| History log of /src/lib/libperfuse/debug.c |
| Revision | | Date | Author | Comments |
| 1.13 |
| 16-Nov-2018 |
manu | Use reclaim2 to fix reclaim/lookup race conditions
The PUFFS reclaim operation had a race condition with lookups: we could
be asked to lookup a node, then to reclaim it before lookup completion.
At lookup completion, we would then create a leaked node.
Enter the PUFFS reclaim2 operation, which features a nlookup argument.
That let us count how many lookups are pending and avoid the above
described scenario. It also makes the codes simplier.
|
| 1.12 |
| 21-Jul-2012 |
manu | branches: 1.12.24; 1.12.30; 1.12.32;
- Fix same vnodes associated with multiple cookies
The scheme used to retreive known nodes on lookup was flawed, as it only
used parent and name. This produced a different cookie for the same file
if it was renamed, when looking up ../ or when dealing with multiple files
associated with the same name through link(2).
We therefore abandon the use of node name and introduce hashed lists of
inodes. This causes a huge rewrite of reclaim code, which do not attempt
to keep parents allocated until all their children are reclaimed
- Fix race conditions in reclaim
There are a few situations where we issue multiple FUSE operations for
a PUFFS operation. On reclaim, we therefore have to wait for all FUSE
operation to complete, not just the current exchanges. We do this by
introducing node reference count with node_ref() and node_rele().
- Detect data loss caused by FAF
VOP_PUTPAGES causes FAF writes where the kernel does not check the
operation result. At least issue a warning on error.
- Enjoy FAF shortcut on setattr
No need to wait for the result if the kernel does not want it. There is
however an exception for setattr that touch the size, we need to wait
for completion because we have other operations queued for after the
resize.
- Fix fchmod() on write-open file
fchmod() on a node open with write privilege will send setattr with both mode and size set. This confuses some FUSE filesystem. Therefore we send two FUSE operations, one for mode, and one for size.
- Remove node TTL handling for netbsd-5 for simplicity sake. The code
still builds on netbsd-5 but does not have the node TTL feature anymore.
It works fine with kernel support on netbsd-6.
|
| 1.11 |
| 21-Mar-2012 |
matt | Use C89 function definition
|
| 1.10 |
| 29-Jan-2012 |
dholland | branches: 1.10.2;
Fix 32-bit build
|
| 1.9 |
| 29-Jan-2012 |
manu | Improve FUSE trace facility
|
| 1.8 |
| 29-Dec-2011 |
riz | Redo previous; remove all the casts I added, and use PRI* macros instead.
(by popular demand - makes sense, too)
|
| 1.7 |
| 29-Dec-2011 |
riz | Cast time_t to intmax_t for printf purposes, and format with %j. Fixes
build on amd64 and probably i386 as well.
|
| 1.6 |
| 28-Dec-2011 |
manu | Add a FUSE trace facility, with statistics collection. This should help
tracking bugs and performance issues
|
| 1.5 |
| 03-Oct-2010 |
manu | branches: 1.5.6;
- Correctly handle rename whith overwritten destination
- Keep track of file name to avoid lookups when we can. This makes sure we
do not have two cookies for the same inode, a situation that cause wreak
havoc when we come to remove or rename a node.
- Do not use PUFFS_FLAG_BUILDPATH at all, since we now track file names
- In open, queue requests after checking for access, as there is no merit
to queue a will-be-denied request while we can deny it immediatly
- request reclaim of removed nodes at inactive stage
|
| 1.4 |
| 29-Sep-2010 |
manu | = Open files =
- Restore open on our own in fsycn and readdir, as the node may not already
be open, and FUSE really wants it to be. No need to close immediatly, it
can be done at inactive time.
= Write operations =
- fix a nasty bug that corrupted files on write (written added twice)
- Keep track of file size in order to honour PUFFS_IO_APPEND
= many fixes in rename =
- handler overwritten nodes correctly
- wait for all operations on the node to drain before doing rename, as
filesystems may not cope with operations on a moving file.
