summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* mtd: uclinux: do not allow to be built as a moduleMike Frysinger2009-06-051-1/+1
| | | | | | | | | | | | | | | There isn't any benefit to building the uClinux MTD map as a module as the rootfs it requires in order to actually be usable is appended to the kernel image, not the module. No known system builds it this way either, so change the option to "bool". Signed-off-by: Mike Frysinger <vapier@gentoo.org> CC: Paul Mundt <lethal@linux-sh.org> CC: Greg Ungerer <gerg@uclinux.org> CC: uclinux-dev@uclinux.org CC: linux-mtd@lists.infradead.org Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: uclinux: allow systems to override map addr/sizeMike Frysinger2009-06-051-4/+6
| | | | | | | | | | | | | | | | Due to a processor anomaly (05000263 to be exact), most Blackfin parts cannot keep the embedded filesystem image directly after the kernel in RAM. Instead, the filesystem needs to be relocated to the end of memory. As such, we need to tweak the map addr/size during boot for Blackfin systems. This can be done in any early arch/board init code. Signed-off-by: Mike Frysinger <vapier@gentoo.org> CC: Paul Mundt <lethal@linux-sh.org> CC: Greg Ungerer <gerg@uclinux.org> CC: uclinux-dev@uclinux.org CC: linux-mtd@lists.infradead.org Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: blackfin NFC: fix hang when using NAND on BF527-EZKITsCliff Cai2009-06-051-2/+11
| | | | | | | | | | | | | | | | The DMAs have different bit sizes on BF52x and BF54x. From the PHRM: "The 16-bit DMA Access Bus (DAB) connects the DMA controller to the on-chip peripherals, PPI, SPI, Ethernet MAC, the SPORTs, NFC, HOSTDP and the UARTs." 32-bit DMA won't work for BF52x. Signed-off-by: Cliff Cai <cliff.cai@analog.com> Signed-off-by: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Bryan Wu <cooloney@kernel.org> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: blackfin NFC: remove pointless return value in bf5xx_nand_dma_rwMike Frysinger2009-06-051-3/+1
| | | | | | | Signed-off-by: Mike Frysinger <vapier@gentoo.org> Signed-off-by: Bryan Wu <cooloney@kernel.org> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: add on-flash BBT support for Atmel NAND driverSimon Polette2009-06-051-1/+10
| | | | | | | | Just add a new on-flash-bbt module parameter. Signed-off-by: Simon Polette <spolette@adetelgroup.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: MXC NAND support for 2KiB page size flashesVladimir Barinov2009-06-051-9/+51
| | | | | | | | | | | | - Add support for 2KiB page size flashes - Fix page address access for large pages - Detect oob layout at runtime - handle pagesize_2k variable - Fix oob16 layout: reserve location 5 of oob area since it's used for bbt Signed-off-by: Vladimir Barinov <vova.barinov@gmail.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: plat_nand: allow platform to set partitionsH Hartley Sweeten2009-06-052-0/+5
| | | | | | | | | | | | | | | | | | | | | | | Add optional callback to allow platform to initialize partitions. Static partitions on a nand device could vary depending on the size of the device. This patch allows an optional platform callback to be used to setup this partition information at runtime. Scan order is: 1) chip.part_probe_types 2) chip.set_parts 3) chip.partitions 4) full mtd device (fallback for no partitions) Some of the existing nand drivers could possibly be replaced by the plat_nand driver by using this patch. These include autcpu12.c and ts7250.c drivers. Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: plat_nand: add platform probe/remove callbacksH Hartley Sweeten2009-06-052-2/+18
| | | | | | | | | | | | | | | Add optional probe and remove callbacks to the plat_nand driver. Some platforms may require additional setup, such as configuring the memory controller, before the nand device can be accessed. This patch provides an optional callback to handle this setup as well as a callback to teardown the setup. Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com> Tested-by: Alexander Clouter <alex@digriz.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: remove pmcmsp-ramroot.cShane McDonald2009-06-053-114/+0
