these two files that are almost-but-not-quite the same leads to false grep
hits, confusion etc.
Only installing one copy with a symlink would be nice but that doesn't
work with SHARED=symlinks (it changes the source tree).
in nfs_vinvalbuf() or the nfs_removeit(), we can have the nfsnode reallocated
from underneath us (eg: replaced by a ufs 'struct inode') which can cause
disk corruption ('freeing free block' when di_db[5] gets trashed).
This is not a cheap fix, but it'll do until the nfsnodes get reference
counting and/or locking.
Apparently NetBSD have a similar fix (apparently from BSDI).
I wish all PR's had this much useful detail. :-)
PR: 6611
Submitted by: Stephen Clawson <sclawson@marker.cs.utah.edu>
---------
Make callers of namei() responsible for releasing references or locks
instead of having the underlying filesystems do it. This eliminates
redundancy in all terminal filesystems and makes it possible for stacked
transport layers such as umapfs or nullfs to operate correctly.
Quality testing was done with testvn, and lat_fs from the lmbench suite.
Some NFS client testing courtesy of Patrik Kudo.
vop_mknod and vop_symlink still release the returned vpp. vop_rename
still releases 4 vnode arguments before it returns. These remaining cases
will be corrected in the next set of patches.
---------
Submitted by: Michael Hancock <michaelh@cet.co.jp>
Reverse the VFS_VRELE patch. Reference counting of vnodes does not need
to be done per-fs. I noticed this while fixing vfs layering violations.
Doing reference counting in generic code is also the preference cited by
John Heidemann in recent discussions with him.
The implementation of alternative vnode management per-fs is still a valid
requirement for some filesystems but will be revisited sometime later,
most likely using a different framework.
Submitted by: Michael Hancock <michaelh@cet.co.jp>
"time" wasn't a atomic variable, so splfoo() protection were needed
around any access to it, unless you just wanted the seconds part.
Most uses of time.tv_sec now uses the new variable time_second instead.
gettime() changed to getmicrotime(0.
Remove a couple of unneeded splfoo() protections, the new getmicrotime()
is atomic, (until Bruce sets a breakpoint in it).
A couple of places needed random data, so use read_random() instead
of mucking about with time which isn't random.
Add a new nfs_curusec() function.
Mark a couple of bogosities involving the now disappeard time variable.
Update ffs_update() to avoid the weird "== &time" checks, by fixing the
one remaining call that passwd &time as args.
Change profiling in ncr.c to use ticks instead of time. Resolution is
the same.
Add new function "tvtohz()" to avoid the bogus "splfoo(), add time, call
hzto() which subtracts time" sequences.
Reviewed by: bde
has been some bitrot and incorrect assumptions in the vfs_bio code. These
problems have manifest themselves worse on NFS type filesystems, but can
still affect local filesystems under certain circumstances. Most of
the problems have involved mmap consistancy, and as a side-effect broke
the vfs.ioopt code. This code might have been committed seperately, but
almost everything is interrelated.
1) Allow (pmap_object_init_pt) prefaulting of buffer-busy pages that
are fully valid.
2) Rather than deactivating erroneously read initial (header) pages in
kern_exec, we now free them.
3) Fix the rundown of non-VMIO buffers that are in an inconsistent
(missing vp) state.
4) Fix the disassociation of pages from buffers in brelse. The previous
code had rotted and was faulty in a couple of important circumstances.
5) Remove a gratuitious buffer wakeup in vfs_vmio_release.
6) Remove a crufty and currently unused cluster mechanism for VBLK
files in vfs_bio_awrite. When the code is functional, I'll add back
a cleaner version.
7) The page busy count wakeups assocated with the buffer cache usage were
incorrectly cleaned up in a previous commit by me. Revert to the
original, correct version, but with a cleaner implementation.
8) The cluster read code now tries to keep data associated with buffers
more aggressively (without breaking the heuristics) when it is presumed
that the read data (buffers) will be soon needed.
9) Change to filesystem lockmgr locks so that they use LK_NOPAUSE. The
delay loop waiting is not useful for filesystem locks, due to the
length of the time intervals.
10) Correct and clean-up spec_getpages.
11) Implement a fully functional nfs_getpages, nfs_putpages.
12) Fix nfs_write so that modifications are coherent with the NFS data on
the server disk (at least as well as NFS seems to allow.)
13) Properly support MS_INVALIDATE on NFS.
14) Properly pass down MS_INVALIDATE to lower levels of the VM code from
vm_map_clean.
15) Better support the notion of pages being busy but valid, so that
fewer in-transit waits occur. (use p->busy more for pageouts instead
of PG_BUSY.) Since the page is fully valid, it is still usable for
reads.
16) It is possible (in error) for cached pages to be busy. Make the
page allocation code handle that case correctly. (It should probably
be a printf or panic, but I want the system to handle coding errors
robustly. I'll probably add a printf.)
17) Correct the design and usage of vm_page_sleep. It didn't handle
consistancy problems very well, so make the design a little less
lofty. After vm_page_sleep, if it ever blocked, it is still important
to relookup the page (if the object generation count changed), and
verify it's status (always.)
18) In vm_pageout.c, vm_pageout_clean had rotted, so clean that up.
19) Push the page busy for writes and VM_PROT_READ into vm_pageout_flush.
20) Fix vm_pager_put_pages and it's descendents to support an int flag
instead of a boolean, so that we can pass down the invalidate bit.
a complement to all ops that return a vpp, VFS_VRELE. This is
initially only for file systems that implement the following ops
that do a WILLRELE:
vop_create, vop_whiteout, vop_mknod, vop_remove, vop_link,
vop_rename, vop_mkdir, vop_rmdir, vop_symlink
This is initial DNA that doesn't do anything yet. VFS_VRELE is
implemented but not called.
