Release notes for v2.5.4
Summary of changes from v2.5.4-pre6 to v2.5.4
============================================
(02/02/09 1.248.1.10)
Sparc64 thread_info implementation.
(02/02/09 1.248.1.11)
Build fixes for sparc64 outside of sparc64 specific files.
(02/02/10 1.248.4.1)
Update alpha defconfig to the options available in 2.5.4-pre5.
(02/02/10 1.248.1.12)
Sparc64 preemption support.
(02/02/10 1.248.1.13)
Add CONFIG_PREEMPTION.
(02/02/10 1.248.1.14)
Clean up sparc64 build
(02/02/10 1.257.1.1)
[PATCH] preempt-kernel on UP compile fix
A recent scheduler change looks to of affected this, and I missed it
since I did my last testing on SMP only. Please, apply.
Robert Love
(02/02/10 1.259)
[PATCH] check copy_from_user return codes in serial drivers.
forward-ported from 2.4
(02/02/10 1.262)
update version
Summary of changes from v2.5.4-pre5 to v2.5.4-pre6
============================================
(02/02/09 1.249)
Update eepro100 net driver pci id list, at the prompting of
Andrew Morton and Hanno Boeck.
Three constants are substituted with their numeric equivalents,
a reverse of the norm, to make the linear progression of PCI ids
more clear, and easier to validate at a glance.
(02/02/09 1.250)
Add pci ids found in 2.4.18-pre9's version of eepro100 net driver
to the current driver.
(02/02/09 1.248.1.1)
[PATCH] Re: [PATCH] Preemptible Kernel for 2.5
On Sat, 2002-02-09 at 01:43, Linus Torvalds wrote:
> That will clean up all your issues with header file ordering.
You are right, it did. I removed all the sched.h dependencies and this
reduced the size of the patch greatly. I now use current_thread_info()
and none of the header or include hackery from before. I've tested this
with and without preemption enabled with success.
I appreciate your help with this.
Again, this is a minimal i386-only patch. I have other arches,
documentation, etc. Patch against 2.5.4-pre5. Enjoy,
Robert Love
(02/02/09 1.248.2.1)
[PATCH] (1/5) ext2_free_blocks() cleanup
new helper function - release_blocks(sb, n); in ext2_free_blocks() we
count the blocks we'd freed and update the fs-wide count of free blocks
in the end - just before unlocking superblock. instead of doing that
step at a time.
reviewed by ext2 folks
(02/02/09 1.248.2.2)
[PATCH] (2/5) ext2_free_blocks() cleanup
we postpone releasing the quota until the end of ext2_free_blocks() and
do that at once.
(02/02/09 1.248.2.3)
[PATCH] (3/5) ext2_free_blocks() cleanup
new helper - group_release_blocks(desc, bh, count); updates free blocks
count for given group ext2_free_blocks now counts the blocks freed in
group and updates the per-group free blocks count at once.
(02/02/09 1.248.2.4)
[PATCH] (4/5) ext2_free_blocks() cleanup
in ext2_free_blocks() inode->i_sb can't be NULL; dead code removed.
(02/02/09 1.248.2.5)
[PATCH] (5/5) ext2_free_blocks() cleanup
minor cleanup: update 'freed' only after we finish with group
(02/02/09 1.248.2.6)
[PATCH] ext2_new_block() cleanup
in ext2_new_block() inode->i_sb can't be NULL. Dead code removed.
merged returns in ext2_new_block()
in ext2_new_block() added a local variable (sbi) for sb->u.ext2_sb;
users converted.
calculation of prealloc_goal moved into beginning of function.
new local variable (block). Initialized to 0, set to the number of
allocated block only when we are known to be successful.
ext2_new_block() returns the value of 'block' now. As the result, out:
is gone - it's identical to out_unlock now.
check for block being chosen past the end of disk had been moved up No
sense to check that after we had done the preallocations...
quota allocation taken into the beginning of ext2_new_block(). we put
the number of blocks we'd reserved into dq_alloc and then decrement it
each time old code would do DQUOT_ALLOC/increment when it would do
DQUOT_FREE. On the exit we release dq_alloc blocks I.e. all
allocations are pushed into the beginning and all freeing - to the end.
now that we had moved the last possible point of failure up, we can do
the same with assignment of final result to block.
code that does allocation within a group had been taken out into a
helper - grab_block(map, size, goal).
decrementing fs-wide free blocks count moved into the beginning of
function. We try to reserve up to dq_alloc blocks - amount we _can_
grab is limited by the actual free space and root-reserved percentage.
That logics had been taken into a helper - reserve_blocks(sb, amount).
Number of blocks we had managed to reserve is put into es_alloc. When
we actually grab a block, es_alloc is decremented. Upon the exit we
release the leftovers. Old code didn't honor 'reserved for root' limit
when doing prealloc - now that had been fixed.
EXT2_BLOCKS_PER_GROUP(sb) put into a local variable (group_size)
similar "reserve and release leftovers in the end" scheme for group
(i.e. we try to reserve as much as we want in the group, then count how
much had been used up and release the leftovers in the end).
cleaned up the preallocation loop.
(02/02/09 1.248.2.7)
[PATCH] fs/ext2/balloc.c cleanup
Mostly cosmetical cleanup - sane variable names, etc.
That's pretty much it for balloc.c cleanups - stuff after that
will be about decent threading.
(02/02/09 1.248.2.8)
[PATCH] driverfs - preparation to fixes
Make dentry and ->i_sem manipulations in driverfs explicit. Something
is seriously rotten there and real fixes will take non-trivial work
(e.g, I suspect that we actually ought to kill driverfs_rename() - at
least cross-directory one for regular files, maybe completely). This
patch simply cleans the ground.
(02/02/09 1.248.3.1)
Make AGP on x86 work temporarily while waiting for the proper fix.
(02/02/09 1.248.3.2)
[PATCH] 2.5.4pre5 scsi/aha1542.c & DMA changes
There are a couple of errors in the DMA changes to the
aha1542.c driver that appeared in 2.5.4pre3:
- a typo (BUF vs BUS)
- a macro handed a page instead of a scatterlist
- printk format length mismatch
Paul.
(02/02/09 1.248.2.10)
update version
(02/02/09 1.248.2.11)
[PATCH] includes cleanup, 2nd try.
Big bits first, I'll redo the smaller bits tomorrow after some sleep.
