CVS log for src/sys/fs/efs/efs_vnops.c

Up to [NetBSD] / src / sys / fs / efs

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

Sync with HEAD.

sync with head.

Fix utterly botched previous commit.  efs compiles now and apprears
to work, but the person doing the original change should verify
that it actually works like before the change.

(hi, elad!)

Where possible, extract the file-system's access() routine to two internal
functions: the first checking if the operation is possible (regardless of
permissions), the second checking file-system permissions, ACLs, etc.

Mailing list reference:

	http://mail-index.netbsd.org/tech-kern/2009/06/21/msg005311.html

Move the implementation of vaccess() to genfs_can_access(), in line with
the other routines of the same spirit.

Adjust file-system code to use it.

Keep vaccess() for KPI compatibility and to keep element of least
surprise. A "diagnostic" message warning that vaccess() is deprecated will
be printed when it's used (obviously, only in DIAGNOSTIC kernels).

No objections on tech-kern@:

	http://mail-index.netbsd.org/tech-kern/2009/06/21/msg005310.html

sync with head.

Sync with HEAD.

Sync with HEAD.

Apply patch (requested by rumble in ticket #216):
Adjust for the netbsd-5 branch by defining UBC_UNMAP_FLAG.

Pull up following revision(s) (requested by rumble in ticket #216):
	sys/fs/efs/efs_vfsops.c: revision 1.17
	sys/fs/efs/efs_vnops.c: revision 1.17
Add specfs/fifofs support.  Not really fully tested, but at least
vfs routines don't crash now.
(didn't have an image with matching device node numbers)
fixes PR kern/40055 by myself

Pull up following revision(s) (requested by rumble in ticket #216):
	sys/fs/efs/efs_vnops.c: revision 1.16
Rototill all remaining file systems to use ubc_uiomove() instead
of the ubc_alloc() - uiomove() - ubc_release() dance.

Pull up following revision(s) (requested by rumble in ticket #216):
	sys/fs/efs/efs_vnops.c: revision 1.15
more <sys/buf.h> police

Update haad-dm branch to haad-dm-base2.

Add specfs/fifofs support.  Not really fully tested, but at least
vfs routines don't crash now.
(didn't have an image with matching device node numbers)

fixes PR kern/40055 by myself

Rototill all remaining file systems to use ubc_uiomove() instead
of the ubc_alloc() - uiomove() - ubc_release() dance.

more <sys/buf.h> police

sync with HEAD

Sync with HEAD.

sync with head.

Remove VOP_LEASE. Discussed on tech-kern.

sync with head

sync with HEAD

Sync with HEAD

Merge vmlocking2 to head.

Sync with HEAD.

sync with head

Pull the vmlocking changes into a new branch.

Sync with HEAD. amd64 Xen support needs testing.

Remove the "struct lwp *" argument from all VFS and VOP interfaces.
The general trend is to remove it from all kernel interfaces and
this is a start.  In case the calling lwp is desired, curlwp should
be used.

quick consensus on tech-kern

sync with HEAD

sync with head.

Sync with HEAD.

Follow the merge of pmap.c on i386 and amd64 and move
pmap_init_tmp_pgtbl into arch/x86/x86/pmap.c. Modify the ACPI wakeup
code to restore CR4 before jumping back into kernel space as the large
page option might cover that.

sync with head.

Merge from vmlocking:

- Split vnode::v_flag into three fields, depending on field locking.
- simple_lock -> kmutex in a few places.
- Fix some simple locking problems.

Sync with head.

Sync with head.

Merge brelse() changes from the vmlocking branch.

sync with head.

Sync with HEAD.

Avoid stack allocation of large dirent structures in foo_readdir().

Checkpoint work in progress on the vnode lifecycle and reference counting
stuff. This makes it work properly without kernel_lock and fixes a few
quite old bugs. See vfs_subr.c 1.283.2.17 for details.

Sync with HEAD.

In efs_readdir don't foolishly allocate and free dirents for each entry
before copying them out, rather just use a single one. Further, follow
the example of tmpfs and others by simply allocating on the stack.

This should have the side-effect of silencing false Coverity reports like
CID 4559 and 4554.

CID 4553:

In efs_lookup, when checking whether a CREATE or RENAME operation would
succeed, be sure to pass the appropriate vnode pointer to VOP_ACCESS.

sync with head.

Sync with HEAD.

- Back out the biodone() changes.
- Eliminate B_ERROR (from HEAD).

Sync with HEAD.

file efs_vnops.c was added on branch yamt-lazymbuf on 2007-09-03 14:40:16 +0000

Only set *ap->a_cookies when we're returning success. Also, do not hard-
code 16, but rather use _DIRENT_MINSIZE as the divisor when determining
the number of cookies to allocate.

Noticed by pooka@.

Support NFS cookies and the eofflag in efs_readdir. Also, be sure to
update uio->uio_offset as not doing so would cause large directory reads
to misbehave.

EFS is now properly NFS-exportable.

It's not a good idea for device drivers to modify b_flags, as they don't
need to understand the locking around that field. Instead of setting
B_ERROR, set b_error instead. b_error is 'owned' by whoever completes
the I/O request.

Sync with head.

file efs_vnops.c was added on branch vmlocking on 2007-07-15 16:15:30 +0000

Additional fixes/enhancements:
    1) Comply with the way buffercache(9) is intended to be used. Now we
       read in single blocks of EFS_BB_SIZE, never taking in variable
       length extents with a single bread() call.

    2) Handle symlinks with more than one extent. There's no reason for
       this to ever happen, but it's handled now.

    3) Finally, add a hint to our iteration initialiser so we can start
       from the desired offset, rather than naively looping through from
       the beginning each time. Since we can binary search the correct
       location quickly, this improves large sequential reads by about
       40% with 128MB files. Improvement should increase with file size.

Fix a significant performance bug in efs_read:

When reading a file, we would erroneously iterate to the next extent
before having filled the entire uio request. This lead to unnecessary
extent iteration and excessive calls to efs_read.

Sequential read performance has doubled in the uncached case and
quadrupled when data is buffered.

Add read-only support for SGI's Extent File System.

Reviewed by pooka@.