A default vfs_vrele was created for fs implementations that use the
standard vnode management routines.
VFS_VRELE implementations were made for the following file systems:
Standard (vfs_vrele)
ffs mfs nfs msdosfs devfs ext2fs
Custom
union umapfs
Just EOPNOTSUPP
fdesc procfs kernfs portal cd9660
These implementations may change as VOP changes are implemented.
In the next phase, in the vop implementations calls to vrele and the vrele
part of vput will be moved to the top layer vfs_vnops and made visible
to all layers. vput will be replaced by unlock in these cases. Unlocking
will still be done in the per fs layer but the refcount decrement will be
triggered at the top because it doesn't hurt to hold a vnode reference a
little longer. This will have minimal impact on the structure of the
existing code.
This will only be done for vnode arguments that are released by the various
fs vop implementations.
Wider use of VFS_VRELE will likely require restructuring of the code.
Reviewed by: phk, dyson, terry et. al.
Submitted by: Michael Hancock <michaelh@cet.co.jp>
to give pollution compatible with <nfs/nqfs.h>. At least mount_nfs.c
previously had to #define KERNEL before including <nfs/nfs.h> to get
this pollution, but this gave other pollution.
Moved comment about NFSINT_SIGMASK to immediately before the code that
it applies to.
depend on <sys/types.h> forward declaring common ones.
Added an underscore to `sin' in prototypes to avoid warnings for the
conflict with the ANSI sin().
MUST be PG_BUSY. It is bogus to free a page that isn't busy,
because it is in a state of being "unavailable" when being
freed. The additional advantage is that the page_remove code
has a better cross-check that the page should be busy and
unavailable for other use. There were some minor problems
with the collapse code, and this plugs those subtile "holes."
Also, the vfs_bio code wasn't checking correctly for PG_BUSY
pages. I am going to develop a more consistant scheme for
grabbing pages, busy or otherwise. For now, we are stuck
with the current morass.
Make vfs_bio buffer mgmt work better.
Buffers were being used after brelse.
Make nfs_getpages work independently of other NFS
interfaces. This eliminates some difficult
recursion problems and decreases pagefault
overhead.
Remove an erroneous vfs_unbusy_pages.
Fix a reentrancy problem, with nfs_vinvalbuf when
vnode is already being rundown.
Reassignbuf wasn't being called when needed under
certain circumstances.
(Thanks to Bill Paul for help.)
Make vfs_bio buffer mgmt work better.
Buffers were being used after brelse.
Make nfs_getpages work independently of other NFS
interfaces. This eliminates some difficult
recursion problems and decreases pagefault
overhead.
Remove an erroneous vfs_unbusy_pages.
Fix a reentrancy problem, with nfs_vinvalbuf when
vnode is already being rundown.
Reassignbuf wasn't being called when needed under
certain circumstances.
(Thanks for help from Bill Paul.)
This will not make any of object files that LINT create change; there
might be differences with INET disabled, but hardly anything compiled
before without INET anyway. Now the 'obvious' things will give a
proper error if compiled without inet - ipx_ip, ipfw, tcp_debug. The
only thing that _should_ work (but can't be made to compile reasonably
easily) is sppp :-(
This commit move struct arpcom from <netinet/if_ether.h> to
<net/if_arp.h>.
original BSD code. The association between the vnode and the vm_object
no longer includes reference counts. The major difference is that
vm_object's are no longer freed gratuitiously from the vnode, and so
once an object is created for the vnode, it will last as long as the
vnode does.
When a vnode object reference count is incremented, then the underlying
vnode reference count is incremented also. The two "objects" are now
more intimately related, and so the interactions are now much less
complex.
When vnodes are now normally placed onto the free queue with an object still
attached. The rundown of the object happens at vnode rundown time, and
happens with exactly the same filesystem semantics of the original VFS
code. There is absolutely no need for vnode_pager_uncache and other
travesties like that anymore.
A side-effect of these changes is that SMP locking should be much simpler,
the I/O copyin/copyout optimizations work, NFS should be more ponderable,
and further work on layered filesystems should be less frustrating, because
of the totally coherent management of the vnode objects and vnodes.
Please be careful with your system while running this code, but I would
greatly appreciate feedback as soon a reasonably possible.
of vnodes and objects. There are some metadata performance improvements
that come along with this. There are also a few prototypes added when
the need is noticed. Changes include:
1) Cleaning up vref, vget.
2) Removal of the object cache.
3) Nuke vnode_pager_uncache and friends, because they aren't needed anymore.
4) Correct some missing LK_RETRY's in vn_lock.
5) Correct the page range in the code for msync.
Be gentle, and please give me feedback asap.
nfs_bio.c code worked better than the 1.44. This commit reverts
the important parts of 1.44 to 1.41, and we will fix it when we
can get a handle on the problem.
a mistake (since softnet interrupts may occur if malloc() waits),
and doing it harmlessly but unnecessarily here interfered with
detection of the mistaken cases.
Ever since I first say the way the mount flags were used I've hated the
fact that modes, and events, internal and exported, and short-term
and long term flags are all thrown together. Finally it's annoyed me enough..
This patch to the entire FreeBSD tree adds a second mount flag word
to the mount struct. it is not exported to userspace. I have moved
some of the non exported flags over to this word. this means that we now
have 8 free bits in the mount flags. There are another two that might
well move over, but which I'm not sure about.
