fuse: allow server to increase max_readahead via FUSE_INIT reply
Subsystems (2)
PATCH 1/1 fuse: allow server to increase max_readahead via FUSE_INIT reply
A FUSE server that advertises a large max_pages and max_write (e.g. max_pages=256, max_write=1MB) cannot currently obtain matching FUSE_READ request sizes from the kernel. Buffered sequential writes arrive at the server at the negotiated max_write size, but buffered sequential reads remain capped at the kernel's default readahead window (VM_READAHEAD_PAGES, 128KB; doubled to 256KB for files marked POSIX_FADV_SEQUENTIAL). A 1MB application read() therefore turns into four sequential 256KB FUSE_READ round-trips instead of one.
This is because process_init_reply() processes the server's max_readahead response as:
ra_pages = arg->max_readahead / PAGE_SIZE; fm->sb->s_bdi->ra_pages = min(fm->sb->s_bdi->ra_pages, ra_pages);
Since the kernel sends its current bdi->ra_pages as init_in->max_readahead, and bdi->ra_pages is the default VM_READAHEAD_PAGES at this point, the server can only ever decrease the readahead window -- never increase it. Even if the server replies with max_readahead=1MB, the min() clamps it back to 128KB.
This clamp dates to commit 9cd684551124 ("[PATCH] fuse: fix async read for legacy filesystems"), which introduced max_readahead at FUSE protocol 7.6 and used min() to preserve legacy (<7.6) filesystem behaviour. Modern filesystems that explicitly advertise a larger max_readahead are silently overridden.
Other filesystems set ra_pages or io_pages directly from negotiated server/device capabilities: cifs sets ra_pages from rsize/rasize, ceph from rasize/rsize mount options, 9p from maxdata, and nfs sets io_pages from rpages.
Use the server's max_readahead response directly, bounded by fc->max_pages (which is itself bounded by fc->max_pages_limit and, for virtio-fs, by the virtqueue descriptor count):
fm->sb->s_bdi->ra_pages = min_t(unsigned int, ra_pages, fc->max_pages);
This is backward compatible:
- Servers that echo init_in->max_readahead back unchanged see the same effective readahead as today.
- Servers that reply with a smaller value still reduce ra_pages.
- Servers that do not negotiate FUSE_MAX_PAGES see no change, since fc->max_pages defaults to FUSE_DEFAULT_MAX_PAGES_PER_REQ (32), matching VM_READAHEAD_PAGES.
- Only servers that both negotiate FUSE_MAX_PAGES and advertise a larger max_readahead see the new behaviour, and in that case fc->max_pages already gates per-request data size.
---
Notes on AI assistance:
The code analysis (tracing the readahead negotiation in process_init_reply(), confirming the behaviour of ractl_max_pages() in mm/readahead.c, and surveying how other filesystems set ra_pages/io_pages) and the bulk of this changelog were drafted with an AI coding assistant (see Assisted-by trailer). The one-line code change was reviewed by me. The motivating performance observation (a 1MB application read producing four 256KB FUSE_READ requests against a server advertising max_pages=256 and max_write=1MB) was observed by me on a real virtio-fs workload prior to any AI involvement, and verification of patched and unpatched behaviour was performed by me.
