crimson/zns: ensure writes happen at write pointer.

For ZNS SSDs, every write to a segment/zone has to happen at the zone's
write pointer. Any write request to an offset which is not at
write pointer will be failed by the drive.
In ZNSSegment::write() error out if write offset is not same as WP.

Signed-off-by: Aravind Ramesh <aravind.ramesh@wdc.com>

diff --git a/src/crimson/os/seastore/segment_manager/zns.cc b/src/crimson/os/seastore/segment_manager/zns.cc
index 4127d0f729e5d9424107e179b0c801bf218ff790..95c35b727317737610eacb9d451b75d0a9a1de2e 100644 (file)
--- a/src/crimson/os/seastore/segment_manager/zns.cc
+++ b/src/crimson/os/seastore/segment_manager/zns.cc
@@ -590,8 +590,8 @@ Segment::write_ertr::future<> ZNSSegment::write(
   seastore_off_t offset, ceph::bufferlist bl)
 {
   LOG_PREFIX(ZNSSegment::write);
-  if (offset < write_pointer || offset % manager.metadata.block_size != 0) {
-    ERROR("invalid segment write on segment {} to offset {}",
+  if (offset != write_pointer || offset % manager.metadata.block_size != 0) {
+    ERROR("Invalid segment write on segment {} to offset {}",
       id,
       offset);
     return crimson::ct_error::invarg::make();