aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* ichspi.c: Check SPI Cycle In-Progress prior start HW SeqSubrata Banik2022-03-091-0/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | As per EDS, SPI controller sets the HSFSTS.bit5 (SCIP) when software sets the Flash Cycle Go (FGO) bit in the Hardware Sequencing Flash Control register. This bit remains set until the cycle completes on the SPI interface. Hardware automatically sets and clears this bit. Software must initiate the next SPI transaction when this bit is 0. Platform Setup: Alder Lake based ChromeOS devices (Brya variants) Replication Steps: Accepting and running firmware Auto Update (AU) on the Brya variants (dogfooder system) is seeing `flashrom` getting timed out. Problem Statement: Evidencing AU (Auto Update) failure while performing firmware update on the Alder Lake based ChromeOS devices. Observation: Based on the initial understanding from the failure log/pattern, it seems like the platform is evidencing multiple `flashrom` access from different source, for example: `futility` accesses flashrom for erase, write and read operation, `crossystem` uses flashrom for updating VBNV, additionally, `set_fw_good` script also uses `crossystem` to update the fw status. Solution: Without this SCIP check being implemented in flashrom, there is no way to ensure multiple instances of flashrom performing different SPI operations are not cancelling each other and running into below error: Erasing and writing flash chip... Timeout error between offset 0x0061c000 and 0x0061c03f (= 0x0061c000 + 63)! FAILED! Uh oh. Erase/write failed. Checking if anything has changed. TEST=Able to flash coreboot image on Alder Lake (Brya variants), Tiger Lake (Volteer variants), Kaby Lake (Eve system), Comet Lake (Hatch variants) and Ivybridge without any failure. Signed-off-by: Subrata Banik <subratabanik@google.com> Change-Id: Ib9265cc20513fd00f32f8fa22e28c312903ca484 Reviewed-on: https://review.coreboot.org/c/flashrom/+/61854 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org>
* dmi: Correctly check for ERROR_PTRNico Huber2022-03-081-1/+1
| | | | | | | | | | | | | For the physmap*() functions, NULL is considered valid return value. Fixes a segmentation fault when DMI tables can't be mapped. Tested on intel/eblake board with broken coreboot. Change-Id: Ic403c2940c2b91acbd113f0acfa3ce9ef6c6bb6c Signed-off-by: Nico Huber <nico.h@gmx.de> Reviewed-on: https://review.coreboot.org/c/flashrom/+/62611 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com>
* ichspi: Add Alder Lake supportEdward O'Callaghan2022-03-075-3/+34
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Does exactly what it says on the tin. BUG=b:220799648 TEST=```localhost ~ # flashrom --flash-name <snip> Found Programmer flash chip "Opaque flash chip" (32768 kB, Programmer-specific) mapped at physical address 0x0000000000000000. vendor="Programmer" name="Opaque flash chip" flashrom -p internal --ifd -i fd -i bios -r /tmp/filename.rom flashrom unknown on Linux 5.15.22 (x86_64) flashrom is free software, get the source code at https://flashrom.org Using clock_gettime for delay loops (clk_id: 1, resolution: 1ns). coreboot table found at 0x768a7000. Found chipset "Intel Alder Lake-N". Enabling flash write... Warning: Setting BIOS Control at 0xdc from 0x8b to 0x89 failed. New value is 0x8b. SPI Configuration is locked down. OK. Found Winbond flash chip "W25Q256JV_M" (32768 kB, Programmer-specific) mapped at physical address 0x0000000000000000. Error accessing W25Q256JV_M, 0x2000000 bytes at 0x00000000fe000000 /dev/mem mmap failed: Resource temporarily unavailable Could not map flash chip W25Q256JV_M at 0x00000000fe000000. Reading ich descriptor... done. Using regions: "bios", "fd". Error accessing W25Q256JV_M, 0x2000000 bytes at 0x00000000fe000000 /dev/mem mmap failed: Resource temporarily unavailable Could not map flash chip W25Q256JV_M at 0x00000000fe000000. Reading flash... done. SUCCESS Also, Reading ich descriptor... Reading 4096 bytes starting at 0x000000. done. Assuming chipset '600 series Alder Point'. Added layout entry 00000000 - 00000fff named fd Added layout entry 00500000 - 01ffffff named bios Added layout entry 00001000 - 004fffff named me ``` Tested on Nivviks/ADL-N and Brya/ADL-P. Change-Id: Ie66cf519df13f3391c41f5016b16a81ef3dfd4bf Signed-off-by: Edward O'Callaghan <quasisec@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/62251 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Subrata Banik <subratabanik@google.com> Reviewed-by: Sam McNally <sammc@google.com>
* pcidev: Move scandev_inclass logic from internal to pcidevEdward O'Callaghan2022-03-046-26/+26