- setback PUFFS_SETBACK_INACT_N1 cannot be used from rename, we therefore
miss the inactive time for an overwritten node. This bounds us to give up
PUFFS_KFLAG_IAONDEMAND.
= Removed files =
- forbid most operations on a removed node, return ENOENT
- setback PUFFS_SETBACK_NOREF_N1 at inactive stage to cause removed
file reclaim
= Misc =
- Update outdated ARGSUSED for lint
- Fix a memory leak (puffs_pn_remove instead of puffs_pn_put)
- Do not use PUFFS_FLAG_BUILDPATH except for debug output. It makes the
lookup code much simplier.
|
| 1.3 |
| 23-Sep-2010 |
manu | == file close operations ==
- use PUFFS_KFLAG_WTCACHE to puffs_init so that all writes are
immediatly send to the filesystem, and we do not have anymore write
after inactive. As a consequence, we can close files at inactive
stage, and there is not any concern left with files opened at
create time. We also do not have anymore to open ourselves in readdir and
fsync.
- Fsync on close (inactive stage). That makes sure we will not need to
do these operations once the file is closed (FUSE want an open file).
short sircuit the request that come after the close, bu not fsinc'ing
closed files,
- Use PUFFS_KFLAG_IAONDEMAND to get less inactive calls
== Removed nodes ==
- more ENOENT retunred for operations on removed node (but there
are probably some still missing): getattr, ooen, setattr, fsync
- set PND_REMOVE before sending the UNLINK/RMDIR operations so that we avoid
races during UNLINK completion. Also set PND_REMOVED on node we overwirte
in rename
== Filehandle fixes ==
- queue open operation to avoid getting two fh for one file
- set FH in getattr, if the file is open
- Just requires a read FH for fsyncdir, as we always opendir in read
mode. Ok, this is misleading :-)
== Misc ==
- do not set FUSE_FATTR_ATIME_NOW in setattr, as we provide the time
- short circuit nilpotent operations in setattr
- add a filename diagnostic flag to dump file names
|
| 1.2 |
| 15-Sep-2010 |
manu | - Use SOCK_DGRAM instead of SOCK_STREAM, as the filesystem seems to
assume datagram semantics: when using SOCK_STREAM, if perfused sends
frames faster than the filesystem consumes them, it will grab multiple
frames at once and discard anything beyond the first one. For now the
code can work both with SOCK_DGRAM and SOCK_STREAM, but SOCK_STREAM
support will probably have to be removed for the sake of readability.
- Remeber to sync parent directories when moving a node
- In debug output, display the requeue type (readdir, write, etc...)
|
| 1.1 |
| 25-Aug-2010 |
manu | libperfuse(3) is a PUFFS relay to FUSE. In order to use it,
FUSE filesystem must be patched to #include <perfuse.h> in the source
files that open /dev/fuse and perform the mount(2) system call. The
FUSE filesystem must be linked with -lperfuse.
libperfuse(3) implements the FUSE kernel interface, on which libfuse or
any FUSE filesystem that opens /dev/fuse directly can be used.
For now, an external daemon called perfused(8) is used. This may change
in the future.
