On 4/17/23 04:27, Luis Chamberlain wrote:
On Fri, Apr 14, 2023 at 03:47:13PM +0200, Hannes Reinecke wrote:@@ -2333,13 +2395,15 @@ int block_read_full_folio(struct folio *folio, get_block_t *get_block) if (IS_ENABLED(CONFIG_FS_VERITY) && IS_VERITY(inode)) limit = inode->i_sb->s_maxbytes; - VM_BUG_ON_FOLIO(folio_test_large(folio), folio); - head = create_folio_buffers(folio, inode, 0); blocksize = head->b_size; bbits = block_size_bits(blocksize); - iblock = (sector_t)folio->index << (PAGE_SHIFT - bbits); + if (WARN_ON(PAGE_SHIFT < bbits)) { + iblock = (sector_t)folio->index >> (bbits - PAGE_SHIFT); + } else { + iblock = (sector_t)folio->index << (PAGE_SHIFT - bbits); + } lblock = (limit+blocksize-1) >> bbits; bh = head; nr = 0;BTW I See this pattern in: fs/mpage.c: do_mpage_readpage() fs/mpage.c: __mpage_writepage() A helper might be in order.
But only _iff_ we decide to stick with buffer_heads and upgrade the buffer_head code to handle multi-page block sizes.
The idea was to move over to iomap, hence I'm not sure into which lengths we should go keeping buffer_heads alive ... Cheers, Hannes -- Dr. Hannes Reinecke Kernel Storage Architect hare@xxxxxxx +49 911 74053 688 SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew Myers, Andrew McDonald, Martje Boudien Moerman
