#include <linux/slab.h>
#include <linux/file.h>
#include <linux/uio.h>
+#include <linux/falloc.h>
#include <linux/sched/mm.h>
-#include <linux/netfs.h>
+#include <trace/events/fscache.h>
#include "internal.h"
struct cachefiles_kiocb {
size_t skipped;
size_t len;
};
+ struct cachefiles_object *object;
netfs_io_terminated_t term_func;
void *term_func_priv;
bool was_async;
+ unsigned int inval_counter; /* Copy of cookie->inval_counter */
+ u64 b_writing;
};
static inline void cachefiles_put_kiocb(struct cachefiles_kiocb *ki)
{
if (refcount_dec_and_test(&ki->ki_refcnt)) {
+ cachefiles_put_object(ki->object, cachefiles_obj_put_ioreq);
fput(ki->iocb.ki_filp);
kfree(ki);
}
static void cachefiles_read_complete(struct kiocb *iocb, long ret)
{
struct cachefiles_kiocb *ki = container_of(iocb, struct cachefiles_kiocb, iocb);
+ struct inode *inode = file_inode(ki->iocb.ki_filp);
_enter("%ld", ret);
+ if (ret < 0)
+ trace_cachefiles_io_error(ki->object, inode, ret,
+ cachefiles_trace_read_error);
+
if (ki->term_func) {
- if (ret >= 0)
- ret += ki->skipped;
+ if (ret >= 0) {
+ if (ki->object->cookie->inval_counter == ki->inval_counter)
+ ki->skipped += ret;
+ else
+ ret = -ESTALE;
+ }
+
ki->term_func(ki->term_func_priv, ret, ki->was_async);
}
static int cachefiles_read(struct netfs_cache_resources *cres,
loff_t start_pos,
struct iov_iter *iter,
- bool seek_data,
+ enum netfs_read_from_hole read_hole,
netfs_io_terminated_t term_func,
void *term_func_priv)
{
+ struct cachefiles_object *object;
struct cachefiles_kiocb *ki;
- struct file *file = cres->cache_priv2;
+ struct file *file;
unsigned int old_nofs;
ssize_t ret = -ENOBUFS;
size_t len = iov_iter_count(iter), skipped = 0;
+ if (!fscache_wait_for_operation(cres, FSCACHE_WANT_READ))
+ goto presubmission_error;
+
+ fscache_count_read();
+ object = cachefiles_cres_object(cres);
+ file = cachefiles_cres_file(cres);
+
_enter("%pD,%li,%llx,%zx/%llx",
file, file_inode(file)->i_ino, start_pos, len,
i_size_read(file_inode(file)));
/* If the caller asked us to seek for data before doing the read, then
* we should do that now. If we find a gap, we fill it with zeros.
*/
- if (seek_data) {
+ if (read_hole != NETFS_READ_HOLE_IGNORE) {
loff_t off = start_pos, off2;
- off2 = vfs_llseek(file, off, SEEK_DATA);
+ off2 = cachefiles_inject_read_error();
+ if (off2 == 0)
+ off2 = vfs_llseek(file, off, SEEK_DATA);
if (off2 < 0 && off2 >= (loff_t)-MAX_ERRNO && off2 != -ENXIO) {
skipped = 0;
ret = off2;
* in the region, so clear the rest of the buffer and
* return success.