| | | | | | | | | | | | | | | BUG=b:220950271 TEST=```sudo ./flashrom -p internal -r /tmp/bios <snip> Found Programmer flash chip "Opaque flash chip" (16384 kB, Programmer-specific) mapped at physical address 0x0000000000000000. Reading flash... done. ``` Change-Id: I1978e178fb73485f1c5c7e732853522847267cee Signed-off-by: Edward O'Callaghan <quasisec@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/59277 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Anastasia Klimchuk <aklm@chromium.org>
* ft2232_spi.c: Add FTDI FT4233HCharles Parent2022-03-031-0/+5
| | | | | | | | | | | | Change tested to probe, read, write and erase a MT25QU256 through a FT4233H FTDI Change-Id: I73cee8fd2a6613a8fbc26508d99bbe67da2b4f72 Signed-off-by: Charles Parent <charles.parent@orolia2s.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/62480 Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Angel Pons <th3fanbus@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
* pcidev: Move pci_get_dev() logic into canonical placeEdward O'Callaghan2022-03-023-9/+15
| | | | | | | | | | | | | | | BUG=b:220950271 TEST=```sudo ./flashrom -p internal --flash-size <snip> Found Programmer flash chip "Opaque flash chip" (16384 kB, Programmer-specific) mapped at physical address 0x0000000000000000. 16777216 ``` Change-Id: Id9ce055d5e5d347520ec5002b8c6548e60eaa0a7 Signed-off-by: Edward O'Callaghan <quasisec@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/59276 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Subrata Banik <subratabanik@google.com>
* writeprotect: add {get,set}_wp_mode()Nikolai Artemiev2022-03-012-7/+47
| | | | | | | | | | | | | | | BUG=b:195381327,b:153800563 BRANCH=none TEST=flashrom --wp-{enable,disable,status} Change-Id: I7b68e940f0e1359281806c98e1da119b4caf8405 Signed-off-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/58483 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org>
* writeprotect: add set_wp_range()Nikolai Artemiev2022-03-012-4/+36
| | | | | | | | | | | | BUG=b:195381327,b:153800563 BRANCH=none TEST=flashrom --wp-{status,range} Change-Id: I7d26f43fb05c5828b9839bb57a28fa1088dcd9a0 Signed-off-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/58482 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
* libflashrom,writeprotect: add flashrom_wp_get_available_ranges()Nikolai Artemiev2022-03-014-0/+281
| | | | | | | | | | | | | | | | Generate list of available ranges by enumerating all possible values that range bits (BPx, TB, ...) can take and using the chip's range decoding function to get the range that is selected by each one. BUG=b:195381327,b:153800563 BRANCH=none TEST=flashrom --wp-list Change-Id: Id51f038f03305c8536d80313e52f77d27835f34d Signed-off-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/58481 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
* writeprotect: add get_wp_range() for decoding rangesNikolai Artemiev2022-03-011-3/+11
| | | | | | | | | | | | | BUG=b:195381327,b:153800563 BRANCH=none TEST=flashrom --wp-{status,range} at end of patch series Change-Id: I5a1dfcf384166b1bac319d286306747e1dcaa000 Signed-off-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/59183 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
* flashchips,writeprotect_ranges: add range decoding functionNikolai Artemiev2022-03-016-1/+107
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Allow chips to specify functions that map status register bits to protection ranges. These are used to enumerate available ranges and determine the protection state of chips. The patch also adds a range decoding function for the example chips. Many other chips can also be handled by it, though some will require different functions (e.g. MX25L6406 and related chips). Another approach that has been tried in cros flashrom is maintaining tables of range data, but it quickly becomes error prone and hard to validate. Using a function to interpret the ranges allows compact encoding with most chips and is flexible enough to allow chips with less predictable ranges to be handled as well. BUG=b:195381327,b:153800563 BRANCH=none TEST=dumped range tables, checked against datasheets Change-Id: Id163ed80938a946a502ed116e48e8236e36eb203 Signed-off-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/58480 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org> Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Anastasia Klimchuk <aklm@chromium.org>
* libflashrom,writeprotect: add functions for reading/writing WP configsNikolai Artemiev2022-03-014-0/+363
