- 23 Aug, 2017 34 commits
-
-
Abhishek Sahu authored
In EBI2, all codeword data will be read in FLASH_BUF_ACC buffer and ADM will copy the data from source (FLASH_BUF_ACC) to destination (memory for data read). In QPIC, there is no FLASH_BUF_ACC and all the codeword data will held in QPIC BAM FIFO buffers. It provides multiple READ_LOCATION registers which will be used for copying the data from FIFO to memory. The READ_LOCATION register will be used to read a specific amount of data from a specific offset within the flash buffer. It supports sequential offset requests. Each request is composed of the following fields: a. Offset within the flash buffer from which data should be read b. Amount of data to be read c. Flag bit specifying the last read request from the flash buffer. Following the last read request the NANDc refers to the buffer as empty. Reviewed-by:
Archit Taneja <architt@codeaurora.org> Signed-off-by:
Abhishek Sahu <absahu@codeaurora.org> Signed-off-by:
Boris Brezillon <boris.brezillon@free-electrons.com>
-
Abhishek Sahu authored
The QPIC NAND BAM has multiple flags to control the transfer. This patch adds flags parameter in register and data transfer DMA helper functions and modifies all these functions call with appropriate flags using following rule 1. Read and write can’t go in single command descriptor so separate SGL should be used. 2. For some of the requests, NWD flag should be set in BAM DMA descriptor. 3. For Data write, the BAM has internal buffer for each codeword. All write request will modify the data in internal buffer and this buffer will be flushed to NAND device once EOT flag is set. So for all the write requests in single codeword, the EOT should be cleared for all tx data descriptors except the last one. Reviewed-by:
-
Abhishek Sahu authored
1. prepare_bam_async_desc is the function which will call all the DMA API’s. It will fetch the outstanding scatter gather list for passed channel and will do the DMA descriptor formation. The DMA flag is dependent upon the type of channel. 2. For ADM DMA, the descriptor is being formed for every DMA request so its sgl count will be always 1 while in BAM DMA, the clubbing of descriptor is being done to increase throughput. 3. ADM DMA uses only one channel while in BAM DMA, data descriptors will be submitted to tx channel (for write) or rx channel (for read) and all the registers read/write descriptors in command channel. Reviewed-by:
-
Abhishek Sahu authored
- The BAM transaction is the core data structure which will be used for all the data transfers in QPIC NAND. Since the core framework in nand_base.c is serializing all the NAND requests so allocating BAM transaction before every transfer will be overhead. The memory for it be allocated during probe time and before every transfer, it will be cleared. - The BAM transaction contains the array of command and data scatter gather list and indexes. For every transfer, all the resource will be taken from BAM transaction. - The size of the buffer used for BAM transactions is calculated based on the NAND device with the maximum page size, among all the devices connected to the controller. Reviewed-by:
-
Abhishek Sahu authored
The EBI2 NAND controller directly remaps register read buffer with dma_map_sg and DMA address of this buffer will be passed to DMA API’s. While, on QPIC NAND controller, which uses BAM DMA, we read the controller registers by preparing a BAM command descriptor. This command descriptor requires the - controller register address - the DMA address in which we want to store the value read back from the controller register. This command descriptor will be remapped with dma_map_sg and its DMA address will be passed to DMA API’s. Therefore, it's required that we also map our register read buffer for DMA (using dma_map_single). We use the returned DMA address for preparing entries in our command descriptor. This patch adds the DMA mapping support for register read buffer. This buffer will be DMA mapped during allocation time. Before starting of any operation, this buffer will be synced for device operation and after operation completion, it will be synced again for CPU. Reviewed-by: -
Abhishek Sahu authored
