add a flag that specify if the file can be truncated safely or not
to nfsm_loadattr and friends. when it isn't safe, just mark the nfsnode
as "should be truncated later".
ok'ed by Frank van der Linden and Chuck Silvers.
close kern/18036.
and confusion about the actual filesize. From Matt Dillon's
similar change in FreeBSD.
XXX n_size is really redundant in -current and must die. This commit
XXX is more of a placeholder for a pullup into the 1.5 branch.
uint32_t namei_hash(const char *p, const char **ep)
which determines the equivalent MI hash32_str() hash for p.
If *ep != NULL, calculate the hash to the character before ep.
If *ep == NULL, calculate the has to the first / or NUL found, and
point *ep to that location.
- Use namei_hash() to calculate cn_hash in lookup() and relookup().
Hash distribution goes from 35-40% to 55-70%, with similar profiled
time spent in cache_lookup() and cache_enter() on my P3-600.
- Use namei_hash() to calculate cn_hash in nfs_readdirplusrpc(),
insetad of homegrown code (that differed from that in lookup() !)
namei_hash() has better spread and is faster than previous code
(which used a non-constant multiplication).
- remove special treatment of pager_map mappings in pmaps. this is
required now, since I've removed the globals that expose the address range.
pager_map now uses pmap_kenter_pa() instead of pmap_enter(), so there's
no longer any need to special-case it.
- eliminate struct uvm_vnode by moving its fields into struct vnode.
- rewrite the pageout path. the pager is now responsible for handling the
high-level requests instead of only getting control after a bunch of work
has already been done on its behalf. this will allow us to UBCify LFS,
which needs tighter control over its pages than other filesystems do.
writing a page to disk no longer requires making it read-only, which
allows us to write wired pages without causing all kinds of havoc.
- use a new PG_PAGEOUT flag to indicate that a page should be freed
on behalf of the pagedaemon when it's unlocked. this flag is very similar
to PG_RELEASED, but unlike PG_RELEASED, PG_PAGEOUT can be cleared if the
pageout fails due to eg. an indirect-block buffer being locked.
this allows us to remove the "version" field from struct vm_page,
and together with shrinking "loan_count" from 32 bits to 16,
struct vm_page is now 4 bytes smaller.
- no longer use PG_RELEASED for swap-backed pages. if the page is busy
because it's being paged out, we can't release the swap slot to be
reallocated until that write is complete, but unlike with vnodes we
don't keep a count of in-progress writes so there's no good way to
know when the write is done. instead, when we need to free a busy
swap-backed page, just sleep until we can get it busy ourselves.
- implement a fast-path for extending writes which allows us to avoid
zeroing new pages. this substantially reduces cpu usage.
- encapsulate the data used by the genfs code in a struct genfs_node,
which must be the first element of the filesystem-specific vnode data
for filesystems which use genfs_{get,put}pages().
- eliminate many of the UVM pagerops, since they aren't needed anymore
now that the pager "put" operation is a higher-level operation.
- enhance the genfs code to allow NFS to use the genfs_{get,put}pages
instead of a modified copy.
- clean up struct vnode by removing all the fields that used to be used by
the vfs_cluster.c code (which we don't use anymore with UBC).
- remove kmem_object and mb_object since they were useless.
instead of allocating pages to these objects, we now just allocate
pages with no object. such pages are mapped in the kernel until they
are freed, so we can use the mapping to find the page to free it.
this allows us to remove splvm() protection in several places.
The sum of all these changes improves write throughput on my
decstation 5000/200 to within 1% of the rate of NetBSD 1.5
and reduces the elapsed time for "make release" of a NetBSD 1.5
source tree on my 128MB pc to 10% less than a 1.5 kernel took.
if we do this for VBLK vnodes which are in use by softdep mounts,
brelse() will mark the buffer B_INVAL as well, which makes the
softdep code very unhappy.
that required to support NFSv2 mounts. Not finished yet, but already
provides some 44k of saving in code size on arm26. More savings, and some
documentation, are still to come.
int lf_advlock __P((struct lockf **,
off_t, caddr_t, int, struct flock *, int));
to
int lf_advlock __P((struct vop_advlock_args *, struct lockf **, off_t));
This matches common usage and is also compatible with similar change
in FreeBSD (though they use u_quad_t as last arg).
case that write verf is changed. Suggested by mycroft@netbsd.org.
- Reset wcred to NULL (i.e., write credential isn't decieded) everytime
before gathering buffer for new commit, so that there is a chance to
the commit request is merged.
default, as the copyright on the main file (ffs_softdep.c) is such
that is has been put into gnusrc. options SOFTDEP will pull this
in. This code also contains the trickle syncer.
Bump version number to 1.4O
call with F_FSCTL set and F_SETFL calls generate calls to a new
fileop fo_fcntl. Add genfs_fcntl() and soo_fcntl() which return 0
for F_SETFL and EOPNOTSUPP otherwise. Have all leaf filesystems
use genfs_fcntl().
