| Message ID | 170404997970.1797094.13056398021609108212.stgit@frogsfrogsfrogs (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | [1/1] xfs: map xfile pages directly into xfs_buf | expand |
On Sun, Dec 31, 2023 at 02:35:23PM -0800, Darrick J. Wong wrote: > From: Darrick J. Wong <djwong@kernel.org> > > Map the xfile pages directly into xfs_buf to reduce memory overhead. > It's silly to use memory to stage changes to shmem pages for ephemeral > btrees that don't care about transactionality. Looking at the users if seems like this is in fact the online use case - PAGE_SIZE sized btree blocks, which by nature of coming from shmem must be aligned. So I'd suggest we remove the non-mapped file path and just always use this one with sufficient sanity checks to trip over early if that assumption doesn't hold true. The users also alway just has a single map per buffer, and a single page per buffer so it really shouldn't support anything else. Writing it directly to shmemfs will probably simply things as well as it just needs to do a shmem_read_mapping_page_gfp to read the one page per buffer, and then a set_page_dirty on the locked page when releasing it. Talking about relasing it: this now gets us back to the old pagebuf problem of competing LRUs, one for the xfs_buf, and then anothr for the shmemfs page in the page cache. I suspect not putting the shmemfs backed buffers onto the LRU at all might be a good thing, but that requires careful benchmarking.
This comment should have been for the kernel version and not the userspace artefact patch.
diff --git a/libxfs/xfs_btree_mem.h b/libxfs/xfs_btree_mem.h index 1f961f3f554..cfb30cb1aab 100644 --- a/libxfs/xfs_btree_mem.h +++ b/libxfs/xfs_btree_mem.h @@ -17,8 +17,14 @@ struct xfbtree_config { /* Owner of this btree. */ unsigned long long owner; + + /* XFBTREE_* flags */ + unsigned int flags; }; +/* buffers should be directly mapped from memory */ +#define XFBTREE_DIRECT_MAP (1U << 0) + #ifdef CONFIG_XFS_BTREE_IN_XFILE unsigned int xfs_btree_mem_head_nlevels(struct xfs_buf *head_bp); diff --git a/libxfs/xfs_rmap_btree.c b/libxfs/xfs_rmap_btree.c index a378bd5daf8..7342623ed5e 100644 --- a/libxfs/xfs_rmap_btree.c +++ b/libxfs/xfs_rmap_btree.c @@ -670,6 +670,7 @@ xfs_rmapbt_mem_create( .btree_ops = &xfs_rmapbt_mem_ops, .target = target, .owner = agno, + .flags = XFBTREE_DIRECT_MAP, }; return xfbtree_create(mp, &cfg, xfbtreep);