Fix kernel-doc warnings in struct pipe_inode_info after @wait was
split into @rd_wait and @wr_wait.
include/linux/pipe_fs_i.h:66: warning: Function parameter or member 'rd_wait' not described in 'pipe_inode_info'
include/linux/pipe_fs_i.h:66: warning: Function parameter or member 'wr_wait' not described in 'pipe_inode_info'
Fixes: 0ddad21d3e99 ("pipe: use exclusive waits when reading or writing")
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This makes the pipe code use separate wait-queues and exclusive waiting
for readers and writers, avoiding a nasty thundering herd problem when
there are lots of readers waiting for data on a pipe (or, less commonly,
lots of writers waiting for a pipe to have space).
While this isn't a common occurrence in the traditional "use a pipe as a
data transport" case, where you typically only have a single reader and
a single writer process, there is one common special case: using a pipe
as a source of "locking tokens" rather than for data communication.
In particular, the GNU make jobserver code ends up using a pipe as a way
to limit parallelism, where each job consumes a token by reading a byte
from the jobserver pipe, and releases the token by writing a byte back
to the pipe.
This pattern is fairly traditional on Unix, and works very well, but
will waste a lot of time waking up a lot of processes when only a single
reader needs to be woken up when a writer releases a new token.
A simplified test-case of just this pipe interaction is to create 64
processes, and then pass a single token around between them (this
test-case also intentionally passes another token that gets ignored to
test the "wake up next" logic too, in case anybody wonders about it):
#include <unistd.h>
int main(int argc, char **argv)
{
int fd[2], counters[2];
pipe(fd);
counters[0] = 0;
counters[1] = -1;
write(fd[1], counters, sizeof(counters));
/* 64 processes */
fork(); fork(); fork(); fork(); fork(); fork();
do {
int i;
read(fd[0], &i, sizeof(i));
if (i < 0)
continue;
counters[0] = i+1;
write(fd[1], counters, (1+(i & 1)) *sizeof(int));
} while (counters[0] < 1000000);
return 0;
}
and in a perfect world, passing that token around should only cause one
context switch per transfer, when the writer of a token causes a
directed wakeup of just a single reader.
But with the "writer wakes all readers" model we traditionally had, on
my test box the above case causes more than an order of magnitude more
scheduling: instead of the expected ~1M context switches, "perf stat"
shows
231,852.37 msec task-clock # 15.857 CPUs utilized
11,250,961 context-switches # 0.049 M/sec
616,304 cpu-migrations # 0.003 M/sec
1,648 page-faults # 0.007 K/sec
1,097,903,998,514 cycles # 4.735 GHz
120,781,778,352 instructions # 0.11 insn per cycle
27,997,056,043 branches # 120.754 M/sec
283,581,233 branch-misses # 1.01% of all branches
14.621273891 seconds time elapsed
0.018243000 seconds user
3.611468000 seconds sys
before this commit.
After this commit, I get
5,229.55 msec task-clock # 3.072 CPUs utilized
1,212,233 context-switches # 0.232 M/sec
103,951 cpu-migrations # 0.020 M/sec
1,328 page-faults # 0.254 K/sec
21,307,456,166 cycles # 4.074 GHz
12,947,819,999 instructions # 0.61 insn per cycle
2,881,985,678 branches # 551.096 M/sec
64,267,015 branch-misses # 2.23% of all branches
1.702148350 seconds time elapsed
0.004868000 seconds user
0.110786000 seconds sys
instead. Much better.
[ Note! This kernel improvement seems to be very good at triggering a
race condition in the make jobserver (in GNU make 4.2.1) for me. It's
a long known bug that was fixed back in June 2017 by GNU make commit
b552b0525198 ("[SV 51159] Use a non-blocking read with pselect to
avoid hangs.").
But there wasn't a new release of GNU make until 4.3 on Jan 19 2020,
so a number of distributions may still have the buggy version. Some
have backported the fix to their 4.2.1 release, though, and even
without the fix it's quite timing-dependent whether the bug actually
is hit. ]
Josh Triplett says:
"I've been hammering on your pipe fix patch (switching to exclusive
wait queues) for a month or so, on several different systems, and I've
run into no issues with it. The patch *substantially* improves
parallel build times on large (~100 CPU) systems, both with parallel
make and with other things that use make's pipe-based jobserver.
All current distributions (including stable and long-term stable
distributions) have versions of GNU make that no longer have the
jobserver bug"
Tested-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This code is ancient, and goes back to when we only had a single page for the pipe buffers. The exact history is hidden in the mists of time (ie "before git", and in fact predates the BK repository too). At that long-ago point in time, it actually helped to try to merge big back-and-forth pipe reads and writes, and not limit pipe reads to the single pipe buffer in length just because that was all we had at a time. However, since then we've expanded the pipe buffers to multiple pages, and this logic really doesn't seem to make sense. And a lot of it is somewhat questionable (ie "hmm, the user asked for a non-blocking read, but we see that there's a writer pending, so let's wait anyway to get the extra data that the writer will have"). But more importantly, it makes the "go to sleep" logic much less obvious, and considering the wakeup issues we've had, I want to make for less of those kinds of things. Cc: David Howells <dhowells@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Split pipe->ring_size into two numbers:
(1) pipe->ring_size - indicates the hard size of the pipe ring.
