- Mar 23, 2006
-
-
Linus Torvalds authored
* master.kernel.org:/home/rmk/linux-2.6-arm: (45 commits) [ARM] 3389/1: typo and grammar fix [ARM] 3386/1: AT91RM9200 Clock update [ARM] 3384/1: AT91RM9200: Timer [ARM] 3382/1: ixp2000: unify defconfigs [ARM] 3381/1: ixp2000: fix slowport write timing control register fields [ARM] 3380/1: ixp2000: simplify ixdp2x00_master_npu() check [ARM] 3379/1: ixp2000: use generic 8250 debug macros [ARM] 3378/1: ixp2000: fix gpio interrupt handling [ARM] Quieten spurious IRQ detection [ARM] Use kcalloc to allocate counter_config array rather than kmalloc [ARM] Oprofile: dynamically allocate counter_config [ARM] Oprofile: Convert semaphore to mutex [ARM] 3376/2: S3C2410 - update defconfig [ARM] 3375/1: S3C2440 - fix osiris machine build [ARM] 3374/1: ep93xx: gpio interrupt support [ARM] 3361/1: S3C24XX - add USB bus clock source [ARM] 3360/1: S3C2440 - add set rate methods and camera clock [ARM] 3359/1: S3C24XX - add support for clk_set_rate [ARM] Convert kmalloc+memset to kzalloc [ARM] 3373/1: move uengine loader to arch/arm/common ...
-
Tejun Heo authored
Now that libata is smart enought to handle both soft and hard resets, add softreset method. Signed-off-by:
Tejun Heo <htejun@gmail.com> Signed-off-by:
Jeff Garzik <jeff@garzik.org>
-
Tejun Heo authored
As libata now can do PIO, don't ignore PIO-only devices. Signed-off-by:
-
Alan Cox authored
Signed-off-by:
Alan Cox <alan@redhat.com> Signed-off-by:
-
Alan Cox authored
Not much to say here except that some drives have fixed and bad firmware Signed-off-by:
-
Alan Cox authored
The current code follows the spec but uses an overlong delay. This would be great if the hardware did. Several vendors however forget the D7 pulldown. Fortunately 0xFF isnt a sane reset state so we can use it to skip detection as is done in drivers/ide. (ie this is a tested solution over a long time) Signed-off-by:
-
Alan Cox authored
Signed-off-by:
-
Jeff Garzik authored
-
Erik Hovland authored
Patch from Erik Hovland I found a typo and what seems to be a run-on sentence in arch/arm/common/dmabounce.c This patch corrects both. Signed-off-by:
Erik Hovland <erik@hovland.org> Signed-off-by:
Russell King <rmk+kernel@arm.linux.org.uk>
-
Andrew Victor authored
Patch from Andrew Victor This patch includes a few changes to the clock support on the AT91RM9200. 1. Added definitions for Ethernet, MMC, TWI, USARTs, and SPI peripheral clocks. 2. Replaced some hard-coded hex values with the text definitions in at91rm9200_sys.h. 3. If the USB96M bit is set for PLLB, then the rate of PLLB is not affected but only the USB Host/Device clocks which are derived from it. Issue reported by Sergei Sharonov. Signed-off-by:
Andrew Victor <andrew@sanpeople.com> Signed-off-by:
-
Andrew Victor authored
Patch from Andrew Victor If the timer interrupt is ever significantly delayed (or after the system was suspended), the system could spin incrementing the time for too long. The fix is to replace the "do {} while" with a "while {}". Orignal patch by Savin Zlobec and Peter Menzebach. Signed-off-by: -
Lennert Buytenhek authored
Patch from Lennert Buytenhek Unify the five existing ixp2000 defconfigs into one defconfig. Signed-off-by:
Lennert Buytenhek <buytenh@wantstofly.org> Signed-off-by:
-
Lennert Buytenhek authored
Patch from Lennert Buytenhek The original version of the chip docs had the PW and SU fields in the slowport write timing control register accidentally reversed. This is mentioned in the errata (documentation change #4) and fixed in newer docs. Signed-off-by:
-
Lennert Buytenhek authored
Patch from Lennert Buytenhek On the IXDP2x00s, the NPU that is PCI master is always the egress (i.e. 'master') NPU. At least on the IXDP2800, both NPUs have flash, so the ixp2000_has_flash() check in ixdp2x00_master_npu() is useless. Signed-off-by:
-
Lennert Buytenhek authored
Patch from Lennert Buytenhek The xscale UART in the ixp2000 is basically just an 8250 UART (with some extra bits and pieces), so we can use the generic 8250 debug macros on the ixp2000. Signed-off-by:
-
Lennert Buytenhek authored
Patch from Lennert Buytenhek ixp2000 used to initially mark GPIO interrupts as invalid, and not mark them valid until set_irq_type() was called, but this doesn't work if you want to use request_irq() with the SA_TRIGGER_* flags. So, just mark the GPIO interrupts valid from the beginning. We configure GPIOs as inputs when set_irq_type() is called anyway, so this shouldn't be a problem. Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/perex/alsaLinus Torvalds authored
* git://git.kernel.org/pub/scm/linux/kernel/git/perex/alsa: (124 commits) [ALSA] version 1.0.11rc4 [PATCH] Intruduce DMA_28BIT_MASK [ALSA] hda-codec - Add support for ASUS P4GPL-X [ALSA] hda-codec - Add support for HP nx9420 laptop [ALSA] Fix memory leaks in error path of control.c [ALSA] AMD Au1x00: AC'97 controller is memory mapped [ALSA] AMD Au1x00: fix DMA init/cleanup [ALSA] hda-codec - Fix generic auto-configurator [ALSA] hda-codec - Fix BIOS auto-configuration [ALSA] Fixes typos in Audiophile-USB.txt [ALSA] ice1712 - typo fixes for dxr_enable module option [ALSA] AMD Au1x00: make driver build after cleanup [ALSA] ice1712 - Fix wrong value types for enum items [ALSA] fix resource leak in usbmixer [ALSA] Fix gus_pcm dereference before NULL [ALSA] Fix seq_clientmgr dereferences before NULL check [ALSA] hda-codec - Fix for Samsung R65 and ASUS A6J [ALSA] hda-codec - Add support for VAIO FE550G and SZ110 [ALSA] usb-audio: add Maya44 mixer control names [ALSA] usb-audio: add Casio PL-40R support ...
