#include "ceph_debug.h"
#include <linux/sched.h>
+#include <linux/slab.h>
#include <linux/file.h>
#include <linux/namei.h>
#include <linux/writeback.h>
/* do the open */
req = prepare_open_request(dir->i_sb, flags, mode);
if (IS_ERR(req))
- return ERR_PTR(PTR_ERR(req));
+ return ERR_CAST(req);
req->r_dentry = dget(dentry);
req->r_num_caps = 2;
if (flags & O_CREAT) {
kfree(cf->dir_info);
dput(cf->dentry);
kmem_cache_free(ceph_file_cachep, cf);
+
+ /* wake up anyone waiting for caps on this inode */
+ wake_up(&ci->i_cap_wq);
return 0;
}
/*
* allocate a vector new pages
*/
-static struct page **alloc_page_vector(int num_pages)
+struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags)
{
struct page **pages;
int i;
- pages = kmalloc(sizeof(*pages) * num_pages, GFP_NOFS);
+ pages = kmalloc(sizeof(*pages) * num_pages, flags);
if (!pages)
return ERR_PTR(-ENOMEM);
for (i = 0; i < num_pages; i++) {
- pages[i] = alloc_page(GFP_NOFS);
+ pages[i] = __page_cache_alloc(flags);
if (pages[i] == NULL) {
ceph_release_page_vector(pages, i);
return ERR_PTR(-ENOMEM);
{
int i = off >> PAGE_CACHE_SHIFT;
+ off &= ~PAGE_CACHE_MASK;
+
dout("zero_page_vector_page %u~%u\n", off, len);
- BUG_ON(len < PAGE_CACHE_SIZE);
/* leading partial page? */
- if (off & ~PAGE_CACHE_MASK) {
+ if (off) {
+ int end = min((int)PAGE_CACHE_SIZE, off + len);
dout("zeroing %d %p head from %d\n", i, pages[i],
- (int)(off & ~PAGE_CACHE_MASK));
- zero_user_segment(pages[i], off & ~PAGE_CACHE_MASK,
- PAGE_CACHE_SIZE);
- off += PAGE_CACHE_SIZE;
- off &= PAGE_CACHE_MASK;
+ (int)off);
+ zero_user_segment(pages[i], off, end);
+ len -= (end - off);
i++;
}
while (len >= PAGE_CACHE_SIZE) {
- dout("zeroing %d %p\n", i, pages[i]);
+ dout("zeroing %d %p len=%d\n", i, pages[i], len);
zero_user_segment(pages[i], 0, PAGE_CACHE_SIZE);
- off += PAGE_CACHE_SIZE;
len -= PAGE_CACHE_SIZE;
i++;
}
*/
static int striped_read(struct inode *inode,
u64 off, u64 len,
- struct page **pages, int num_pages)
+ struct page **pages, int num_pages,
+ int *checkeof)
{
struct ceph_client *client = ceph_inode_to_client(inode);
struct ceph_inode_info *ci = ceph_inode(inode);
u64 pos, this_len;
- int page_off = off & ~PAGE_CACHE_SIZE; /* first byte's offset in page */
+ int page_off = off & ~PAGE_CACHE_MASK; /* first byte's offset in page */
int left, pages_left;
int read;
struct page **page_pos;
dout("zero tail\n");
zero_page_vector_range(page_off + read, len - read,
pages);
+ read = len;
goto out;
}
/* check i_size */
- ret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE);
- if (ret < 0)
- goto out;
-
- /* hit EOF? */
- if (pos >= inode->i_size)
- goto out;
-
- goto more;
+ *checkeof = 1;
}
out:
* If the read spans object boundary, just do multiple reads.
*/
static ssize_t ceph_sync_read(struct file *file, char __user *data,
- unsigned len, loff_t *poff)
+ unsigned len, loff_t *poff, int *checkeof)
{
struct inode *inode = file->f_dentry->d_inode;
struct page **pages;
* but it will at least behave sensibly when they are
* in sequence.
