GitHub/exynos8895/android_kernel_samsung_universal8895.git
9 years agomtd: nand: fsmc_nand: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:47 +0000 (22:38 +0200)]
mtd: nand: fsmc_nand: show parent device in sysfs

Fix a bug where parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: fsl_ifc_nand: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:45 +0000 (22:38 +0200)]
mtd: nand: fsl_ifc_nand: show parent device in sysfs

Fix a bug where parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: docg4: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:43 +0000 (22:38 +0200)]
mtd: nand: docg4: show parent device in sysfs

Fix a bug where parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: denali: drop owner assignment
Frans Klaver [Wed, 10 Jun 2015 20:38:42 +0000 (22:38 +0200)]
mtd: nand: denali: drop owner assignment

Owner is automatically set by mtdcore. Make use of that.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: davinci_nand: drop owner and name assignment
Frans Klaver [Wed, 10 Jun 2015 20:38:41 +0000 (22:38 +0200)]
mtd: nand: davinci_nand: drop owner and name assignment

Owner and name are automatically set by mtdcore. Make use of that.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: cafe_nand: drop owner assignment
Frans Klaver [Wed, 10 Jun 2015 20:38:40 +0000 (22:38 +0200)]
mtd: nand: cafe_nand: drop owner assignment

Owner is automatically set by mtdcore. Make use of that.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: bf5xx_nand: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:39 +0000 (22:38 +0200)]
mtd: nand: bf5xx_nand: show parent device in sysfs

Fix a bug where parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: bcm47xxnflash: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:38 +0000 (22:38 +0200)]
mtd: nand: bcm47xxnflash: show parent device in sysfs

Fix a bug where parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: au1550nd: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:37 +0000 (22:38 +0200)]
mtd: nand: au1550nd: show parent device in sysfs

Fix a bug where parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: atmel_nand: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:36 +0000 (22:38 +0200)]
mtd: nand: atmel_nand: show parent device in sysfs

Fix a bug where parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: maps: sa1100-flash: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:35 +0000 (22:38 +0200)]
mtd: maps: sa1100-flash: show parent device in sysfs

Fix a bug where mtd parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.
Incidentally, it seems the owner field in the concatenated mtds is not
actually used, so this shouldn't make much of a difference anyway.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: maps: rbtx4939-flash: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:34 +0000 (22:38 +0200)]
mtd: maps: rbtx4939-flash: show parent device in sysfs

Fix a bug where mtd parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: maps: pxa2xx-flash: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:33 +0000 (22:38 +0200)]
mtd: maps: pxa2xx-flash: show parent device in sysfs

Fix a bug where mtd parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: maps: plat_ram: drop owner assignment
Frans Klaver [Wed, 10 Jun 2015 20:38:32 +0000 (22:38 +0200)]
mtd: maps: plat_ram: drop owner assignment

Owner is automatically set by mtdcore. Make use of that.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: maps: physmap_of: drop owner assignment
Frans Klaver [Wed, 10 Jun 2015 20:38:31 +0000 (22:38 +0200)]
mtd: maps: physmap_of: drop owner assignment

Owner is automatically set by mtdcore. Make use of that.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: maps: physmap: drop owner assignment
Frans Klaver [Wed, 10 Jun 2015 20:38:30 +0000 (22:38 +0200)]
mtd: maps: physmap: drop owner assignment

Owner is automatically set by mtdcore. Make use of that.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: maps: latch-addr-flash: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:29 +0000 (22:38 +0200)]
mtd: maps: latch-addr-flash: show parent device in sysfs

Fix a bug where mtd parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: maps: lantiq-flash: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:28 +0000 (22:38 +0200)]
mtd: maps: lantiq-flash: show parent device in sysfs

Fix a bug where mtd parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: maps: ixp4xx: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:27 +0000 (22:38 +0200)]
mtd: maps: ixp4xx: show parent device in sysfs

Fix a bug where mtd parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: maps: intel_vr_nor: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:26 +0000 (22:38 +0200)]
mtd: maps: intel_vr_nor: show parent device in sysfs

Fix a bug where mtd parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: maps: gpio-addr-flash: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:25 +0000 (22:38 +0200)]
mtd: maps: gpio-addr-flash: show parent device in sysfs

Fix a bug where mtd parent device symlinks aren't shown in sysfs.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: lpddr: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:23 +0000 (22:38 +0200)]
mtd: lpddr: show parent device in sysfs

Fix a bug where mtd parent device symlinks aren't shown in sysfs.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: devices: sst251: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:22 +0000 (22:38 +0200)]
mtd: devices: sst251: show parent device in sysfs

Fix a bug where mtd parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: devices: spear_smi: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:21 +0000 (22:38 +0200)]
mtd: devices: spear_smi: show parent device in sysfs

Fix a bug where mtd parent device symlinks aren't shown in sysfs.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: devices: mtd_dataflash: drop owner assignment
Frans Klaver [Wed, 10 Jun 2015 20:38:20 +0000 (22:38 +0200)]
mtd: devices: mtd_dataflash: drop owner assignment

Owner is automatically set by mtdcore. Make use of that.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: devices: docg3: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:18 +0000 (22:38 +0200)]
mtd: devices: docg3: show parent device in sysfs

Fix a bug where mtd parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: devices: bcm47xxflash: show parent device in sysfs
Frans Klaver [Wed, 10 Jun 2015 20:38:17 +0000 (22:38 +0200)]
mtd: devices: bcm47xxflash: show parent device in sysfs

Fix a bug where mtd parent device symlinks aren't shown in sysfs.

While at it, make use of the default owner value set by mtdcore.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: core: set some defaults when dev.parent is set
Frans Klaver [Wed, 10 Jun 2015 20:38:16 +0000 (22:38 +0200)]
mtd: core: set some defaults when dev.parent is set

If a parent device is set, add_mtd_device() has enough knowledge to fill
in some sane default values for the module name and owner. Do so if they
aren't already set.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: core: tone down suggestion that dev.parent should be set
Frans Klaver [Wed, 10 Jun 2015 20:38:15 +0000 (22:38 +0200)]
mtd: core: tone down suggestion that dev.parent should be set

add_mtd_device() has a comment suggesting that the caller should have
set dev.parent. This is required to have the parent device symlink show
up in sysfs, but not for proper operation of the mtd device itself.
Currently we have five drivers registering mtd devices during module
initialization, so they don't actually provide a parent device to link
to. That means we cannot WARN_ON() here, as it would trigger false
positives.

Make the comment a bit less firm in its assertion that dev.parent should
be set.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: spi-nor: Add support for Micron n25q032a
Aurelien Chanot [Wed, 7 Oct 2015 19:10:08 +0000 (12:10 -0700)]
mtd: spi-nor: Add support for Micron n25q032a

The N25Q032A is identical to the N25Q032 except it has a different
supply voltage range. Therefore, it has a new JEDEC ID.

Signed-off-by: Aurelien Chanot <chanot.a@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: Allow MTD_NAND_BRCMNAND to be selected for ARM64
Anup Patel [Fri, 2 Oct 2015 17:56:43 +0000 (23:26 +0530)]
mtd: nand: Allow MTD_NAND_BRCMNAND to be selected for ARM64

The BRCM NAND driver can be re-used for Broadcom ARM64 SoCs hence
this patch updates Kconfig to allow selection of MTD_NAND_BRCMNAND
for ARM64.

Signed-off-by: Anup Patel <anup.patel@broadcom.com>
Reviewed-by: Vikram Prakash <vikramp@broadcom.com>
Reviewed-by: Ray Jui <rjui@broadcom.com>
Reviewed-by: Pramod KUMAR <pramodku@broadcom.com>
Reviewed-by: Scott Branden <sbranden@broadcom.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: brcmnand: Fix pointer type-cast in brcmnand_write()
Anup Patel [Fri, 2 Oct 2015 17:56:42 +0000 (23:26 +0530)]
mtd: brcmnand: Fix pointer type-cast in brcmnand_write()

We should always type-cast pointer to "long" or "unsigned long"
because size of pointer is same as machine word size. This will
avoid pointer type-cast issues on both 32bit and 64bit systems.

This patch fixes pointer type-cast issue in brcmnand_write()
as-per above info.