The only user visible change would have been in pstat -v, except
that davidg has disabled it anyhow.
I'd still like to move the state flags and the 'command' flags
apart from each other.. e.g. MNT_FORCE really doesn't have the
same semantics as MNT_RDONLY, but that's left for another day.
it in struct proc instead.
This fixes a boatload of compiler warning, and removes a lot of cruft
from the sources.
I have not removed the /*ARGSUSED*/, they will require some looking at.
libkvm, ps and other userland struct proc frobbing programs will need
recompiled.
Rename vn_default_error to vop_defaultop all over the place.
Move vn_bwrite from vfs_bio.c to vfs_default.c and call it vop_stdbwrite.
Use vop_null instead of nullop.
Move vop_nopoll from vfs_subr.c to vfs_default.c
Move vop_sharedlock from vfs_subr.c to vfs_default.c
Move vop_nolock from vfs_subr.c to vfs_default.c
Move vop_nounlock from vfs_subr.c to vfs_default.c
Move vop_noislocked from vfs_subr.c to vfs_default.c
Use vop_ebadf instead of *_ebadf.
Add vop_defaultop for getpages on master vnode in MFS.
1. Add defaults for more VOPs
VOP_LOCK vop_nolock
VOP_ISLOCKED vop_noislocked
VOP_UNLOCK vop_nounlock
and remove direct reference in filesystems.
2. Rename the nfsv2 vnop tables to improve sorting order.
1. Remove VOP_UPDATE, it is (also) an UFS/{FFS,LFS,EXT2FS,MFS}
intereface function, and now lives in the ufsmount structure.
2. Remove VOP_SEEK, it was unused.
3. Add mode default vops:
VOP_ADVLOCK vop_einval
VOP_CLOSE vop_null
VOP_FSYNC vop_null
VOP_IOCTL vop_enotty
VOP_MMAP vop_einval
VOP_OPEN vop_null
VOP_PATHCONF vop_einval
VOP_READLINK vop_einval
VOP_REALLOCBLKS vop_eopnotsupp
And remove identical functionality from filesystems
4. Add vop_stdpathconf, which returns the canonical stuff. Use
it in the filesystems. (XXX: It's probably wrong that specfs
and fifofs sets this vop, shouldn't it come from the "host"
filesystem, for instance ufs or cd9660 ?)
5. Try to make system wide VOP functions have vop_* names.
6. Initialize the um_* vectors in LFS.
(Recompile your LKMS!!!)
1. Add new file "sys/kern/vfs_default.c" where default actions for
VOPs go. Implement proper defaults for ABORTOP, BWRITE, LEASE,
POLL, REVOKE and STRATEGY. Various stuff spread over the entire
tree belongs here.
2. Change VOP_BLKATOFF to a normal function in cd9660.
3. Kill VOP_BLKATOFF, VOP_TRUNCATE, VOP_VFREE, VOP_VALLOC. These
are private interface functions between UFS and the underlying
storage manager layer (FFS/LFS/MFS/EXT2FS). The functions now
live in struct ufsmount instead.
4. Remove a kludge of VOP_ functions in all filesystems, that did
nothing but obscure the simplicity and break the expandability.
If a filesystem doesn't implement VOP_FOO, it shouldn't have an
entry for it in its vnops table. The system will try to DTRT
if it is not implemented. There are still some cruft left, but
the bulk of it is done.
5. Fix another VCALL in vfs_cache.c (thanks Bruce!)
1. Use the default function to access all the specfs operations.
2. Use the default function to access all the fifofs operations.
3. Use the default function to access all the ufs operations.
4. Fix VCALL usage in vfs_cache.c
5. Use VOCALL to access specfs functions in devfs_vnops.c
6. Staticize most of the spec and fifofs vnops functions.
7. Make UFS panic if it lacks bits of the underlying storage handling.
1. Remove comment stating the blatantly obvious.
2. Align in two columns.
3. Sort all but the default element alphabetically.
4. Remove XXX comments pointing out entries not needed.
Distribute all but the most fundamental malloc types. This time I also
remembered the trick to making things static: Put "static" in front of
them.
A couple of finer points by: bde
plus the previous changes to use the zone allocator decrease the useage
of malloc by half. The Zone allocator will be upgradeable to be able
to use per CPU-pools, and has more intelligent usage of SPLs. Additionally,
it has reasonable stats gathering capabilities, while making most calls
inline.
This unifies several times in theory indentical 50 lines of code.
The filesystems have a new method: vop_cachedlookup, which is the
meat of the lookup, and use vfs_cache_lookup() for their vop_lookup
method. vfs_cache_lookup() will check the namecache and pass on
to the vop_cachedlookup method in case of a miss.
It's still the task of the individual filesystems to populate the
namecache with cache_enter().
Filesystems that do not use the namecache will just provide the
vop_lookup method as usual.
socket addresses in mbufs. (Socket buffers are the one exception.) A number
of kernel APIs needed to get fixed in order to make this happen. Also,
fix three protocol families which kept PCBs in mbufs to not malloc them
instead. Delete some old compatibility cruft while we're at it, and add
some new routines in the in_cksum family.
code that says this:
nfsm_request(vp, NFSPROC_FSSTAT, p, cred);
if (v3)
nfsm_postop_attr(vp, retattr);
if (!error)
nfsm_dissect(sfp, struct nfs_statfs *, NFSX_STATFS(v3));
The problem here is that if error != 0, nfsm_dissect() will not be
called, which leaves sfp == NULL. But nfs_statfs() does not bail out
at this point: it continues processing until it tries to dereference
sfp, which causes a panic. I was able to generate this crash under
the following conditions:
1) Set up a machine as an NFS server and NFS client, with amd running
(using NIS maps). /usr/local is exported, though any exported fs
can can be used to trigger the bug.