*/
+ ret = -ENODATA;
+ if (read_hole == NETFS_READ_HOLE_FAIL)
+ goto presubmission_error;
+
iov_iter_zero(len, iter);
skipped = len;
ret = 0;
iov_iter_zero(skipped, iter);
}
- ret = -ENOBUFS;
+ ret = -ENOMEM;
ki = kzalloc(sizeof(struct cachefiles_kiocb), GFP_KERNEL);
if (!ki)
goto presubmission_error;
ki->iocb.ki_hint = ki_hint_validate(file_write_hint(file));
ki->iocb.ki_ioprio = get_current_ioprio();
ki->skipped = skipped;
+ ki->object = object;
+ ki->inval_counter = cres->inval_counter;
ki->term_func = term_func;
ki->term_func_priv = term_func_priv;
ki->was_async = true;
ki->iocb.ki_complete = cachefiles_read_complete;
get_file(ki->iocb.ki_filp);
+ cachefiles_grab_object(object, cachefiles_obj_get_ioreq);
+ trace_cachefiles_read(object, file_inode(file), ki->iocb.ki_pos, len - skipped);
old_nofs = memalloc_nofs_save();
- ret = vfs_iocb_iter_read(file, &ki->iocb, iter);
+ ret = cachefiles_inject_read_error();
+ if (ret == 0)
+ ret = vfs_iocb_iter_read(file, &ki->iocb, iter);
memalloc_nofs_restore(old_nofs);
switch (ret) {
case -EIOCBQUEUED:
static void cachefiles_write_complete(struct kiocb *iocb, long ret)
{
struct cachefiles_kiocb *ki = container_of(iocb, struct cachefiles_kiocb, iocb);
+ struct cachefiles_object *object = ki->object;
struct inode *inode = file_inode(ki->iocb.ki_filp);
_enter("%ld", ret);
__sb_writers_acquired(inode->i_sb, SB_FREEZE_WRITE);
__sb_end_write(inode->i_sb, SB_FREEZE_WRITE);
+ if (ret < 0)
+ trace_cachefiles_io_error(object, inode, ret,
+ cachefiles_trace_write_error);
+
+ atomic_long_sub(ki->b_writing, &object->volume->cache->b_writing);
+ set_bit(FSCACHE_COOKIE_HAVE_DATA, &object->cookie->flags);
if (ki->term_func)
ki->term_func(ki->term_func_priv, ret, ki->was_async);
-
cachefiles_put_kiocb(ki);
}
netfs_io_terminated_t term_func,
void *term_func_priv)
{
+ struct cachefiles_object *object;
+ struct cachefiles_cache *cache;
struct cachefiles_kiocb *ki;
struct inode *inode;
- struct file *file = cres->cache_priv2;
+ struct file *file;
unsigned int old_nofs;
ssize_t ret = -ENOBUFS;
size_t len = iov_iter_count(iter);
+ if (!fscache_wait_for_operation(cres, FSCACHE_WANT_WRITE))
+ goto presubmission_error;
+ fscache_count_write();
+ object = cachefiles_cres_object(cres);
+ cache = object->volume->cache;
+ file = cachefiles_cres_file(cres);
+
_enter("%pD,%li,%llx,%zx/%llx",
file, file_inode(file)->i_ino, start_pos, len,
i_size_read(file_inode(file)));
+ ret = -ENOMEM;
ki = kzalloc(sizeof(struct cachefiles_kiocb), GFP_KERNEL);
if (!ki)
goto presubmission_error;
ki->iocb.ki_flags = IOCB_DIRECT | IOCB_WRITE;
ki->iocb.ki_hint = ki_hint_validate(file_write_hint(file));
ki->iocb.ki_ioprio = get_current_ioprio();
+ ki->object = object;
+ ki->inval_counter = cres->inval_counter;
ki->start = start_pos;
ki->len = len;
ki->term_func = term_func;
ki->term_func_priv = term_func_priv;
ki->was_async = true;
+ ki->b_writing = (len + (1 << cache->bshift)) >> cache->bshift;
if (ki->term_func)
ki->iocb.ki_complete = cachefiles_write_complete;
