summaryrefslogtreecommitdiffstats
path: root/drivers/mtd/ubi
Commit message (Collapse)AuthorAgeFilesLines
* Merge git://git.infradead.org/mtd-2.6Linus Torvalds2011-11-076-18/+18
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * git://git.infradead.org/mtd-2.6: (226 commits) mtd: tests: annotate as DANGEROUS in Kconfig mtd: tests: don't use mtd0 as a default mtd: clean up usage of MTD_DOCPROBE_ADDRESS jffs2: add compr=lzo and compr=zlib options jffs2: implement mount option parsing and compression overriding mtd: nand: initialize ops.mode mtd: provide an alias for the redboot module name mtd: m25p80: don't probe device which has status of 'disabled' mtd: nand_h1900 never worked mtd: Add DiskOnChip G3 support mtd: m25p80: add EON flash EN25Q32B into spi flash id table mtd: mark block device queue as non-rotational mtd: r852: make r852_pm_ops static mtd: m25p80: add support for at25df321a spi data flash mtd: mxc_nand: preset_v1_v2: unlock all NAND flash blocks mtd: nand: switch `check_pattern()' to standard `memcmp()' mtd: nand: invalidate cache on unaligned reads mtd: nand: do not scan bad blocks with NAND_BBT_NO_OOB set mtd: nand: wait to set BBT version mtd: nand: scrub BBT on ECC errors ... Fix up trivial conflicts: - arch/arm/mach-at91/board-usb-a9260.c Merged into board-usb-a926x.c - drivers/mtd/maps/lantiq-flash.c add_mtd_partitions -> mtd_device_register vs changed to use mtd_device_parse_register.
| * mtd: utilize `mtd_is_*()' functionsBrian Norris2011-09-216-18/+18
| | | | | | | | | | Signed-off-by: Brian Norris <computersforpeace@gmail.com> Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
* | mtd: Add export.h for EXPORT_SYMBOL/THIS_MODULE where neededPaul Gortmaker2011-10-311-0/+1
| | | | | | | | | | | | | | These two common macros will be no longer present everywhere. Call out the include needs of them explicitly where required. Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
* | UBI: do not link debug messages when debugging is disabledArtem Bityutskiy2011-08-191-1/+1
|/ | | | | | | | | | | | | | | Michal Marek spotted the same issue in UBIFS and this patch fixes UBI, see "UBIFS: not build debug messages with CONFIG_UBIFS_FS_DEBUG disabled" When UBI debugging is disabled, we have debugging messages defined as: if (0) pr_debug() But pr_debug macro defines data structures with debugging data and makes the linux binary larger, even though we have "if (0)". Signed-off-by: Artem Bityutskiy <artem.bityutskiy@intel.com>
* Merge branch 'for-linus' of ↵Linus Torvalds2011-07-221-3/+7
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6 * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6: (107 commits) vfs: use ERR_CAST for err-ptr tossing in lookup_instantiate_filp isofs: Remove global fs lock jffs2: fix IN_DELETE_SELF on overwriting rename() killing a directory fix IN_DELETE_SELF on overwriting rename() on ramfs et.al. mm/truncate.c: fix build for CONFIG_BLOCK not enabled fs:update the NOTE of the file_operations structure Remove dead code in dget_parent() AFS: Fix silly characters in a comment switch d_add_ci() to d_splice_alias() in "found negative" case as well simplify gfs2_lookup() jfs_lookup(): don't bother with . or .. get rid of useless dget_parent() in btrfs rename() and link() get rid of useless dget_parent() in fs/btrfs/ioctl.c fs: push i_mutex and filemap_write_and_wait down into ->fsync() handlers drivers: fix up various ->llseek() implementations fs: handle SEEK_HOLE/SEEK_DATA properly in all fs's that define their own llseek Ext4: handle SEEK_HOLE/SEEK_DATA generically Btrfs: implement our own ->llseek fs: add SEEK_HOLE and SEEK_DATA flags reiserfs: make reiserfs default to barrier=flush ... Fix up trivial conflicts in fs/xfs/linux-2.6/xfs_super.c due to the new shrinker callout for the inode cache, that clashed with the xfs code to start the periodic workers later.