2) Log in as normal user, with home directory mounted from a SunOS 4.1.3
NFS server via amd (along with a few other NFS filesystems from same
machine).
3) Su to root and type the following:
# mount localhost:/usr/local /mnt
# df
To fix the panic, I changed the code to read:
if (!error) {
nfsm_dissect(sfp, struct nfs_statfs *, NFSX_STATFS(v3));
} else
goto nfsmout;
This is a bit kludgy in that nfsmout is a label defined by the nfsm_subs.h
macros, but these macros are themselves more than a little kludgy. This
stops the machine from crashing, but does not fix the overall bug: 'error'
somehow becomes 5 (EIO) when a statfs() is performed on the locally mounted
NFS filesystem. This seems to only happen the first time the filesystem
is accesed: on subsequent accesses, it seems to work fine again.
Now, I know there's no practical use in mounting a local filesystem
via NFS, but doing it shouldn't cause the system to melt down.
writes sent to the server were synchronous and therefore no commits are
needed. This is the same as the vfs.nfs.async variable on the server but
allows each client to choose whether to work this way.
Also make the vfs.nfs.async variable do the 'right' thing for NFSv3, i.e.
pretend that the write was synchronous.
attached to the vnode, some of them could be re-written synchronously
(if they overflowed the fixed size array nfs_flush had for them). The
fix involves mallocing an array if there are more than its limited
size stack buffer.
Reviewed by: Hidetoshi Shimokawa <simokawa@sat.t.u-tokyo.ac.jp>
and b_validend. The changes to vfs_bio.c are a bit ugly but hopefully
can be tidied up later by a slight redesign.
PR: kern/2573, kern/2754, kern/3046 (possibly)
Reviewed by: dyson
compatible with boot proms made from the 2.2 source.
Convert the nfs arguments when copying to the new diskless structure.
Copy the gateway field in the diskless structure.
to fill in the nfs_diskless structure, at the cost of some kernel
bloat. The advantage is that this code works on a wider range of
network adapters than netboot. Several new kernel options are
documented in LINT.
Obtained from: parts of the code comes from NetBSD.
accessing files which it shouldn't be able to. This required a better
approximation of VOP_ACCESS for NFSv2 (NFSv3 already has an ACCESS rpc
which is a better solution) and adding a call to VOP_ACCESS from VOP_LOOKUP.
PR: kern/876, kern/2635
Submitted by: David Malone <dwmalone@maths.tcd.ie> (for kern/2635)
".." vnode. This is cheaper storagewise than keeping it in the
namecache, and it makes more sense since it's a 1:1 mapping.
2. Also handle the case of "." more intelligently rather than stuff
the namecache with pointless entries.
3. Add two lists to the vnode and hang namecache entries which go from
or to this vnode. When cleaning a vnode, delete all namecache
entries it invalidates.
4. Never reuse namecache enties, malloc new ones when we need it, free
old ones when they die. No longer a hard limit on how many we can
have.
5. Remove the upper limit on namelength of namecache entries.
6. Make a global list for negative namecache entries, limit their number
to a sysctl'able (debug.ncnegfactor) fraction of the total namecache.
Currently the default fraction is 1/16th. (Suggestions for better
default wanted!)
7. Assign v_id correctly in the face of 32bit rollover.
8. Remove the LRU list for namecache entries, not needed. Remove the
#ifdef NCH_STATISTICS stuff, it's not needed either.
9. Use the vnode freelist as a true LRU list, also for namecache accesses.
10. Reuse vnodes more aggresively but also more selectively, if we can't
reuse, malloc a new one. There is no longer a hard limit on their
number, they grow to the point where we don't reuse potentially
usable vnodes. A vnode will not get recycled if still has pages in
core or if it is the source of namecache entries (Yes, this does
indeed work :-) "." and ".." are not namecache entries any longer...)
11. Do not overload the v_id field in namecache entries with whiteout
information, use a char sized flags field instead, so we can get
rid of the vpid and v_id fields from the namecache struct. Since
we're linked to the vnodes and purged when they're cleaned, we don't
have to check the v_id any more.
12. NFS knew about the limitation on name length in the namecache, it
shouldn't and doesn't now.
Bugs:
The namecache statistics no longer includes the hits for ".."
and "." hits.
Performance impact:
Generally in the +/- 0.5% for "normal" workstations, but
I hope this will allow the system to be selftuning over a
bigger range of "special" applications. The case where
RAM is available but unused for cache because we don't have
any vnodes should be gone.
Future work:
Straighten out the namecache statistics.
"desiredvnodes" is still used to (bogusly ?) size hash
tables in the filesystems.
I have still to find a way to safely free unused vnodes
back so their number can shrink when not needed.
There is a few uses of the v_id field left in the filesystems,
scheduled for demolition at a later time.
Maybe a one slot cache for unused namecache entries should
be implemented to decrease the malloc/free frequency.
This commit includes the following changes:
1) Old-style (pr_usrreq()) protocols are no longer supported, the compatibility
glue for them is deleted, and the kernel will panic on boot if any are compiled
in.
2) Certain protocol entry points are modified to take a process structure,
so they they can easily tell whether or not it is possible to sleep, and
also to access credentials.
