nksingh85@xxxxxxxxx writes: > From: Neeraj Singh <neerajsi@xxxxxxxxxxxxx> > > GGG closed this series erroneously, so I'm trying out git-send-email. Apologies for any mistakes. > > This series is also available at https://github.com/neerajsi-msft/git/git.git ns/batched-fsync-v6. > > V6 changes: > > * Based on master at faa21c1 to pick up ns/fsync-or-die-message-fix. Also resolved a conflict with 8aa0209 in t/perf/p7519-fsmonitor.sh. > > * Some independent patches were submitted separately on-list. This series is now dependent on ns/fsync-or-die-message-fix. > > * Rename bulk_checkin_state to bulk_checkin_packfile to discourage future authors from adding any non-packfile related stuff to it. Each individual component of bulk_checkin should have its own state variable(s) going forward, and they should only be tied together by odb_transaction_nesting. > > * Rename finish_bulk_checkin and do_batch_fsync to flush_bulk_checkin and flush_batch_fsync. The "finish" step is going to be the end_odb_transaction. The "flush" terminology should be consistently used for making changes visible. > > * Add flush_odb_transaction and use it in update-index before printing verbose output to mitigate risk of missing objects for a tricky stdin feeder. > > * Re-add shell "local with assignment": now these are all on a separate line with quotes around any values, to comply with dash. I'm running on ubuntu 20.04 LTS where I saw some of the dash issues before. Thanks. These looked all sensible. Wil queue.
