On 17/08/2020 13:21, Greg KH wrote:

On Mon, Aug 17, 2020 at 12:36:25PM -0300, Guilherme G. Piccoli wrote:

...

Hi Greg / Thomas and all involved here. First, apologies for necro-bumping this thread, but I'm working a backport of this patch to kernel 4.15 (Ubuntu) and then I noticed we have it on stable, but only in 4.19+.

Since the fixes tag presents an old commit (since ~3.19), I'm curious if we have a special reason to not have it on long-term stables, like 4.9 or 4.14. It's a subtle portion of arch code, so I'm afraid I didn't see something that prevents its backport for previous versions.

What is the git commit id of this patch you are referring to, you didn't provide any context here :(

thanks,

greg k-h

I'm sorry, I hoped the subject + thread would suffice heh

So, the mainline commit is: f8a8fe61fec8 ("x86/irq: Seperate unused system vectors from spurious entry again") [0]. The backport to 4.19 stable tree has the following id: fc6975ee932b .

Thanks,

Guilherme

[0] http://git.kernel.org/linus/f8a8fe61fec8

On 17/08/2020 13:49, Greg KH wrote:

[...]

...

I'm sorry, I hoped the subject + thread would suffice heh

There is no thread here :(

Wow, that's odd. I've sent with In-Reply-To, I'd expect it'd get threaded with the original message. Looking in lore archive [1], it seems my first message wasn't threaded but the others were...apologies for that, not sure what happened...

...

So, the mainline commit is: f8a8fe61fec8 ("x86/irq: Seperate unused system vectors from spurious entry again") [0]. The backport to 4.19 stable tree has the following id: fc6975ee932b .

Wow, over 1 1/2 years old, can you remember individual patches that long ago?

Anyway, did you try to backport the patch to older kernels to see if it was possible and could work?

If so, great, please feel free to submit it to the stable@vger.kernel.org list and I will be glad to pick it up.

I'm working on it, it is feasible. But I'm seeking here, in this message, what is the reason it wasn't backported for pre-4.19 - is there anything on these patches that is known to break old kernels? I'll certainly submit my backported patch to stable 4.14 once I'm sure it's working and there's no specific reason to not have it done before.

Thanks again!

[1] https://lore.kernel.org/stable/a2788632-5690-932b-90de-14bd9cabedec@canonica...

On Mon, Aug 17, 2020 at 2:05 PM Greg KH gregkh@linuxfoundation.org wrote:

On Mon, Aug 17, 2020 at 01:59:00PM -0300, Guilherme G. Piccoli wrote:

...

On 17/08/2020 13:49, Greg KH wrote:

...

[...]

...

I'm sorry, I hoped the subject + thread would suffice heh

There is no thread here :(

Wow, that's odd. I've sent with In-Reply-To, I'd expect it'd get threaded with the original message. Looking in lore archive [1], it seems my first message wasn't threaded but the others were...apologies for that, not sure what happened...

reply to is fine, but how do you know what my email client has (hint, not a copy of 1.5 years of history sitting around in it at the moment...) So there is no "thread" here as far as it is concerned...

Anyway, not a big deal, just properly quote emails in the future, that's good to get used to no matter what :)

Sure, will do - specially for super old threads like this.

...

...

...

So, the mainline commit is: f8a8fe61fec8 ("x86/irq: Seperate unused system vectors from spurious entry again") [0]. The backport to 4.19 stable tree has the following id: fc6975ee932b .

Wow, over 1 1/2 years old, can you remember individual patches that long ago?

Anyway, did you try to backport the patch to older kernels to see if it was possible and could work?

If so, great, please feel free to submit it to the stable@vger.kernel.org list and I will be glad to pick it up.

I'm working on it, it is feasible. But I'm seeking here, in this message, what is the reason it wasn't backported for pre-4.19

Try reading the stable mailing list archives, again, you are asking about a patch 1.5 years ago. I can't remember information about patches sent _yesterday_ given the quantity we go through...

thanks,

greg k-h

OK, thanks Greg. If Thomas or anybody involved here knows a reason to not backport it to older kernels, please let me know - I'd really appreciate that. Cheers,

Guilherme