| | | | | | | | | | | | | | | The RAMROOT function was a successful but non-portable attempt to append the root filesystem to the end of the kernel image. The preferred and portable solution is to use an initramfs instead. The only user of this function was the msp71xx configuration in the MIPS architecture; as the use of the RAMROOT has been removed from that configuration, there are no more users, so this code can be removed. Signed-off-by: Shane McDonald <mcdonald.shane@gmail.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: OneNAND: add missing __devexit_pMika Korhonen2009-06-051-2/+2
| | | | | | | | | Add missing __devexit_p wrapper and no more mark shutdown with __devexit. Fixes build in configurations where devexit functions get discarded. Signed-off-by: Mika Korhonen <mika.j.korhonen@gmail.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: m25p80: add support for Macronix MX25L12805DLennert Buytenhek2009-06-051-0/+3
| | | | | | Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: CFI 1.0 and CFI 1.1Daniel Ribeiro2009-06-051-4/+10
| | | | | | | | | | | | | | | This patch allows otpinfo for CFI >= 1.0 and burst read for CFI >= 1.1. references: 1.0: http://www.datasheetcatalog.org/datasheets2/81/816884_1.pdf 1.1: http://milkymist.org/doc/MT28F640J3.pdf http://www.delorie.com/agenda/specs/29066709.pdf Signed-off-by: Daniel Ribeiro <drwyrm@gmail.com> Acked-by: Nicolas Pitre <nico@cam.org> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: CFI: quirk for PF38F4476.Daniel Ribeiro2009-06-051-0/+12
| | | | | | | | | | | | This chip reports CFI 1.3, but the CFI PRI is like CFI 1.1. Add a quirk to pass probe on this chip. This patch depends on "MTD: CFI 1.0 and CFI 1.1" Signed-off-by: Daniel Ribeiro <drwyrm@gmail.com> Acked-by: Nicolas Pitre <nico@cam.org> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: extend plat_nand for (read|write)_bufAlexander Clouter2009-06-052-0/+8
| | | | | | | | | | | | | This patch adds (write|read)_buf callbacks to plat_nand. The NAND on the TS-7800 provisioned by the FPGA allows readw() and readl() to be used which gives a 2.5x speed up. To be able to use this from the plat_nand driver a hook for read_buf (and also write_buf whilst we are in there) need to be made available. This patch adds the hook. Signed-off-by: Alexander Clouter <alex@digriz.org.uk> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: add SST39SF040 chip to jedec_probeMichał Mirosław2009-06-051-0/+13
| | | | | | | | Add SST39SF040 chip (like SST39SF020A but bigger - 4Mbit). Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: davinci_nand: cmdlinepart uses MTD IDsDavid Brownell2009-06-051-11/+0
| | | | | | | | | | | | | Remove some legacy code from the davinci_nand driver, which made cmdlinepart ignore the the MTD ID passed to it. Boards can have multiple NAND chips, and some do (like the DM357 EVM), so this dated hack is undesirable. Correct labels are like "davinci_nand.0" (for chipselect 0). Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: OneNAND: add support for OneNAND manufactured by NumonyxAdrian Hunter2009-06-052-1/+3
| | | | | | | | | | | In addition to adding the Numonyx manufacturer code, this patch also ensures 'sync. write' is disabled when reading identification data - something that the Numonyx chip objects to, but the Samsung chip seems to ignore. Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: m25p80 nand: add m45pe10 idsMatthieu CASTET2009-06-051-0/+1
| | | | | | | | | | this patch add m45pe10 [1] chip support to the m25p80 driver. [1] http://www.numonyx.com/Documents/Datasheets/M45PE10.pdf Signed-off-by: Matthieu CASTET <matthieu.castet@parrot.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: mxc_nand: add correct dev_id parameter to free_irq() callsMagnus Lilja2009-06-051-2/+2
| | | | | | | | | | Make sure to pass the same dev_id data to free_irq() that was used when calling request_irq(), otherwise we get a warning about freeing an already free IRQ. Signed-off-by: Magnus Lilja <lilja.magnus@gmail.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: nand: fix 512 byte software ecc supportVimal Singh2009-06-051-2/+2
| | | | | | | | | Type of 'byte_addr' needes to be 'unsigned int' for 512 byte ECC support. Signed-off-by: Vimal Singh <vimalsingh@ti.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: nand: davinci_nand, 4-bit ECC for smallpageDavid Brownell2009-06-052-15/+297
