Linus Torvalds [Fri, 2 May 2008 15:20:43 +0000 (08:20 -0700)]
Merge branch 'for-linus' of git://www.linux-m32r.org/git/takata/linux-2.6_dev
* 'for-linus' of git://www.linux-m32r.org/git/takata/linux-2.6_dev:
m32r: cleanup: drop .data.idt section in vmlinux.lds script
m32r: use KBUILD_DEFCONFIG
* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus:
lguest: make Launcher see device status updates
lguest: remove bogus NULL cpu check
lguest: avoid using NR_CPUS as a bounds check.
virtio: add virtio disk geometry feature
virtio: explicit advertisement of driver features
virtio: change config to guest endian.
virtio: finer-grained features for virtio_net
virtio: wean net driver off NETDEV_TX_BUSY
virtio-blk: fix remove oops
virtio: fix scatterlist sizing in net driver.
virtio: de-structify virtio_block status byte
virtio: export more headers to userspace
virtio: fix sparse return void-valued expression warnings
virtio: fix tx_ stats in virtio_net
virtio: ignore corrupted virtqueues rather than spinning.
Rusty Russell [Sat, 3 May 2008 02:50:51 +0000 (21:50 -0500)]
lguest: avoid using NR_CPUS as a bounds check.
NR_CPUS (being a host number) is an arbitrary limit for the Guest.
Using the array size directly (which currently happes to be NR_CPUS)
is more futureproof.
Ryan Harper [Wed, 16 Apr 2008 18:56:37 +0000 (13:56 -0500)]
virtio: add virtio disk geometry feature
Rather than faking up some geometry, allow the backend to push the disk
geometry via virtio pci config option. Keep the old geo code around for
compatibility.
Rusty Russell [Sat, 3 May 2008 02:50:50 +0000 (21:50 -0500)]
virtio: explicit advertisement of driver features
A recent proposed feature addition to the virtio block driver revealed
some flaws in the API: in particular, we assume that feature
negotiation is complete once a driver's probe function returns.
There is nothing in the API to require this, however, and even I
didn't notice when it was violated.
So instead, we require the driver to specify what features it supports
in a table, we can then move the feature negotiation into the virtio
core. The intersection of device and driver features are presented in
a new 'features' bitmap in the struct virtio_device.
Note that this highlights the difference between Linux unsigned-long
bitmaps where each unsigned long is in native endian, and a
straight-forward little-endian array of bytes.
Drivers can still remove feature bits in their probe routine if they
really have to.
API changes:
- dev->config->feature() no longer gets and acks a feature.
- drivers should advertise their features in the 'feature_table' field
- use virtio_has_feature() for extra sanity when checking feature bits
Rusty Russell [Sat, 3 May 2008 02:50:49 +0000 (21:50 -0500)]
virtio: change config to guest endian.
A recent proposed feature addition to the virtio block driver revealed
some flaws in the API, in particular how easy it is to break big
endian machines.
The virtio config space was originally chosen to be little-endian,
because we thought the config might be part of the PCI config space
for virtio_pci. It's actually a separate mmio region, so that
argument holds little water; as only x86 is currently using the virtio
mechanism, we can change this (but must do so now, before the
impending s390 merge).
API changes:
- __virtio_config_val() just becomes a striaght vdev->config_get() call.
Rusty Russell [Sat, 3 May 2008 02:50:46 +0000 (21:50 -0500)]
virtio: finer-grained features for virtio_net
So, we previously had a 'VIRTIO_NET_F_GSO' bit which meant that 'the
host can handle csum offload, and any TSO (v4&v6 incl ECN) or UFO
packets you might want to send. I thought this was good enough for
Linux, but it actually isn't, since we don't do UFO in software.
So, add separate feature bits for what the host can handle. Add
equivalent ones for the guest to say what it can handle, because LRO
is coming too (thanks Herbert!).
Rusty Russell [Sat, 3 May 2008 02:50:45 +0000 (21:50 -0500)]
virtio: fix scatterlist sizing in net driver.
Herbert Xu points out (within another patch) that my scatterlists are
too short: one entry for the gso header, one for the skb->data, and
MAX_SKB_FRAGS for all the fragments.
Fix both xmit and recv sides (recv currently unused, coming in later
patch).
is there a reason why we dont export the virtio headers for
9p, balloon, console, pci, and virtio_ring? kvm uses make sync,
but I think it is still useful to heave these headers exported
as they might be useful for other userspace tools.