| | | | | | | | | | | | | | | | | | | | | | | New functions are exposed through the libflashrom API for reading/writing chip's WP settins: `flashrom_wp_{read,write}_cfg()`. They read/write an opaque `struct flashrom_wp_cfg` instance, which includes the flash protection range and status register protection mode. This commit also adds `{read,write}_wp_bits()` helper functions that read/write chip-specific WP configuration bits. BUG=b:195381327,b:153800563 BRANCH=none TEST=flashrom --wp-{enable,disable,range,list,status} at end of patch series Change-Id: I3ad25708c3321b8fb0216c3eaf6ffc07616537ad Signed-off-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/58479 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
* writeprotect.h: add structure to represent chip wp configuration bitsNikolai Artemiev2022-03-011-0/+36
| | | | | | | | | | | | | | | | | | | | | | | Add `struct wp_bits` for representing values of all WP bits in a chip's status/config register(s). It allows most WP code to store and manipulate a chip's configuration without knowing the exact layout of bits in the chip's status registers. Supporting other chips may require additional fields to be added to the structure. BUG=b:195381327,b:153800563 BRANCH=none TEST=flashrom --wp-{enable,disable,range,list,status} at end of patch series Change-Id: I17dee630248ce7b51e624a6e46d7097d5d0de809 Signed-off-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/58478 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
* flash.h,flashchips.c: add writeprotect bit layout map to chipsNikolai Artemiev2022-03-013-0/+77
| | | | | | | | | | | | | | | | | | | | | | | | This patch adds a register bit map `struct reg_bit_info`, with fields for storing the register, bit index, and writability of each bit that affects the chip's write protection. This allows writeprotect code to be independent of the register layout of any specific chip. The new fields have been filled out for example chips. The representation is centered around describing how bits can be accessed and modified, rather than the layout of registers. This is generally easier to work with in code that needs to access specific bits and typically requires specifying the locations of fewer bits overall. BUG=b:195381327,b:153800563 BRANCH=none TEST=flashrom --wp-{enable,disable,range,list,status} at end of patch series Change-Id: Id08d77e6d4ca5109c0d698271146d026dbc21284 Signed-off-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/58477 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Anastasia Klimchuk <aklm@chromium.org>
* spi25_statusreg,flashchips: add SR2 read/write supportNikolai Artemiev2022-02-284-3/+53
| | | | | | | | | | | | | | | | | | | | | This patch adds support for reading and writing the second status register and enables it on a limited set of flash chips. Chip support for RDSR2/WRSR2/extended WRSR is represented using feature flags to be consistent with how other SPI capabilities are represented. BUG=b:195381327,b:153800563 BRANCH=none TEST=flashrom -{r,w,E} TEST=flashrom --wp-{enable,disable,range,list,status} at end of patch series TEST=logged SR2 read/write values during wp commands Change-Id: I34a503b0958e8f2f22a2a993a6ea529eb46b41db Signed-off-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/58570 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
* spi25_statusreg: make register read/write functions genericNikolai Artemiev2022-02-283-39/+95
| | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch adds new spi_{read,write}_register() functions that take the source/destination register as an argument. Currently they can only access SR1, support for other registers will be added in another patch. Since we're refactoring things, this commit also makes spi_read_register() return an error code, making it possible to identify error conditions that spi_read_status_register() concealed. This also removes the initial 100ms delay between writing a register and the first attempt to check the chip's status. An initial delay was added to avoid needing to read the status register multiple times, but that is unlikely to cause problems on modern flash chips. BUG=b:195381327,b:153800563 BRANCH=none TEST=flashrom -{r,w,E} TEST=flashrom --wp-{enable,disable,range,list,status} at end of patch series Change-Id: I0a3951bbf993f2d8d830143b29d3ce16cc6901d7 Signed-off-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/58475 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org>
* writeprotect, cli_classic: delete old writeprotect codeNikolai Artemiev2022-02-285-571/+7
| | | | | | | | | | | | | | | | | | | | | Delete writeprotect code that was previously extracted from the cros tree. This is the first of a series of commits adding writeprotect support. Following commits incrementally implement writeprotect operations, culminating in writeprotect support for three example chips: GD25LQ128, GD25Q32, and GD25Q256. BUG=b:195381327,b:153800563 BRANCH=none TEST=flashrom -{r,w,E} TEST=flashrom --wp-{enable,disable,range,list,status} at end of patch series Change-Id: I67e9b31f86465e5a8f7d3def637198671ee818a8 Signed-off-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/58474 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org> Reviewed-by: Anastasia Klimchuk <aklm@chromium.org>
