- Flashrom-stable - mail.coreboot.org

Announcement: flashrom-stable continues as flashprog
by Nico Huber 06 Dec '23

06 Dec '23

Hi everybody! as you may know already, flashrom-stable was removed from review.coreboot.org a while ago. flashrom-stable was a short- lived semi-fork, based on flashrom v1.2 [1]. The effort to keep developing a stable flash programming tool for everyone, however, shall continue. I have hence renamed flashrom-stable to *flashprog*. Now it's definitely a fork. The main reason for the fork is that I saw the flashrom master (now main) branch to become unmaintainable. Probably because we merged too much code together, without actually managing to work together on a product. Beside names, I tried to keep everything as usual. There's a wiki (slowly filled with contents from the old one) https://flashprog.org/ an IRC channel #flashprog on libera.chat and a mailing list, of course flashprog(a)flashprog.org managed at https://mail.sourcearcade.org/postorius/lists/flashprog.flashprog.org/ Review continues on a Gerrit instance, albeit on https://review.sourcearcade.org/ And you can clone the repository from a GitHub mirror https://github.com/SourceArcade/flashprog.git Now that all this is done, I hope to return to work on the source code again. There are still many patches open that were pending for flashrom-stable. And it looks like the next release candidate is only one commit away :) So stay tuned! Nico [1] https://mail.coreboot.org/hyperkitty/list/flashrom-stable@flashrom.org/thre…

1 0

Re: [flashrom] AMD FP4 - FCH device found but SMBus revision 0x61 does not match known values.
by Nico Huber 28 Apr '23

28 Apr '23

Hi Daniel, Ayden, On 25.04.23 19:03, Daniel Rozsnyó wrote: > I have an Epyc 2nd gen (Rome) system and I am getting this error: ... > Enabling flash write... FCH device found but SMBus revision 0x61 does > not match known values. there's a good chance that the flashrom-stable fork (list in CC) works for your systems. The source code is available via Git $ git clone -b v1.1 https://review.coreboot.org/flashrom-stable.git or as a tarball: https://download.flashrom.org/flashrom-stable/releases/flashrom-stable-v1.1… Daniel, I've never looked into ebuilds, I suppose it could be easily adapted, though. If you want to build it manually, check [1] for prerequisites. I suggest that you make a backup first and try to confirm its consistency before any write attempt. With some luck, there are some markers visible in a hexdump. At *20000 you may see "aa 55 aa 55". And probably in the very last line some pattern with 90* e9. The latter is legacy, though, and may not be there. This is from one of my AM4 boards: $ hexdump -C /tmp/dump | grep -E '0000 aa 55 aa 55|fff0 (90 )+.*e9' 00020000 aa 55 aa 55 00 00 00 00 00 00 00 00 00 10 02 ff |.U.U............| 00fffff0 90 90 e9 03 e9 00 00 00 fc 00 00 00 00 00 d0 ff |................| Cheers, Nico [1] https://www.flashrom.org/Downloads

1 0

[announce] flashrom-stable v1.1
by Nico Huber 03 Apr '23

03 Apr '23

Hi folks, after a little more (boring) testing, the v1.1 release is ready and tagged. 281 commits went into this release. About 200 of them were cherry-picks from the original master branch. After those, regular development continued, mostly adding more programmer drivers. # What changed * The buildsystem integration was almost re-written. The Makefile looks a lot cleaner now, and Meson is ready to be used on more than just Linux. * Two new programmer drivers were added: `ch347_spi`, `linux_gpio_spi` * The `dirtyjtag_spi` driver is DJTAG2 ready. * The `internal` programmer has a new driver for AMD's SPI100 controller now. This was tested with different chipset confi- gurations on Pinnacle Ridge, Raven Ridge, Matisse, Vermeer, and Genoa(!) :) * And the usual smaller refactorings and other changes. # Download The official tarball + signature live on flashrom.org now: https://download.flashrom.org/flashrom-stable/releases/flashrom-stable-v1.1… https://download.flashrom.org/flashrom-stable/releases/flashrom-stable-v1.1… And of course, there's the Git tag `v1.1`. # What was tested recently: * external SPI: ch347_spi, ft2232_spi, dirtyjtag_spi, buspirate_spi, serprog (w/ stm32-vserprog, SPI flash), pickit2_spi, jlink_spi, usbblaster_spi, ch341a_spi, linux_spi and linux_gpio_spi * internal: SPI on Intel ICH7, ICH9 (hw&sw sequencing), PCH7, and APL, Parallel through VIA VT82C686 southbridge Cheers, Nico

