* [call for review] netdev-related stuff in vfs.git
@ 2015-04-11 21:17 Al Viro
2015-04-11 21:18 ` [PATCH 01/17] fs: remove ki_nbytes Al Viro
` (16 more replies)
0 siblings, 17 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:17 UTC (permalink / raw)
To: David Miller; +Cc: netdev
There are several commits sitting in vfs.git that probably ought to go in
via net-next.git. First of all, there's merge with vfs.git#iocb - that's
Christoph's aio rework, which has triggered conflicts with the ->sendmsg()
and ->recvmsg() patches a while ago. It's not so much Christoph's stuff
that ought to be in net-next, as (pretty simple) conflict resolution on merge.
The next chunk is switch to {compat_,}import_iovec/import_single_range - new
safer primitives for initializing iov_iter. The primitives themselves come
from vfs/git#iov_iter (and they are used quite a lot in vfs part of queue),
conversion of net/socket.c syscalls belongs in net-next, IMO. Next there's
afs and rxrpc stuff from dhowells. And then there's sanitizing kernel_sendmsg
et.al. + missing inlined helper for "how much data is left in msg->msg_iter" -
this stuff is used in e.g. cifs stuff, but it belongs in net-next.
That pile is pullable from
git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git for-davem
I'll post the individual patches in there in followups; could you take a look
and tell if everything in there is OK with you?
Shortlog:
Al Viro (9):
saner iov_iter initialization primitives
Merge branch 'iocb' into for-davem
Merge branch 'iov_iter' into for-davem
net: switch sendto() and recvfrom() to import_single_range()
net: switch importing msghdr from userland to {compat_,}import_iovec()
switch kernel_sendmsg() and kernel_recvmsg() to iov_iter_kvec()
get rid of the size argument of sock_sendmsg()
Merge remote-tracking branch 'dh/afs' into for-davem
new helper: msg_data_left()
Christoph Hellwig (5):
fs: remove ki_nbytes
fuse: handle synchronous iocbs internally
fs: don't allow to complete sync iocbs through aio_complete
fs: split generic and aio kiocb
fs: move struct kiocb to fs.h
David Howells (5):
RxRPC: Fix the conversion to iov_iter
RxRPC: Don't call skb_add_data() if there's no data to copy
RxRPC: Use iov_iter_count() in rxrpc_send_data() instead of the len argument
AFS: afs_send_empty_reply() doesn't require an iovec array
RxRPC: Handle VERSION Rx protocol packets
Nathaniel Wesley Filardo (1):
kafs: Add more "unified AFS" error codes
Diffstat without #iocb part:
crypto/algif_hash.c | 4 +-
crypto/algif_skcipher.c | 4 +-
drivers/vhost/net.c | 4 +-
fs/afs/misc.c | 16 +++++++-
fs/afs/rxrpc.c | 5 +--
include/linux/net.h | 2 +-
include/linux/socket.h | 5 +++
include/linux/uio.h | 14 +++++++
include/net/compat.h | 2 +-
include/rxrpc/packet.h | 3 +-
lib/iov_iter.c | 57 ++++++++++++++++++++++++++++
net/compat.c | 18 ++++-----
net/core/datagram.c | 2 +-
net/ipv4/tcp.c | 8 ++--
net/rxrpc/ar-input.c | 23 +++++++++++-
net/rxrpc/ar-internal.h | 2 +
net/rxrpc/ar-local.c | 98 +++++++++++++++++++++++++++++++++++++++++++++++++
net/rxrpc/ar-output.c | 53 +++++++++++++-------------
net/socket.c | 96 +++++++++++++++++-------------------------------
net/sunrpc/svcsock.c | 2 +-
20 files changed, 296 insertions(+), 122 deletions(-)
Full diffstat:
arch/s390/hypfs/inode.c | 2 +-
crypto/algif_hash.c | 4 +-
crypto/algif_skcipher.c | 6 +-
drivers/char/mem.c | 2 +-
drivers/char/tile-srom.c | 1 -
drivers/infiniband/hw/ipath/ipath_file_ops.c | 1 -
drivers/infiniband/hw/qib/qib_file_ops.c | 1 -
drivers/misc/mei/amthif.c | 1 -
drivers/misc/mei/main.c | 1 -
drivers/misc/mei/pci-me.c | 1 -
drivers/scsi/sg.c | 2 +-
drivers/staging/unisys/include/timskmod.h | 1 -
drivers/usb/gadget/function/f_fs.c | 6 +-
drivers/usb/gadget/legacy/inode.c | 6 +-
drivers/vhost/net.c | 4 +-
fs/9p/vfs_addr.c | 2 +-
fs/affs/file.c | 2 +-
fs/afs/misc.c | 16 ++-
fs/afs/rxrpc.c | 5 +-
fs/afs/write.c | 1 -
fs/aio.c | 150 +++++++++++++++------------
fs/bfs/inode.c | 1 +
fs/block_dev.c | 1 -
fs/btrfs/file.c | 2 +-
fs/btrfs/inode.c | 2 +-
fs/ceph/file.c | 3 +-
fs/direct-io.c | 5 +-
fs/ecryptfs/file.c | 7 --
fs/ext2/inode.c | 2 +-
fs/ext3/inode.c | 2 +-
fs/ext4/file.c | 2 +-
fs/ext4/indirect.c | 2 +-
fs/ext4/inode.c | 1 -
fs/ext4/page-io.c | 1 -
fs/f2fs/data.c | 2 +-
fs/fat/inode.c | 1 -
fs/fuse/cuse.c | 2 +-
fs/fuse/dev.c | 1 -
fs/fuse/file.c | 55 ++++++----
fs/fuse/fuse_i.h | 1 +
fs/gfs2/aops.c | 2 +-
fs/gfs2/file.c | 1 -
fs/hfs/inode.c | 2 +-
fs/hfsplus/inode.c | 2 +-
fs/jfs/inode.c | 2 +-
fs/nfs/direct.c | 4 +-
fs/nfs/file.c | 1 -
fs/nilfs2/inode.c | 2 +-
fs/ntfs/file.c | 1 -
fs/ntfs/inode.c | 1 -
fs/ocfs2/aops.c | 1 +
fs/ocfs2/aops.h | 2 +-
fs/ocfs2/file.c | 8 +-
fs/pipe.c | 1 -
fs/read_write.c | 35 ++-----
fs/reiserfs/inode.c | 2 +-
fs/splice.c | 1 -
fs/ubifs/file.c | 1 -
fs/udf/file.c | 4 +-
fs/udf/inode.c | 2 +-
fs/xfs/xfs_aops.c | 1 -
fs/xfs/xfs_file.c | 1 -
include/linux/aio.h | 70 +------------
include/linux/fs.h | 22 ++++
include/linux/net.h | 2 +-
include/linux/socket.h | 5 +
include/linux/uio.h | 14 +++
include/net/compat.h | 2 +-
include/net/sock.h | 1 -
include/rxrpc/packet.h | 3 +-
kernel/printk/printk.c | 4 +-
kernel/sysctl.c | 1 +
lib/iov_iter.c | 57 ++++++++++
mm/filemap.c | 1 -
mm/page_io.c | 3 +-
mm/shmem.c | 2 +-
net/compat.c | 18 ++--
net/core/datagram.c | 2 +-
net/ipv4/raw.c | 1 -
net/ipv4/tcp.c | 8 +-
net/rxrpc/ar-input.c | 23 +++-
net/rxrpc/ar-internal.h | 2 +
net/rxrpc/ar-local.c | 98 +++++++++++++++++
net/rxrpc/ar-output.c | 53 +++++-----
net/socket.c | 100 +++++++-----------
net/sunrpc/svcsock.c | 2 +-
sound/core/pcm_native.c | 2 +-
87 files changed, 494 insertions(+), 387 deletions(-)
^ permalink raw reply [flat|nested] 36+ messages in thread
* [PATCH 01/17] fs: remove ki_nbytes
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-11 21:18 ` [PATCH 02/17] fuse: handle synchronous iocbs internally Al Viro
` (15 subsequent siblings)
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: Christoph Hellwig <hch@lst.de>
There is no need to pass the total request length in the kiocb, as
we already get passed in through the iov_iter argument.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
fs/aio.c | 34 ++++++++++++++++++----------------
fs/ceph/file.c | 2 +-
fs/nfs/direct.c | 2 +-
fs/ocfs2/file.c | 8 +++-----
fs/read_write.c | 8 --------
fs/udf/file.c | 2 +-
include/linux/aio.h | 1 -
kernel/printk/printk.c | 2 +-
mm/page_io.c | 1 -
net/socket.c | 6 +++---
10 files changed, 28 insertions(+), 38 deletions(-)
diff --git a/fs/aio.c b/fs/aio.c
index 118a2e0..667054c 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -1344,12 +1344,13 @@ typedef ssize_t (rw_iter_op)(struct kiocb *, struct iov_iter *);
static ssize_t aio_setup_vectored_rw(struct kiocb *kiocb,
int rw, char __user *buf,
unsigned long *nr_segs,
+ size_t *len,
struct iovec **iovec,
bool compat)
{
ssize_t ret;
- *nr_segs = kiocb->ki_nbytes;
+ *nr_segs = *len;
#ifdef CONFIG_COMPAT
if (compat)
@@ -1364,21 +1365,22 @@ static ssize_t aio_setup_vectored_rw(struct kiocb *kiocb,
if (ret < 0)
return ret;
- /* ki_nbytes now reflect bytes instead of segs */
- kiocb->ki_nbytes = ret;
+ /* len now reflect bytes instead of segs */
+ *len = ret;
return 0;
}
static ssize_t aio_setup_single_vector(struct kiocb *kiocb,
int rw, char __user *buf,
unsigned long *nr_segs,
+ size_t len,
struct iovec *iovec)
{
- if (unlikely(!access_ok(!rw, buf, kiocb->ki_nbytes)))
+ if (unlikely(!access_ok(!rw, buf, len)))
return -EFAULT;
iovec->iov_base = buf;
- iovec->iov_len = kiocb->ki_nbytes;
+ iovec->iov_len = len;
*nr_segs = 1;
return 0;
}
@@ -1388,7 +1390,7 @@ static ssize_t aio_setup_single_vector(struct kiocb *kiocb,
* Performs the initial checks and io submission.
*/
static ssize_t aio_run_iocb(struct kiocb *req, unsigned opcode,
- char __user *buf, bool compat)
+ char __user *buf, size_t len, bool compat)
{
struct file *file = req->ki_filp;
ssize_t ret;
@@ -1423,21 +1425,21 @@ rw_common:
if (!rw_op && !iter_op)
return -EINVAL;
- ret = (opcode == IOCB_CMD_PREADV ||
- opcode == IOCB_CMD_PWRITEV)
- ? aio_setup_vectored_rw(req, rw, buf, &nr_segs,
- &iovec, compat)
- : aio_setup_single_vector(req, rw, buf, &nr_segs,
- iovec);
+ if (opcode == IOCB_CMD_PREADV || opcode == IOCB_CMD_PWRITEV)
+ ret = aio_setup_vectored_rw(req, rw, buf, &nr_segs,
+ &len, &iovec, compat);
+ else
+ ret = aio_setup_single_vector(req, rw, buf, &nr_segs,
+ len, iovec);
if (!ret)
- ret = rw_verify_area(rw, file, &req->ki_pos, req->ki_nbytes);
+ ret = rw_verify_area(rw, file, &req->ki_pos, len);
if (ret < 0) {
if (iovec != inline_vecs)
kfree(iovec);
return ret;
}
- req->ki_nbytes = ret;
+ len = ret;
/* XXX: move/kill - rw_verify_area()? */
/* This matches the pread()/pwrite() logic */
@@ -1450,7 +1452,7 @@ rw_common:
file_start_write(file);
if (iter_op) {
- iov_iter_init(&iter, rw, iovec, nr_segs, req->ki_nbytes);
+ iov_iter_init(&iter, rw, iovec, nr_segs, len);
ret = iter_op(req, &iter);
} else {
ret = rw_op(req, iovec, nr_segs, req->ki_pos);
@@ -1553,10 +1555,10 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
req->ki_obj.user = user_iocb;
req->ki_user_data = iocb->aio_data;
req->ki_pos = iocb->aio_offset;
- req->ki_nbytes = iocb->aio_nbytes;
ret = aio_run_iocb(req, iocb->aio_lio_opcode,
(char __user *)(unsigned long)iocb->aio_buf,
+ iocb->aio_nbytes,
compat);
if (ret)
goto out_put_req;
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 905986d..081c4e3 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -807,7 +807,7 @@ static ssize_t ceph_read_iter(struct kiocb *iocb, struct iov_iter *to)
{
struct file *filp = iocb->ki_filp;
struct ceph_file_info *fi = filp->private_data;
- size_t len = iocb->ki_nbytes;
+ size_t len = iov_iter_count(to);
struct inode *inode = file_inode(filp);
struct ceph_inode_info *ci = ceph_inode(inode);
struct page *pinned_page = NULL;
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 7077521..27cebf1 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -265,7 +265,7 @@ ssize_t nfs_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, loff_t
return -EINVAL;
#else
- VM_BUG_ON(iocb->ki_nbytes != PAGE_SIZE);
+ VM_BUG_ON(iov_iter_count(iter) != PAGE_SIZE);
if (rw == READ)
return nfs_file_direct_read(iocb, iter, pos);
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 46e0d4e..266845d 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -2280,7 +2280,7 @@ static ssize_t ocfs2_file_write_iter(struct kiocb *iocb,
file->f_path.dentry->d_name.name,
(unsigned int)from->nr_segs); /* GRRRRR */
- if (iocb->ki_nbytes == 0)
+ if (count == 0)
return 0;
appending = file->f_flags & O_APPEND ? 1 : 0;
@@ -2330,8 +2330,7 @@ relock:
}
can_do_direct = direct_io;
- ret = ocfs2_prepare_inode_for_write(file, ppos,
- iocb->ki_nbytes, appending,
+ ret = ocfs2_prepare_inode_for_write(file, ppos, count, appending,
&can_do_direct, &has_refcount);
if (ret < 0) {
mlog_errno(ret);
@@ -2339,8 +2338,7 @@ relock:
}
if (direct_io && !is_sync_kiocb(iocb))
- unaligned_dio = ocfs2_is_io_unaligned(inode, iocb->ki_nbytes,
- *ppos);
+ unaligned_dio = ocfs2_is_io_unaligned(inode, count, *ppos);
/*
* We can't complete the direct I/O as requested, fall back to
diff --git a/fs/read_write.c b/fs/read_write.c
index 8e1b687..f8b8fc1 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -343,7 +343,6 @@ ssize_t vfs_iter_read(struct file *file, struct iov_iter *iter, loff_t *ppos)
init_sync_kiocb(&kiocb, file);
kiocb.ki_pos = *ppos;
- kiocb.ki_nbytes = iov_iter_count(iter);
iter->type |= READ;
ret = file->f_op->read_iter(&kiocb, iter);
@@ -366,7 +365,6 @@ ssize_t vfs_iter_write(struct file *file, struct iov_iter *iter, loff_t *ppos)
init_sync_kiocb(&kiocb, file);
kiocb.ki_pos = *ppos;
- kiocb.ki_nbytes = iov_iter_count(iter);
iter->type |= WRITE;
ret = file->f_op->write_iter(&kiocb, iter);
@@ -426,7 +424,6 @@ ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *pp
init_sync_kiocb(&kiocb, filp);
kiocb.ki_pos = *ppos;
- kiocb.ki_nbytes = len;
ret = filp->f_op->aio_read(&kiocb, &iov, 1, kiocb.ki_pos);
if (-EIOCBQUEUED == ret)
@@ -446,7 +443,6 @@ ssize_t new_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *p
init_sync_kiocb(&kiocb, filp);
kiocb.ki_pos = *ppos;
- kiocb.ki_nbytes = len;
iov_iter_init(&iter, READ, &iov, 1, len);
ret = filp->f_op->read_iter(&kiocb, &iter);
@@ -510,7 +506,6 @@ ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, lof
init_sync_kiocb(&kiocb, filp);
kiocb.ki_pos = *ppos;
- kiocb.ki_nbytes = len;
ret = filp->f_op->aio_write(&kiocb, &iov, 1, kiocb.ki_pos);
if (-EIOCBQUEUED == ret)
@@ -530,7 +525,6 @@ ssize_t new_sync_write(struct file *filp, const char __user *buf, size_t len, lo
init_sync_kiocb(&kiocb, filp);
kiocb.ki_pos = *ppos;
- kiocb.ki_nbytes = len;
iov_iter_init(&iter, WRITE, &iov, 1, len);
ret = filp->f_op->write_iter(&kiocb, &iter);
@@ -719,7 +713,6 @@ static ssize_t do_iter_readv_writev(struct file *filp, int rw, const struct iove
init_sync_kiocb(&kiocb, filp);
kiocb.ki_pos = *ppos;
- kiocb.ki_nbytes = len;
iov_iter_init(&iter, rw, iov, nr_segs, len);
ret = fn(&kiocb, &iter);
@@ -737,7 +730,6 @@ static ssize_t do_sync_readv_writev(struct file *filp, const struct iovec *iov,
init_sync_kiocb(&kiocb, filp);
kiocb.ki_pos = *ppos;
- kiocb.ki_nbytes = len;
ret = fn(&kiocb, iov, nr_segs, kiocb.ki_pos);
if (ret == -EIOCBQUEUED)
diff --git a/fs/udf/file.c b/fs/udf/file.c
index 08f3555..9c0b6da 100644
--- a/fs/udf/file.c
+++ b/fs/udf/file.c
@@ -122,7 +122,7 @@ static ssize_t udf_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
struct file *file = iocb->ki_filp;
struct inode *inode = file_inode(file);
int err, pos;
- size_t count = iocb->ki_nbytes;
+ size_t count = iov_iter_count(from);
struct udf_inode_info *iinfo = UDF_I(inode);
mutex_lock(&inode->i_mutex);
diff --git a/include/linux/aio.h b/include/linux/aio.h
index d9c92da..132d1ec 100644
--- a/include/linux/aio.h
+++ b/include/linux/aio.h
@@ -42,7 +42,6 @@ struct kiocb {
__u64 ki_user_data; /* user's data for completion */
loff_t ki_pos;
- size_t ki_nbytes; /* copy of iocb->aio_nbytes */
struct list_head ki_list; /* the aio core uses this
* for cancellation */
diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
index c06df7d..60b2aa2 100644
--- a/kernel/printk/printk.c
+++ b/kernel/printk/printk.c
@@ -521,7 +521,7 @@ static ssize_t devkmsg_write(struct kiocb *iocb, struct iov_iter *from)
int i;
int level = default_message_loglevel;
int facility = 1; /* LOG_USER */
- size_t len = iocb->ki_nbytes;
+ size_t len = iov_iter_count(from);
ssize_t ret = len;
if (len > LOG_LINE_MAX)
diff --git a/mm/page_io.c b/mm/page_io.c
index e604580..7ef2157 100644
--- a/mm/page_io.c
+++ b/mm/page_io.c
@@ -274,7 +274,6 @@ int __swap_writepage(struct page *page, struct writeback_control *wbc,
iov_iter_bvec(&from, ITER_BVEC | WRITE, &bv, 1, PAGE_SIZE);
init_sync_kiocb(&kiocb, swap_file);
kiocb.ki_pos = page_file_offset(page);
- kiocb.ki_nbytes = PAGE_SIZE;
set_page_writeback(page);
unlock_page(page);
diff --git a/net/socket.c b/net/socket.c
index bbedbfc..f921455 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -858,11 +858,11 @@ static ssize_t sock_read_iter(struct kiocb *iocb, struct iov_iter *to)
if (iocb->ki_pos != 0)
return -ESPIPE;
- if (iocb->ki_nbytes == 0) /* Match SYS5 behaviour */
+ if (!iov_iter_count(to)) /* Match SYS5 behaviour */
return 0;
res = __sock_recvmsg(iocb, sock, &msg,
- iocb->ki_nbytes, msg.msg_flags);
+ iov_iter_count(to), msg.msg_flags);
*to = msg.msg_iter;
return res;
}
@@ -883,7 +883,7 @@ static ssize_t sock_write_iter(struct kiocb *iocb, struct iov_iter *from)
if (sock->type == SOCK_SEQPACKET)
msg.msg_flags |= MSG_EOR;
- res = __sock_sendmsg(iocb, sock, &msg, iocb->ki_nbytes);
+ res = __sock_sendmsg(iocb, sock, &msg, iov_iter_count(from));
*from = msg.msg_iter;
return res;
}
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 02/17] fuse: handle synchronous iocbs internally
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
2015-04-11 21:18 ` [PATCH 01/17] fs: remove ki_nbytes Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-11 21:18 ` [PATCH 03/17] fs: don't allow to complete sync iocbs through aio_complete Al Viro
` (14 subsequent siblings)
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: Christoph Hellwig <hch@lst.de>
Based on a patch from Maxim Patlasov <MPatlasov@parallels.com>.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
fs/fuse/file.c | 51 +++++++++++++++++++++++++++++++--------------------
fs/fuse/fuse_i.h | 1 +
2 files changed, 32 insertions(+), 20 deletions(-)
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index c01ec3b..f81d83e 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -528,6 +528,17 @@ static void fuse_release_user_pages(struct fuse_req *req, int write)
}
}
+static ssize_t fuse_get_res_by_io(struct fuse_io_priv *io)
+{
+ if (io->err)
+ return io->err;
+
+ if (io->bytes >= 0 && io->write)
+ return -EIO;
+
+ return io->bytes < 0 ? io->size : io->bytes;
+}
+
/**
* In case of short read, the caller sets 'pos' to the position of
* actual end of fuse request in IO request. Otherwise, if bytes_requested
@@ -546,6 +557,7 @@ static void fuse_release_user_pages(struct fuse_req *req, int write)
*/
static void fuse_aio_complete(struct fuse_io_priv *io, int err, ssize_t pos)
{
+ bool is_sync = is_sync_kiocb(io->iocb);
int left;
spin_lock(&io->lock);
@@ -555,27 +567,21 @@ static void fuse_aio_complete(struct fuse_io_priv *io, int err, ssize_t pos)
io->bytes = pos;
left = --io->reqs;
+ if (!left && is_sync)
+ complete(io->done);
spin_unlock(&io->lock);
- if (!left) {
- long res;
+ if (!left && !is_sync) {
+ ssize_t res = fuse_get_res_by_io(io);
- if (io->err)
- res = io->err;
- else if (io->bytes >= 0 && io->write)
- res = -EIO;
- else {
- res = io->bytes < 0 ? io->size : io->bytes;
+ if (res >= 0) {
+ struct inode *inode = file_inode(io->iocb->ki_filp);
+ struct fuse_conn *fc = get_fuse_conn(inode);
+ struct fuse_inode *fi = get_fuse_inode(inode);
- if (!is_sync_kiocb(io->iocb)) {
- struct inode *inode = file_inode(io->iocb->ki_filp);
- struct fuse_conn *fc = get_fuse_conn(inode);
- struct fuse_inode *fi = get_fuse_inode(inode);
-
- spin_lock(&fc->lock);
- fi->attr_version = ++fc->attr_version;
- spin_unlock(&fc->lock);
- }
+ spin_lock(&fc->lock);
+ fi->attr_version = ++fc->attr_version;
+ spin_unlock(&fc->lock);
}
aio_complete(io->iocb, res, 0);
@@ -2801,6 +2807,7 @@ static ssize_t
fuse_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter,
loff_t offset)
{
+ DECLARE_COMPLETION_ONSTACK(wait);
ssize_t ret = 0;
struct file *file = iocb->ki_filp;
struct fuse_file *ff = file->private_data;
@@ -2852,6 +2859,9 @@ fuse_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter,
if (!is_sync_kiocb(iocb) && (offset + count > i_size) && rw == WRITE)
io->async = false;
+ if (io->async && is_sync_kiocb(iocb))
+ io->done = &wait;
+
if (rw == WRITE)
ret = __fuse_direct_write(io, iter, &pos);
else
@@ -2864,11 +2874,12 @@ fuse_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter,
if (!is_sync_kiocb(iocb))
return -EIOCBQUEUED;
- ret = wait_on_sync_kiocb(iocb);
- } else {
- kfree(io);
+ wait_for_completion(&wait);
+ ret = fuse_get_res_by_io(io);
}
+ kfree(io);
+
if (rw == WRITE) {
if (ret > 0)
fuse_write_update_size(inode, pos);
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index 1cdfb07..7354dc1 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -263,6 +263,7 @@ struct fuse_io_priv {
int err;
struct kiocb *iocb;
struct file *file;
+ struct completion *done;
};
/**
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 03/17] fs: don't allow to complete sync iocbs through aio_complete
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
2015-04-11 21:18 ` [PATCH 01/17] fs: remove ki_nbytes Al Viro
2015-04-11 21:18 ` [PATCH 02/17] fuse: handle synchronous iocbs internally Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-14 18:00 ` Tadeusz Struk
2015-04-11 21:18 ` [PATCH 04/17] fs: split generic and aio kiocb Al Viro
` (13 subsequent siblings)
16 siblings, 1 reply; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: Christoph Hellwig <hch@lst.de>
The AIO interface is fairly complex because it tries to allow
filesystems to always work async and then wakeup a synchronous
caller through aio_complete. It turns out that basically no one
was doing this to avoid the complexity and context switches,
and we've already fixed up the remaining users and can now
get rid of this case.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
fs/aio.c | 24 +-----------------------
fs/ecryptfs/file.c | 6 ------
fs/read_write.c | 26 ++++++++------------------
include/linux/aio.h | 4 ----
net/socket.c | 9 +++------
5 files changed, 12 insertions(+), 57 deletions(-)
diff --git a/fs/aio.c b/fs/aio.c
index 667054c..8ca8df1 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -778,22 +778,6 @@ static int kill_ioctx(struct mm_struct *mm, struct kioctx *ctx,
return 0;
}
-/* wait_on_sync_kiocb:
- * Waits on the given sync kiocb to complete.
