New header file lifecycle.h need to gather all things shared among
lifecycle tests.
This is one step to the goal of splitting lifecycle tests into
separate per-programmer file.
BUG=b:237606255
TEST=ninja test
Change-Id: I93d0db943d9c96e2c36e9f7dce5c885c959745a0
Signed-off-by: Anastasia Klimchuk <aklm@chromium.org>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/65541
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Singer <felixsinger@posteo.net>
Reviewed-by: Thomas Heijligen <src@posteo.de>
Move global singleton states into a struct and store within
the par_master data field for the life-time of the driver.
This is one of the steps on the way to move par_master data
memory management behind the initialisation API, for more
context see other patches under the same topic "register_master_api".
BUG=b:185191942
TEST=builds
Change-Id: I00877e3cc359996e3aa59649f62c76e521ab119b
Signed-off-by: Alexander Goncharov <chat@joursoir.net>
Ticket: https://ticket.coreboot.org/issues/391
Reviewed-on: https://review.coreboot.org/c/flashrom/+/65342
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Anastasia Klimchuk <aklm@chromium.org>
Reviewed-by: Thomas Heijligen <src@posteo.de>
Avoid confusion from the comment. While technically a GPIO
can do anything, like drive a LED. The GPIO pin 88 *is*
meant to drive the WP line of the SPI flash, that is its
purpose.
Change-Id: If718d41a27931380e5f7ebdb75b9863da0c61559
Signed-off-by: Edward O'Callaghan <quasisec@google.com>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/65546
Reviewed-by: Peter Marheine <pmarheine@chromium.org>
Reviewed-by: Thomas Heijligen <src@posteo.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This code was originally introduced by ITE for now exceedingly old
Chromebooks. The code has had very little attention to maintain it,
unlikely tested for a long time and now seems to be just a technical
burden to the flashrom project.
If someone is later interested it could be resurrected for reference
from git history. However, it needs quite a bit of work to bring it back
into maintainable order.
BUG=b:156143896,b:170689483
TEST=tree builds under meson+make and unit tests pass.
Change-Id: I5e8cafd73db837941c518f0e2d72d8192274fd79
Signed-off-by: Edward O'Callaghan <quasisec@google.com>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/65378
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Anastasia Klimchuk <aklm@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This patch changes the function name of extract_programmer_param() to
extract_programmer_param_str() as this function name will clearly
specify that it returns the value of the given parameter as a string.
Signed-off-by: Chinmay Lonkar <chinmay20220@gmail.com>
Change-Id: Id7b9fff4d3e1de22abd31b8123a1d237cd0f5c97
Reviewed-on: https://review.coreboot.org/c/flashrom/+/65521
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Singer <felixsinger@posteo.net>
Reviewed-by: Thomas Heijligen <src@posteo.de>
Move global singleton states into a struct and store within
the par_master data field for the life-time of the driver.
This is one of the steps on the way to move par_master data
memory management behind the initialisation API, for more
context see other patches under the same topic "register_master_api".
Implements: https://ticket.coreboot.org/issues/391
BUG=b:185191942
TEST=builds
Change-Id: I3dd35eceadb9dbca8e526705b7be977564ed7318
Signed-off-by: Alexander Goncharov <chat@joursoir.net>
Signed-off-by: Anastasia Klimchuk <aklm@chromium.org>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/65194
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Edward O'Callaghan <quasisec@chromium.org>
In commit a721181a08 dummyflasher became an opaque master too, and
now registers prog bus by default. This patch upgrades a dummy unit
test which uses all buses as programmer param, and adds a unit test
which covers specific use case for opaque programmer.
BUG=b:233816068
TEST=ninja test
Change-Id: I61a5333b61ea84fb91c7f8310d52b64213c62f83
Signed-off-by: Anastasia Klimchuk <aklm@chromium.org>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/65236
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Felix Singer <felixsinger@posteo.net>
Reviewed-by: Joursoir <chat@joursoir.net>
Reviewed-by: Thomas Heijligen <src@posteo.de>
There is a function, spi_get_erasefn_from_opcode, which returns the
erase function for given opcode. Add a function which does the opposite
i.e. returns the opcode for given erase function.
Change-Id: Ia3aefc9b9465efdd16b1678bb2ada9a23f00d316
Signed-off-by: Aarya Chaumal <aarya.chaumal@gmail.com>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/65355
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Thomas Heijligen <src@posteo.de>
Simplify ich9_handle_frap() to do the translation to the logical
representation of the ich_access_protection enum in one place and
work from there. This removes some unnecessary branch complexity
and the possibility of out of bounds array accesses.
