Commit 580f2367 authored by Trond Myklebust's avatar Trond Myklebust

NFS: Adjust the amount of readahead performed by NFS readdir

The current NFS readdir code will always try to maximise the amount of
readahead it performs on the assumption that we can cache anything that
isn't immediately read by the process.
There are several cases where this assumption breaks down, including
when the 'ls -l' heuristic kicks in to try to force use of readdirplus
as a batch replacement for lookup/getattr.

This patch therefore tries to tone down the amount of readahead we
perform, and adjust it to try to match the amount of data being
requested by user space.
Signed-off-by: default avatarTrond Myklebust <trond.myklebust@hammerspace.com>
parent c8f0523b
...@@ -69,6 +69,8 @@ const struct address_space_operations nfs_dir_aops = { ...@@ -69,6 +69,8 @@ const struct address_space_operations nfs_dir_aops = {
.freepage = nfs_readdir_clear_array, .freepage = nfs_readdir_clear_array,
}; };
#define NFS_INIT_DTSIZE PAGE_SIZE
static struct nfs_open_dir_context * static struct nfs_open_dir_context *
alloc_nfs_open_dir_context(struct inode *dir) alloc_nfs_open_dir_context(struct inode *dir)
{ {
...@@ -78,6 +80,7 @@ alloc_nfs_open_dir_context(struct inode *dir) ...@@ -78,6 +80,7 @@ alloc_nfs_open_dir_context(struct inode *dir)
ctx = kzalloc(sizeof(*ctx), GFP_KERNEL_ACCOUNT); ctx = kzalloc(sizeof(*ctx), GFP_KERNEL_ACCOUNT);
if (ctx != NULL) { if (ctx != NULL) {
ctx->attr_gencount = nfsi->attr_gencount; ctx->attr_gencount = nfsi->attr_gencount;
ctx->dtsize = NFS_INIT_DTSIZE;
spin_lock(&dir->i_lock); spin_lock(&dir->i_lock);
if (list_empty(&nfsi->open_files) && if (list_empty(&nfsi->open_files) &&
(nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER)) (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER))
...@@ -154,6 +157,7 @@ struct nfs_readdir_descriptor { ...@@ -154,6 +157,7 @@ struct nfs_readdir_descriptor {
struct page *page; struct page *page;
struct dir_context *ctx; struct dir_context *ctx;
pgoff_t page_index; pgoff_t page_index;
pgoff_t page_index_max;
u64 dir_cookie; u64 dir_cookie;
u64 last_cookie; u64 last_cookie;
u64 dup_cookie; u64 dup_cookie;
...@@ -166,12 +170,36 @@ struct nfs_readdir_descriptor { ...@@ -166,12 +170,36 @@ struct nfs_readdir_descriptor {
unsigned long gencount; unsigned long gencount;
unsigned long attr_gencount; unsigned long attr_gencount;
unsigned int cache_entry_index; unsigned int cache_entry_index;
unsigned int buffer_fills;
unsigned int dtsize;
signed char duped; signed char duped;
bool plus; bool plus;
bool eob; bool eob;
bool eof; bool eof;
}; };
static void nfs_set_dtsize(struct nfs_readdir_descriptor *desc, unsigned int sz)
{
struct nfs_server *server = NFS_SERVER(file_inode(desc->file));
unsigned int maxsize = server->dtsize;
if (sz > maxsize)
sz = maxsize;
if (sz < NFS_MIN_FILE_IO_SIZE)
sz = NFS_MIN_FILE_IO_SIZE;
desc->dtsize = sz;
}
static void nfs_shrink_dtsize(struct nfs_readdir_descriptor *desc)
{
nfs_set_dtsize(desc, desc->dtsize >> 1);
}
static void nfs_grow_dtsize(struct nfs_readdir_descriptor *desc)
{
nfs_set_dtsize(desc, desc->dtsize << 1);
}
static void nfs_readdir_array_init(struct nfs_cache_array *array) static void nfs_readdir_array_init(struct nfs_cache_array *array)
{ {
memset(array, 0, sizeof(struct nfs_cache_array)); memset(array, 0, sizeof(struct nfs_cache_array));
...@@ -784,6 +812,7 @@ static int nfs_readdir_page_filler(struct nfs_readdir_descriptor *desc, ...@@ -784,6 +812,7 @@ static int nfs_readdir_page_filler(struct nfs_readdir_descriptor *desc,
