Intel Graphics Driver Likely To Re-Enable Fastboot By Default

[phoronix]

Phoronix: Intel Graphics Driver Likely To Re-Enable Fastboot By Default

For the past five years or so has been work on Intel DRM "Fastboot" support and it's looking like this feature may finally be re-enabled by default...

http://www.phoronix.com/scan.php?page=news_item&px=i915-Fastboot-Default-2017

[uid313]

So it took them five years to fix this?

[caligula]

Amazing. Maybe in few years the Atom mobile SoCs won't crash all the time anymore.

[wdb974]

Amazing. Maybe in few years the Atom mobile SoCs won't crash all the time anymore.

What examples do you have in mind?

[yoshi314]

So it took them five years to fix this?

i had a bug around kernel 3.0 which caused modesetting to lock up the machine. it got fixed in 3.5, broken again and fixed in 3.7. take into account how long does it take to release just one major kernel version and do the math.

and since witout modesetting intel driver would not work at all, i had to use vesa driver on subpar resolution and no 2d acceleration with a full hd capable display all that time.

ever since then, i am somewhat wary of intel graphics. they have enterprise bug fixing cycle, even in their opensource projects. when stuff works, it's okay. when it breaks, better get a backup non-intel card.

[caligula]

What examples do you have in mind?

How about this https://bugzilla.kernel.org/show_bug.cgi?id=109051

[Junix]

I'm still waiting for Intel to fix broken Wake on LAN on Skylake NUCs, probably will happen with Kernel 5.20

[uid313]

i had a bug around kernel 3.0 which caused modesetting to lock up the machine. it got fixed in 3.5, broken again and fixed in 3.7. take into account how long does it take to release just one major kernel version and do the math.

and since witout modesetting intel driver would not work at all, i had to use vesa driver on subpar resolution and no 2d acceleration with a full hd capable display all that time.

ever since then, i am somewhat wary of intel graphics. they have enterprise bug fixing cycle, even in their opensource projects. when stuff works, it's okay. when it breaks, better get a backup non-intel card.

I've only used Intel Core i7 4770K "Haswell" on Ubuntu, and I've never had any issues.

[linuxgeex]

So it took them five years to fix this?

I think it's more to do with moving Xorg's i915 driver to KMS, which has been a distro policy thing more than a 'how long did it take them'. I could be wrong. There's been several generations of i915 hardware and Intel has been trying to make it support several releases of X, Mir and Wayland, with and without kernel modesetting, so I really wouldn't blame them if they took 5 years just to kiss a bunch of legacy kernel and userspace drivers goodbye. They needed to keep it compatible for a minimum of 4 years plus an LTS kernel plus the release period of the major distros... 5 starts to seem very fast.

[smartalgorithm]

where should i find this option to enable it?
thanks