- */
-ssize_t wait_on_sync_kiocb(struct kiocb *req)
-{
- while (!req->ki_ctx) {
- set_current_state(TASK_UNINTERRUPTIBLE);
- if (req->ki_ctx)
- break;
- io_schedule();
- }
- __set_current_state(TASK_RUNNING);
- return req->ki_user_data;
-}
-EXPORT_SYMBOL(wait_on_sync_kiocb);
-
/*
* exit_aio: called when the last user of mm goes away. At this point, there is
* no way for any new requests to be submited or any of the io_* syscalls to be
@@ -1025,13 +1009,7 @@ void aio_complete(struct kiocb *iocb, long res, long res2)
* ref, no other paths have a way to get another ref
* - the sync task helpfully left a reference to itself in the iocb
*/
- if (is_sync_kiocb(iocb)) {
- iocb->ki_user_data = res;
- smp_wmb();
- iocb->ki_ctx = ERR_PTR(-EXDEV);
- wake_up_process(iocb->ki_obj.tsk);
- return;
- }
+ BUG_ON(is_sync_kiocb(iocb));
if (iocb->ki_list.next) {
unsigned long flags;
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index 6f4e659..a36da88 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -52,12 +52,6 @@ static ssize_t ecryptfs_read_update_atime(struct kiocb *iocb,
struct file *file = iocb->ki_filp;
rc = generic_file_read_iter(iocb, to);
- /*
- * Even though this is a async interface, we need to wait
- * for IO to finish to update atime
- */
- if (-EIOCBQUEUED == rc)
- rc = wait_on_sync_kiocb(iocb);
if (rc >= 0) {
path = ecryptfs_dentry_to_lower_path(file->f_path.dentry);
touch_atime(path);
diff --git a/fs/read_write.c b/fs/read_write.c
index f8b8fc1..76e324e 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -346,9 +346,7 @@ ssize_t vfs_iter_read(struct file *file, struct iov_iter *iter, loff_t *ppos)
iter->type |= READ;
ret = file->f_op->read_iter(&kiocb, iter);
- if (ret == -EIOCBQUEUED)
- ret = wait_on_sync_kiocb(&kiocb);
-
+ BUG_ON(ret == -EIOCBQUEUED);
if (ret > 0)
*ppos = kiocb.ki_pos;
return ret;
@@ -368,9 +366,7 @@ ssize_t vfs_iter_write(struct file *file, struct iov_iter *iter, loff_t *ppos)
iter->type |= WRITE;
ret = file->f_op->write_iter(&kiocb, iter);
- if (ret == -EIOCBQUEUED)
- ret = wait_on_sync_kiocb(&kiocb);
-
+ BUG_ON(ret == -EIOCBQUEUED);
if (ret > 0)
*ppos = kiocb.ki_pos;
return ret;
@@ -426,8 +422,7 @@ ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *pp
kiocb.ki_pos = *ppos;
ret = filp->f_op->aio_read(&kiocb, &iov, 1, kiocb.ki_pos);
- if (-EIOCBQUEUED == ret)
- ret = wait_on_sync_kiocb(&kiocb);
+ BUG_ON(ret == -EIOCBQUEUED);
*ppos = kiocb.ki_pos;
return ret;
}
@@ -446,8 +441,7 @@ ssize_t new_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *p
iov_iter_init(&iter, READ, &iov, 1, len);
ret = filp->f_op->read_iter(&kiocb, &iter);
- if (-EIOCBQUEUED == ret)
- ret = wait_on_sync_kiocb(&kiocb);
+ BUG_ON(ret == -EIOCBQUEUED);
*ppos = kiocb.ki_pos;
return ret;
}
@@ -508,8 +502,7 @@ ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, lof
kiocb.ki_pos = *ppos;
ret = filp->f_op->aio_write(&kiocb, &iov, 1, kiocb.ki_pos);
- if (-EIOCBQUEUED == ret)
- ret = wait_on_sync_kiocb(&kiocb);
+ BUG_ON(ret == -EIOCBQUEUED);
*ppos = kiocb.ki_pos;
return ret;
}
@@ -528,8 +521,7 @@ ssize_t new_sync_write(struct file *filp, const char __user *buf, size_t len, lo
iov_iter_init(&iter, WRITE, &iov, 1, len);
ret = filp->f_op->write_iter(&kiocb, &iter);
- if (-EIOCBQUEUED == ret)
- ret = wait_on_sync_kiocb(&kiocb);
+ BUG_ON(ret == -EIOCBQUEUED);
*ppos = kiocb.ki_pos;
return ret;
}
@@ -716,8 +708,7 @@ static ssize_t do_iter_readv_writev(struct file *filp, int rw, const struct iove
iov_iter_init(&iter, rw, iov, nr_segs, len);
ret = fn(&kiocb, &iter);
- if (ret == -EIOCBQUEUED)
- ret = wait_on_sync_kiocb(&kiocb);
+ BUG_ON(ret == -EIOCBQUEUED);
*ppos = kiocb.ki_pos;
return ret;
}
@@ -732,8 +723,7 @@ static ssize_t do_sync_readv_writev(struct file *filp, const struct iovec *iov,
kiocb.ki_pos = *ppos;
ret = fn(&kiocb, iov, nr_segs, kiocb.ki_pos);
- if (ret == -EIOCBQUEUED)
- ret = wait_on_sync_kiocb(&kiocb);
+ BUG_ON(ret == -EIOCBQUEUED);
*ppos = kiocb.ki_pos;
return ret;
}
diff --git a/include/linux/aio.h b/include/linux/aio.h
index 132d1ec..f851643 100644
--- a/include/linux/aio.h
+++ b/include/linux/aio.h
@@ -37,7 +37,6 @@ struct kiocb {
union {
void __user *user;
- struct task_struct *tsk;
} ki_obj;
__u64 ki_user_data; /* user's data for completion */
@@ -63,13 +62,11 @@ static inline void init_sync_kiocb(struct kiocb *kiocb, struct file *filp)
*kiocb = (struct kiocb) {
.ki_ctx = NULL,
.ki_filp = filp,
- .ki_obj.tsk = current,
};
}
/* prototypes */
#ifdef CONFIG_AIO
-extern ssize_t wait_on_sync_kiocb(struct kiocb *iocb);
extern void aio_complete(struct kiocb *iocb, long res, long res2);
struct mm_struct;
extern void exit_aio(struct mm_struct *mm);
@@ -77,7 +74,6 @@ extern long do_io_submit(aio_context_t ctx_id, long nr,
struct iocb __user *__user *iocbpp, bool compat);
void kiocb_set_cancel_fn(struct kiocb *req, kiocb_cancel_fn *cancel);
#else
-static inline ssize_t wait_on_sync_kiocb(struct kiocb *iocb) { return 0; }
static inline void aio_complete(struct kiocb *iocb, long res, long res2) { }
struct mm_struct;
static inline void exit_aio(struct mm_struct *mm) { }
diff --git a/net/socket.c b/net/socket.c
index f921455..f6c519d 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -633,8 +633,7 @@ static int do_sock_sendmsg(struct socket *sock, struct msghdr *msg,
init_sync_kiocb(&iocb, NULL);
ret = nosec ? __sock_sendmsg_nosec(&iocb, sock, msg, size) :
__sock_sendmsg(&iocb, sock, msg, size);
- if (-EIOCBQUEUED == ret)
- ret = wait_on_sync_kiocb(&iocb);
+ BUG_ON(ret == -EIOCBQUEUED);
return ret;
}
@@ -766,8 +765,7 @@ int sock_recvmsg(struct socket *sock, struct msghdr *msg,
init_sync_kiocb(&iocb, NULL);
ret = __sock_recvmsg(&iocb, sock, msg, size, flags);
- if (-EIOCBQUEUED == ret)
- ret = wait_on_sync_kiocb(&iocb);
+ BUG_ON(ret == -EIOCBQUEUED);
return ret;
}
EXPORT_SYMBOL(sock_recvmsg);
@@ -780,8 +778,7 @@ static int sock_recvmsg_nosec(struct socket *sock, struct msghdr *msg,
init_sync_kiocb(&iocb, NULL);
ret = __sock_recvmsg_nosec(&iocb, sock, msg, size, flags);
- if (-EIOCBQUEUED == ret)
- ret = wait_on_sync_kiocb(&iocb);
+ BUG_ON(ret == -EIOCBQUEUED);
return ret;
}
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 04/17] fs: split generic and aio kiocb
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (2 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 03/17] fs: don't allow to complete sync iocbs through aio_complete Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-11 21:18 ` [PATCH 05/17] fs: move struct kiocb to fs.h Al Viro
` (12 subsequent siblings)
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: Christoph Hellwig <hch@lst.de>
Most callers in the kernel want to perform synchronous file I/O, but
still have to bloat the stack with a full struct kiocb. Split out
the parts needed in filesystem code from those in the aio code, and
only allocate those needed to pass down argument on the stack. The
aio code embedds the generic iocb in the one it allocates and can
easily get back to it by using container_of.
Also add a ->ki_complete method to struct kiocb, this is used to call
into the aio code and thus removes the dependency on aio for filesystems
impementing asynchronous operations. It will also allow other callers
to substitute their own completion callback.
We also add a new ->ki_flags field to work around the nasty layering
violation recently introduced in commit 5e33f6 ("usb: gadget: ffs: add
eventfd notification about ffs events").
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
drivers/usb/gadget/function/f_fs.c | 5 +-
drivers/usb/gadget/legacy/inode.c | 5 +-
fs/aio.c | 94 ++++++++++++++++++++++++++------------
fs/direct-io.c | 4 +-
fs/fuse/file.c | 2 +-
fs/nfs/direct.c | 2 +-
include/linux/aio.h | 46 +++----------------
7 files changed, 81 insertions(+), 77 deletions(-)
diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c
index 175c995..b64538b 100644
--- a/drivers/usb/gadget/function/f_fs.c
+++ b/drivers/usb/gadget/function/f_fs.c
@@ -655,9 +655,10 @@ static void ffs_user_copy_worker(struct work_struct *work)
unuse_mm(io_data->mm);
}
- aio_complete(io_data->kiocb, ret, ret);
+ io_data->kiocb->ki_complete(io_data->kiocb, ret, ret);
- if (io_data->ffs->ffs_eventfd && !io_data->kiocb->ki_eventfd)
+ if (io_data->ffs->ffs_eventfd &&
+ !(io_data->kiocb->ki_flags & IOCB_EVENTFD))
eventfd_signal(io_data->ffs->ffs_eventfd, 1);
usb_ep_free_request(io_data->ep, io_data->req);
diff --git a/drivers/usb/gadget/legacy/inode.c b/drivers/usb/gadget/legacy/inode.c
index 200f9a5..a4a8069 100644
--- a/drivers/usb/gadget/legacy/inode.c
+++ b/drivers/usb/gadget/legacy/inode.c
@@ -469,7 +469,7 @@ static void ep_user_copy_worker(struct work_struct *work)
ret = -EFAULT;
/* completing the iocb can drop the ctx and mm, don't touch mm after */
- aio_complete(iocb, ret, ret);
+ iocb->ki_complete(iocb, ret, ret);
kfree(priv->buf);
kfree(priv->to_free);
@@ -497,7 +497,8 @@ static void ep_aio_complete(struct usb_ep *ep, struct usb_request *req)
kfree(priv);
iocb->private = NULL;
/* aio_complete() reports bytes-transferred _and_ faults */
- aio_complete(iocb, req->actual ? req->actual : req->status,
+
+ iocb->ki_complete(iocb, req->actual ? req->actual : req->status,
req->status);
} else {
/* ep_copy_to_user() won't report both; we hide some faults */
diff --git a/fs/aio.c b/fs/aio.c
index 8ca8df1..9582865 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -151,6 +151,38 @@ struct kioctx {
unsigned id;
};
+/*
+ * We use ki_cancel == KIOCB_CANCELLED to indicate that a kiocb has been either
+ * cancelled or completed (this makes a certain amount of sense because
+ * successful cancellation - io_cancel() - does deliver the completion to
+ * userspace).
+ *
+ * And since most things don't implement kiocb cancellation and we'd really like
+ * kiocb completion to be lockless when possible, we use ki_cancel to
+ * synchronize cancellation and completion - we only set it to KIOCB_CANCELLED
+ * with xchg() or cmpxchg(), see batch_complete_aio() and kiocb_cancel().
+ */
+#define KIOCB_CANCELLED ((void *) (~0ULL))
+
+struct aio_kiocb {
+ struct kiocb common;
+
+ struct kioctx *ki_ctx;
+ kiocb_cancel_fn *ki_cancel;
+
+ struct iocb __user *ki_user_iocb; /* user's aiocb */
+ __u64 ki_user_data; /* user's data for completion */
+
+ struct list_head ki_list; /* the aio core uses this
+ * for cancellation */
+
+ /*
+ * If the aio_resfd field of the userspace iocb is not zero,
+ * this is the underlying eventfd context to deliver events to.
+ */
+ struct eventfd_ctx *ki_eventfd;
+};
+
/*------ sysctl variables----*/
static DEFINE_SPINLOCK(aio_nr_lock);
unsigned long aio_nr; /* current system wide number of aio requests */
@@ -220,7 +252,7 @@ static int __init aio_setup(void)
if (IS_ERR(aio_mnt))
panic("Failed to create aio fs mount.");
- kiocb_cachep = KMEM_CACHE(kiocb, SLAB_HWCACHE_ALIGN|SLAB_PANIC);
+ kiocb_cachep = KMEM_CACHE(aio_kiocb, SLAB_HWCACHE_ALIGN|SLAB_PANIC);
kioctx_cachep = KMEM_CACHE(kioctx,SLAB_HWCACHE_ALIGN|SLAB_PANIC);
pr_debug("sizeof(struct page) = %zu\n", sizeof(struct page));
@@ -480,8 +512,9 @@ static int aio_setup_ring(struct kioctx *ctx)
#define AIO_EVENTS_FIRST_PAGE ((PAGE_SIZE - sizeof(struct aio_ring)) / sizeof(struct io_event))
#define AIO_EVENTS_OFFSET (AIO_EVENTS_PER_PAGE - AIO_EVENTS_FIRST_PAGE)
-void kiocb_set_cancel_fn(struct kiocb *req, kiocb_cancel_fn *cancel)
+void kiocb_set_cancel_fn(struct kiocb *iocb, kiocb_cancel_fn *cancel)
{
+ struct aio_kiocb *req = container_of(iocb, struct aio_kiocb, common);
struct kioctx *ctx = req->ki_ctx;
unsigned long flags;
@@ -496,7 +529,7 @@ void kiocb_set_cancel_fn(struct kiocb *req, kiocb_cancel_fn *cancel)
}
EXPORT_SYMBOL(kiocb_set_cancel_fn);
-static int kiocb_cancel(struct kiocb *kiocb)
+static int kiocb_cancel(struct aio_kiocb *kiocb)
{
kiocb_cancel_fn *old, *cancel;
@@ -514,7 +547,7 @@ static int kiocb_cancel(struct kiocb *kiocb)
cancel = cmpxchg(&kiocb->ki_cancel, old, KIOCB_CANCELLED);
} while (cancel != old);
- return cancel(kiocb);
+ return cancel(&kiocb->common);
}
static void free_ioctx(struct work_struct *work)
@@ -550,13 +583,13 @@ static void free_ioctx_reqs(struct percpu_ref *ref)
static void free_ioctx_users(struct percpu_ref *ref)
{
struct kioctx *ctx = container_of(ref, struct kioctx, users);
- struct kiocb *req;
+ struct aio_kiocb *req;
spin_lock_irq(&ctx->ctx_lock);
while (!list_empty(&ctx->active_reqs)) {
req = list_first_entry(&ctx->active_reqs,
- struct kiocb, ki_list);
+ struct aio_kiocb, ki_list);
list_del_init(&req->ki_list);
kiocb_cancel(req);
@@ -932,9 +965,9 @@ static void user_refill_reqs_available(struct kioctx *ctx)
* Allocate a slot for an aio request.
* Returns NULL if no requests are free.