-
git://git.kernel.org/pub/scm/linux/kernel/git/bunk/trivialLinus Torvalds authored
* git://git.kernel.org/pub/scm/linux/kernel/git/bunk/trivial: fixed path to moved file in include/linux/device.h Fix spelling in E1000_DISABLE_PACKET_SPLIT Kconfig description Documentation/dvb/get_dvb_firmware: fix firmware URL Documentation: Update to BUG-HUNTING Remove superfluous NOTIFY_COOKIE_LEN define add "tags" to .gitignore Fix "frist", "fisrt", typos fix rwlock usage example It's UTF-8
-
Linus Torvalds authored
* master.kernel.org:/pub/scm/linux/kernel/git/davem/sparc-2.6: [SPARC64]: Add a secondary TSB for hugepage mappings. [SPARC]: Respect vm_page_prot in io_remap_page_range().
-
Linus Torvalds authored
* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6: [TG3]: Bump driver version and reldate. [TG3]: Skip phy power down on some devices [TG3]: Fix SRAM access during tg3_init_one() [X25]: dte facilities 32 64 ioctl conversion [X25]: allow ITU-T DTE facilities for x25 [X25]: fix kernel error message 64 bit kernel [X25]: ioctl conversion 32 bit user to 64 bit kernel [NET]: socket timestamp 32 bit handler for 64 bit kernel [NET]: allow 32 bit socket ioctl in 64 bit kernel [BLUETOOTH]: Return negative error constant
-
Linus Torvalds authored
* master.kernel.org:/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6: (138 commits) [SCSI] libata: implement minimal transport template for ->eh_timed_out [SCSI] eliminate rphy allocation in favour of expander/end device allocation [SCSI] convert mptsas over to end_device/expander allocations [SCSI] allow displaying and setting of cache type via sysfs [SCSI] add scsi_mode_select to scsi_lib.c [SCSI] 3ware 9000 add big endian support [SCSI] qla2xxx: update MAINTAINERS [SCSI] scsi: move target_destroy call [SCSI] fusion - bump version [SCSI] fusion - expander hotplug suport in mptsas module [SCSI] fusion - exposing raid components in mptsas [SCSI] fusion - memory leak, and initializing fields [SCSI] fusion - exclosure misspelled [SCSI] fusion - cleanup mptsas event handling functions [SCSI] fusion - removing target_id/bus_id from the VirtDevice structure [SCSI] fusion - static fix's [SCSI] fusion - move some debug firmware event debug msgs to verbose level [SCSI] fusion - loginfo header update [SCSI] add scsi_reprobe_device [SCSI] megaraid_sas: fix extended timeout handling ...