| | | | | | | | | | | | | | | | Minimal support for the 4-bit ECC engine found on DM355, DM365, DA830/OMAP-L137, and similar recent DaVinci-family chips. This is limited to small-page flash for now; there are some page layout issues for large page chips. Note that most boards using this engine (like the DM355 EVM) include 2GiB large page chips. Sanity tested on DM355 EVM after swapping the socketed NAND for a small-page one. Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: nand: minor davinci_nand cleanupDavid Brownell2009-06-051-16/+11
| | | | | | | | | | | | | | | | Make the DaVinci NAND driver require platform_data with board-specific configuration. We can't actually do any kind of sane job of configuring it otherwise. Also fix the comment about picking the "best" ECC mode. We can't do those any more; that relied on knowing what kind of CPU we're using (they don't all support 4-bit ECC), and current policy is that drivers not have cpu_is_*() checks. Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: nand: don't walk past end of oobfree[]David Brownell2009-06-051-1/+2
| | | | | | | | | | | | Resolve issue noted by Sneha: when computing oobavail from the list of free areas in the OOB, don't assume there will always be an unused slot at the end. With ECC_HW_SYNDROME and 4KiB page chips, it's fairly likely there *won't* be one. Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Cc: "Narnakaje, Snehaprabha" <nsnehaprabha@ti.com>" Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* Kill jffs2-user.hArtem Bityutskiy2009-06-053-36/+0
| | | | | | | | | | | | | | | This file does not define any kernel-userspace API, all it does it defines few helpers for userspace. Instead, userspace should have a private copy of this file. The main (if not the only) user is the mtd-utils package, but it already has a private copy of this file. This patch also removes references to 'jffs2-user.h' from 'Kbuild' and MAINTAINERS' files. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: plat_nand: fix section errorThomas Chou2009-06-051-1/+1
| | | | | | | | | | | | | With CONFIG_HOTPLUG=n, the following eror occurred during link: local symbol 0: discarded in section `.devexit.text' from drivers/built-in.o It was caused by improper section reference. The __devexit_p() should be added to the .remove function. Signed-off-by: Thomas Chou <thomas@wytron.com.tw> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: Fix pointer handling in compat ioctls to use compat_ptr()David Woodhouse2009-05-291-2/+2
| | | | Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: Handle compat ioctls directly; remove all trace from compat_ioctl.cKevin Cernekee2009-05-292-23/+2
| | | | | | | | | Remove all references to MTD ioctls from fs/compat_ioctl.c and let them all be handled by mtd_compat_ioctl(). Signed-off-by: Kevin Cernekee <kpc.mtd@gmail.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: add OOB ioctls for >4GiB devicesKevin Cernekee2009-05-293-0/+39
| | | | | | Signed-off-by: Kevin Cernekee <kpc.mtd@gmail.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: compat_ioctl cleanupKevin Cernekee2009-05-292-126/+180
| | | | | | | | | | | | | | | 1) Move the MEMREADOOB/MEMWRITEOOB compat_ioctl wrappers from fs/compat_ioctl.c into mtdchar.c . Original request was here: http://lkml.org/lkml/2009/4/1/295 2) Add missing COMPATIBLE_IOCTL lines, so that mtd-utils does not error out when running in 64/32 compatibility mode. LKML-Reference: <200904011650.22928.arnd@arndb.de> Signed-off-by: Kevin Cernekee <kpc.mtd@gmail.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: add MEMERASE64 ioctl for >4GiB devicesKevin Cernekee2009-05-293-8/+28
| | | | | | | | | New MEMERASE/MEMREADOOB/MEMWRITEOOB ioctls are needed in order to support 64-bit offsets into large NAND flash devices. Signed-off-by: Kevin Cernekee <kpc.mtd@gmail.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* Documentation: add MTD sysfs docsKevin Cernekee2009-05-291-0/+125
| | | | | | Signed-off-by: Kevin Cernekee <kpc.mtd@gmail.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: Fix handling of mtdname in txx9ndfmc.cDavid Woodhouse2009-05-291-4/+12