*/
-static inline struct kiocb *aio_get_req(struct kioctx *ctx)
+static inline struct aio_kiocb *aio_get_req(struct kioctx *ctx)
{
- struct kiocb *req;
+ struct aio_kiocb *req;
if (!get_reqs_available(ctx)) {
user_refill_reqs_available(ctx);
@@ -955,10 +988,10 @@ out_put:
return NULL;
}
-static void kiocb_free(struct kiocb *req)
+static void kiocb_free(struct aio_kiocb *req)
{
- if (req->ki_filp)
- fput(req->ki_filp);
+ if (req->common.ki_filp)
+ fput(req->common.ki_filp);
if (req->ki_eventfd != NULL)
eventfd_ctx_put(req->ki_eventfd);
kmem_cache_free(kiocb_cachep, req);
@@ -994,8 +1027,9 @@ out:
/* aio_complete
* Called when the io request on the given iocb is complete.
*/
-void aio_complete(struct kiocb *iocb, long res, long res2)
+static void aio_complete(struct kiocb *kiocb, long res, long res2)
{
+ struct aio_kiocb *iocb = container_of(kiocb, struct aio_kiocb, common);
struct kioctx *ctx = iocb->ki_ctx;
struct aio_ring *ring;
struct io_event *ev_page, *event;
@@ -1009,7 +1043,7 @@ void aio_complete(struct kiocb *iocb, long res, long res2)
* ref, no other paths have a way to get another ref
* - the sync task helpfully left a reference to itself in the iocb
*/
- BUG_ON(is_sync_kiocb(iocb));
+ BUG_ON(is_sync_kiocb(kiocb));
if (iocb->ki_list.next) {
unsigned long flags;
@@ -1035,7 +1069,7 @@ void aio_complete(struct kiocb *iocb, long res, long res2)
ev_page = kmap_atomic(ctx->ring_pages[pos / AIO_EVENTS_PER_PAGE]);
event = ev_page + pos % AIO_EVENTS_PER_PAGE;
- event->obj = (u64)(unsigned long)iocb->ki_obj.user;
+ event->obj = (u64)(unsigned long)iocb->ki_user_iocb;
event->data = iocb->ki_user_data;
event->res = res;
event->res2 = res2;
@@ -1044,7 +1078,7 @@ void aio_complete(struct kiocb *iocb, long res, long res2)
flush_dcache_page(ctx->ring_pages[pos / AIO_EVENTS_PER_PAGE]);
pr_debug("%p[%u]: %p: %p %Lx %lx %lx\n",
- ctx, tail, iocb, iocb->ki_obj.user, iocb->ki_user_data,
+ ctx, tail, iocb, iocb->ki_user_iocb, iocb->ki_user_data,
res, res2);
/* after flagging the request as done, we
@@ -1091,7 +1125,6 @@ void aio_complete(struct kiocb *iocb, long res, long res2)
percpu_ref_put(&ctx->reqs);
}
-EXPORT_SYMBOL(aio_complete);
/* aio_read_events_ring
* Pull an event off of the ioctx's event ring. Returns the number of
@@ -1480,7 +1513,7 @@ rw_common:
static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
struct iocb *iocb, bool compat)
{
- struct kiocb *req;
+ struct aio_kiocb *req;
ssize_t ret;
/* enforce forwards compatibility on users */
@@ -1503,11 +1536,14 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
if (unlikely(!req))
return -EAGAIN;
- req->ki_filp = fget(iocb->aio_fildes);
- if (unlikely(!req->ki_filp)) {
+ req->common.ki_filp = fget(iocb->aio_fildes);
+ if (unlikely(!req->common.ki_filp)) {
ret = -EBADF;
goto out_put_req;
}
+ req->common.ki_pos = iocb->aio_offset;
+ req->common.ki_complete = aio_complete;
+ req->common.ki_flags = 0;
if (iocb->aio_flags & IOCB_FLAG_RESFD) {
/*
@@ -1522,6 +1558,8 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
req->ki_eventfd = NULL;
goto out_put_req;
}
+
+ req->common.ki_flags |= IOCB_EVENTFD;
}
ret = put_user(KIOCB_KEY, &user_iocb->aio_key);
@@ -1530,11 +1568,10 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
goto out_put_req;
}
- req->ki_obj.user = user_iocb;
+ req->ki_user_iocb = user_iocb;
req->ki_user_data = iocb->aio_data;
- req->ki_pos = iocb->aio_offset;
- ret = aio_run_iocb(req, iocb->aio_lio_opcode,
+ ret = aio_run_iocb(&req->common, iocb->aio_lio_opcode,
(char __user *)(unsigned long)iocb->aio_buf,
iocb->aio_nbytes,
compat);
@@ -1623,10 +1660,10 @@ SYSCALL_DEFINE3(io_submit, aio_context_t, ctx_id, long, nr,
/* lookup_kiocb
* Finds a given iocb for cancellation.
*/
-static struct kiocb *lookup_kiocb(struct kioctx *ctx, struct iocb __user *iocb,
- u32 key)
+static struct aio_kiocb *
+lookup_kiocb(struct kioctx *ctx, struct iocb __user *iocb, u32 key)
{
- struct list_head *pos;
+ struct aio_kiocb *kiocb;
assert_spin_locked(&ctx->ctx_lock);
@@ -1634,9 +1671,8 @@ static struct kiocb *lookup_kiocb(struct kioctx *ctx, struct iocb __user *iocb,
return NULL;
/* TODO: use a hash or array, this sucks. */
- list_for_each(pos, &ctx->active_reqs) {
- struct kiocb *kiocb = list_kiocb(pos);
- if (kiocb->ki_obj.user == iocb)
+ list_for_each_entry(kiocb, &ctx->active_reqs, ki_list) {
+ if (kiocb->ki_user_iocb == iocb)
return kiocb;
}
return NULL;
@@ -1656,7 +1692,7 @@ SYSCALL_DEFINE3(io_cancel, aio_context_t, ctx_id, struct iocb __user *, iocb,
struct io_event __user *, result)
{
struct kioctx *ctx;
- struct kiocb *kiocb;
+ struct aio_kiocb *kiocb;
u32 key;
int ret;
diff --git a/fs/direct-io.c b/fs/direct-io.c
index e181b6b..c38b460 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -265,7 +265,7 @@ static ssize_t dio_complete(struct dio *dio, loff_t offset, ssize_t ret,
ret = err;
}
- aio_complete(dio->iocb, ret, 0);
+ dio->iocb->ki_complete(dio->iocb, ret, 0);
}
kmem_cache_free(dio_cache, dio);
@@ -1056,7 +1056,7 @@ static inline int drop_refcount(struct dio *dio)
* operation. AIO can if it was a broken operation described above or
* in fact if all the bios race to complete before we get here. In
* that case dio_complete() translates the EIOCBQUEUED into the proper
- * return code that the caller will hand to aio_complete().
+ * return code that the caller will hand to ->complete().
*
* This is managed by the bio_lock instead of being an atomic_t so that
* completion paths can drop their ref and use the remaining count to
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index f81d83e..a5c5e38 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -584,7 +584,7 @@ static void fuse_aio_complete(struct fuse_io_priv *io, int err, ssize_t pos)
spin_unlock(&fc->lock);
}
- aio_complete(io->iocb, res, 0);
+ io->iocb->ki_complete(io->iocb, res, 0);
kfree(io);
}
}
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 27cebf1..5db3385 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -393,7 +393,7 @@ static void nfs_direct_complete(struct nfs_direct_req *dreq, bool write)
long res = (long) dreq->error;
if (!res)
res = (long) dreq->count;
- aio_complete(dreq->iocb, res, 0);
+ dreq->iocb->ki_complete(dreq->iocb, res, 0);
}
complete_all(&dreq->completion);
diff --git a/include/linux/aio.h b/include/linux/aio.h
index f851643..5c40b61 100644
--- a/include/linux/aio.h
+++ b/include/linux/aio.h
@@ -14,67 +14,38 @@ struct kiocb;
#define KIOCB_KEY 0
-/*
- * We use ki_cancel == KIOCB_CANCELLED to indicate that a kiocb has been either
- * cancelled or completed (this makes a certain amount of sense because
- * successful cancellation - io_cancel() - does deliver the completion to
- * userspace).
- *
- * And since most things don't implement kiocb cancellation and we'd really like
- * kiocb completion to be lockless when possible, we use ki_cancel to
- * synchronize cancellation and completion - we only set it to KIOCB_CANCELLED
- * with xchg() or cmpxchg(), see batch_complete_aio() and kiocb_cancel().
- */
-#define KIOCB_CANCELLED ((void *) (~0ULL))
-
typedef int (kiocb_cancel_fn)(struct kiocb *);
+#define IOCB_EVENTFD (1 << 0)
+
struct kiocb {
struct file *ki_filp;
- struct kioctx *ki_ctx; /* NULL for sync ops */
- kiocb_cancel_fn *ki_cancel;
- void *private;
-
- union {
- void __user *user;
- } ki_obj;
-
- __u64 ki_user_data; /* user's data for completion */
loff_t ki_pos;
-
- struct list_head ki_list; /* the aio core uses this
- * for cancellation */
-
- /*
- * If the aio_resfd field of the userspace iocb is not zero,
- * this is the underlying eventfd context to deliver events to.
- */
- struct eventfd_ctx *ki_eventfd;
+ void (*ki_complete)(struct kiocb *iocb, long ret, long ret2);
+ void *private;
+ int ki_flags;
};
static inline bool is_sync_kiocb(struct kiocb *kiocb)
{
- return kiocb->ki_ctx == NULL;
+ return kiocb->ki_complete == NULL;
}
static inline void init_sync_kiocb(struct kiocb *kiocb, struct file *filp)
{
*kiocb = (struct kiocb) {
- .ki_ctx = NULL,
.ki_filp = filp,
};
}
/* prototypes */
#ifdef CONFIG_AIO
-extern void aio_complete(struct kiocb *iocb, long res, long res2);
struct mm_struct;
extern void exit_aio(struct mm_struct *mm);
extern long do_io_submit(aio_context_t ctx_id, long nr,
struct iocb __user *__user *iocbpp, bool compat);
void kiocb_set_cancel_fn(struct kiocb *req, kiocb_cancel_fn *cancel);
#else
-static inline void aio_complete(struct kiocb *iocb, long res, long res2) { }
struct mm_struct;
static inline void exit_aio(struct mm_struct *mm) { }
static inline long do_io_submit(aio_context_t ctx_id, long nr,
@@ -84,11 +55,6 @@ static inline void kiocb_set_cancel_fn(struct kiocb *req,
kiocb_cancel_fn *cancel) { }
#endif /* CONFIG_AIO */
-static inline struct kiocb *list_kiocb(struct list_head *h)
-{
- return list_entry(h, struct kiocb, ki_list);
-}
-
/* for sysctl: */
extern unsigned long aio_nr;
extern unsigned long aio_max_nr;
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 05/17] fs: move struct kiocb to fs.h
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (3 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 04/17] fs: split generic and aio kiocb Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-11 21:18 ` [PATCH 06/17] saner iov_iter initialization primitives Al Viro
` (11 subsequent siblings)
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: Christoph Hellwig <hch@lst.de>
struct kiocb now is a generic I/O container, so move it to fs.h.
Also do a #include diet for aio.h while we're at it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
arch/s390/hypfs/inode.c | 2 +-
drivers/char/mem.c | 2 +-
drivers/char/tile-srom.c | 1 -
drivers/infiniband/hw/ipath/ipath_file_ops.c | 1 -
drivers/infiniband/hw/qib/qib_file_ops.c | 1 -
drivers/misc/mei/amthif.c | 1 -
drivers/misc/mei/main.c | 1 -
drivers/misc/mei/pci-me.c | 1 -
drivers/scsi/sg.c | 2 +-
drivers/staging/unisys/include/timskmod.h | 1 -
drivers/usb/gadget/function/f_fs.c | 1 +
drivers/usb/gadget/legacy/inode.c | 1 +
fs/9p/vfs_addr.c | 2 +-
fs/affs/file.c | 2 +-
fs/afs/write.c | 1 -
fs/bfs/inode.c | 1 +
fs/block_dev.c | 1 -
fs/btrfs/file.c | 2 +-
fs/btrfs/inode.c | 2 +-
fs/ceph/file.c | 1 -
fs/direct-io.c | 1 -
fs/ecryptfs/file.c | 1 -
fs/ext2/inode.c | 2 +-
fs/ext3/inode.c | 2 +-
fs/ext4/file.c | 2 +-
fs/ext4/indirect.c | 2 +-
fs/ext4/inode.c | 1 -
fs/ext4/page-io.c | 1 -
fs/f2fs/data.c | 2 +-
fs/fat/inode.c | 1 -
fs/fuse/cuse.c | 2 +-
fs/fuse/dev.c | 1 -
fs/fuse/file.c | 2 +-
fs/gfs2/aops.c | 2 +-
fs/gfs2/file.c | 1 -
fs/hfs/inode.c | 2 +-
fs/hfsplus/inode.c | 2 +-
fs/jfs/inode.c | 2 +-
fs/nfs/file.c | 1 -
fs/nilfs2/inode.c | 2 +-
fs/ntfs/file.c | 1 -
fs/ntfs/inode.c | 1 -
fs/ocfs2/aops.c | 1 +
fs/ocfs2/aops.h | 2 +-
fs/pipe.c | 1 -
fs/read_write.c | 1 -
fs/reiserfs/inode.c | 2 +-
fs/splice.c | 1 -
fs/ubifs/file.c | 1 -
fs/udf/file.c | 2 +-
fs/udf/inode.c | 2 +-
fs/xfs/xfs_aops.c | 1 -
fs/xfs/xfs_file.c | 1 -
include/linux/aio.h | 31 +---------------------------
include/linux/fs.h | 22 ++++++++++++++++++++
include/net/sock.h | 1 -
kernel/printk/printk.c | 2 +-
kernel/sysctl.c | 1 +
mm/filemap.c | 1 -
mm/page_io.c | 2 +-
mm/shmem.c | 2 +-
net/ipv4/raw.c | 1 -
sound/core/pcm_native.c | 2 +-
63 files changed, 55 insertions(+), 86 deletions(-)
diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
index 4c8008d..ad66b07 100644
--- a/arch/s390/hypfs/inode.c
+++ b/arch/s390/hypfs/inode.c
@@ -21,7 +21,7 @@
#include <linux/module.h>
#include <linux/seq_file.h>
#include <linux/mount.h>
-#include <linux/aio.h>
+#include <linux/uio.h>
#include <asm/ebcdic.h>
#include "hypfs.h"
diff --git a/drivers/char/mem.c b/drivers/char/mem.c
index 297110c..9c4fd7a 100644
--- a/drivers/char/mem.c
+++ b/drivers/char/mem.c
@@ -26,7 +26,7 @@
#include <linux/pfn.h>
#include <linux/export.h>
#include <linux/io.h>
-#include <linux/aio.h>
+#include <linux/uio.h>
#include <linux/uaccess.h>
diff --git a/drivers/char/tile-srom.c b/drivers/char/tile-srom.c
index 02e76ac..69f6b4a 100644
--- a/drivers/char/tile-srom.c
+++ b/drivers/char/tile-srom.c
@@ -27,7 +27,6 @@
#include <linux/types.h> /* size_t */
#include <linux/proc_fs.h>
#include <linux/fcntl.h> /* O_ACCMODE */
-#include <linux/aio.h>
#include <linux/pagemap.h>
#include <linux/hugetlb.h>
#include <linux/uaccess.h>
diff --git a/drivers/infiniband/hw/ipath/ipath_file_ops.c b/drivers/infiniband/hw/ipath/ipath_file_ops.c
index 6d7f453..aed8afe 100644
--- a/drivers/infiniband/hw/ipath/ipath_file_ops.c
+++ b/drivers/infiniband/hw/ipath/ipath_file_ops.c
@@ -40,7 +40,6 @@
#include <linux/slab.h>
#include <linux/highmem.h>
#include <linux/io.h>
-#include <linux/aio.h>
#include <linux/jiffies.h>
#include <linux/cpu.h>
#include <asm/pgtable.h>
diff --git a/drivers/infiniband/hw/qib/qib_file_ops.c b/drivers/infiniband/hw/qib/qib_file_ops.c
index b15e34e..826c17e 100644
--- a/drivers/infiniband/hw/qib/qib_file_ops.c
+++ b/drivers/infiniband/hw/qib/qib_file_ops.c
@@ -39,7 +39,6 @@
#include <linux/vmalloc.h>
#include <linux/highmem.h>
#include <linux/io.h>
-#include <linux/aio.h>
#include <linux/jiffies.h>
#include <asm/pgtable.h>
#include <linux/delay.h>
diff --git a/drivers/misc/mei/amthif.c b/drivers/misc/mei/amthif.c
index c4cb9a9..40ea639 100644
--- a/drivers/misc/mei/amthif.c
+++ b/drivers/misc/mei/amthif.c
@@ -19,7 +19,6 @@
#include <linux/errno.h>
#include <linux/types.h>
#include <linux/fcntl.h>
-#include <linux/aio.h>
#include <linux/ioctl.h>
#include <linux/cdev.h>
#include <linux/list.h>
diff --git a/drivers/misc/mei/main.c b/drivers/misc/mei/main.c
index 3c019c0..47680c8 100644
--- a/drivers/misc/mei/main.c
+++ b/drivers/misc/mei/main.c
@@ -22,7 +22,6 @@
#include <linux/errno.h>
#include <linux/types.h>
#include <linux/fcntl.h>
-#include <linux/aio.h>
#include <linux/poll.h>
#include <linux/init.h>
#include <linux/ioctl.h>
diff --git a/drivers/misc/mei/pci-me.c b/drivers/misc/mei/pci-me.c
index bd3039a..af44ee2 100644
--- a/drivers/misc/mei/pci-me.c
+++ b/drivers/misc/mei/pci-me.c
@@ -21,7 +21,6 @@
#include <linux/errno.h>
#include <linux/types.h>
#include <linux/fcntl.h>
-#include <linux/aio.h>
#include <linux/pci.h>
#include <linux/poll.h>
#include <linux/ioctl.h>
diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c
index 0cbc1fb..c78a6f7 100644
--- a/drivers/scsi/sg.c
+++ b/drivers/scsi/sg.c
@@ -33,7 +33,6 @@ static int sg_version_num = 30536; /* 2 digits for each component */
#include <linux/sched.h>
#include <linux/string.h>
#include <linux/mm.h>
-#include <linux/aio.h>
#include <linux/errno.h>
#include <linux/mtio.h>
#include <linux/ioctl.h>
@@ -51,6 +50,7 @@ static int sg_version_num = 30536; /* 2 digits for each component */
#include <linux/mutex.h>
#include <linux/atomic.h>
#include <linux/ratelimit.h>
+#include <linux/uio.h>
#include "scsi.h"
#include <scsi/scsi_dbg.h>
diff --git a/drivers/staging/unisys/include/timskmod.h b/drivers/staging/unisys/include/timskmod.h
index 4019a0d..52648d4 100644
--- a/drivers/staging/unisys/include/timskmod.h
+++ b/drivers/staging/unisys/include/timskmod.h
@@ -46,7 +46,6 @@
#include <linux/module.h>
#include <linux/moduleparam.h>
#include <linux/fcntl.h>
-#include <linux/aio.h>
#include <linux/workqueue.h>
#include <linux/kthread.h>
#include <linux/seq_file.h>
diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c
index b64538b..a12315a78 100644
--- a/drivers/usb/gadget/function/f_fs.c
+++ b/drivers/usb/gadget/function/f_fs.c
@@ -23,6 +23,7 @@
#include <linux/export.h>
#include <linux/hid.h>
#include <linux/module.h>
+#include <linux/uio.h>
#include <asm/unaligned.h>
#include <linux/usb/composite.h>
diff --git a/drivers/usb/gadget/legacy/inode.c b/drivers/usb/gadget/legacy/inode.c
index a4a8069..662ef2c 100644
--- a/drivers/usb/gadget/legacy/inode.c
+++ b/drivers/usb/gadget/legacy/inode.c
@@ -26,6 +26,7 @@
#include <linux/poll.h>
#include <linux/mmu_context.h>
#include <linux/aio.h>
+#include <linux/uio.h>
#include <linux/device.h>
#include <linux/moduleparam.h>
diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c
index eb14e05..ff1a5ba 100644
--- a/fs/9p/vfs_addr.c
+++ b/fs/9p/vfs_addr.c
@@ -33,7 +33,7 @@
#include <linux/pagemap.h>
#include <linux/idr.h>
#include <linux/sched.h>
-#include <linux/aio.h>
+#include <linux/uio.h>
#include <net/9p/9p.h>
#include <net/9p/client.h>
diff --git a/fs/affs/file.c b/fs/affs/file.c
index d2468bf..33eaa67 100644
--- a/fs/affs/file.c
+++ b/fs/affs/file.c
@@ -12,7 +12,7 @@
* affs regular file handling primitives
*/
-#include <linux/aio.h>
+#include <linux/uio.h>
#include "affs.h"
static struct buffer_head *affs_get_extblock_slow(struct inode *inode, u32 ext);
diff --git a/fs/afs/write.c b/fs/afs/write.c
index c13cb08..0714abc 100644
--- a/fs/afs/write.c
+++ b/fs/afs/write.c
@@ -14,7 +14,6 @@
#include <linux/pagemap.h>
#include <linux/writeback.h>
#include <linux/pagevec.h>
-#include <linux/aio.h>
#include "internal.h"
static int afs_write_back_from_locked_page(struct afs_writeback *wb,
diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
index 90bc079..fdcb4d6 100644
--- a/fs/bfs/inode.c
+++ b/fs/bfs/inode.c
@@ -15,6 +15,7 @@
#include <linux/buffer_head.h>
#include <linux/vfs.h>
#include <linux/writeback.h>
+#include <linux/uio.h>
#include <asm/uaccess.h>
#include "bfs.h"
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 975266b..2e522ae 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -27,7 +27,6 @@
#include <linux/namei.h>
#include <linux/log2.h>
#include <linux/cleancache.h>
-#include <linux/aio.h>
#include <asm/uaccess.h>
#include "internal.h"
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index b78bbba..69c9508 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -24,7 +24,6 @@
#include <linux/string.h>
#include <linux/backing-dev.h>
#include <linux/mpage.h>
-#include <linux/aio.h>
#include <linux/falloc.h>
#include <linux/swap.h>
#include <linux/writeback.h>
@@ -32,6 +31,7 @@
#include <linux/compat.h>
#include <linux/slab.h>
#include <linux/btrfs.h>
+#include <linux/uio.h>
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 54bcf63..b214ab1 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -32,7 +32,6 @@
#include <linux/writeback.h>
#include <linux/statfs.h>
#include <linux/compat.h>
-#include <linux/aio.h>
