Linux 4.0 Kernel Released

[phoronix]

Phoronix: Linux 4.0 Kernel Released

Linus Torvalds went ahead and released the Linux 4.0 kernel today as expected...

http://www.phoronix.com/scan.php?page=news_item&px=Linux-4.0-Kernel

[edoantonioco]

Phoronix: Linux 4.0 Kernel Released

"So I decided to release 4.0 as per the normal schedule, because there really weren't any known issues..."

http://www.phoronix.com/scan.php?pag...nux-4.0-Kernel

"because there really weren't any known issues"? so kernel devs doesnt know the issue than shows me a black screen when I start the kernel 3.19 and make it useless? Thats why Im stuck in kernel 3.16. I have read about other people with similar problems, so I wonder how a bug than breaks the user-space its not a known issue (I have seen the bug in freedesktop, that's why I havent reported it)

[dungeon]

Kernel has 4K+ drivers, your issue is probably related to one. I guess issue is with one graphic driver, one graphic driver often supports hundreds of asics. Issue might be related only to that one or couple of the same families...

Blah, blah, i guess you got an idea

[nanonyme]

"because there really weren't any known issues"? so kernel devs doesnt know the issue than shows me a black screen when I start the kernel 3.19 and make it useless? Thats why Im stuck in kernel 3.16. I have read about other people with similar problems, so I wonder how a bug than breaks the user-space its not a known issue (I have seen the bug in freedesktop, that's why I havent reported it)

Have you tried bisecting?

[bridgman]

"because there really weren't any known issues"? so kernel devs doesnt know the issue than shows me a black screen when I start the kernel 3.19 and make it useless? Thats why Im stuck in kernel 3.16. I have read about other people with similar problems, so I wonder how a bug than breaks the user-space its not a known issue (I have seen the bug in freedesktop, that's why I havent reported it)

I looked through the kernel and FDO bugzillas and didn't see anything talking about black screen appearing between 3.16 and 3.18 for your hardware. There was one blackscreen bug related to dpm (66963) but it started several kernel versions earlier, and one related to dp (73530) but I don't think that would apply to your hardware (you have the 4250, right ?).

Note that when Linus says "known issues" he usually means "known regressions relative to the previous kernel version" -- and AFAICS there is a pretty good chance the regression you are seeing is not known to the kernel devs. I might have missed the ticket you mentioned though, searching the bugzilla db's is a bit of a black art, but sounds like it might be worth trying to pinpoint exactly where the kernel stopped working (ideally bisecting to find the commit) and file a ticket for it. I think you mentioned in another post that 3.18 had the problem, not just 3.19 ?

[Azrael5]

Excellent news! Just 10 days before newer distros launch.

[edoantonioco]

Have you tried bisecting?

No, I havent, but I just started investigating how to do it and I found interesting information, after all I have very clear than the las working properly wash 3.17 and the first not working (or randomly not working) was 3.18, but it requires me to download 4GB of data before bisecting, so due to slow internet connection I'm unable right now to do it. But im interested in trying.

I looked through the kernel and FDO bugzillas and didn't see anything talking about black screen appearing between 3.16 and 3.18 for your hardware. There was one blackscreen bug related to dpm (66963) but it started several kernel versions earlier, and one related to dp (73530) but I don't think that would apply to your hardware (you have the 4250, right ?).

Note that when Linus says "known issues" he usually means "known regressions relative to the previous kernel version" -- and AFAICS there is a pretty good chance the regression you are seeing is not known to the kernel devs. I might have missed the ticket you mentioned though, searching the bugzilla db's is a bit of a black art, but sounds like it might be worth trying to pinpoint exactly where the kernel stopped working (ideally bisecting to find the commit) and file a ticket for it. I think you mentioned in another post that 3.18 had the problem, not just 3.19 ?

Thats correct, the bug started appearing in 3.18 (and that is my card too, mobility 4200 if that matters). I think this is the report of the bug http://lists.freedesktop.org/archive...er/073948.html http://filibusta.crema.unimi.it/~cavok/kbts/kr626.html but seems to be the dev than wrote the patch is aware of it, so I wonder why still not any fix.

I guess the only thing than I can do is trying with the bisecting thing to see what happen. Thanks for the info, at least now I know how to proceed.

[bridgman]

Thats correct, the bug started appearing in 3.18 (and that is my card too, mobility 4200 if that matters). I think this is the report of the bug http://lists.freedesktop.org/archive...er/073948.html http://filibusta.crema.unimi.it/~cavok/kbts/kr626.html but seems to be the dev than wrote the patch is aware of it, so I wonder why still not any fix.

Heinz filed a bug ticket against the Intel gfx driver...

https://bugs.freedesktop.org/show_bug.cgi?id=87330

... which was marked as a duplicate of a similar ticket...

https://bugs.freedesktop.org/show_bug.cgi?id=86679

... and both are marked as "FIXED".

Since you are using AMD hardware rather than Intel hardware, and since the problem Heinz reported seemed to be caused by Intel-specific driver code, I think you have a different problem and need a different bugzilla ticket.

[edoantonioco]

Heinz filed a bug ticket against the Intel gfx driver...
Since you are using AMD hardware rather than Intel hardware, and since the problem Heinz reported seemed to be caused by Intel-specific driver code, I think you have a different problem and need a different bugzilla ticket.

Fair enough. I have seen others having the same problem on amd hardware (on free driver), so I guess that is something to take a look. Then I guess this is actually a unknown bug for the kernel devs, what makes null my first comment.