Commit graph

53 commits

Author SHA1 Message Date
Jens Axboe
9e94cd4fd1 [PATCH] splice: retrieve mapping after locking the page
Otherwise we could be racing with truncate/mapping removal.

Problem found/fixed by Nick Piggin <npiggin@suse.de>, logic rewritten
by me.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-06-23 17:10:39 +02:00
Jens Axboe
a0548871ed [PATCH] splice: redo page lookup if add_to_page_cache() returns -EEXIST
This can happen quite easily, if several processes are trying to splice
the same file at the same time. It's not a failure, it just means someone
raced with us in allocating this file page. So just dump the allocated
page and relookup the original.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-05-04 06:55:12 +02:00
Jens Axboe
76ad4d1110 [PATCH] splice: rename remaining info variables to pipe
Same thing was done in fs/pipe.c and most of fs/splice.c, but we had
a few missing still.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-05-04 06:55:12 +02:00
Jens Axboe
1432873af7 [PATCH] splice: LRU fixups
Nick says that the current construct isn't safe. This goes back to the
original, but sets PIPE_BUF_FLAG_LRU on user pages as well as they all
seem to be on the LRU in the first place.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-05-04 06:55:12 +02:00
Jens Axboe
bfc4ee39fd [PATCH] splice: fix unlocking of page on error ->prepare_write()
Looking at generic_file_buffered_write(), we need to unlock_page() if
prepare write fails and it isn't due to racing with truncate().

Also trim the size if ->prepare_write() fails, if we have to.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-05-04 06:55:12 +02:00
Jens Axboe
330ab71619 [PATCH] vmsplice: restrict stealing a little more
Apply the same rules as the anon pipe pages, only allow stealing
if no one else is using the page.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-05-02 15:29:57 +02:00
Jens Axboe
a893b99be7 [PATCH] splice: fix page LRU accounting
Currently we rely on the PIPE_BUF_FLAG_LRU flag being set correctly
to know whether we need to fiddle with page LRU state after stealing it,
however for some origins we just don't know if the page is on the LRU
list or not.

So remove PIPE_BUF_FLAG_LRU and do this check/add manually in pipe_to_file()
instead.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-05-02 15:03:27 +02:00
Jens Axboe
7591489a8f [PATCH] vmsplice: fix badly placed end paranthesis
We need to use the minium of {len, PAGE_SIZE-off}, not {len, PAGE_SIZE}-off.
The latter doesn't make any sense, and could cause us to attempt negative
length transfers...

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-05-02 12:57:18 +02:00
Jens Axboe
7afa6fd037 [PATCH] vmsplice: allow user to pass in gift pages
If SPLICE_F_GIFT is set, the user is basically giving this pages away to
the kernel. That means we can steal them for eg page cache uses instead
of copying it.

The data must be properly page aligned and also a multiple of the page size
in length.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-05-01 20:02:33 +02:00
Jens Axboe
f6762b7ad8 [PATCH] pipe: enable atomic copying of pipe data to/from user space
The pipe ->map() method uses kmap() to virtually map the pages, which
is both slow and has known scalability issues on SMP. This patch enables
atomic copying of pipe pages, by pre-faulting data and using kmap_atomic()
instead.

lmbench bw_pipe and lat_pipe measurements agree this is a Good Thing. Here
are results from that on a UP machine with highmem (1.5GiB of RAM), running
first a UP kernel, SMP kernel, and SMP kernel patched.

Vanilla-UP:
Pipe bandwidth: 1622.28 MB/sec
Pipe bandwidth: 1610.59 MB/sec
Pipe bandwidth: 1608.30 MB/sec
Pipe latency: 7.3275 microseconds
Pipe latency: 7.2995 microseconds
Pipe latency: 7.3097 microseconds

Vanilla-SMP:
Pipe bandwidth: 1382.19 MB/sec
Pipe bandwidth: 1317.27 MB/sec
Pipe bandwidth: 1355.61 MB/sec
Pipe latency: 9.6402 microseconds
Pipe latency: 9.6696 microseconds
Pipe latency: 9.6153 microseconds

Patched-SMP:
Pipe bandwidth: 1578.70 MB/sec
Pipe bandwidth: 1579.95 MB/sec
Pipe bandwidth: 1578.63 MB/sec
Pipe latency: 9.1654 microseconds
Pipe latency: 9.2266 microseconds
Pipe latency: 9.1527 microseconds