* libflashrom: Allow NULL-pointer argument in flashrom_flash_release()Nico Huber2022-02-281-0/+3
| | | | | | | | | | | | | | free() allows NULL and it makes error paths easier to handle when one just needs to write `free(x);` without needing to care if `x` was allocated already. Let's follow this rule in flashrom_flash_release(). flashrom_layout_release() already checks for NULL. Change-Id: Id119c2e4f3aa1b11313059f11aac73c3e583185c Signed-off-by: Nico Huber <nico.h@gmx.de> Reviewed-on: https://review.coreboot.org/c/flashrom/+/62340 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-by: Felix Singer <felixsinger@posteo.net>
* hwaccess_x86_io: merge error message in rget_io_perms()Thomas Heijligen2022-02-271-5/+2
| | | | | | | | | | | Have the same error message on all platforms. Change-Id: I7aae096deedd9b78f5fd38a73390cd8a33528545 Signed-off-by: Thomas Heijligen <thomas.heijligen@secunet.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61966 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Nico Huber <nico.h@gmx.de>
* meson.build: Don't depend on getrevision.shSamuel R. Messner2022-02-251-2/+2
| | | | | | | | | | | | | | | | `getrevision.sh` isn't included in exported source code (including GitHub's auto-generated tarballs and ZIPs). Per issue #95, the build shouldn't depend on getrevision.sh for this reason. Previously, however, Flashrom would not build from exported source using Meson due to it requiring `getrevision.sh`. This patch has Meson use the intended `getversion.sh` instead. Signed-off-by: Samuel R. Messner <powpowd@gmail.com> Change-Id: Id8601155b35f0299200c27d0278606127410ff16 Reviewed-on: https://review.coreboot.org/c/flashrom/+/62061 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Nico Huber <nico.h@gmx.de> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
* pcidev: Avoid internal programmer relying on pacc globalEdward O'Callaghan2022-02-243-23/+25
| | | | | | | | | | | | | | | | | | | Make progress towards the goal of removing pacc from global state as noted in the FIXME of programmer.h BUG=b:220950271 TEST=```sudo ./flashrom -p internal --flash-size <snip> Found Programmer flash chip "Opaque flash chip" (16384 kB, Programmer-specific) mapped at physical address 0x0000000000000000. 16777216 ``` Change-Id: Id83bfd41f785f907e52a65a6689e8c7016fc1b77 Signed-off-by: Edward O'Callaghan <quasisec@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/59275 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Anastasia Klimchuk <aklm@chromium.org>
* cli_classic: Use flashrom_init() from APIEdward O'Callaghan2022-02-241-4/+2
| | | | | | | | | | | | | | | | | | | | | flashrom_init() already does delay calibration and self-checking via the canonical libflashrom API. Port the cli implementation to go via the libflashrom API entry-point natively. BUG=b:208132085 TEST=```sudo ./flashrom -p internal --flash-size <snip> Found Programmer flash chip "Opaque flash chip" (16384 kB, Programmer-specific) mapped at physical address 0x0000000000000000. 16777216 ``` Change-Id: I07faeed876f678c35355621a080c7852eed16824 Signed-off-by: Edward O'Callaghan <quasisec@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61582 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com>
* Makefile: Rework the EXEC_SUFFIX determinationThomas Heijligen2022-02-231-2/+1
| | | | | | | | | | | | | Use a conditional function for the statement. This limits the decision to one line instead of multiple places. Change-Id: Iee66dbc609bd5c6eb9d04b457f4508911b2e6560 Signed-off-by: Thomas Heijligen <thomas.heijligen@secunet.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/62195 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Felix Singer <felixsinger@posteo.net>
* cli_classic: Replace programmer_shutdown() with libflashrom callEdward O'Callaghan2022-02-221-1/+1
| | | | | | | | | | | | | | | | flashrom_programmer_shutdown(NULL) is an equiv call to programmer_shutdown() however this further decouples cli from flashrom core logic at link-time, prefering to instead enter via libflashrom instead. BUG=none TEST=`make`. Change-Id: Ie194fa2e891797a29d05d7e9d0c7226fd62c0679 Signed-off-by: Edward O'Callaghan <quasisec@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61584 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com>
* cli_classic: Use flashrom_flash_getsize() over raw accessEdward O'Callaghan2022-02-221-5/+5
| | | | | | | | | | | | | | | Use native libflashrom entry-points for getting flash size over accessing internal struct members within cli code. BUG=none TEST=`sudo ./flashrom -p internal --get-size`. Change-Id: Iff70a0de1dee517c4e38c45b1540c42326a13340 Signed-off-by: Edward O'Callaghan <quasisec@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61583 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com>