break; break;
arrays++; arrays++;
*arrays = page = new; *arrays = page = new;
desc->page_index_max++;
} else { } else {
new = nfs_readdir_page_get_next(mapping, new = nfs_readdir_page_get_next(mapping,
page->index + 1, page->index + 1,
...@@ -793,6 +822,7 @@ static int nfs_readdir_page_filler(struct nfs_readdir_descriptor *desc, ...@@ -793,6 +822,7 @@ static int nfs_readdir_page_filler(struct nfs_readdir_descriptor *desc,
if (page != *arrays) if (page != *arrays)
nfs_readdir_page_unlock_and_put(page); nfs_readdir_page_unlock_and_put(page);
page = new; page = new;
desc->page_index_max = new->index;
} }
status = nfs_readdir_add_to_array(entry, page); status = nfs_readdir_add_to_array(entry, page);
} while (!status && !entry->eof); } while (!status && !entry->eof);
...@@ -858,7 +888,7 @@ static int nfs_readdir_xdr_to_array(struct nfs_readdir_descriptor *desc, ...@@ -858,7 +888,7 @@ static int nfs_readdir_xdr_to_array(struct nfs_readdir_descriptor *desc,
struct nfs_entry *entry; struct nfs_entry *entry;
size_t array_size; size_t array_size;
struct inode *inode = file_inode(desc->file); struct inode *inode = file_inode(desc->file);
size_t dtsize = NFS_SERVER(inode)->dtsize; unsigned int dtsize = desc->dtsize;
int status = -ENOMEM; int status = -ENOMEM;
entry = kzalloc(sizeof(*entry), GFP_KERNEL); entry = kzalloc(sizeof(*entry), GFP_KERNEL);
...@@ -894,6 +924,7 @@ static int nfs_readdir_xdr_to_array(struct nfs_readdir_descriptor *desc, ...@@ -894,6 +924,7 @@ static int nfs_readdir_xdr_to_array(struct nfs_readdir_descriptor *desc,
status = nfs_readdir_page_filler(desc, entry, pages, pglen, status = nfs_readdir_page_filler(desc, entry, pages, pglen,
arrays, narrays); arrays, narrays);
desc->buffer_fills++;
} while (!status && nfs_readdir_page_needs_filling(page) && } while (!status && nfs_readdir_page_needs_filling(page) &&
page_mapping(page)); page_mapping(page));
...@@ -941,6 +972,10 @@ static int find_and_lock_cache_page(struct nfs_readdir_descriptor *desc) ...@@ -941,6 +972,10 @@ static int find_and_lock_cache_page(struct nfs_readdir_descriptor *desc)
if (!desc->page) if (!desc->page)
return -ENOMEM; return -ENOMEM;
if (nfs_readdir_page_needs_filling(desc->page)) { if (nfs_readdir_page_needs_filling(desc->page)) {
/* Grow the dtsize if we had to go back for more pages */
if (desc->page_index == desc->page_index_max)
nfs_grow_dtsize(desc);
desc->page_index_max = desc->page_index;
res = nfs_readdir_xdr_to_array(desc, nfsi->cookieverf, verf, res = nfs_readdir_xdr_to_array(desc, nfsi->cookieverf, verf,
&desc->page, 1); &desc->page, 1);
if (res < 0) { if (res < 0) {
...@@ -1075,6 +1110,7 @@ static int uncached_readdir(struct nfs_readdir_descriptor *desc) ...@@ -1075,6 +1110,7 @@ static int uncached_readdir(struct nfs_readdir_descriptor *desc)
desc->cache_entry_index = 0; desc->cache_entry_index = 0;
desc->last_cookie = desc->dir_cookie; desc->last_cookie = desc->dir_cookie;
desc->duped = 0; desc->duped = 0;
desc->page_index_max = 0;
status = nfs_readdir_xdr_to_array(desc, desc->verf, verf, arrays, sz); status = nfs_readdir_xdr_to_array(desc, desc->verf, verf, arrays, sz);
...@@ -1084,10 +1120,22 @@ static int uncached_readdir(struct nfs_readdir_descriptor *desc) ...@@ -1084,10 +1120,22 @@ static int uncached_readdir(struct nfs_readdir_descriptor *desc)
} }
desc->page = NULL; desc->page = NULL;
/*
* Grow the dtsize if we have to go back for more pages,
* or shrink it if we're reading too many.