1 0

[announce] RC1 for v1.1 ready
by Nico Huber 29 Mar '23

29 Mar '23

Hi folks, after a lot of additional cherry-picking from the last few years of flashrom development and some new development, finally a new RC is ready. After flashrom-stable v1.0, about 200 more patches were cherry picked from the original flashrom master branch. These were all changing things under the hood: Integration of both build systems, Make and Meson, was almost re-written to align the two closer to another. Which brings us a huge step closer to eventually replace everything with Meson. Much refactoring took place, most of it to reduce the use of global symbols in programmer drivers. The internal API for handling layouts was re-designed to better encapsulate the implementation. Which is using linked lists now. On top, some new programmers were added: * `linux_gpio_spi`, a bit-banging driver utilizing libgpiod, * `ch347_spi`, a new high-speed USB-SPI bridge, * and finally `amd_spi100` with AMD Zen support! The tag `v1.1-rc1` is available via Git: git+https://review.coreboot.org/flashrom-stable.git And an official tarball is also available: https://download.flashrom.org/flashrom-stable/releases/flashrom-stable-v1.1… Happy testing, Nico

1 1

[announce] flashrom-stable v1.0
by Nico Huber 20 Jan '23

20 Jan '23

Hi all, after roughly one week of settling and testing the RC1 proved itself well (see "What was tested" below) and I'm happy to announce v1.0! The v1.0 release comprises about 300 patches picked from the original flashrom master branch on top of flashrom v1.2.1. As planned, these patches mostly add support for new flash chips, new programmers and related new features. We skipped invasive patches, so hopefully it will still build on all platforms where flashrom v1.2 built. # What changed * The flash-chip database grew by 31 entries * One new programmer driver for DirtyJTAG * Many updates to the Intel chipset driver, including support for Jasper, Elkhart, Comet, Tiger and Alder Lake * Preliminary support for the block-protection of SPI flashes in libflashrom * And lots of smaller updates Again, many thanks to all the contributors and reviewers! # Notes on building The official build system is still the `Makefile` which requires GNU make. On some older or rarer platforms, a `WARNERROR=no` ar- gument to `[g]make` will be necessary. One programmer `jlink_spi` is still disabled by default because of its rare dependency on `libjaylink`. If the latter is available, it is preferred to build with `CONFIG_JLINK_SPI=yes`. There has been rudimentary support for the Meson build system for a while. It's still in a rather experimental state. However where packages for flashrom v1.2 were built with Meson it should work the same. Meson support should get a more official status in the next release. # Download Due to time constraints, flashrom-stable has no official place for signed tarballs yet. So the preferred option is to use the Git tag `v1.0` that is signed with the following key: 2853079C9C66AB7E82C64966A5C163B7E557CAEB e.g. $ git clone -b v1.0 https://review.coreboot.org/flashrom-stable.git $ cd flashrom-stable $ git tag -v v1.0 An automatically generated tarball is available here: https://review.coreboot.org/plugins/gitiles/flashrom-stable/+archive/5a12b9… # What was tested Beside some massive build testing [1] (that's probably still not enough given the amount of supported platforms), the following programmers were tested recently: * external SPI: buspirate_spi, ch341a_spi, dediprog, ft2232_spi, jlink_spi, linux_spi, pickit2_spi, serprog, usbblaster_spi * internal: SPI with hw&sw sequencing on Intel ICH9 and PCH7, Parallel through VIA VT82C686 southbridge Happy flashing! Nico [1] The list is long, the following should provide an idea: https://review.coreboot.org/plugins/gitiles/flashrom-stable/+/refs/tags/v1.…

