Andrew Hewus Fresh <and...@afresh1.com> wrote: > On Thu, Feb 10, 2022 at 12:40:43PM +0100, Alexander Bluhm wrote: > > Does the new fw_update not check the version number of the file? Does > > it always overwrite with the tgz from firmware.openbsd.org? > > It only checks whether the versions match, one of the caveats I > explained early in the process was that it would not be able to check > for "newer" or "older", just "different". > > If someone wants to supply a patch that supports comparing version > numbers in the shell the way the package tools do, > I will run away screaming :-)
I think the entire idea of 'there are multiple files, pick one' is completely insane. We don't need such a scheme. We curate a singular choice of each firmware. We supply that. End of story. If anything we want to reduce the choices people make, because those choices are not disclosed to us during the normal bug reporting choices. With firmwares, we can make such a stand, it is completely normal. I also think there is an extreme amount of drama over "intel cpu firmware are terrible, surely they are going to break someone's machine". Yes, that has happened. When it happens, the game of "go back to another firmware" isn't for us to play. Intel issues a new one before long, and then we move forward again. I don't see any reason for us to play a this back-and-forth game at firmware.openbsd.org, to engage the public in that diagnosis, it is a thing for experts to do, and those experts can stop running fw_update and do it by hand.