*/
- filemap_write_and_wait(inode->i_mapping);
} else {
- pages = alloc_page_vector(num_pages);
+ pages = ceph_alloc_page_vector(num_pages, GFP_NOFS);
}
if (IS_ERR(pages))
return PTR_ERR(pages);
- ret = striped_read(inode, off, len, pages, num_pages);
+ ret = filemap_write_and_wait(inode->i_mapping);
+ if (ret < 0)
+ goto done;
+
+ ret = striped_read(inode, off, len, pages, num_pages, checkeof);
if (ret >= 0 && (file->f_flags & O_DIRECT) == 0)
ret = copy_page_vector_to_user(pages, data, off, ret);
if (ret >= 0)
*poff = off + ret;
+done:
if (file->f_flags & O_DIRECT)
put_page_vector(pages, num_pages);
else
else
pos = *offset;
+ ret = filemap_write_and_wait_range(inode->i_mapping, pos, pos + left);
+ if (ret < 0)
+ return ret;
+
+ ret = invalidate_inode_pages2_range(inode->i_mapping,
+ pos >> PAGE_CACHE_SHIFT,
+ (pos + left) >> PAGE_CACHE_SHIFT);
+ if (ret < 0)
+ dout("invalidate_inode_pages2_range returned %d\n", ret);
+
flags = CEPH_OSD_FLAG_ORDERSNAP |
CEPH_OSD_FLAG_ONDISK |
CEPH_OSD_FLAG_WRITE;
do_sync,
ci->i_truncate_seq, ci->i_truncate_size,
&mtime, false, 2);
- if (IS_ERR(req))
- return PTR_ERR(req);
+ if (!req)
+ return -ENOMEM;
num_pages = calc_pages_for(pos, len);
* throw out any page cache pages in this range. this
* may block.
*/
- truncate_inode_pages_range(inode->i_mapping, pos, pos+len);
+ truncate_inode_pages_range(inode->i_mapping, pos,
+ (pos+len) | (PAGE_CACHE_SIZE-1));
} else {
- pages = alloc_page_vector(num_pages);
+ pages = ceph_alloc_page_vector(num_pages, GFP_NOFS);
if (IS_ERR(pages)) {
ret = PTR_ERR(pages);
goto out;
size_t len = iov->iov_len;
struct inode *inode = filp->f_dentry->d_inode;
struct ceph_inode_info *ci = ceph_inode(inode);
+ void *base = iov->iov_base;
ssize_t ret;
int got = 0;
+ int checkeof = 0, read = 0;
dout("aio_read %p %llx.%llx %llu~%u trying to get caps on %p\n",
inode, ceph_vinop(inode), pos, (unsigned)len, inode);
+again:
__ceph_do_pending_vmtruncate(inode);
ret = ceph_get_caps(ci, CEPH_CAP_FILE_RD, CEPH_CAP_FILE_CACHE,
&got, -1);
(iocb->ki_filp->f_flags & O_DIRECT) ||
(inode->i_sb->s_flags & MS_SYNCHRONOUS))
/* hmm, this isn't really async... */
- ret = ceph_sync_read(filp, iov->iov_base, len, ppos);
+ ret = ceph_sync_read(filp, base, len, ppos, &checkeof);
else
ret = generic_file_aio_read(iocb, iov, nr_segs, pos);
dout("aio_read %p %llx.%llx dropping cap refs on %s = %d\n",
inode, ceph_vinop(inode), ceph_cap_string(got), (int)ret);
ceph_put_cap_refs(ci, got);
+
+ if (checkeof && ret >= 0) {
+ int statret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE);
+
+ /* hit EOF or hole? */
+ if (statret == 0 && *ppos < inode->i_size) {
+ dout("aio_read sync_read hit hole, reading more\n");
+ read += ret;
+ base += ret;
+ len -= ret;
+ checkeof = 0;
+ goto again;
+ }
+ }
+ if (ret >= 0)
+ ret += read;
+
return ret;
}
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_dentry->d_inode;
struct ceph_inode_info *ci = ceph_inode(inode);
- struct ceph_osd_client *osdc = &ceph_client(inode->i_sb)->osdc;
+ struct ceph_osd_client *osdc = &ceph_sb_to_client(inode->i_sb)->osdc;
loff_t endoff = pos + iov->iov_len;
int got = 0;
- int ret;
+ int ret, err;
if (ceph_snap(inode) != CEPH_NOSNAP)
return -EROFS;
if ((ret >= 0 || ret == -EIOCBQUEUED) &&
((file->f_flags & O_SYNC) || IS_SYNC(file->f_mapping->host)
- || ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_NEARFULL)))
- ret = vfs_fsync_range(file, file->f_path.dentry,
+ || ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_NEARFULL))) {
+ err = vfs_fsync_range(file, file->f_path.dentry,
pos, pos + ret - 1, 1);
+ if (err < 0)
+ ret = err;
+ }
}
if (ret >= 0) {
spin_lock(&inode->i_lock);
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff