You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
After commit 10b43f1 ("net/rds: Use DMA memory pool allocation
for rds_header"), the time to free the RDS headers at connection tear
down becomes longer. At connection fail over time, hundreds of
connections are shut down sequentially. A little extra time spent for
each connection shutdown adds up. To reduce this fail over time, the
headers free work is offloaded to a workqueue.
Orabug: 30434704
Fixes: 10b43f1 ("net/rds: Use DMA memory pool allocation for
rds_header")
Signed-off-by: Ka-Cheong Poon <[email protected]>
Reviewed-by: Håkon Bugge <[email protected]>
Tested-by: Håkon Bugge <[email protected]>
Signed-off-by: Somasundaram Krishnasamy <[email protected]>
0 commit comments