* Makefile: enable internal programmer for x86 and all linux platformsThomas Heijligen2022-02-211-12/+20
| | | | | | | | | | | | | | The internal programmer has platform independent code for x86 and linux based code for mipsel. Furthermore the internal programmer can call the linux mtd programmer when available. Enable the internal programmer on x86 or linux. Change-Id: Ia607ea60c3d7d15fe231fa412595992dadc535ad Signed-off-by: Thomas Heijligen <thomas.heijligen@secunet.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61300 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
* libflashrom/fmap: Don't use off_t for flash offsetsJulius Werner2022-02-203-6/+6
| | | | | | | | | | | | | | | | | | | | | | | | | off_t is a special POSIX type that is used to represent file offsets in certain APIs (e.g. lseek(), mmap()), and should not be reused to represent anything else (such as flash offsets). In particular, the width of the type may change based on the definition of the _FILE_OFFSET_BITS macro. Using such a type at the libflashrom interface is particularly dangerous, because if a program is built with a different _FILE_OFFSET_BITS value than libflashrom, the resulting ABI corruption will cause very very nasty and confusing bugs. This patch replaces all instances of off_t that are not related to file offsets with (s)size_t. BUG=b:219811851 TEST=`elogtool list` on cherry. Signed-off-by: Julius Werner <jwerner@chromium.org> Change-Id: I68a386973f79ea634f63dfcd7d95a63400e1fdee Reviewed-on: https://review.coreboot.org/c/flashrom/+/61943 Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org> Reviewed-by: Nikolai Artemiev <nartemiev@google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
* tests: Add tests for verify operationAnastasia Klimchuk2022-02-183-0/+100
| | | | | | | | | | | | | | | This patch adds two tests which cover verify operation, and adds io_mock for fread. BUG=b:181803212 TEST=ninja test Change-Id: I1cc6f73f9b1e385eb963adccf20759c13a40ed3b Signed-off-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-on: https://review.coreboot.org/c/flashrom/+/59239 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org> Reviewed-by: Felix Singer <felixsinger@posteo.net>
* internal.c: Seperate out get_params() from internal_init()Edward O'Callaghan2022-02-171-15/+35
| | | | | | | | | | | | | | | | Move all programmer parameter wrangling into its own function. BUG=none TEST=`flashrom -p internal` on dooly DUT. Change-Id: I66bb370eb0466c5c838621762a6ba825c44567d4 Tested-by: Nikolai Artemiev <nartemiev@google.com> Signed-off-by: Edward O'Callaghan <quasisec@google.com> Signed-off-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61578 Reviewed-by: Anastasia Klimchuk <aklm@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
* Makefile: remove CPPFLAGS and LDFLAGS to default directoriesThomas Heijligen2022-02-161-15/+0
| | | | | | | | | | All CPPFLAGS and LDFLAGS for dependencies are handled by pkg-config Change-Id: Ib7c11a0c8a7918562256480c4be0c95355f981c5 Signed-off-by: Thomas Heijligen <thomas.heijligen@secunet.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61526 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
* ich_descriptors.c Invert the meaning of 'dual_output' bitWerner Zeh2022-02-161-1/+1
| | | | | | | | | | | | | | | | | | In the Flash Component description register (FLCOMP) bit 30 reports the capability of using dual output for fast read operation on the flash component. According to various SPI Programming Guides (checked for Panther Point, Lewisburg C620, Apollo Lake and Elkhart Lake) the dual output is enabled when this bit is set and disabled if not. Currently the logic displays it the other way around when parsing the descriptor. This patch changes this so now if bit 30 in FLCOMP is not set, dual read support for fast read operation is shown as disabled. Change-Id: If6282ac8326ab0b92e9c70c09dba0299bf0deb6f Signed-off-by: Werner Zeh <werner.zeh@siemens.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61362 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com>
* platform/pci.h: adapt NetBSD include path to pkg-configThomas Heijligen2022-02-151-1/+1
| | | | | | | | | | | | The pkg-config include path resolves the pciutils directory. So the include is only the pci.h file. Change-Id: I69dc8184d1d012fb695770cbf6f7c64e5a024453 Signed-off-by: Thomas Heijligen <thomas.heijligen@secunet.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61525 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Angel Pons <th3fanbus@gmail.com>