Same as last time, rediffed against pre5
(02/02/09 1.248.1.3)
defconfig:
update
(02/02/09 1.223.4.1)
Various bugfixes to sync with 2.4 version:
- Fix bug in scripts that was causing CDRW hang
- add sense buffer length initialisation
- update lasi700 to new driver model and fix rmmod hang
- add correct MCA reporting to D700 driver
(02/02/09 1.248.1.4)
[PATCH] Make cardbus device_register work
It compiles in -pre5 but doesn't work unless you also apply the patch
below. Without this patch, bus_id will be empty which makes
device_register fail.
(02/02/09 1.248.1.5)
[PATCH] fix for typo in ext2_new_inode()
Please, apply. Otherwise we eat quota like there's no tomorrow ;-/
(02/02/09 1.248.1.6)
[PATCH] /proc/slabinfo cleanup (seq_...)
Straightforward rewrite to seq_file.
(02/02/09 1.248.1.7)
[PATCH] /proc/modules cleanup (seq_file, again)
/proc/modules switched to use of seq_file, cleaned up.
(02/02/09 1.248.1.8)
[PATCH] affs cleanup
affs digs in blk_size[] to find the size of device (it
has its analog of superblock living in the middle of disk). That
information is available in ->s_bdev->bd_inode->i_size - no need
to mess with major/minor/arrays to get it.
(02/02/09 1.248.1.9)
[PATCH] ->s_id cleanups
Patch converts a bunch of places that hadn't been caught when
->s_id was introduced. Please, apply.
(02/02/10 1.248.5.1)
remove CONFIG_ISDN entries from arch/Config.help - The real entry is
in drivers/isdn/Config.help
(02/02/10 1.248.5.2)
add missing "n" in printk statements.
(02/02/10 1.248.6.1)
rename built-in ISDN object to vmlinux-obj.o
(was isdn.a before, but there's no need to use an archive here)
(02/02/10 1.248.6.2)
Link ISDN after USB. Otherwise, when builtin, the ST5481 USB ISDN
adapter driver initializes before the USB layer -> Oops
(02/02/10 1.256)
[PATCH] yet another lseek fix
Another lseek fix, spotted by Dave Hansen @ IBM. Missing a
corresponding lock_kernel for an unlock_kernel. Patch against
2.5.4-pre5, please apply. Thanks,
Robert Love
(02/02/10 1.257)
[PATCH] preemptible kernel documentation, etc
Linus,
The attached patch adds a Documentation/preempt-locking.txt file which
describes the new locking rules wrt preemptive kernels (ie, watch
per-CPU data, etc). It also updates a CREDITS entry and adds some
comments.
Patch is against 2.5.4-pre5, please apply.
Robert Love
Summary of changes from v2.5.4-pre4 to v2.5.4-pre5
============================================
(02/02/08 1.229.4.1)
typo fix for the name of the pci hotplug filesystem caused by the superblock changes.
(02/02/08 1.234)
[PATCH] handle out of spec SMP athlons.
Newer Athlons have means of checking if they are SMP capable or not.
This code adds checks that printk a warning on systems not intended
for SMP, and set the taint flag that modutils is already aware of.
The taint code is also improved to use defines instead of magic numbers.
(02/02/08 1.235)
[PATCH] text.lock -> subsection changes.
Make spinlocks etc use subsections of their parent sections instead of
an ELF section of their own - needed for newer binutils when the parent
sector is removed.
(02/02/08 1.236)
[PATCH] AGPGart update
Mostly from 2.4.17 / 18pre.
- Code cleanup
- Fix suspend/resume on several chipsets
- Add support for i820.
(02/02/08 1.237)
[PATCH] fix serial close hang
2.4/2.5 kernels suffer from an infinitely long hang when a serial tty device
is closed, and there are characters waiting to be sent. The hang occurs in
tty_wait_until_sent.
There is a timeout 'closing_wait' which defines how long to wait for the TX
buffers to empty; the problem is that the serial layer totally ignores it.
It is stored in two structures, 'info' and 'state'. It is initialised in
the 'state' structure, but used from the 'info' structure.
It turns out that 'hub6' was also missing.
I'm not currently clear what the expected behaviour should be when the
timeout is changed via setserial, and others have the port open - I've
opted to preserve the timeout until all users close the port. It's
trivial to change this behaviour though.
(02/02/08 1.238)
[PATCH] Power management support for opl3sa2 driver
(02/02/08 1.239)
[PATCH] Config.in updates
- Reorder mainmenu items so CONFIG_EXPERIMENTAL is at the top again.
- Various updates from 2.4
(02/02/08 1.240)
[PATCH] remove ancient workaround.
unneeded these days..
(02/02/08 1.241)
[PATCH] Extra config.help for x86
Additional descriptions that got lost along the way somewhere.
Also updates some extra IDE options that shared the same fate.
(02/02/08 1.242)
[PATCH] remove duplicate zconf.h
Change to use the new include/linux/zconf.h
(02/02/08 1.243)
[PATCH] knfsd cleanups
exports hash switched to use of struct list_head, code cleaned up.
(02/02/08 1.244)
[PATCH] knfsd cleanups
all exports for given client are put on the cyclic list
(->cl_list/->ex_list). Code cleaned up.
(02/02/08 1.245)
[PATCH] knfsd cleanups
exp_procfs_exports() switched to seq_..., missing lock added.
(02/02/08 1.229.3.2)
Doh!
struct device has no ->sysdata
and ->device should be ->dev
(02/02/08 1.247)
update version
(02/02/08 1.248)
[PATCH] fix for drivers/input/input.c
Fix lost argument to "connect()"
Summary of changes from v2.5.4-pre3 to v2.5.4-pre4
============================================
(02/02/07 1.224)
scatterlist address breakage in task file ioctl building
(02/02/07 1.225)
scatterlist address fixup
(02/02/07 1.220.1.66)
eepro100 net driver fix:
Second half of fix (first half of which added an embarrassingly
long time ago) which works around hardware bug at 10 mbit.
Contributors: Steve Parker, Arjan van de Ven
(02/02/07 1.220.1.67)
Add PCI id constants for afavlab serial cards.
Contributor: Russell King
(02/02/07 1.220.1.68)
New tulip net driver PCI id, for a new "comet" board
Contributor: Uwe Bonnes
(02/02/07 1.220.1.69)
eepro100 net driver link beat detection improvement.
Control interface running flag correctly, via netif_carrier_{on,off},
rather than manipulating the IFF_RUNNING bit directly from within
the driver (ug).