Signed-off-by: Anup Patel <anup.patel@broadcom.com>
Reviewed-by: Vikram Prakash <vikramp@broadcom.com>
Reviewed-by: Ray Jui <rjui@broadcom.com>
Reviewed-by: Scott Branden <sbranden@broadcom.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: pxa2xx-flash: switch from ioremap_cache to memremap
Dan Williams [Fri, 9 Oct 2015 22:16:51 +0000 (18:16 -0400)]
mtd: pxa2xx-flash: switch from ioremap_cache to memremap

In preparation for deprecating ioremap_cache() convert its usage in
pxa2xx-flash to memremap.

Cc: David Woodhouse <dwmw2@infradead.org>
[brian: also convert iounmap to memunmap]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand_bbt: set the smallest size of bbt table
Sheng Yong [Fri, 31 Jul 2015 01:12:44 +0000 (01:12 +0000)]
mtd: nand_bbt: set the smallest size of bbt table

When using nandsim to simulate a 128K block nand with `overridesize = 1',
the size of mtd device is too small (mtd_size = 4 * block_size) to get the
right length of bbt. Then when creating bbt, kzmalloc() will return
ZERO_SIZE_PTR. This causes a NULL pointer oops when scanning bbt.

[  952.156166] BUG: unable to handle kernel NULL pointer dereference at 0000000000000010
[  952.157064] IP: [<ffffffff8148ad4a>] nand_isreserved_bbt+0x2a/0x40
[  952.157064] PGD 0
[  952.157064] Oops: 0000 [#1] SMP
[  952.157064] Modules linked in: nandsim(+) [last unloaded: nandsim]
[  952.157064] CPU: 1 PID: 7103 Comm: modprobe Not tainted 4.2.0-rc3-next-20150724 #4
[  952.157064] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006
[  952.157064] task: ffff88003e24b980 ti: ffff88003d274000 task.ti: ffff88003d274000
[  952.157064] RIP: 0010:[<ffffffff8148ad4a>]  [<ffffffff8148ad4a>] nand_isreserved_bbt+0x2a/0x40
[  952.157064] RSP: 0018:ffff88003d277b90  EFLAGS: 00010246
[  952.157064] RAX: 0000000000000010 RBX: ffff88003d5a1000 RCX: 0000000000000000
[  952.157064] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff88003d919000
[  952.157064] RBP: ffff88003d277b98 R08: 0000000000020000 R09: 0000000000000000
[  952.157064] R10: 0000000000000000 R11: 0000000000000195 R12: ffff88003d919000
[  952.157064] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
[  952.157064] FS:  00007fada4d07700(0000) GS:ffff88003fd00000(0000) knlGS:0000000000000000
[  952.157064] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[  952.157064] CR2: 0000000000000010 CR3: 0000000037924000 CR4: 00000000000006a0
[  952.157064] Stack:
[  952.157064]  ffffffff814851ec ffff88003d277ba8 ffffffff8147e35f ffff88003d277bf8
[  952.157064]  ffffffff814816f3 ffff88003d277c08 ffff88003d277bc8 0000000000000282
[  952.157064]  0000000000000001 0000000000000000 ffff88003d209540 0000000000000001
[  952.157064] Call Trace:
[  952.157064]  [<ffffffff814851ec>] ? nand_block_isreserved+0x1c/0x20
[  952.157064]  [<ffffffff8147e35f>] mtd_block_isreserved+0x1f/0x30
[  952.157064]  [<ffffffff814816f3>] allocate_partition+0x463/0x6a0
[  952.157064]  [<ffffffff81481b3b>] add_mtd_partitions+0x4b/0xe0
[  952.157064]  [<ffffffff8147f14c>] mtd_device_parse_register+0x4c/0xe0
[  952.157064]  [<ffffffffa0013daf>] ns_init_module+0xdaf/0xde4 [nandsim]
[  952.157064]  [<ffffffff8128d7c8>] ? kasprintf+0x38/0x40
[  952.157064]  [<ffffffffa0013000>] ? 0xffffffffa0013000
[  952.157064]  [<ffffffff810002c3>] do_one_initcall+0x83/0x1b0
[  952.157064]  [<ffffffff8113afab>] ? kmem_cache_alloc_trace+0x6b/0x120
[  952.157064]  [<ffffffff8160b503>] do_init_module+0x5c/0x1dd
[  952.157064]  [<ffffffff810aa4db>] load_module+0x1bbb/0x20b0
[  952.157064]  [<ffffffff810a6fc0>] ? __symbol_put+0x30/0x30
[  952.157064]  [<ffffffff810aaac9>] SyS_init_module+0xf9/0x110
[  952.157064]  [<ffffffff810aa9d1>] ? SyS_init_module+0x1/0x110
[  952.157064]  [<ffffffff81615f57>] entry_SYSCALL_64_fastpath+0x12/0x6a
[  952.157064] Code: 00 55 48 8b 87 80 01 00 00 48 89 e5 8b 88 cc 00 00 00 48 8b 80 f0 03 00 00 5d 48 d3 fe 89 f2 83 e6 03 c1 fa 02 8d 0c 36 48 63 d2 <0f> b6 04 10 d3 f8 83 e0 03 3c 02 0f 94 c0 0f b6 c0 c3 0f 1f 40
[  952.157064] RIP  [<ffffffff8148ad4a>] nand_isreserved_bbt+0x2a/0x40
[  952.157064]  RSP <ffff88003d277b90>
[  952.157064] CR2: 0000000000000010
[  952.204010] ---[ end trace 6ca2e1c041fdba36 ]---

This patch gives a smallest length to bbt, 1 byte, which is enough to
represent up to 4 blocks.

Signed-off-by: Sheng Yong <shengyong1@huawei.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: mtdpart: add debug prints to partition parser.
Michal Suchanek [Tue, 18 Aug 2015 15:34:07 +0000 (15:34 +0000)]
mtd: mtdpart: add debug prints to partition parser.

The probe of a mtd device can fail when a partition parser returns
error. The failure due to partition parsing can be quite mysterious when
multiple partitioning schemes are compiled in and any of them can fail
the probe.

Add debug prints which show what parsers were tried and what they
returned.

Signed-off-by: Michal Suchanek <hramrach@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agojffs2: fix a memleak in read_direntry()
Wei Fang [Tue, 29 Sep 2015 03:23:55 +0000 (11:23 +0800)]
jffs2: fix a memleak in read_direntry()

Need to free the memory allocated for 'fd' if failed to read all
of the remainder name.

Signed-off-by: Wei Fang <fangwei1@huawei.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: lpc32xx_slc: fix calculation of timing arcs from given values
Vladimir Zapolskiy [Wed, 30 Sep 2015 23:23:37 +0000 (02:23 +0300)]
mtd: nand: lpc32xx_slc: fix calculation of timing arcs from given values

According to LPC32xx User's Manual all values measured in clock cycles
are programmable from 1 to 16 clocks (4 bits) starting from 0 in
bitfield, the current version of calculated clock cycles is too
conservative.

Correctness of 0 bitfield value (i.e. programmed 1 clock
timing) is proven with actual NAND chip devices.

Signed-off-by: Vladimir Zapolskiy <vz@mleia.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: lpc32xx_slc: fix potential overflow over 4 bits
Vladimir Zapolskiy [Wed, 30 Sep 2015 23:23:36 +0000 (02:23 +0300)]
mtd: nand: lpc32xx_slc: fix potential overflow over 4 bits

In case if quotient of controller clock rate to device clock rate does
not fit into 4 bit value, choose the maximum acceptable value 0xF, which
stands for 16 clocks.

Signed-off-by: Vladimir Zapolskiy <vz@mleia.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: lpc32xx_slc: improve SLCTAC_*() macro definitions
Vladimir Zapolskiy [Wed, 30 Sep 2015 23:23:35 +0000 (02:23 +0300)]
mtd: nand: lpc32xx_slc: improve SLCTAC_*() macro definitions

No functional change, move bitfield calculations to macro
definitions with added clock rate argument, which are in turn defined
by new common SLCTAC_CLOCKS(c, n, s) macro definition.

Signed-off-by: Vladimir Zapolskiy <vz@mleia.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: sunxi: fix bitflips in erased pages
Boris BREZILLON [Wed, 30 Sep 2015 21:45:29 +0000 (23:45 +0200)]
mtd: nand: sunxi: fix bitflips in erased pages

Use the nand_check_erased_ecc_chunk() function to test if the ECC error
was triggered by an erased page containing a few bitflips.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: sunxi: replace the NFC_BUF_TO_USER_DATA() macro by an inline function
Boris BREZILLON [Wed, 30 Sep 2015 21:45:28 +0000 (23:45 +0200)]
mtd: nand: sunxi: replace the NFC_BUF_TO_USER_DATA() macro by an inline function

sunxi_nfc_user_data_to_buf() is exposed as an inline function, replace the
NFC_BUF_TO_USER_DATA() macro by an inline function to be consistent.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: sunxi: retrieve corrected OOB bytes
Boris BREZILLON [Wed, 30 Sep 2015 21:45:27 +0000 (23:45 +0200)]
mtd: nand: sunxi: retrieve corrected OOB bytes

The ECC engine is protecting a few OOB bytes. Retrieve them from the
USER_DATA register instead of reading them in raw mode (ie without the ECC
protection).

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: sunxi: factorize extra OOB bytes handling
Boris BREZILLON [Wed, 30 Sep 2015 21:45:26 +0000 (23:45 +0200)]
mtd: nand: sunxi: factorize extra OOB bytes handling

Add helper functions to factorize the code dealing extra OOB bytes in the
normal and syndrome ECC implementations.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: sunxi: make use of sunxi_nfc_hw_ecc_read/write_chunk()
Boris BREZILLON [Wed, 30 Sep 2015 21:45:25 +0000 (23:45 +0200)]
mtd: nand: sunxi: make use of sunxi_nfc_hw_ecc_read/write_chunk()

The sunxi_nfc_hw_ecc_read/write_chunk() functions have been created to
factorize the code in the normal and syndrome ECC implementation.
Make use of them where appropriate.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: sunxi: introduce sunxi_nfc_hw_ecc_read/write_chunk()
Boris BREZILLON [Wed, 30 Sep 2015 21:45:24 +0000 (23:45 +0200)]
mtd: nand: sunxi: introduce sunxi_nfc_hw_ecc_read/write_chunk()

The logic behind normal and syndrome ECC handling is pretty much the same,
the only difference is the ECC bytes placement.
Create two functions to read/write ECC chunks. Those functions will later
be used by the sunxi_nfc_hw_ecc_read/write_page() and
sunxi_nfc_hw_syndrome_ecc_read/write_page() functions.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: sunxi: create sunxi_nfc_hw_ecc_enable()/disable() functions
Boris BREZILLON [Wed, 30 Sep 2015 21:45:23 +0000 (23:45 +0200)]
mtd: nand: sunxi: create sunxi_nfc_hw_ecc_enable()/disable() functions

The code used to enable/disable the hardware ECC engine is repeated in a
lot of places. Create two functions to avoid code duplication.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: fsmc: Remove BUG macros
Stefan Roese [Fri, 2 Oct 2015 10:40:21 +0000 (12:40 +0200)]
mtd: nand: fsmc: Remove BUG macros

Remove the BUG macros and return with error (if possible) instead.

Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: fsmc: Small whitespace cleanup
Stefan Roese [Fri, 2 Oct 2015 10:40:20 +0000 (12:40 +0200)]
mtd: nand: fsmc: Small whitespace cleanup

Remove tab in empty line.

Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: remove unused ->init_size() hook
Boris BREZILLON [Thu, 1 Oct 2015 14:58:27 +0000 (16:58 +0200)]
mtd: nand: remove unused ->init_size() hook

The ->init_size() hook was introduced to let NAND controller drivers
support NAND devices that could not be described in the nand_ids table.
Since then, the core has added support for extended-id parsing and
full-id description, thus allowing to describe pretty much all existing
NANDs.
Moreover, this hook is not used by any mainline driver, and should not be
used by new drivers, because detecting the NAND chip is not something
controller specific.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agosh: mach-rsk: remove unnecessary MTD partition probe specification
Brian Norris [Mon, 18 May 2015 23:24:30 +0000 (16:24 -0700)]
sh: mach-rsk: remove unnecessary MTD partition probe specification

The cmdlinepart parser is already supported in the default probe.

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: brcmnand: remove unnecessary fields from brcmnand_soc
Brian Norris [Thu, 21 May 2015 00:05:06 +0000 (17:05 -0700)]
mtd: brcmnand: remove unnecessary fields from brcmnand_soc

These really aren't needed, especially now that we embed the soc struct
in our private struct, so we can stash things there if needed.

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: brcmnand: refactor iProc SoC layering
Brian Norris [Thu, 21 May 2015 00:05:05 +0000 (17:05 -0700)]
mtd: brcmnand: refactor iProc SoC layering

Removes an unnecessary allocation and saves a little bit of pointer
chasing.

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: brcmnand: refactor bcm63138 SoC layering
Brian Norris [Thu, 21 May 2015 00:05:04 +0000 (17:05 -0700)]
mtd: brcmnand: refactor bcm63138 SoC layering

Removes an unnecessary allocation and saves a little bit of pointer
chasing.

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: sunxi: rework macros
Boris BREZILLON [Wed, 16 Sep 2015 07:46:36 +0000 (09:46 +0200)]
mtd: nand: sunxi: rework macros

Suffix mask macros with _MSK and add new helper macros to avoid manually
shifting values.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agoMerge MTD 4.3-rc updates into -next
Brian Norris [Wed, 30 Sep 2015 18:12:14 +0000 (11:12 -0700)]
Merge MTD 4.3-rc updates into -next

9 years agomtd: nand: vf610_nfc: include missing pincrl/consumer.h
Brian Norris [Wed, 30 Sep 2015 16:54:26 +0000 (09:54 -0700)]
mtd: nand: vf610_nfc: include missing pincrl/consumer.h

This must have been implicitly included on the builds I tested. Reported
by numerous test bots:

   drivers/mtd/nand/vf610_nfc.c: In function 'vf610_nfc_resume':
   drivers/mtd/nand/vf610_nfc.c:660:2: error: implicit declaration of function 'pinctrl_pm_select_default_state' [-Werror=implicit-function-declaration]
     pinctrl_pm_select_default_state(dev);
     ^

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Acked-by: Stefan Agner <stefan@agner.ch>
9 years agomtd: mtdpart: fix add_mtd_partitions error path
Boris BREZILLON [Thu, 30 Jul 2015 10:18:03 +0000 (12:18 +0200)]
mtd: mtdpart: fix add_mtd_partitions error path

If we fail to allocate a partition structure in the middle of the partition
creation process, the already allocated partitions are never removed, which
means they are still present in the partition list and their resources are
never freed.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Cc: stable@vger.kernel.org
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agoRevert "mtd: mtdram: check offs and len in mtdram->erase"
Brian Norris [Wed, 30 Sep 2015 00:28:44 +0000 (17:28 -0700)]
Revert "mtd: mtdram: check offs and len in mtdram->erase"

This reverts commit 7827e3acad2df1c6537e5fe7211d216dabc60399.

There are some 64-bit arithmetic issues on some architectures, so let's
wait until we get a better patch for this.

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: mtdram: check offs and len in mtdram->erase
Dongsheng Yang [Tue, 21 Jul 2015 08:30:20 +0000 (16:30 +0800)]
mtd: mtdram: check offs and len in mtdram->erase

We should prevent user to erasing mtd device with
an unaligned offset or length.

Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: vf610_nfc: add device tree bindings
Stefan Agner [Thu, 3 Sep 2015 01:06:35 +0000 (18:06 -0700)]
mtd: nand: vf610_nfc: add device tree bindings

Signed-off-by: Bill Pringlemeir <bpringlemeir@nbsps.com>
Acked-by: Shawn Guo <shawnguo@kernel.org>
Signed-off-by: Stefan Agner <stefan@agner.ch>
[Brian: fixup #size-cells in example]
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: vf610_nfc: add hardware BCH-ECC support
Stefan Agner [Thu, 3 Sep 2015 01:06:34 +0000 (18:06 -0700)]
mtd: nand: vf610_nfc: add hardware BCH-ECC support

This adds hardware ECC support using the BCH encoder in the NFC IP.
The ECC encoder supports up to 32-bit correction by using 60 error
correction bytes. There is no sub-page ECC step, ECC is calculated
always across the whole page (up to 2k pages).

