned

Labels

core runtime

Milestone

No milestone