|
| 1.5.6.2 |
| 30-Oct-2012 |
yamt | sync with head
|
| 1.5.6.1 |
| 17-Apr-2012 |
yamt | sync with head
|
| 1.10.2.2 |
| 12-Aug-2012 |
martin | Pull up following revision(s) (requested by manu in ticket #438):
lib/libperfuse/perfuse_priv.h: revision 1.31
sys/fs/puffs/puffs_msgif.h: revision 1.80
sys/fs/puffs/puffs_vnops.c: revision 1.171
lib/libpuffs/puffs_ops.3: revision 1.31
sys/fs/puffs/puffs_vnops.c: revision 1.172
sys/fs/puffs/puffs_vnops.c: revision 1.173
sys/fs/puffs/puffs_vnops.c: revision 1.174
usr.sbin/perfused/perfused.c: revision 1.24
sys/fs/puffs/puffs_sys.h: revision 1.80
sys/fs/puffs/puffs_sys.h: revision 1.81
sys/fs/puffs/puffs_sys.h: revision 1.82
lib/libperfuse/subr.c: revision 1.19
lib/libperfuse/perfuse.c: revision 1.30
sys/fs/puffs/puffs_msgif.c: revision 1.90
sys/fs/puffs/puffs_msgif.c: revision 1.91
sys/fs/puffs/puffs_msgif.c: revision 1.92
lib/libperfuse/ops.c: revision 1.59
lib/libpuffs/puffs.3: revision 1.53
lib/libperfuse/debug.c: revision 1.12
lib/libpuffs/puffs.3: revision 1.54
sys/fs/puffs/puffs_vnops.c: revision 1.167
sys/fs/puffs/puffs_msgif.h: revision 1.79
usr.sbin/perfused/msg.c: revision 1.21
sys/fs/puffs/puffs_vfsops.c: revision 1.102
sys/fs/puffs/puffs_vfsops.c: revision 1.103
sys/fs/puffs/puffs_vfsops.c: revision 1.105
lib/libpuffs/puffs.h: revision 1.123
lib/libperfuse/perfuse_if.h: revision 1.20
lib/libperfuse/perfuse.c: revision 1.29
lib/libpuffs/dispatcher.c: revision 1.42
lib/libpuffs/dispatcher.c: revision 1.43
- Fix same vnodes associated with multiple cookies
The scheme used to retreive known nodes on lookup was flawed, as it only
used parent and name. This produced a different cookie for the same file
if it was renamed, when looking up ../ or when dealing with multiple files
associated with the same name through link(2).
We therefore abandon the use of node name and introduce hashed lists of
inodes. This causes a huge rewrite of reclaim code, which do not attempt
to keep parents allocated until all their children are reclaimed
- Fix race conditions in reclaim
There are a few situations where we issue multiple FUSE operations for
a PUFFS operation. On reclaim, we therefore have to wait for all FUSE
operation to complete, not just the current exchanges. We do this by
introducing node reference count with node_ref() and node_rele().
- Detect data loss caused by FAF
VOP_PUTPAGES causes FAF writes where the kernel does not check the
operation result. At least issue a warning on error.
- Enjoy FAF shortcut on setattr
No need to wait for the result if the kernel does not want it. There is
however an exception for setattr that touch the size, we need to wait
for completion because we have other operations queued for after the
resize.
- Fix fchmod() on write-open file
fchmod() on a node open with write privilege will send setattr with both mode
and size set. This confuses some FUSE filesystem. Therefore we send two FUSE
operations, one for mode, and one for size.
- Remove node TTL handling for netbsd-5 for simplicity sake. The code
still builds on netbsd-5 but does not have the node TTL feature anymore.
It works fine with kernel support on netbsd-6.
- Improve PUFFS_KFLAG_CACHE_FS_TTL by reclaiming older inactive nodes.
The normal kernel behavior is to retain inactive nodes in the freelist
until it runs out of vnodes. This has some merit for local filesystems,
where the cost of an allocation is about the same as the cost of a
lookup. But that situation is not true for distributed filesystems.
On the other hand, keeping inactive nodes for a long time hold memory
in the file server process, and when the kernel runs out of vnodes, it
produce reclaim avalanches that increase lattency for other operations.
We do not reclaim inactive vnodes immediatly either, as they may be
looked up again shortly. Instead we introduce a grace time and we
reclaim nodes that have been inactive beyond the grace time.
- Fix lookup/reclaim race condition.
The above improvement undercovered a race condition between lookup and
reclaim. If we reclaimed a vnode associated with a userland cookie while
a lookup returning that same cookiewas inprogress, then the kernel ends
up with a vnode associated with a cookie that has been reclaimed in
userland. Next operation on the cookie will crash (or at least confuse)
the filesystem.
We fix this by introducing a lookup count in kernel and userland. On
reclaim, the kernel sends the count, which enable userland to detect
situation where it initiated a lookup that is not completed in kernel.
In such a situation, the reclaim must be ignored, as the node is about
to be looked up again.
Fix hang unmount bug introduced by last commit.