3) SS_PRIV is no more, and with it goes the SO_PRIVSTATE setsockopt()
call. Protocols should use the process pointer they are now passed.
4) The PF_LOCAL and PF_ROUTE families have been updated to use the new
style, as has the `raw' skeleton family.
5) PF_LOCAL sockets now obey the process's umask when creating a socket
in the filesystem.
As a result, LINT is now broken. I'm hoping that some enterprising hacker
with a bit more time will either make the broken bits work (should be
easy for netipx) or dike them out.
wind up writing zeros instead of real data when the file is on an NFSv2
mounted directory.
While tracking this bug down, I noticed that nfs_asyncio was waking *all*
the iods when a block was written instead of just one per block. Fixing this
gives a 25% performance improvment for writes on v2 (less for v3).
Both are 2.2 candidates.
PR: kern/2774
Zero the b_dirty{off,end} after cluster-comitting a group of buffers.
With these fixes, I was able to complete a 'make world' with remote src
and obj directories.
Use the same value of 512 (ufs actually uses DEV_BSIZE). There are
too many versions of DIRBLKSIZ, one for ufs, one for ext2fs, one for
nfs, one for ibcs2, one for linux, one for applications, ... I think
nfs's DIRBLKSIZ needs to be a divisor of the directory blocks sizes
of all supported file systems. There is also NFS_DIRBLKSIZ, which is
different from nfs's DIRBLKSIZ but is sometimes confused with it in
comments.
Removed a bogus #ifdef KERNEL that hid the tunable constants for nfs.
This came in undocumented with the Lite2 merge although it isn't in
Lite2. It required more-bogus #define KERNEL's in fstat and pstat
to make the constants visible.
Restored a spelling fix from rev.1.17.
Removed duplicate #defines of all the the NFS mount option flags.
form `tv = time'. Use a new function gettime(). The current version
just forces atomicicity without fixing precision or efficiency bugs.
Simplified some related valid accesses by using the central function.
missing-parentheses bug, but this exposed a misplaced vfs_busy_pages().
This bug cost a factor of 2.5-3 in nfsv3 write performance! It should
be fixed in 2.2.
Removed some debugging code that gets triggered often in normal
operation. There are still many backwards diagnostics (#define
DIAGNOSTIC gives no diagnostics).
Submitted by: vfs_busy_pages() fix by dfr
changes, so don't expect to be able to run the kernel as-is (very well)
without the appropriate Lite/2 userland changes.
The system boots and can mount UFS filesystems.
Untested: ext2fs, msdosfs, NFS
Known problems: Incorrect Berkeley ID strings in some files.
Mount_std mounts will not work until the getfsent
library routine is changed.
Reviewed by: various people
Submitted by: Jeffery Hsu <hsu@freebsd.org>
This will make a number of things easier in the future, as well as (finally!)
avoiding the Id-smashing problem which has plagued developers for so long.
Boy, I'm glad we're not using sup anymore. This update would have been
insane otherwise.
if you do:
% cd /nfsdir
% mkdir -p foo/foo
% mv foo/foo .
nfs_sillyrename() self-destructs if you try to sillyrename a directory,
however nfs_rename() can be coerced into doing just that by the above
sequence of commands. To avoid this, nfs_rename() now checks that
v_type of the 'destination' vnode != VDIR before attempting the
sillyrename. The server correctly handles this particular situation
by returning ENOTEMPTY on the rename() attempt.
I asked if this was the correct fix for this on -hackers but nobody
ever answered.
This is a 2.2 candidate.
to TAILQs. Fix places which referenced these for no good reason
that I can see (the references remain, but were fixed to compile
again; they are still questionable).
existing mechanism uses a global queue for some buffers and the
vp->b_dirtyblkhd queue for others. This turns sequential writes into
randomly ordered writes to the server, affecting both read and write
performance. The existing mechanism also copes badly with hung
servers, tending to block accesses to other servers when all the iods
are waiting for a hung server.
The new mechanism uses a queue for each mount point. All asynchronous
i/o goes through this queue which preserves the ordering of requests.
A simple mechanism ensures that the iods are shared out fairly between
active mount points. This removes the sysctl variable vfs.nfs.dwrite
since the new queueing mechanism removes the old delayed write code
completely.
This should go into the 2.2 branch.
contents are discarded, including the cached seek cookies.
Unfortunately, if the directory was larger than NFS_DIRBLKSIZ, then
this confused nfs_readdirrpc(), making it appear as if the directory
was truncated.
Reviewed by: Karl Denninger <karl@Mcs.Net>
if a single process is performing a large number of requests (in this
case writing a large file). The writing process could monopolise the
recieve lock and prevent any other processes from recieving their
replies.
It also adds a new sysctl variable 'vfs.nfs.dwrite' which controls the
behaviour which originally pointed out the problem. When a process
writes to a file over NFS, it usually arranges for another process
(the 'iod') to perform the request. If no iods are available, then it
turns the write into a 'delayed write' which is later picked up by the
next iod to do a write request for that file. This can cause that
particular iod to do a disproportionate number of requests from a
single process which can harm performance on some NFS servers. The
alternative is to perform the write synchronously in the context of
the original writing process if no iod is avaiable for asynchronous
writing.
The 'delayed write' behaviour is selected when vfs.nfs.dwrite=1 and
the non-delayed behaviour is selected when vfs.nfs.dwrite=0. The
default is vfs.nfs.dwrite=1; if many people tell me that performance
is better if vfs.nfs.dwrite=0 then I will change the default.