* Makefile: use pkg-config --libs --static for LDFLAGSThomas Heijligen2022-02-152-2/+1
| | | | | | | | | | | | | | The --static flag of pkg-config returns also the LDFLAGS which are required to link the library static. Use this flag to successfully link against static libraries when the shared variant is not available. This is the case in OpenBSD with libpci. Change-Id: I6029a096c1ceca625789d18c88119d912d79bc0e Signed-off-by: Thomas Heijligen <thomas.heijligen@secunet.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61524 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Angel Pons <th3fanbus@gmail.com>
* Makefile: check if librt is an external libraryThomas Heijligen2022-02-152-0/+8
| | | | | | | | | | | | Some systems, e.g. OpenBSD, have clock_gettime / librt build into the libc and therefore fail to link against it with -lrt. Thus, detect this and link only if needed. Change-Id: I2c1668a350aa0806fccfb4e9cd8b04861f085ee9 Signed-off-by: Thomas Heijligen <thomas.heijligen@secunet.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61523 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com>
* tests: Add comprehensive comment for chip.cAnastasia Klimchuk2022-02-141-0/+12
| | | | | | | | | | | | | | | The following describes the two mechanisms of testing done for flash chip operations. BUG=b:181803212 TEST=ninja test Change-Id: Ie498ec55cce8460fc0b2e1fe27254d3a9f763fac Signed-off-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-on: https://review.coreboot.org/c/flashrom/+/59238 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Felix Singer <felixsinger@posteo.net> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org>
* tests: Set up mock chip memory in consistent and predictable wayAnastasia Klimchuk2022-02-141-0/+2
| | | | | | | | | | | | | | | | | | | | | This patch adds a macro MOCK_CHIP_CONTENT which represents a memory state of a mock chip. The macro is used to initialise mock chip memory at the beginning of a test (in setup_chip function). Previously mock chip memory was not reset between tests. For existing tests that did not matter, however new test for verify operation (added later in this chain) needs mock chip memory to be setup in a predictable way. BUG=b:181803212 TEST=ninja test Change-Id: I0d7623a601c207bfc62d54ab89d94cda56d85871 Signed-off-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-on: https://review.coreboot.org/c/flashrom/+/59237 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Felix Singer <felixsinger@posteo.net> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org>
* tests: add fprintf() mock with redirection to io_mockNikolai Artemiev2022-02-143-0/+16
| | | | | | | | | | | | | | | | | | | We've seen somewhat obscure test failures where the real fprintf() function was passed a fake file returned by the fopen() mock. Although the code that caused the specific failure was cros-specific, adding an fprintf() mock should help avoid future debugging. TEST=ninja test BRANCH=none BUG=b:217661133 Change-Id: I3f8594ea24d17436a7932732d9d05416b804dc93 Signed-off-by: Nikolai Artemiev <nartemiev@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61708 Reviewed-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
* Add Elkhart Lake supportWerner Zeh2022-02-045-6/+47
| | | | | | | | | | | | | | | | Elkhart Lake has a chipset called Mule Creek Canyon which is quite compatible with 300 series chipsets. There are a few differences though, e.g. different encoding for the SPI clock values for read and write in the FLCOMP register. In addition Elkhart Lake has a new PCI device ID for the SPI controller which is added, too. TEST=Read and flash complete flash on Siemens MC EHL1 Change-Id: I711e39a3ec9cd7098389231eaa1cb864d615a475 Signed-off-by: Werner Zeh <werner.zeh@siemens.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/60711 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
* fmap.c: Avoid undefined behaviour with fmap_lsearch([len:=0])Edward O'Callaghan2022-02-031-0/+3
| | | | | | | | | | | | | | | | | | Calling libflashrom entry-points that internally dispatch to fmap_lsearch() can result in a integer overflow. Therefore validate the length paramter before attempting to use it. BUG=none TEST=`make` Change-Id: Ifb408c55c3b69ddff453dcc704b7389298050473 Signed-off-by: Edward O'Callaghan <quasisec@google.com> Spotted-by: Julius Werner <jwerner@chromium.org> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61545 Reviewed-by: Julius Werner <jwerner@chromium.org> Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Angel Pons <th3fanbus@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
* ichspi: Remove unneeded line breaks, add useful line breaks and tabsAnastasia Klimchuk2022-02-021-58/+36