We introduced a slow queue for delayed reclaims, while the existing
queue for unmount, flush and exist has been renamed fast queue. Both
queues had timestamp for when an operation should be done, but it was
useless for the fast queue, which is always used to run an operation
ASAP. And the timestamp test had an error that turned ASAP into "at next
tick", but nobody what there to wake the thread at next tick, hence
the hang. The fix is to remove the useless and buggy timestamp test for
fast queue.
Rename slow sopreq queue into node sopreq queue, to refet the fact that
is only intended for postponed node reclaims.
When purging the node sopreq queue, do not call puffs_msg_sendresp(), as
it makes no sense.
Fix race condition between (create|mknod|mkdir|symlino) and reclaim, just
like we did it between lookup and reclaim.
Missing bit in previous commit (prevent race between create|mknod|mkdir|symlink
and reclaim)
Bump date for previous.
New sentence, new line; remove trailing whitespace; fix typos;
punctuation nits.
Add PUFFS_KFLAG_CACHE_DOTDOT so that vnodes hold a reference on their
parent, keeping them active, and allowing to lookup .. without sending
a request to the filesystem.
Enable the featuure for perfused, as this is how FUSE works.
Missing bit in previous commit (PUFFS_KFLAG_CACHE_DOTDOT option to avoid
looking up ..)
|
| 1.10.2.1 |
| 23-Apr-2012 |
riz | Pull up following revision(s) (requested by manu in ticket #195):
lib/libskey/skeysubr.c: revision 1.27
lib/libkvm/kvm_getloadavg.c: revision 1.11
lib/libwrap/update.c: revision 1.9
lib/liby/yyerror.c: revision 1.9
lib/libpuffs/puffs_ops.3: revision 1.30
lib/libwrap/misc.c: revision 1.10
lib/libwrap/hosts_access.c: revision 1.20
lib/libpuffs/pnode.c: revision 1.11
lib/libperfuse/subr.c: revision 1.17
lib/libpuffs/pnode.c: revision 1.12
lib/libperfuse/subr.c: revision 1.18
lib/libwrap/options.c: revision 1.15
lib/libwrap/fix_options.c: revision 1.11
lib/libperfuse/ops.c: revision 1.52
lib/libperfuse/ops.c: revision 1.53
lib/libperfuse/ops.c: revision 1.54
lib/libwrap/hosts_ctl.c: revision 1.5
lib/libintl/gettext.c: revision 1.27
lib/libwrap/shell_cmd.c: revision 1.6
lib/libpuffs/dispatcher.c: revision 1.39
lib/libperfuse/perfuse_priv.h: revision 1.27
lib/libwrap/socket.c: revision 1.19
lib/libpuffs/puffs.3: revision 1.50
lib/libperfuse/perfuse_priv.h: revision 1.28
lib/libpuffs/puffs_priv.h: revision 1.45
lib/libpuffs/puffs.3: revision 1.51
lib/libperfuse/perfuse_priv.h: revision 1.29
lib/libwrap/percent_x.c: revision 1.5
lib/libpuffs/puffs.3: revision 1.52
lib/libperfuse/debug.c: revision 1.11
sys/fs/puffs/puffs_vnops.c: revision 1.165
lib/libwrap/tcpd.h: revision 1.13
sys/fs/puffs/puffs_vnops.c: revision 1.166
lib/libwrap/eval.c: revision 1.7
sys/fs/puffs/puffs_msgif.h: revision 1.78
sys/fs/puffs/puffs_vfsops.c: revision 1.101
lib/libwrap/rfc931.c: revision 1.9
lib/libwrap/clean_exit.c: revision 1.5
lib/libpuffs/puffs.h: revision 1.120
lib/libc/stdlib/jemalloc.c: revision 1.27
lib/librmt/rmtlib.c: revision 1.26
lib/libpuffs/puffs.h: revision 1.121
sys/fs/puffs/puffs_sys.h: revision 1.79
lib/librumpclient/rumpclient.c: revision 1.48
lib/libwrap/refuse.c: revision 1.5
lib/libperfuse/perfuse.c: revision 1.26
lib/libperfuse/perfuse.c: revision 1.27
tests/fs/puffs/t_fuzz.c: revision 1.5
lib/libperfuse/perfuse.c: revision 1.28
lib/libpuffs/dispatcher.c: revision 1.40
sys/fs/puffs/puffs_node.c: revision 1.24
lib/libwrap/diag.c: revision 1.9
lib/libintl/textdomain.c: revision 1.13
Use C89 function definition
Add name and atttribute cache with filesytem provided TTL.
lookup, create, mknod, mkdir, symlink, getattr and setattr messages
have been extended so that attributes and their TTL can be provided
by the filesytem. lookup, create, mknod, mkdir, and symlink messages
are also extended so that the filesystem can provide name TTL.
