lkml.org 
[lkml]   [2023]   [Jul]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [PATCH] fuse: enable larger read buffers for readdir.
From


On 7/26/23 12:59, Jaco Kroon wrote:
> Signed-off-by: Jaco Kroon <jaco@uls.co.za>
> ---
> fs/fuse/Kconfig | 16 ++++++++++++++++
> fs/fuse/readdir.c | 42 ++++++++++++++++++++++++------------------
> 2 files changed, 40 insertions(+), 18 deletions(-)
>
> diff --git a/fs/fuse/Kconfig b/fs/fuse/Kconfig
> index 038ed0b9aaa5..0783f9ee5cd3 100644
> --- a/fs/fuse/Kconfig
> +++ b/fs/fuse/Kconfig
> @@ -18,6 +18,22 @@ config FUSE_FS
> If you want to develop a userspace FS, or if you want to use
> a filesystem based on FUSE, answer Y or M.
>
> +config FUSE_READDIR_ORDER
> + int
> + range 0 5
> + default 5
> + help
> + readdir performance varies greatly depending on the size of the read.
> + Larger buffers results in larger reads, thus fewer reads and higher
> + performance in return.
> +
> + You may want to reduce this value on seriously constrained memory
> + systems where 128KiB (assuming 4KiB pages) cache pages is not ideal.
> +
> + This value reprents the order of the number of pages to allocate (ie,
> + the shift value). A value of 0 is thus 1 page (4KiB) where 5 is 32
> + pages (128KiB).
> +

I like the idea of a larger readdir size, but shouldn't that be a
server/daemon/library decision which size to use, instead of kernel
compile time? So should be part of FUSE_INIT negotiation?

> config CUSE
> tristate "Character device in Userspace support"
> depends on FUSE_FS
> diff --git a/fs/fuse/readdir.c b/fs/fuse/readdir.c
> index dc603479b30e..98c62b623240 100644
> --- a/fs/fuse/readdir.c
> +++ b/fs/fuse/readdir.c
> @@ -13,6 +13,12 @@
> #include <linux/pagemap.h>
> #include <linux/highmem.h>
>
> +#define READDIR_PAGES_ORDER CONFIG_FUSE_READDIR_ORDER
> +#define READDIR_PAGES (1 << READDIR_PAGES_ORDER)
> +#define READDIR_PAGES_SIZE (PAGE_SIZE << READDIR_PAGES_ORDER)
> +#define READDIR_PAGES_MASK (READDIR_PAGES_SIZE - 1)
> +#define READDIR_PAGES_SHIFT (PAGE_SHIFT + READDIR_PAGES_ORDER)
> +
> static bool fuse_use_readdirplus(struct inode *dir, struct dir_context *ctx)
> {
> struct fuse_conn *fc = get_fuse_conn(dir);
> @@ -52,10 +58,10 @@ static void fuse_add_dirent_to_cache(struct file *file,
> }
> version = fi->rdc.version;
> size = fi->rdc.size;
> - offset = size & ~PAGE_MASK;
> - index = size >> PAGE_SHIFT;
> + offset = size & ~READDIR_PAGES_MASK;
> + index = size >> READDIR_PAGES_SHIFT;
> /* Dirent doesn't fit in current page? Jump to next page. */
> - if (offset + reclen > PAGE_SIZE) {
> + if (offset + reclen > READDIR_PAGES_SIZE) {
> index++;
> offset = 0;
> }
> @@ -83,7 +89,7 @@ static void fuse_add_dirent_to_cache(struct file *file,
> }
> memcpy(addr + offset, dirent, reclen);
> kunmap_local(addr);
> - fi->rdc.size = (index << PAGE_SHIFT) + offset + reclen;
> + fi->rdc.size = (index << READDIR_PAGES_SHIFT) + offset + reclen;
> fi->rdc.pos = dirent->off;
> unlock:
> spin_unlock(&fi->rdc.lock);
> @@ -104,7 +110,7 @@ static void fuse_readdir_cache_end(struct file *file, loff_t pos)
> }
>
> fi->rdc.cached = true;
> - end = ALIGN(fi->rdc.size, PAGE_SIZE);
> + end = ALIGN(fi->rdc.size, READDIR_PAGES_SIZE);
> spin_unlock(&fi->rdc.lock);
>
> /* truncate unused tail of cache */
> @@ -328,25 +334,25 @@ static int fuse_readdir_uncached(struct file *file, struct dir_context *ctx)
> struct fuse_mount *fm = get_fuse_mount(inode);
> struct fuse_io_args ia = {};
> struct fuse_args_pages *ap = &ia.ap;
> - struct fuse_page_desc desc = { .length = PAGE_SIZE };
> + struct fuse_page_desc desc = { .length = READDIR_PAGES_SIZE };
> u64 attr_version = 0;
> bool locked;
>
> - page = alloc_page(GFP_KERNEL);
> + page = alloc_pages(GFP_KERNEL, READDIR_PAGES_ORDER);

