Christoph Hellwig <hch@xxxxxxxxxxxxx> writes: > On Tue, Jun 06, 2023 at 05:13:49PM +0530, Ritesh Harjani (IBM) wrote: >> Also refactors and adds iof->state handling functions for uptodate >> state. > > What does this mean? It is this part. +static inline bool iomap_iof_is_fully_uptodate(struct folio *folio, + struct iomap_folio *iof) +static inline bool iomap_iof_is_block_uptodate(struct iomap_folio *iof, + unsigned int block) +static void iomap_iof_set_range_uptodate(struct folio *folio, + struct iomap_folio *iof, size_t off, size_t len) +static void iomap_set_range_uptodate(struct folio *folio, size_t off, + size_t len) > And please don't mix renames and other changes in > a single patch. All of this is related to uptodate bitmap handling code. i.e. - Renaming "uptodate" bitmap to "state" bitmap in struct iomap_page. - Renaming "uptodate_lock" to "state_lock" in struct iomap_page - Adding helper routines for uptodate bitmap handling - A small refactoring of iomap_set_range_uptodate() function to drop "iop" as a function argument. And move it's function definition above iomap_iof_alloc() Ok, so would you prefer if this has been split into 3 seperate patches? 1. Renaming of uptodate and uptodate_lock to state and state_Lock. 2. Refactor iomap_set_range_uptodate() function to drop struct iomap_page from it's argument and move it above iomap_iof_alloc() (or iomap_ibs_alloc or iomap_fbs_alloc whichever name we settle with) 3. Add uptodate bitmap helper routines e.g. iomap_iof_is_block_uptodate(), iomap_iof_is_fully_uptodate(). -ritesh