Change-Id: I1eda067c44a84d662713475d13902c85534a59fe
Signed-off-by: Edward O'Callaghan <quasisec@google.com>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/65189
Reviewed-by: Sam McNally <sammc@google.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nikolai Artemiev <nartemiev@google.com>
0 is an invalid value for freq parameter and caused floating point
exception. This patch checks that freq is not 0 during
initialisation.
Fixes: https://ticket.coreboot.org/issues/366
TEST=the following scenarios
1) error
$ ./flashrom -p dummy:emulate=W25Q128FV,freq=0 -V
<...>
init_data: invalid value 0 for freq parameter
Unhandled programmer parameters (possibly due to another failure): emulate=W25Q128FV,
Error: Programmer initialization failed.
2) successful
$ ./flashrom -p dummy:emulate=W25Q128FV,freq=10 -V
Found Winbond flash chip "W25Q128.V" (16384 kB, SPI).
3) default is also successful
$ ./flashrom -p dummy:emulate=W25Q128FV -V
Found Winbond flash chip "W25Q128.V" (16384 kB, SPI).
Change-Id: I0a95495de0a677f0d4d7f4c2fc61dcbc00d6ad4c
Signed-off-by: Anastasia Klimchuk <aklm@chromium.org>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/65240
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Edward O'Callaghan <quasisec@chromium.org>
Reviewed-by: Felix Singer <felixsinger@posteo.net>
According to its datasheet, Spansion S25FL512S supports writing/
reading its extended address register via 0x17/0x16 opcodes. With
that enabled, we can also enable the EAR7 feature, i.e. toggling
4BA mode via bit 7 of that register.
S25FL512S did not advertise EAR support at all, so we set it to
TEST_UNTESTED again.
Change-Id: Ib214e509a5c294ab60460a2b5d00a713a119ab3f
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/65265
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Thomas Heijligen <src@posteo.de>
According to their datasheets, ISSI IS25LP256 and IS25WP256 support
both 0xc5/0xc8 and 0x17/0x16 opcodes to write / read their extended
address register. Flashrom will use 0xc5 by default if available,
so adding the FEATURE_4BA_EAR_1716 flag makes no difference for now
(FEATURE_4BA_EAR_C5C8 is included in the already selected FEATURE_4BA
set). It's better to have a comprehensive description of the chips,
though, in case somebody wants to use them in the future with a
master that restricts available opcodes.
Change-Id: I03e4ff825c7742e7ff79b51b75293d53a091d4d4
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/65264
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Thomas Heijligen <src@posteo.de>
There are two competing sets of instructions to access the extended
address register of 4BA SPI chips. Some chips even support both sets.
So far, we assumed the 0xc5/0xc8 instructions by default and allowed
to override the write instructions with the `.wrea_override` field.
This has some disadvantages:
* The additional field is easily overlooked. So when adding a new
flash chip, one might assume only 0xc5/0xc8 are supported.
* We cannot describe flash chips completely that allow both
instructions (and some programmers may be picky about which
instructions can be used).
Therefore, replace the `.wrea_override` field with a feature flag.
Signed-off-by: Nico Huber <nico.h@gmx.de>
Change-Id: I6d82f24898acd0789203516a7456fd785907bc10
Ticket: https://ticket.coreboot.org/issues/357
Reviewed-on: https://review.coreboot.org/c/flashrom/+/64636
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Thomas Heijligen <src@posteo.de>
There are two competing sets of instructions to access the extended
address register of 4BA SPI chips. Some chips even support both sets.
To prepare for other instructions than the default 0xc5/0xc8, rename
the original feature flag.
Signed-off-by: Nico Huber <nico.h@gmx.de>
Change-Id: Iacb7b68a9e3444fe28873ff0fe5e3fab16643c8c
Ticket: https://ticket.coreboot.org/issues/357
Reviewed-on: https://review.coreboot.org/c/flashrom/+/64635
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Thomas Heijligen <src@posteo.de>
As discovered earlier[1], the `vendor_id` and `device_id` fields are not
always automatically set. However, we use these fields throughout flash-
rom. To not lose track when we actually fetched them, let's always call
pci_fill_info(PCI_FILL_IDENT) before returning a `pci_dev` handle.
[1] Commit ca2e3bce0 (pcidev.c: populate IDs with pci_fill_info())
Signed-off-by: Nico Huber <nico.h@gmx.de>
Change-Id: Iae2511178bec44343cbe902722fdca9eda036059
Ticket: https://ticket.coreboot.org/issues/367
Reviewed-on: https://review.coreboot.org/c/flashrom/+/64573
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Daniel Verkamp <dverkamp@chromium.org>
Reviewed-by: Thomas Heijligen <src@posteo.de>
The ichspi hwseq path has a opaque master specialisation that
allows for reading and writing STATUS1 registers. Hook the callbacks
with a implementation to allow for this so that writeprotect maybe
supported though this path.