1. QPIC NAND controller uses 3 BAM channels: command, data tx and data rx while EBI2 NAND controller uses only single ADM channel. 2. CRCI is only required for ADM DMA and it's not required for BAM DMA. Reviewed-by:
-
Abhishek Sahu authored
The current driver only supports EBI2 NAND controller which uses ADM DMA. The latest QCOM SoC uses QPIC NAND controller with BAM DMA. NAND registers and programming sequence are same for EBI2 and QPIC NAND so the same driver can support QPIC NAND also by adding the BAM DMA support. This patch adds the is_bam in NAND property which will be checked for determining the DMA engine type. Reviewed-by:
-
Abhishek Sahu authored
Currently driver data is being assigned directly with ECC modes. Now, the plan is to add more NAND controller versions which will have different properties. This patch reorganizes the current driver data assignment by creating NAND controller properties structure which will contain all properties specific to NAND controller. Reviewed-by:
-
Abhishek Sahu authored
The NAND page read fails without complete boot chain since NAND_DEV_CMD_VLD value is not proper. The default power on reset value for this register is 0xe - ERASE_START_VALID | WRITE_START_VALID | READ_STOP_VALID The READ_START_VALID should be enabled for sending PAGE_READ command. READ_STOP_VALID should be cleared since normal NAND page read does not require READ_STOP command. Fixes: c76b78d8 ("mtd: nand: Qualcomm NAND controller driver") Cc: stable@vger.kernel.org Reviewed-by: -
Gustavo A. R. Silva authored
platform_get_irq() returns an error code, but the mtk_ecc driver ignores it and always returns -EINVAL. This is not correct and, prevents -EPROBE_DEFER from being propagated properly. Print and propagate the return value of platform_get_irq on failure. This issue was detected with the help of Coccinelle. Signed-off-by:
Gustavo A. R. Silva <gustavo@embeddedor.com> Signed-off-by:
-
Gustavo A. R. Silva authored
platform_get_irq() returns an error code, but the sh_flctl driver ignores it and always returns -ENXIO. This is not correct and, prevents -EPROBE_DEFER from being propagated properly. Print and propagate the return value of platform_get_irq on failure. This issue was detected with the help of Coccinelle. Signed-off-by:
-
Pan Bian authored
The return value of dma_map_single() should be checked by dma_mapping_error(). However, in function flctl_dma_fifo0_transfer(), its return value is checked against NULL, which could result in failures. Signed-off-by:
Pan Bian <bianpan2016@163.com> Signed-off-by:
-
Sergei Shtylyov authored
of_irq_get() may return 0 as well as negative error number on failure, while the driver only checks for the negative values. The driver would then call devm_request_irq() for IRQ0 in its probe method and never get a valid interrupt. Check for 'nc->irq <= 0' instead and return -ENXIO from the driver's probe if of_irq_get() returned 0. Fixes: f88fc122 ("mtd: nand: Cleanup/rework the atmel_nand driver") Signed-off-by:
Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Acked-by:
Wenyou Yang <Wenyou.yang@microchip.com> Signed-off-by:
-
Martin Blumenstingl authored
According to the datasheet of the H27UCG8T2BTR the NAND Technology field (6th byte of the "Device Identifier Description", bits 0-2) the following values are possible: - 0x0 = 48nm - 0x1 = 41nm - 0x2 = 32nm - 0x3 = 26nm - 0x4 = 20nm - (all others are reserved) Fix this by extending the mask for this field to allow detecting value 0x4 (20nm) as valid NAND technology. Without this the detection of the ECC requirements fails, because the code assumes that the device is a 48nm device (0x4 & 0x3 = 0x0) and aborts with "Invalid ECC requirements" because it cannot map the "ECC Level". Extending the mask makes the ECC requirement detection code recognize this chip as <= 26nm and sets up the ECC step size and ECC strength correctly. Signed-off-by:
Martin Blumenstingl <martin.blumenstingl@googlemail.com> Fixes: 78f3482d ("mtd: nand: hynix: Rework NAND ID decoding to extract more information") Cc: <stable@vger.kernel.org> Signed-off-by:
-
Boris Brezillon authored
Commit a894cf6c ("mtd: nand: mxc: switch to mtd_ooblayout_ops") introduced a bug in the OOB layout description. Even if the driver claims that 3 ECC bytes are reserved to protect 512 bytes of data, it's actually 5 ECC bytes to protect 512+6 bytes of data (some OOB bytes are also protected using extra ECC bytes). Fix the mxc_v1_ooblayout_{free,ecc}() functions to reflect this behavior. Signed-off-by:
-
Philipp Zabel authored
Commit a53e35db ("reset: Ensure drivers are explicit when requesting reset lines") started to transition the reset control request API calls to explicitly state whether the driver needs exclusive or shared reset control behavior. Convert all drivers requesting exclusive resets to the explicit API call so the temporary transition helpers can be removed. No functional changes. Cc: Boris Brezillon <boris.brezillon@free-electrons.com> Cc: Richard Weinberger <richard@nod.at> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Brian Norris <computersforpeace@gmail.com> Cc: Marek Vasut <marek.vasut@gmail.com> Cc: Cyrille Pitchen <cyrille.pitchen@wedev4u.fr> Cc: Maxime Ripard <maxime.ripard@free-electrons.com> Cc: Chen-Yu Tsai <wens@csie.org> Cc: linux-mtd@lists.infradead.org Signed-off-by:
Philipp Zabel <p.zabel@pengutronix.de> Signed-off-by:
-
Arvind Yadav authored
- clk_prepare_enable() can fail here and we must check its return value. - stfsm_probe() can fail here and we must disable clock. Signed-off-by:
Arvind Yadav <arvind.yadav.cs@gmail.com> Signed-off-by:
-
Arvind Yadav authored
clk_prepare_enable() can fail here and we must check its return value. Signed-off-by:
-
Arvind Yadav authored
clk_prepare_enable() can fail here and we must check its return value. Signed-off-by:
-
Arvind Yadav authored
- clk_prepare_enable() can fail here and we must check its return value. - oxnas_nand_probe() can fail here and we must disable clock. Signed-off-by:
Neil Armstrong <narmstrong@baylibre.com> Signed-off-by:
-
Arvind Yadav authored
clk_prepare_enable() can fail here and we must check its return value. Signed-off-by:
-
Arnd Bergmann authored
orion_nand_read_buf uses an inline assembly with the "ldrd" instruction, which is only available from ARMv5 upwards. This used to be fine, since all users have an ARMv5 or ARMv7 CPU, but now we can also build a multiplatform kernel with ARMv4 support enabled in addition to the "kirkwood" (mvebu) platform. This provides an alternative to call the readsl() function that is supposed to have the same effect and is also optimized for performance. I first posted a version of this patch back in 2014, and there was some discussion about it then. This fixes the bugs identified back then and should be a reasonable alternative for the rare corner case. Link: https://patchwork.kernel.org/patch/4144791/ Cc: Jingoo Han <jg1.han@samsung.com> Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
Gregory CLEMENT authored
The controller supported by the pxa3xx_nand driver is also available on the mvebu 64-bit SoCs, such as the Armada 7K and Armada 8K SoCs. This patch updates the Kconfig dependency to allow building the kernel for this SoC family too. Signed-off-by:
Gregory CLEMENT <gregory.clement@free-electrons.com> Signed-off-by:
-
Abhishek Sahu authored
The NAND controller can support multiple NAND devices having different page sizes. Future code will require us to allocate memory based on the maximum number of codewords among all the devices. We reorganize the NAND device probing such that the ONFI parameters are first read for each connected device to identify the maximum number of codewords possible, and only then proceed with MTD device registration (i.e, call nand_scan_tail and mtd_device_register). This is a reorganization of the existing code and will not change any functionality. Signed-off-by:
-
Abhishek Sahu authored
The memset in clear_read_regs is overhead. All the register data will be filled by DMA during NAND operation so making these register variables zero is not required. Signed-off-by:
-
Abhishek Sahu authored
Each NAND page consist of multiple codewords. Following is sequence for NAND page write according to hardware guide. 1. Program Power-up configuration, page row, page column address and flash configuration registers. 2. Write NAND_FLASH_CMD followed by NANC_EXEC_CMD for each codeword. 3. Read NAND_FLASH_STATUS for each codeword. The step 1 should be done once for each page and step 2,3 should be done for each codeword. Currently, all the 3 steps are being done for each codeword which is wrong. Now this patch reorganizes page write functions to configure page specific register once and per codeword specific registers for each NAND ECC step. Signed-off-by:
-
Abhishek Sahu authored
Each NAND page consist of multiple codewords. Following is sequence for NAND page read according to hardware guide. 1. Program Power-up configuration, page row, page column address and flash configuration registers. 2. Write NAND_FLASH_CMD followed by NANC_EXEC_CMD for each codeword. 3. Read NAND_FLASH_STATUS for each codeword. The step 1 should be done once for each page and step 2,3 should be done for each codeword. Currently, all the 3 steps are being done for each codeword which is wrong. Now this patch reorganizes read page functions to configure page specific register once and per codeword specific registers for each NAND ECC step. Signed-off-by:
-
Abhishek Sahu authored
Currently the compatible “qcom,nandcs” is being used for each connected NAND device to support for multiple NAND devices in the same bus. The same thing can be achieved by looking reg property for each sub nodes which contains the chip select number so this patch removes the use of “qcom,nandcs” for specifying NAND device sub nodes. Since there is no user for this driver currently in so changing compatible string is safe. Signed-off-by:
Rob Herring <robh@kernel.org> Signed-off-by:
-
Abhishek Sahu authored
Currently the compatible “qcom,nandcs” is being used for each connected NAND device to support for multiple NAND devices in the same bus. The same thing can be achieved by looking reg property for each sub nodes which contains the chip select number so this patch removes the use of “qcom,nandcs” for specifying NAND device sub nodes. Since there is no user for this driver currently in so changing compatible string is safe. Signed-off-by:
-
Abhishek Sahu authored
The configuration for BCH is not correct in the current driver. The ECC_CFG_ECC_DISABLE bit defines whether to enable or disable the BCH ECC in which 0x1 : BCH_DISABLED 0x0 : BCH_ENABLED But currently host->bch_enabled is being assigned to BCH_DISABLED. Fixes: c76b78d8 ("mtd: nand: Qualcomm NAND controller driver") Cc: stable@vger.kernel.org Signed-off-by:
-
Fabio Estevam authored
pinctrl_pm_select_default_state() is already the default pinctrl state and since pinctrl_pm_select_sleep_state() is not used in this driver, there is no need to explicitly call pinctrl_pm_select_default_state(). Signed-off-by:
Fabio Estevam <fabio.estevam@nxp.com> Signed-off-by:
-
Fabio Estevam authored
clk_prepare_enable() may fail, so we should better check its return value and propagate it in the case of error. Signed-off-by:
-
Jean-Louis Thekekara authored
This commit removes hard-coded '8' used for looping into struct nand_chip.id.data array. NAND_MAX_ID_LEN has been introduced by Artem Bityutskiy in 53552d22 for defining ids length in nand_flash_ids[] list. This commit unifies ids length in nand base driver. Signed-off-by:
Jean-Louis Thekekara <jeanlouis.thekekara@parrot.com> Signed-off-by:
-
Boris Brezillon authored
The nand_scan_ident() function is not expected to allocate resources, and people are usually not calling nand_cleanup() if something fails between nand_scan_ident() and nand_scan_tail(). Move all functions that may allocate resource to the nand_scan_tail() path to prevent such resource leaks. Signed-off-by:
-
- 03 Aug, 2017 1 commit
-
-
Boris Brezillon authored
Commit 7d70f334 ("mtd: nand: add lock/unlock routines") introduced support for the Micron LOCK/UNLOCK commands but no one ever used the nand_lock/unlock() functions. Remove support for these vendor-specific operations from the core. If one ever wants to add them back they should be put in nand_micron.c and mtd->_lock/_unlock should be directly assigned from there instead of exporting the functions. Signed-off-by:
-
- 15 Jul, 2017 5 commits
-
-
Linus Torvalds authored
-
git://git.lwn.net/linuxLinus Torvalds authored