+ atomic_long_add(ki->b_writing, &cache->b_writing);
/* Open-code file_start_write here to grab freeze protection, which
* will be released by another thread in aio_complete_rw(). Fool
__sb_writers_release(inode->i_sb, SB_FREEZE_WRITE);
get_file(ki->iocb.ki_filp);
+ cachefiles_grab_object(object, cachefiles_obj_get_ioreq);
+ trace_cachefiles_write(object, inode, ki->iocb.ki_pos, len);
old_nofs = memalloc_nofs_save();
- ret = vfs_iocb_iter_write(file, &ki->iocb, iter);
+ ret = cachefiles_inject_write_error();
+ if (ret == 0)
+ ret = vfs_iocb_iter_write(file, &ki->iocb, iter);
memalloc_nofs_restore(old_nofs);
switch (ret) {
case -EIOCBQUEUED:
presubmission_error:
if (term_func)
- term_func(term_func_priv, -ENOMEM, false);
- return -ENOMEM;
+ term_func(term_func_priv, ret, false);
+ return ret;
}
/*
static enum netfs_read_source cachefiles_prepare_read(struct netfs_read_subrequest *subreq,
loff_t i_size)
{
- struct fscache_retrieval *op = subreq->rreq->cache_resources.cache_priv;
+ enum cachefiles_prepare_read_trace why;
+ struct netfs_read_request *rreq = subreq->rreq;
+ struct netfs_cache_resources *cres = &rreq->cache_resources;
struct cachefiles_object *object;
struct cachefiles_cache *cache;
+ struct fscache_cookie *cookie = fscache_cres_cookie(cres);
const struct cred *saved_cred;
- struct file *file = subreq->rreq->cache_resources.cache_priv2;
+ struct file *file = cachefiles_cres_file(cres);
+ enum netfs_read_source ret = NETFS_DOWNLOAD_FROM_SERVER;
loff_t off, to;
+ ino_t ino = file ? file_inode(file)->i_ino : 0;
_enter("%zx @%llx/%llx", subreq->len, subreq->start, i_size);
- object = container_of(op->op.object,
- struct cachefiles_object, fscache);
- cache = container_of(object->fscache.cache,
- struct cachefiles_cache, cache);
+ if (subreq->start >= i_size) {
+ ret = NETFS_FILL_WITH_ZEROES;
+ why = cachefiles_trace_read_after_eof;
+ goto out_no_object;
+ }
- if (!file)
- goto cache_fail_nosec;
+ if (test_bit(FSCACHE_COOKIE_NO_DATA_TO_READ, &cookie->flags)) {
+ __set_bit(NETFS_SREQ_WRITE_TO_CACHE, &subreq->flags);
+ why = cachefiles_trace_read_no_data;
+ goto out_no_object;
+ }
- if (subreq->start >= i_size)
- return NETFS_FILL_WITH_ZEROES;
+ /* The object and the file may be being created in the background. */
+ if (!file) {
+ why = cachefiles_trace_read_no_file;
+ if (!fscache_wait_for_operation(cres, FSCACHE_WANT_READ))
+ goto out_no_object;
+ file = cachefiles_cres_file(cres);
+ if (!file)
+ goto out_no_object;
+ ino = file_inode(file)->i_ino;
+ }
+ object = cachefiles_cres_object(cres);
+ cache = object->volume->cache;
cachefiles_begin_secure(cache, &saved_cred);
- off = vfs_llseek(file, subreq->start, SEEK_DATA);
+ off = cachefiles_inject_read_error();
+ if (off == 0)
+ off = vfs_llseek(file, subreq->start, SEEK_DATA);
if (off < 0 && off >= (loff_t)-MAX_ERRNO) {
- if (off == (loff_t)-ENXIO)
+ if (off == (loff_t)-ENXIO) {
+ why = cachefiles_trace_read_seek_nxio;
goto download_and_store;
- goto cache_fail;
+ }
+ trace_cachefiles_io_error(object, file_inode(file), off,
