September 2026 - flashrom - mail.coreboot.org

flashrom for windows, maintainers wanted
by Anastasia Klimchuk Sept. 25, 2026

Sept. 25, 2026

Good day everyone, We are looking for people who would be the "flashrom for windows" maintainers (and it can be multiple people). At the moment, people use flashrom on windows, and apparently it works (good! :)) but no one officially takes care of this. Support mostly comes from our amazing community: someone notices something is broken, tests it, and hopefully sends us a patch. We would like to make this more organized, and we need people willing to own the Windows platform. You do not need to know everything about flashrom, or every programmer we support. You need to understand coding, be willing to spend some time testing on real Windows hardware, and help us make sure that what we claim to support actually works. This is a partial support, there are no plans to make every flashrom feature work on Windows. We plan to keep all the windows-specific code in the tree until the end of 2026. If no one wants to maintain it, the code will be deleted and it won't be included in the release v2.0.0. If you would like to volunteer, help with testing, or just ask questions before committing to anything, please reply to this thread on the mailing list. For reference, here is our Development Guide: https://flashrom.org/dev_guide/development_guide.html Thank you for your time reading this email :) -- Anastasia.

4 6

MX25L4005(A/C)/MX25L4006E on mstarddc_spi: WP status/disable "not implemented";
by Carlos Bragatto Sept. 24, 2026

Sept. 24, 2026

*Subject:* MX25L4005(A/C)/MX25L4006E on mstarddc_spi: WP status/disable "not implemented"; erase fails on protected top sector during multi-sector write *Chip:* Macronix MX25L4005(A/C)/MX25L4006E (512kB, SPI), probed correctly via mstarddc_spi *Programmer/setup:* mstarddc_spi:dev=/dev/i2c-1:49 — MStar-based scaler (TSUMOP88CDT9-3) on a Dell P1917S monitor mainboard, accessed via the ISP-over-DDC bridge on the VGA connector (not direct SPI wiring). *Issue 1:* --wp-status and --wp-disable both fail immediately with WP operations are not implemented for this chip. No protection range information is available through flashrom's WP interface for this part. *Issue 2:* During a -w multi-sector write, the implicit unprotect step prints Block protection could not be disabled!, then the write proceeds and fails specifically at one 4KB-aligned sector at the very top of the chip ( 0x7e000–0x7efff), with Expected=0xff, Found=0x00 — i.e., the erase of that specific sector did not take effect, while every other sector in the same write (including sectors both before and after it in address order) erased and wrote successfully and verified correctly. This is consistent with a standard Macronix BP (block-protect) status-register configuration protecting the top region of the chip, but I can't confirm this since WP reporting isn't implemented for this chip. *Consequence / near-miss:* because the overall write operation aborted after this sector's erase failure, one *earlier*, unrelated, unprotected sector (0x3000–0x3fff) had already been erased in preparation for writing but never received its new data before the abort, leaving it fully blank ( 0xFF) until manually repaired from a known-good backup. Worth flagging as a real data-loss risk for anyone hitting this same protected-sector failure mode mid-write on a multi-sector image, not just a cosmetic warning. *Request:* any guidance on whether WP support could be added for this chip, and/or whether flashrom's write logic could avoid leaving unrelated already-erased sectors unwritten when a later sector's erase fails, would be very welcome. Happy to provide full logs/dumps on request.

1 0

SPDX adoption final ?
by Abdelkader Boudih Sept. 21, 2026

Sept. 21, 2026

Hi there I was working on adding a module for my linter for flashrom and noticed > * This file is part of the flashrom project. I think we should remove this. We have SPDX headers in most files, that string add just noise. I added it in some files i authored, because i thought there is a documented reason for it i didnt read yet.. I can't find it. SPDX and git history is the SBOM way to know files origin. If we ever vendor something, a folder vendor/ or contrib/ is the proper way to do it. Best, Abdelkader

2 1

Flashrom & Modern Bus Pirate (2026)
by David Reguera Garcia Sept. 18, 2026

Sept. 18, 2026

Hi everyone, For those who don't know me, I'm the maintainer of the Bus Pirate support in flashrom, but I'm also the developer working on the flashrom side of the modern Bus Pirate firmware. Basically, I try to cover both sides so everything works smoothly, and I fix whatever needs fixing along the way. Over the last few days I've been improving the tests and, especially, the physical setup I use for testing different chips and manufacturers. I've attached a photo showing how I have it set up. https://photos.app.goo.gl/ac8nkVC8ffo5Kad57 I have a modern Bus Pirate connected to a collection of SPI flash chips from different manufacturers, and for every flashrom release I make sure that at least these chips can be read and written successfully. I label the chips using electrical tape and a permanent marker. I put a little cyanoacrylate over the writing so that IPA residue or simply handling the chips doesn't slowly rub the labels off. I also write down the voltage I want to test each chip at. I have different sockets and adapters for QFN, DIP and other package types as well. I'm posting this in case anyone is running into any weird problems or bugs when using flashrom with the modern Bus Pirate. If you find something that doesn't work as expected, just let me know and I'll take a look. I can also check what is going wrong on the firmware side and fix it if needed. For anything related to flashrom and the modern Bus Pirate, I'm around on the mailing list, so feel free to ask me anything. Thanks! Dreg

1 0

Windows and DOS support drop ?
by Abdelkader Boudih Sept. 15, 2026

Sept. 15, 2026

Hi all, While working on adding colored output (red -> errors and green for successful operations) I noticed that the Windows and DOS support appears to be only partially implemented. I tested the DOS code in a virtual machine and found that it does not currently work. Do DOS even have USB support ? Since I don't want to maintain or fix this support, I would like to propose removing DOS support along with the programmers. This would simplify the codebase and remove support for dead platform. Personally, I think our maintenance effort would be better spent supporting platforms that are still actively used or developed. I would prefer see support for platforms such as AmigaOS, 9front, HaikuOS, or UEFI than continue compiling non-functional DOS support. Also Windows... Windows has Powershell now, and linux emulation. And people with Windows will prefer a GUI. Thoughts ? Best, Abdelkader

4 9