I dont export virtio.h, because it does not seem to have useful
information for userspace and it requires scatterlist.h which is
also not exported. See also my other mail about your "virtio:
change config to guest endian." patch.
Rusty Russell [Sat, 3 May 2008 02:50:43 +0000 (21:50 -0500)]
virtio: ignore corrupted virtqueues rather than spinning.
A corrupt virtqueue (caused by the other end screwing up) can have
strange results such as a driver spinning: just bail when we try to
get a buffer from a known-broken queue.
Al Viro [Thu, 1 May 2008 21:36:36 +0000 (14:36 -0700)]
[IA64] fix file and descriptor handling in perfmon
Races galore... General rule: as soon as it's in descriptor table,
it's over; another thread might have started IO on it/dup2() it
elsewhere/dup2() something *over* it/etc. fd_install() is the very
last step one should take - it's a point of no return.
Besides, the damn thing leaked on failure exits...
Roland McGrath [Wed, 30 Apr 2008 21:40:14 +0000 (14:40 -0700)]
[IA64] TS_RESTORE_SIGMASK
Replace TIF_RESTORE_SIGMASK with TS_RESTORE_SIGMASK and define
our own set_restore_sigmask() function. This saves the costly
SMP-safe set_bit operation, which we do not need for the sigmask
flag since TIF_SIGPENDING always has to be set too.
Linus Torvalds [Thu, 1 May 2008 18:31:38 +0000 (11:31 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6:
firewire: fw-sbp2: log scsi_target ID at release
ieee1394: fix NULL pointer dereference in sysfs access
Jared Hulbert [Wed, 30 Apr 2008 06:26:49 +0000 (23:26 -0700)]
[MTD][NOR] Add physical address to point() method
Adding the ability to get a physical address from point() in addition
to virtual address. This physical address is required for XIP of
userspace code from flash.
Stefan Richter [Thu, 1 May 2008 08:43:04 +0000 (10:43 +0200)]
ieee1394: fix NULL pointer dereference in sysfs access
Regression since "ieee1394: prevent device binding of raw1394,
video1394, dv1394", commit d2ace29fa44589da51fedc06a67b3f05301f3bfd:
$ cat /sys/bus/ieee1394/drivers/raw1394/device_ids
triggers a NULL pointer dereference in fw_show_drv_device_ids.
Reported by Miles Lane.
David Woodhouse [Thu, 1 May 2008 17:47:17 +0000 (18:47 +0100)]
[JFFS2] Track parent inode for directories (for NFS export)
To support NFS export, we need to know the parent inode of directories.
Rather than growing the jffs2_inode_cache structure, share space with
the nlink field -- which was always set to 1 for directories anyway.
With s390 the last arch switched to the generic sys_ptrace yesterday so
we can now kill the ifdef around it to enforce every new port it using
it instead of introducing new weirdo versions.
Linus Torvalds [Thu, 1 May 2008 15:28:26 +0000 (08:28 -0700)]
Merge branch 'release' of git://lm-sensors.org/kernel/mhoffman/hwmon-2.6
* 'release' of git://lm-sensors.org/kernel/mhoffman/hwmon-2.6:
hwmon: (adt7473) minor cleanup / refactoring
hwmon: (asb100) Remove some dead code
hwmon: (lm75) Fix an incorrect comment
hwmon: (w83793) VID and VRM handling cleanups
hwmon: (w83l785ts) Don't ask the user to report failures
hwmon: (smsc47b397) add a new chip id (0x8c)
linux-2.6-mk68/drivers/net/fec.c: In function 'fec_enet_module_init':
linux-2.6-mk68/drivers/net/fec.c:2627: warning: unused variable 'j'
linux-2.6-mk68/drivers/net/fec.c: At top level:
linux-2.6-mk68/drivers/net/fec.c:2136: warning: 'mii_link_interrupt' defined but not used
Renamed the 5272 hash_table registers to match the "grp" hash_table
registers of the other ColdFire parts. They are actually a group hash.
The makes for consistent setup across all ColdFire parts.
m68knommu: ColdFire add support for kernel preemption
As the subject says this patch adds the support for kernel preemption
on m68knommu Coldfire. I thing the same changes could be applied to
68360 & 68328 but since I don't have the HW for testing, I don't touch it.
Wilson Callan [Thu, 1 May 2008 02:16:28 +0000 (12:16 +1000)]
m68knommu: fix signal handling return path
The return from software signal handling pushes code on the stack
that system calls to the kernels cleanup code. This is borrowed
directly from the m68k linux signal handler.