Contributors: Joerg Mayer, Stefans Rumpf
(02/02/07 1.223.1.15)
[PATCH] the rest of ->get_sb() switchover
Switch over ext3, ufs, reiserfs, msdos, vfat, isofs, qnx4, hpfs, efs,
udf, romfs, bfs, affs, hfs, adfs, cramfs, ntfs, jffs, jffs2 and freevxfs
to new ->get_sb().
Kill ->read_super() that is thus no longer used.
(02/02/07 1.223.1.16)
[PATCH] another llseek change fix
Fixe another typo in the recent llseek cleanups. Typo was spotted by
Miloslav Trmac ... hopefully the last.
(02/02/08 1.227)
[PATCH] thread information block
The first fix changes get_wchan() and the second one fixed what DaveM noticed.
(02/02/08 1.228)
[PATCH] Fix the replacement mistake of xxx_thread_flag
- fix the replacement mistake of xxx_thread_flag
- fix the miss test of PT_PTRACED flag
OGAWA Hirofumi
(02/02/08 1.229)
update Makefile and x86 defconfig
(02/02/08 1.230)
After Al Viro's recent swapfile cleanup, the swap_device member of
swap_info_struct became pretty much superfluous. As we are minimizing
kdev_t usage anyway, I took the opportunity to remove swap_device
member, and replace the remaining usages with SWP_BLOCKDEV bit flag.
Adding SWP_BLOCKDEV in turn motivated a small cleanup of the
SWP_xxx bit flags and their usage.
Patch has been in light testing for a couple weeks, and
has been glanced at by Al. "looks sane"
(02/02/08 1.229.1.1)
Fix typo in the winbond-840 net driver which doubled
the size of the Tx data buffer list without cause.
Spotted by Dave Jones.
(02/02/08 1.229.1.2)
Add config option to enable natsemi net driver hardware bug workaround.
"some" systems with "some" cables see a large amount of errors,
due to a hardware bug. This bug is (apparently) not probe-able;
however it only appears on rare reference boards and the like,
so we simply add a config option and default the option to OFF.
Further detail:
When CONFIG_NATSEMI_CABLE_MAGIC option is enabled, PMDCSR_VAL
register value becomes 0x1898, a value provided by a NatSemi
app note. This enables a workaround for a hardware bug
which is (apparently) not probe-able. Luckily the hardware bug
is (apparently) not common either, so we default to disabling
this workaround.
Contributor: Tim Hockin
(02/02/08 1.229.1.3)
Fix naming conflict with pcnet32 net driver and ethtool,
by cleaning up the pcnet32 namespace a bit.
s/PORT_/PCNET32_PORT_/ for local constants, to avoid conflicting
with linux/ethtool.h.
Contributor: William Lee Irwin III
(02/02/08 1.229.1.4)
tulip net driver updates:
* Add support for Conexant tulip clones.
* Do not store eeprom data on stack (128 or 512 bytes), it's a
large object, and also, we already have a copy in kmalloc'd RAM.
Contributor: Pavel Roskin
(02/02/08 1.229.1.5)
Update tulip net driver to indicate link beat to system,
via netif_carrier_{on,off}. Some placeholders to do this were
already in the code, making this an even easier and more obvious patch.
Also, decrease time until next link beat check, if link beat
is not present. (previously the code would wait 60 seconds until
next check, regardless of current link state)
Contributor: Stefan Rompf, with changes from me
(02/02/08 1.229.1.6)
Fix natsemi net driver rx-related hang, by polling for RX events
on all RX interrupts. Prior to this fix, RX FIFO overrun and RX
buffer overrun interrupts did not trigger an RX poll; now they do.
Contributor: Manfred Spraul
(02/02/08 1.229.1.7)
Update eepro100 net driver link state tracking:
* Initialize interface carrier state in speedo_open.
* Update previous netif_carrier_{on,off} change to use
linux/mii.h constants.
Contributor: Andrew Morton, with modifications from me
(02/02/08 1.229.2.1)
[PATCH] knfsd cleanups
Get rid of NFSCTL_GETFH.
(02/02/08 1.229.2.2)
[PATCH] knfsd cleanups
unused arguments (and dead code) removed from exp_rootfh().
(02/02/08 1.229.2.3)
[PATCH] knfsd cleanups
exp_find() is gone (never used).
(02/02/08 1.229.2.4)
[PATCH] knfsd cleanups
exp_child() and exp_parent() take super_block instead of kdev_t.
(02/02/08 1.229.2.5)
[PATCH] knfsd cleanups
logics in exp_get() cleaned up (it's kernel, not IOCCC, damnit...)
(02/02/08 1.229.2.6)
[PATCH] knfsd cleanups
new helper - exp_get_by_name(client, vfsmount, dentry). Used
in places where we travers mountpoint.
(02/02/08 1.229.2.7)
[PATCH] knfsd cleanups
logics in nfsd_lookup() cleaned up (see above, only more so ;-/)
(02/02/08 1.229.2.8)
[PATCH] de_put: entry meminfo already free!
initialize proc_inode properly for /proc/self/
(02/02/08 1.229.2.9)
[PATCH] knfsd cleanups
nfsd_read() slightly cleaned up.
(02/02/08 1.229.3.1)
Small driver model/driverfs update.
Summary of changes from v2.5.4-pre2 to v2.5.4-pre3
============================================
(02/02/06 1.221)
fix multi write and pio issues
(02/02/06 1.222)
kill struct scatterlist address member
(02/02/07 1.220.6.1)
Add support for basic ethtool ioctls to 3com net drivers,
3c501, 3c503, 3c505, 3c507, 3c509, 3c515, 3c523, 3c527.
(02/02/06 1.220.1.45)
Fix up tree oddities that Al pointed out.
(02/02/07 1.220.1.47)
Set up 'bk ignore' for standard files created in an ia32 build.
(02/02/07 1.220.1.48)
Obvious updates to the alpha port, for recent 2.5.[123] API changes.
More changes are needed to the task struct/scheduler area before
the alpha port will boot again; this flushes the easy stuff.
(02/02/07 1.220.1.49)
de4x5 net driver endian-related fixes.
Caught by Ralf and the mips crowd.
(02/02/07 1.220.1.50)
Fix several bugs in DL2K Gigabit Ethernet driver, updating to
vendor revision 1.08.
Contributed by the vendor, D-Link.
(02/02/07 1.220.1.51)
epic100 net driver updates:
* Fix wake-up sequence for some clone boards.
* Support media-related ethtool ioctls, using MII ethtool helper lib.
(02/02/07 1.220.1.52)
epic100 now uses the MII ethtool helper lib, so link against it.
(02/02/07 1.220.1.53)
Jiffies wrap cleanup for the slip net driver.
