Message ID | 20210617082617.971602-6-david@fromorbit.com (mailing list archive) |
---|---|
State | Superseded |
Headers | show |
Series | xfs: log fixes for for-next | expand |
On Thu, Jun 17, 2021 at 06:26:14PM +1000, Dave Chinner wrote: > From: Dave Chinner <dchinner@redhat.com> > > So we can use it for start record ordering as well as commit record > ordering in future. > > Signed-off-by: Dave Chinner <dchinner@redhat.com> This tricked me for a second until I realized that xlog_cil_order_write is the chunk of code just prior to the xlog_cil_write_commit_record call. Reviewed-by: Darrick J. Wong <djwong@kernel.org> --D > --- > fs/xfs/xfs_log_cil.c | 89 ++++++++++++++++++++++++++------------------ > 1 file changed, 52 insertions(+), 37 deletions(-) > > diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c > index 35fc3e57d870..f993ec69fc97 100644 > --- a/fs/xfs/xfs_log_cil.c > +++ b/fs/xfs/xfs_log_cil.c > @@ -784,9 +784,54 @@ xlog_cil_build_trans_hdr( > } > > /* > - * Write out the commit record of a checkpoint transaction associated with the > - * given ticket to close off a running log write. Return the lsn of the commit > - * record. > + * Ensure that the order of log writes follows checkpoint sequence order. This > + * relies on the context LSN being zero until the log write has guaranteed the > + * LSN that the log write will start at via xlog_state_get_iclog_space(). > + */ > +static int > +xlog_cil_order_write( > + struct xfs_cil *cil, > + xfs_csn_t sequence) > +{ > + struct xfs_cil_ctx *ctx; > + > +restart: > + spin_lock(&cil->xc_push_lock); > + list_for_each_entry(ctx, &cil->xc_committing, committing) { > + /* > + * Avoid getting stuck in this loop because we were woken by the > + * shutdown, but then went back to sleep once already in the > + * shutdown state. > + */ > + if (XLOG_FORCED_SHUTDOWN(cil->xc_log)) { > + spin_unlock(&cil->xc_push_lock); > + return -EIO; > + } > + > + /* > + * Higher sequences will wait for this one so skip them. > + * Don't wait for our own sequence, either. > + */ > + if (ctx->sequence >= sequence) > + continue; > + if (!ctx->commit_lsn) { > + /* > + * It is still being pushed! Wait for the push to > + * complete, then start again from the beginning. > + */ > + xlog_wait(&cil->xc_commit_wait, &cil->xc_push_lock); > + goto restart; > + } > + } > + spin_unlock(&cil->xc_push_lock); > + return 0; > +} > + > +/* > + * Write out the commit record of a checkpoint transaction to close off a > + * running log write. These commit records are strictly ordered in ascending CIL > + * sequence order so that log recovery will always replay the checkpoints in the > + * correct order. > */ > int > xlog_cil_write_commit_record( > @@ -816,6 +861,10 @@ xlog_cil_write_commit_record( > if (XLOG_FORCED_SHUTDOWN(log)) > return -EIO; > > + error = xlog_cil_order_write(ctx->cil, ctx->sequence); > + if (error) > + return error; > + > /* account for space used by record data */ > ctx->ticket->t_curr_res -= reg.i_len; > error = xlog_write(log, ctx, &lv_chain, ctx->ticket, iclog, reg.i_len); > @@ -1048,40 +1097,6 @@ xlog_cil_push_work( > if (error) > goto out_abort_free_ticket; > > - /* > - * now that we've written the checkpoint into the log, strictly > - * order the commit records so replay will get them in the right order. > - */ > -restart: > - spin_lock(&cil->xc_push_lock); > - list_for_each_entry(new_ctx, &cil->xc_committing, committing) { > - /* > - * Avoid getting stuck in this loop because we were woken by the > - * shutdown, but then went back to sleep once already in the > - * shutdown state. > - */ > - if (XLOG_FORCED_SHUTDOWN(log)) { > - spin_unlock(&cil->xc_push_lock); > - goto out_abort_free_ticket; > - } > - > - /* > - * Higher sequences will wait for this one so skip them. > - * Don't wait for our own sequence, either. > - */ > - if (new_ctx->sequence >= ctx->sequence) > - continue; > - if (!new_ctx->commit_lsn) { > - /* > - * It is still being pushed! Wait for the push to > - * complete, then start again from the beginning. > - */ > - xlog_wait(&cil->xc_commit_wait, &cil->xc_push_lock); > - goto restart; > - } > - } > - spin_unlock(&cil->xc_push_lock); > - > error = xlog_cil_write_commit_record(ctx, &commit_iclog); > if (error) > goto out_abort_free_ticket; > -- > 2.31.1 >
On Thu, Jun 17, 2021 at 12:59:04PM -0700, Darrick J. Wong wrote: > On Thu, Jun 17, 2021 at 06:26:14PM +1000, Dave Chinner wrote: > > From: Dave Chinner <dchinner@redhat.com> > > > > So we can use it for start record ordering as well as commit record > > ordering in future. > > > > Signed-off-by: Dave Chinner <dchinner@redhat.com> > > This tricked me for a second until I realized that xlog_cil_order_write > is the chunk of code just prior to the xlog_cil_write_commit_record > call. Yeah, moving the caller at the same time as the factoring is a trick test for every reader. I think this needs to be documented in the commit log. Or even better moved to a separate log, but it seems you get shot for that kind of suggestion on the xfs list these days..