| * fs: push i_mutex and filemap_write_and_wait down into ->fsync() handlersJosef Bacik2011-07-201-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Btrfs needs to be able to control how filemap_write_and_wait_range() is called in fsync to make it less of a painful operation, so push down taking i_mutex and the calling of filemap_write_and_wait() down into the ->fsync() handlers. Some file systems can drop taking the i_mutex altogether it seems, like ext3 and ocfs2. For correctness sake I just pushed everything down in all cases to make sure that we keep the current behavior the same for everybody, and then each individual fs maintainer can make up their mind about what to do from there. Thanks, Acked-by: Jan Kara <jack@suse.cz> Signed-off-by: Josef Bacik <josef@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
* | UBI: remove dead codeArtem Bityutskiy2011-06-071-13/+3
| | | | | | | | | | | | | | | | | | | | | | Thanks to new gcc 4.6 for issuing the following warning: drivers/mtd/ubi/vtbl.c: In function ‘create_vtbl’: drivers/mtd/ubi/vtbl.c:311:33: warning: variable ‘old_seb’ set but not used [-Wunused-but-set-variable] This patch removes some dead code and fixes the warning. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: dump stack when switching to R/O modeArtem Bityutskiy2011-06-011-0/+1
| | | | | | | | | | | | | | | | If we have debugging enabled and switching to R/O mode because of an error - dump the stack to improve UBI error reporting and make the further diagnostics easier to do. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: fix oops in error pathArtem Bityutskiy2011-06-011-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch fixes an oops in the error path of 'ubi_attach_mtd_dev()'. If anything after 'uif_init()' fails, we get an oops in 'cancel_pending()'. The reason is that 'uif_close()' drops the last reference count for 'ubi->dev' and whole 'struct ubi_device' is freed. And then 'ubi_wl_close()'->'cancel_pending()' tries to access the 'ubi' pointer and problems begin. Note, in 'ubi_detach_mtd_dev()' function we get a device reference to work-around this issue. Do the same in the error path of 'ubi_attach_mtd_dev()'. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: switch debugging tests knobs to debugfsArtem Bityutskiy2011-06-012-28/+49
| | | | | | | | | | | | | | | | | | Kill the UBI 'debug_tsts' module parameter and switch to debugfs. Create per-test mode files there. E.g., to enable bit-flips emulation you may just do: echo 1 > /sys/kernel/debug/ubi/ubi0/tst_emulate_bitflips Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: make it possible to use struct ubi_device in debug.hArtem Bityutskiy2011-06-012-9/+2
| | | | | | | | | | | | | | | | | | | | Current layout does not allow us to add inline functions to debug.h which use the 'struct ubi_device' object, because it is undefined there. Move '#include "debug.h"' in "ubi.h" down so to make 'struct ubi_device" be defined. Additionally, this makes it possible to remove a bunch of forward declarations in "debug.h". This is a preparation to the next patch. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: prepare debugging stuff to further debugfs conversionArtem Bityutskiy2011-06-013-13/+20
| | | | | | | | | | | | | | | | We'll need the 'struct ubi_device *ubi' pointer in every debugging function (to access the ->dbg field), so add this pointer to all the functions implementing UBI debugging test modes like 'ubi_dbg_is_bitflip()' etc. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: use debugfs for the extra checks knobsArtem Bityutskiy2011-06-019-36/+310
| | | | | | | | | | | | | | | | | | | | This patch introduces debugfs support to UBI. All the UBI stuff is kept in the "ubi" debugfs directory, which contains per-UBI device "ubi/ubiX" sub-directories, containing debugging files. This file also creates "ubi/ubiX/chk_gen" and "ubi/ubiX/chk_io" knobs for switching general and I/O extra checks on and off. And it removes the 'debug_chks' UBI module parameters. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: change the interface of a debugging check functionArtem Bityutskiy2011-05-311-13/+19
|/ | | | | | | | | | | This is a minor preparational patch which changes the 'paranoid_check_in_wl_tree()' function interface by adding the 'ubi' parameter which will be needed there in the next patch. And while on it, add "const" qualifier to the 'ubi' parameter of the 'paranoid_check_in_pq()' function. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* Merge git://git.infradead.org/mtd-2.6Linus Torvalds2011-05-271-3/+3
