mirror of
https://github.com/torvalds/linux.git
synced 2024-11-24 13:11:40 +00:00
1122f40072
For now, enqueuing and dequeuing on-demand requests all start from
idx 0, this makes request distribution unfair. In the weighty
concurrent I/O scenario, the request stored in higher idx will starve.
Searching requests cyclically in cachefiles_ondemand_daemon_read,
makes distribution fairer.
Fixes:
|
||
---|---|---|
.. | ||
cache.c | ||
daemon.c | ||
error_inject.c | ||
interface.c | ||
internal.h | ||
io.c | ||
Kconfig | ||
key.c | ||
main.c | ||
Makefile | ||
namei.c | ||
ondemand.c | ||
security.c | ||
volume.c | ||
xattr.c |