Contributor: somebody from lkml, Tim ?.
(02/02/07 1.220.1.54)
Update pci-skeleton net driver:
Replace MII-related magic numbers with meaningful constants
listed in linux/mii.h.
(02/02/07 1.220.1.55)
Add basic ethtool ioctl support to pcmcia net drivers
3c589_cs, aironet4500_cs, and fmvj18x_cs.
(02/02/07 1.220.1.56)
sundance net driver updates:
* Properly identify 4-port server NIC.
* Add new NIC to PCI ids list.
* Fix multicast bug.
Contributed by the vendor, D-Link.
(02/02/07 1.220.1.57)
winbond-840 (tulip clone) net driver updates:
* Support MII ethtool helper interface, and related ioctls.
* Replace some MII-related magic numbers with constants.
(02/02/07 1.220.1.58)
via-rhine net driver updates:
* Support MII ethtool helper library, and associated ioctls.
* Replace some MII-related magic numbers with constants from linux/mii.h.
* Remove deprecated SIOCDEVPRIVATE ioctl usage, SIOCxMIIxxx is the
supported (and already present) replacement.
(02/02/07 1.220.1.59)
Additional media-related ethtool ioctls for net drivers
ns82830, starfire.
(02/02/07 1.220.1.60)
Fix kernel "taint" issue with mii.o helper module, by
adding MODULE_{AUTHOR,DESC,LICENSE} tags.
(02/02/07 1.220.1.61)
s/suser/capable/ cleanup for the LMC WAN driver.
Contributed by one of the kernel janitors.
(02/02/07 1.220.1.62)
8139cp net driver updates:
* Support MII ethtool helper interface, and associated ioctls.
* Remove unused CP_CHIP_VERSION constant.
* Update to-do list in driver header.
(02/02/07 1.220.1.63)
fealnx net driver updates:
* Support MII ethtool helper interface, and associated ioctls.
* Replace some MII-related magic numbers with constants from linux/mii.h.
* Remove deprecated SIOCDEVPRIVATE ioctls, SIOCxMIIxxx ioctl replacements
have been in place since 2.4.x.
(02/02/07 1.220.1.64)
8139too net driver updates:
* Support new clone board.
* Count rx-lost-in-ring events.
* Support ethtool-compatible bitmapped printk message enable number.
* Support MII ethtool helper interface, and associated ioctls.
* Replace some MII-related magic numbers with linux/mii.h constants.
* Add in optional (and hopefully preferred... after testing) code
for resetting RX DMA engine on error.
(02/02/07 1.220.1.65)
Remove deprecated SIOCDEVPRIVATE ioctls from several net drivers. The same capabilities have been accessible via SIOCxMIIxxx ioctls
since their introduction in 2.4.x, and it's DaveM's intention to
eventually remove SIOCDEVPRIVATE during the 2.5.x series.
(02/02/07 1.220.7.1)
[PATCH] 2.5: bluetooth compile fix
The llseek cleanup contained a typo in bluetooth/hci_vhci.c that
prevents compile. Thankfully it seems to be the only of this type.
Robert Love
(02/02/07 1.223.1.2)
[PATCH] Input drivers, step #3
This patch updates the input core to the current version, fixing a bunch
of bugs, adding hotplug support and a listing of available input devices
in /proc. The later is very useful for troubleshooting.
Vojtech Pavlik
SuSE Labs
(02/02/07 1.223.1.3)
[PATCH] (1/6) more ->get_sb()
smbfs switched to ->get_sb()
(02/02/07 1.223.1.4)
[PATCH] (2/6) more ->get_sb()
ncpfs.
(02/02/07 1.223.1.5)
[PATCH] (3/6) more ->get_sb()
coda.
(02/02/07 1.223.1.6)
[PATCH] (4/6) more ->get_sb()
we are done with nodev filesystems - removed dead code in super.c and
fs.h
(02/02/07 1.223.1.7)
[PATCH] (5/6) more ->get_sb()
sysvfs and v7 switched.
* super.c: switched to ->get_sb()
* ChangeLog: fixed dates ;-)
(02/02/07 1.223.1.8)
[PATCH] (6/6) more ->get_sb()
minixfs switched.
(02/02/07 1.223.1.9)
[PATCH] BLKGETSIZE64 (bytes not sectors)
Linus, below is a patch to return the size of a device in bytes, not
in units of 512-byte sectors (which is incorrect per function
definition). Tim Pepper submitted an equivalent patch for 2.4.x, now
included in 2.4.18-pre4. Please apply.
(02/02/07 1.223.1.10)
[PATCH] tkill ia64 syscall table entry
minor nit, but it seems the tkill() patch added a sys_tkill entry without
removing an ia64_ni_syscall entry.
(02/02/07 1.220.8.1)
fix compile time bug in usbfs caused by a typo in the superblock changes.
(02/02/07 1.220.9.1)
typo fix, fixing a compile time bug due to the change over to the new
filesystem definition patches in -pre2 (changes pcihpfs_fs_type to
pcihpfs_type).
(02/02/07 1.220.8.2)
Added driverfs support for the USB subsystem.
(02/02/07 1.223.1.11)
[PATCH] thread information block
syscall latency improvement
* There's now an asm/thread_info.h header file with the basic structure
def and asm offsets in it.
* There's now a linux/thread_info.h header file which includes the asm
version and wraps some bitops calls to make convenience functions for
accessing the low-level flags.
* The task_struct has had some fields removed (and some flags), and has
acquired a pointer to the thread_info struct.
* task_struct's are now allocated on slabs in kernel/fork.c, whereas
thread_info structs are allocated at the bottom of the stack pages.
* Some more convenience functions are provided at the end of linux/sched.h to
access flags in other tasks (these are here because they need to access the
task_struct).
(02/02/07 1.223.1.13)
update version
Summary of changes from v2.5.4-pre1 to v2.5.4-pre2
============================================
(02/02/06 1.220.1.1)
[PATCH] Third version of signal changes for thread groups
During the course of developing our pthread library (the NGPT pthread
library) it became clear we needed some kernel support for handling
signals. This patch helps the library by redirecting all signals sent
to tasks in a thread group to the thread group leader. It also defines
the tkill() system call so the library can signal a specific task if
necessary.
Given that as far as I know NGPT is the only user of thread groups, and
that this change would benefit any other user of thread groups, I'm
submitting this for inclusion in the 2.5 kernel.
Note that this patch also adds support for sys_gettid() for the
architectures that don't have it. While this could have been split into a
spearate patch, it would create conflicts since this patch also adds
sys_tkill(), so I felt it was cleaner to leave them together.