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * git://git.infradead.org/mtd-2.6: (97 commits) mtd: kill CONFIG_MTD_PARTITIONS mtd: remove add_mtd_partitions, add_mtd_device and friends mtd: convert remaining users to mtd_device_register() mtd: samsung onenand: convert to mtd_device_register() mtd: omap2 onenand: convert to mtd_device_register() mtd: txx9ndfmc: convert to mtd_device_register() mtd: tmio_nand: convert to mtd_device_register() mtd: socrates_nand: convert to mtd_device_register() mtd: sharpsl: convert to mtd_device_register() mtd: s3c2410 nand: convert to mtd_device_register() mtd: ppchameleonevb: convert to mtd_device_register() mtd: orion_nand: convert to mtd_device_register() mtd: omap2: convert to mtd_device_register() mtd: nomadik_nand: convert to mtd_device_register() mtd: ndfc: convert to mtd_device_register() mtd: mxc_nand: convert to mtd_device_register() mtd: mpc5121_nfc: convert to mtd_device_register() mtd: jz4740_nand: convert to mtd_device_register() mtd: h1910: convert to mtd_device_register() mtd: fsmc_nand: convert to mtd_device_register() ... Fixed up trivial conflicts in - drivers/mtd/maps/integrator-flash.c: removed in ARM tree - drivers/mtd/maps/physmap.c: addition of afs partition probe type clashing with removal of CONFIG_MTD_PARTITIONS
| * mtd: convert remaining users to mtd_device_register()Jamie Iles2011-05-251-3/+3
| | | | | | | | | | | | | | | | | | | | The older add_mtd_device()/add_mtd_partitions() and their removal counterparts will soon be gone. Replace uses with mtd_device_register() and mtd_device_unregister(). Signed-off-by: Jamie Iles <jamie@jamieiles.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
* | UBI: switch to dynamic printksArtem Bityutskiy2011-05-202-58/+28
| | | | | | | | | | | | | | Remove custom dynamic prints and the module parameter to toggle them and use the generic kernel dynamic printk infrastructure. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: turn some macros into static inlineArtem Bityutskiy2011-05-201-25/+38
| | | | | | | | | | | | | | | | Similarly as we have done for UBIFS in 1dcffad74183bb00e8129ba1c5bb2c9931d31bd7, turn the debugging macros into static inline functions, for the same reasons: to avoid gcc 4.5 warnings. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: improve checking in debugging printsArtem Bityutskiy2011-05-201-8/+25
| | | | | | | | | | | | | | | | When debugging is disabled, define debugging prints as if (0) printk() to make sure that the compiler still the format string in debugging messages even if debugging is disabled. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: fix typo in a messageArtem Bityutskiy2011-04-141-1/+1
| | | | | | | | | | | | | | | | When a PEB passes the torture test, UBI prints "do not mark it a bad", but should print "do not mark it as bad". This patch corrects the typo. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: fix minor stylistic issuesArtem Bityutskiy2011-04-146-17/+18
| | | | | | | | | | | | | | | | | | | | Fix checkpatch.pl errors and warnings: * space before tab * line over 80 characters * include linux/ioctl.h instead of asm/ioctl.h Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: use __packed instead of __attribute__((packed))Artem Bityutskiy2011-04-141-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There was an attempt to standartize various "__attribute__" and other macros in order to have potentially portable and more consistent code, see commit 82ddcb040570411fc2d421d96b3e69711c670328. Note, that commit refers Rober Love's blog post, but the URL is broken, the valid URL is: http://blog.rlove.org/2005/10/with-little-help-from-your-compiler.html Moreover, nowadays checkpatch.pl warns about using __attribute__((packed)): "WARNING: __packed is preferred over __attribute__((packed))" It is not a big deal for UBI to use __packed, so let's do it. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: re-name set volume properties ioctlArtem Bityutskiy2011-04-141-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Rename the ioctl which sets volume properties from 'UBI_IOCSETPROP' to 'UBI_IOCSETVOLPROP' to reflect the fact that this ioctl is about volume properties, not device properties. This is also consistent with the other volume ioctl name - 'UBI_IOCVOLUP'. The main motivation for the re-name, however, is that we are going to introduce the per-UBI device "set properties" ioctl, so we need good and logical naming. At the same time, re-name the "set volume properties request" data structure from 'struct ubi_set_prop_req' to 'struct ubi_set_vol_prop_req'. And re-name 'UBI_PROP_DIRECT_WRITE' to 'UBI_VOL_PROP_DIRECT_WRITE'. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | UBI: make the control character device non-seekableArtem Bityutskiy2011-04-141-1/+1