| | | | | | | | | | | As pointed out by Kay Sievers, the name size limit is gone from the driver-core, and BUS_ID_SIZE is obsolescent. Rather than just papering over the problem by replacing the mtdname array size with an arbitrary '20 + 2', fix the problem properly and handle arbitrary name sizes. Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: Add armflash support for multiple blocks of flashCatalin Marinas2009-05-291-62/+164
| | | | | | | | | | This patch adds MTD concatenation support to integrator-flash.c for platforms with more than one block of flash memory (e.g. RealView PB11MPCore). The implementation is based on the sa1100-flash.c one. Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* mtd: DIL/NetPC broken for nowDavid Woodhouse2009-05-291-1/+1
| | | | | | We'll fix it up again, but for now I don't think anyone really cares. Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* [MTD] Remove option for add_mtd_partitions() to not register partitions.David Woodhouse2009-05-262-15/+4
| | | | | | | | | | | This breaks the dilnetpc map driver, but it could be fixed not to use that option. We want to simplify the partition handling, and this is a step towards that. Remove superfluous 'index' field from private struct mtd_part too, while we're at it. Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* [MTD] Only set partition suspend/resume method if parent not registeredDavid Woodhouse2009-05-261-1/+1
| | | | Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* [MTD] Remove mtd->{suspend,resume} calls from board driversDavid Woodhouse2009-05-265-113/+0
| | | | | | | | | Now the MTD core will do this for us, we don't need to hook it up from the board drivers. Shame we can't do shutdown from the class too... Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* [MTD] Restore suspend/resume support for mtd devicesDavid Woodhouse2009-05-261-10/+37
| | | | | | | This is intended to suspend/resume the _chip_, while we leave board drivers to handle their own suspend/resume for the controller. Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* [MTD] set blkdev parent to the mtd device, not its parentDavid Woodhouse2009-05-261-1/+1
| | | | Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* Linux 2.6.30-rc5v2.6.30-rc5Linus Torvalds2009-05-081-1/+1
|
* Merge git://git.infradead.org/mtd-2.6Linus Torvalds2009-05-082-15/+29
|\ | | | | | | | | | | | | | | * git://git.infradead.org/mtd-2.6: mtd: fix timeout in M25P80 driver mtd: Bug in m25p80.c during whole-chip erase mtd: expose subpage size via sysfs mtd: mtd in mtd_release is unused without CONFIG_MTD_CHAR
| * mtd: fix timeout in M25P80 driverPeter Horton2009-05-081-8/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | Extend erase timeout in M25P80 SPI Flash driver. The M25P80 drivers fails erasing sectors on a M25P128 because the ready wait timeout is too short. Change the timeout from a simple loop count to a suitable number of seconds. Signed-off-by: Peter Horton <zero@colonel-panic.org> Tested-by: Martin Michlmayr <tbm@cyrius.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
| * mtd: Bug in m25p80.c during whole-chip eraseSteven A. Falco2009-04-291-5/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | There is a logic error in "whole chip erase" for the m25p80 family. If the whole device is successfully erased, erase_chip() will return 0, and the code will fall through to the "else" clause, and do sector-by-sector erase in addition to the whole-chip erase. This patch corrects that. Also, the MAX_READY_WAIT_COUNT is insufficient for an m25p16 connected to a 400 MHz powerpc. Increasing it allows me to successfully program the device on my board. Signed-off-by: Steven A. Falco <sfalco@harris.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
| * mtd: expose subpage size via sysfsArtem Bityutskiy2009-04-191-0/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | MTD has got sysfs support in 2.6.30-rc1. But subpage size is not exposed there - do this. UBI utilities badly need this parameter. At the moment there is no way to get subpage size - ioctls do not return it. And we just got sysfs support, so we can easilly extend it with this additional parameter. This can be merged late in the development cycle because: 1. sysfs support has been just added - there are no users for it so far, even. 2. UBI utilities really need this parameter, and it is better not to delay this. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