The rt signal case is not quite right for the restricted instruction
set of the ColdFire parts. And neither the normal signal case or rt
signal case properly flushes/pushes the appropriate cache lines.
Rework the return path to just call back through some code fragments
in the kernel proper (with no MMU in the way we can do this). No
cache problems, and less code overall.
Andrew Morton [Thu, 1 May 2008 11:35:20 +0000 (04:35 -0700)]
drivers/scsi/mvsas.c: fix printk warnings
drivers/scsi/mvsas.c: In function `mvs_update_phyinfo':
drivers/scsi/mvsas.c:2822: warning: long long unsigned int format, u64 arg (arg 5)
drivers/scsi/mvsas.c:2822: warning: long long unsigned int format, u64 arg (arg 6)
We do not know what type the arch uses to implement u64.
Andrew Morton [Thu, 1 May 2008 11:35:20 +0000 (04:35 -0700)]
drivers/scsi/ncr53c8xx.c: fix warning
drivers/scsi/ncr53c8xx.c: In function 'process_waiting_list':
drivers/scsi/ncr53c8xx.c:8225: warning: suggest parentheses around assignment used as truth value
Signed-off-by: Alan Cox <[email protected]> Cc: Paul Fulghum <[email protected]> Cc: Jiri Slaby <[email protected]>
drivers/char/synclink_gt.c: In function 'put_char':
drivers/char/synclink_gt.c:919: warning: 'ret' may be used uninitialized in this function
and do some whitespace repair and unneeded-cast-removal in there as well.
Andrew Liu [Thu, 1 May 2008 11:35:14 +0000 (04:35 -0700)]
workqueue: remove redundant function invocation
timer_stats_timer_set_start_info is invoked twice, additionally, the
invocation of this function can be moved to where it is only called when a
delay is really required.
Jan Blunck [Thu, 1 May 2008 11:35:10 +0000 (04:35 -0700)]
autofs: path_{get,put}() cleanups
Here are some more places where path_{get,put}() can be used instead of
dput()/mntput() pair. Besides that it fixes a bug in autofs4_mount_busy()
where mntput() was called before dput().
Jeff Moyer [Thu, 1 May 2008 11:35:09 +0000 (04:35 -0700)]
autofs4: fix incorrect return from root.c:try_to_fill_dentry()
Jeff Moyer has identified a case where the autofs4 function
root.c:try_to_fill_dentry() can return -EBUSY when it should return 0.
Jeff's description of the way this happens is:
"automount starts an expire for directory d. after the callout to the daemon,
but before the rmdir, another process tries to walk into the same directory.
It puts itself onto the waitq, pending the expiration.
When the expire finishes, the second process is woken up. In
try_to_fill_dentry, it does this check:
status = d_invalidate(dentry);
if (status != -EBUSY)
return -EAGAIN;
And status is EBUSY. The dentry still has a non-zero d_inode, and the
flags do not contain LOOKUP_CONTINUE or LOOKUP_DIRECTORY
So, we fall through and return -EBUSY to the caller."
Jeff Moyer [Thu, 1 May 2008 11:35:08 +0000 (04:35 -0700)]
autofs4: fix execution order race in mount request code
Jeff Moyer has identified a race in due to an execution order dependency
in the autofs4 function root.c:try_to_fill_dentry().
Jeff's description of this race is:
"P1 does a lookup of /mount/submount/foo. Since the VFS can't find an entry
for "foo" under /mount/submount, it calls into the autofs4 kernel module to
allocate a new dentry, D1. The kernel creates a new waitq for this lookup and
calls the daemon to perform the mount.
The daemon performs a mkdir of the "foo" directory under /mount/submount,
which ends up creating a *new* dentry, D2.
Then, P2 does a lookup of /mount/submount/foo. The VFS path walking logic
finds a dentry in the dcache, D2, and calls the revalidate function with this.
In the autofs4 revalidate code, we then trigger a mount, since the dentry is
an empty directory that isn't a mountpoint, and so set DCACHE_AUTOFS_PENDING
and call into the wait code to trigger the mount.
The wait code finds our existing waitq entry (since it is keyed off of the
directory name) and adds itself to the list of waiters.
After the daemon finishes the mount, it calls back into the kernel to release
the waiters. When this happens, P1 is woken up and goes about clearing the
DCACHE_AUTOFS_PENDING flag, but it does this in D1! So, given that P1 in our
case is a program that will immediately try to access a file under
/mount/submount/foo, we end up finding the dentry D2 which still has the
pending flag set, and we set out to wait for a mount *again*!
So, one way to address this is to re-do the lookup at the end of
try_to_fill_dentry, and to clear the pending flag on the hashed dentry. This
seems a sane approach to me."
Matthew Wilcox [Thu, 1 May 2008 11:35:04 +0000 (04:35 -0700)]
spi_s3c24xx signedness fix
On Fri, Apr 18, 2008 at 09:08:55PM +0200, Julia Lawall wrote:
> I found 63 occurrences of this problem with the following semantic match
> (http://www.emn.fr/x-info/coccinelle/):
>
> @@ unsigned int i; @@
>
> * i < 0
>
Since this one's always in the range 0-255, it could probably be made
signed, but it's just as easy to make it work unsigned.
Jean Delvare [Thu, 1 May 2008 11:35:01 +0000 (04:35 -0700)]
dmi: clean-up dmi helper declarations
The declaration of dmi helper functions is a bit messy and inconsistent at the
moment:
* On ia64 they are declared in <asm/io.h>.
* On x86-64 they are declared in <asm/dmi.h>.
* On i386 they are declared both in <asm/io.h> and <asm/dmi.h>.
Fix the header files so that the dmi helper functions are consistently
defined in <asm/dmi.h>.
Randy Dunlap [Thu, 1 May 2008 11:34:55 +0000 (04:34 -0700)]
PNP: fix printk format warnings
next-20080430/drivers/pnp/pnpbios/rsparser.c:594: warning: format '%d' expects type 'int', but argument 4 has type 'resource_size_t'
next-20080430/drivers/pnp/pnpbios/rsparser.c:605: warning: format '%d' expects type 'int', but argument 4 has type 'resource_size_t'
Sam Ravnborg [Thu, 1 May 2008 11:34:51 +0000 (04:34 -0700)]
pcmcia: silence section mismatch warnings from pci_driver variables
Silence following warnings:
WARNING: drivers/pcmcia/built-in.o(.data+0x14e0): Section mismatch in reference from the variable pd6729_pci_drv to the function .devinit.text:pd6729_pci_probe()
WARNING: drivers/pcmcia/built-in.o(.data+0x14e8): Section mismatch in reference from the variable pd6729_pci_drv to the function .devexit.text:pd6729_pci_remove()
WARNING: drivers/pcmcia/built-in.o(.data+0x16c0): Section mismatch in reference from the variable i82092aa_pci_drv to the function .devinit.text:i82092aa_pci_probe()
WARNING: drivers/pcmcia/built-in.o(.data+0x16c8): Section mismatch in reference from the variable i82092aa_pci_drv to the function .devexit.text:i82092aa_pci_remove()
Rename the variables from *_drv to *_driver so modpost ignore the OK
references to __devinit/__devexit functions.
Sam Ravnborg [Thu, 1 May 2008 11:34:50 +0000 (04:34 -0700)]
pcmcia: silence section mismatch warnings from class_interface variables
Silence the following warnings:
WARNING: drivers/pcmcia/built-in.o(.data+0x6e8): Section mismatch in reference from the variable pcmcia_bus_interface to the function .devinit.text:pcmcia_bus_add_socket()
WARNING: drivers/pcmcia/built-in.o(.data+0xa88): Section mismatch in reference from the variable pccard_rsrc_interface to the function .devinit.text:pccard_sysfs_add_rsrc()
WARNING: drivers/pcmcia/built-in.o(.data+0xa90): Section mismatch in reference from the variable pccard_rsrc_interface to the function .devexit.text:pccard_sysfs_remove_rsrc()
The variables of type class_interface contains references
to __devinit and __devexit functions which is OK.
Silence warnings by annotating the variables with __refdata.
kexec: make extended crashkernel= syntax less confusing
The extended crashkernel syntax is a little confusing in the way it handles
ranges. eg:
crashkernel=512M-2G:64M,2G-:128M
Means if the machine has between 512M and 2G of memory the crash region should
be 64M, and if the machine has 2G of memory the region should be 64M. Only if
the machine has more than 2G memory will 128M be allocated.
Although that semantic is correct, it is somewhat baffling. Instead I propose
that the end of the range means the first address past the end of the range,
ie: 512M up to but not including 2G.
Ingo Molnar [Thu, 1 May 2008 11:34:47 +0000 (04:34 -0700)]
isdn: hysdn_procconf.c build fix
x86.git randconfig testing found the following build error in latest
-git:
CC [M] drivers/isdn/hysdn/hysdn_procconf.o
CC [M] drivers/isdn/hysdn/hysdn_init.o
drivers/isdn/hysdn/hysdn_procconf.c: In function 'hysdn_procconf_init':
drivers/isdn/hysdn/hysdn_procconf.c:408: error: too few arguments to function 'proc_create'
Miklos Szeredi [Thu, 1 May 2008 11:34:45 +0000 (04:34 -0700)]
vfs: fix permission checking in sys_utimensat
If utimensat() is called with both times set to UTIME_NOW or one of them to
UTIME_NOW and the other to UTIME_OMIT, then it will update the file time
without any permission checking.
I don't think this can be used for anything other than a local DoS, but could
be quite bewildering at that (e.g. "Why was that large source tree rebuilt
when I didn't modify anything???")
This affects all kernels from 2.6.22, when the utimensat() syscall was
introduced.
Fix by doing the same permission checking as for the "times == NULL" case.
Thanks to Michael Kerrisk, whose utimensat-non-conformances-and-fixes.patch in
-mm also fixes this (and breaks other stuff), only he didn't realize the
security implications of this bug.
Roman Zippel [Thu, 1 May 2008 11:34:41 +0000 (04:34 -0700)]
ntp: handle leap second via timer
Remove the leap second handling from second_overflow(), which doesn't have to
check for it every second anymore. With CONFIG_NO_HZ this also makes sure the
leap second is handled close to the full second. Additionally this makes it
possible to abort a leap second properly by resetting the STA_INS/STA_DEL
status bits.
Roman Zippel [Thu, 1 May 2008 11:34:39 +0000 (04:34 -0700)]
ntp: remove current_tick_length()
current_tick_length used to do a little more, but now it just returns
tick_length, which we can also access directly at the few places, where it's
needed.
Roman Zippel [Thu, 1 May 2008 11:34:37 +0000 (04:34 -0700)]
ntp: support for TAI
This adds support for setting the TAI value (International Atomic Time). The
value is reported back to userspace via timex (as we don't have a
ntp_gettime() syscall).
Roman Zippel [Thu, 1 May 2008 11:34:34 +0000 (04:34 -0700)]
ntp: increase time_freq resolution
This changes time_freq to a 64bit value and makes it static (the only outside
user had no real need to modify it). Intermediate values were already 64bit,
so the change isn't that big, but it saves a little in shifts by replacing
SHIFT_NSEC with TICK_LENGTH_SHIFT. PPM_SCALE is then used to convert between
user space and kernel space representation.
Roman Zippel [Thu, 1 May 2008 11:34:33 +0000 (04:34 -0700)]
ntp: NTP4 user space bits update
This adds a few more things from the ntp nanokernel related to user space.
It's now possible to select the resolution used of some values via STA_NANO
and the kernel reports in which mode it works (pll/fll).
If some values for adjtimex() are outside the acceptable range, they are now
simply normalized instead of letting the syscall fail. I removed
MOD_CLKA/MOD_CLKB as the mapping didn't really makes any sense, the kernel
doesn't support setting the clock.
Roman Zippel [Thu, 1 May 2008 11:34:32 +0000 (04:34 -0700)]
ntp: cleanup ntp.c
This is mostly a style cleanup of ntp.c and extracts part of do_adjtimex as
ntp_update_offset(). Otherwise the functionality is still the same as before.
Roman Zippel [Thu, 1 May 2008 11:34:31 +0000 (04:34 -0700)]
remove div_long_long_rem
x86 is the only arch right now, which provides an optimized for
div_long_long_rem and it has the downside that one has to be very careful that
the divide doesn't overflow.
The API is a little akward, as the arguments for the unsigned divide are
signed. The signed version also doesn't handle a negative divisor and
produces worse code on 64bit archs.
There is little incentive to keep this API alive, so this converts the few
users to the new API.
Roman Zippel [Thu, 1 May 2008 11:34:28 +0000 (04:34 -0700)]
rename div64_64 to div64_u64
Rename div64_64 to div64_u64 to make it consistent with the other divide
functions, so it clearly includes the type of the divide. Move its definition
to math64.h as currently no architecture overrides the generic implementation.
They can still override it of course, but the duplicated declarations are
avoided.
projects / linux.git / log