#include <linux/bit_spinlock.h>
#include <linux/xattr.h>
#include <linux/posix_acl.h>
@@ -43,6 +42,7 @@
#include <linux/btrfs.h>
#include <linux/blkdev.h>
#include <linux/posix_acl_xattr.h>
+#include <linux/uio.h>
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 081c4e3..98e257c 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -7,7 +7,6 @@
#include <linux/mount.h>
#include <linux/namei.h>
#include <linux/writeback.h>
-#include <linux/aio.h>
#include <linux/falloc.h>
#include "super.h"
diff --git a/fs/direct-io.c b/fs/direct-io.c
index c38b460..6fb00e3 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -37,7 +37,6 @@
#include <linux/uio.h>
#include <linux/atomic.h>
#include <linux/prefetch.h>
-#include <linux/aio.h>
/*
* How many user pages to map in one call to get_user_pages(). This determines
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index a36da88..273d36e 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -31,7 +31,6 @@
#include <linux/security.h>
#include <linux/compat.h>
#include <linux/fs_stack.h>
-#include <linux/aio.h>
#include "ecryptfs_kernel.h"
/**
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index 6434bc0..df9d6af 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -31,7 +31,7 @@
#include <linux/mpage.h>
#include <linux/fiemap.h>
#include <linux/namei.h>
-#include <linux/aio.h>
+#include <linux/uio.h>
#include "ext2.h"
#include "acl.h"
#include "xattr.h"
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 2c6ccc4..db07ffb 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -27,7 +27,7 @@
#include <linux/writeback.h>
#include <linux/mpage.h>
#include <linux/namei.h>
-#include <linux/aio.h>
+#include <linux/uio.h>
#include "ext3.h"
#include "xattr.h"
#include "acl.h"
diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index 33a09da..598abbb 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -23,9 +23,9 @@
#include <linux/jbd2.h>
#include <linux/mount.h>
#include <linux/path.h>
-#include <linux/aio.h>
#include <linux/quotaops.h>
#include <linux/pagevec.h>
+#include <linux/uio.h>
#include "ext4.h"
#include "ext4_jbd2.h"
#include "xattr.h"
diff --git a/fs/ext4/indirect.c b/fs/ext4/indirect.c
index 6b9878a..8611640 100644
--- a/fs/ext4/indirect.c
+++ b/fs/ext4/indirect.c
@@ -20,9 +20,9 @@
* (sct@redhat.com), 1993, 1998
*/
-#include <linux/aio.h>
#include "ext4_jbd2.h"
#include "truncate.h"
+#include <linux/uio.h>
#include <trace/events/ext4.h>
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 85404f1..6325d2c 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -37,7 +37,6 @@
#include <linux/printk.h>
#include <linux/slab.h>
#include <linux/ratelimit.h>
-#include <linux/aio.h>
#include <linux/bitops.h>
#include "ext4_jbd2.h"
diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c
index b24a254..4649842 100644
--- a/fs/ext4/page-io.c
+++ b/fs/ext4/page-io.c
@@ -18,7 +18,6 @@
#include <linux/pagevec.h>
#include <linux/mpage.h>
#include <linux/namei.h>
-#include <linux/aio.h>
#include <linux/uio.h>
#include <linux/bio.h>
#include <linux/workqueue.h>
diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 985ed02..497f851 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -12,12 +12,12 @@
#include <linux/f2fs_fs.h>
#include <linux/buffer_head.h>
#include <linux/mpage.h>
-#include <linux/aio.h>
#include <linux/writeback.h>
#include <linux/backing-dev.h>
#include <linux/blkdev.h>
#include <linux/bio.h>
#include <linux/prefetch.h>
+#include <linux/uio.h>
#include "f2fs.h"
#include "node.h"
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 497c7c5..8521207 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -19,7 +19,6 @@
#include <linux/mpage.h>
#include <linux/buffer_head.h>
#include <linux/mount.h>
-#include <linux/aio.h>
#include <linux/vfs.h>
#include <linux/parser.h>
#include <linux/uio.h>
diff --git a/fs/fuse/cuse.c b/fs/fuse/cuse.c
index 28d0c7a..b3fa050 100644
--- a/fs/fuse/cuse.c
+++ b/fs/fuse/cuse.c
@@ -38,7 +38,6 @@
#include <linux/device.h>
#include <linux/file.h>
#include <linux/fs.h>
-#include <linux/aio.h>
#include <linux/kdev_t.h>
#include <linux/kthread.h>
#include <linux/list.h>
@@ -48,6 +47,7 @@
#include <linux/slab.h>
#include <linux/stat.h>
#include <linux/module.h>
+#include <linux/uio.h>
#include "fuse_i.h"
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index ed19a7d..8c92c72 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -19,7 +19,6 @@
#include <linux/pipe_fs_i.h>
#include <linux/swap.h>
#include <linux/splice.h>
-#include <linux/aio.h>
MODULE_ALIAS_MISCDEV(FUSE_MINOR);
MODULE_ALIAS("devname:fuse");
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index a5c5e38..ff102cb 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -15,8 +15,8 @@
#include <linux/module.h>
#include <linux/compat.h>
#include <linux/swap.h>
-#include <linux/aio.h>
#include <linux/falloc.h>
+#include <linux/uio.h>
static const struct file_operations fuse_direct_io_file_operations;
diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c
index 4ad4f94..fe6634d 100644
--- a/fs/gfs2/aops.c
+++ b/fs/gfs2/aops.c
@@ -20,7 +20,7 @@
#include <linux/swap.h>
#include <linux/gfs2_ondisk.h>
#include <linux/backing-dev.h>
-#include <linux/aio.h>
+#include <linux/uio.h>
#include <trace/events/writeback.h>
#include "gfs2.h"
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 3e32bb8..f6fc412 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -25,7 +25,6 @@
#include <asm/uaccess.h>
#include <linux/dlm.h>
#include <linux/dlm_plock.h>
-#include <linux/aio.h>
#include <linux/delay.h>
#include "gfs2.h"
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index d0929bc..98d4ea4 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -14,7 +14,7 @@
#include <linux/pagemap.h>
#include <linux/mpage.h>
#include <linux/sched.h>
-#include <linux/aio.h>
+#include <linux/uio.h>
#include "hfs_fs.h"
#include "btree.h"
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index 0cf786f..f541196 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -14,7 +14,7 @@
#include <linux/pagemap.h>
#include <linux/mpage.h>
#include <linux/sched.h>
-#include <linux/aio.h>
+#include <linux/uio.h>
#include "hfsplus_fs.h"
#include "hfsplus_raw.h"
diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c
index bd3df1c..3197aed 100644
--- a/fs/jfs/inode.c
+++ b/fs/jfs/inode.c
@@ -22,8 +22,8 @@
#include <linux/buffer_head.h>
#include <linux/pagemap.h>
#include <linux/quotaops.h>
+#include <linux/uio.h>
#include <linux/writeback.h>
-#include <linux/aio.h>
#include "jfs_incore.h"
#include "jfs_inode.h"
#include "jfs_filsys.h"
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 94712fc..5d8b89c 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -26,7 +26,6 @@
#include <linux/nfs_mount.h>
#include <linux/mm.h>
#include <linux/pagemap.h>
-#include <linux/aio.h>
#include <linux/gfp.h>
#include <linux/swap.h>
diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c
index 8b59695..ab4987b 100644
--- a/fs/nilfs2/inode.c
+++ b/fs/nilfs2/inode.c
@@ -26,7 +26,7 @@
#include <linux/mpage.h>
#include <linux/pagemap.h>
#include <linux/writeback.h>
-#include <linux/aio.h>
+#include <linux/uio.h>
#include "nilfs.h"
#include "btnode.h"
#include "segment.h"
diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c
index 1da9b2d..f16f2d8 100644
--- a/fs/ntfs/file.c
+++ b/fs/ntfs/file.c
@@ -28,7 +28,6 @@
#include <linux/swap.h>
#include <linux/uio.h>
#include <linux/writeback.h>
-#include <linux/aio.h>
#include <asm/page.h>
#include <asm/uaccess.h>
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index 898b994..1d0c21d 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -28,7 +28,6 @@
#include <linux/quotaops.h>
#include <linux/slab.h>
#include <linux/log2.h>
-#include <linux/aio.h>
#include "aops.h"
#include "attrib.h"
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index 44db180..e1bf18c 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -29,6 +29,7 @@
#include <linux/mpage.h>
#include <linux/quotaops.h>
#include <linux/blkdev.h>
+#include <linux/uio.h>
#include <cluster/masklog.h>
diff --git a/fs/ocfs2/aops.h b/fs/ocfs2/aops.h
index 6cae155..dd59599 100644
--- a/fs/ocfs2/aops.h
+++ b/fs/ocfs2/aops.h
@@ -22,7 +22,7 @@
#ifndef OCFS2_AOPS_H
#define OCFS2_AOPS_H
-#include <linux/aio.h>
+#include <linux/fs.h>
handle_t *ocfs2_start_walk_page_trans(struct inode *inode,
struct page *page,
diff --git a/fs/pipe.c b/fs/pipe.c
index 21981e5..2d084f2 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -21,7 +21,6 @@
#include <linux/audit.h>
#include <linux/syscalls.h>
#include <linux/fcntl.h>
-#include <linux/aio.h>
#include <asm/uaccess.h>
#include <asm/ioctls.h>
diff --git a/fs/read_write.c b/fs/read_write.c
index 76e324e..99a6ef9 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -9,7 +9,6 @@
#include <linux/fcntl.h>
#include <linux/file.h>
#include <linux/uio.h>
-#include <linux/aio.h>
#include <linux/fsnotify.h>
#include <linux/security.h>
#include <linux/export.h>
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index e72401e..9312b78 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -18,7 +18,7 @@
#include <linux/writeback.h>
#include <linux/quotaops.h>
#include <linux/swap.h>
-#include <linux/aio.h>
+#include <linux/uio.h>
int reiserfs_commit_write(struct file *f, struct page *page,
unsigned from, unsigned to);
diff --git a/fs/splice.c b/fs/splice.c
index 7968da9..4bbfa95 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -32,7 +32,6 @@
#include <linux/gfp.h>
#include <linux/socket.h>
#include <linux/compat.h>
-#include <linux/aio.h>
#include "internal.h"
/*
diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index e627c0a..c3d15fe 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -50,7 +50,6 @@
*/
#include "ubifs.h"
-#include <linux/aio.h>
#include <linux/mount.h>
#include <linux/namei.h>
#include <linux/slab.h>
diff --git a/fs/udf/file.c b/fs/udf/file.c
index 9c0b6da..7f885cc 100644
--- a/fs/udf/file.c
+++ b/fs/udf/file.c
@@ -34,7 +34,7 @@
#include <linux/errno.h>
#include <linux/pagemap.h>
#include <linux/buffer_head.h>
-#include <linux/aio.h>
+#include <linux/uio.h>
#include "udf_i.h"
#include "udf_sb.h"
diff --git a/fs/udf/inode.c b/fs/udf/inode.c
index a445d59..9c1fbd2 100644
--- a/fs/udf/inode.c
+++ b/fs/udf/inode.c
@@ -38,7 +38,7 @@
#include <linux/slab.h>
#include <linux/crc-itu-t.h>
#include <linux/mpage.h>
-#include <linux/aio.h>
+#include <linux/uio.h>
#include "udf_i.h"
#include "udf_sb.h"
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index 3a9b7a1..4f8cdc5 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -31,7 +31,6 @@
#include "xfs_bmap.h"
#include "xfs_bmap_util.h"
#include "xfs_bmap_btree.h"
-#include <linux/aio.h>
#include <linux/gfp.h>
#include <linux/mpage.h>
#include <linux/pagevec.h>
diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index 1cdba95..f527618 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c
@@ -37,7 +37,6 @@
#include "xfs_log.h"
#include "xfs_icache.h"
-#include <linux/aio.h>
#include <linux/dcache.h>
#include <linux/falloc.h>
#include <linux/pagevec.h>
diff --git a/include/linux/aio.h b/include/linux/aio.h
index 5c40b61..9eb42db 100644
--- a/include/linux/aio.h
+++ b/include/linux/aio.h
@@ -1,52 +1,23 @@
#ifndef __LINUX__AIO_H
#define __LINUX__AIO_H
-#include <linux/list.h>
-#include <linux/workqueue.h>
#include <linux/aio_abi.h>
-#include <linux/uio.h>
-#include <linux/rcupdate.h>
-
-#include <linux/atomic.h>
struct kioctx;
struct kiocb;
+struct mm_struct;
#define KIOCB_KEY 0
typedef int (kiocb_cancel_fn)(struct kiocb *);
-#define IOCB_EVENTFD (1 << 0)
-
-struct kiocb {
- struct file *ki_filp;
- loff_t ki_pos;
- void (*ki_complete)(struct kiocb *iocb, long ret, long ret2);
- void *private;
- int ki_flags;
-};
-
-static inline bool is_sync_kiocb(struct kiocb *kiocb)
-{
- return kiocb->ki_complete == NULL;
-}
-
-static inline void init_sync_kiocb(struct kiocb *kiocb, struct file *filp)
-{
- *kiocb = (struct kiocb) {
- .ki_filp = filp,
- };
-}
-
/* prototypes */
#ifdef CONFIG_AIO
-struct mm_struct;
extern void exit_aio(struct mm_struct *mm);
extern long do_io_submit(aio_context_t ctx_id, long nr,
struct iocb __user *__user *iocbpp, bool compat);
void kiocb_set_cancel_fn(struct kiocb *req, kiocb_cancel_fn *cancel);
#else
-struct mm_struct;
static inline void exit_aio(struct mm_struct *mm) { }
static inline long do_io_submit(aio_context_t ctx_id, long nr,
struct iocb __user * __user *iocbpp,
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 447932a..48c1472 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -314,6 +314,28 @@ struct page;
struct address_space;
struct writeback_control;
+#define IOCB_EVENTFD (1 << 0)
+
+struct kiocb {
+ struct file *ki_filp;
+ loff_t ki_pos;
+ void (*ki_complete)(struct kiocb *iocb, long ret, long ret2);
+ void *private;
+ int ki_flags;
+};
+
+static inline bool is_sync_kiocb(struct kiocb *kiocb)
+{
+ return kiocb->ki_complete == NULL;
+}
+
+static inline void init_sync_kiocb(struct kiocb *kiocb, struct file *filp)
+{
+ *kiocb = (struct kiocb) {
+ .ki_filp = filp,
+ };
+}
+
/*
* "descriptor" for what we're up to with a read.
* This allows us to use the same read code yet
diff --git a/include/net/sock.h b/include/net/sock.h
index ab186b1..71c1300 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -57,7 +57,6 @@
#include <linux/page_counter.h>
#include <linux/memcontrol.h>
#include <linux/static_key.h>
-#include <linux/aio.h>
#include <linux/sched.h>
#include <linux/filter.h>
diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
index 60b2aa2..40d50cc 100644
--- a/kernel/printk/printk.c
+++ b/kernel/printk/printk.c
@@ -32,7 +32,6 @@
#include <linux/security.h>
#include <linux/bootmem.h>
#include <linux/memblock.h>
-#include <linux/aio.h>
#include <linux/syscalls.h>
#include <linux/kexec.h>
#include <linux/kdb.h>
@@ -46,6 +45,7 @@
#include <linux/irq_work.h>
#include <linux/utsname.h>
#include <linux/ctype.h>
+#include <linux/uio.h>
#include <asm/uaccess.h>
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 88ea2d6..83d907a 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -19,6 +19,7 @@
*/
#include <linux/module.h>
+#include <linux/aio.h>
#include <linux/mm.h>
#include <linux/swap.h>
#include <linux/slab.h>
diff --git a/mm/filemap.c b/mm/filemap.c
index ad72420..876f4e6 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -13,7 +13,6 @@
#include <linux/compiler.h>
#include <linux/fs.h>
#include <linux/uaccess.h>
-#include <linux/aio.h>
#include <linux/capability.h>
#include <linux/kernel_stat.h>
#include <linux/gfp.h>
diff --git a/mm/page_io.c b/mm/page_io.c
index 7ef2157..a96c856 100644
--- a/mm/page_io.c
+++ b/mm/page_io.c
@@ -20,8 +20,8 @@
#include <linux/buffer_head.h>
#include <linux/writeback.h>
#include <linux/frontswap.h>
-#include <linux/aio.h>
#include <linux/blkdev.h>
+#include <linux/uio.h>
#include <asm/pgtable.h>
static struct bio *get_swap_bio(gfp_t gfp_flags,
diff --git a/mm/shmem.c b/mm/shmem.c
index a63031f..944b940 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -31,7 +31,7 @@
#include <linux/mm.h>
#include <linux/export.h>
#include <linux/swap.h>
-#include <linux/aio.h>
+#include <linux/uio.h>
static struct vfsmount *shm_mnt;
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index f027a70..4a356b7 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -46,7 +46,6 @@
#include <linux/stddef.h>
#include <linux/slab.h>
#include <linux/errno.h>
-#include <linux/aio.h>
#include <linux/kernel.h>
#include <linux/export.h>
#include <linux/spinlock.h>
diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c
index b03a638..9ecff24 100644
--- a/sound/core/pcm_native.c
+++ b/sound/core/pcm_native.c
@@ -25,7 +25,6 @@
#include <linux/slab.h>
#include <linux/time.h>
#include <linux/pm_qos.h>
-#include <linux/aio.h>
#include <linux/io.h>
#include <linux/dma-mapping.h>
#include <sound/core.h>
@@ -35,6 +34,7 @@
#include <sound/pcm_params.h>
#include <sound/timer.h>
#include <sound/minors.h>
+#include <linux/uio.h>
/*
* Compatibility
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 06/17] saner iov_iter initialization primitives
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (4 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 05/17] fs: move struct kiocb to fs.h Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-11 21:18 ` [PATCH 07/17] RxRPC: Fix the conversion to iov_iter Al Viro
` (10 subsequent siblings)
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: Al Viro <viro@zeniv.linux.org.uk>
iovec-backed iov_iter instances are assumed to satisfy several properties:
* no more than UIO_MAXIOV elements in iovec array
* total size of all ranges is no more than MAX_RW_COUNT
* all ranges pass access_ok().
The problem is, invariants of data structures should be established in the
primitives creating those data structures, not in the code using those
primitives. And iov_iter_init() violates that principle. For a while we
managed to get away with that, but once the use of iov_iter started to
spread, it didn't take long for shit to hit the fan - missed check in
sys_sendto() had introduced a roothole.
We _do_ have primitives for importing and validating iovecs (both native and
compat ones) and those primitives are almost always followed by shoving the
resulting iovec into iov_iter. Life would be considerably simpler (and safer)
if we combined those primitives with initializing iov_iter.
That gives us two new primitives - import_iovec() and compat_import_iovec().
Calling conventions:
iovec = iov_array;
err = import_iovec(direction, uvec, nr_segs,
ARRAY_SIZE(iov_array), &iovec,
&iter);
imports user vector into kernel space (into iov_array if it fits, allocated
if it doesn't fit or if iovec was NULL), validates it and sets iter up to
refer to it. On success 0 is returned and allocated kernel copy (or NULL
if the array had fit into caller-supplied one) is returned via iovec.
On failure all allocations are undone and -E... is returned. If the total
size of ranges exceeds MAX_RW_COUNT, the excess is silently truncated.
compat_import_iovec() expects uvec to be a pointer to user array of compat_iovec;
otherwise it's identical to import_iovec().
