lightnvm: pblk: garbage collect lines with failed writes
authorHans Holmberg <hans.holmberg@cnexlabs.com>
Fri, 1 Jun 2018 14:41:06 +0000 (16:41 +0200)
committerJens Axboe <axboe@kernel.dk>
Fri, 1 Jun 2018 15:02:53 +0000 (09:02 -0600)
commit48b8d20895f8a489e1527e9bdc5e372808542fa3
tree754ad9c349589759af22adad8286e7fa2b7095bc
parent6a3abf5beef6ae46381c1fb6976e6f313c40f0c1
lightnvm: pblk: garbage collect lines with failed writes

Write failures should not happen under normal circumstances,
so in order to bring the chunk back into a known state as soon
as possible, evacuate all the valid data out of the line and let the
fw judge if the block can be written to in the next reset cycle.

Do this by introducing a new gc list for lines with failed writes,
and ensure that the rate limiter allocates a small portion of
the write bandwidth to get the job done.

The lba list is saved in memory for use during gc as we
cannot gurantee that the emeta data is readable if a write
error occurred.

Signed-off-by: Hans Holmberg <hans.holmberg@cnexlabs.com>
Reviewed-by: Javier González <javier@cnexlabs.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
drivers/lightnvm/pblk-core.c
drivers/lightnvm/pblk-gc.c
drivers/lightnvm/pblk-init.c
drivers/lightnvm/pblk-rl.c
drivers/lightnvm/pblk-sysfs.c
drivers/lightnvm/pblk-write.c
drivers/lightnvm/pblk.h