On Fri, May 21, 2021 at 09:35:47AM +0200, Christoph Hellwig wrote: > On Fri, May 21, 2021 at 03:31:05PM +0800, Ming Lei wrote: > > > iomap_ioend_bioset is sized to make sure we can always complete up > > > to 4 pages, and the list is only used inside a page, so we're fine. > > > > The number itself does not matter, because there isn't any limit on how > > many ioends can be allocated before submitting, for example, it can be > > observed that 64 ioends is allocated before submitting when writing > > 5GB file to ext4. So far the reserved pool size is 32. > > How do you manage to allocate iomap ioends when writing to ext4? ext4 > doesn't use iomap for buffered I/O. Just double check, the multiple ioends allocation is from root XFS and not from big file write to ext4, so looks it can be triggered easily in background writeback. > > > > fs_bio_set always has two entries to allow exactly for the common > > > chain and submit pattern. > > > > It is easy to trigger dozens of chained bios in one ioend when writing > > big file to XFS. > > Yes, we can still have one chained bio per ioend, so we need a bioset > with the same size as iomap_ioend_bioset. That still should not be > dozends for a common setup, though. Yeah, that can be one solution. Just wondering why the ioend isn't submitted out after it becomes full? Thanks, Ming