Finally, import_single_range() sets iov_iter backed by single-element iovec
covering a user-supplied range -
err = import_single_range(direction, address, size, iovec, &iter);
does validation and sets iter up. Again, size in excess of MAX_RW_COUNT gets
silently truncated.
Next commits will be switching the things up to use of those and reducing
the amount of iov_iter_init() instances.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
include/linux/uio.h | 14 +++++++++++++
lib/iov_iter.c | 57 +++++++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 71 insertions(+)
diff --git a/include/linux/uio.h b/include/linux/uio.h
index 7188029..1f4a37f 100644
--- a/include/linux/uio.h
+++ b/include/linux/uio.h
@@ -139,4 +139,18 @@ static inline void iov_iter_reexpand(struct iov_iter *i, size_t count)
size_t csum_and_copy_to_iter(void *addr, size_t bytes, __wsum *csum, struct iov_iter *i);
size_t csum_and_copy_from_iter(void *addr, size_t bytes, __wsum *csum, struct iov_iter *i);
+int import_iovec(int type, const struct iovec __user * uvector,
+ unsigned nr_segs, unsigned fast_segs,
+ struct iovec **iov, struct iov_iter *i);
+
+#ifdef CONFIG_COMPAT
+struct compat_iovec;
+int compat_import_iovec(int type, const struct compat_iovec __user * uvector,
+ unsigned nr_segs, unsigned fast_segs,
+ struct iovec **iov, struct iov_iter *i);
+#endif
+
+int import_single_range(int type, void __user *buf, size_t len,
+ struct iovec *iov, struct iov_iter *i);
+
#endif
diff --git a/lib/iov_iter.c b/lib/iov_iter.c
index 9d96e283..fc6e33f 100644
--- a/lib/iov_iter.c
+++ b/lib/iov_iter.c
@@ -766,3 +766,60 @@ const void *dup_iter(struct iov_iter *new, struct iov_iter *old, gfp_t flags)
flags);
}
EXPORT_SYMBOL(dup_iter);
+
+int import_iovec(int type, const struct iovec __user * uvector,
+ unsigned nr_segs, unsigned fast_segs,
+ struct iovec **iov, struct iov_iter *i)
+{
+ ssize_t n;
+ struct iovec *p;
+ n = rw_copy_check_uvector(type, uvector, nr_segs, fast_segs,
+ *iov, &p);
+ if (n < 0) {
+ if (p != *iov)
+ kfree(p);
+ *iov = NULL;
+ return n;
+ }
+ iov_iter_init(i, type, p, nr_segs, n);
+ *iov = p == *iov ? NULL : p;
+ return 0;
+}
+EXPORT_SYMBOL(import_iovec);
+
+#ifdef CONFIG_COMPAT
+#include <linux/compat.h>
+
+int compat_import_iovec(int type, const struct compat_iovec __user * uvector,
+ unsigned nr_segs, unsigned fast_segs,
+ struct iovec **iov, struct iov_iter *i)
+{
+ ssize_t n;
+ struct iovec *p;
+ n = compat_rw_copy_check_uvector(type, uvector, nr_segs, fast_segs,
+ *iov, &p);
+ if (n < 0) {
+ if (p != *iov)
+ kfree(p);
+ *iov = NULL;
+ return n;
+ }
+ iov_iter_init(i, type, p, nr_segs, n);
+ *iov = p == *iov ? NULL : p;
+ return 0;
+}
+#endif
+
+int import_single_range(int rw, void __user *buf, size_t len,
+ struct iovec *iov, struct iov_iter *i)
+{
+ if (len > MAX_RW_COUNT)
+ len = MAX_RW_COUNT;
+ if (unlikely(!access_ok(!rw, buf, len)))
+ return -EFAULT;
+
+ iov->iov_base = buf;
+ iov->iov_len = len;
+ iov_iter_init(i, rw, iov, 1, len);
+ return 0;
+}
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 07/17] RxRPC: Fix the conversion to iov_iter
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (5 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 06/17] saner iov_iter initialization primitives Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-11 21:18 ` [PATCH 08/17] RxRPC: Don't call skb_add_data() if there's no data to copy Al Viro
` (9 subsequent siblings)
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: David Howells <dhowells@redhat.com>
This commit:
commit af2b040e470b470bfc881981db3c796072853eae
Author: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu Nov 27 21:44:24 2014 -0500
Subject: rxrpc: switch rxrpc_send_data() to iov_iter primitives
incorrectly changes a do-while loop into a while loop in rxrpc_send_data().
Unfortunately, at least one pass through the loop is required - even if
there is no data - so that the packet the closes the send phase can be
sent if MSG_MORE is not set.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/ar-output.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/rxrpc/ar-output.c b/net/rxrpc/ar-output.c
index 8331c95..833a33b 100644
--- a/net/rxrpc/ar-output.c
+++ b/net/rxrpc/ar-output.c
@@ -548,7 +548,7 @@ static int rxrpc_send_data(struct kiocb *iocb,
copied = 0;
if (len > iov_iter_count(&msg->msg_iter))
len = iov_iter_count(&msg->msg_iter);
- while (len) {
+ do {
int copy;
if (!skb) {
@@ -689,7 +689,7 @@ static int rxrpc_send_data(struct kiocb *iocb,
rxrpc_queue_packet(call, skb, !iov_iter_count(&msg->msg_iter) && !more);
skb = NULL;
}
- }
+ } while (len > 0);
success:
ret = copied;
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 08/17] RxRPC: Don't call skb_add_data() if there's no data to copy
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (6 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 07/17] RxRPC: Fix the conversion to iov_iter Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-11 21:18 ` [PATCH 09/17] RxRPC: Use iov_iter_count() in rxrpc_send_data() instead of the len argument Al Viro
` (8 subsequent siblings)
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: David Howells <dhowells@redhat.com>
Don't call skb_add_data() in rxrpc_send_data() if there's no data to copy and
also skip the calculations associated with it in such a case.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/ar-output.c | 38 +++++++++++++++++++-------------------
1 file changed, 19 insertions(+), 19 deletions(-)
diff --git a/net/rxrpc/ar-output.c b/net/rxrpc/ar-output.c
index 833a33b..f48dc1a 100644
--- a/net/rxrpc/ar-output.c
+++ b/net/rxrpc/ar-output.c
@@ -549,8 +549,6 @@ static int rxrpc_send_data(struct kiocb *iocb,
if (len > iov_iter_count(&msg->msg_iter))
len = iov_iter_count(&msg->msg_iter);
do {
- int copy;
-
if (!skb) {
size_t size, chunk, max, space;
@@ -616,23 +614,25 @@ static int rxrpc_send_data(struct kiocb *iocb,
sp = rxrpc_skb(skb);
/* append next segment of data to the current buffer */
- copy = skb_tailroom(skb);
- ASSERTCMP(copy, >, 0);
- if (copy > len)
- copy = len;
- if (copy > sp->remain)
- copy = sp->remain;
-
- _debug("add");
- ret = skb_add_data(skb, &msg->msg_iter, copy);
- _debug("added");
- if (ret < 0)
- goto efault;
- sp->remain -= copy;
- skb->mark += copy;
- copied += copy;
-
- len -= copy;
+ if (len > 0) {
+ int copy = skb_tailroom(skb);
+ ASSERTCMP(copy, >, 0);
+ if (copy > len)
+ copy = len;
+ if (copy > sp->remain)
+ copy = sp->remain;
+
+ _debug("add");
+ ret = skb_add_data(skb, &msg->msg_iter, copy);
+ _debug("added");
+ if (ret < 0)
+ goto efault;
+ sp->remain -= copy;
+ skb->mark += copy;
+ copied += copy;
+
+ len -= copy;
+ }
/* check for the far side aborting the call or a network error
* occurring */
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 09/17] RxRPC: Use iov_iter_count() in rxrpc_send_data() instead of the len argument
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (7 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 08/17] RxRPC: Don't call skb_add_data() if there's no data to copy Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-11 21:18 ` [PATCH 10/17] AFS: afs_send_empty_reply() doesn't require an iovec array Al Viro
` (7 subsequent siblings)
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: David Howells <dhowells@redhat.com>
Use iov_iter_count() in rxrpc_send_data() to get the remaining data length
instead of using the len argument as the len argument is now redundant.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/ar-output.c | 24 +++++++++++-------------
1 file changed, 11 insertions(+), 13 deletions(-)
diff --git a/net/rxrpc/ar-output.c b/net/rxrpc/ar-output.c
index f48dc1a..de8d2f1 100644
--- a/net/rxrpc/ar-output.c
+++ b/net/rxrpc/ar-output.c
@@ -546,8 +546,6 @@ static int rxrpc_send_data(struct kiocb *iocb,
call->tx_pending = NULL;
copied = 0;
- if (len > iov_iter_count(&msg->msg_iter))
- len = iov_iter_count(&msg->msg_iter);
do {
if (!skb) {
size_t size, chunk, max, space;
@@ -570,8 +568,8 @@ static int rxrpc_send_data(struct kiocb *iocb,
max &= ~(call->conn->size_align - 1UL);
chunk = max;
- if (chunk > len && !more)
- chunk = len;
+ if (chunk > iov_iter_count(&msg->msg_iter) && !more)
+ chunk = iov_iter_count(&msg->msg_iter);
space = chunk + call->conn->size_align;
space &= ~(call->conn->size_align - 1UL);
@@ -614,11 +612,11 @@ static int rxrpc_send_data(struct kiocb *iocb,
sp = rxrpc_skb(skb);
/* append next segment of data to the current buffer */
- if (len > 0) {
+ if (iov_iter_count(&msg->msg_iter) > 0) {
int copy = skb_tailroom(skb);
ASSERTCMP(copy, >, 0);
- if (copy > len)
- copy = len;
+ if (copy > iov_iter_count(&msg->msg_iter))
+ copy = iov_iter_count(&msg->msg_iter);
if (copy > sp->remain)
copy = sp->remain;
@@ -630,8 +628,6 @@ static int rxrpc_send_data(struct kiocb *iocb,
sp->remain -= copy;
skb->mark += copy;
copied += copy;
-
- len -= copy;
}
/* check for the far side aborting the call or a network error
@@ -640,7 +636,8 @@ static int rxrpc_send_data(struct kiocb *iocb,
goto call_aborted;
/* add the packet to the send queue if it's now full */
- if (sp->remain <= 0 || (!len && !more)) {
+ if (sp->remain <= 0 ||
+ (iov_iter_count(&msg->msg_iter) == 0 && !more)) {
struct rxrpc_connection *conn = call->conn;
uint32_t seq;
size_t pad;
@@ -670,7 +667,7 @@ static int rxrpc_send_data(struct kiocb *iocb,
sp->hdr.serviceId = conn->service_id;
sp->hdr.flags = conn->out_clientflag;
- if (len == 0 && !more)
+ if (iov_iter_count(&msg->msg_iter) == 0 && !more)
sp->hdr.flags |= RXRPC_LAST_PACKET;
else if (CIRC_SPACE(call->acks_head, call->acks_tail,
call->acks_winsz) > 1)
@@ -686,10 +683,11 @@ static int rxrpc_send_data(struct kiocb *iocb,
memcpy(skb->head, &sp->hdr,
sizeof(struct rxrpc_header));
- rxrpc_queue_packet(call, skb, !iov_iter_count(&msg->msg_iter) && !more);
+ rxrpc_queue_packet(call, skb,
+ iov_iter_count(&msg->msg_iter) == 0 && !more);
skb = NULL;
}
- } while (len > 0);
+ } while (iov_iter_count(&msg->msg_iter) > 0);
success:
ret = copied;
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 10/17] AFS: afs_send_empty_reply() doesn't require an iovec array
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (8 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 09/17] RxRPC: Use iov_iter_count() in rxrpc_send_data() instead of the len argument Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-11 21:18 ` [PATCH 11/17] RxRPC: Handle VERSION Rx protocol packets Al Viro
` (6 subsequent siblings)
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: David Howells <dhowells@redhat.com>
afs_send_empty_reply() doesn't require an iovec array with which to initialise
the msghdr, but can pass NULL instead.
Suggested-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David Howells <dhowells@redhat.com>
---
fs/afs/rxrpc.c | 5 +----
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c
index dbc732e..3a57a1b 100644
--- a/fs/afs/rxrpc.c
+++ b/fs/afs/rxrpc.c
@@ -770,15 +770,12 @@ static int afs_deliver_cm_op_id(struct afs_call *call, struct sk_buff *skb,
void afs_send_empty_reply(struct afs_call *call)
{
struct msghdr msg;
- struct kvec iov[1];
_enter("");
- iov[0].iov_base = NULL;
- iov[0].iov_len = 0;
msg.msg_name = NULL;
msg.msg_namelen = 0;
- iov_iter_kvec(&msg.msg_iter, WRITE | ITER_KVEC, iov, 0, 0); /* WTF? */
+ iov_iter_kvec(&msg.msg_iter, WRITE | ITER_KVEC, NULL, 0, 0);
msg.msg_control = NULL;
msg.msg_controllen = 0;
msg.msg_flags = 0;
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 11/17] RxRPC: Handle VERSION Rx protocol packets
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (9 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 10/17] AFS: afs_send_empty_reply() doesn't require an iovec array Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-11 21:18 ` [PATCH 12/17] kafs: Add more "unified AFS" error codes Al Viro
` (5 subsequent siblings)
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: David Howells <dhowells@redhat.com>
Handle VERSION Rx protocol packets. We should respond to a VERSION packet
with a string indicating the Rx version. This is a maximum of 64 characters
and is padded out to 65 chars with NUL bytes.
Note that other AFS clients use the version request as a NAT keepalive so we
need to handle it rather than returning an abort.
The standard formulation seems to be:
<project> <version> built <yyyy>-<mm>-<dd>
for example:
" OpenAFS 1.6.2 built 2013-05-07 "
(note the three extra spaces) as obtained with:
rxdebug grand.mit.edu -version
from the openafs package.
Signed-off-by: David Howells <dhowells@redhat.com>
---
include/rxrpc/packet.h | 3 +-
net/rxrpc/ar-input.c | 23 +++++++++++-
net/rxrpc/ar-internal.h | 2 +
net/rxrpc/ar-local.c | 98 +++++++++++++++++++++++++++++++++++++++++++++++++
4 files changed, 124 insertions(+), 2 deletions(-)
diff --git a/include/rxrpc/packet.h b/include/rxrpc/packet.h
index f2902ef..4dce116 100644
--- a/include/rxrpc/packet.h
+++ b/include/rxrpc/packet.h
@@ -47,7 +47,8 @@ struct rxrpc_header {
#define RXRPC_PACKET_TYPE_CHALLENGE 6 /* connection security challenge (SRVR->CLNT) */
#define RXRPC_PACKET_TYPE_RESPONSE 7 /* connection secutity response (CLNT->SRVR) */
#define RXRPC_PACKET_TYPE_DEBUG 8 /* debug info request */
-#define RXRPC_N_PACKET_TYPES 9 /* number of packet types (incl type 0) */
+#define RXRPC_PACKET_TYPE_VERSION 13 /* version string request */
+#define RXRPC_N_PACKET_TYPES 14 /* number of packet types (incl type 0) */
uint8_t flags; /* packet flags */
#define RXRPC_CLIENT_INITIATED 0x01 /* signifies a packet generated by a client */
diff --git a/net/rxrpc/ar-input.c b/net/rxrpc/ar-input.c
index 481f89f..4505a69 100644
--- a/net/rxrpc/ar-input.c
+++ b/net/rxrpc/ar-input.c
@@ -28,7 +28,7 @@
const char *rxrpc_pkts[] = {
"?00",
"DATA", "ACK", "BUSY", "ABORT", "ACKALL", "CHALL", "RESP", "DEBUG",
- "?09", "?10", "?11", "?12", "?13", "?14", "?15"
+ "?09", "?10", "?11", "?12", "VERSION", "?14", "?15"
};
/*
@@ -593,6 +593,20 @@ static void rxrpc_post_packet_to_conn(struct rxrpc_connection *conn,
rxrpc_queue_conn(conn);
}
+/*
+ * post endpoint-level events to the local endpoint
+ * - this includes debug and version messages
+ */
+static void rxrpc_post_packet_to_local(struct rxrpc_local *local,
+ struct sk_buff *skb)
+{
+ _enter("%p,%p", local, skb);
+
+ atomic_inc(&local->usage);
+ skb_queue_tail(&local->event_queue, skb);
+ rxrpc_queue_work(&local->event_processor);
+}
+
static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local,
struct sk_buff *skb,
struct rxrpc_skb_priv *sp)
@@ -699,6 +713,11 @@ void rxrpc_data_ready(struct sock *sk)
goto bad_message;
}
+ if (sp->hdr.type == RXRPC_PACKET_TYPE_VERSION) {
+ rxrpc_post_packet_to_local(local, skb);
+ goto out;
+ }
+
if (sp->hdr.type == RXRPC_PACKET_TYPE_DATA &&
(sp->hdr.callNumber == 0 || sp->hdr.seq == 0))
goto bad_message;
@@ -731,6 +750,8 @@ void rxrpc_data_ready(struct sock *sk)
else
goto cant_route_call;
}
+
+out:
rxrpc_put_local(local);
return;
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index ba9fd36..9a4f7a2 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -152,11 +152,13 @@ struct rxrpc_local {
struct work_struct destroyer; /* endpoint destroyer */
struct work_struct acceptor; /* incoming call processor */
struct work_struct rejecter; /* packet reject writer */
+ struct work_struct event_processor; /* endpoint event processor */
struct list_head services; /* services listening on this endpoint */
struct list_head link; /* link in endpoint list */
struct rw_semaphore defrag_sem; /* control re-enablement of IP DF bit */
struct sk_buff_head accept_queue; /* incoming calls awaiting acceptance */
struct sk_buff_head reject_queue; /* packets awaiting rejection */
+ struct sk_buff_head event_queue; /* endpoint event packets awaiting processing */
spinlock_t lock; /* access lock */
rwlock_t services_lock; /* lock for services list */
atomic_t usage;
diff --git a/net/rxrpc/ar-local.c b/net/rxrpc/ar-local.c
index 87f7135..ca904ed 100644
--- a/net/rxrpc/ar-local.c
+++ b/net/rxrpc/ar-local.c
@@ -13,16 +13,22 @@
#include <linux/net.h>
#include <linux/skbuff.h>
#include <linux/slab.h>
+#include <linux/udp.h>
+#include <linux/ip.h>
#include <net/sock.h>
#include <net/af_rxrpc.h>
+#include <generated/utsrelease.h>
#include "ar-internal.h"
+static const char rxrpc_version_string[65] = "linux-" UTS_RELEASE " AF_RXRPC";
+
static LIST_HEAD(rxrpc_locals);
DEFINE_RWLOCK(rxrpc_local_lock);
static DECLARE_RWSEM(rxrpc_local_sem);
static DECLARE_WAIT_QUEUE_HEAD(rxrpc_local_wq);
static void rxrpc_destroy_local(struct work_struct *work);
+static void rxrpc_process_local_events(struct work_struct *work);
/*
* allocate a new local
@@ -37,11 +43,13 @@ struct rxrpc_local *rxrpc_alloc_local(struct sockaddr_rxrpc *srx)
INIT_WORK(&local->destroyer, &rxrpc_destroy_local);
INIT_WORK(&local->acceptor, &rxrpc_accept_incoming_calls);
INIT_WORK(&local->rejecter, &rxrpc_reject_packets);
+ INIT_WORK(&local->event_processor, &rxrpc_process_local_events);
INIT_LIST_HEAD(&local->services);
INIT_LIST_HEAD(&local->link);
init_rwsem(&local->defrag_sem);
skb_queue_head_init(&local->accept_queue);
skb_queue_head_init(&local->reject_queue);
+ skb_queue_head_init(&local->event_queue);
spin_lock_init(&local->lock);
rwlock_init(&local->services_lock);
atomic_set(&local->usage, 1);
@@ -264,10 +272,12 @@ static void rxrpc_destroy_local(struct work_struct *work)
ASSERT(list_empty(&local->services));
ASSERT(!work_pending(&local->acceptor));
ASSERT(!work_pending(&local->rejecter));
+ ASSERT(!work_pending(&local->event_processor));
/* finish cleaning up the local descriptor */
rxrpc_purge_queue(&local->accept_queue);
rxrpc_purge_queue(&local->reject_queue);
+ rxrpc_purge_queue(&local->event_queue);
kernel_sock_shutdown(local->socket, SHUT_RDWR);
sock_release(local->socket);
@@ -308,3 +318,91 @@ void __exit rxrpc_destroy_all_locals(void)
_leave("");
}
+
+/*
+ * Reply to a version request
+ */
+static void rxrpc_send_version_request(struct rxrpc_local *local,
+ struct rxrpc_header *hdr,
+ struct sk_buff *skb)
+{
+ struct sockaddr_in sin;
+ struct msghdr msg;
+ struct kvec iov[2];
+ size_t len;
+ int ret;
+
+ _enter("");
+
+ sin.sin_family = AF_INET;
+ sin.sin_port = udp_hdr(skb)->source;
+ sin.sin_addr.s_addr = ip_hdr(skb)->saddr;
+
+ msg.msg_name = &sin;
+ msg.msg_namelen = sizeof(sin);
+ msg.msg_control = NULL;
+ msg.msg_controllen = 0;
+ msg.msg_flags = 0;
+
+ hdr->seq = 0;
+ hdr->serial = 0;
+ hdr->type = RXRPC_PACKET_TYPE_VERSION;
+ hdr->flags = RXRPC_LAST_PACKET | (~hdr->flags & RXRPC_CLIENT_INITIATED);
+ hdr->userStatus = 0;
+ hdr->_rsvd = 0;
+
+ iov[0].iov_base = hdr;
+ iov[0].iov_len = sizeof(*hdr);
+ iov[1].iov_base = (char *)rxrpc_version_string;
+ iov[1].iov_len = sizeof(rxrpc_version_string);
+
+ len = iov[0].iov_len + iov[1].iov_len;
+
+ _proto("Tx VERSION (reply)");
+
+ ret = kernel_sendmsg(local->socket, &msg, iov, 2, len);
+ if (ret < 0)
+ _debug("sendmsg failed: %d", ret);
+
+ _leave("");
+}
+
+/*
+ * Process event packets targetted at a local endpoint.