Limitations:
- HW ECC: Only 2K page with 64+ OOB.
- HW ECC: Only 24 and 32-bit error correction implemented.

Raw writes have been tested using the generic nand_write_page_raw
implementation. However, raw reads are currently not possible
because the controller need to know whether we are going to use
the ECC mode already at NAND_CMD_READ0 command time. At this point
we do not have the information whether it is a raw read or a
regular read at driver level...

Signed-off-by: Bill Pringlemeir <bpringlemeir@nbsps.com>
Signed-off-by: Stefan Agner <stefan@agner.ch>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: vf610_nfc: Freescale NFC for VF610, MPC5125 and others
Stefan Agner [Thu, 3 Sep 2015 01:06:33 +0000 (18:06 -0700)]
mtd: nand: vf610_nfc: Freescale NFC for VF610, MPC5125 and others

This driver supports Freescale NFC (NAND flash controller) found on
Vybrid (VF610), MPC5125, MCF54418 and Kinetis K70. The driver has
been tested using 8-bit and 16-bit NAND interface on the ARM based
Vybrid SoC VF500 and VF610 platform.
parameter page reading.

Limitations:
- Untested on MPC5125 and M54418.
- DMA and pipelining not used.
- 2K pages or less.
- No chip select, one NAND chip per controller.
- No hardware ECC.

Some paths have been hand-optimized and evaluated by measurements
made using mtd_speedtest.ko on a 100MB MTD partition.

Colibri VF50
        eb write     %   eb read     %   page write      %   page read     %
rel/opt     5175           11537                4560             11039
opt         5164 -0.21     11420 -1.01          4737 +3.88       10918 -1.10
none        5113 -1.20     11352 -1.60          4490 -1.54       10865 -1.58

Colibri VF61
        eb write     %   eb read     %   page write      %   page read     %
rel/opt     5766           13096                5459             12846
opt         5883 +2.03     13064 -0.24          5561 +1.87       12802 -0.34
none        5701 -1.13     12980 -0.89          5488 +0.53       12735 -0.86

rel = using readl_relaxed/writel_relaxed in optimized paths
opt = hand-optimized by combining multiple accesses into one read/write

The measurements have not been statistically verfied, hence use them
with care. The author came to the conclusion that using the relaxed
variants of readl/writel are not worth the additional code.

Signed-off-by: Bill Pringlemeir <bpringlemeir@nbsps.com>
Tested-by: Albert ARIBAUD <albert.aribaud@3adev.fr>
Signed-off-by: Stefan Agner <stefan@agner.ch>
Reviewed-by: Alexey Klimov <klimov.linux@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: provide proper 32/64-bit compat_ioctl() support for BLKPG
Brian Norris [Mon, 21 Sep 2015 20:26:59 +0000 (13:26 -0700)]
mtd: provide proper 32/64-bit compat_ioctl() support for BLKPG

After a bit of poking around wondering why my 32-bit user-space can't
seem to send a proper ioctl(BLKPG) to an MTD on my 64-bit kernel
(ARM64), I noticed that struct blkpg_ioctl_arg is actually pretty
unsuitable for use in the ioctl() ABI, due to its use of raw pointers,
and its lack of alignment/packing restrictions (32-bit arch'es tend to
pack the 4 fields into 4 32-bit words, whereas 64-bit arch'es would add
padding after the third int, and make this 6 32-bit words).

Anyway, this means BLKPG deserves some special compat_ioctl handling. Do
the conversion in a small shim for MTD.

block/compat_ioctl.c already has compat support for the block subsystem,
but it does so by a re-marshalling data to/from user-space (see
compat_blkpg_ioctl()). Personally, I think this approach is cleaner.

Tested only on MTD, with an ARM32 user space on an ARM64 kernel.

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: spi-nor: add support for w25q128fw
Brian Norris [Fri, 18 Sep 2015 22:08:14 +0000 (15:08 -0700)]
mtd: spi-nor: add support for w25q128fw

Tested only with single I/O, but the datasheet says it supports dual and
quad.

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: spi-nor: scale up timeout for full-chip erase
Furquan Shaikh [Fri, 18 Sep 2015 21:59:17 +0000 (14:59 -0700)]
mtd: spi-nor: scale up timeout for full-chip erase

This patch fixes timeout issues seen on large NOR flash (e.g., 16MB
w25q128fw) when using ioctl(MEMERASE) with offset=0 and length=16M. The
input parameters matter because spi_nor_erase() uses a different code
path for full-chip erase, where we use the SPINOR_OP_CHIP_ERASE (0xc7)
opcode.

Fix: use a different timeout for full-chip erase than for other
commands.

While most operations can be expected to perform relatively similarly
across a variety of NOR flash types and sizes (and therefore might as
well use a similar timeout to keep things simple), full-chip erase is
unique, because the time it typically takes to complete:
(1) is much larger than most operations and
(2) scales with the size of the flash.

Let's base our timeout on the original comments stuck here -- that a 2MB
flash requires max 40s to erase.

Small survey of a few flash datasheets I have lying around:

  Chip         Size (MB)   Max chip erase (seconds)
  ----         --------    ------------------------
  w25q32fw     4           50
  w25q64cv     8           30
  w25q64fw     8           100
  w25q128fw    16          200
  s25fl128s    16          ~256
  s25fl256s    32          ~512

From this data, it seems plenty sufficient to say we need to wait for
40 seconds for each 2MB of flash.

After this change, it might make some sense to decrease the timeout for
everything else, as even the most extreme operations (single block
erase?) shouldn't take more than a handful of seconds. But for safety,
let's leave it as-is. It's only an error case, after all, so we don't
exactly need to optimize it.

Signed-off-by: Furquan Shaikh <furquan@google.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: spi-nor: Add support for sst25wf040b
Yao Yuan [Wed, 16 Sep 2015 09:59:45 +0000 (17:59 +0800)]
mtd: spi-nor: Add support for sst25wf040b

It is a 512KiB flash with 4 KiB erase sectors.

Signed-off-by: Yuan Yao <yao.yuan@freescale.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agopcmciamtd: Add id for PRETEC 4MB SRAM
Tom Englund [Fri, 18 Sep 2015 19:42:59 +0000 (21:42 +0200)]
pcmciamtd: Add id for PRETEC 4MB SRAM

The module pcmciamtd doesn't generate a mtd node for PRETEC 4MB SRAM
cards without the id and hash added to pcmciamtd.c

Tested on 3 different 4MB pretec sram cards.

Signed-off-by: Tom Englund <tomenglund26@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: denali: max_banks calculation changed in revision 5.1
Graham Moore [Tue, 21 Jul 2015 14:39:31 +0000 (09:39 -0500)]
mtd: nand: denali: max_banks calculation changed in revision 5.1

Read Denali hardware revision number and use it to
calculate max_banks,  The encoding of max_banks changed
in Denali revision 5.1.

Signed-off-by: Graham Moore <grmoore@opensource.altera.com>
[Brian: parentheses around macro arg]
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: denali: pass col argument to READID operation
Enrico Jorns [Fri, 18 Sep 2015 08:02:41 +0000 (10:02 +0200)]
mtd: nand: denali: pass col argument to READID operation

A read id operation followed by 0x00 reads the device ID while
a read id operation followed by 0x20 reads the possible ONFI identifier.

As the READID function did not propagate the second id parameter but had
a hard-coded call for 0x90 0x00, reading the ONFI identifier was not
possible and thus chips werde not detected (tested with
MT29F8G08ABABAWP)

Signed-off-by: Enrico Jorns <ejo@pengutronix.de>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agojffs2: remove unneeded kfree
fangwei [Tue, 7 Jul 2015 02:54:26 +0000 (10:54 +0800)]
jffs2: remove unneeded kfree

c->oobbuf hasn't been kmalloced in jffs2_dataflash_setup, so
there is no need to free it.

Signed-off-by: Wei Fang <fangwei1@huawei.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: cmdlinepart: convert printk() to pr_*()
Brian Norris [Fri, 19 Jun 2015 20:10:07 +0000 (13:10 -0700)]
mtd: cmdlinepart: convert printk() to pr_*()

This driver uses some custom macros for printing. Let's use the standard
pr_fmt()/pr_{err,warn}().