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-05-01 20:02:05 +02:00
Jens Axboe
e27dedd84c [PATCH] splice: call handle_ra_miss() on failure to lookup page
Notify the readahead logic of the missing page. Suggested by
Oleg Nesterov.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-05-01 19:59:54 +02:00
Jens Axboe
f84d751994 [PATCH] pipe: introduce ->pin() buffer operation
The ->map() function is really expensive on highmem machines right now,
since it has to use the slower kmap() instead of kmap_atomic(). Splice
rarely needs to access the virtual address of a page, so it's a waste
of time doing it.

Introduce ->pin() to take over the responsibility of making sure the
page data is valid. ->map() is then reduced to just kmap(). That way we
can also share a most of the pipe buffer ops between pipe.c and splice.c

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-05-01 19:59:03 +02:00
Jens Axboe
0568b409c7 [PATCH] splice: fix bugs in pipe_to_file()
Found by Oleg Nesterov <oleg@tv-sign.ru>, fixed by me.

- Only allow full pages to go to the page cache.
- Check page != buf->page instead of using PIPE_BUF_FLAG_STOLEN.
- Remember to clear 'stolen' if add_to_page_cache() fails.

And as a cleanup on that:

- Make the bottom fall-through logic a little less convoluted. Also make
  the steal path hold an extra reference to the page, so we don't have
  to differentiate between stolen and non-stolen at the end.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-05-01 19:50:48 +02:00
Jens Axboe
46e678c96b [PATCH] splice: fix bugs with stealing regular pipe pages
- Check that page has suitable count for stealing in the regular pipes.
- pipe_to_file() assumes that the page is locked on succesful steal, so
  do that in the pipe steal hook
- Missing unlock_page() in add_to_page_cache() failure.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-30 16:36:32 +02:00
Jens Axboe
eb20796bf6 [PATCH] splice: make the read-side do batched page lookups
Use the new find_get_pages_contig() to potentially look up the entire
splice range in one single call. This speeds up generic_file_splice_read()
quite a bit.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-27 11:05:22 +02:00
Jens Axboe
eb645a24de [PATCH] splice: switch to using page_cache_readahead()
Avoids doing useless work, when the file is fully cached.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-27 08:59:48 +02:00
Jens Axboe
00522fb41a [PATCH] splice: rearrange moving to/from pipe helpers
We need these for people writing their own ->splice_read/write hooks.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-26 14:39:29 +02:00
Jens Axboe
912d35f867 [PATCH] Add support for the sys_vmsplice syscall
sys_splice() moves data to/from pipes with a file input/output. sys_vmsplice()
moves data to a pipe, with the input being a user address range instead.

This uses an approach suggested by Linus, where we can hold partial ranges
inside the pages[] map. Hopefully this will be useful for network
receive support as well.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-26 10:59:21 +02:00
Jens Axboe
016b661e2f [PATCH] splice: fix offset problems
Make the move_from_pipe() actors return number of bytes processed, then
move_from_pipe() can decide more cleverly when to move on to the next
buffer.

This fixes problems with pipe offset and differing file offset.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-26 10:33:34 +02:00
Andrew Morton
ba5f5d90c4 [PATCH] splice: fix min() warning
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-26 10:33:34 +02:00
Jens Axboe
82aa5d6183 [PATCH] splice: fix smaller sized splice reads
Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-20 13:05:48 +02:00
Jens Axboe
9e0267c26e [PATCH] splice: fixup writeout path after ->map changes
Since ->map() no longer locks the page, we need to adjust the handling
of those pages (and stealing) a little. This now passes full regressions
again.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-19 15:57:31 +02:00
Jens Axboe
a4514ebd8e [PATCH] splice: offset fixes
- We need to adjust *ppos for writes as well.
- Copy back modified offset value if one was passed in, similar to
  what sendfile does.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-19 15:57:05 +02:00
Jens Axboe
2a27250e6c [PATCH] tee: link_pipe() must be careful when dropping one of the pipe locks
We need to ensure that we only drop a lock that is ordered last, to avoid
ABBA deadlocks with competing processes.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-19 15:56:40 +02:00
Jens Axboe
c4f895cbe1 [PATCH] splice: cleanup the SPLICE_F_NONBLOCK handling
- generic_file_splice_read() more readable and correct
- Don't bail on page allocation with NONBLOCK set, just don't allow
  direct blocking on IO (eg lock_page).