Reviewed by: thorpej
Tested by: wrstuden
UVM was written by chuck cranor <chuck@maria.wustl.edu>, with some
minor portions derived from the old Mach code. i provided some help
getting swap and paging working, and other bug fixes/ideas. chuck
silvers <chuq@chuq.com> also provided some other fixes.
this is the rest of the MI portion changes.
this will be KNF'd shortly. :-)
the directory cache as translation table. See nfs_subs.c for comments.
Makes the code a bit more complex to look at than I would have liked,
but doesn't affect the speed of the default behavior.
* Optimize caching behavior a bit when buffers are invalidated.
* Save some RPCs in readdir operations by not bothering if there is
a small amount left to do to fill the buffer. It'll be done in the
next RPC with a larger chunk anyway. Wastes a bit of buffer space
but is faster.
* Make n_vattr an allocated vattr struct. This avoids nfsnode bloat,
and is friendlier to the malloc routines.
directory cookie that may be thrown back at us from userspace, up
to a size limit. Fixes double entry problem.
* Split flags for internal and external use in the NFS mount structure.
* Fix some buffer structure fields that weren're being used correctly.
* Fix missing directory cache inval call in nfs_open.
* Limit on NFS_DIRBLKSIZ no longer needed, bumped to the more reasonable
value of 8k.
* Various other things that I forget, all related to the dir caching
somehow, though.
From Olaf Seibert <rhialto@polder.ubc.kun.nl> (PR 3687)
* Make an attempt to check the maximum filesize before attempting
a write to the server, as write RPCs will typically happen
asynchronously, and the process will not see the error.
Fixes problems with unexpectly truncated files at 4G
* Pass up errors in nfs_writerpc correctly
In readdirplus, don't keep such pointers but store the file attributes
in a variable instead until they are needed. Change nfsm_loadattr*
a bit so it can accept a direct pointer to an nfs_fattr structure.
architectures), truncate them intelligently instead.
The truncation is done centralized in vnode_pager.c.
This prevents from wrap-over effects when parts of large (>2^32 byte) files
are mmapped.
Don't allow to mmap above the numerical range of vm_offset_t.
This is considered a temporary solution until the vm system handles the
object sizes/offsets more cleanly.
date: 1996/07/23 17:14:46; author: donn; state: Exp; lines: +6 -4
Be sure to push out the last page of the file before truncating it.
date: 1996/10/14 22:41:20; author: donn; state: Exp; lines: +2 -2
From Chris: Nfs_link() called vput() on the wrong vnode when aborting
from a cross-device link, which could (and did) lead to crashes.
date: 1996/10/24 16:43:43; author: pjd; state: Exp; lines: +6 -2
Return EOPNOTSUPP when trying to do a setattr with flags.
===
Also (from BSDI too, but the RCS message did not quite describe the change
to this particular file well): move the EROFS a bit further down to
let VOP_ACCESS do it's work and return an 'expected' error value to
a possible layered filesystem.
Improve the queuing algorithms used by NFS' asynchronous i/o. The
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.
Reviewed/integrated/approved by Frank van der Linden <fvdl@netbsd.org>
struct member cn_nameptr 'const', since they should never be used to
modify the path name. (Only the pathname buffer, cn_pnbuf, should be
modified.) Propagate the const poisoning to code that uses the namei
and componentname structs.
* Make 2nd and 3rd args timespecs, not timevals.
* Consistently pass a Boolean as the 4th arg (except in LFS).
Also, fix ffs_update() and lfs_update() to actually change the nsec fields.
* Change the argument names to vop_link so they actually make sense.
* Implement vop_link and vop_symlink for all file systems, so they do proper
cleanup.
* Require the file system to decide whether or not linking and unlinking of
directories is allowed, and disable it for all current file systems.
each entry, and read them out in nfs_readdir().
Caveat: our current caching method for directory blocks uses the
server offset of the first directory entry as an identifier, so a
Linux emulation getdirentries() will wind up retrieving one block from
the NFS server for each directory entry, unnecessarily thrashing the
cache. The situation isn't as bad for other emulations.
Instead of getblk(), we need to write a routine to scan each cache
block associated with vp to find a cookie that matches at some
directory entry. Some later time.
(1) it's unnecessary
(2) it causes machines to hang (yup!)
(3) it'd be gone in a few days anyway (it'd been yanked out
of 4.4-Lite by macklem long ago)
It was only there because macklem couldn't originally decide if things
should be locked, or not...
Remove comment talking about nfsiomaps that we don't have.
Always use credentials that are in the buffer header, in stead of trying
to get them from pageproc, which may once have been necessary to push pages
to swap (cannot imaging anyone having exercised this over NFS though).
John Woods, jfwfrom: @ksr.com. also, fixes the following problems:
the va_gen field is in a similar position
(Suns are going to be reporting the change-date microseconds as their
"generation"), I've supplied my own set of diffs below for your inspection.
Note these aren't even compiled, but they're pretty similar to what I had
to do to our older version of OSF/1 here. (There's also an unrelated change
supplied for xdr_subs.h; the pointer types supplied to the fxdr_time() and
txdr_time() macros are not, in fact, both struct timevals. That turns out
to be one of many tips-of-the-iceberg facing those porting the (old) Berkeley
NFS code to 64-bit machines...)