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: cmdlinepart: allow small partitions
Brian Norris [Fri, 19 Jun 2015 20:10:06 +0000 (13:10 -0700)]
mtd: cmdlinepart: allow small partitions

I'm not sure why we have a PAGE_SIZE restriction on this partition
parser.

If we really wanted the restriction, I would expect it to be a
restriction for *all* parsers, so we'd move it to the MTD core

At any rate, while small partitions may not be useful (they'll often be
smaller than the eraseblock size and therefore can only be used
read-only), they still have use as a read-only partition.

This restriction is especially annoying because it aborts the entire
MTD's cmdline parsing, leaving it unpartitioned.

So, let's kill the restriction and only check for zero-sized partitions,
which I expect we don't want to allow.

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: spi-nor: s25fl008k and s25fl016k supports dual/quad mode
Joachim Eastwood [Thu, 9 Jul 2015 20:30:57 +0000 (22:30 +0200)]
mtd: spi-nor: s25fl008k and s25fl016k supports dual/quad mode

s25fl016k can be found on Embedded Artists' LPC4357 Developer's Kit
where is used in quad mode by the LPC4357 SPIFI controller.

Signed-off-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: Destroy mtd_idr on module_exit
Johannes Thumshirn [Wed, 8 Jul 2015 15:15:34 +0000 (17:15 +0200)]
mtd: Destroy mtd_idr on module_exit

Destroy mtd_idr on module_exit, reclaiming the allocated memory.

This was detected by the following semantic patch (written by Luis Rodriguez
<mcgrof@suse.com>)
<SmPL>
@ defines_module_init @
declarer name module_init, module_exit;
declarer name DEFINE_IDR;
identifier init;
@@

module_init(init);

@ defines_module_exit @
identifier exit;
@@

module_exit(exit);

@ declares_idr depends on defines_module_init && defines_module_exit @
identifier idr;
@@

DEFINE_IDR(idr);

@ on_exit_calls_destroy depends on declares_idr && defines_module_exit @
identifier declares_idr.idr, defines_module_exit.exit;
@@

exit(void)
{
 ...
  idr_destroy(&idr);
  ...
}

@ missing_module_idr_destroy depends on declares_idr && defines_module_exit && !on_exit_calls_destroy @
identifier declares_idr.idr, defines_module_exit.exit;
@@

exit(void)
{
 ...
 +idr_destroy(&idr);
}
</SmPL>

Signed-off-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: pxa3xx-nand: prevent DFI bus lockup on removal
Robert Jarzmik [Mon, 28 Sep 2015 20:56:51 +0000 (22:56 +0200)]
mtd: nand: pxa3xx-nand: prevent DFI bus lockup on removal

After the conversion of pxa architecture to common clock framework, the
NAND clock can be disabled on driver exit.

In this case, it happens that if the driver used the NAND and set the
DFI arbitration bit, the next access to a static memory controller area,
such as an ethernet card, will stall the system bus, and the core will
be stalled forever.

This is especially true on pxa31x SoCs, where the NDCR was augmented
with a new bit to prevent this lockups by giving full ownership of the
DFI arbiter to the SMC, in change SCr#6.

Fix this by clearing the DFI arbritration bit in driver exit. This
effectively prevents a lockup on zylonite when removing pxa3xx-nand
module, and using ethernet afterwards.

Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: mxc_nand: Fix module autoload for OF platform driver
Luis de Bethencourt [Thu, 17 Sep 2015 22:13:28 +0000 (00:13 +0200)]
mtd: mxc_nand: Fix module autoload for OF platform driver

This platform driver has a OF device ID table but the OF module
alias information is not created so module autoloading won't work.

Signed-off-by: Luis de Bethencourt <luisbg@osg.samsung.com>
Acked-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: blktrans: fix multiplication overflow
Peng Fan [Fri, 11 Sep 2015 13:41:47 +0000 (21:41 +0800)]
mtd: blktrans: fix multiplication overflow

In drivers/mtd/mtd_blkdevs.c:
406 set_capacity(gd, (new->size * tr->blksize) >> 9);
The type of new->size is unsigned long and the type of tr->blksize is int,
the result of 'new->size * tr->blksize' may exceed ULONG_MAX on 32bit
machines.

I use nand chip MT29F32G08CBADBWP which is 4GB and the parameters passed
to kernel is 'mtdparts=gpmi-nand:-(user)', the whole nand chip will be
treated as a 4GB mtd partition. new->size is 0x800000 and tr->blksize is
0x200, 'new->size * tr->blksize' however is 0. This is what we do not want
to see.

Using type cast u64 to fix the multiplication overflow issue.

Signed-off-by: Peng Fan <van.freenix@gmail.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: mpc5121_nfc: Fix module autoload for OF platform driver
Luis de Bethencourt [Thu, 17 Sep 2015 22:13:01 +0000 (00:13 +0200)]
mtd: mpc5121_nfc: Fix module autoload for OF platform driver

This platform driver has a OF device ID table but the OF module
alias information is not created so module autoloading won't work.

Signed-off-by: Luis de Bethencourt <luisbg@osg.samsung.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: fsl_ifc_nand: Fix module autoload for OF platform driver
Luis de Bethencourt [Thu, 17 Sep 2015 22:12:30 +0000 (00:12 +0200)]
mtd: fsl_ifc_nand: Fix module autoload for OF platform driver

This platform driver has a OF device ID table but the OF module
alias information is not created so module autoloading won't work.

Signed-off-by: Luis de Bethencourt <luisbg@osg.samsung.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: fsl_elbc_nand: Fix module autoload for OF platform driver
Luis de Bethencourt [Thu, 17 Sep 2015 22:11:59 +0000 (00:11 +0200)]
mtd: fsl_elbc_nand: Fix module autoload for OF platform driver

This platform driver has a OF device ID table but the OF module
alias information is not created so module autoloading won't work.

Signed-off-by: Luis de Bethencourt <luisbg@osg.samsung.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: orion_nand: Fix module autoload for OF platform driver
Luis de Bethencourt [Thu, 17 Sep 2015 22:14:02 +0000 (00:14 +0200)]
mtd: orion_nand: Fix module autoload for OF platform driver

This platform driver has a OF device ID table but the OF module
alias information is not created so module autoloading won't work.

Signed-off-by: Luis de Bethencourt <luisbg@osg.samsung.com>
Acked-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agojffs2: remove unnecessary new_valid_dev check
Yaowei Bai [Mon, 28 Sep 2015 14:15:38 +0000 (22:15 +0800)]
jffs2: remove unnecessary new_valid_dev check

As new_valid_dev always returns 1, so !new_valid_dev check is not
needed, remove it.

Signed-off-by: Yaowei Bai <bywxiaobai@163.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: pxa3xx-nand: switch to dmaengine
Robert Jarzmik [Sun, 6 Sep 2015 13:12:47 +0000 (15:12 +0200)]
mtd: nand: pxa3xx-nand: switch to dmaengine

Now pxa architecture has a dmaengine driver, remove the access to direct
dma registers in favor of the more generic dmaengine code.

This should be also applicable for mmp and orion, provided they work in
device-tree environment.

This patch also removes the previous hack which was necessary to make
the driver work in a devicetree environment.

Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
Reviewed-by: Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
Tested-by: Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
[Brian: fixup use of 'enum dma_transfer_direction']
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomxc_nand: fix copy_spare
Eric Benard [Wed, 23 Sep 2015 15:07:28 +0000 (17:07 +0200)]
mxc_nand: fix copy_spare

it was broken by 35d5d20efad8a04c8c002c7f31241dff973977a6
"mtd: mxc_nand: cleanup copy_spare function"

else we get the following error :
[   22.709507] ubi0: attaching mtd3
[   23.613470] ubi0: scanning is finished
[   23.617278] ubi0: empty MTD device detected
[   23.623219] Unhandled fault: imprecise external abort (0x1c06) at 0x9e62f0ec
[   23.630291] pgd = 9df80000
[   23.633005] [9e62f0ec] *pgd=8e60041e(bad)
[   23.637064] Internal error: : 1c06 [#1] SMP ARM
[   23.641605] Modules linked in:
[   23.644687] CPU: 0 PID: 99 Comm: ubiattach Not tainted 4.2.0-dirty #22
[   23.651222] Hardware name: Freescale i.MX53 (Device Tree Support)
[   23.657322] task: 9e687300 ti: 9dcfc000 task.ti: 9dcfc000
[   23.662744] PC is at memcpy16_toio+0x4c/0x74
[   23.667026] LR is at mxc_nand_command+0x484/0x640
[   23.671739] pc : [<803f9c08>]    lr : [<803faeb0>]    psr: 60000013
[   23.671739] sp : 9dcfdb10  ip : 9e62f0ea  fp : 9dcfdb1c
[   23.683222] r10: a09c1000  r9 : 0000001a  r8 : ffffffff
[   23.688453] r7 : ffffffff  r6 : 9e674810  r5 : 9e674810  r4 : 000000b6
[   23.694985] r3 : a09c16a4  r2 : a09c16a4  r1 : a09c16a4  r0 : 0000ffff
[   23.701521] Flags: nZCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment user
[   23.708662] Control: 10c5387d  Table: 8df80019  DAC: 00000015
[   23.714413] Process ubiattach (pid: 99, stack limit = 0x9dcfc210)
[   23.720514] Stack: (0x9dcfdb10 to 0x9dcfe000)
[   23.724881] db00:                                     9dcfdb6c 9dcfdb20 803faeb0 803f9bc8
[   23.733069] db20: 803f227c 803f9b74 ffffffff 9e674810 9e674810 9e674810 00000040 9e62f010
[   23.741255] db40: 803faa2c 9e674b40 9e674810 803faa2c 00000400 803faa2c 00000000 9df42800
[   23.749441] db60: 9dcfdb9c 9dcfdb70 803f2024 803faa38 9e4201cc 00000000 803f0a78 9e674b40
[   23.757627] db80: 803f1f80 9e674810 00000400 00000400 9dcfdc14 9dcfdba0 803f3bd8 803f1f8c
[   23.765814] dba0: 9e4201cc 00000000 00000580 00000000 00000000 800718c0 0000007f 00001000
[   23.774000] dbc0: 9df42800 000000e0 00000000 00000000 9e4201cc 00000000 00000000 00000000
[   23.782186] dbe0: 00000580 00000580 00000000 9e674810 9dcfdc20 9dcfdce8 9df42800 00580000
[   23.790372] dc00: 00000000 00000400 9dcfdc6c 9dcfdc18 803f3f94 803f39a4 9dcfdc20 00000000
[   23.798558] dc20: 00000000 00000400 00000000 00000000 00000000 00000000 9df42800 00000000
[   23.806744] dc40: 9dcfdd0c 00580000 00000000 00000400 00000000 9df42800 9dee1000 9d802000
[   23.814930] dc60: 9dcfdc94 9dcfdc70 803eb63c 803f3f38 00000400 9dcfdce8 9df42800 dead4ead
[   23.823116] dc80: 803eb5f4 00000000 9dcfdcc4 9dcfdc98 803e82ac 803eb600 00000400 9dcfdce8
[   23.831301] dca0: 9df42800 00000400 9dee0000 00000000 00000400 00000000 9dcfdd1c 9dcfdcc8
[   23.839488] dcc0: 80406048 803e8230 00000400 9dcfdce8 9df42800 9dcfdc78 00000008 00000000
[   23.847673] dce0: 00000000 00000000 00000000 00000004 00000000 9df42800 9dee0000 00000000
[   23.855859] dd00: 9d802030 00000000 9dc8b214 9d802000 9dcfdd44 9dcfdd20 804066cc 80405f50
[   23.864047] dd20: 00000400 9dc8b200 9d802030 9df42800 9dee0000 9dc8b200 9dcfdd84 9dcfdd48
[   23.872233] dd40: 8040a544 804065ac 9e401c80 000080d0 9dcfdd84 00000001 800fc828 9df42400
[   23.880418] dd60: 00000000 00000080 9dc8b200 9dc8b200 9dc8b200 9dee0000 9dcfdddc 9dcfdd88
[   23.888605] dd80: 803fb560 8040a440 9dcfddc4 9dcfdd98 800f1428 9dee1000 a0acf000 00000000
[   23.896792] dda0: 00000000 ffffffff 00000006 00000000 9dee0000 9dee0000 00005600 00000080
[   23.904979] ddc0: 9dc8b200 a0acf000 9dc8b200 8112514c 9dcfde24 9dcfdde0 803fc08c 803fb4f0
[   23.913165] dde0: 9e401c80 00000013 9dcfde04 9dcfddf8 8006bbf8 8006ba00 9dcfde24 00000000
[   23.921351] de00: 9dee0000 00000065 9dee0000 00000001 9dc8b200 8112514c 9dcfde84 9dcfde28
[   23.929538] de20: 8040afa0 803fb948 ffffffff 00000000 9dc8b214 9dcfde40 800f1428 800f11dc
[   23.937724] de40: 9dc8b21c 9dc8b20c 9dc8b204 9dee1000 9dc8b214 8069bb60 fffff000 fffff000
[   23.945911] de60: 9e7b5400 00000000 9dee0000 9dee1000 00001000 9e7b5400 9dcfdecc 9dcfde88
[   23.954097] de80: 803ff1bc 8040a630 9dcfdea4 9dcfde98 00000800 00000800 9dcfdecc 9dcfdea8
[   23.962284] dea0: 803e8f6c 00000000 7e87ab70 9e7b5400 80113e30 00000003 9dcfc000 00000000
[   23.970470] dec0: 9dcfdf04 9dcfded0 804008cc 803feb98 ffffffff 00000003 00000000 00000000
[   23.978656] dee0: 00000000 00000000 9e7cb000 9dc193e0 7e87ab70 9dd92140 9dcfdf7c 9dcfdf08
[   23.986842] df00: 80113b5c 8040080c 800fbed8 8006bbf0 9e7cb000 00000003 9e7cb000 9dd92140
[   23.995029] df20: 9dc193e0 9dd92148 9dcfdf4c 9dcfdf38 8011022c 800fbe78 8000f9cc 9e687300
[   24.003216] df40: 9dcfdf6c 9dcfdf50 8011f798 8007ffe8 7e87ab70 9dd92140 00000003 9dd92140
[   24.011402] df60: 40186f40 7e87ab70 9dcfc000 00000000 9dcfdfa4 9dcfdf80 80113e30 8011373c
[   24.019588] df80: 7e87ab70 7e87ab70 7e87aea9 00000036 8000fb84 9dcfc000 00000000 9dcfdfa8
[   24.027775] dfa0: 8000f9a0 80113e00 7e87ab70 7e87ab70 00000003 40186f40 7e87ab70 00000000
[   24.035962] dfc0: 7e87ab70 7e87ab70 7e87aea9 00000036 00000000 00000000 76fd1f70 00000000
[   24.044148] dfe0: 76f80f8c 7e87ab28 00009810 76f80fc4 60000010 00000003 00000000 00000000
[   24.052328] Backtrace:
[   24.054806] [<803f9bbc>] (memcpy16_toio) from [<803faeb0>] (mxc_nand_command+0x484/0x640)
[   24.062996] [<803faa2c>] (mxc_nand_command) from [<803f2024>] (nand_write_page+0xa4/0x154)
[   24.071264]  r10:9df42800 r9:00000000 r8:803faa2c r7:00000400 r6:803faa2c r5:9e674810
[   24.079180]  r4:9e674b40
[   24.081738] [<803f1f80>] (nand_write_page) from [<803f3bd8>] (nand_do_write_ops+0x240/0x444)
[   24.090180]  r8:00000400 r7:00000400 r6:9e674810 r5:803f1f80 r4:9e674b40
[   24.096970] [<803f3998>] (nand_do_write_ops) from [<803f3f94>] (nand_write+0x68/0x88)
[   24.104804]  r10:00000400 r9:00000000 r8:00580000 r7:9df42800 r6:9dcfdce8 r5:9dcfdc20
[   24.112719]  r4:9e674810
[   24.115287] [<803f3f2c>] (nand_write) from [<803eb63c>] (part_write+0x48/0x50)
[   24.122514]  r10:9d802000 r9:9dee1000 r8:9df42800 r7:00000000 r6:00000400 r5:00000000
[   24.130429]  r4:00580000
[   24.132989] [<803eb5f4>] (part_write) from [<803e82ac>] (mtd_write+0x88/0xa0)
[   24.140129]  r5:00000000 r4:803eb5f4
[   24.143748] [<803e8224>] (mtd_write) from [<80406048>] (ubi_io_write+0x104/0x65c)
[   24.151235]  r7:00000000 r6:00000400 r5:00000000 r4:9dee0000
[   24.156968] [<80405f44>] (ubi_io_write) from [<804066cc>] (ubi_io_write_ec_hdr+0x12c/0x190)
[   24.165323]  r10:9d802000 r9:9dc8b214 r8:00000000 r7:9d802030 r6:00000000 r5:9dee0000
[   24.173239]  r4:9df42800
[   24.175798] [<804065a0>] (ubi_io_write_ec_hdr) from [<8040a544>] (ubi_early_get_peb+0x110/0x1f0)
[   24.184587]  r6:9dc8b200 r5:9dee0000 r4:9df42800
[   24.189262] [<8040a434>] (ubi_early_get_peb) from [<803fb560>] (create_vtbl+0x7c/0x238)
[   24.197271]  r10:9dee0000 r9:9dc8b200 r8:9dc8b200 r7:9dc8b200 r6:00000080 r5:00000000
[   24.205187]  r4:9df42400
[   24.207746] [<803fb4e4>] (create_vtbl) from [<803fc08c>] (ubi_read_volume_table+0x750/0xa64)
[   24.216187]  r10:8112514c r9:9dc8b200 r8:a0acf000 r7:9dc8b200 r6:00000080 r5:00005600
[   24.224103]  r4:9dee0000
[   24.226662] [<803fb93c>] (ubi_read_volume_table) from [<8040afa0>] (ubi_attach+0x97c/0x152c)
[   24.235103]  r10:8112514c r9:9dc8b200 r8:00000001 r7:9dee0000 r6:00000065 r5:9dee0000
[   24.243018]  r4:00000000
[   24.245579] [<8040a624>] (ubi_attach) from [<803ff1bc>] (ubi_attach_mtd_dev+0x630/0xbac)
[   24.253673]  r10:9e7b5400 r9:00001000 r8:9dee1000 r7:9dee0000 r6:00000000 r5:9e7b5400
[   24.261588]  r4:fffff000
[   24.264148] [<803feb8c>] (ubi_attach_mtd_dev) from [<804008cc>] (ctrl_cdev_ioctl+0xcc/0x1cc)
[   24.272589]  r10:00000000 r9:9dcfc000 r8:00000003 r7:80113e30 r6:9e7b5400 r5:7e87ab70
[   24.280505]  r4:00000000
[   24.283070] [<80400800>] (ctrl_cdev_ioctl) from [<80113b5c>] (do_vfs_ioctl+0x42c/0x6c4)
[   24.291077]  r6:9dd92140 r5:7e87ab70 r4:9dc193e0
[   24.295753] [<80113730>] (do_vfs_ioctl) from [<80113e30>] (SyS_ioctl+0x3c/0x64)
[   24.303066]  r10:00000000 r9:9dcfc000 r8:7e87ab70 r7:40186f40 r6:9dd92140 r5:00000003
[   24.310981]  r4:9dd92140
[   24.313549] [<80113df4>] (SyS_ioctl) from [<8000f9a0>] (ret_fast_syscall+0x0/0x54)
[   24.321123]  r9:9dcfc000 r8:8000fb84 r7:00000036 r6:7e87aea9 r5:7e87ab70 r4:7e87ab70
[   24.328957] Code: e1c300b0 e1510002 e1a03001 1afffff9 (e89da800)
[   24.335066] ---[ end trace ab1cb17887f21bbb ]---
[   24.340249] Unhandled fault: imprecise external abort (0x1c06) at 0x7ee8bcf0
[   24.347310] pgd = 9df3c000
[   24.350023] [7ee8bcf0] *pgd=8dcbf831, *pte=8eb3334f, *ppte=8eb3383f
Segmentation fault