Submitted by: Hidetoshi Shimokawa <simokawa@sat.t.u-tokyo.ac.jp>
/*
* Structure defined by POSIX.4 to be like a timeval.
*/
struct timespec {
time_t ts_sec; /* seconds */
long ts_nsec; /* and nanoseconds */
};
The correct names of the fields are tv_sec and tv_nsec.
Reminded by: James Drobina <jdrobina@infinet.com>
The interface into the "VMIO" system has changed to be more consistant
and robust. Essentially, it is now no longer necessary to call vn_open
to get merged VM/Buffer cache operation, and exceptional conditions
such as merged operation of VBLK devices is simpler and more correct.
This code corrects a potentially large set of problems including the
problems with ktrace output and loaded systems, file create/deletes,
etc.
Most of the changes to NFS are cosmetic and name changes, eliminating
a layer of subroutine calls. The direct calls to vput/vrele have
been re-instituted for better cross platform compatibility.
Reviewed by: davidg
rick@snowhite.cis.uoguelph.ca:
1. Clear B_NEEDCOMMIT in nfs_write to make sure that dirty data is
correctly send to the server. If a buffer was dirtied when it was in
the B_DELWRI+B_NEEDCOMMIT state, the state of the buffer was left
unchanged and when the buffer was later cleaned, just a commit rpc was
made to the server to complete the previous write. Clearing
B_NEEDCOMMIT ensures that another write is made to the server.
2. If a server returned a server (for whatever reason) returned an
answer to a write RPC that implied that fewer bytes than requested
were written, bad things would happen.
3. The setattr operation passed on the atime in stead of the mtime to
the server. The fix is trivial.
4. XIDs always started at 0, but this caused some servers (older DEC
OSF/1 3.0 so I've been told) who had very long-lasting XID caches to
get confused if, after a reboot of a BSD client, RPCs came in with a
XID that had in the past been used before from that client. Patch is
to use the current time in seconds as a starting point for XIDs. The
patch below is not perfect, because it requires the root fs to be
mounted first. This is because of the check BSD systems do, comparing
FS time to system time.
Reviewed by: Bruce Evans, Terry Lambert.
Obtained from: frank@fwi.uva.nl (Frank van der Linden) via rick@snowhite.cis.uoguelph.ca
pr_usrreq mechanism which was poorly designed and error-prone. This
commit renames pr_usrreq to pr_ousrreq so that old code which depended on it
would break in an obvious manner. This commit also implements the new
interface for TCP, although the old function is left as an example
(#ifdef'ed out). This commit ALSO fixes a longstanding bug in the
TCP timer processing (introduced by davidg on 1995/04/12) which caused
timer processing on a TCB to always stop after a single timer had
expired (because it misinterpreted the return value from tcp_usrreq()
to indicate that the TCB had been deleted). Finally, some code
related to polling has been deleted from if.c because it is not
relevant t -current and doesn't look at all like my current code.
Poul mentioned that he thought this was some kind of timing problem, and
that started me thinking. After a little poking around, I found that
nfs_timer() was completely disabled when NFS_NOSERVER was #defined.
But after looking at nfs_timer(), it seemed like it was something
required by both the client and server code, and disabling it outright
just didn't seem to make any sense. Parts of it relate only to the
NFS server side code, so I disabled those, but I re-enabled the rest
of the function and made sure that it would be called from nfs_init()
(in nfs_subs.c).
With nfs_timer() re-enabled, everything seems to work again. The only
other changes I made were to #ifdef away some variable declarations
in the NFS_NOSERVER case so that gcc would stop complaining about
unused variables.
Reviewed by: phk
Submitted by: Bill Paul <wpaul@skynet.ctr.columbia.edu>
process won't possibly block before filling in the fsnode pointer (v_data)
which might be dereferenced during a sync since the vnode is put on the
mnt_vnodelist by getnewvnode.
Pointed out by Matt Day <mday@artisoft.com>
released until the file system was unmounted. This bug also affected
kern/vfs_syscalls.c but was fixed in rev.1.18 and rev.1.20 there.
Reviewed by: davidg
a panic due to an attaempt to allocate a buffer for a terabyte or
so of data when an attempt is made to create sparse data (e.g.
a holey file) more than 1 block past the end of the file.
Note: some other areas of this code need to be looked at,
since they might cause problems when the file size exceeds 2GB,
due to storing results in ints when the computations are being
done with quad sized variables.
Reviewed by: bde
Speed up for vfs_bio -- addition of a routine bqrelse to greatly diminish
overhead for merged cache.
Efficiency improvement for vfs_cluster. It used to do alot of redundant
calls to cluster_rbuild.
Correct the ordering for vrele of .text and release of credentials.
Use the selective tlb update for 486/586/P6.
Numerous fixes to the size of objects allocated for files. Additionally,
fixes in the various pagers.
Fixes for proper positioning of vnode_pager_setsize in msdosfs and ext2fs.
Fixes in the swap pager for exhausted resources. The pageout code
will not as readily thrash.
Change the page queue flags (PG_ACTIVE, PG_INACTIVE, PG_FREE, PG_CACHE) into
page queue indices (PQ_ACTIVE, PQ_INACTIVE, PQ_FREE, PQ_CACHE),
thereby improving efficiency of several routines.
Eliminate even more unnecessary vm_page_protect operations.
Significantly speed up process forks.
Make vm_object_page_clean more efficient, thereby eliminating the pause
that happens every 30seconds.
Make sequential clustered writes B_ASYNC instead of B_DELWRI even in the
case of filesystems mounted async.
Fix a panic with busy pages when write clustering is done for non-VMIO
buffers.
structs and prototypes for syscalls.
