forked from Minki/linux
f42e8e5088
The recovery write thread started out as a normal pwrite thread and when the filesystem was told about potential media error in the range, filesystem turns the normal pwrite to a dax_recovery_write. The recovery write consists of clearing media poison, clearing page HWPoison bit, reenable page-wide read-write permission, flush the caches and finally write. A competing pread thread will be held off during the recovery process since data read back might not be valid, and this is achieved by clearing the badblock records after the recovery write is complete. Competing recovery write threads are already serialized by writer lock held by dax_iomap_rw(). Signed-off-by: Jane Chu <jane.chu@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Link: https://lore.kernel.org/r/165247997655.53156.8381418704988035976.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com> |
||
---|---|---|
.. | ||
badrange.c | ||
btt_devs.c | ||
btt.c | ||
btt.h | ||
bus.c | ||
claim.c | ||
core.c | ||
dax_devs.c | ||
dimm_devs.c | ||
dimm.c | ||
e820.c | ||
Kconfig | ||
label.c | ||
label.h | ||
Makefile | ||
namespace_devs.c | ||
nd_perf.c | ||
nd_virtio.c | ||
nd-core.h | ||
nd.h | ||
of_pmem.c | ||
pfn_devs.c | ||
pfn.h | ||
pmem.c | ||
pmem.h | ||
region_devs.c | ||
region.c | ||
security.c | ||
virtio_pmem.c | ||
virtio_pmem.h |