Re: [PATCH 0/2] ext4: reduce fast-commit write amplification for scattered writes

[Theodore Tso]

On Thu, Jun 11, 2026 at 01:47:33PM -0500, Daejun Park wrote:
> ext4 fast commit tracks a single coalesced logical range per inode.  When
> an inode is dirtied at several disjoint offsets between two commits (e.g.
> sparse/scattered random writes), that range is widened to span [min, max]
> of all the touched offsets, and ext4_fc_write_inode_data() then re-logs
> every extent inside that span -- including the unmodified ones.  On sparse
> allocation this inflates fast-commit traffic and frequently overflows the
> fast-commit area, forcing a fallback to a full jbd2 commit.
> 
> This series replaces the single range with a small, bounded set of disjoint
> ranges so that only the actually-modified regions are logged, while keeping
> the per-inode memory cost negligible:
> 
> Based on v6.17-rc3.

Hi Daejun,

Unfortunately, this patch series doesn't apply on top of the latest
development branch of ext4.  Li Chen has also been submitting changes
to the fast commit code and your changes are conflicting with his
work.  Could you look at rebasing your patches against the mainline
development of the Linux kernel?

Many thanks,

						- Ted