*/
if (!desc->eof) {
if (!desc->eob)
nfs_grow_dtsize(desc);
else if (desc->buffer_fills == 1 &&
i < (desc->page_index_max >> 1))
nfs_shrink_dtsize(desc);
}
for (i = 0; i < sz && arrays[i]; i++) for (i = 0; i < sz && arrays[i]; i++)
nfs_readdir_page_array_free(arrays[i]); nfs_readdir_page_array_free(arrays[i]);
out: out:
desc->page_index_max = -1;
kfree(arrays); kfree(arrays);
dfprintk(DIRCACHE, "NFS: %s: returns %d\n", __func__, status); dfprintk(DIRCACHE, "NFS: %s: returns %d\n", __func__, status);
return status; return status;
...@@ -1126,6 +1174,7 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) ...@@ -1126,6 +1174,7 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx)
desc->file = file; desc->file = file;
desc->ctx = ctx; desc->ctx = ctx;
desc->plus = nfs_use_readdirplus(inode, ctx); desc->plus = nfs_use_readdirplus(inode, ctx);
desc->page_index_max = -1;
spin_lock(&file->f_lock); spin_lock(&file->f_lock);
desc->dir_cookie = dir_ctx->dir_cookie; desc->dir_cookie = dir_ctx->dir_cookie;
...@@ -1136,6 +1185,7 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) ...@@ -1136,6 +1185,7 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx)
desc->last_cookie = dir_ctx->last_cookie; desc->last_cookie = dir_ctx->last_cookie;
desc->attr_gencount = dir_ctx->attr_gencount; desc->attr_gencount = dir_ctx->attr_gencount;
desc->eof = dir_ctx->eof; desc->eof = dir_ctx->eof;
nfs_set_dtsize(desc, dir_ctx->dtsize);
memcpy(desc->verf, dir_ctx->verf, sizeof(desc->verf)); memcpy(desc->verf, dir_ctx->verf, sizeof(desc->verf));
spin_unlock(&file->f_lock); spin_unlock(&file->f_lock);
...@@ -1187,6 +1237,7 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) ...@@ -1187,6 +1237,7 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx)
dir_ctx->attr_gencount = desc->attr_gencount; dir_ctx->attr_gencount = desc->attr_gencount;
dir_ctx->page_index = desc->page_index; dir_ctx->page_index = desc->page_index;
dir_ctx->eof = desc->eof; dir_ctx->eof = desc->eof;
dir_ctx->dtsize = desc->dtsize;
memcpy(dir_ctx->verf, desc->verf, sizeof(dir_ctx->verf)); memcpy(dir_ctx->verf, desc->verf, sizeof(dir_ctx->verf));
spin_unlock(&file->f_lock); spin_unlock(&file->f_lock);
out_free: out_free:
......
...@@ -107,6 +107,7 @@ struct nfs_open_dir_context { ...@@ -107,6 +107,7 @@ struct nfs_open_dir_context {
__u64 dup_cookie; __u64 dup_cookie;
__u64 last_cookie; __u64 last_cookie;
pgoff_t page_index; pgoff_t page_index;
unsigned int dtsize;
signed char duped; signed char duped;
bool eof; bool eof;
}; };
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment