[PATCH] aio: kiocb locking to serialise retry and cancel
Implement a per-kiocb lock to serialise retry operations and cancel. This
is done using wait_on_bit_lock() on the KIF_LOCKED bit of kiocb->ki_flags.
Also, make the cancellation path lock the kiocb and subsequently release
all references to it if the cancel was successful. This version includes a
fix for the deadlock with __aio_run_iocbs.
Signed-off-by: Benjamin LaHaise <bcrl@linux.intel.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] change io_cancel return code for no cancel case
Note that other than few exceptions, most of the current filesystem and/or
drivers do not have aio cancel specifically defined (kiob->ki_cancel field
is mostly NULL). However, sys_io_cancel system call universally sets
return code to -EAGAIN. This gives applications a wrong impression that
this call is implemented but just never works. We have customer inquires
about this issue.
Changed by Benjamin LaHaise to EINVAL instead of ENOSYS
Signed-off-by: S. Wendy Cheng <wcheng@redhat.com> Acked-by: Benjamin LaHaise <bcrl@kvack.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] Prefetch kernel stacks to speed up context switch
For architecture like ia64, the switch stack structure is fairly large
(currently 528 bytes). For context switch intensive application, we found
that significant amount of cache misses occurs in switch_to() function.
The following patch adds a hook in the schedule() function to prefetch
switch stack structure as soon as 'next' task is determined. This allows
maximum overlap in prefetch cache lines for that structure.
Signed-off-by: Ken Chen <kenneth.w.chen@intel.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: "Luck, Tony" <tony.luck@intel.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Jason Baron [Fri, 9 Sep 2005 20:02:01 +0000 (13:02 -0700)]
[PATCH] fix disassociate_ctty vs. fork race
Race is as follows. Process A forks process B, both being part of the same
session. Then, A calls disassociate_ctty while B forks C:
A B
==== ====
fork()
copy_signal()
dissasociate_ctty() ....
attach_pid(p, PIDTYPE_SID, p->signal->session);
Now, C can have current->signal->tty pointing to a freed tty structure, as
it hasn't yet been added to the session group (to have its controlling tty
cleared on the diassociate_ctty() call).
This has shown up as an oops but could be even more serious. I haven't
tried to create a test case, but a customer has verified that the patch
below resolves the issue, which was occuring quite frequently. I'll try
and post the test case if i can.
The patch simply checks for a NULL tty *after* it has been attached to the
proper session group and clears it as necessary. Alternatively, we could
simply do the tty assignment after the the process is added to the proper
session group.
Signed-off-by: Jason Baron <jbaron@redhat.com> Cc: Roland McGrath <roland@redhat.com> Cc: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch for old autofs (version 3) cleans dentries which are not putted
after killing the automount daemon (it's analogue of recent patch for
autofs4).
Signed-off-by: Alexander Krizhanovsky <klx@yandex.ru> Cc: Ian Kent <raven@themaw.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Samuel Thibault [Fri, 9 Sep 2005 20:01:58 +0000 (13:01 -0700)]
[PATCH] vga text console and stty cols/rows
Some people use 66-cells braille devices for reading the console, and hence
would like to reduce the width of the screen by using:
stty cols 66
However, the vga text console doesn't behave correctly: the 14 first
characters of the second line are put on the right of the first line and so
forth.
Here is a patch to correct that. It corrects the disp_end and offset
registers of the vga board on console resize and console switch.
On usual screens, you then correctly get a right and/or bottom blank
margin. On some laptop panels, the output is resized so that text actually
gets magnified, which can be great for some people (see
http://dept-info.labri.fr/~thibault/ls.jpg ).
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Jason Baron [Fri, 9 Sep 2005 20:01:57 +0000 (13:01 -0700)]
[PATCH] pty_chars_in_buffer oops fix
The idea of this patch is to lock both sides of a ptmx/pty pair during line
discipline changing. This is needed to ensure that say a poll on one side of
the pty doesn't occur while the line discipline is actively being changed.
This resulted in an oops reported on lkml, see:
A 'hacky' approach was previously implmemented which served to eliminate the
poll vs. line discipline changing race. However, this patch takes a more
general approach to the issue. The patch only adds locking on a less often
used path, the line-discipline changing path, as opposed to locking the
ptmx/pty pair on read/write/poll paths.
The patch below, takes both ldisc locks in either order b/c the locks are both
taken under the same spinlock(). I thought about locking the ptmx/pty
separately, such as master always first but that introduces a 3 way deadlock.
For example, process 1 does a blocking read on the slave side. Then, process
2 does an ldisc change on the slave side, which acquires the master ldisc lock
but not the slave's. Finally, process 3 does a write which blocks on the
process 2's ldisc reference.
This patch does introduce some changes in semantics. For example, a line
discipline change on side 'a' of a ptmx/pty pair, will now wait for a
read/write to complete on the other side, or side 'b'. The current behavior
is to simply wait for any read/writes on only side 'a', not both sides 'a' and
'b'. I think this behavior makes sense, but I wanted to point it out.
I've tested the patch with a bunch of read/write/poll while changing the line
discipline out from underneath.
This patch obviates the need for the above "hide the problem" patch.
Signed-off-by: Jason Baron <jbaron@redhat.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] alpha: process_reloc_for_got confuses r_offset and r_addend
arch/alpha/kernel/module.c:process_reloc_for_got(), which figures out how big
the .got section for a module should be, appears to be confusing r_offset (the
file offset that the relocation needs to be applied to) with r_addend (the
offset of the relocation's actual target address from the address of the
relocation's symbol). Because of this, one .got entry is allocated for each
relocation instead of one each unique symbol/addend.
In the module I am working with, this causes the .got section to be almost 10
times larger than it needs to be (75544 bytes instead of 7608 bytes). As the
.got is accessed with global-pointer-relative instructions, it needs to be
within the 64k gp "zone", and a 75544 byte .got clearly does not fit. The
result of this is that relocation overflows are detected during module load
and the load is aborted.
Change struct got_entry/process_reloc_for_got to fix this.
Acked-by: Richard Henderson <rth@twiddle.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] x86_64: Don't call enforce_max_cpus when hotplug is enabled
enforce_max_cpus nukes out cpu_present_map and cpu_possible_map making it
impossible to add new cpus in the system. Since it doesnt provide any
additional value apart this call and reference is removed.
[PATCH] x86_64: Don't do broadcast IPIs when hotplug is enabled in flat mode.
The use of non-shortcut version of routines breaking CPU hotplug. The option
to select this via cmdline also is deleted with the physflat patch, hence
directly placing this code under CONFIG_HOTPLUG_CPU.
We dont want to use broadcast mode IPI's when hotplug is enabled. This causes
bad effects in send IPI to a cpu that is offline which can trip when the cpu
is in the process of being kicked alive.
This is the same issue as ppc64 before, when returning to userland we
shouldn't re-compute the seccomp check or the task could be killed during
sigreturn when orig_eax is overwritten by the sigreturn syscall. This was
found by Roland.
This was harmless from a security standpoint, but some i686 users reported
failures with auditing enabled system wide (some distro surprisingly makes
it the default) and I reproduced it too by keeping the whole workload under
strace -f.
Patch is tested and works for me under strace -f.
nobody@athlon:~/cpushare> strace -o /tmp/o -f python seccomp_test.py
make: Nothing to be done for `seccomp_test'.
Starting computing some malicious bytecode
init
load
start
stop
receive_data failure
kill
exit_code 0 signal 9
The malicious bytecode has been killed successfully by seccomp
Starting computing some safe bytecode
init
load
start
stop
174 counts
kill
exit_code 0 signal 0
The seccomp_test.py completed successfully, thank you for testing.
(akpm: collaterally cleaned up a bit of do_syscall_trace() too)
Signed-off-by: Andrea Arcangeli <andrea@cpushare.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Tom Rini [Fri, 9 Sep 2005 20:01:48 +0000 (13:01 -0700)]
[PATCH] ppc32: Kill PVR_440* defines
The following patch changes the usages of PVR_440* into strcmp's with the
cpu_name field, and removes the defines altogether. The Ebony portion was
briefly tested long ago. One benefit of moving from PVR-tests to string
tests in general is that not all CPUs can be on and be able to do this type
of comparison.
See http://patchwork.ozlabs.org/linuxppc/patch?id=1250 for the original
thread.
Signed-off-by: Tom Rini <trini@kernel.crashing.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Tom Rini [Fri, 9 Sep 2005 20:01:47 +0000 (13:01 -0700)]
[PATCH] ppc32: In the boot code, don't rely on BASE_BAUD directly
Modifies serial_init to get base baud rate from the rs_table entry instead
of BAUD_BASE. This patch eliminates duplication between the
SERIAL_PORT_DFNS macro and BAUD_BASE. Without the patch, if a port set the
baud rate in SERIAL_PORT_DFNS, but did not update BASE_BAUD, the BASE_BAUD
value would still be used.
Signed-off-by: Grant Likely <grant.likely@gdcanada.com> Signed-off-by: Tom Rini <trini@kernel.crashing.org> Cc: Russell King <rmk@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch removes the inode_post_create/mkdir/mknod/symlink LSM hooks as
they are obsoleted by the new inode_init_security hook that enables atomic
inode security labeling.
If anyone sees any reason to retain these hooks, please speak now. Also,
is anyone using the post_rename/link hooks; if not, those could also be
removed.
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch modifies tmpfs to call the inode_init_security LSM hook to set
up the incore inode security state for new inodes before the inode becomes
accessible via the dcache.
As there is no underlying storage of security xattrs in this case, it is
not necessary for the hook to return the (name, value, len) triple to the
tmpfs code, so this patch also modifies the SELinux hook function to
correctly handle the case where the (name, value, len) pointers are NULL.
The hook call is needed in tmpfs in order to support proper security
labeling of tmpfs inodes (e.g. for udev with tmpfs /dev in Fedora). With
this change in place, we should then be able to remove the
security_inode_post_create/mkdir/... hooks safely.
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov> Cc: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch modifies ext3 to call the inode_init_security LSM hook to obtain
the security attribute for a newly created inode and to set the resulting
attribute on the new inode as part of the same transaction. This parallels
the existing processing for setting ACLs on newly created inodes.
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch modifies ext2 to call the inode_init_security LSM hook to obtain
the security attribute for a newly created inode and to set the resulting
attribute on the new inode. This parallels the existing processing for
setting ACLs on newly created inodes.
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The following patch set enables atomic security labeling of newly created
inodes by altering the fs code to invoke a new LSM hook to obtain the security
attribute to apply to a newly created inode and to set up the incore inode
security state during the inode creation transaction. This parallels the
existing processing for setting ACLs on newly created inodes. Otherwise, it
is possible for new inodes to be accessed by another thread via the dcache
prior to complete security setup (presently handled by the
post_create/mkdir/... LSM hooks in the VFS) and a newly created inode may be
left unlabeled on the disk in the event of a crash. SELinux presently works
around the issue by ensuring that the incore inode security label is
initialized to a special SID that is inaccessible to unprivileged processes
(in accordance with policy), thereby preventing inappropriate access but
potentially causing false denials on legitimate accesses. A simple test
program demonstrates such false denials on SELinux, and the patch solves the
problem. Similar such false denials have been encountered in real
applications.
This patch defines a new inode_init_security LSM hook to obtain the security
attribute to apply to a newly created inode and to set up the incore inode
security state for it, and adds a corresponding hook function implementation
to SELinux.
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Mark Fasheh [Fri, 9 Sep 2005 20:01:31 +0000 (13:01 -0700)]
[PATCH] update filesystems for new delete_inode behavior
Update the file systems in fs/ implementing a delete_inode() callback to
call truncate_inode_pages(). One implementation note: In developing this
patch I put the calls to truncate_inode_pages() at the very top of those
filesystems delete_inode() callbacks in order to retain the previous
behavior. I'm guessing that some of those could probably be optimized.
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com> Acked-by: Christoph Hellwig <hch@infradead.org> Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Mark Fasheh [Fri, 9 Sep 2005 20:01:29 +0000 (13:01 -0700)]
[PATCH] move truncate_inode_pages() into ->delete_inode()
Allow file systems supporting ->delete_inode() to call
truncate_inode_pages() on their own. OCFS2 wants this so it can query the
cluster before making a final decision on whether to wipe an inode from
disk or not. In some corner cases an inode marked on the local node via
voting may not actually get orphaned. A good example is node death before
the transaction moving the inode to the orphan dir commits to the journal.
Without this patch, the truncate_inode_pages() call in
generic_delete_inode() would discard valid data for such inodes.
During earlier discussion in the 2.6.13 merge plan thread, Christoph
Hellwig indicated that other file systems might also find this useful.
IMHO, the best solution would be to just allow ->drop_inode() to do the
cluster query but it seems that would require a substantial reworking of
that section of the code. Assuming it is safe to call write_inode_now() in
ocfs2_delete_inode() for those inodes which won't actually get wiped, this
solution should get us by for now.
Trivial testing of this patch (and a related OCFS2 update) has shown this
to avoid the corruption I'm seeing.
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com> Acked-by: Christoph Hellwig <hch@infradead.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andrew Morton [Fri, 9 Sep 2005 20:01:21 +0000 (13:01 -0700)]
[PATCH] x86: MP_processor_info fix
Remove the weird and apparently unnecessary logic in MP_processor_info() which
assumes that the BSP is the first one to run MP_processor_info(). On one of
my boxes that isn't true and cpu_possible_map gets the wrong value.
Cc: Zwane Mwaikambo <zwane@arm.linux.org.uk> Cc: Alexander Nyberg <alexn@telia.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This makes ACPI_BLACKLIST_YEAR be consistently defined when ACPI is
enabled, regardless of whether we're on x86 or not, and thus avoids
bogus -Wundef warnings on ia64.
Removed gratuitous includes of asm/serial.h in synklinkmp and ip2main.
Allows to remove the rest of "broken on sparc32" in drivers/char - this
stuff doesn't break the build anymore. Since it got zero testing, it almost
certainly won't work there, though...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
<qualifier> void * is not the same as void <qualifier> *... Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Paul Mackerras [Fri, 9 Sep 2005 13:02:36 +0000 (23:02 +1000)]
Allow PCI config space syscalls to be used by 64-bit processes.
The pciconfig_iobase, pciconfig_read and pciconfig_write system calls
were only implemented for 32-bit processes; for 64-bit processes they
returned an ENOSYS error. This allows them to be used by 64-bit
processes as well. The X server uses pciconfig_iobase at least, and
this change is necessary to allow a 64-bit X server to work on my G5.
Paul Mackerras [Tue, 6 Sep 2005 03:17:54 +0000 (13:17 +1000)]
[PATCH] Separate pci bits out of struct device_node
This patch pulls the PCI-related junk out of struct device_node and
puts it in a separate structure, struct pci_dn. The device_node now
just has a void * pointer in it, which points to a struct pci_dn for
nodes that represent PCI devices. It could potentially be used in
future for device-specific data for other sorts of devices, such as
virtual I/O devices.
Mark Bellon [Tue, 6 Sep 2005 22:50:02 +0000 (15:50 -0700)]
[PATCH] PPC64: large INITRD causes kernel not to boot
In PPC64 there are number of problems in arch/ppc64/boot/main.c that
prevent a kernel from making use of a large (greater than ~16MB) INITRD.
This is 64 bit architecture and really large INITRD images should be
possible.
Simply put the existing code has a fixed reservation (claim) address and
once the kernel plus initrd image are large enough to pass this address
all sorts of bad things occur. The fix is the dynamically establish the
first claim address above the loaded kernel plus initrd (plus some
"padding" and rounding). If PROG_START is defined this will be used as
the minimum safe address - currently known to be 0x01400000 for the
firmwares tested so far.
Signed-off-by: Mark Bellon <mbellon@mvista.com> Signed-off-by: Paul Mackerras <paulus@samba.org>
Merge a few asm-ppc and asm-ppc64 header files.
Note: the merge of setup.h intentionally does not carry
forward the m68k cruft. That means this patch continues
to break the already broken amiga on the ppc32.
Signed-off-by: Jon Loeliger <jdl@freescale.com> Signed-off-by: Kumar Gala <kumar.gala@freescale.com> Signed-off-by: Paul Mackerras <paulus@samba.org>
The earlier commit 8d9273918635f0301368c01b56c03a6f339e8d51
(Consolidate early console and PPCDBG code) broke iSeries because
it caused unregister_console(&udbg_console) to be called
unconditionally. iSeries never registers the udbg_console.
This just reverts part of the change.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Paul Mackerras <paulus@samba.org>
Arnd Bergmann [Thu, 18 Aug 2005 17:35:21 +0000 (19:35 +0200)]
[PATCH] ppc64: fix IPI on bpa_iic
This fixes a severe bug in the bpa_iic driver that caused
all sorts of problems.
We had been using incorrect priority values for inter processor
interrupts, which resulted in always doing CALL_FUNCTION
instead of RESCHEDULE or DEBUGGER_BREAK.
The symptoms cured by this patch include bad performance on
SMP systems spurious kernel panics in the IPI code.
Signed-off-by: Arnd Bergmann <arndb@de.ibm.com> Signed-off-by: Paul Mackerras <paulus@samba.org>
[PATCH] m68knommu: extract common timer code for 68EZ328 processor
Rework the 68x328 configuration and setup code. All 68x328 varients
share the same timer hardware. So extract that into its own file,
instead of keeping copies in each processors setup code.
[PATCH] m68knommu: register map setup for MOD5272 board
The boot loader on the MOD5272 board doesn't set the register maping,
so set it in the 5272 init code. There was code in there to support
this, but we had never needed to use it before.
[PATCH] m68knommu: extract common timer code for 68328 processor
Rework the 68x328 configuration and setup code. All 68x328 varient
share the same timer hardware, so extract that into its own file,
instead of keeping copies in each processors setup code.
Henk [Tue, 16 Aug 2005 14:17:43 +0000 (16:17 +0200)]
[PATCH] input-driver-yealink-P1K-usb-phone
This patch aggregates all modifications in the -mm tree and adds
complete ringtone support.
The following features are supported:
- keyboard full support
- LCD full support
- LED full support
- dialtone full support
- ringtone full support
- audio playback via generic usb audio diver
- audio record via generic usb audio diver
For driver documentation see: Documentation/input/yealink.txt
For vendor documentation see: http://yealink.com
Nick Sillik [Wed, 17 Aug 2005 17:37:34 +0000 (13:37 -0400)]
[PATCH] USB Storage: code cleanups for onetouch.c
As sugested by Alan Stern here are a few code cleanups for onetouch.c:
-Check number of endpoints before directly referencing intf->endpoint[2]
-Use defined constants instead of magic numbers
-Revmove the non-ascii characters from copyright notice
-Make registration and deregistration messages more similar
Signed-off-by: Nick Sillik <n.sillik@temple.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Pete Zaitcev [Mon, 15 Aug 2005 23:53:57 +0000 (16:53 -0700)]
[PATCH] usbmon in 2.6.13: peeking into DMA areas
This code looks at urb->transfer_dma, maps the page and takes the data.
I am looking for volunteers to contribute architectures other than i386
or to develop an architecure-neutral API for it (or point me that it
was done already).
Signed-off-by: Pete Zaitcev <zaitcev@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This adds the field tt_usecs to ehci_qh and ehci_iso_stream, and sets it
appropriately when setting them up as periodic endpoints. It records
the transation translator's think_time (added in last patch) plus the
downstream (i.e. low or full speed) bustime of the transfer associated
with each interrupt or iso frame, as calculated by usb_calc_bus_time.
Signed-off-by: Dan Streetman <ddstreet@ieee.org> Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David Brownell [Wed, 31 Aug 2005 16:54:50 +0000 (09:54 -0700)]
[PATCH] USB: usbnet (9/9) module for pl2301/2302 cables
This wraps up the conversion of the "usbnet" driver structure, by
moving the Prolific PL-2201/2302 minidriver to a module of its own.
It also includes some minor cleanups to the remaining "usbnet" file,
notably removing that long changelog at the top.
Minor historical note: Linux 2.2 first called the driver for
this hardware "plusb".
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David Brownell [Wed, 31 Aug 2005 16:54:36 +0000 (09:54 -0700)]
[PATCH] USB: usbnet (8/9) module for RNDIS devices
This adds host-side RNDIS support to the "usbnet" driver, so Linux can talk
to various devices (often based on WinCE) that otherwise only Windows could
talk to.
Tested with little-endian Linux talking to a Linux-USB Ethernet/RNDIS based
peripheral. This also includes updates from Eddie C. Dost <ecd@brainaid.de>
for big-endian SPARC Linux talking to a Nokia 9500 Communicator.
It's still marked as EXPERIMENTAL because this code is so young. This
ought to let Linux to work with various cable modems that previously
would have been "Windows Only".
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David Brownell [Wed, 31 Aug 2005 16:54:20 +0000 (09:54 -0700)]
[PATCH] USB: usbnet (7/9) module for CDC Ethernet
Makes the CDC Ethernet support live in a separate driver module.
This module is a bit special since it exports utility functions
that are reused by the the Zaurus and RNDIS drivers, but it's
not "core" like usbnet itself.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David Brownell [Wed, 31 Aug 2005 16:53:58 +0000 (09:53 -0700)]
[PATCH] USB: usbnet (6/9) module for Zaurii and compatibles
This moves usbnet support for Zaurus and compatibles into its own module.
Other than exporting a couple of helper functions, this just involved
shuffling some code and updating the comments.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David Brownell [Wed, 31 Aug 2005 16:53:42 +0000 (09:53 -0700)]
[PATCH] USB: usbnet (5/9) module for genesys gl620a cables
This moves the GeneSys GL620USB-A support into its own driver file.
It also fixes a "return wrong skb" glitch in the rx unbatching, as
recently reported, and adds some missing byteswaps in the special
"genelink" headers (so it might now work on big-endian Linux).
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David Brownell [Wed, 31 Aug 2005 16:53:26 +0000 (09:53 -0700)]
[PATCH] USB: usbnet (4/9) module for net1080 cables
As with the "cdc_subset" and "asix" drivers, this just moves the net1080
support into its one driver module. In this case there's a small bit of
extra cleanup involved, moving some funky framing logic into the tx_fixup()
routine (resolving a long overdue FIXME).
Minor historical note: "usbnet" started out as "net1080", then got
generalized to make it easier for other network drivers to reuse the
urb queueing and fault management code here.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David Brownell [Wed, 31 Aug 2005 16:53:10 +0000 (09:53 -0700)]
[PATCH] USB: usbnet (3/9) module for ASIX Ethernet adapters
This patch moves the ASIX AX8817x driver into its own file, just using
the "usbnet" infrastructure as a utility library.
- As with "cdc_subset" this involved minor Kconfig/kbuild tweaks,
moving code from one file to another, and exporting a few functions.
- This includes updates from Jamie Painter to add (and use) a new hook
to handle the different maximum transfer sizes for rx and tx sides.
- Also from Jamie, some bugfixes:
* MDIO byteorder (to address some PPC media negotiation problems);
* Force alignment at key spots when using ax88772 framing (on some
embedded hardware, the network stack will break otherwise);
* Address some link reset problems.
It also makes this driver use the standard (5 seconds vs half second)
control timeouts used elsewhere in USB; and wraps a few lines before
the 80th column (which previously needed it).
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David Brownell [Wed, 31 Aug 2005 16:52:45 +0000 (09:52 -0700)]
[PATCH] USB: usbnet (2/9) module for simple network links
This patch creates the first of several separate "minidriver" modules
for "usbnet". This one handles only the very simplest hardware, which
can be handled almost entirely by the "usbnet" core.
- Move device-specific bits into new "cdc_subset.c" driver,
shrinking "usbnet" by a bunch;
- Export the functions needed to support this minidriver
(with EXPORT_SYMBOL_GPL);
- Update Kconfig and kbuild accordingly.
This one handles about a dozen different device types, with the most
notable ones being Gumstix and most Linux-based PDAs (except Zaurus
running that ancient code from Sharp).
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David Brownell [Wed, 31 Aug 2005 16:52:31 +0000 (09:52 -0700)]
[PATCH] USB: usbnet (1/9) clean up framing
This starts to prepare the core of "usbnet" to know less about various
framing protocols that map Ethernet packets onto USB, so "minidrivers"
can be modules that just plug into the core.
- Remove some framing-specific code that cluttered the core:
* net->hard_header_len records how much space to preallocate;
now drivers that add their own framing (Net1080, GeneLink,
Zaurus, and RNDIS) will have smoother TX paths. Even for
the drivers (Zaurus, Net1080) that need trailers.
* defines new dev->hard_mtu, using this "hardware" limit to
check changes to the link's settable "software" mtu.
* now net->hard_header_len and dev->hard_mtu are set up in the
driver bind() routines, if needed.
- Transaction ID is no longer specific to the Net1080 framing;
RNDIS needs one too.
- Creates a new "usbnet.h" header with declarations that are shared
between the core and what will be separate modules.
- Plus a couple other minor tweaks, like recognizing -ESHUTDOWN
means the keventd work should just shut itself down asap.
The core code is only about 1/3 of this large file. Splitting out the
minidrivers into separate modules (e.g. ones for ASIX adapters,
Zaurii and similar, CDC Ethernet, etc), in later patches, will
improve maintainability and shrink typical runtime footprints.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Alexey Dobriyan [Sun, 14 Aug 2005 13:24:26 +0000 (17:24 +0400)]
[PATCH] USB ldusb: fmt warnings fixes for 64-bit platforms
Fix
drivers/usb/misc/ldusb.c: In function `ld_usb_read':
drivers/usb/misc/ldusb.c:467: warning: int format, different type arg (arg 4)
drivers/usb/misc/ldusb.c: In function `ld_usb_write':
drivers/usb/misc/ldusb.c:531: warning: int format, different type arg (arg 4)
drivers/usb/misc/ldusb.c:532: warning: int format, different type arg (arg 5)
drivers/usb/misc/ldusb.c:532: warning: int format, different type arg (arg 6)
Dale Farnsworth [Thu, 11 Aug 2005 00:25:25 +0000 (17:25 -0700)]
[PATCH] USB: remove include of asm/usb.h in ohci-ppc-soc.c
ohci-ppc-soc.c provides for a platform-specific callback mechanism for
when the HC is successfully probed or removed. It turned out that none
of the 3 platforms using it need this facility. Also the required
include/asm-ppc/usb.h has never been accepted. This patch removes the
callback feature and the include of <asm/usb.h>.
Alan Stern [Thu, 11 Aug 2005 19:50:32 +0000 (15:50 -0400)]
[PATCH] USB: Support unbinding of the usb_generic driver
This patch (as556) adds support for unbinding the usb_generic "driver".
That driver only binds to USB devices, as opposed to interfaces, and it
does nothing much besides marking which struct device's go with an
overall USB device plus providing suspend/resume methods. Now that
users can unbind drivers at will using the sysfs "unbind" attribute, we
need a rational way of dealing with USB devices that are no longer under
full control of the USB stack. The patch handles this by unconfiguring
the device, thereby removing all the interfaces and their associated
drivers and children.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
projects / linux-2.6 / log