Dave McCracken
======================================================================
Dave McCracken IBM Linux Base Kernel Team 1-512-838-3059
[email protected] T/L 678-3059
(02/02/06 1.220.1.2)
[PATCH] (1/2) inode trimming
more inode->u trimming - socket_i is killed.
(02/02/06 1.220.1.3)
[PATCH] (2/2) inode trimming
procfs switched to alloc_inode/destroy_inode.
(02/02/06 1.220.1.4)
[PATCH] (1/5) beginning of getattr series.
added new helpers - vfs_stat(), vfs_lstat() and vfs_fstat().
fs/stat.c switched to use them.
Following patches will
stat(2) variants in arch/* that used to copy inode fields manually
switched to vfs_*stat() and partially cleaned up
irix_...() switched from sys_new*stat() to vfs_*stat() and cleaned
up. Missing LFS check added.
similar for solaris ones
ditto for x86 compatibility ones on ia64.
We are almost ready to switch to ->getattr() - let filesystem decide what
values should go into ->st_... (e.g. for CODA life would become much
easier if it could just use ->i_size of caching file, for supermount
we want ->i_ino inherited from underlying fs, etc.)
Another thing that needs to be done is fixing the rest of LFS/uid size
fsckups in architecture-specific variants of stat() - I've fixed several,
but quite a few are still there.
(02/02/06 1.220.1.5)
[PATCH] (2/5) beginning of getattr series.
stat(2) variants in arch/* that used to copy inode fields manually
switched to vfs_*stat() and partially cleaned up
(02/02/06 1.220.1.6)
[PATCH] (3/5) beginning of getattr series.
irix_...() switched from sys_new*stat() to vfs_*stat() and cleaned up.
Missing LFS check added
(02/02/06 1.220.1.7)
[PATCH] (4/5) beginning of getattr series.
solaris: switched from sys_new*stat() to vfs_*stat() and cleaned up.
Missing LFS check added.
(02/02/06 1.220.1.8)
[PATCH] (5/5) beginning of getattr series.
ia64: switched from sys_new*stat() to vfs_*stat() and cleaned up.
(02/02/06 1.220.1.9)
[PATCH] 2.5.4-pre1: further llseek cleanup (1/3)
This is the first of three patches implementing further llseek cleanup,
against 2.5.4-pre1.
The 'push locking into llseek methods' patch was integrated into 2.5.3.
The networking filesystems, however, do not protect i_size and can not
rely on the inode semaphore used in generic_file_llseek.
This patch implements a remote_llseek method, which is basically the
pre-2.5.3 version of generic_file_llseek. Locking is done via the BKL.
When we have a saner locking system in place, we can push it into this
function in lieu.
Ncpfs, nfs, and smbfs have been converted to use this new llseek.
Note this is updated over the previous posted patch.
Robert Love
(02/02/06 1.220.1.10)
[PATCH] 2.5.4-pre1: further llseek cleanup (2/3)
This is the second patch of three implementing further llseek cleanups,
against 2.5.4-pre1. It does not depend on the other patches.
This patch cleans up various code and quite nicely removes much more
code than it adds. Specifically:
- remove static lseek method which merely reimplements
the standard no_llseek in the following seven files:
hci_vhci.c, ite8172.c, nec_vrc5477.c, auerswald.c,
pipe.c, netlink_dev.c, and socket.c
- remove fs/ufs/file.c::ufs_file_lseek -- Al says it is
reimplementing generic_file_llseek, so let's use that
instead (the comment about 32-bit sizes shouldn't be
an issue, the generic method checks size)
- include smp_lock.h in 3 files missed from previous
'remove bkl' patch
- Documentation/filesystem/Locking update
Please, apply.
Robert Love
(02/02/06 1.220.1.11)
[PATCH] 2.5.4-pre1: further llseek cleanup (3/3)
The previous patch did not provide protection for device lseek methods
(drivers/* stuff). This patch pushes the BKL into each of the remaining
lseek methods -- without them we have a race.
I'd much prefer to have a a better lock to push down than the BKL, but
that will have to wait.
Before you balk at the size, remember patch #2 in this series which
removed much code ;-)
Thanks to Al for assistance, especially a listing of affected files.
Robert Love
(02/02/06 1.220.1.12)
[PATCH] hpfs cleanup (switch from sleep_on() to semaphore)
Pair (flag, wait_queue) in hpfs replaced with semaphore.
(02/02/06 1.220.1.13)
[PATCH] death of hfs rename_lock
Useless (lock,waitqueue) pair killed - we have fs-wide
exclusion between renames enforced by VFS, so hfs rename_{lock,wait}
can be removed.
(02/02/06 1.220.1.14)
[PATCH] more sleep_on() removals
Yet another flag/wait_queue pair bites the dust - hfs bitmap_{lock,wait}
was trying to implement a semaphore.
(02/02/06 1.220.1.15)
[PATCH] missing ->i_zombie in hpux_getdents()
hpux_getdents() contains a years-old copy of code from fs/readdir.c.
Switched to use of vfs_readdir(), which does correct locking.
(02/02/06 1.220.1.16)
[PATCH] fix for exit_fat_fs()
Kudos to OGAWA Hirofumi for finding that one - gcc barfs on the
current code, trying to define an alias to external symbol.
(02/02/06 1.220.1.17)
[PATCH] reiserfs fix for inodes with wrong item versions (2.5)
This is hopefully last bugfix for a bug introduced by struct inode splitting.
Because of setting i_flags to some value and then cleaning the i_flags
contents later, on-disk items received wrong item version ob v3.6 filesystems
(02/02/06 1.220.1.18)
fatfs_syms.c:
Al Viro: fix a typo (destory->destroy)
(02/02/06 1.220.2.1)
patch from Wolfgang Mües for the usb auerswald.c driver:
- Code-Review from Oliver Neukum: remove SMP races.
- Added some wake_up calls after auerbuf_releasebuf to wake up tasks waiting
for cp buffers.
- Change the module count handling to automatic (owner: THIS_MODULE).
(02/02/06 1.220.3.1)
patch from dimitry for the usb ibmcam driver that does:
- Updates the documentation
- Adds Veo Stingray support
- Fixes hotplug table dependency upon now-defunct symbol
- deletes drivers/usb/ibmcam.h
(02/02/06 1.220.4.1)
usb core:
- initial add of the "int mem_flags" call to the usb_submit_urb() call
(02/02/06 1.220.4.2)
usb host controllers:
- add mem_flags support
- portions of this patch by Oliver Neukum and David Brownell.