+ */
+static void rxrpc_process_local_events(struct work_struct *work)
+{
+ struct rxrpc_local *local = container_of(work, struct rxrpc_local, event_processor);
+ struct sk_buff *skb;
+ char v;
+
+ _enter("");
+
+ atomic_inc(&local->usage);
+
+ while ((skb = skb_dequeue(&local->event_queue))) {
+ struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
+
+ kdebug("{%d},{%u}", local->debug_id, sp->hdr.type);
+
+ switch (sp->hdr.type) {
+ case RXRPC_PACKET_TYPE_VERSION:
+ if (skb_copy_bits(skb, 0, &v, 1) < 0)
+ return;
+ _proto("Rx VERSION { %02x }", v);
+ if (v == 0)
+ rxrpc_send_version_request(local, &sp->hdr, skb);
+ break;
+
+ default:
+ /* Just ignore anything we don't understand */
+ break;
+ }
+
+ rxrpc_put_local(local);
+ rxrpc_free_skb(skb);
+ }
+
+ rxrpc_put_local(local);
+ _leave("");
+}
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 12/17] kafs: Add more "unified AFS" error codes
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (10 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 11/17] RxRPC: Handle VERSION Rx protocol packets Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-11 21:18 ` [PATCH 13/17] net: switch sendto() and recvfrom() to import_single_range() Al Viro
` (4 subsequent siblings)
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: Nathaniel Wesley Filardo <nwf@cs.jhu.edu>
This should cover the set emitted by viced and the volume server.
Signed-off-by: Nathaniel Wesley Filardo <nwf@cs.jhu.edu>
Signed-off-by: David Howells <dhowells@redhat.com>
---
fs/afs/misc.c | 16 +++++++++++++++-
1 file changed, 15 insertions(+), 1 deletion(-)
diff --git a/fs/afs/misc.c b/fs/afs/misc.c
index 0dd4daf..91ea1aa 100644
--- a/fs/afs/misc.c
+++ b/fs/afs/misc.c
@@ -22,9 +22,12 @@
int afs_abort_to_error(u32 abort_code)
{
switch (abort_code) {
+ /* low errno codes inserted into abort namespace */
case 13: return -EACCES;
case 27: return -EFBIG;
case 30: return -EROFS;
+
+ /* VICE "special error" codes; 101 - 111 */
case VSALVAGE: return -EIO;
case VNOVNODE: return -ENOENT;
case VNOVOL: return -ENOMEDIUM;
@@ -36,11 +39,18 @@ int afs_abort_to_error(u32 abort_code)
case VOVERQUOTA: return -EDQUOT;
case VBUSY: return -EBUSY;
case VMOVED: return -ENXIO;
- case 0x2f6df0a: return -EWOULDBLOCK;
+
+ /* Unified AFS error table; ET "uae" == 0x2f6df00 */
+ case 0x2f6df00: return -EPERM;
+ case 0x2f6df01: return -ENOENT;
+ case 0x2f6df04: return -EIO;
+ case 0x2f6df0a: return -EAGAIN;
+ case 0x2f6df0b: return -ENOMEM;
case 0x2f6df0c: return -EACCES;
case 0x2f6df0f: return -EBUSY;
case 0x2f6df10: return -EEXIST;
case 0x2f6df11: return -EXDEV;
+ case 0x2f6df12: return -ENODEV;
case 0x2f6df13: return -ENOTDIR;
case 0x2f6df14: return -EISDIR;
case 0x2f6df15: return -EINVAL;
@@ -54,8 +64,12 @@ int afs_abort_to_error(u32 abort_code)
case 0x2f6df23: return -ENAMETOOLONG;
case 0x2f6df24: return -ENOLCK;
case 0x2f6df26: return -ENOTEMPTY;
+ case 0x2f6df28: return -EWOULDBLOCK;
+ case 0x2f6df69: return -ENOTCONN;
+ case 0x2f6df6c: return -ETIMEDOUT;
case 0x2f6df78: return -EDQUOT;
+ /* RXKAD abort codes; from include/rxrpc/packet.h. ET "RXK" == 0x1260B00 */
case RXKADINCONSISTENCY: return -EPROTO;
case RXKADPACKETSHORT: return -EPROTO;
case RXKADLEVELFAIL: return -EKEYREJECTED;
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 13/17] net: switch sendto() and recvfrom() to import_single_range()
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (11 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 12/17] kafs: Add more "unified AFS" error codes Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-11 21:18 ` [PATCH 14/17] net: switch importing msghdr from userland to {compat_,}import_iovec() Al Viro
` (3 subsequent siblings)
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
net/socket.c | 24 ++++++++----------------
1 file changed, 8 insertions(+), 16 deletions(-)
diff --git a/net/socket.c b/net/socket.c
index 989b1ae..46f0e1d 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -1650,18 +1650,14 @@ SYSCALL_DEFINE6(sendto, int, fd, void __user *, buff, size_t, len,
struct iovec iov;
int fput_needed;
- if (len > INT_MAX)
- len = INT_MAX;
- if (unlikely(!access_ok(VERIFY_READ, buff, len)))
- return -EFAULT;
+ err = import_single_range(WRITE, buff, len, &iov, &msg.msg_iter);
+ if (unlikely(err))
+ return err;
sock = sockfd_lookup_light(fd, &err, &fput_needed);
if (!sock)
goto out;
- iov.iov_base = buff;
- iov.iov_len = len;
msg.msg_name = NULL;
- iov_iter_init(&msg.msg_iter, WRITE, &iov, 1, len);
msg.msg_control = NULL;
msg.msg_controllen = 0;
msg.msg_namelen = 0;
@@ -1675,7 +1671,7 @@ SYSCALL_DEFINE6(sendto, int, fd, void __user *, buff, size_t, len,
if (sock->file->f_flags & O_NONBLOCK)
flags |= MSG_DONTWAIT;
msg.msg_flags = flags;
- err = sock_sendmsg(sock, &msg, len);
+ err = sock_sendmsg(sock, &msg, iov_iter_count(&msg.msg_iter));
out_put:
fput_light(sock->file, fput_needed);
@@ -1710,26 +1706,22 @@ SYSCALL_DEFINE6(recvfrom, int, fd, void __user *, ubuf, size_t, size,
int err, err2;
int fput_needed;
- if (size > INT_MAX)
- size = INT_MAX;
- if (unlikely(!access_ok(VERIFY_WRITE, ubuf, size)))
- return -EFAULT;
+ err = import_single_range(READ, ubuf, size, &iov, &msg.msg_iter);
+ if (unlikely(err))
+ return err;
sock = sockfd_lookup_light(fd, &err, &fput_needed);
if (!sock)
goto out;
msg.msg_control = NULL;
msg.msg_controllen = 0;
- iov.iov_len = size;
- iov.iov_base = ubuf;
- iov_iter_init(&msg.msg_iter, READ, &iov, 1, size);
/* Save some cycles and don't copy the address if not needed */
msg.msg_name = addr ? (struct sockaddr *)&address : NULL;
/* We assume all kernel code knows the size of sockaddr_storage */
msg.msg_namelen = 0;
if (sock->file->f_flags & O_NONBLOCK)
flags |= MSG_DONTWAIT;
- err = sock_recvmsg(sock, &msg, size, flags);
+ err = sock_recvmsg(sock, &msg, iov_iter_count(&msg.msg_iter), flags);
if (err >= 0 && addr != NULL) {
err2 = move_addr_to_user(&address,
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 14/17] net: switch importing msghdr from userland to {compat_,}import_iovec()
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (12 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 13/17] net: switch sendto() and recvfrom() to import_single_range() Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-11 21:18 ` [PATCH 15/17] switch kernel_sendmsg() and kernel_recvmsg() to iov_iter_kvec() Al Viro
` (2 subsequent siblings)
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
include/net/compat.h | 2 +-
net/compat.c | 18 +++++++-----------
net/socket.c | 31 ++++++++++++-------------------
3 files changed, 20 insertions(+), 31 deletions(-)
diff --git a/include/net/compat.h b/include/net/compat.h
index 42a9c84..48103cf 100644
--- a/include/net/compat.h
+++ b/include/net/compat.h
@@ -40,7 +40,7 @@ int compat_sock_get_timestampns(struct sock *, struct timespec __user *);
#define compat_mmsghdr mmsghdr
#endif /* defined(CONFIG_COMPAT) */
-ssize_t get_compat_msghdr(struct msghdr *, struct compat_msghdr __user *,
+int get_compat_msghdr(struct msghdr *, struct compat_msghdr __user *,
struct sockaddr __user **, struct iovec **);
asmlinkage long compat_sys_sendmsg(int, struct compat_msghdr __user *,
unsigned int);
diff --git a/net/compat.c b/net/compat.c
index c4b6b0f..5cfd26a 100644
--- a/net/compat.c
+++ b/net/compat.c
@@ -31,10 +31,10 @@
#include <asm/uaccess.h>
#include <net/compat.h>
-ssize_t get_compat_msghdr(struct msghdr *kmsg,
- struct compat_msghdr __user *umsg,
- struct sockaddr __user **save_addr,
- struct iovec **iov)
+int get_compat_msghdr(struct msghdr *kmsg,
+ struct compat_msghdr __user *umsg,
+ struct sockaddr __user **save_addr,
+ struct iovec **iov)
{
compat_uptr_t uaddr, uiov, tmp3;
compat_size_t nr_segs;
@@ -81,13 +81,9 @@ ssize_t get_compat_msghdr(struct msghdr *kmsg,
kmsg->msg_iocb = NULL;
- err = compat_rw_copy_check_uvector(save_addr ? READ : WRITE,
- compat_ptr(uiov), nr_segs,
- UIO_FASTIOV, *iov, iov);
- if (err >= 0)
- iov_iter_init(&kmsg->msg_iter, save_addr ? READ : WRITE,
- *iov, nr_segs, err);
- return err;
+ return compat_import_iovec(save_addr ? READ : WRITE,
+ compat_ptr(uiov), nr_segs,
+ UIO_FASTIOV, iov, &kmsg->msg_iter);
}
/* Bleech... */
diff --git a/net/socket.c b/net/socket.c
index 46f0e1d..e5669ce 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -1841,10 +1841,10 @@ struct used_address {
unsigned int name_len;
};
-static ssize_t copy_msghdr_from_user(struct msghdr *kmsg,
- struct user_msghdr __user *umsg,
- struct sockaddr __user **save_addr,
- struct iovec **iov)
+static int copy_msghdr_from_user(struct msghdr *kmsg,
+ struct user_msghdr __user *umsg,
+ struct sockaddr __user **save_addr,
+ struct iovec **iov)
{
struct sockaddr __user *uaddr;
struct iovec __user *uiov;
@@ -1890,13 +1890,8 @@ static ssize_t copy_msghdr_from_user(struct msghdr *kmsg,
kmsg->msg_iocb = NULL;
- err = rw_copy_check_uvector(save_addr ? READ : WRITE,
- uiov, nr_segs,
- UIO_FASTIOV, *iov, iov);
- if (err >= 0)
- iov_iter_init(&kmsg->msg_iter, save_addr ? READ : WRITE,
- *iov, nr_segs, err);
- return err;
+ return import_iovec(save_addr ? READ : WRITE, uiov, nr_segs,
+ UIO_FASTIOV, iov, &kmsg->msg_iter);
}
static int ___sys_sendmsg(struct socket *sock, struct user_msghdr __user *msg,
@@ -1921,8 +1916,8 @@ static int ___sys_sendmsg(struct socket *sock, struct user_msghdr __user *msg,
else
err = copy_msghdr_from_user(msg_sys, msg, NULL, &iov);
if (err < 0)
- goto out_freeiov;
- total_len = err;
+ return err;
+ total_len = iov_iter_count(&msg_sys->msg_iter);
err = -ENOBUFS;
@@ -1988,8 +1983,7 @@ out_freectl:
if (ctl_buf != ctl)
sock_kfree_s(sock->sk, ctl_buf, ctl_len);
out_freeiov:
- if (iov != iovstack)
- kfree(iov);
+ kfree(iov);
return err;
}
@@ -2114,8 +2108,8 @@ static int ___sys_recvmsg(struct socket *sock, struct user_msghdr __user *msg,
else
err = copy_msghdr_from_user(msg_sys, msg, &uaddr, &iov);
if (err < 0)
- goto out_freeiov;
- total_len = err;
+ return err;
+ total_len = iov_iter_count(&msg_sys->msg_iter);
cmsg_ptr = (unsigned long)msg_sys->msg_control;
msg_sys->msg_flags = flags & (MSG_CMSG_CLOEXEC|MSG_CMSG_COMPAT);
@@ -2153,8 +2147,7 @@ static int ___sys_recvmsg(struct socket *sock, struct user_msghdr __user *msg,
err = len;
out_freeiov:
- if (iov != iovstack)
- kfree(iov);
+ kfree(iov);
return err;
}
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 15/17] switch kernel_sendmsg() and kernel_recvmsg() to iov_iter_kvec()
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (13 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 14/17] net: switch importing msghdr from userland to {compat_,}import_iovec() Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-14 16:21 ` David Laight
2015-04-11 21:18 ` [PATCH 16/17] get rid of the size argument of sock_sendmsg() Al Viro
2015-04-11 21:18 ` [PATCH 17/17] new helper: msg_data_left() Al Viro
16 siblings, 1 reply; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: Al Viro <viro@zeniv.linux.org.uk>
For kernel_sendmsg() that eliminates the need to play with setfs();
for kernel_recvmsg() it does *not* - a couple of callers are using
it with non-NULL ->msg_control, which would be treated as userland
address on recvmsg side of things.
In all cases we are really setting a kvec-backed iov_iter, though.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
net/socket.c | 20 +++-----------------
1 file changed, 3 insertions(+), 17 deletions(-)
diff --git a/net/socket.c b/net/socket.c
index e5669ce..b6ceeda 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -627,18 +627,8 @@ EXPORT_SYMBOL(sock_sendmsg);
int kernel_sendmsg(struct socket *sock, struct msghdr *msg,
struct kvec *vec, size_t num, size_t size)
{
- mm_segment_t oldfs = get_fs();
- int result;
-
- set_fs(KERNEL_DS);
- /*
- * the following is safe, since for compiler definitions of kvec and
- * iovec are identical, yielding the same in-core layout and alignment
- */
- iov_iter_init(&msg->msg_iter, WRITE, (struct iovec *)vec, num, size);
- result = sock_sendmsg(sock, msg, size);
- set_fs(oldfs);
- return result;
+ iov_iter_kvec(&msg->msg_iter, WRITE | ITER_KVEC, vec, num, size);
+ return sock_sendmsg(sock, msg, size);
}
EXPORT_SYMBOL(kernel_sendmsg);
@@ -755,12 +745,8 @@ int kernel_recvmsg(struct socket *sock, struct msghdr *msg,
mm_segment_t oldfs = get_fs();
int result;
+ iov_iter_kvec(&msg->msg_iter, READ | ITER_KVEC, vec, num, size);
set_fs(KERNEL_DS);
- /*
- * the following is safe, since for compiler definitions of kvec and
- * iovec are identical, yielding the same in-core layout and alignment
- */
- iov_iter_init(&msg->msg_iter, READ, (struct iovec *)vec, num, size);
result = sock_recvmsg(sock, msg, size, flags);
set_fs(oldfs);
return result;
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 16/17] get rid of the size argument of sock_sendmsg()
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (14 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 15/17] switch kernel_sendmsg() and kernel_recvmsg() to iov_iter_kvec() Al Viro
@ 2015-04-11 21:18 ` Al Viro
2015-04-14 16:25 ` David Laight
2015-04-11 21:18 ` [PATCH 17/17] new helper: msg_data_left() Al Viro
16 siblings, 1 reply; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: Al Viro <viro@zeniv.linux.org.uk>
it's equal to iov_iter_count(&msg->msg_iter) in all cases
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
include/linux/net.h | 2 +-
net/socket.c | 27 ++++++++++++++-------------
net/sunrpc/svcsock.c | 2 +-
3 files changed, 16 insertions(+), 15 deletions(-)
diff --git a/include/linux/net.h b/include/linux/net.h
index e74114b..738ea48 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -211,7 +211,7 @@ int sock_create(int family, int type, int proto, struct socket **res);
int sock_create_kern(int family, int type, int proto, struct socket **res);
int sock_create_lite(int family, int type, int proto, struct socket **res);
void sock_release(struct socket *sock);
-int sock_sendmsg(struct socket *sock, struct msghdr *msg, size_t len);
+int sock_sendmsg(struct socket *sock, struct msghdr *msg);
int sock_recvmsg(struct socket *sock, struct msghdr *msg, size_t size,
int flags);
struct file *sock_alloc_file(struct socket *sock, int flags, const char *dname);
diff --git a/net/socket.c b/net/socket.c
index b6ceeda..21676e4 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -610,17 +610,19 @@ void __sock_tx_timestamp(const struct sock *sk, __u8 *tx_flags)
}
EXPORT_SYMBOL(__sock_tx_timestamp);
-static inline int sock_sendmsg_nosec(struct socket *sock, struct msghdr *msg,
- size_t size)
+static inline int sock_sendmsg_nosec(struct socket *sock, struct msghdr *msg)
{
- return sock->ops->sendmsg(sock, msg, size);
+ int ret = sock->ops->sendmsg(sock, msg, iov_iter_count(&msg->msg_iter));
+ BUG_ON(ret == -EIOCBQUEUED);
+ return ret;
}
-int sock_sendmsg(struct socket *sock, struct msghdr *msg, size_t size)
+int sock_sendmsg(struct socket *sock, struct msghdr *msg)
{
- int err = security_socket_sendmsg(sock, msg, size);
+ int err = security_socket_sendmsg(sock, msg,
+ iov_iter_count(&msg->msg_iter));
- return err ?: sock_sendmsg_nosec(sock, msg, size);
+ return err ?: sock_sendmsg_nosec(sock, msg);
}
EXPORT_SYMBOL(sock_sendmsg);
@@ -628,7 +630,7 @@ int kernel_sendmsg(struct socket *sock, struct msghdr *msg,
struct kvec *vec, size_t num, size_t size)
{
iov_iter_kvec(&msg->msg_iter, WRITE | ITER_KVEC, vec, num, size);
- return sock_sendmsg(sock, msg, size);
+ return sock_sendmsg(sock, msg);
}
EXPORT_SYMBOL(kernel_sendmsg);
@@ -819,7 +821,7 @@ static ssize_t sock_write_iter(struct kiocb *iocb, struct iov_iter *from)
if (sock->type == SOCK_SEQPACKET)
msg.msg_flags |= MSG_EOR;
- res = sock_sendmsg(sock, &msg, iov_iter_count(from));
+ res = sock_sendmsg(sock, &msg);
*from = msg.msg_iter;
return res;
}
@@ -1657,7 +1659,7 @@ SYSCALL_DEFINE6(sendto, int, fd, void __user *, buff, size_t, len,
if (sock->file->f_flags & O_NONBLOCK)
flags |= MSG_DONTWAIT;
msg.msg_flags = flags;
- err = sock_sendmsg(sock, &msg, iov_iter_count(&msg.msg_iter));
+ err = sock_sendmsg(sock, &msg);
out_put:
fput_light(sock->file, fput_needed);
@@ -1892,7 +1894,7 @@ static int ___sys_sendmsg(struct socket *sock, struct user_msghdr __user *msg,
__attribute__ ((aligned(sizeof(__kernel_size_t))));
/* 20 is size of ipv6_pktinfo */
unsigned char *ctl_buf = ctl;
- int ctl_len, total_len;
+ int ctl_len;
ssize_t err;
msg_sys->msg_name = &address;
@@ -1903,7 +1905,6 @@ static int ___sys_sendmsg(struct socket *sock, struct user_msghdr __user *msg,
err = copy_msghdr_from_user(msg_sys, msg, NULL, &iov);
if (err < 0)
return err;
- total_len = iov_iter_count(&msg_sys->msg_iter);
err = -ENOBUFS;
@@ -1950,10 +1951,10 @@ static int ___sys_sendmsg(struct socket *sock, struct user_msghdr __user *msg,
used_address->name_len == msg_sys->msg_namelen &&
!memcmp(&used_address->name, msg_sys->msg_name,
used_address->name_len)) {
- err = sock_sendmsg_nosec(sock, msg_sys, total_len);
+ err = sock_sendmsg_nosec(sock, msg_sys);
goto out_freectl;
}
- err = sock_sendmsg(sock, msg_sys, total_len);
+ err = sock_sendmsg(sock, msg_sys);
/*
* If this is sendmmsg() and sending to current destination address was
* successful, remember it.