I guess that should become folio alloc(), one way or the other. Now I
think order 0 was chosen before to avoid risk of allocation failure. I
guess it might work to try a large size and to fall back to 0 when that
failed. Or fail back to the slower vmalloc.

> if (!page)
> return -ENOMEM;
>
> plus = fuse_use_readdirplus(inode, ctx);
> ap->args.out_pages = true;
> - ap->num_pages = 1;
> + ap->num_pages = READDIR_PAGES;
> ap->pages = &page;
> ap->descs = &desc;
> if (plus) {
> attr_version = fuse_get_attr_version(fm->fc);
> - fuse_read_args_fill(&ia, file, ctx->pos, PAGE_SIZE,
> + fuse_read_args_fill(&ia, file, ctx->pos, READDIR_PAGES_SIZE,
> FUSE_READDIRPLUS);
> } else {
> - fuse_read_args_fill(&ia, file, ctx->pos, PAGE_SIZE,
> + fuse_read_args_fill(&ia, file, ctx->pos, READDIR_PAGES_SIZE,
> FUSE_READDIR);
> }
> locked = fuse_lock_inode(inode);
> @@ -383,7 +389,7 @@ static enum fuse_parse_result fuse_parse_cache(struct fuse_file *ff,
> void *addr, unsigned int size,
> struct dir_context *ctx)
> {
> - unsigned int offset = ff->readdir.cache_off & ~PAGE_MASK;
> + unsigned int offset = ff->readdir.cache_off & ~READDIR_PAGES_MASK;
> enum fuse_parse_result res = FOUND_NONE;
>
> WARN_ON(offset >= size);
> @@ -504,16 +510,16 @@ static int fuse_readdir_cached(struct file *file, struct dir_context *ctx)
>
> WARN_ON(fi->rdc.size < ff->readdir.cache_off);
>
> - index = ff->readdir.cache_off >> PAGE_SHIFT;
> + index = ff->readdir.cache_off >> READDIR_PAGES_SHIFT;
>
> - if (index == (fi->rdc.size >> PAGE_SHIFT))
> - size = fi->rdc.size & ~PAGE_MASK;
> + if (index == (fi->rdc.size >> READDIR_PAGES_SHIFT))
> + size = fi->rdc.size & ~READDIR_PAGES_MASK;
> else
> - size = PAGE_SIZE;
> + size = READDIR_PAGES_SIZE;
> spin_unlock(&fi->rdc.lock);
>
> /* EOF? */
> - if ((ff->readdir.cache_off & ~PAGE_MASK) == size)
> + if ((ff->readdir.cache_off & ~READDIR_PAGES_MASK) == size)
> return 0;
>
> page = find_get_page_flags(file->f_mapping, index,
> @@ -559,9 +565,9 @@ static int fuse_readdir_cached(struct file *file, struct dir_context *ctx)
> if (res == FOUND_ALL)
> return 0;
>
> - if (size == PAGE_SIZE) {
> + if (size == READDIR_PAGES_SIZE) {
> /* We hit end of page: skip to next page. */
> - ff->readdir.cache_off = ALIGN(ff->readdir.cache_off, PAGE_SIZE);
> + ff->readdir.cache_off = ALIGN(ff->readdir.cache_off, READDIR_PAGES_SIZE);
> goto retry;
> }
>

Thanks,
Bernd

\
 
 \ /
  Last update: 2023-07-26 15:56    [W:0.096 / U:1.016 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site