(02/02/06 1.220.4.3)
"mem_flags" support for the cris usb host controller.
These changes have been approved by the CRIS maintainer.
(02/02/06 1.220.4.4)
usb_submit_urb() changes for the drivers/usb/*.c drivers.
Some driver changes were done by Oliver Neukum and David Brownell.
(02/02/06 1.220.4.5)
usb_submit_urb() changes for all drivers/usb/serial/*.c drivers
(02/02/06 1.220.4.6)
usb_submit_urb() changes.
Patch done by Oliver Neukum.
(02/02/06 1.220.4.7)
usb_submit_urb() changes for USB drivers that live outside of the drivers/usb subdirectory.
(02/02/06 1.220.1.20)
[PATCH] (1/8) ->get_sb() switchover
OK, here comes the long-promised switchover to ->get_sb().
New method added - ->get_sb(type, flags, dev, data). At this point we
still keep ->read_super(), but it will go away in the end of series.
->get_sb() is a generalization of get_sb_{bdev,nodev,single}() - as the
matter of fact, these will become common helpers used by filesystems in
their ->get_sb().
(02/02/06 1.220.1.21)
[PATCH] (2/8) ->get_sb() switchover
get_sb_bdev() et.al. get a new argument (fill_super) and become
exported.
(02/02/06 1.220.1.22)
[PATCH] (3/8) ->get_sb() switchover
ext2 switched to ->get_sb(). For that we had
a) turned ext2_read_super() into ext2_fill_super() (instead of
"return NULL if failed and sb if succeeded" it's "return error value or
0").
b) added ext2_get_sb(type, flags, dev, data) - all it does is
return get_sb_bdev(type, flags, dev, data, ext2_fill_super);
c) replaced DECLARE_...() with explicit initializer.
That's it.
(02/02/06 1.220.1.23)
[PATCH] (4/8) ->get_sb() switchover
convert procfs to ->get_sb(). Similar to ext2, except that we use
get_sb_single() instead of get_sb_bdev(). Notice the we _don't_ keep
FS_SINGLE in flags - as the matter of fact, with ->get_sb() FS_SINGLE is
useless.
(02/02/06 1.220.1.24)
[PATCH] (5/8) ->get_sb() switchover
converte binfmt_misc to ->get_sb()
(02/02/06 1.220.1.25)
[PATCH] (6/8) ->get_sb() switchover
converte usb to ->get_sb()
(02/02/06 1.220.1.26)
[PATCH] (7/8) ->get_sb() switchover
converted ramfs/rootfs to ->get_sb(). Use get_sb_nodev() as helper.
(02/02/06 1.220.1.27)
[PATCH] (8/8) ->get_sb() switchover
converted fs/block_dev.c ("bdev") to ->get_sb. use get_sb_nodev() as
helper.
(02/02/06 1.220.1.28)
[PATCH] (1/6) more ->get_sb() stuff
Fix for idiotic typo in __get_sb_single().
(02/02/06 1.220.1.29)
[PATCH] (2/6) more ->get_sb() stuff
devpts converted.
(02/02/06 1.220.1.30)
[PATCH] (3/6) more ->get_sb() stuff
pcihpfs converted.
(02/02/06 1.220.1.31)
[PATCH] (4/6) more ->get_sb() stuff
driverfs converted.
(02/02/06 1.220.1.32)
[PATCH] (5/6) more ->get_sb() stuff
devfs converted.
(02/02/06 1.220.1.33)
[PATCH] (6/6) more ->get_sb() stuff
Now that all FS_SINGLE filesystems are using ->get_sb(), FS_SINGLE
can die.
(02/02/06 1.220.5.1)
[PATCH] Slab name cleanup - last try
Hallo Linus,
This is the slab name cleanup you originally requested when the reiserfs
slab name too long boot bug was introduced. It replaces the static
name limit for slab cache names with the requirement for the caller
to supply a stable name. This version addresses all concerns that were given
over earlier versions of the patch.
As I do not feel very strongly about the slab cache name limit and you
also do not seem to be interested in it and the original bug of reiserfs
not booting anymore has been long fixed in your tree with the patch
you originally rejected from me I will not resend this patch after this
try.
Patch against 2.5.4pre1.
-Andi
(02/02/06 1.220.5.2)
[PATCH] Bootmem speedup
This patch speeds the bootmem freeing up a bit. Not particularly important,
but helps on some slow simulators where the loop can chew up significant
CPU time.
(02/02/06 1.220.5.3)
[PATCH] Automatic file-max sizing
The default for NR_FILES of 8192 is far too low for many workloads. This
patch does dynamic sizing for it instead. It assumes file+inode+dentry
are roughly 1K and will use upto 10% of the memory for it.
Also removes two obsolete prototypes.
(02/02/06 1.220.5.4)
[PATCH] Fix page cache limit wrapping in filesystems
Several file systems in tree that nominally support files >2GB set their
s_maxbytes value to ~0ULL. This has the nasty side effect on 32bit machines
that when a file write reaches the page cache limit (e.g. 2^43) it'll silently
wrap and destroy data at the beginning of the file.
This patch changes the file systems in question to fill in a proper limit.
I also have an alternate patch that adds a check for this generically
in super.c, but preliminary comments from Al suggested that he prefered
to do it in the file systems, so it is done this way way.
(02/02/06 1.220.5.5)
[PATCH] Fix mount hash table
On my 512MB machine with 6 mount points the mount hash table uses 64K.
This patch brings it to a more reasonable size by limiting it to one
page.
(02/02/06 1.220.1.35)
update version number
(02/02/06 1.220.1.36)
[PATCH] (1/9) more ->get_sb() stuff
More ->get_sb() patches.
shmem/tmpfs switched
(02/02/06 1.220.1.37)
[PATCH] (2/9) more ->get_sb() stuff
More ->get_sb() patches: fix for usbdevfs - forgot to remove
DECLARE_FSTYPE() after adding explicit initializer.
(02/02/06 1.220.1.38)
[PATCH] (3/9) more ->get_sb() stuff
More ->get_sb() patches: pipefs
(02/02/06 1.220.1.39)
[PATCH] (4/9) more ->get_sb() stuff
More ->get_sb() patches: sockfs
(02/02/06 1.220.1.40)
[PATCH] (5/9) more ->get_sb() stuff
More ->get_sb() patches: openpromfs
(02/02/06 1.220.1.41)
[PATCH] (6/9) more ->get_sb() stuff
More ->get_sb() patches: capifs
(02/02/06 1.220.1.42)
[PATCH] (7/9) more ->get_sb() stuff
More ->get_sb() patches: autofs
(02/02/06 1.220.1.43)
[PATCH] (8/9) more ->get_sb() stuff
More ->get_sb() patches: autofs4
(02/02/06 1.220.1.44)
[PATCH] (9/9) more ->get_sb() stuff
More ->get_sb() patches: nfs. It still acts as nodev - what we want to
(and can) do now is to give it enough smarts to recognize that same tree
might be already mounted.