Ifdefed duplicated decentralized declarations of args structs. It's
convenient to have this visible but they are hard to maintain. Some
are already different from the central declarations. 4.4lite2 puts
them in comments in the function headers but I wanted to avoid the
large changes for that.
it 1138 times (:-() in casts and a few more times in declarations.
This change is null for the i386.
The type has to be `typedef int vop_t(void *)' and not `typedef
int vop_t()' because `gcc -Wstrict-prototypes' warns about the
latter. Since vnode op functions are called with args of different
(struct pointer) types, neither of these function types is any use
for type checking of the arg, so it would be preferable not to use
the complete function type, especially since using the complete
type requires adding 1138 casts to avoid compiler warnings and
another 40+ casts to reverse the function pointer conversions before
calling the functions.
These functions went away:
enosys (hasn't been used for some time)
enxio
enodev
enoioctl (was used only once, actually for a vop)
if_tun.c:
Continued cleaning up...
conf.h:
Probably fixed the type of d_reset_t. It is hard to tell the correct
type because there are no non-dummy device reset functions.
Removed last vestige of ambiguous sleep message strings.
filesystem layer, as was done in lite-2. Merged in some other cosmetic
changes while I was at it. Rewrote most of msdosfs_access() to be more
like ufs_access() and to include the FS read-only check.
Obtained from: partially from 4.4BSD-lite2
nfsm_rpchead() has been called with the wrong number of args and misplaced
args since someone added new args in the middle for nfsv3.
Here's another one that would be important on 64-bit systems. VOP_READDIR
takes a `u_int **cookies' arg.
Submitted by: Bruce Evans <bde@zeta.org.au>
Detect in nfsrv_readdirplus when a filesystem soes not support VFS_VGET and
return NFSERR_NOTSUPP so that the client will use ordinary readdir instead.
wrong vp's ops vector being used by changing the VOP_LINK's argument order.
The special-case hack doesn't go far enough and breaks the generic
bypass routine used in some non-leaf filesystems. Pointed out by Kirk
McKusick.
when it is moved to an NFS filesystem from from another filesystem and /bin/mv
failed to set the file ownership during the move.
I believe that this bug is present in STABLE but I have not tested it. The fix
would be the same in STABLE even though the code has changed quite considerably
in CURRENT.
the problem "when a file is truncated on the server after being written on
a client under NFSv3, the client doesn't see the size drop to zero".
(As you noted, the problem is that NMODIFIED wasn't being cleared by nfs_close
when it flushed the buffers. After checking through the code, the only place
where NMODIFIED was used to test for the possibility of dirty blocks was in
nfs_setattr(). The two cases are safe to do when there aren't dirty blocks,
so I just took out the tests. Unfortunately, testing for
v_dirtyblkhd.lh_first being non-null is not sufficient, since there are
times when the code moves blocks to the clean list and then back to the
dirty list.)
Submitted by: rick@snowhite.cis.uoguelph.ca
proc or any VM system structure will have to be rebuilt!!!
Much needed overhaul of the VM system. Included in this first round of
changes:
1) Improved pager interfaces: init, alloc, dealloc, getpages, putpages,
haspage, and sync operations are supported. The haspage interface now
provides information about clusterability. All pager routines now take
struct vm_object's instead of "pagers".
2) Improved data structures. In the previous paradigm, there is constant
confusion caused by pagers being both a data structure ("allocate a
pager") and a collection of routines. The idea of a pager structure has
escentially been eliminated. Objects now have types, and this type is
used to index the appropriate pager. In most cases, items in the pager
structure were duplicated in the object data structure and thus were
unnecessary. In the few cases that remained, a un_pager structure union
was created in the object to contain these items.
3) Because of the cleanup of #1 & #2, a lot of unnecessary layering can now
be removed. For instance, vm_object_enter(), vm_object_lookup(),
vm_object_remove(), and the associated object hash list were some of the
things that were removed.
4) simple_lock's removed. Discussion with several people reveals that the
SMP locking primitives used in the VM system aren't likely the mechanism
that we'll be adopting. Even if it were, the locking that was in the code
was very inadequate and would have to be mostly re-done anyway. The
locking in a uni-processor kernel was a no-op but went a long way toward
making the code difficult to read and debug.
5) Places that attempted to kludge-up the fact that we don't have kernel
thread support have been fixed to reflect the reality that we are really
dealing with processes, not threads. The VM system didn't have complete
thread support, so the comments and mis-named routines were just wrong.
We now use tsleep and wakeup directly in the lock routines, for instance.
6) Where appropriate, the pagers have been improved, especially in the
pager_alloc routines. Most of the pager_allocs have been rewritten and
are now faster and easier to maintain.
7) The pagedaemon pageout clustering algorithm has been rewritten and
now tries harder to output an even number of pages before and after
the requested page. This is sort of the reverse of the ideal pagein
algorithm and should provide better overall performance.
8) Unnecessary (incorrect) casts to caddr_t in calls to tsleep & wakeup
have been removed. Some other unnecessary casts have also been removed.
9) Some almost useless debugging code removed.
10) Terminology of shadow objects vs. backing objects straightened out.
The fact that the vm_object data structure escentially had this
backwards really confused things. The use of "shadow" and "backing
object" throughout the code is now internally consistent and correct
in the Mach terminology.