BUG=none
BRANCH=none
TEST=flashrom --wp-status on AMD and Intel DUTs
Change-Id: I7ecbe8491ecea3697922c91af26ca62276e86317
Signed-off-by: Edward O'Callaghan <quasisec@google.com>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/64540
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Subrata Banik <subratabanik@google.com>
Allow specialisation in opaque masters, such as ichspi hwseq, to
write to status registers.
Also update the dispatch logic in libflashrom to call wp code when
status register access functions are provided by an opaque master.
BUG=none
BRANCH=none
TEST=flashrom --wp-status on AMD and Intel DUTs
Change-Id: I3ab0d7f5f48338c8ecb118a69651c203fbc516ac
Signed-off-by: Edward O'Callaghan <quasisec@google.com>
Signed-off-by: Nikolai Artemiev <nartemiev@google.com>
Co-Authored-by: Nikolai Artemiev <nartemiev@google.com>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/64375
Reviewed-by: Anastasia Klimchuk <aklm@chromium.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
hwaccess functions used to be static inline functions and needed
a special treatment so that they could be mocked for unit tests.
This has changed, see include/hwaccess_x86_io.h now the functions
are not static inline anymore, and it is possible to use regular
cmocka wraps.
Fixes https://ticket.coreboot.org/issues/385
BUG=b:181803212
TEST=ninja test
Change-Id: Iafce071ea7ad5bcfdebbba968699d5743705f8e0
Signed-off-by: Anastasia Klimchuk <aklm@chromium.org>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/64881
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Joursoir <chat@joursoir.net>
Reviewed-by: Thomas Heijligen <src@posteo.de>
These chips seem to be rather regular, supporting 2.7V..3.6V, the
common erase block sizes 4KiB, 32KiB, 64KiB and the usual block-
protection bits.
Status/configuration register naming differs from other vendors,
though. These chips have 2 status registers plus 3 configuration
registers. Configuration registers 1 & 2 match status registers
2 & 3 of what we are used from other vendors. Read opcodes match
too, however writes are always done through the WRSR instruction
which can write up to 4 bytes (SR1, CR1, CR2, CR3).
S25FL256L supports native 4BA commands and entering a 4BA mode.
However, it uses an unusual opcode (0x53) for the 32KiB 4BA block
erase.
Signed-off-by: Nico Huber <nico.h@gmx.de>
Change-Id: I356df6649f29e50879a4da4183f1164a81cb0a09
Reviewed-on: https://review.coreboot.org/c/flashrom/+/64747
Reviewed-by: Thomas Heijligen <src@posteo.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Spansion flash chips S25FL128L and S25FL256L use the WRSR instruction to
write more than 2 registers. So align SR2 and SR3 support: The current
FEATURE_WRSR_EXT is renamed to FEATURE_WRSR_EXT2 and FEATURE_WRSR_EXT3
is added. Also, WRSR3 needs a separate flag now.
Verified that FEATURE_WRSR_EXT2 still works using the `dummy_flasher`.
Signed-off-by: Nico Huber <nico.h@gmx.de>
Change-Id: Ibdfc6eb3d2cfecbf8da0493d067031ddb079a094
Reviewed-on: https://review.coreboot.org/c/flashrom/+/64746
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nikolai Artemiev <nartemiev@google.com>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-by: Thomas Heijligen <src@posteo.de>
Add a check so that the erase functions for all flashchips are in
increasing order of their respective eraseblock sizes. This is required
for the implentation of the improved erasing algorithm. The patch uses
the count of eraseblocks in each erase function to determine the order
(More eraseblocks means that the function has smaller eraseblock size).
Also fix the structs in flashchips.c which were found to be not
conforming to this test.
TEST = make && ./flashrom
Change-Id: I137cb40483fa690ecc6c7eaece2d9d3f7a851bb4
Signed-off-by: Aarya Chaumal <aarya.chaumal@gmail.com>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/64961
Reviewed-by: Thomas Heijligen <src@posteo.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Move global singleton states into a struct and store within
the par_master data field for the life-time of the driver.
This is one of the steps on the way to move par_master data
memory management behind the initialisation API, for more
context see other patches under the same topic "register_master_api".
Implements: https://ticket.coreboot.org/issues/391
BUG=b:185191942
TEST=builds
Change-Id: I82e4c82916dc835e9462b80750b06f9d78701edf
Signed-off-by: Alexander Goncharov <chat@joursoir.net>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/64963
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Anastasia Klimchuk <aklm@chromium.org>
Reviewed-by: Thomas Heijligen <src@posteo.de>
Reviewed-by: Edward O'Callaghan <quasisec@chromium.org>
Wire "variable size" feature in dummy programmer via opaque infra.