We are almost done with nodev filesystems by now.
Summary of changes from v2.5.3 to v2.5.4-pre1
============================================
(02/02/05 1.168)
[PATCH] Dup in drivers/net/Config.in
Don't offer SunLANCE twice.
(02/02/05 1.169)
[PATCH] Fix ESP thinko in 2.5.3-final
I think I told you to revert this bit from 2.5.3, but here
it is in patch form anyways. Whoever made this change didn't
read the driver, and well... didn't even build test it either :-)
(02/02/05 1.170)
[PATCH] Fix IDE printf formatting
The usual "u64 is long long only on some platforms" problem.
(02/02/05 1.171)
[PATCH] OSST kdev_t fixes
MINOR --> minor
MKDEV --> mk_kdev
(02/02/05 1.172)
[PATCH] Fix typo in i386 PCI header
I made a typo the other weeks while renaming the interfaces for you,
oops. Please apply, thanks.
(02/02/05 1.173)
[PATCH] malloc.h references
linux/malloc.h --> linux/slab.h
(02/02/05 1.174)
[PATCH] Fix UFS build
Missing smp_lock.h inclusion.
(02/02/05 1.175)
[PATCH] Missing ZLIB export
(02/02/05 1.176)
[PATCH] Sparc updates
Gets sparc64 in sync with 2.5.3 final changes.
(02/02/05 1.177)
[PATCH] 2.5.3-pre6: ecr
This patch (from 2.4.x) cleans up the use of the ECR in parport_pc.
* drivers/parport/parport_pc.c: Integrate fixes and cleanups
from Damian Gruszka (VScom).
* drivers/parport/ChangeLog: Updated.
(02/02/05 1.178)
[PATCH] 2.5.3-pre6: getmodes
This patch prevents ppdev from oopsing when the PPGETMODES ioctl is
used before a PPCLAIM.
* drivers/char/ppdev.c: Fix an oops in PPGETMODES handling.
(02/02/05 1.179)
[PATCH] 2.5.3-pre6: console
I finally found the reason that printer console sometimes acted up
(duh):
* drivers/char/lp.c: Fix printer console.
(02/02/05 1.180)
[PATCH] 2.5.3-pre6: deadlock
This patch fixes a potential deadlock in ppdev.
* drivers/char/ppdev.c: Watch out for errors from
parport_claim_or_block.
* drivers/parport/share.c: Watch out for signals.
* drivers/parport/ChangeLog: Updated.
(02/02/05 1.181)
[PATCH] 2.5.3-pre6: mode
This patch paves the way for a new driver which needs the
functionality. Now parport_daisy_select actually _uses_ its mode
parameter.
* drivers/parport/daisy.c: Make parport_daisy_select aware of
its 'mode' parameter.
* drivers/parport/ChangeLog: Updated.
(02/02/05 1.182)
[PATCH] nbd in 2.5.3 does not work, and can cause severe damage when read-write
Hi Linus,
I've got strange idea and tried to build diskless machine around
2.5.3... Besides problem with segfaulting crc32 (it is initialized after
net/ipv4/ipconfig.c due to lib/lib.a being a library... I had to hardcode
lib/crc32.o before --start-group in main Makefile, but it is another
story) there is bad problem with NBD caused by BIO changes:
(1) request flags were immediately put into on-wire request format.
In the past, we had 0=READ, !0=WRITE. Now only REQ_RW bit determines
direction. As nbd-server from nbd distribution package treats any
non-zero value as write, it performs writes instead of read. Fortunately
it will die due to other consistency checks on incoming request, but...
(2) nbd servers handle only up to 10240 byte requests. So setting max_sectors
to 20 is needed, as otherwise nbd server commits suicide. Maximum request size
should be handshaked during nbd initialization, but currently just use
hardwired 20 sectors, so it will behave like it did in the past.
(02/02/05 1.183)
[PATCH] typo in drivers/scsi/megaraid.h
A trivial patch that fixes this irritation in my dmesg, 2.5.3:
megaraid: v1.18 (Release Date: Thu Oct 11 15:02:53 EDT 2001
)<5>megaraid: found 0x8086:0x1960:idx 0:bus 2:slot 5:func 1
scsi0 : Found a MegaRAID controller at 0xe089c000, IRQ: 12
Please apply.
(02/02/05 1.184)
[PATCH] fix xconfig for new help system
Here is a patch to enhance xconfig to read the new Config.help files.
Olaf Dietsche wrote this, and Steven Cole passed it on to me.
Testing: Steven Cole tested it, and I tested it.
(02/02/05 1.185)
[PATCH] driver model updates (1/5)
Patch 1: Make device_driver_init() an initcall.
It declares it as subsys_initcall and removes the explicit call from
init/main.c::do_basic_setup().
(02/02/05 1.186)
[PATCH] driver model updates (3/5)
Patch 3: Make default callbacks simpler.
I want to move as much to a 1 file/1 value model as possible. I haven't
come up with a clean way to enforce it except via social pressure.
This patch is a step in that direction. It:
- Reduces the output of 'power' to just the decimal state of the device
- Adds a 'name' file which exports just the device name
- Reduces the 'status' file to just export the bus ID. (This will change,
since the bus ID is obvious based on what directory you're in, but it's
another patch at another time)
(02/02/05 1.187)
[PATCH] driver model updates (4/5)
Patch 4: Add some default files for PCI devices.
This adds two files for PCI devices: 'irq' and 'resources'. They display
just those things and currently do nothing on write. These are the
examples for other subsystems to use for creating files ('Hey, look how
simple it is!')
(02/02/05 1.188)
[PATCH] driver model updates (5/5)
Remove struct iobus.
There is a lot of duplication between struct device and struct iobus, both
in their members and the code in their interfaces. Waxing struct iobus
removes this duplication and makes things a bit simpler.
(02/02/05 1.189)
[PATCH] reiserfs patchset, patch 1 of 9 01-pick_correct_key_version.diff
01-pick_correct_key_version.diff
This is to fix certain cases where items may get its keys to be interpreted
wrong, or to be inserted into the tree in wrong order. This bug was only
observed live on 2.5.3, though it is present in 2.4, too.