11) Several minor bug fixes, including one in the vm daemon that caused
0 RSS objects to not get purged as intended.
12) A "default pager" has now been created which cleans up the transition
of objects to the "swap" type. The previous checks throughout the code
for swp->pg_data != NULL were really ugly. This change also provides
the rudiments for future backing of "anonymous" memory by something
other than the swap pager (via the vnode pager, for example), and it
allows the decision about which of these pagers to use to be made
dynamically (although will need some additional decision code to do
this, of course).
13) (dyson) MAP_COPY has been deprecated and the corresponding "copy
object" code has been removed. MAP_COPY was undocumented and non-
standard. It was furthermore broken in several ways which caused its
behavior to degrade to MAP_PRIVATE. Binaries that use MAP_COPY will
continue to work correctly, but via the slightly different semantics
of MAP_PRIVATE.
14) (dyson) Sharing maps have been removed. It's marginal usefulness in a
threads design can be worked around in other ways. Both #12 and #13
were done to simplify the code and improve readability and maintain-
ability. (As were most all of these changes)
TODO:
1) Rewrite most of the vnode pager to use VOP_GETPAGES/PUTPAGES. Doing
this will reduce the vnode pager to a mere fraction of its current size.
2) Rewrite vm_fault and the swap/vnode pagers to use the clustering
information provided by the new haspage pager interface. This will
substantially reduce the overhead by eliminating a large number of
VOP_BMAP() calls. The VOP_BMAP() filesystem interface should be
improved to provide both a "behind" and "ahead" indication of
contiguousness.
3) Implement the extended features of pager_haspage in swap_pager_haspage().
It currently just says 0 pages ahead/behind.
4) Re-implement the swap device (swstrategy) in a more elegant way, perhaps
via a much more general mechanism that could also be used for disk
striping of regular filesystems.
5) Do something to improve the architecture of vm_object_collapse(). The
fact that it makes calls into the swap pager and knows too much about
how the swap pager operates really bothers me. It also doesn't allow
for collapsing of non-swap pager objects ("unnamed" objects backed by
other pagers).
that call vnode_pager_alloc() so that a failure return can be dealt with.
This fixes a panic seen on NFS clients when a file being opened is deleted
on the server before the open completes.
2) Removed unnecessary vm_object_lookup()/pager_cache(object, TRUE) pairs
after vnode_pager_alloc() calls - the object is already guaranteed to be
persistent.
3) Removed some gratuitous casts.
The version 2 support has been tested (client+server) against FreeBSD-2.0,
IRIX 5.3 and FreeBSD-current (using a loopback mount). The version 2 support
is stable AFAIK.
The version 3 support has been tested with a loopback mount and minimally
against an IRIX 5.3 server. It needs more testing and may have problems.
I have patched amd to support the new variable length filehandles although
it will still only use version 2 of the protocol.
Before booting a kernel with these changes, nfs clients will need to at least
build and install /usr/sbin/mount_nfs. Servers will need to build and
install /usr/sbin/mountd.
NFS diskless support is untested.
Obtained from: Rick Macklem <rick@snowhite.cis.uoguelph.ca>
is an ambiguity in the NFS version 2 protocol.
VREG should be taken literally as a regular file. If a
server intents to return some type information differently
in the upper bits of the mode field (e.g. for sockets, or
FIFOs), NFSv2 mandates fa_type to be VNON. Anyway, we
leave the examination of the mode bits even in the VREG
case to avoid breakage for bogus servers, but we make sure
that there are actually type bits set in the upper part of
fa_mode (and failing that, trust the va_type field).
NFSv3 cleared the issue, and requires fa_mode to not
contain any type information (while also introduing sockets
and FIFOs for fa_type).
The fix has been tested against a variety of NFS servers.
It fixes problems with the ``Tropic'' NFS server for Windows,
while apparently not breaking anything.
Pointed-out by: scott@zorch.sf-bay.org (Scott Hazen Mueller)
handled correctly. This would manifest itself as "object deallocated too
many times" panics and perhaps other strange inconsistencies on NFS servers.
Reviewed by: me, of course
Submitted by: John Dyson
1) Files weren't properly synced on filesystems other than UFS. In some
cases, this lead to lost data. Most likely would be noticed on NFS.
The fix is to make the VM page sync/object_clean general rather than
in each filesystem.
2) Mixing regular and mmaped file I/O on NFS was very broken. It caused
chunks of files to end up as zeroes rather than the intended contents.
The fix was to fix several race conditions and to kludge up the
"b_dirtyoff" and "b_dirtyend" that NFS relies upon - paying attention
to page modifications that occurred via the mmapping.
Reviewed by: David Greenman
Submitted by: John Dyson
require specific partitions be mentioned in the kernel config
file ("swap on foo" is now obsolete).
From Poul-Henning:
The visible effect is this:
As default, unless
options "NSWAPDEV=23"
is in your config, you will have four swap-devices.
You can swapon(2) any block device you feel like, it doesn't have
to be in the kernel config.
There is a performance/resource win available by getting the NSWAPDEV right
(but only if you have just one swap-device ??), but using that as default
would be too restrictive.
The invisible effect is that:
Swap-handling disappears from the $arch part of the kernel.
It gets a lot simpler (-145 lines) and cleaner.
Reviewed by: John Dyson, David Greenman
Submitted by: Poul-Henning Kamp, with minor changes by me.
1) Rewrote screwy code that uses an incore buffer without making it busy.
2) Use B_CACHE instead of B_DONE in cases where it is appropriate.
3) Minor code optimization.
This *might* fix kern/345 submitted by Heikki Suonsivu.
happen normally when there is heavy write activity to a file since the
vnode isn't locked (NFS plays fast and loose with vnode locks). This change
"fixes" PR#267.