Fixes: 35d5d20efad8 ("mtd: mxc_nand: cleanup copy_spare function")
Signed-off-by: Eric Bénard <eric@eukrea.com>
Reviewed-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Reviewed-by: Baruch Siach <baruch@tkos.co.il>
Cc: <stable@vger.kernel.org>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: sunxi: fix sunxi_nand_chips_cleanup()
Boris BREZILLON [Sun, 13 Sep 2015 16:14:43 +0000 (18:14 +0200)]
mtd: nand: sunxi: fix sunxi_nand_chips_cleanup()

The sunxi_nand_chips_cleanup() function is missing a call to list_del()
which generates a double free error.

Reported-by: Priit Laes <plaes@plaes.org>
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Cc: <stable@vger.kernel.org> # 3.19+
Fixes: 1fef62c1423b ("mtd: nand: add sunxi NAND flash controller support")
Tested-by: Priit Laes <plaes@plaes.org>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agojffs2: drop null test before destroy functions
Julia Lawall [Sun, 13 Sep 2015 12:15:29 +0000 (14:15 +0200)]
jffs2: drop null test before destroy functions

Remove unneeded NULL test.

The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@@ expression x; @@
-if (x != NULL)
  \(kmem_cache_destroy\|mempool_destroy\|dma_pool_destroy\)(x);
// </smpl>

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nandsim: drop null test before destroy functions
Julia Lawall [Sun, 13 Sep 2015 12:14:54 +0000 (14:14 +0200)]
mtd: nandsim: drop null test before destroy functions

Remove unneeded NULL test.

The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@@ expression x; @@
-if (x != NULL)
  \(kmem_cache_destroy\|mempool_destroy\|dma_pool_destroy\)(x);
// </smpl>

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: spi-nor: remove unused read_xfer/write_xfer hooks
Cyrille Pitchen [Fri, 18 Sep 2015 15:49:25 +0000 (17:49 +0200)]
mtd: spi-nor: remove unused read_xfer/write_xfer hooks

struct spi_nor_xfer_cfg and read_xfer/write_xfer hooks were never used by
any driver. Do some cleanup by removing them.

Signed-off-by: Cyrille Pitchen <cyrille.pitchen@atmel.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: add nand_check_erased helper functions
Boris BREZILLON [Thu, 3 Sep 2015 16:03:38 +0000 (18:03 +0200)]
mtd: nand: add nand_check_erased helper functions

Add two helper functions to help NAND controller drivers test whether a
specific NAND region is erased or not.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agomtd: nand: sunxi: fix OOB handling in ->write_xxx() functions
Boris BREZILLON [Mon, 14 Sep 2015 08:41:03 +0000 (10:41 +0200)]
mtd: nand: sunxi: fix OOB handling in ->write_xxx() functions

The USER_DATA register cannot be accessed using byte accessors on A13
SoCs, thus triggering a bug when using memcpy_toio on this register.
Declare an helper macros to convert an OOB buffer into a suitable
USER_DATA value and vice-versa.