|/ | | | | | | | | | | | | | This patch makes the UBI control device (/dev/ubi_ctrl) non-seekable. The seek operation does is not applicable to this file, so it is cleaner to explicitly return error (which the added 'no_llseek()') does than trying to change the position (which the removed 'default_llseek()' does). This is an API break, but the only known user of this interface is mtd-utils which does not need the seeking functionality. And any app which relies on this is broken, but I'm not aware of such apps. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* Merge branch 'for-linus' of git://git.infradead.org/ubifs-2.6Linus Torvalds2011-04-073-7/+8
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | * 'for-linus' of git://git.infradead.org/ubifs-2.6: UBI: do not select KALLSYMS_ALL UBI: do not compare array with NULL UBI: check if we are in RO mode in the erase routine UBIFS: fix debugging failure in dbg_check_space_info UBIFS: fix error path in dbg_debugfs_init_fs UBIFS: unify error path dbg_debugfs_init_fs UBIFS: do not select KALLSYMS_ALL UBIFS: fix assertion warnings UBIFS: fix oops on error path in read_pnode UBIFS: do not read flash unnecessarily
| * UBI: do not select KALLSYMS_ALLArtem Bityutskiy2011-04-051-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | All UBI needs is to make sure we stacktraces when UBI debugging is enabled. It is enough to select KALLSYMS for this, KALLSYMS_ALL is not necessary. And the current Kconfig line we have: select KALLSYMS_ALL if KALLSYMS && DEBUG_KERNEL is just too complex to be sane and right. But this "if" part there is needed to prevent "unmet direct dependency" warnings, because KALLSYMS_ALL depends on KALLSYMS and DEBUG_KERNEL, so we cannot just select KALLSYMS_ALL. Anyway, this feels messy, and we do not seem to really need KALLSYMS_ALL, so select KALLSYMS instead. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com> Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
| * UBI: do not compare array with NULLArtem Bityutskiy2011-04-051-5/+0
| | | | | | | | | | | | | | | | Coverity spotted that UBI debugging code tries to compare an array and NULL, which obviously makes little sense. Kill this check. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * UBI: check if we are in RO mode in the erase routineArtem Bityutskiy2011-04-051-1/+7
| | | | | | | | | | | | | | | | | | | | 'do_sync_erase()' has to check whether we are in R/O mode before erasing the PEB. This patch adds the check and while on it, adds an assertion which validates the 'pnum' argument, as well as removes a check which is always true because it has already been done few lines before. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | Fix common misspellingsLucas De Marchi2011-03-311-1/+1
|/ | | | | | Fixes generated by 'codespell' and manually reviewed. Signed-off-by: Lucas De Marchi <lucas.demarchi@profusion.mobi>
* UBI: use GFP_NOFS properlyArtem Bityutskiy2011-03-241-2/+2
| | | | | | | | | This patch fixes a brown-paperbag bug which was introduced by me: I used incorrect "GFP_KERNEL | GFP_NOFS" allocation flags to make sure my allocations do not cause write-back. But the correct form is "GFP_NOFS". Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* Merge branch 'linux-next' of git://git.infradead.org/ubifs-2.6Linus Torvalds2011-03-183-0/+19
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * 'linux-next' of git://git.infradead.org/ubifs-2.6: (25 commits) UBIFS: clean-up commentaries UBIFS: save 128KiB or more RAM UBIFS: allocate orphans scan buffer on demand UBIFS: allocate lpt dump buffer on demand UBIFS: allocate ltab checking buffer on demand UBIFS: allocate scanning buffer on demand UBIFS: allocate dump buffer on demand UBIFS: do not check data crc by default UBIFS: simplify UBIFS Kconfig menu UBIFS: print max. index node size UBIFS: handle allocation failures in UBIFS write path UBIFS: use max_write_size during recovery UBIFS: use max_write_size for write-buffers UBIFS: introduce write-buffer size field UBI: incorporate LEB offset information UBIFS: incorporate maximum write size UBI: provide LEB offset information UBI: incorporate maximum write size UBIFS: fix LEB number in printk UBIFS: restrict world-writable debugfs files ...