On Fri, Jun 18, 2021 at 03:27:49PM +0100, Christoph Hellwig wrote: > On Thu, Jun 17, 2021 at 12:59:04PM -0700, Darrick J. Wong wrote: > > On Thu, Jun 17, 2021 at 06:26:14PM +1000, Dave Chinner wrote: > > > From: Dave Chinner <dchinner@redhat.com> > > > > > > So we can use it for start record ordering as well as commit record > > > ordering in future. > > > > > > Signed-off-by: Dave Chinner <dchinner@redhat.com> > > > > This tricked me for a second until I realized that xlog_cil_order_write > > is the chunk of code just prior to the xlog_cil_write_commit_record > > call. > > Yeah, moving the caller at the same time as the factoring is a trick > test for every reader. I think this needs to be documented in the > commit log. Or even better moved to a separate log, but it seems you > get shot for that kind of suggestion on the xfs list these days.. Sorry, what? This should be a straight factoring - the place we do the ordering check must not change because that'll break shit. Ngggh. Yeah, thanks git. When I rebased the patch, it's merged the hunk into the wrong place. It gets fixed up later when I move the ordering inside the xlog_cil_write_commit_record() function, but this patch by itself was silently broken by the tooling. -Dave.
diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c index 35fc3e57d870..f993ec69fc97 100644 --- a/fs/xfs/xfs_log_cil.c +++ b/fs/xfs/xfs_log_cil.c @@ -784,9 +784,54 @@ xlog_cil_build_trans_hdr( } /* - * Write out the commit record of a checkpoint transaction associated with the - * given ticket to close off a running log write. Return the lsn of the commit - * record. + * Ensure that the order of log writes follows checkpoint sequence order. This + * relies on the context LSN being zero until the log write has guaranteed the + * LSN that the log write will start at via xlog_state_get_iclog_space(). + */ +static int +xlog_cil_order_write( + struct xfs_cil *cil, + xfs_csn_t sequence) +{ + struct xfs_cil_ctx *ctx; + +restart: + spin_lock(&cil->xc_push_lock); + list_for_each_entry(ctx, &cil->xc_committing, committing) { + /* + * Avoid getting stuck in this loop because we were woken by the + * shutdown, but then went back to sleep once already in the + * shutdown state. + */ + if (XLOG_FORCED_SHUTDOWN(cil->xc_log)) { + spin_unlock(&cil->xc_push_lock); + return -EIO; + } + + /* + * Higher sequences will wait for this one so skip them. + * Don't wait for our own sequence, either. + */ + if (ctx->sequence >= sequence) + continue; + if (!ctx->commit_lsn) { + /* + * It is still being pushed! Wait for the push to + * complete, then start again from the beginning. + */ + xlog_wait(&cil->xc_commit_wait, &cil->xc_push_lock); + goto restart; + } + } + spin_unlock(&cil->xc_push_lock); + return 0; +} + +/* + * Write out the commit record of a checkpoint transaction to close off a + * running log write. These commit records are strictly ordered in ascending CIL + * sequence order so that log recovery will always replay the checkpoints in the + * correct order. */ int xlog_cil_write_commit_record( @@ -816,6 +861,10 @@ xlog_cil_write_commit_record( if (XLOG_FORCED_SHUTDOWN(log)) return -EIO; + error = xlog_cil_order_write(ctx->cil, ctx->sequence); + if (error) + return error; + /* account for space used by record data */ ctx->ticket->t_curr_res -= reg.i_len; error = xlog_write(log, ctx, &lv_chain, ctx->ticket, iclog, reg.i_len); @@ -1048,40 +1097,6 @@ xlog_cil_push_work( if (error) goto out_abort_free_ticket; - /* - * now that we've written the checkpoint into the log, strictly - * order the commit records so replay will get them in the right order. - */ -restart: - spin_lock(&cil->xc_push_lock); - list_for_each_entry(new_ctx, &cil->xc_committing, committing) { - /* - * Avoid getting stuck in this loop because we were woken by the - * shutdown, but then went back to sleep once already in the - * shutdown state. - */ - if (XLOG_FORCED_SHUTDOWN(log)) { - spin_unlock(&cil->xc_push_lock); - goto out_abort_free_ticket; - } - - /* - * Higher sequences will wait for this one so skip them. - * Don't wait for our own sequence, either. - */ - if (new_ctx->sequence >= ctx->sequence) - continue; - if (!new_ctx->commit_lsn) { - /* - * It is still being pushed! Wait for the push to - * complete, then start again from the beginning. - */ - xlog_wait(&cil->xc_commit_wait, &cil->xc_push_lock); - goto restart; - } - } - spin_unlock(&cil->xc_push_lock); - error = xlog_cil_write_commit_record(ctx, &commit_iclog); if (error) goto out_abort_free_ticket;