(02/02/05 1.190)
[PATCH] reiserfs patchset, patch 2 of 9 02-prealloc_list_init.diff
02-prealloc_list_init.diff
prealloc list was forgotten to be initialised.
(02/02/05 1.191)
[PATCH] reiserfs patchset, patch 3 of 9 03-key_output_fix.diff
03-key_output_fix.diff
Fix all the places where cpu key is attempted to be printed as ondisk key
(02/02/05 1.192)
[PATCH] reiserfs patchset, patch 4 of 9 04-nfs_stale_inode_access.diff
04-nfs_stale_inode_access.diff
This is to fix a case where stale NFS handles are correctly detected as
stale, but inodes assotiated with them are still valid and present in cache,
hence there is no way to deal with files, these handles are attached to.
Bug was found and explained by
Anne Milicia
(02/02/05 1.193)
[PATCH] reiserfs patchset, patch 5 of 9 05-kernel-reiserfs_fs_h-offset_v2.diff
05-kernel-reiserfs_fs_h-offset_v2.diff
Convert erroneous le64_to_cpu to cpu_to_le64
(02/02/05 1.194)
[PATCH] reiserfs patchset, patch 6 of 9 06-return_braindamage_removal.diff
06-return_braindamage_removal.diff
Kill stupid code like 'goto label ; return 1;'
(02/02/05 1.195)
[PATCH] reiserfs patchset, patch 7 of 9 07-remove_nospace_warnings.diff
07-remove_nospace_warnings.diff
Do not print scary warnings in out of free space situations.
(02/02/05 1.196)
[PATCH] reiserfs patchset, patch 8 of 9 08-unfinished_rebuildtree_message.diff
08-unfinished_rebuildtree_message.diff
Give a proper explanation if unfinished reiserfsck --rebuild-tree
run on a fs was detected.
(02/02/05 1.197)
[PATCH] reiserfs patchset, patch 9 of 9 09-64bit_bitops_fix-1.diff
09-64bit_bitops_fix-1.diff
Bitopts arguments must be long, not int.
(02/02/05 1.198)
[PATCH] USB Kawasaki driver maintainer change
Here's a patch against 2.5.3 that changes the maintainer of the USB
Kawasaki driver to Oliver Neukum.
(02/02/05 1.199)
[PATCH] USB Config.help update
Here's a patch against 2.5.3 that updates the Config.help entries for
the USB microtek and hpusbscsi drivers.
This patch was done by Oliver Neukum.
(02/02/05 1.200)
[PATCH] USB Kaweth driver update
Here's a patch against 2.5.3 for the USB kaweth driver that does the
following:
- removes SMP deadlock
- removes nfs deadlock
- fixes a memory leak when the firmware is not loaded.
- few other minor cleanups.
This patch was done by Oliver Neukum.
(02/02/05 1.201)
[PATCH] USB pegasus driver update
Here's a patch against 2.5.3 for the USB pegasus driver that does the
following:
- fixes __FUNCTION__ warnings on gcc-3.0.3 and up
- added 3 more devices
- fixed memory leak
This patch was done by Petko Manolov and Oliver Neukum.
(02/02/05 1.202)
[PATCH] USB printer driver update
Here's a patch against 2.5.3 for the USB printer driver that does the
following:
- removes the races inherent in sleep_on
- uses 2.5 style of module usage counting
- kills a lockup on failure of usb_submit_urb
This patch was done by Oliver Neukum.
(02/02/05 1.203)
[PATCH] USB stv680 driver update
Here's a patch against 2.5.3 for the USB stv680 driver that fixes two
bugs in the existing driver. This patch was done by Kevin Sisson.
(02/02/05 1.204)
[PATCH] USB core update
Here's a patch against 2.5.3 for the USB core that fixes a possible
initialization bug for some platforms when allocating a new usb, and
changes the warning level on a message (it isn't an error.) This patch
was done by Oliver Neukum and David Brownell.
(02/02/05 1.205)
[PATCH] USB vicam driver update
Here's a patch against 2.5.3 for the USB vicam driver that removes the
use of interruptible_sleep_on() in the driver. This patch was done by
Oliver Neukum.
(02/02/05 1.206)
[PATCH] USB ohci-hcd driver update
Here's a patch against 2.5.3 for the USB ohci-hcd driver that does the
following:
- doesn't assume CONFIG_DEBUG_SLAB
- unlink from interrupt completions now work
- doesn't force debugging on
- updated copyright / license statements
- slightly smaller object size
- fewer inlined magic numbers
- removes unused fields from data structures
- header file reorg, doc fixup
This patch was done by David Brownell.
(02/02/05 1.207)
[PATCH] NFS lookup code rewrite w/o open(".") fix...
This is a resend of the NFS lookup code rewrite, but with the open(".")
VFS fix removed. (I'll resend the 'uses d_revalidate()' version
separately after a suitable delay to allow for comments.)
Issues fixed by this patch:
- Use the directory mtime in order to give us a hint when we should
check for namespace changes.
- Add support for the 'nocto' flag, in order to turn off the strict
attribute cache revalidation on file open().
- Simplify inode lookup. Don't check the 'fsid' field (which appears
to be buggy in too many servers in order to be reliable). Instead
we only rely on the inode number (a.k.a. 'fileid') and the
(supposedly unique) filehandle.
(02/02/05 1.208)
[PATCH] Fix spurious ETXTBSY errors due to late release of struct file
The following patch should fix a problem of ETXTBSY sometimes
occurring if one tries to run a file straight after compilation.
The problem is that both NFS read and write requests can currently
hold a count on the struct file. This is done partly so as to be able
to pass along the RPC credential (which is cached in the struct file),
and partly so that asynchronous writes can report any errors via the
file->f_error mechanism.
The problem is that both the read and write requests may persist even
after file close() occurs. For O_RDONLY files, this is not a problem,
but for O_WRONLY, and O_RDWR files, the fact that the struct file is
not released until the last call to nfs_release_request() means that
inode->i_writecount does not necessarily get cleared upon file
close().
The following patch fixes both these issues.
- NFS read requests no longer hold the struct file. They take a
count on the the RPC credential itself.
- NFS write requests still hold the struct file, since they want to
report errors to sys_close() using the file->f_error mechanism.
However they are made to release the page, creden
Linux 2.5.4 Released
By
Get the Free Newsletter!
Subscribe to Developer Insider for top news, trends, & analysis