Pull documentation format standardization from Jonathan Corbet: "This series converts a number of top-level documents to the RST format without incorporating them into the Sphinx tree. The hope is to bring some uniformity to kernel documentation and, perhaps more importantly, have our existing docs serve as an example of the desired formatting for those that will be added later. Mauro has gone through and fixed up a lot of top-level documentation files to make them conform to the RST format, but without moving or renaming them in any way. This will help when we incorporate the ones we want to keep into the Sphinx doctree, but the real purpose is to bring a bit of uniformity to our documentation and let the top-level docs serve as examples for those writing new ones" * tag 'standardize-docs' of git://git.lwn.net/linux: (84 commits) docs: kprobes.txt: Fix whitespacing tee.txt: standardize document format cgroup-v2.txt: standardize document format dell_rbu.txt: standardize document format zorro.txt: standardize document format xz.txt: standardize document format xillybus.txt: standardize document format vfio.txt: standardize document format vfio-mediated-device.txt: standardize document format unaligned-memory-access.txt: standardize document format this_cpu_ops.txt: standardize document format svga.txt: standardize document format static-keys.txt: standardize document format smsc_ece1099.txt: standardize document format SM501.txt: standardize document format siphash.txt: standardize document format sgi-ioc4.txt: standardize document format SAK.txt: standardize document format rpmsg.txt: standardize document format robust-futexes.txt: standardize document format ...
-
git://git.kernel.org/pub/scm/linux/kernel/git/tytso/randomLinus Torvalds authored
Pull random updates from Ted Ts'o: "Add wait_for_random_bytes() and get_random_*_wait() functions so that callers can more safely get random bytes if they can block until the CRNG is initialized. Also print a warning if get_random_*() is called before the CRNG is initialized. By default, only one single-line warning will be printed per boot. If CONFIG_WARN_ALL_UNSEEDED_RANDOM is defined, then a warning will be printed for each function which tries to get random bytes before the CRNG is initialized. This can get spammy for certain architecture types, so it is not enabled by default" * tag 'random_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/random: random: reorder READ_ONCE() in get_random_uXX random: suppress spammy warnings about unseeded randomness random: warn when kernel uses unseeded randomness net/route: use get_random_int for random counter net/neighbor: use get_random_u32 for 32-bit hash random rhashtable: use get_random_u32 for hash_rnd ceph: ensure RNG is seeded before using iscsi: ensure RNG is seeded before use cifs: use get_random_u32 for 32-bit lock random random: add get_random_{bytes,u32,u64,int,long,once}_wait family random: add wait_for_random_bytes() API
-
git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfsLinus Torvalds authored
Pull ->s_options removal from Al Viro: "Preparations for fsmount/fsopen stuff (coming next cycle). Everything gets moved to explicit ->show_options(), killing ->s_options off + some cosmetic bits around fs/namespace.c and friends. Basically, the stuff needed to work with fsmount series with minimum of conflicts with other work. It's not strictly required for this merge window, but it would reduce the PITA during the coming cycle, so it would be nice to have those bits and pieces out of the way" * 'work.mount' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: isofs: Fix isofs_show_options() VFS: Kill off s_options and helpers orangefs: Implement show_options 9p: Implement show_options isofs: Implement show_options afs: Implement show_options affs: Implement show_options befs: Implement show_options spufs: Implement show_options bpf: Implement show_options ramfs: Implement show_options pstore: Implement show_options omfs: Implement show_options hugetlbfs: Implement show_options VFS: Don't use save/replace_mount_options if not using generic_show_options VFS: Provide empty name qstr VFS: Make get_filesystem() return the affected filesystem VFS: Clean up whitespace in fs/namespace.c and fs/super.c Provide a function to create a NUL-terminated string from unterminated data
-
git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfsLinus Torvalds authored
Pull more __copy_.._user elimination from Al Viro. * 'work.__copy_to_user' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: drm_dp_aux_dev: switch to read_iter/write_iter
-