| * UBI: provide LEB offset informationArtem Bityutskiy2011-03-081-0/+1
| | | | | | | | | | | | | | | | | | | | | | Provide the LEB offset information in the UBI device information data structure. This piece of information is required by UBIFS to find out what are the LEB offsets which are aligned to the max. write size. If LEB offset not aligned to max. write size, then UBIFS has to take this into account to write more optimally. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * UBI: incorporate maximum write sizeArtem Bityutskiy2011-03-083-0/+18
| | | | | | | | | | | | | | | | Incorporate MTD write buffer size into UBI device information because UBIFS needs this field. UBI does not use it ATM, just provides to upper layers in 'struct ubi_device_info'. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | Merge branch 'linux-next' of git://git.infradead.org/ubi-2.6Linus Torvalds2011-03-1812-234/+297
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * 'linux-next' of git://git.infradead.org/ubi-2.6: UBI: make tests modes dynamic UBI: make self-checks dynamic UBI: make debugging messages dynamic UBI: remove UBI_IO_DEBUG macro UBI: kill debugging buffer UBI: allocate erase checking buffer on demand UBI: allocate write checking buffer on demand UBI: always re-read in case of read failures UBI: cleanup comments about corrupted PEBs UBI: add slab cache for ubi_scan_leb objects UBI: use raw mtd read function in debugging code UBI: try to reveal buggy MTD drivers UBI: add a commentary about allocating VID header buffer on stack UBI: cleanup LEB start calculations UBI: fix NOR erase preparation quirk
| * | UBI: make tests modes dynamicArtem Bityutskiy2011-03-166-62/+49
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Similarly to the debugging checks and message, make the test modes be dynamically selected via the "debug_tsts" module parameter or via the "/sys/module/ubi/parameters/debug_tsts" sysfs file. This is consistent with UBIFS as well. And now, since all the Kconfig knobs became dynamic, we can remove the Kconfig.debug file completely. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: make self-checks dynamicArtem Bityutskiy2011-03-168-25/+70
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch adds a possibility to dynamically switch UBI self-checks on and off, instead of toggling them compile-time from the configuration menu. This is much more flexible, and consistent with UBIFS, and this also simplifies UBI Kconfig menu and the code. This patch introduces two levels of self-checks - general, which includes all self-checks which are relatively fast, and I/O, which includes write-verify checks and erase-verify checks, which are relatively slow and involve flash I/O. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: make debugging messages dynamicArtem Bityutskiy2011-03-163-55/+37
| | | | | | | | | | | | | | | | | | | | | | | | | | | This patch adds a possibility to dynamically select UBI debugging messages, instead of selecting them compile-time from the configuration menu. This is much more flexible, and consistent with UBIFS, and this also simplifies UBI Kconfig menu and the code. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: remove UBI_IO_DEBUG macroArtem Bityutskiy2011-03-162-21/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | This additional little macro is used to print a bit more messages while scanning the media. However, we have the 'dbg_bld()' macro for this, so we better us 'dbg_bld()' and kill UBI_IO_DEBUG. This simplifies the code a tiny bit. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: kill debugging bufferArtem Bityutskiy2011-03-162-19/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch kills the 'ubi->dbg_peb_buf' debugging buffer and the associated mutex, because all users of this buffer are now gone. We are killing this buffer because we are going to switch to dynamic debugging control, just like in UBIFS, which means that CONFIG_MTD_UBI_DEBUG_PARANOID will be removed. In this case we'd end up always allocating 'ubi->dbg_peb_buf', which is rather large (128KiB or more), and this would be wasteful. Thus, we are just killing it. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: allocate erase checking buffer on demandArtem Bityutskiy2011-03-161-7/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of using pre-allocated 'ubi->dbg_peb_buf' buffer in 'ubi_dbg_check_all_ff()', dynamically allocate it when needed. The intend is to get rid of the pre-allocated 'ubi->dbg_peb_buf' buffer completely. And the need for this arises because we want to change to dynamic debugging control instead of compile-time control, i.e., we are going to kill the CONFIG_MTD_UBI_DEBUG_PARANOID Kconfig option, which would mean that 'ubi->dbg_peb_buf' is always allocated, which would be wasteful. Thus, we are getting rid of 'ubi->dbg_peb_buf', and this is a preparation for that. signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: allocate write checking buffer on demandArtem Bityutskiy2011-03-162-9/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of using pre-allocated 'ubi->dbg_peb_buf' buffer in 'ubi_dbg_check_write()', dynamically allocate it when needed. The intend is to get rid of the pre-allocated 'ubi->dbg_peb_buf' buffer completely. And the need for this arises because we want to change to dynamic debugging control instead of compile-time control, i.e., we are going to kill the CONFIG_MTD_UBI_DEBUG_PARANOID Kconfig option, which would mean that 'ubi->dbg_peb_buf' is always allocated, which would be wasteful. Thus, we are getting rid of 'ubi->dbg_peb_buf', and this is a preparation for that. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: always re-read in case of read failuresArtem Bityutskiy2011-02-061-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When the read operation fails, UBI tries to re-read several times in a hope that one of the subsequent reads may succeed. However, currently UBI re-reads only if MTD failed to read all data, but does not re-reads if all the data were read, but with an integrity error (-EBADMSB). This patch makes UBI to always re-try reading. This should be useful for reading NAND pages with unstable bits - re-reading may help to get correct data. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: cleanup comments about corrupted PEBsArtem Bityutskiy2011-02-061-20/+34
| | | | | | | | | | | | | | | | | | Just make them a bit more readable and explanatory. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: add slab cache for ubi_scan_leb objectsArtem Bityutskiy2011-02-063-11/+25
| | | | | | | | | | | | | | | | | | | | | | | | | | | During scanning UBI allocates one struct ubi_scan_leb object for each PEB, so it can end up allocating thousands of them. Use slab cache to reduce memory consumption for these 48-byte objects, because currently used 'kmalloc()' ends up allocating 64 bytes per object, instead of 48. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: use raw mtd read function in debugging codeArtem Bityutskiy2011-02-061-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | This change affects only the debugging code. Namely, use mtd->read() function instead of ubi_io_read() to avoid bit-flips injection (ubi_dbg_is_bitflip()) which we do not want on the debugging path. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: try to reveal buggy MTD driversArtem Bityutskiy2011-02-061-0/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When reading data from the flash, corrupt the buffer we are about to read to. The idea is to fix the following possible situation: 1. The buffer contains data from previous operation, e.g., read from another PEB previously. The data looks like expected, e.g., if we just do not read anything and return - the caller would not notice this. E.g., if we are reading a VID header, the buffer may contain a valid VID header from another PEB. 2. The driver is buggy and returns use success or -EBADMSG or -EUCLEAN, but it does not actually put any data to the buffer. This may confuse UBI or upper layers - they may think the buffer contains valid data while in fact it is just old data. Thus, try to reveal such buggy MTD drivers with simple debugging code which fills the read buffer with 0x12 constant. Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: add a commentary about allocating VID header buffer on stackArtem Bityutskiy2011-01-261-0/+7
| | | | | | | | | | | | Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: cleanup LEB start calculationsJohn Ogness2011-01-261-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Wrong macro was used in calculating the data offset: UBI_EC_HDR_SIZE instead of UBI_VID_HDR_SIZE. The data offset should be VID header offset + VID header size (aligned to the minimum I/O unit). This was not a bug only because currently UBI_EC_HDR_SIZE and UBI_VID_HDR_SIZE have the same value of 64 bytes. Commit message was amended by Artem. Signed-off-by: John Ogness <john.ogness@linutronix.de> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
| * | UBI: fix NOR erase preparation quirkHolger Brunck2011-01-261-2/+4
| |/ | | | | | | | | | | | | | | | | | | In 'nor_erase_prepare()' we want to make sure the UBI headers are corrupted. But it is possible that one of the headers just contains all 0xFFs, which is also OK, because UBI will erase it in case of a power cut. Signed-off-by: Holger Brunck <holger.brunck@keymile.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
* | Revert "UBI: use mtd->writebufsize to set minimal I/O unit size"Artem Bityutskiy2011-01-291-27/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit a121f643993474548fe98144514c50dd4f3dbe76. Unfortunately, this commit breaks UBIFS backward compatibility and makes new UBIFS refuse older UBIFS-formatted media: UBIFS error: validate_sb: min. I/O unit mismatch: 8 in superblock, 64 real Thus, we have to revert this patch and work on a better solution. Reported-by: Holger Brunck <holger.brunck@keymile.com> Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
OpenPOWER on IntegriCloud