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index cc331b6..0c81202 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -257,7 +257,7 @@ static int svc_sendto(struct svc_rqst *rqstp, struct xdr_buf *xdr)
svc_set_cmsg_data(rqstp, cmh);
- if (sock_sendmsg(sock, &msg, 0) < 0)
+ if (sock_sendmsg(sock, &msg) < 0)
goto out;
}
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* [PATCH 17/17] new helper: msg_data_left()
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
` (15 preceding siblings ...)
2015-04-11 21:18 ` [PATCH 16/17] get rid of the size argument of sock_sendmsg() Al Viro
@ 2015-04-11 21:18 ` Al Viro
16 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-11 21:18 UTC (permalink / raw)
To: davem; +Cc: netdev
From: Al Viro <viro@zeniv.linux.org.uk>
convert open-coded instances
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
crypto/algif_hash.c | 4 ++--
crypto/algif_skcipher.c | 4 ++--
drivers/vhost/net.c | 4 ++--
include/linux/socket.h | 5 +++++
net/core/datagram.c | 2 +-
net/ipv4/tcp.c | 8 ++++----
net/rxrpc/ar-output.c | 19 +++++++++----------
net/socket.c | 4 ++--
8 files changed, 27 insertions(+), 23 deletions(-)
diff --git a/crypto/algif_hash.c b/crypto/algif_hash.c
index 0a465e0..1396ad0 100644
--- a/crypto/algif_hash.c
+++ b/crypto/algif_hash.c
@@ -56,8 +56,8 @@ static int hash_sendmsg(struct socket *sock, struct msghdr *msg,
ctx->more = 0;
- while (iov_iter_count(&msg->msg_iter)) {
- int len = iov_iter_count(&msg->msg_iter);
+ while (msg_data_left(msg)) {
+ int len = msg_data_left(msg);
if (len > limit)
len = limit;
diff --git a/crypto/algif_skcipher.c b/crypto/algif_skcipher.c
index 8f903b6..9450752 100644
--- a/crypto/algif_skcipher.c
+++ b/crypto/algif_skcipher.c
@@ -641,7 +641,7 @@ static int skcipher_recvmsg_sync(struct socket *sock, struct msghdr *msg,
long copied = 0;
lock_sock(sk);
- while (iov_iter_count(&msg->msg_iter)) {
+ while (msg_data_left(msg)) {
sgl = list_first_entry(&ctx->tsgl,
struct skcipher_sg_list, list);
sg = sgl->sg;
@@ -655,7 +655,7 @@ static int skcipher_recvmsg_sync(struct socket *sock, struct msghdr *msg,
goto unlock;
}
- used = min_t(unsigned long, ctx->used, iov_iter_count(&msg->msg_iter));
+ used = min_t(unsigned long, ctx->used, msg_data_left(msg));
used = af_alg_make_sg(&ctx->rsgl, &msg->msg_iter, used);
err = used;
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 18f05bf..7d137a4 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -357,13 +357,13 @@ static void handle_tx(struct vhost_net *net)
iov_iter_init(&msg.msg_iter, WRITE, vq->iov, out, len);
iov_iter_advance(&msg.msg_iter, hdr_size);
/* Sanity check */
- if (!iov_iter_count(&msg.msg_iter)) {
+ if (!msg_data_left(&msg)) {
vq_err(vq, "Unexpected header len for TX: "
"%zd expected %zd\n",
len, hdr_size);
break;
}
- len = iov_iter_count(&msg.msg_iter);
+ len = msg_data_left(&msg);
zcopy_used = zcopy && len >= VHOST_GOODCOPY_LEN
&& (nvq->upend_idx + 1) % UIO_MAXIOV !=
diff --git a/include/linux/socket.h b/include/linux/socket.h
index c9852ef..5bf59c8 100644
--- a/include/linux/socket.h
+++ b/include/linux/socket.h
@@ -139,6 +139,11 @@ static inline struct cmsghdr * cmsg_nxthdr (struct msghdr *__msg, struct cmsghdr
return __cmsg_nxthdr(__msg->msg_control, __msg->msg_controllen, __cmsg);
}
+static inline size_t msg_data_left(struct msghdr *msg)
+{
+ return iov_iter_count(&msg->msg_iter);
+}
+
/* "Socket"-level control message types: */
#define SCM_RIGHTS 0x01 /* rw: access rights (array of int) */
diff --git a/net/core/datagram.c b/net/core/datagram.c
index df493d6..b80fb91 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -673,7 +673,7 @@ int skb_copy_and_csum_datagram_msg(struct sk_buff *skb,
if (!chunk)
return 0;
- if (iov_iter_count(&msg->msg_iter) < chunk) {
+ if (msg_data_left(msg) < chunk) {
if (__skb_checksum_complete(skb))
goto csum_error;
if (skb_copy_datagram_msg(skb, hlen, msg, chunk))
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 094a682..18e3a12 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1119,7 +1119,7 @@ int tcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
sg = !!(sk->sk_route_caps & NETIF_F_SG);
- while (iov_iter_count(&msg->msg_iter)) {
+ while (msg_data_left(msg)) {
int copy = 0;
int max = size_goal;
@@ -1163,8 +1163,8 @@ new_segment:
}
/* Try to append data to the end of skb. */
- if (copy > iov_iter_count(&msg->msg_iter))
- copy = iov_iter_count(&msg->msg_iter);
+ if (copy > msg_data_left(msg))
+ copy = msg_data_left(msg);
/* Where to copy to? */
if (skb_availroom(skb) > 0) {
@@ -1221,7 +1221,7 @@ new_segment:
tcp_skb_pcount_set(skb, 0);
copied += copy;
- if (!iov_iter_count(&msg->msg_iter)) {
+ if (!msg_data_left(msg)) {
tcp_tx_timestamp(sk, skb);
goto out;
}
diff --git a/net/rxrpc/ar-output.c b/net/rxrpc/ar-output.c
index 7a31a39..c004280 100644
--- a/net/rxrpc/ar-output.c
+++ b/net/rxrpc/ar-output.c
@@ -564,8 +564,8 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
max &= ~(call->conn->size_align - 1UL);
chunk = max;
- if (chunk > iov_iter_count(&msg->msg_iter) && !more)
- chunk = iov_iter_count(&msg->msg_iter);
+ if (chunk > msg_data_left(msg) && !more)
+ chunk = msg_data_left(msg);
space = chunk + call->conn->size_align;
space &= ~(call->conn->size_align - 1UL);
@@ -608,11 +608,11 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
sp = rxrpc_skb(skb);
/* append next segment of data to the current buffer */
- if (iov_iter_count(&msg->msg_iter) > 0) {
+ if (msg_data_left(msg) > 0) {
int copy = skb_tailroom(skb);
ASSERTCMP(copy, >, 0);
- if (copy > iov_iter_count(&msg->msg_iter))
- copy = iov_iter_count(&msg->msg_iter);
+ if (copy > msg_data_left(msg))
+ copy = msg_data_left(msg);
if (copy > sp->remain)
copy = sp->remain;
@@ -633,7 +633,7 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
/* add the packet to the send queue if it's now full */
if (sp->remain <= 0 ||
- (iov_iter_count(&msg->msg_iter) == 0 && !more)) {
+ (msg_data_left(msg) == 0 && !more)) {
struct rxrpc_connection *conn = call->conn;
uint32_t seq;
size_t pad;
@@ -663,7 +663,7 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
sp->hdr.serviceId = conn->service_id;
sp->hdr.flags = conn->out_clientflag;
- if (iov_iter_count(&msg->msg_iter) == 0 && !more)
+ if (msg_data_left(msg) == 0 && !more)
sp->hdr.flags |= RXRPC_LAST_PACKET;
else if (CIRC_SPACE(call->acks_head, call->acks_tail,
call->acks_winsz) > 1)
@@ -679,11 +679,10 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
memcpy(skb->head, &sp->hdr,
sizeof(struct rxrpc_header));
- rxrpc_queue_packet(call, skb,
- iov_iter_count(&msg->msg_iter) == 0 && !more);
+ rxrpc_queue_packet(call, skb, !msg_data_left(msg) && !more);
skb = NULL;
}
- } while (iov_iter_count(&msg->msg_iter) > 0);
+ } while (msg_data_left(msg) > 0);
success:
ret = copied;
diff --git a/net/socket.c b/net/socket.c
index 21676e4..5b01262 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -612,7 +612,7 @@ EXPORT_SYMBOL(__sock_tx_timestamp);
static inline int sock_sendmsg_nosec(struct socket *sock, struct msghdr *msg)
{
- int ret = sock->ops->sendmsg(sock, msg, iov_iter_count(&msg->msg_iter));
+ int ret = sock->ops->sendmsg(sock, msg, msg_data_left(msg));
BUG_ON(ret == -EIOCBQUEUED);
return ret;
}
@@ -620,7 +620,7 @@ static inline int sock_sendmsg_nosec(struct socket *sock, struct msghdr *msg)
int sock_sendmsg(struct socket *sock, struct msghdr *msg)
{
int err = security_socket_sendmsg(sock, msg,
- iov_iter_count(&msg->msg_iter));
+ msg_data_left(msg));
return err ?: sock_sendmsg_nosec(sock, msg);
}
--
2.1.4
^ permalink raw reply related [flat|nested] 36+ messages in thread
* RE: [PATCH 15/17] switch kernel_sendmsg() and kernel_recvmsg() to iov_iter_kvec()
2015-04-11 21:18 ` [PATCH 15/17] switch kernel_sendmsg() and kernel_recvmsg() to iov_iter_kvec() Al Viro
@ 2015-04-14 16:21 ` David Laight
2015-04-14 16:34 ` Al Viro
0 siblings, 1 reply; 36+ messages in thread
From: David Laight @ 2015-04-14 16:21 UTC (permalink / raw)
To: 'Al Viro', davem; +Cc: netdev
From: Al Viro
> Sent: 11 April 2015 22:18
> From: Al Viro <viro@zeniv.linux.org.uk>
>
> For kernel_sendmsg() that eliminates the need to play with setfs();
> for kernel_recvmsg() it does *not* - a couple of callers are using
> it with non-NULL ->msg_control, which would be treated as userland
> address on recvmsg side of things.
>
> In all cases we are really setting a kvec-backed iov_iter, though.
>
> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
> ---
> net/socket.c | 20 +++-----------------
> 1 file changed, 3 insertions(+), 17 deletions(-)
>
> diff --git a/net/socket.c b/net/socket.c
> index e5669ce..b6ceeda 100644
> --- a/net/socket.c
> +++ b/net/socket.c
> @@ -627,18 +627,8 @@ EXPORT_SYMBOL(sock_sendmsg);
> int kernel_sendmsg(struct socket *sock, struct msghdr *msg,
> struct kvec *vec, size_t num, size_t size)
> {
> - mm_segment_t oldfs = get_fs();
> - int result;
> -
> - set_fs(KERNEL_DS);
> - /*
> - * the following is safe, since for compiler definitions of kvec and
> - * iovec are identical, yielding the same in-core layout and alignment
> - */
> - iov_iter_init(&msg->msg_iter, WRITE, (struct iovec *)vec, num, size);
> - result = sock_sendmsg(sock, msg, size);
> - set_fs(oldfs);
> - return result;
> + iov_iter_kvec(&msg->msg_iter, WRITE | ITER_KVEC, vec, num, size);
> + return sock_sendmsg(sock, msg, size);
Massive NAK.
This breaks any code that is using msg_control to set SCTP parameters
when sending data.
David
^ permalink raw reply [flat|nested] 36+ messages in thread
* RE: [PATCH 16/17] get rid of the size argument of sock_sendmsg()
2015-04-11 21:18 ` [PATCH 16/17] get rid of the size argument of sock_sendmsg() Al Viro
@ 2015-04-14 16:25 ` David Laight
2015-04-14 16:35 ` Al Viro
0 siblings, 1 reply; 36+ messages in thread
From: David Laight @ 2015-04-14 16:25 UTC (permalink / raw)
To: 'Al Viro', davem; +Cc: netdev
From: Al Viro
> Sent: 11 April 2015 22:18
> it's equal to iov_iter_count(&msg->msg_iter) in all cases
I don't know whether this is guaranteed for iov[] that come from
userspace.
In any case iov_iter_count() is non-trivial and you don't
really want to call it when unnecessary.
David
^ permalink raw reply [flat|nested] 36+ messages in thread
* Re: [PATCH 15/17] switch kernel_sendmsg() and kernel_recvmsg() to iov_iter_kvec()
2015-04-14 16:21 ` David Laight
@ 2015-04-14 16:34 ` Al Viro
2015-04-14 16:36 ` David Laight
0 siblings, 1 reply; 36+ messages in thread
From: Al Viro @ 2015-04-14 16:34 UTC (permalink / raw)
To: David Laight; +Cc: davem, netdev
On Tue, Apr 14, 2015 at 04:21:02PM +0000, David Laight wrote:
> Massive NAK.
> This breaks any code that is using msg_control to set SCTP parameters
> when sending data.
Huh? ->sendmsg() expects ->msg_control already in kernel space;
it's ->recvmsg() that plays silly buggers with userland pointers there.
^ permalink raw reply [flat|nested] 36+ messages in thread
* Re: [PATCH 16/17] get rid of the size argument of sock_sendmsg()
2015-04-14 16:25 ` David Laight
@ 2015-04-14 16:35 ` Al Viro
2015-04-14 16:44 ` Al Viro
2015-04-14 17:55 ` David Miller
0 siblings, 2 replies; 36+ messages in thread
From: Al Viro @ 2015-04-14 16:35 UTC (permalink / raw)
To: David Laight; +Cc: davem, netdev
On Tue, Apr 14, 2015 at 04:25:24PM +0000, David Laight wrote:
> From: Al Viro
> > Sent: 11 April 2015 22:18
> > it's equal to iov_iter_count(&msg->msg_iter) in all cases
>
> I don't know whether this is guaranteed for iov[] that come from
> userspace.
>
> In any case iov_iter_count() is non-trivial and you don't
> really want to call it when unnecessary.
Really?
static inline size_t iov_iter_count(struct iov_iter *i)
{
return i->count;
}
^ permalink raw reply [flat|nested] 36+ messages in thread
* RE: [PATCH 15/17] switch kernel_sendmsg() and kernel_recvmsg() to iov_iter_kvec()
2015-04-14 16:34 ` Al Viro
@ 2015-04-14 16:36 ` David Laight
2015-04-14 16:59 ` Al Viro
0 siblings, 1 reply; 36+ messages in thread
From: David Laight @ 2015-04-14 16:36 UTC (permalink / raw)
To: 'Al Viro'; +Cc: davem, netdev
From: Al Viro
> Sent: 14 April 2015 17:34
> On Tue, Apr 14, 2015 at 04:21:02PM +0000, David Laight wrote:
>
> > Massive NAK.
> > This breaks any code that is using msg_control to set SCTP parameters
> > when sending data.
>
> Huh? ->sendmsg() expects ->msg_control already in kernel space;
> it's ->recvmsg() that plays silly buggers with userland pointers there.
I read your commit message as implying that you hadn't found any
users of kernel_sendmsg() that used msg_control.
Not that the data was always read from kernel space.
David
^ permalink raw reply [flat|nested] 36+ messages in thread
* Re: [PATCH 16/17] get rid of the size argument of sock_sendmsg()
2015-04-14 16:35 ` Al Viro
@ 2015-04-14 16:44 ` Al Viro
2015-04-14 17:55 ` David Miller
1 sibling, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-14 16:44 UTC (permalink / raw)
To: David Laight; +Cc: davem, netdev
On Tue, Apr 14, 2015 at 05:35:34PM +0100, Al Viro wrote:
> On Tue, Apr 14, 2015 at 04:25:24PM +0000, David Laight wrote:
> > From: Al Viro
> > > Sent: 11 April 2015 22:18
> > > it's equal to iov_iter_count(&msg->msg_iter) in all cases
> >
> > I don't know whether this is guaranteed for iov[] that come from
> > userspace.
> >
> > In any case iov_iter_count() is non-trivial and you don't
> > really want to call it when unnecessary.
>
> Really?
>
> static inline size_t iov_iter_count(struct iov_iter *i)
> {
> return i->count;
> }
Incidentally, have you even looked at the patch? sock_write_iter():
- res = sock_sendmsg(sock, &msg, iov_iter_count(from));
+ res = sock_sendmsg(sock, &msg);
sys_sendto():
- err = sock_sendmsg(sock, &msg, iov_iter_count(&msg.msg_iter));
+ err = sock_sendmsg(sock, &msg);
__sys_sendmsg():
- total_len = iov_iter_count(&msg_sys->msg_iter);
- err = sock_sendmsg(sock, msg_sys, total_len);
+ err = sock_sendmsg(sock, msg_sys);
kernel_sendmsg():
iov_iter_kvec(&msg->msg_iter, WRITE | ITER_KVEC, vec, num, size);
- return sock_sendmsg(sock, msg, size);
+ return sock_sendmsg(sock, msg);
... and iov_iter_kvec() sets ->count to its last argument, so after the
first line we'll have iov_iter_count(&msg->msg_iter) == size.
svc_sendto():
- if (sock_sendmsg(sock, &msg, 0) < 0)
+ if (sock_sendmsg(sock, &msg) < 0)
and msg->msg_iter is all-zeroes there, including ->msg_iter.count.
In cases when iov comes from userland we used to have iov_iter_count(...)
passed there; no need to even look its definition up. It's an equivalent
transformation, no matter what...
^ permalink raw reply [flat|nested] 36+ messages in thread
* Re: [PATCH 15/17] switch kernel_sendmsg() and kernel_recvmsg() to iov_iter_kvec()
2015-04-14 16:36 ` David Laight
@ 2015-04-14 16:59 ` Al Viro
2015-04-15 9:08 ` David Laight
0 siblings, 1 reply; 36+ messages in thread
From: Al Viro @ 2015-04-14 16:59 UTC (permalink / raw)
To: David Laight; +Cc: davem, netdev
On Tue, Apr 14, 2015 at 04:36:36PM +0000, David Laight wrote:
> From: Al Viro
> > Sent: 14 April 2015 17:34
> > On Tue, Apr 14, 2015 at 04:21:02PM +0000, David Laight wrote:
> >
> > > Massive NAK.
> > > This breaks any code that is using msg_control to set SCTP parameters
> > > when sending data.
> >
> > Huh? ->sendmsg() expects ->msg_control already in kernel space;
> > it's ->recvmsg() that plays silly buggers with userland pointers there.
>
> I read your commit message as implying that you hadn't found any
> users of kernel_sendmsg() that used msg_control.
> Not that the data was always read from kernel space.
Sigh... The situation is:
* ->sendmsg() expects ->msg_control copied to userland. sendmsg(2),
sendto(2), etc. do that copying. See ___sys_sendmsg() - there we have
/*
* Careful! Before this, msg_sys->msg_control contains a user pointer.
* Afterwards, it will be a kernel pointer. Thus the compiler-assisted
* checking falls down on this.
*/
if (copy_from_user(ctl_buf,
(void __user __force *)msg_sys->msg_control,
ctl_len))
goto out_freectl;
msg_sys->msg_control = ctl_buf;
As the result, ->sendmsg() instances access ->msg_control contents as normal
kernel data.
* ->recvmsg() expects ->msg_control to point to userland. See
net/core/scm.c for the helpers used to store into it. recvmsg(2) et.al.
simply leave the userland pointer there; worse, that pointer might be
to native or to compat variants, and layouts _are_ different. Thus those
if (MSG_CMSG_COMPAT & msg->msg_flags) in net/core/scm.c...
* kernel-side users of ->sendmsg() do not depend on setfs() for
access to their ->msg_control, simply because ->sendmsg() won't be using
copy_from_user()/get_user() to access it anyway.
* kernel-side users of ->recvmsg() are less lucky - most of them
don't give a damn either (they have NULL ->msg_control), but there's an
exception (somewhere in sunrpc, IIRC). So there we need to keep
playing with setfs(), even though the data side would be just fine without
that.