Add PUFFS_KFLAG_CACHE_FS_TTL flag to puffs_init(3) to use name and
attribute cache with filesystem provided TTL.
lookup, create, mknod, mkdir, symlink, getattr and setattr messages
have been extended so that attributes and their TTL can be provided
by the filesytem. lookup, create, mknod, mkdir, and symlink messages
are also extended so that the filesystem can provide name TTL.
The filesystem updates attributes and TTL using
puffs_pn_getvap(3), puffs_pn_getvattl(3), and puffs_pn_getcnttl(3)
Use new PUFFS_KFLAG_CACHE_FS_TTL option to puffs_init(3) so that
FUSE TTL on name and attributes are used. This save many PUFFS
operations and improves performances.
PUFFS_KFLAG_CACHE_FS_TTL is #ifdef'ed in many places for now so that
libperfuse can still be used on netbsd-5.
Split file system.
Comma fixes.
Remove dangling "and".
Bump date for previous.
- Makesure update_va does not change vnode size when it should not. For
instance when doing a fault-issued VOP_GETPAGES within VOP_WRITE, changing
size leads to panic: genfs_getpages: past eof.
-Handle ticks wrap around for vnode name andattribute timeout
- When using PUFFS_KFLAG_CACHE_FS_TTL, do not use puffs_node to carry
attribute and TTL fora newly created node. Instead extend puffs_newinfo
and add puffs_newinfo_setva() and puffs_newinfo_setttl()
- Remove node_mk_common_final in libperfuse. It used to set uid/gid for
a newly created vnode but has been made redundant along time ago since
uid and gid are properly set in FUSE header.
- In libperfuse, check for corner case where opc = 0 on INACTIVE and RECLAIM
(how is it possible? Check for it to avoid a crash anyway)
- In libperfuse, make sure we unlimit RLIMIT_AS and RLIMIT_DATA so that
we do notrun out of memory because the kernel is lazy at reclaiming vnodes.
- In libperfuse, cleanup style of perfuse_destroy_pn()
Do not set PUFFS_KFLAG_CACHE_FS_TTL for PUFFS tests
|
| 1.12.32.1 |
| 10-Jun-2019 |
christos | Sync with HEAD
|
| 1.12.30.1 |
| 26-Nov-2018 |
pgoyette | Sync with HEAD, resolve a couple of conflicts
|
| 1.12.24.1 |
| 10-Feb-2019 |
martin | Pull up following revision(s) (requested by manu in ticket #1186):
lib/libperfuse/perfuse.c: revision 1.41
lib/libperfuse/perfuse_priv.h: revision 1.37
lib/libperfuse/ops.c: revision 1.85
lib/libperfuse/ops.c: revision 1.86
lib/libperfuse/debug.c: revision 1.13
Use reclaim2 to fix reclaim/lookup race conditions
The PUFFS reclaim operation had a race condition with lookups: we could
be asked to lookup a node, then to reclaim it before lookup completion.
At lookup completion, we would then create a leaked node.
Enter the PUFFS reclaim2 operation, which features a nlookup argument.
That let us count how many lookups are pending and avoid the above
described scenario. It also makes the codes simplier.
-
Fix directory filehandle usage with libufse. Fix lookup count
libfuse does not use filehandle the same way for directories and other
objects. As a result, filehandles obtained by OPENDIR should not be
sent on non-directory related operations like READ/WRITE/GETATTR...
While there, fix the lookup count sent to the FORGET operation, which
led to leaked nodes.
|