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-19 15:56:12 +02:00
Jens Axboe
91ad66ef44 [PATCH] splice: close i_size truncate races on read
We need to check i_size after doing a blocking readpage.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-19 15:55:10 +02:00
Jens Axboe
70524490ee [PATCH] splice: add support for sys_tee()
Basically an in-kernel implementation of tee, which uses splice and the
pipe buffers as an intelligent way to pass data around by reference.

Where the user space tee consumes the input and produces a stdout and
file output, this syscall merely duplicates the data inside a pipe to
another pipe. No data is copied, the output just grabs a reference to the
input pipe data.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-11 15:51:17 +02:00
Jens Axboe
cbb7e577e7 [PATCH] splice: pass offset around for ->splice_read() and ->splice_write()
We need not use ->f_pos as the offset for the file input/output. If the
user passed an offset pointer in through sys_splice(), just use that and
leave ->f_pos alone.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-11 15:47:07 +02:00
Ingo Molnar
73d62d83ec [PATCH] splice: comment styles
- capitalize consistently
 - end sentences in one way or another
 - update comment text to match the implementation

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-11 13:57:21 +02:00
Jens Axboe
c2058e0611 [PATCH] splice: add Ingo as addition copyright holder
The comment is also somewhat out of date, correct that as well.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-11 13:56:34 +02:00
Jens Axboe
49570e9b29 [PATCH] splice: unlikely() optimizations
Also corrects a few comments. Patch mainly from Ingo, changes by me.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-11 13:56:09 +02:00
Jens Axboe
6f767b0425 [PATCH] splice: speedups and optimizations
- Kill the local variables that cache ->nrbufs, they just take up space.

- Only set do_wakeup for a real pipe. This is a big win for direct splicing.

- Kill i_mutex lock around ->f_pos update, regular io paths don't do this
  either.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-11 13:53:56 +02:00
Jens Axboe
7480a90435 [PATCH] splice: speedup __generic_file_splice_read
Using find_get_page() is a lot faster than find_or_create_page(). This
gets splice a lot closer to sendfile() for fd -> socket transfers.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-11 13:52:47 +02:00
Jens Axboe
b92ce55893 [PATCH] splice: add direct fd <-> fd splicing support
It's more efficient for sendfile() emulation. Basically we cache an
internal private pipe and just use that as the intermediate area for
pages. Direct splicing is not available from sys_splice(), it is only
meant to be used for sendfile() emulation.

Additional patch from Ingo Molnar to avoid the PIPE_BUFFERS loop at
exit for the normal fast path.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-11 13:52:07 +02:00
Ingo Molnar
529565dcb1 [PATCH] splice: add optional input and output offsets
add optional input and output offsets to sys_splice(), for seekable file
descriptors:

 asmlinkage long sys_splice(int fd_in, loff_t __user *off_in,
                            int fd_out, loff_t __user *off_out,
                            size_t len, unsigned int flags);

semantics are straightforward: f_pos will be updated with the offset
provided by user-space, before the splice transfer is about to begin.
Providing a NULL offset pointer means the existing f_pos will be used
(and updated in situ).  Providing an offset for a pipe results in
-ESPIPE. Providing an invalid offset pointer results in -EFAULT.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-10 15:18:58 +02:00
Ingo Molnar
3a326a2ce8 [PATCH] introduce a "kernel-internal pipe object" abstraction
separate out the 'internal pipe object' abstraction, and make it
usable to splice. This cleans up and fixes several aspects of the
internal splice APIs and the pipe code:

 - pipes: the allocation and freeing of pipe_inode_info is now more symmetric
   and more streamlined with existing kernel practices.