^ permalink raw reply [flat|nested] 36+ messages in thread
* Re: [PATCH 16/17] get rid of the size argument of sock_sendmsg()
2015-04-14 16:35 ` Al Viro
2015-04-14 16:44 ` Al Viro
@ 2015-04-14 17:55 ` David Miller
2015-04-15 8:37 ` David Laight
1 sibling, 1 reply; 36+ messages in thread
From: David Miller @ 2015-04-14 17:55 UTC (permalink / raw)
To: viro; +Cc: David.Laight, netdev
From: Al Viro <viro@ZenIV.linux.org.uk>
Date: Tue, 14 Apr 2015 17:35:34 +0100
> On Tue, Apr 14, 2015 at 04:25:24PM +0000, David Laight wrote:
>> From: Al Viro
>> > Sent: 11 April 2015 22:18
>> > it's equal to iov_iter_count(&msg->msg_iter) in all cases
>>
>> I don't know whether this is guaranteed for iov[] that come from
>> userspace.
>>
>> In any case iov_iter_count() is non-trivial and you don't
>> really want to call it when unnecessary.
>
> Really?
>
> static inline size_t iov_iter_count(struct iov_iter *i)
> {
> return i->count;
> }
This just made my day.
^ permalink raw reply [flat|nested] 36+ messages in thread
* Re: [PATCH 03/17] fs: don't allow to complete sync iocbs through aio_complete
2015-04-11 21:18 ` [PATCH 03/17] fs: don't allow to complete sync iocbs through aio_complete Al Viro
@ 2015-04-14 18:00 ` Tadeusz Struk
2015-04-14 18:26 ` Al Viro
0 siblings, 1 reply; 36+ messages in thread
From: Tadeusz Struk @ 2015-04-14 18:00 UTC (permalink / raw)
To: Al Viro, davem; +Cc: netdev, linux-crypto
On 04/11/2015 02:18 PM, Al Viro wrote:
> @@ -766,8 +765,7 @@ int sock_recvmsg(struct socket *sock, struct msghdr *msg,
>
> init_sync_kiocb(&iocb, NULL);
> ret = __sock_recvmsg(&iocb, sock, msg, size, flags);
> - if (-EIOCBQUEUED == ret)
> - ret = wait_on_sync_kiocb(&iocb);
> + BUG_ON(ret == -EIOCBQUEUED);
NACK - PF_ALG type sockets support async operations and return -EIOCBQUEUED
See skcipher_recvmsg_async() in crypto/algif_skcipher.c in net-next
^ permalink raw reply [flat|nested] 36+ messages in thread
* Re: [PATCH 03/17] fs: don't allow to complete sync iocbs through aio_complete
2015-04-14 18:00 ` Tadeusz Struk
@ 2015-04-14 18:26 ` Al Viro
2015-04-14 18:37 ` Tadeusz Struk
0 siblings, 1 reply; 36+ messages in thread
From: Al Viro @ 2015-04-14 18:26 UTC (permalink / raw)
To: Tadeusz Struk; +Cc: davem, netdev, linux-crypto
On Tue, Apr 14, 2015 at 11:00:49AM -0700, Tadeusz Struk wrote:
> On 04/11/2015 02:18 PM, Al Viro wrote:
> > @@ -766,8 +765,7 @@ int sock_recvmsg(struct socket *sock, struct msghdr *msg,
> >
> > init_sync_kiocb(&iocb, NULL);
> > ret = __sock_recvmsg(&iocb, sock, msg, size, flags);
> > - if (-EIOCBQUEUED == ret)
> > - ret = wait_on_sync_kiocb(&iocb);
> > + BUG_ON(ret == -EIOCBQUEUED);
>
> NACK - PF_ALG type sockets support async operations and return -EIOCBQUEUED
> See skcipher_recvmsg_async() in crypto/algif_skcipher.c in net-next
Its only caller is
static int skcipher_recvmsg(struct socket *sock, struct msghdr *msg,
size_t ignored, int flags)
{
return (msg->msg_iocb && !is_sync_kiocb(msg->msg_iocb)) ?
skcipher_recvmsg_async(sock, msg, flags) :
skcipher_recvmsg_sync(sock, msg, flags);
}
Note that !is_sync_kiocb() in there. Compare with init_sync_kiocb() in
sock_recvmsg()...
^ permalink raw reply [flat|nested] 36+ messages in thread
* Re: [PATCH 03/17] fs: don't allow to complete sync iocbs through aio_complete
2015-04-14 18:26 ` Al Viro
@ 2015-04-14 18:37 ` Tadeusz Struk
2015-04-14 19:22 ` Al Viro
0 siblings, 1 reply; 36+ messages in thread
From: Tadeusz Struk @ 2015-04-14 18:37 UTC (permalink / raw)
To: Al Viro; +Cc: davem, netdev, linux-crypto
On 04/14/2015 11:26 AM, Al Viro wrote:
>>> @@ -766,8 +765,7 @@ int sock_recvmsg(struct socket *sock, struct msghdr *msg,
>>> > >
>>> > > init_sync_kiocb(&iocb, NULL);
>>> > > ret = __sock_recvmsg(&iocb, sock, msg, size, flags);
>>> > > - if (-EIOCBQUEUED == ret)
>>> > > - ret = wait_on_sync_kiocb(&iocb);
>>> > > + BUG_ON(ret == -EIOCBQUEUED);
>> >
>> > NACK - PF_ALG type sockets support async operations and return -EIOCBQUEUED
>> > See skcipher_recvmsg_async() in crypto/algif_skcipher.c in net-next
> Its only caller is
>
> static int skcipher_recvmsg(struct socket *sock, struct msghdr *msg,
> size_t ignored, int flags)
> {
> return (msg->msg_iocb && !is_sync_kiocb(msg->msg_iocb)) ?
> skcipher_recvmsg_async(sock, msg, flags) :
> skcipher_recvmsg_sync(sock, msg, flags);
> }
>
> Note that !is_sync_kiocb() in there. Compare with init_sync_kiocb() in
> sock_recvmsg()...
The only problem is it calls init_sync_kiocb(&iocb, NULL) on a different iocb.
The one that isn't even passed to skcipher_recvmsg()
skcipher_recvmsg() checks is_sync_kiocb(msg->msg_iocb).
I just want to make sure that after the merging window is closed I can still
trigger aio_read on PF_ALG socket.
thanks
^ permalink raw reply [flat|nested] 36+ messages in thread
* Re: [PATCH 03/17] fs: don't allow to complete sync iocbs through aio_complete
2015-04-14 18:37 ` Tadeusz Struk
@ 2015-04-14 19:22 ` Al Viro
0 siblings, 0 replies; 36+ messages in thread
From: Al Viro @ 2015-04-14 19:22 UTC (permalink / raw)
To: Tadeusz Struk; +Cc: davem, netdev, linux-crypto
On Tue, Apr 14, 2015 at 11:37:01AM -0700, Tadeusz Struk wrote:
> The only problem is it calls init_sync_kiocb(&iocb, NULL) on a different iocb.
> The one that isn't even passed to skcipher_recvmsg()
> skcipher_recvmsg() checks is_sync_kiocb(msg->msg_iocb).
> I just want to make sure that after the merging window is closed I can still
> trigger aio_read on PF_ALG socket.
> thanks
... and after the merge with net-next#master that BUG_ON() is gone, so
you are still just fine.
FWIW, in the current for-davem (or in net-next#master, now that Dave has
pulled it) the functions in question are:
static inline int sock_recvmsg_nosec(struct socket *sock, struct msghdr *msg,
size_t size, int flags)
{
return sock->ops->recvmsg(sock, msg, size, flags);
}
int sock_recvmsg(struct socket *sock, struct msghdr *msg, size_t size,
int flags)
{
int err = security_socket_recvmsg(sock, msg, size, flags);
return err ?: sock_recvmsg_nosec(sock, msg, size, flags);
}
EXPORT_SYMBOL(sock_recvmsg);
static ssize_t sock_read_iter(struct kiocb *iocb, struct iov_iter *to)
{
struct file *file = iocb->ki_filp;
struct socket *sock = file->private_data;
struct msghdr msg = {.msg_iter = *to,
.msg_iocb = iocb};
ssize_t res;
if (file->f_flags & O_NONBLOCK)
msg.msg_flags = MSG_DONTWAIT;
if (iocb->ki_pos != 0)
return -ESPIPE;
if (!iov_iter_count(to)) /* Match SYS5 behaviour */
return 0;
res = sock_recvmsg(sock, &msg, iov_iter_count(to), msg.msg_flags);
*to = msg.msg_iter;
return res;
}
AFAICS, everything looks fine. Now, _another_ BUG_ON() might be worth
eventual removal (one in sock_sendmsg_nosec()), but right now we don't
have async ->sendmsg() instances, so it's not urgent.
^ permalink raw reply [flat|nested] 36+ messages in thread
* RE: [PATCH 16/17] get rid of the size argument of sock_sendmsg()
2015-04-14 17:55 ` David Miller
@ 2015-04-15 8:37 ` David Laight
2015-04-15 10:15 ` Eric Dumazet
2015-04-15 16:06 ` David Miller
0 siblings, 2 replies; 36+ messages in thread
From: David Laight @ 2015-04-15 8:37 UTC (permalink / raw)
To: 'David Miller', viro; +Cc: netdev
From: David Miller
> Sent: 14 April 2015 18:56
> > On Tue, Apr 14, 2015 at 04:25:24PM +0000, David Laight wrote:
> >> From: Al Viro
> >> > Sent: 11 April 2015 22:18
> >> > it's equal to iov_iter_count(&msg->msg_iter) in all cases
> >>
> >> I don't know whether this is guaranteed for iov[] that come from
> >> userspace.
> >>
> >> In any case iov_iter_count() is non-trivial and you don't
> >> really want to call it when unnecessary.
> >
> > Really?
> >
> > static inline size_t iov_iter_count(struct iov_iter *i)
> > {
> > return i->count;
> > }
>
> This just made my day.
I hate accessor functions
David
^ permalink raw reply [flat|nested] 36+ messages in thread
* RE: [PATCH 15/17] switch kernel_sendmsg() and kernel_recvmsg() to iov_iter_kvec()
2015-04-14 16:59 ` Al Viro
@ 2015-04-15 9:08 ` David Laight
2015-04-15 9:36 ` Daniel Borkmann
0 siblings, 1 reply; 36+ messages in thread
From: David Laight @ 2015-04-15 9:08 UTC (permalink / raw)
To: 'Al Viro'; +Cc: davem, netdev
From: Al Viro
> Sent: 14 April 2015 17:59
> On Tue, Apr 14, 2015 at 04:36:36PM +0000, David Laight wrote:
> > From: Al Viro
> > > Sent: 14 April 2015 17:34
> > > On Tue, Apr 14, 2015 at 04:21:02PM +0000, David Laight wrote:
> > >
> > > > Massive NAK.
> > > > This breaks any code that is using msg_control to set SCTP parameters
> > > > when sending data.
> > >
> > > Huh? ->sendmsg() expects ->msg_control already in kernel space;
> > > it's ->recvmsg() that plays silly buggers with userland pointers there.
> >
> > I read your commit message as implying that you hadn't found any
> > users of kernel_sendmsg() that used msg_control.
> > Not that the data was always read from kernel space.
>
> Sigh... The situation is:
> * ->sendmsg() expects ->msg_control copied to userland. sendmsg(2),
> sendto(2), etc. do that copying. See ___sys_sendmsg() - there we have
> /*
> * Careful! Before this, msg_sys->msg_control contains a user pointer.
> * Afterwards, it will be a kernel pointer. Thus the compiler-assisted
> * checking falls down on this.
> */
> if (copy_from_user(ctl_buf,
> (void __user __force *)msg_sys->msg_control,
> ctl_len))
> goto out_freectl;
> msg_sys->msg_control = ctl_buf;
> As the result, ->sendmsg() instances access ->msg_control contents as normal
> kernel data.
> * ->recvmsg() expects ->msg_control to point to userland. See
> net/core/scm.c for the helpers used to store into it. recvmsg(2) et.al.
> simply leave the userland pointer there; worse, that pointer might be
> to native or to compat variants, and layouts _are_ different. Thus those
> if (MSG_CMSG_COMPAT & msg->msg_flags) in net/core/scm.c...
> * kernel-side users of ->sendmsg() do not depend on setfs() for
> access to their ->msg_control, simply because ->sendmsg() won't be using
> copy_from_user()/get_user() to access it anyway.
> * kernel-side users of ->recvmsg() are less lucky - most of them
> don't give a damn either (they have NULL ->msg_control), but there's an
> exception (somewhere in sunrpc, IIRC). So there we need to keep
> playing with setfs(), even though the data side would be just fine without
> that.
Apart from any other code that is using the interface.
I know you guys don't do anything to help out of tree code, but removing the setfs()
stuff from the kernel_recvmsg() code would break anything using sctp.
It shouldn't need some code lurking in sunrpc for you to leave the setfs().
In any case, how much does the setfs() cost?
I suspect it is just modifying a flag in 'current'.
A comment in kernel_recvmsg() saying that the setfs() is for msg_control
might be useful.
Then one in kelnel_sendmsg() saying that setfs() isn't needed because
msg_control is always kernel - just to avoid any confusion.
David
^ permalink raw reply [flat|nested] 36+ messages in thread
* Re: [PATCH 15/17] switch kernel_sendmsg() and kernel_recvmsg() to iov_iter_kvec()
2015-04-15 9:08 ` David Laight
@ 2015-04-15 9:36 ` Daniel Borkmann
2015-04-15 9:53 ` David Laight
0 siblings, 1 reply; 36+ messages in thread
From: Daniel Borkmann @ 2015-04-15 9:36 UTC (permalink / raw)
To: David Laight, 'Al Viro'; +Cc: davem, netdev
On 04/15/2015 11:08 AM, David Laight wrote:
...
> Apart from any other code that is using the interface.
> I know you guys don't do anything to help out of tree code, but removing the setfs()
> stuff from the kernel_recvmsg() code would break anything using sctp.
Then that might just be one more incentive to work towards upstreaming
your out-of-tree SCTP bits ... ;)
^ permalink raw reply [flat|nested] 36+ messages in thread
* RE: [PATCH 15/17] switch kernel_sendmsg() and kernel_recvmsg() to iov_iter_kvec()
2015-04-15 9:36 ` Daniel Borkmann
@ 2015-04-15 9:53 ` David Laight
0 siblings, 0 replies; 36+ messages in thread
From: David Laight @ 2015-04-15 9:53 UTC (permalink / raw)
To: 'Daniel Borkmann', 'Al Viro'; +Cc: davem, netdev
From: Daniel Borkmann
> Sent: 15 April 2015 10:37
> On 04/15/2015 11:08 AM, David Laight wrote:
> ...
> > Apart from any other code that is using the interface.
> > I know you guys don't do anything to help out of tree code, but removing the setfs()
> > stuff from the kernel_recvmsg() code would break anything using sctp.
>
> Then that might just be one more incentive to work towards upstreaming
> your out-of-tree SCTP bits ... ;)
You really wouldn't want the many MB of code that generates several MB
of driver object to support a telephony applications using custom hardware.
Almost all the code is completely OS agnostic, the linux 'glue' code is
a few 1000 lines of wrapper functions (compiled as part of the installation
process).
Much the same wrappers exist for windows and solaris.
David
^ permalink raw reply [flat|nested] 36+ messages in thread
* Re: [PATCH 16/17] get rid of the size argument of sock_sendmsg()
2015-04-15 8:37 ` David Laight
@ 2015-04-15 10:15 ` Eric Dumazet
2015-04-15 16:06 ` David Miller
1 sibling, 0 replies; 36+ messages in thread
From: Eric Dumazet @ 2015-04-15 10:15 UTC (permalink / raw)
To: David Laight; +Cc: 'David Miller', viro, netdev
On Wed, 2015-04-15 at 08:37 +0000, David Laight wrote:
> From: David Miller
> > Sent: 14 April 2015 18:56
> > > On Tue, Apr 14, 2015 at 04:25:24PM +0000, David Laight wrote:
> > >> From: Al Viro
> > >> > Sent: 11 April 2015 22:18
> > >> > it's equal to iov_iter_count(&msg->msg_iter) in all cases
> > >>
> > >> I don't know whether this is guaranteed for iov[] that come from
> > >> userspace.
> > >>
> > >> In any case iov_iter_count() is non-trivial and you don't
> > >> really want to call it when unnecessary.
> > >
> > > Really?
> > >
> > > static inline size_t iov_iter_count(struct iov_iter *i)
> > > {
> > > return i->count;
> > > }
> >
> > This just made my day.
>
> I hate accessor functions
It is quite clear you do not really tweak linux code, do you ?
Typical linux developers love them when they want to add in accessor a
debugging facility or logic.
^ permalink raw reply [flat|nested] 36+ messages in thread
* Re: [PATCH 16/17] get rid of the size argument of sock_sendmsg()
2015-04-15 8:37 ` David Laight
2015-04-15 10:15 ` Eric Dumazet
@ 2015-04-15 16:06 ` David Miller
1 sibling, 0 replies; 36+ messages in thread
From: David Miller @ 2015-04-15 16:06 UTC (permalink / raw)
To: David.Laight; +Cc: viro, netdev
From: David Laight <David.Laight@ACULAB.COM>
Date: Wed, 15 Apr 2015 08:37:01 +0000
> From: David Miller
>> Sent: 14 April 2015 18:56
>> > On Tue, Apr 14, 2015 at 04:25:24PM +0000, David Laight wrote:
>> >> From: Al Viro
>> >> > Sent: 11 April 2015 22:18
>> >> > it's equal to iov_iter_count(&msg->msg_iter) in all cases
>> >>
>> >> I don't know whether this is guaranteed for iov[] that come from
>> >> userspace.
>> >>
>> >> In any case iov_iter_count() is non-trivial and you don't
>> >> really want to call it when unnecessary.
>> >
>> > Really?
>> >
>> > static inline size_t iov_iter_count(struct iov_iter *i)
>> > {
>> > return i->count;
>> > }
>>
>> This just made my day.
>
> I hate accessor functions
I hate people who jump to conclusions and don't actually read the code
in question before commenting. :-)
^ permalink raw reply [flat|nested] 36+ messages in thread
end of thread, other threads:[~2015-04-15 16:06 UTC | newest]
Thread overview: 36+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-04-11 21:17 [call for review] netdev-related stuff in vfs.git Al Viro
2015-04-11 21:18 ` [PATCH 01/17] fs: remove ki_nbytes Al Viro
2015-04-11 21:18 ` [PATCH 02/17] fuse: handle synchronous iocbs internally Al Viro
2015-04-11 21:18 ` [PATCH 03/17] fs: don't allow to complete sync iocbs through aio_complete Al Viro
2015-04-14 18:00 ` Tadeusz Struk
2015-04-14 18:26 ` Al Viro
2015-04-14 18:37 ` Tadeusz Struk
2015-04-14 19:22 ` Al Viro
2015-04-11 21:18 ` [PATCH 04/17] fs: split generic and aio kiocb Al Viro
2015-04-11 21:18 ` [PATCH 05/17] fs: move struct kiocb to fs.h Al Viro
2015-04-11 21:18 ` [PATCH 06/17] saner iov_iter initialization primitives Al Viro
2015-04-11 21:18 ` [PATCH 07/17] RxRPC: Fix the conversion to iov_iter Al Viro
2015-04-11 21:18 ` [PATCH 08/17] RxRPC: Don't call skb_add_data() if there's no data to copy Al Viro
2015-04-11 21:18 ` [PATCH 09/17] RxRPC: Use iov_iter_count() in rxrpc_send_data() instead of the len argument Al Viro
2015-04-11 21:18 ` [PATCH 10/17] AFS: afs_send_empty_reply() doesn't require an iovec array Al Viro
2015-04-11 21:18 ` [PATCH 11/17] RxRPC: Handle VERSION Rx protocol packets Al Viro
2015-04-11 21:18 ` [PATCH 12/17] kafs: Add more "unified AFS" error codes Al Viro
2015-04-11 21:18 ` [PATCH 13/17] net: switch sendto() and recvfrom() to import_single_range() Al Viro
2015-04-11 21:18 ` [PATCH 14/17] net: switch importing msghdr from userland to {compat_,}import_iovec() Al Viro
2015-04-11 21:18 ` [PATCH 15/17] switch kernel_sendmsg() and kernel_recvmsg() to iov_iter_kvec() Al Viro
2015-04-14 16:21 ` David Laight
2015-04-14 16:34 ` Al Viro
2015-04-14 16:36 ` David Laight
2015-04-14 16:59 ` Al Viro
2015-04-15 9:08 ` David Laight
2015-04-15 9:36 ` Daniel Borkmann
2015-04-15 9:53 ` David Laight
2015-04-11 21:18 ` [PATCH 16/17] get rid of the size argument of sock_sendmsg() Al Viro
2015-04-14 16:25 ` David Laight
2015-04-14 16:35 ` Al Viro
2015-04-14 16:44 ` Al Viro
2015-04-14 17:55 ` David Miller
2015-04-15 8:37 ` David Laight
2015-04-15 10:15 ` Eric Dumazet
2015-04-15 16:06 ` David Miller
2015-04-11 21:18 ` [PATCH 17/17] new helper: msg_data_left() Al Viro
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.