(2) pipe->max_usage - indicates the maximum number of pipe ring slots that
userspace orchestrated events can fill.
This allows for a pipe that is both writable by the general kernel
notification facility and by userspace, allowing plenty of ring space for
notifications to be added whilst preventing userspace from being able to
pin too much unswappable kernel space.
Signed-off-by: David Howells <dhowells@redhat.com>
Convert pipes to use head and tail pointers for the buffer ring rather than
pointer and length as the latter requires two atomic ops to update (or a
combined op) whereas the former only requires one.
(1) The head pointer is the point at which production occurs and points to
the slot in which the next buffer will be placed. This is equivalent
to pipe->curbuf + pipe->nrbufs.
The head pointer belongs to the write-side.
(2) The tail pointer is the point at which consumption occurs. It points
to the next slot to be consumed. This is equivalent to pipe->curbuf.
The tail pointer belongs to the read-side.
(3) head and tail are allowed to run to UINT_MAX and wrap naturally. They
are only masked off when the array is being accessed, e.g.:
pipe->bufs[head & mask]
This means that it is not necessary to have a dead slot in the ring as
head == tail isn't ambiguous.
(4) The ring is empty if "head == tail".
A helper, pipe_empty(), is provided for this.
(5) The occupancy of the ring is "head - tail".
A helper, pipe_occupancy(), is provided for this.
(6) The number of free slots in the ring is "pipe->ring_size - occupancy".
A helper, pipe_space_for_user() is provided to indicate how many slots
userspace may use.
(7) The ring is full if "head - tail >= pipe->ring_size".
A helper, pipe_full(), is provided for this.
Signed-off-by: David Howells <dhowells@redhat.com>
Pull tracing fixes from Steven Rostedt:
"Three tracing fixes:
- Use "nosteal" for ring buffer splice pages
- Memory leak fix in error path of trace_pid_write()
- Fix preempt_enable_no_resched() (use preempt_enable()) in ring
buffer code"
* tag 'trace-v5.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
trace: Fix preempt_enable_no_resched() abuse
tracing: Fix a memory leak by early error exit in trace_pid_write()
tracing: Fix buffer_ref pipe ops
This fixes multiple issues in buffer_pipe_buf_ops:
- The ->steal() handler must not return zero unless the pipe buffer has
the only reference to the page. But generic_pipe_buf_steal() assumes
that every reference to the pipe is tracked by the page's refcount,
which isn't true for these buffers - buffer_pipe_buf_get(), which
duplicates a buffer, doesn't touch the page's refcount.
Fix it by using generic_pipe_buf_nosteal(), which refuses every
attempted theft. It should be easy to actually support ->steal, but the
only current users of pipe_buf_steal() are the virtio console and FUSE,
and they also only use it as an optimization. So it's probably not worth
the effort.
- The ->get() and ->release() handlers can be invoked concurrently on pipe
buffers backed by the same struct buffer_ref. Make them safe against
concurrency by using refcount_t.
- The pointers stored in ->private were only zeroed out when the last
reference to the buffer_ref was dropped. As far as I know, this
shouldn't be necessary anyway, but if we do it, let's always do it.
Link: http://lkml.kernel.org/r/20190404215925.253531-1-jannh@google.com
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: stable@vger.kernel.org
Fixes: 73a757e63114d ("ring-buffer: Return reader page back into existing ring buffer")
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Merge page ref overflow branch. Jann Horn reported that he can overflow the page ref count with sufficient memory (and a filesystem that is intentionally extremely slow). Admittedly it's not exactly easy. To have more than four billion references to a page requires a minimum of 32GB of kernel memory just for the pointers to the pages, much less any metadata to keep track of those pointers. Jann needed a total of 140GB of memory and a specially crafted filesystem that leaves all reads pending (in order to not ever free the page references and just keep adding more). Still, we have a fairly straightforward way to limit the two obvious user-controllable sources of page references: direct-IO like page references gotten through get_user_pages(), and the splice pipe page duplication. So let's just do that. * branch page-refs: fs: prevent page refcount overflow in pipe_buf_get mm: prevent get_user_pages() from overflowing page refcount mm: add 'try_get_page()' helper function mm: make page ref count overflow check tighter and more explicit
Change pipe_buf_get() to return a bool indicating whether it succeeded in raising the refcount of the page (if the thing in the pipe is a page). This removes another mechanism for overflowing the page refcount. All callers converted to handle a failure. Reported-by: Jann Horn <jannh@google.com> Signed-off-by: Matthew Wilcox <willy@infradead.org> Cc: stable@kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Al Viro pointed out that since there is only one pipe buffer type to which new data can be appended, it isn't necessary to have a ->can_merge field in struct pipe_buf_operations, we can just check for a magic type. Suggested-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Jann Horn <jannh@google.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>