1 0

[announce] First flashrom-stable RC
by Nico Huber 13 Jan '23

13 Jan '23

Hi, pleased to announce that there is now a v1.0-rc1 tagged. Alas, much delayed. It was hard to settle on a point amidst all the patches that were / will be picked from flashrom/master. About 300 patches were picked for the first flashrom-stable release. Another 200 are already prepared in the queue for the following release. What changed? * The flash-chip database grew by 31 entries * One new programmer driver for DirtyJTAG * Many updates to the Intel chipset driver, including support for Jasper, Elkhart, Comet, Tiger and Alder Lake * Preliminary support for the block-protection of SPI flashes in libflashrom * And lots of smaller updates Big thanks to everybody helping with the review! The source code is available via Git: git+https://review.coreboot.org/flashrom-stable.git or as tarball: https://review.coreboot.org/plugins/gitiles/flashrom-stable/+archive/14326b… And for Arch users, there's an AUR package: https://aur.archlinux.org/packages/flashrom-stable Cheers, Nico

1 0

Progress update
by Nico Huber 23 Dec '22

23 Dec '22

Hi all, we've started work on a new `main` branch, and thanks to some help from Angel, the first round of patches (flash chip related changes) is merged. \o/ Patches for a second round of picks are also prepared and build tested: https://review.coreboot.org/q/hashtag:fr-2nd_round They contain mostly programmer related changes, a part of the Makefile overhaul and smaller refactorings. These will take a little longer to review compared to the first round. The original patches seemed in good shape, however picking needed much more manual merging. And it's also twice as many patches oO I wish everybody some nice holidays! :) Cheers, Nico

1 0

Kicking off flashrom-stable
by Nico Huber 17 Dec '22

17 Dec '22

Hi folks, welcome to fashrom-stable! What's it going to be? Most of all, a continuation of the effort that ceased ~3 years ago. We tried to reduce technical debt in the code and to provide regular releases. I hope we can create something sustainable, something that can be maintained without requiring permanent attention. I see some traits in flashrom that I grew particularly fond of over the years: * It compiles and runs on an incredible amount of platforms. * It doesn't have to drop old hardware support. * It's for everyone, even humble users. We try our best not to brick any machine by accident. I'd like to keep it that way. Lacking any bigger discussion about the development process, I came up with a tight plan for the first release and a rough one for the second. I'll start a `main` branch soon that will be based on flash- rom 1.2.x. The idea is to start by picking cherries from the original flashrom `master`. For the first release, flashrom-stable 1.0, I'll make two runs, picking changes and additions related to flash chip support and then changes and additions in programmer support. This will also draw in some related work to avoid merge conflicts. Apart from that only fixes and infra- structure changes should be merged. This will hopefully keep the wide platform support of flashrom 1.2 in tact. Then, flashrom-stable 1.0 should be able to fill the 3 year gap for older platforms. Once we have branched for the 1.0 release, development should be more open for all kinds of changes. However, I'd like to focus on some par- ticular things until branching for the 1.1 release: * build-system changes that happened on flashrom master (revised Makefile and decent Meson support) * useful refactorings from flashrom master * CLI for the write-protect support * updated internal flashing support for AMD systems * Aarya's still unmerged erase overhaul 1.0 should be ready around Christmas/New Year. 1.1 hopefully around FOSDEM. I'd much appreciate any help with the review. I'll set up a group on Gerrit for reviewers (+2 rights) and add a few folks that I see as experts in (an area of) flashrom. If anyone feels like they are mis- sing there, please write to me and we can discuss things. I'll try to write down some rules about this after the first release. +1 re- views are also appreciated, just note down what aspects are reviewed and if you consider a patch as ready. Well, I tried to keep the mail short. And hope it's enough to give flashrom-stable a good start. Cheers, Nico

1 0

2024

2023

2022

Flashrom-stable