| * mtd: mtd in mtd_release is unused without CONFIG_MTD_CHARDenis V. Lunev2009-04-191-3/+3
| | | | | | | | | | | | | | | | | | | | drivers/mtd/mtdcore.c: In function 'mtd_release': drivers/mtd/mtdcore.c:51: warning: unused variable 'mtd' [akpm: make it actually build] Signed-off-by: Denis V. Lunev <den@openvz.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* | Merge branch 'x86-fixes-for-linus' of ↵Linus Torvalds2009-05-088-15/+40
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip * 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: x86: MCE: make cmci_discover_lock irq-safe x86: xen, i386: reserve Xen pagetables x86, kexec: fix crashdump panic with CONFIG_KEXEC_JUMP x86-64: finish cleanup_highmaps()'s job wrt. _brk_end x86: fix boot hang in early_reserve_e820() x86: Fix a typo in a printk message x86, srat: do not register nodes beyond e820 map
| * | x86: MCE: make cmci_discover_lock irq-safeHidetoshi Seto2009-05-081-4/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Lockdep reports the warning below when Li tries to offline one cpu: [ 110.835487] ================================= [ 110.835616] [ INFO: inconsistent lock state ] [ 110.835688] 2.6.30-rc4-00336-g8c9ed89 #52 [ 110.835757] --------------------------------- [ 110.835828] inconsistent {HARDIRQ-ON-W} -> {IN-HARDIRQ-W} usage. [ 110.835908] swapper/0 [HC1[1]:SC0[0]:HE0:SE1] takes: [ 110.835982] (cmci_discover_lock){?.+...}, at: [<ffffffff80236dc0>] cmci_clear+0x30/0x9b cmci_clear() can be called via smp_call_function_single(). It is better to disable interrupt while holding cmci_discover_lock, to turn it into an irq-safe lock - we can deadlock otherwise. [ Impact: fix possible deadlock in the MCE code ] Reported-by: Shaohua Li <shaohua.li@intel.com> Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: Andrew Morton <akpm@linux-foundation.org> LKML-Reference: <4A03ED38.8000700@jp.fujitsu.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> Reported-by: Shaohua Li<shaohua.li@intel.com>
| * | x86: xen, i386: reserve Xen pagetablesJeremy Fitzhardinge2009-05-081-0/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The Xen pagetables are no longer implicitly reserved as part of the other i386_start_kernel reservations, so make sure we explicitly reserve them. This prevents them from being released into the general kernel free page pool and reused. [ Impact: fix Xen guest crash ] Also-Bisected-by: Bryan Donlan <bdonlan@gmail.com> Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com> Cc: Xen-devel <xen-devel@lists.xensource.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> LKML-Reference: <4A032EEC.30509@goop.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * | x86, kexec: fix crashdump panic with CONFIG_KEXEC_JUMPHuang Ying2009-05-072-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Tim Starling reported that crashdump will panic with kernel compiled with CONFIG_KEXEC_JUMP due to null pointer deference in machine_kexec_32.c: machine_kexec(), when deferencing kexec_image. Refering to: http://bugzilla.kernel.org/show_bug.cgi?id=13265 This patch fixes the BUG via replacing global variable reference: kexec_image in machine_kexec() with local variable reference: image, which is more appropriate, and will not be null. Same BUG is in machine_kexec_64.c too, so fixed too in the same way. [ Impact: fix crash on kexec ] Reported-by: Tim Starling <tstarling@wikimedia.org> Signed-off-by: Huang Ying <ying.huang@intel.com> LKML-Reference: <1241751101.6259.85.camel@yhuang-dev.sh.intel.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
| * | x86-64: finish cleanup_highmaps()'s job wrt. _brk_endJan Beulich2009-05-071-1/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | With the introduction of the .brk section, special care must be taken that no unused page table entries remain if _brk_end and _end are separated by a 2M page boundary. cleanup_highmap() runs very early and hence cannot take care of that, hence potential entries needing to be removed past _brk_end must be cleared once the brk allocator has done its job. [ Impact: avoids undesirable TLB aliases ] Signed-off-by: Jan Beulich <jbeulich@novell.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
OpenPOWER on IntegriCloud