+ cachefiles_trace_seek_error);
+ why = cachefiles_trace_read_seek_error;
+ goto out;
}
- if (off >= subreq->start + subreq->len)
+ if (off >= subreq->start + subreq->len) {
+ why = cachefiles_trace_read_found_hole;
goto download_and_store;
+ }
if (off > subreq->start) {
off = round_up(off, cache->bsize);
subreq->len = off - subreq->start;
+ why = cachefiles_trace_read_found_part;
goto download_and_store;
}
- to = vfs_llseek(file, subreq->start, SEEK_HOLE);
- if (to < 0 && to >= (loff_t)-MAX_ERRNO)
- goto cache_fail;
+ to = cachefiles_inject_read_error();
+ if (to == 0)
+ to = vfs_llseek(file, subreq->start, SEEK_HOLE);
+ if (to < 0 && to >= (loff_t)-MAX_ERRNO) {
+ trace_cachefiles_io_error(object, file_inode(file), to,
+ cachefiles_trace_seek_error);
+ why = cachefiles_trace_read_seek_error;
+ goto out;
+ }
if (to < subreq->start + subreq->len) {
if (subreq->start + subreq->len >= i_size)
subreq->len = to - subreq->start;
}
- cachefiles_end_secure(cache, saved_cred);
- return NETFS_READ_FROM_CACHE;
+ why = cachefiles_trace_read_have_data;
+ ret = NETFS_READ_FROM_CACHE;
+ goto out;
download_and_store:
- if (cachefiles_has_space(cache, 0, (subreq->len + PAGE_SIZE - 1) / PAGE_SIZE) == 0)
- __set_bit(NETFS_SREQ_WRITE_TO_CACHE, &subreq->flags);
-cache_fail:
+ __set_bit(NETFS_SREQ_WRITE_TO_CACHE, &subreq->flags);
+out:
cachefiles_end_secure(cache, saved_cred);
-cache_fail_nosec:
- return NETFS_DOWNLOAD_FROM_SERVER;
+out_no_object:
+ trace_cachefiles_prep_read(subreq, ret, why, ino);
+ return ret;
}
/*
* Prepare for a write to occur.
*/
-static int cachefiles_prepare_write(struct netfs_cache_resources *cres,
- loff_t *_start, size_t *_len, loff_t i_size)
+static int __cachefiles_prepare_write(struct netfs_cache_resources *cres,
+ loff_t *_start, size_t *_len, loff_t i_size,
+ bool no_space_allocated_yet)
{
- loff_t start = *_start;
+ struct cachefiles_object *object = cachefiles_cres_object(cres);
+ struct cachefiles_cache *cache = object->volume->cache;
+ struct file *file = cachefiles_cres_file(cres);
+ loff_t start = *_start, pos;
size_t len = *_len, down;
+ int ret;
/* Round to DIO size */
down = start - round_down(start, PAGE_SIZE);
*_start = start - down;
*_len = round_up(down + len, PAGE_SIZE);
- return 0;
+
+ /* We need to work out whether there's sufficient disk space to perform
+ * the write - but we can skip that check if we have space already
+ * allocated.
+ */
+ if (no_space_allocated_yet)
+ goto check_space;
+
+ pos = cachefiles_inject_read_error();
+ if (pos == 0)
+ pos = vfs_llseek(file, *_start, SEEK_DATA);
+ if (pos < 0 && pos >= (loff_t)-MAX_ERRNO) {
+ if (pos == -ENXIO)
+ goto check_space; /* Unallocated tail */
+ trace_cachefiles_io_error(object, file_inode(file), pos,
+ cachefiles_trace_seek_error);
+ return pos;
+ }
+ if ((u64)pos >= (u64)*_start + *_len)
+ goto check_space; /* Unallocated region */
+
+ /* We have a block that's at least partially filled - if we're low on
+ * space, we need to see if it's fully allocated. If it's not, we may
+ * want to cull it.