-
- Mar 22, 2006
-
-
James Morris authored
Add a slab cache for the SELinux inode security struct, one of which is allocated for every inode instantiated by the system. The memory savings are considerable. On 64-bit, instead of the size-128 cache, we have a slab object of 96 bytes, saving 32 bytes per object. After booting, I see about 4000 of these and then about 17,000 after a kernel compile. With this patch, we save around 530KB of kernel memory in the latter case. On 32-bit, the savings are about half of this. Signed-off-by:
James Morris <jmorris@namei.org> Acked-by:
Stephen Smalley <sds@tycho.nsa.gov> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
James Morris authored
Remove an unneded pointer variable in selinux_inode_init_security(). Signed-off-by:
-
James Morris authored
A further fix is needed for selinuxfs link count management, to ensure that the count is correct for the parent directory when a subdirectory is created. This is only required for the root directory currently, but the code has been updated for the general case. Signed-off-by:
-
James Morris authored
Fix copy & paste error in sel_make_avc_files(), removing a supurious call to d_genocide() in the error path. All of this will be cleaned up by kill_litter_super(). Signed-off-by:
-
James Morris authored
Remove the call to sel_make_bools() from sel_fill_super(), as policy needs to be loaded before the boolean files can be created. Policy will never be loaded during sel_fill_super() as selinuxfs is kernel mounted during init and the only means to load policy is via selinuxfs. Also, the call to d_genocide() on the error path of sel_make_bools() is incorrect and replaced with sel_remove_bools(). Signed-off-by:
-
James Morris authored
Unify the error path of sel_fill_super() so that all errors pass through the same point and generate an error message. Also, removes a spurious dput() in the error path which breaks the refcounting for the filesystem (litter_kill_super() will correctly clean things up itself on error). Signed-off-by:
-
James Morris authored
Use existing sel_make_dir() helper to create booleans directory rather than duplicating the logic. Signed-off-by:
-
James Morris authored
Fix the hard link count for selinuxfs directories, which are currently one short. Signed-off-by:
-
Stephen Smalley authored
Simplify sel_read_bool to use the simple_read_from_buffer helper, like the other selinuxfs functions. Signed-off-by:
-
Ingo Molnar authored
Semaphore to mutex conversion. The conversion was generated via scripts, and the result was validated automatically via a script as well. Signed-off-by:
Ingo Molnar <mingo@elte.hu> Cc: Stephen Smalley <sds@epoch.ncsc.mil> Cc: James Morris <jmorris@namei.org> Cc: David Howells <dhowells@redhat.com> Signed-off-by:
-
Stephen Smalley authored
This patch disables the automatic labeling of new inodes on disk when no policy is loaded. Discussion is here: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=180296 In short, we're changing the behavior so that when no policy is loaded, SELinux does not label files at all. Currently it does add an 'unlabeled' label in this case, which we've found causes problems later. SELinux always maintains a safe internal label if there is none, so with this patch, we just stick with that and wait until a policy is loaded before adding a persistent label on disk. The effect is simply that if you boot with SELinux enabled but no policy loaded and create a file in that state, SELinux won't try to set a security extended attribute on the new inode on the disk. This is the only sane behavior for SELinux in that state, as it cannot determine the right label to assign in the absence of a policy. That state usually doesn't occur, but the rawhide installer seemed to be misbehaving temporarily so it happened to show up on a test install. Signed-off-by:
-
Christoph Lameter authored
Centralize the page migration functions in anticipation of additional tinkering. Creates a new file mm/migrate.c 1. Extract buffer_migrate_page() from fs/buffer.c 2. Extract central migration code from vmscan.c 3. Extract some components from mempolicy.c 4. Export pageout() and remove_from_swap() from vmscan.c 5. Make it possible to configure NUMA systems without page migration and non-NUMA systems with page migration. I had to so some #ifdeffing in mempolicy.c that may need a cleanup. Signed-off-by:
Christoph Lameter <clameter@sgi.com> Signed-off-by:
-
Paul Jackson authored
The alien cache rotor in mm/slab.c assumes that the first online node is node 0. Eventually for some archs, especially with hotplug, this will no longer be true. Fix the interleave rotor to handle the general case of node numbering. Signed-off-by:
Paul Jackson <pj@sgi.com> Acked-by:
Christoph Lameter <clameter@engr.sgi.com> Signed-off-by:
-
Paul Jackson authored
Fix bogus node loop in hugetlb.c alloc_fresh_huge_page(), which was assuming that nodes are numbered contiguously from 0 to num_online_nodes(). Once the hotplug folks get this far, that will be false. Signed-off-by:
-
Akinobu Mita authored
When we've allocated SWAPFILE_CLUSTER pages, ->cluster_next should be the first index of swap cluster. But current code probably sets it wrong offset. Signed-off-by:
Akinobu Mita <mita@miraclelinux.com> Acked-by:
Hugh Dickins <hugh@veritas.com> Signed-off-by:
-
Christoph Lameter authored
1. Only disable interrupts if there is actually something to free 2. Only dirty the pcp cacheline if we actually freed something. 3. Disable interrupts for each single pcp and not for cleaning all the pcps in all zones of a node. drain_node_pages is called every 2 seconds from cache_reap. This fix should avoid most disabling of interrupts. Signed-off-by:
-
Christoph Lameter authored
The list_lock also protects the shared array and we call drain_array() with the shared array. Therefore we cannot go as far as I wanted to but have to take the lock in a way so that it also protects the array_cache in drain_pages. (Note: maybe we should make the array_cache locking more consistent? I.e. always take the array cache lock for shared arrays and disable interrupts for the per cpu arrays?) Signed-off-by:
-
Christoph Lameter authored
Remove drain_array_locked and use that opportunity to limit the time the l3 lock is taken further. Signed-off-by:
-
Christoph Lameter authored
And a parameter to drain_array to control the freeing of all objects and then use drain_array() to replace instances of drain_array_locked with drain_array. Doing so will avoid taking locks in those locations if the arrays are empty. Signed-off-by:
-
