DRM Panic "Screen of Death" Support Being Extended To All Recent AMD GPUs

[intelfx]

Please. Don't. Turn. This. Forum. Into. MORONIX.

You should have directed this reply to the great-grandparent comment.

(That said, judging by the incredibly thoughtful subsequent "hurr durr QR malware" comment of yours, you don't seem to require any help in turning this forum into Moronix )

[intelfx]

As I said, a panic message that *needs to be supported* is bad a priori,

You should educate yourself on the matter prior to “saying” anything.

[intelfx]

It's good you have your priorities straight. Don't care about the hundreds of people being maimed and killed every day (on both sides), but make a few IT professionals submit patches through other people instead of directly and it's just brutal to have to live through such a betrayal, isn't it?

There's a many great deal of wars happening on Earth at every point in time. Why don't you care about every single one of them, about the Israelites and Palestinians, about starving children and mutilated women in every single of the African civil wars, etc, etc? Out of 100 wars happening on Earth, you don't care about 99 of them, so even by your own standards, your priorities are 99/100ths as misguided as the GP's

[aerospace]

You should educate yourself on the matter prior to “saying” anything.

I've designed and I'm designing quite a few systems. Dealing with single points of failures is a design concept which is above the panic message specifics.

[primary]

For some reason, since the recent outrage concerning the brutal banning of several kernel maintainers, Linux news feel much less exciting.
I have a sharp feeling of betrayal because of what happened. And yet, I am French.
Who knows what'll happen next.
We live in strange times, that is for sure.

go and whine somewhere else.

[intelfx]

I've designed and I'm designing quite a few systems. Dealing with single points of failures is a design concept which is above the panic message specifics.

That does not lend any credence to your point. It's very common for system designers to become so enamored with "design concepts" and so dismissive of "lowly specifics" that they become completely detached from reality.

What exactly makes you think that anything mentioned in the article is a single point of failure?

[archkde]

As I said, a panic message that *needs to be supported* is bad a priori, It doesn't matter how fancy it is. Making it vurnerable to more and more failure points is plain bad engineering, anything in the panic path should work from a vesa card to an A100. Tomorrow someone will propose a "tiny" LLM model running on the GPU to produce the most informative panic message ever, and promises of speed and useful information will follow. Mark my words.

Why does anything in the panic path need to work on a vesa card? The system is already extremely broken when the panic happens, so you might as well do the best you can on each system to be as helpful to the user as possible.

[archkde]

Out of topic but...So you're OK with the "poor souls" working for the western offense industry to be kept as maintainers?

No, they should be shamed or (in particularly bad cases) banned too. The difference is that the maintainers from Russia have to be banned because of the law (I do admittedly not completely understand why sanctions apply to this situation, but I don't make these laws).

[Blisterexe]

For some reason, since the recent outrage concerning the brutal banning of several kernel maintainers, Linux news feel much less exciting.
I have a sharp feeling of betrayal because of what happened. And yet, I am French.
Who knows what'll happen next.
We live in strange times, that is for sure.

fyi they didnt ban russian maintainers, they banned maintainers working for russian companies, there is a difference.

[F.Ultra]

fyi they didnt ban russian maintainers, they banned maintainers working for russian companies, there is a difference.

also they where not banned, simply removed from the maintainers file.