fs/fuse/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/fuse/inode.c +1 -1
@@ -1494,7 +1494,7 @@ static void process_init_reply(struct fuse_mount *fm, struct fuse_args *args,init_server_timeout (fc, timeout); fm->sb->s_bdi->ra_pages = -min (fm->sb->s_bdi->ra_pages, ra_pages); + min_t(unsigned int, ra_pages, fc->max_pages); fc->minor = arg->minor; fc->max_write = arg->minor <5 ?4096 : arg->max_write; fc->max_write =max_t (unsigned,4096 , fc->max_write);
Signature · 1 line
2.43.0
Quoted 77 lines · Jim Harris
A FUSE server that advertises a large
max_pagesandmax_write(e.g. max_pages=256, max_write=1MB) cannot currently obtain matchingFUSE_READrequest sizes from the kernel. Buffered sequential writes arrive at the server at the negotiatedmax_writesize, but buffered sequential reads remain capped at the kernel's default readahead window (VM_READAHEAD_PAGES, 128KB; doubled to 256KB for files markedPOSIX_FADV_SEQUENTIAL). A 1MB applicationread()therefore turns into four sequential 256KBFUSE_READround-trips instead of one.This is because
process_init_reply()processes the server'smax_readaheadresponse as:ra_pages = arg->max_readahead / PAGE_SIZE; fm->sb->s_bdi->ra_pages = min(fm->sb->s_bdi->ra_pages, ra_pages);Since the kernel sends its current
bdi->ra_pagesasinit_in->max_readahead, andbdi->ra_pagesis the defaultVM_READAHEAD_PAGESat this point, the server can only ever decrease the readahead window -- never increase it. Even if the server replies with max_readahead=1MB, themin()clamps it back to 128KB.This clamp dates to commit
9cd684551124("[PATCH] fuse: fix async read for legacy filesystems"), which introducedmax_readaheadat FUSE protocol 7.6 and usedmin()to preserve legacy (<7.6) filesystem behaviour. Modern filesystems that explicitly advertise a largermax_readaheadare silently overridden.Other filesystems set
ra_pagesorio_pagesdirectly from negotiated server/device capabilities: cifs setsra_pagesfrom rsize/rasize, ceph from rasize/rsize mount options, 9p from maxdata, and nfs setsio_pagesfrom rpages.Use the server's
max_readaheadresponse directly, bounded byfc->max_pages(which is itself bounded byfc->max_pages_limitand, for virtio-fs, by the virtqueue descriptor count):fm->sb->s_bdi->ra_pages = min_t(unsigned int, ra_pages, fc->max_pages);This is backward compatible:
- Servers that echo
init_in->max_readaheadback unchanged see the same effective readahead as today.
- Servers that reply with a smaller value still reducera_pages.
- Servers that do not negotiateFUSE_MAX_PAGESsee no change, sincefc->max_pagesdefaults toFUSE_DEFAULT_MAX_PAGES_PER_REQ(32), matchingVM_READAHEAD_PAGES.
- Only servers that both negotiateFUSE_MAX_PAGESand advertise a largermax_readaheadsee the new behaviour, and in that casefc->max_pagesalready gates per-request data size.Signed-off-by Jim Harris <[email protected]>Assisted-by Cursor:claude-opus-4.7---
Notes on AI assistance:The code analysis (tracing the readahead negotiation in
process_init_reply(), confirming the behaviour ofractl_max_pages()inmm/readahead.c, and surveying how other filesystems set ra_pages/io_pages) and the bulk of this changelog were drafted with an AI coding assistant (see Assisted-by trailer). The one-line code change was reviewed by me. The motivating performance observation (a 1MB application read producing four 256KBFUSE_READrequests against a server advertising max_pages=256 and max_write=1MB) was observed by me on a real virtio-fs workload prior to any AI involvement, and verification of patched and unpatched behaviour was performed by me.fs/fuse/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)fs/fuse/inode.c +0
-0
fs/fuse/inode.c static void process_init_reply(struct fuse_mount *fm, struct fuse_args *args, · line 1497
Quoted 5 lines
fs/fuse/inode.c +1
-1@@ -1494,7 +1494,7 @@ static void process_init_reply(struct fuse_mount *fm, struct fuse_args *args,init_server_timeout (fc, timeout); fm->sb->s_bdi->ra_pages = -min (fm->sb->s_bdi->ra_pages, ra_pages); + min_t(unsigned int, ra_pages, fc->max_pages);
Looking at how the mm code uses ra_pages, I think this will also have the side effect of upping the number of pages read into the page cache for speculative readahead. I don't think this is safe, at least not for unprivileged servers.
I think setting s_bdi->io_pages would be a better fit here. fromthe logic in page_cache_sync_ra() -> ractl_max_pages(), this would only exceed the readahead window size limit for non-speculative readahead.
Thanks,
Joanne
Quoted 7 lines
fs/fuse/inode.c +0
-1fc->minor = arg->minor; fc->max_write = arg->minor <5 ?4096 : arg->max_write; fc->max_write =max_t (unsigned,4096 , fc->max_write); --2.43.0