This patch also fixes an error in the oob_required logic (some OOB data
are not written even if the user required it) by removing the
oob_required condition, which is perfectly valid since the core already
fill ->oob_poi with FFs when oob_required is false.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Cc: <stable@vger.kernel.org> # 3.19+
Fixes: 1fef62c1423b ("mtd: nand: add sunxi NAND flash controller support")
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
9 years agoMerge tag 'v4.3-rc1' into MTD -next development
Brian Norris [Mon, 14 Sep 2015 01:41:29 +0000 (18:41 -0700)]
Merge tag 'v4.3-rc1' into MTD -next development

9 years agoLinux 4.3-rc1
Linus Torvalds [Sat, 12 Sep 2015 23:35:56 +0000 (16:35 -0700)]
Linux 4.3-rc1

9 years agoMerge tag 'cris-for-4.3' of git://git.kernel.org/pub/scm/linux/kernel/git/jesper...
Linus Torvalds [Sat, 12 Sep 2015 19:24:29 +0000 (12:24 -0700)]
Merge tag 'cris-for-4.3' of git://git./linux/kernel/git/jesper/cris

Pull CRIS updates from Jesper Nilsson:
 "Mostly removal of old cruft of which we can use a generic version, or
  fixes for code not commonly run in the cris port, but also additions
  to enable some good debug"

* tag 'cris-for-4.3' of git://git.kernel.org/pub/scm/linux/kernel/git/jesper/cris: (25 commits)
  CRISv10: delete unused lib/dmacopy.c
  CRISv10: delete unused lib/old_checksum.c
  CRIS: fix switch_mm() lockdep splat
  CRISv32: enable LOCKDEP_SUPPORT
  CRIS: add STACKTRACE_SUPPORT
  CRISv32: annotate irq enable in idle loop
  CRISv32: add support for irqflags tracing
  CRIS: UAPI: use generic types.h
  CRIS: UAPI: use generic shmbuf.h
  CRIS: UAPI: use generic msgbuf.h
  CRIS: UAPI: use generic socket.h
  CRIS: UAPI: use generic sembuf.h
  CRIS: UAPI: use generic sockios.h
  CRIS: UAPI: use generic auxvec.h
  CRIS: UAPI: use generic headers via Kbuild
  CRIS: UAPI: fix elf.h export
  CRIS: don't make asm/elf.h depend on asm/user.h
  CRIS: UAPI: fix ptrace.h
  CRISv32: Squash compile warnings for axisflashmap
  CRISv32: Add GPIO driver to the default configs
  ...

9 years agoblk: rq_data_dir() should not return a boolean
Linus Torvalds [Wed, 27 May 2015 22:32:15 +0000 (15:32 -0700)]
blk: rq_data_dir() should not return a boolean

rq_data_dir() returns either READ or WRITE (0 == READ, 1 == WRITE), not
a boolean value.

Now, admittedly the "!= 0" doesn't really change the value (0 stays as
zero, 1 stays as one), but it's not only redundant, it confuses gcc, and
causes gcc to warn about the construct

    switch (rq_data_dir(req)) {
        case READ:
            ...
        case WRITE:
            ...

that we have in a few drivers.

Now, the gcc warning is silly and stupid (it seems to warn not about the
switch value having a different type from the case statements, but about
_any_ boolean switch value), but in this case the code itself is silly
and stupid too, so let's just change it, and get rid of warnings like
this:

  drivers/block/hd.c: In function ‘hd_request’:
  drivers/block/hd.c:630:11: warning: switch condition has boolean value [-Wswitch-bool]
     switch (rq_data_dir(req)) {

The odd '!= 0' came in when "cmd_flags" got turned into a "u64" in
commit 5953316dbf90 ("block: make rq->cmd_flags be 64-bit") and is
presumably because the old code (that just did a logical 'and' with 1)
would then end up making the type of rq_data_dir() be u64 too.

But if we want to retain the old regular integer type, let's just cast
the result to 'int' rather than use that rather odd '!= 0'.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoMerge branch 'writeback-plugging'
Linus Torvalds [Sat, 12 Sep 2015 18:19:01 +0000 (11:19 -0700)]
Merge branch 'writeback-plugging'

Fix up the writeback plugging introduced in commit d353d7587d02
("writeback: plug writeback at a high level") that then caused problems
due to the unplug happening with a spinlock held.

* writeback-plugging:
  writeback: plug writeback in wb_writeback() and writeback_inodes_wb()
  Revert "writeback: plug writeback at a high level"

9 years agowriteback: plug writeback in wb_writeback() and writeback_inodes_wb()
Linus Torvalds [Fri, 11 Sep 2015 20:37:19 +0000 (13:37 -0700)]
writeback: plug writeback in wb_writeback() and writeback_inodes_wb()

We had to revert the pluggin in writeback_sb_inodes() because the
wb->list_lock is held, but we could easily plug at a higher level before
taking that lock, and unplug after releasing it.  This does that.

Chris will run performance numbers, just to verify that this approach is
comparable to the alternative (we could just drop and re-take the lock
around the blk_finish_plug() rather than these two commits.

I'd have preferred waiting for actual performance numbers before picking
one approach over the other, but I don't want to release rc1 with the
known "sleeping function called from invalid context" issue, so I'll
pick this cleanup version for now.  But if the numbers show that we
really want to plug just at the writeback_sb_inodes() level, and we
should just play ugly games with the spinlock, we'll switch to that.

Cc: Chris Mason <clm@fb.com>
Cc: Josef Bacik <jbacik@fb.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Neil Brown <neilb@suse.de>
Cc: Jan Kara <jack@suse.cz>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agothermal: fix intel PCH thermal driver mismerge
Linus Torvalds [Sat, 12 Sep 2015 03:06:59 +0000 (20:06 -0700)]
thermal: fix intel PCH thermal driver mismerge

I didn't notice this when merging the thermal code from Zhang, but his
merge (commit 5a924a07f882: "Merge branches 'thermal-core' and
'thermal-intel' of .git into next") of the thermal-core and
thermal-intel branches was wrong.

In thermal-core, commit 17e8351a7739 ("thermal: consistently use int for
temperatures") converted the thermal layer to use "int" for
temperatures.

But in parallel, in the thermal-intel branch commit d0a12625d2ff
("thermal: Add Intel PCH thermal driver") added support for the intel
PCH thermal sensor using the old interfaces that used "unsigned long"
pointers.

This resulted in warnings like this:

  drivers/thermal/intel_pch_thermal.c:184:14: warning: initialization from incompatible pointer type [-Wincompatible-pointer-types]
    .get_temp = pch_thermal_get_temp,
                ^
  drivers/thermal/intel_pch_thermal.c:184:14: note: (near initialization for ‘tzd_ops.get_temp’)
  drivers/thermal/intel_pch_thermal.c:186:19: warning: initialization from incompatible pointer type [-Wincompatible-pointer-types]
    .get_trip_temp = pch_get_trip_temp,
                     ^
  drivers/thermal/intel_pch_thermal.c:186:19: note: (near initialization for ‘tzd_ops.get_trip_temp’)

This fixes it.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoMerge branch 'akpm' (patches from Andrew)
Linus Torvalds [Sat, 12 Sep 2015 02:34:09 +0000 (19:34 -0700)]
Merge branch 'akpm' (patches from Andrew)

Merge fourth patch-bomb from Andrew Morton:

 - sys_membarier syscall

 - seq_file interface changes

 - a few misc fixups

* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
  revert "ocfs2/dlm: use list_for_each_entry instead of list_for_each"
  mm/early_ioremap: add explicit #include of asm/early_ioremap.h
  fs/seq_file: convert int seq_vprint/seq_printf/etc... returns to void
  selftests: enhance membarrier syscall test
  selftests: add membarrier syscall test
  sys_membarrier(): system-wide memory barrier (generic, x86)
  MODSIGN: fix a compilation warning in extract-cert

9 years agoARCv2: [axs103_smp] Reduce clk for SMP FPGA configs
Vineet Gupta [Fri, 11 Sep 2015 23:32:22 +0000 (16:32 -0700)]
ARCv2: [axs103_smp] Reduce clk for SMP FPGA configs

Newer bitfiles needs the reduced clk even for SMP builds

Cc: <stable@vger.kernel.org> #4.2
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
9 years agoMerge tag 'ntb-4.3' of git://github.com/jonmason/ntb
Linus Torvalds [Sat, 12 Sep 2015 02:29:00 +0000 (19:29 -0700)]
Merge tag 'ntb-4.3' of git://github.com/jonmason/ntb

Pull NTB fixes from Jon Mason:
 "NTB bug and documentation fixes, new device IDs, performance
  improvements, and adding a mailing list to MAINTAINERS for NTB"

* tag 'ntb-4.3' of git://github.com/jonmason/ntb:
  NTB: Fix range check on memory window index
  NTB: Improve index handling in B2B MW workaround
  NTB: Fix documentation for ntb_peer_db_clear.
  NTB: Fix documentation for ntb_link_is_up
  NTB: Use unique DMA channels for TX and RX
  NTB: Remove dma_sync_wait from ntb_async_rx
  NTB: Clean up QP stats info
  NTB: Make the transport list in order of discovery
  NTB: Add PCI Device IDs for Broadwell Xeon
  NTB: Add flow control to the ntb_netdev
  NTB: Add list to MAINTAINERS