+ */
+ if (cachefiles_has_space(cache, 0, *_len / PAGE_SIZE,
+ cachefiles_has_space_check) == 0)
+ return 0; /* Enough space to simply overwrite the whole block */
+
+ pos = cachefiles_inject_read_error();
+ if (pos == 0)
+ pos = vfs_llseek(file, *_start, SEEK_HOLE);
+ if (pos < 0 && pos >= (loff_t)-MAX_ERRNO) {
+ trace_cachefiles_io_error(object, file_inode(file), pos,
+ cachefiles_trace_seek_error);
+ return pos;
+ }
+ if ((u64)pos >= (u64)*_start + *_len)
+ return 0; /* Fully allocated */
+
+ /* Partially allocated, but insufficient space: cull. */
+ fscache_count_no_write_space();
+ ret = cachefiles_inject_remove_error();
+ if (ret == 0)
+ ret = vfs_fallocate(file, FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE,
+ *_start, *_len);
+ if (ret < 0) {
+ trace_cachefiles_io_error(object, file_inode(file), ret,
+ cachefiles_trace_fallocate_error);
+ cachefiles_io_error_obj(object,
+ "CacheFiles: fallocate failed (%d)\n", ret);
+ ret = -EIO;
+ }
+
+ return ret;
+
+check_space:
+ return cachefiles_has_space(cache, 0, *_len / PAGE_SIZE,
+ cachefiles_has_space_for_write);
+}
+
+static int cachefiles_prepare_write(struct netfs_cache_resources *cres,
+ loff_t *_start, size_t *_len, loff_t i_size,
+ bool no_space_allocated_yet)
+{
+ struct cachefiles_object *object = cachefiles_cres_object(cres);
+ struct cachefiles_cache *cache = object->volume->cache;
+ const struct cred *saved_cred;
+ int ret;
+
+ if (!cachefiles_cres_file(cres)) {
+ if (!fscache_wait_for_operation(cres, FSCACHE_WANT_WRITE))
+ return -ENOBUFS;
+ if (!cachefiles_cres_file(cres))
+ return -ENOBUFS;
+ }
+
+ cachefiles_begin_secure(cache, &saved_cred);
+ ret = __cachefiles_prepare_write(cres, _start, _len, i_size,
+ no_space_allocated_yet);
+ cachefiles_end_secure(cache, saved_cred);
+ return ret;
}
/*
*/
static void cachefiles_end_operation(struct netfs_cache_resources *cres)
{
- struct fscache_retrieval *op = cres->cache_priv;
- struct file *file = cres->cache_priv2;
-
- _enter("");
+ struct file *file = cachefiles_cres_file(cres);
if (file)
fput(file);
- if (op) {
- fscache_op_complete(&op->op, false);
- fscache_put_retrieval(op);
- }
-
- _leave("");
+ fscache_end_cookie_access(fscache_cres_cookie(cres), fscache_access_io_end);
}
static const struct netfs_cache_ops cachefiles_netfs_cache_ops = {
/*
* Open the cache file when beginning a cache operation.
*/
-int cachefiles_begin_read_operation(struct netfs_read_request *rreq,
- struct fscache_retrieval *op)
+bool cachefiles_begin_operation(struct netfs_cache_resources *cres,
+ enum fscache_want_state want_state)
{
- struct cachefiles_object *object;
- struct cachefiles_cache *cache;
- struct path path;
- struct file *file;
+ struct cachefiles_object *object = cachefiles_cres_object(cres);
+
+ if (!cachefiles_cres_file(cres)) {
+ cres->ops = &cachefiles_netfs_cache_ops;
+ if (object->file) {
+ spin_lock(&object->lock);
+ if (!cres->cache_priv2 && object->file)
+ cres->cache_priv2 = get_file(object->file);
+ spin_unlock(&object->lock);
+ }
+ }
- _enter("");
-
- object = container_of(op->op.object,
- struct cachefiles_object, fscache);
- cache = container_of(object->fscache.cache,
- struct cachefiles_cache, cache);
-
- path.mnt = cache->mnt;
- path.dentry = object->backer;
- file = open_with_fake_path(&path, O_RDWR | O_LARGEFILE | O_DIRECT,
- d_inode(object->backer), cache->cache_cred);
- if (IS_ERR(file))
- return PTR_ERR(file);
- if (!S_ISREG(file_inode(file)->i_mode))
- goto error_file;
- if (unlikely(!file->f_op->read_iter) ||
- unlikely(!file->f_op->write_iter)) {
- pr_notice("Cache does not support read_iter and write_iter\n");
- goto error_file;
+ if (!cachefiles_cres_file(cres) && want_state != FSCACHE_WANT_PARAMS) {
+ pr_err("failed to get cres->file\n");
+ return false;
}
- fscache_get_retrieval(op);
- rreq->cache_resources.cache_priv = op;
- rreq->cache_resources.cache_priv2 = file;
- rreq->cache_resources.ops = &cachefiles_netfs_cache_ops;
- rreq->cache_resources.debug_id = object->fscache.debug_id;
- _leave("");
- return 0;
-
-error_file:
- fput(file);
- return -EIO;
+ return true;
}