| | | | | | | | | | | | | | | | BUG=b:204488958 TEST=Check that the following scenarios still behave properly: 1) probe-read-verify-erase section-write-reboot on Intel octopus board with GD25LQ128C/GD25LQ128D/GD25LQ128E 2) flashrom binary built before and after this patch with command `make clean && make CONFIG_EVERYTHING=yes VERSION=none` is the same Change-Id: I7ca2902b7caaa95418b828b068c661afafdcd171 Signed-off-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-on: https://review.coreboot.org/c/flashrom/+/60272 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
* ichspi: Extract initialisation of swseq and hwseq into a functionAnastasia Klimchuk2022-02-021-34/+41
| | | | | | | | | | | | | | | | | | | | | | | | Initialisation of swseq_data and hwseq_data gets its own function, which is called from init_ich_default. This makes init_ich_default more readable. This patch also gives a name to (previously anonymous) struct swseq_data. Its sibling struct hwseq_data already has a name. Structs need names to be able to declare function parameters. BUG=b:204488958 TEST=Check that the following scenarios still behave properly: 1) probe-read-verify-erase section-write-reboot on Intel octopus board with GD25LQ128C/GD25LQ128D/GD25LQ128E 2) probe and read on Panther Point (7 series PCH) Change-Id: I7d62b1b380e497b82dcae1284d752204cc541bd3 Tested-by: Anastasia Klimchuk <aklm@chromium.org> Tested-by: Nico Huber <nico.h@gmx.de> Signed-off-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-on: https://review.coreboot.org/c/flashrom/+/58737 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
* ichspi: Extract handling programmer param into a functionAnastasia Klimchuk2022-02-021-27/+37
| | | | | | | | | | | | | | | | | | | | | Extract processing of ich_spi_mode into a separate function which is called from init_ich_default. This makes init_ich_default more readable and avoids one local variable. BUG=b:204488958 TEST=Check that the following scenarios still behave properly: 1) probe-read-verify-erase section-write-reboot on Intel octopus board with GD25LQ128C/GD25LQ128D/GD25LQ128E 2) probe and read on Panther Point (7 series PCH) Change-Id: I20e2379a6fd58c9346f0a2d6daf2b8decf1f6976 Tested-by: Anastasia Klimchuk <aklm@chromium.org> Tested-by: Nico Huber <nico.h@gmx.de> Signed-off-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-on: https://review.coreboot.org/c/flashrom/+/58736 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Nico Huber <nico.h@gmx.de>
* ichspi: Split very long init function into twoAnastasia Klimchuk2022-02-021-242/+255
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ich_init_spi is very long, but logically it can be split. Init function detects the chipset and then the rest of operations depends on the chipset. Init function is more readable now, it consists of only a switch. Initialisation of hwseq and swseq that used to happen in the beginning of init function now moved to init_ich_default, because hwseq and swseq are only used for chipsets served by init_ich_default. BUG=b:204488958 TEST=Check that the following scenarios still behave properly: 1) probe-read-verify-erase section-write-reboot on Intel octopus board with GD25LQ128C/GD25LQ128D/GD25LQ128E 2) probe and read on Panther Point (7 series PCH) 3) on machine with ich7 chipset, output from probe and read is the same between master and this patch Change-Id: I6789bc456a4878e6555831ae0b80ecbdbf62938b Signed-off-by: Anastasia Klimchuk <aklm@chromium.org> Tested-by: Anastasia Klimchuk <aklm@chromium.org> Tested-by: Nico Huber <nico.h@gmx.de> Tested-by: Nicholas Chin <nic.c3.14@gmail.com> Signed-off-by: Anastasia Klimchuk <aklm@chromium.org> Reviewed-on: https://review.coreboot.org/c/flashrom/+/58735 Reviewed-by: Nico Huber <nico.h@gmx.de> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
* Add mediatek_i2c_spi interfaceNeill Corlett2022-02-016-0/+531
| | | | | | | | | | | | | | | Add a spi_master interface supporting MediaTek MST9U ISP mode. Autodetect the bus type via I2C_FUNC_I2C, and use the appropriate read/write commands, in case the MST9U is attached to smbus. TEST=Successfully programmed SPI on test hardware. Change-Id: I24adb14e7b4f7160e1c3ff941774064d5a81e820 Signed-off-by: Neill Corlett <corlett@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61288 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org>
* flashrom.c: Move do_*() helpers into cli_classic.cEdward O'Callaghan2022-01-273-101/+97
| | | | | | | | | | | | | | | These helpers are only used by the CLI logic and so we localise them here to move towards cli_classic being a pure libflashrom user. BUG=b:208132085 TEST=`make` Change-Id: If1112155e2421e0178fd73f847cbb80868387433 Signed-off-by: Edward O'Callaghan <quasisec@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/60070 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Anastasia Klimchuk <aklm@chromium.org>