 - splice: small micro-optimization: less pointer dereferencing in splice
   methods

Signed-off-by: Ingo Molnar <mingo@elte.hu>

Update XFS for the ->splice_read/->splice_write changes.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-10 15:18:35 +02:00
Jens Axboe
0b749ce380 [PATCH] splice: be smarter about calling do_page_cache_readahead()
We don't want to call into the read-ahead logic unless we are at the
start of a page, _or_ we have multiple pages to read.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-10 09:05:04 +02:00
Jens Axboe
49d0b21be2 [PATCH] splice: optimize the splice buffer mapping
We don't really need to lock down the pages, just make sure they
are uptodate.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-10 09:04:41 +02:00
Jens Axboe
16c523ddab [PATCH] splice: cleanup __generic_file_splice_read()
The whole shadow/pages logic got overly complex, and this simpler
approach is actually faster in testing.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-10 09:03:58 +02:00
Jens Axboe
c0bd1f650b [PATCH] splice: only call wake_up_interruptible() when we really have to
__wake_up_common() is pretty heavy in the kernel profiles, this brings
it down to a more acceptable level.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-10 09:03:32 +02:00
Dave Jones
9aefe431f5 [PATCH] splice: potential !page dereference
We can get to out: with a NULL page, which we probably
don't want to be calling page_cache_release() on.

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-10 09:02:40 +02:00
Jens Axboe
c7f21e4f5a [PATCH] splice: mark the io page as accessed
We should do that, since we do the LRU manipulation ourselves now. Suggested
by Nick Piggin.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-10 09:01:01 +02:00
Jens Axboe
3e7ee3e7b3 [PATCH] splice: fix page stealing LRU handling.
Originally from Nick Piggin, just adapted to the newer branch.

You can't check PageLRU without holding zone->lru_lock.  The page
release code can get away with it only because the page refcount is 0 at
that point. Also, you can't reliably remove pages from the LRU unless
the refcount is 0. Ever.

Signed-off-by: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-02 23:11:04 +02:00
Jens Axboe
ad8d6f0a78 [PATCH] splice: page stealing needs to wait_on_page_writeback()
Thanks to Andrew for the good explanation of why this is so. akpm writes:

If a page is under writeback and we remove it from pagecache, it's still
going to get written to disk.  But the VFS no longer knows about that page,
nor that this page is about to modify disk blocks.

So there might be scenarios in which those
blocks-which-are-about-to-be-written-to get reused for something else.
When writeback completes, it'll scribble on those blocks.

This won't happen in ext2/ext3-style filesystems in normal mode because the
page has buffers and try_to_release_page() will fail.

But ext2 in nobh mode doesn't attach buffers at all - it just sticks the
page in a BIO, finds some new blocks, points the BIO at those blocks and
lets it rip.

While that write IO's in flight, someone could truncate the file.  Truncate
won't block on the writeout because the page isn't in pagecache any more.
So truncate will the free the blocks from the file under the page's feet.
Then something else can reallocate those blocks.  Then write data to them.

Now, the original write completes, corrupting the filesystem.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-02 23:10:32 +02:00
Jens Axboe
059a8f3734 [PATCH] splice: export generic_splice_sendpage
Forgot that one, thanks Jeff. Also move the other EXPORT_SYMBOL
to right below the functions.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-02 23:06:05 +02:00
Jens Axboe
b2b39fa478 [PATCH] splice: add a SPLICE_F_MORE flag
This lets userspace indicate whether more data will be coming in a
subsequent splice call.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-02 23:05:41 +02:00
Jens Axboe
83f9135bdd [PATCH] splice: add comments documenting more of the code
Hopefully this will make Andrew a little more happy.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-02 23:05:09 +02:00
Jens Axboe
4f6f0bd2ff [PATCH] splice: improve writeback and clean up page stealing
By cleaning up the writeback logic (killing write_one_page() and the manual
set_page_dirty()), we can get rid of ->stolen inside the pipe_buffer and
just keep it local in pipe_to_file().

This also adds dirty page balancing logic and O_SYNC handling.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-02 23:04:46 +02:00
Jens Axboe
53cd9ae886 [PATCH] splice: fix shadow[] filling logic
Clear the entire range, and don't increment pidx or we keep filling
the same position again and again.

Thanks to KAMEZAWA Hiroyuki.

Signed-off-by: Jens Axboe <axboe@suse.de>
2006-04-02 23:04:21 +02:00
Linus Torvalds
29e350944f splice: add SPLICE_F_NONBLOCK flag
It doesn't make the splice itself necessarily nonblocking (because the
actual file descriptors that are spliced from/to may block unless they
have the O_NONBLOCK flag set), but it makes the splice pipe operations
nonblocking.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-04-02 12:46:35 -07:00