This patch fixes the broken build with CONFIG_DUMMY=no.
Dummyflasher registers opaque master for the case when it is
initialised with EMULATE_VARIABLE_SIZE. Dummy opaque master emulates
read/write/erase as simple memory operations over
`data->flashchip_contents`.
The feature works via "Opaque flash chip" in flashchips.c which has
one block eraser at the moment. If this changes in future, each block
eraser needs to be updated in `probe_variable_size`.
Fixes: https://ticket.coreboot.org/issues/365
TEST=the following scenarious run successfully
Testing build
$ make clean && make CONFIG_DUMMY=no
$ flashrom -h : dummy is not in the list
$ make clean && make CONFIG_EVERYTHING=yes
$ flashrom -h : dummy is in the list
Testing "variable size" feature
$ flashrom -p dummy:size=8388608,emulate=VARIABLE_SIZE -V
$ flashrom -p dummy:size=8388608,emulate=VARIABLE_SIZE
-r /tmp/dump.bin -V
$ head -c 8388608 </dev/urandom >/tmp/image.bin
$ flashrom
-p dummy:image=/tmp/image.bin,size=8388608,emulate=VARIABLE_SIZE
-w /tmp/dump.bin -V
also same as above with erase_to_zero=yes
Testing standard flow
$ flashrom -p dummy:emulate=W25Q128FV -V
$ flashrom -p dummy:emulate=W25Q128FV -r /tmp/dump.bin -V
$ head -c 16777216 </dev/urandom >/tmp/image.bin
$ flashrom -p dummy:image=/tmp/image.bin,emulate=W25Q128FV
-w /tmp/dump.bin -V
Testing invalid combination of programmer params (`init_data` fails
and prints error message which is WAI)
$ flashrom -p dummy:size=8388608 -V
-> init_data: size parameter is only valid for VARIABLE_SIZE chip.
$ flashrom -p dummy:emulate=VARIABLE_SIZE -V
-> init_data: the size parameter is not given.
$ flashrom -p dummy:emulate=W25Q128FV,erase_to_zero=yes -V
-> init_data: erase_to_zero parameter is not valid for real chip.
Change-Id: I76402bfdf8b1a75489e4509fec92c9a777d0cf58
Signed-off-by: Anastasia Klimchuk <aklm@chromium.org>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/64488
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Edward O'Callaghan <quasisec@chromium.org>
Currently, WARNERROR is hardcoded to 'yes' which is what we want most of
the time, but there are cases (such as when building with -fanalyzer)
that we don't want it enabled, so we see the entire output of the build,
and it doesn't halt at the first error. Removing the WARNERROR line
from the script allows the variable to be overridden from an environment
variable.
Signed-off-by: Martin Roth <gaumless@gmail.com>
Change-Id: Iea931e57f2a6992762566dc3dbaae8bb8df5b226
Reviewed-on: https://review.coreboot.org/c/flashrom/+/62745
Reviewed-by: Felix Singer <felixsinger@posteo.net>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Projects using libflashrom like fwupd expect the user to wait for the
operation to complete. To avoid the user thinking the process has
"hung" or "got stuck" report back the progress complete of the erase,
write and read operations.
Add a new --progress flag to the CLI to report progress of operations.
Include a test for the dummy spi25 device.
TEST=./test_build.sh; ./flashrom -p lspcon_i2c_spi:bus=7 -r /dev/null --progress
Change-Id: I7197572bb7f19e3bdb2bde855d70a0f50fd3854c
Signed-off-by: Richard Hughes <richard@hughsie.com>
Signed-off-by: Daniel Campello <campello@chromium.org>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/49643
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Edward O'Callaghan <quasisec@chromium.org>
Reviewed-by: Anastasia Klimchuk <aklm@chromium.org>
Reviewed-by: Thomas Heijligen <src@posteo.de>
Set `flash->chip_restore_fn_count` to zero before calling the chip's
unlock funciton in `prepare_flash_access()`.
Previously `flash->chip_restore_fn_count` was uninitialized before
calling `chip->unlock()` and subsequently reset after the dispatch by
initializing it. This caused the restore handler that is registered
within `spi_disable_blockprotect_generic()` to be lost.
BUG=b:228945411
BRANCH=none
TEST=enable wp; flashrom -w; check wp still enabled.
Change-Id: I4c7df424bd2ae2b5fb2a2ab6b47a3c9ff3233acf
Signed-off-by: Nikolai Artemiev <nartemiev@google.com>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/63881
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Edward O'Callaghan <quasisec@chromium.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