* flashrom.c: Make {read,write}_buf_from_include_args() publicEdward O'Callaghan2022-01-272-2/+4
| | | | | | | | | | | | | | | This allows for moving all the do_*() helper functions use for the cli from flashrom.c within static local functions in cli_classic.c BUG=b:208132085 TEST=`make` Change-Id: Ia0abec655a682ca449d0e8ba620886a2d616b86d Signed-off-by: Edward O'Callaghan <quasisec@google.com> Reviewed-on: https://review.coreboot.org/c/flashrom/+/60069 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Anastasia Klimchuk <aklm@chromium.org>
* SFDP: make mandatory table length check work with newer SFDP revisionsMichael Niewöhner2022-01-261-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The JEDEC SFDP specification JESD216A (1.5) adds five new DWORDs to the Basic Flash Parameter Table. Later versions of the spec add even more fields. This increases the table being read from 36 bytes to currently 64 bytes and makes flashrom bail out for any SFDP version >= 1.5 due to a static table length check. This was discovered on a GigaDevice GD25B127DSIGR from 2021 with SFDP revision 1.6, while another flash of the same model from 2020 with SFDP revision 1.0 was detected fine by flashrom. GD25B127DSIGR - 2020 version: Probing for Unknown SFDP-capable chip, 0 kB: SFDP revision = 1.0 SFDP number of parameter headers is 2 (NPH = 1). SFDP parameter table header 0/1: ID 0x00, version 1.0 Length 36 B, Parameter Table Pointer 0x000030 GD25B127DSIGR - 2021 version: Probing for Unknown SFDP-capable chip, 0 kB: SFDP revision = 1.6 SFDP number of parameter headers is 2 (NPH = 1). SFDP parameter table header 0/1: ID 0x00, version 1.6 Length 64 B, Parameter Table Pointer 0x000030 ... Length of the mandatory JEDEC SFDP parameter table is wrong (64 B), skipping it. The specification says that changes of the minor SFDP revision will maintain compatibility. Thus, simply check for the minimal required table length, which is 16 bytes for legacy Intel pre-SFDP and 36 bytes for SFDP. Change-Id: Id84cde4ebc805d68e2984e8041fbc48d7ceebe34 Signed-off-by: Michael Niewöhner <foss@mniewoehner.de> Reviewed-on: https://review.coreboot.org/c/flashrom/+/60055 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
* sfdp: drop redundant check of the mandatory table sizeMichael Niewöhner2022-01-261-4/+0
| | | | | | | | Change-Id: I464856612a6d21c682f1d9ad5110fa11a0a276c2 Signed-off-by: Michael Niewöhner <foss@mniewoehner.de> Reviewed-on: https://review.coreboot.org/c/flashrom/+/61379 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nico Huber <nico.h@gmx.de>
* meson: sync programmer dependencies from MakefilePeter Marheine2022-01-251-44/+65
| | | | | | | | | | | | | | | | | | | | | | The Makefile recently gained finer-grained programmer dependency lists allowing it to track which enabled programmers assume various things about the system, like availability of libraries or the CPU architecture. This change implements the same changes in the Meson configuration file. This fixes a number of programmers to correctly build on non-x86 systems, because they were previously misclassified as dependent on x86 architectural features but actually only used PCI. BUG=none TEST=meson build succeeds on both x86 and ARM Signed-off-by: Peter Marheine <pmarheine@chromium.org> Change-Id: Iae93111fd48865f3fe8dd0eb637349b9a0c4affc Reviewed-on: https://review.coreboot.org/c/flashrom/+/61287 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Edward O'Callaghan <quasisec@chromium.org> Reviewed-by: Thomas Heijligen <src@posteo.de>
* hwaccess: fix build on non-x86 targetsPeter Marheine2022-01-254-11/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | The changes to hwaccess in commit 49d758698a0dd166679c48b1a2785e50e9b0cc83 cause build failure on non-x86 systems because the hwaccess_x86_* headers are included in some files that are built for all platforms (particularly those in the internal programmer) and those headers in turn include <sys/io.h> which only exists on x86. This change avoids including those headers on non-x86 platforms so the internal programmer can be built without errors. The comment on the stub implementation of rget_io_perms() is also modified to remove references to non-x86 platforms, since that file is only built on x86 now. BUG=None TEST=meson build succeeds for both x86 and ARM targets Signed-off-by: Peter Marheine <pmarheine@chromium.org> Change-Id: I20f122679c30340b2c73afd7419e79644ddc3c4e Reviewed-on: https://review.coreboot.org/c/flashrom/+/61194 Reviewed-by: Nico Huber <nico.h@gmx.de> Reviewed-by: Thomas Heijligen <src@posteo.de> Reviewed-by: Angel Pons <th3fanbus@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
generated by cgit v1.2.3 (git 2.25.1) at 2025-04-25 10:55:17 +0000