c/mirror_zfs
1
0
Fork 0
mirror of https://git.proxmox.com/git/mirror_zfs.git synced 2024-11-17 01:51:00 +03:00
Code Issues Packages Projects Releases Wiki Activity

Increase Linux pipe buffer size on 'zfs receive'

Browse Source 
I noticed when reviewing documentation that it is possible for user
space to use fctnl(fd, F_SETPIPE_SZ, (unsigned long) size) to change
the kernel pipe buffer size on Linux to increase the pipe size up to
the value specified in /proc/sys/fs/pipe-max-size. There are users using
mbuffer to improve zfs recv performance when piping over the network, so
it seems advantageous to integrate such functionality directly into the
zfs recv tool. This avoids the addition of two buffers and two copies
(one for the buffer mbuffer adds and another for the additional pipe),
so it should be more efficient.

This could have been made configurable and/or this could have changed
the value back to the original after we were done with the file
descriptor, but I do not see a strong case for doing either, so I
went with a simple implementation.

Signed-off-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1161
This commit is contained in:
Richard Yao 2015-03-11 14:24:46 -04:00 committed by Brian Behlendorf
parent b1a3e93217
commit 5c3f61eb49
1 changed files with 43 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

43
lib/libzfs/libzfs_sendrecv.c
View File

@ -43,6 +43,7 @@
#include <sys/mnttab.h>
#include <sys/avl.h>
#include <sys/debug.h>
#include <sys/stat.h>
#include <stddef.h>
#include <pthread.h>
#include <umem.h>
@ -3297,6 +3298,48 @@ zfs_receive(libzfs_handle_t *hdl, const char *tosnap, recvflags_t *flags,
int err;
int cleanup_fd;
uint64_t action_handle = 0;
struct stat sb;
/*
* The only way fstat can fail is if we do not have a valid file
* descriptor.
*/
if (fstat(infd, &sb) == -1) {
perror("fstat");
return (-2);
}
#ifdef __linux__
#ifndef F_SETPIPE_SZ
#define F_SETPIPE_SZ (F_SETLEASE + 7)
#endif /* F_SETPIPE_SZ */
#ifndef F_GETPIPE_SZ
#define F_GETPIPE_SZ (F_GETLEASE + 7)
#endif /* F_GETPIPE_SZ */
/*
* It is not uncommon for gigabytes to be processed in zfs receive.
* Speculatively increase the buffer size via Linux-specific fcntl()
* call.
*/
if (S_ISFIFO(sb.st_mode)) {
FILE *procf = fopen("/proc/sys/fs/pipe-max-size", "r");
if (procf != NULL) {
unsigned long max_psize;
long cur_psize;
if (fscanf(procf, "%lu", &max_psize) > 0) {
cur_psize = fcntl(infd, F_GETPIPE_SZ);
if (cur_psize > 0 &&
max_psize > (unsigned long) cur_psize)
(void) fcntl(infd, F_SETPIPE_SZ,
max_psize);
}
fclose(procf);
}
}
#endif /* __linux__ */
cleanup_fd = open(ZFS_DEV, O_RDWR);
VERIFY(cleanup_fd >= 0);