anyone know of a resolution for the intel i915 "black screen" issue?
i haven't been following closely enough -- has there been a solution checked in for the (common) problem of recent git trees producing a totally black screen shortly after boot with onboard intel i915 graphics? i'd really like to start building kernels again for my ubuntu system but everything lately still gives me a black screen. rday -- ======================================================================== Robert P. J. Day Waterloo, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================
On Fri, Jan 14, 2011 at 11:15 AM, Robert P. J. Day <rpjday@crashcourse.ca> wrote:
i haven't been following closely enough -- has there been a solution checked in for the (common) problem of recent git trees producing a totally black screen shortly after boot with onboard intel i915 graphics?
i'd really like to start building kernels again for my ubuntu system but everything lately still gives me a black screen.
rday
robert, I haven't seen any reports of that on the opensuse list and they are using 2.6.37 kernels for factory. They did have issues like that back in 2.6.34 timeframe (ie. when os 11.3 was released), so one had to disable the new KMS driver if you had an intel graphics adapter. So I'm not sure which kernel you're worried about. Greg
On Fri, 14 Jan 2011, Greg Freemyer wrote:
On Fri, Jan 14, 2011 at 11:15 AM, Robert P. J. Day <rpjday@crashcourse.ca> wrote:
i haven't been following closely enough -- has there been a solution checked in for the (common) problem of recent git trees producing a totally black screen shortly after boot with onboard intel i915 graphics?
i'd really like to start building kernels again for my ubuntu system but everything lately still gives me a black screen.
rday
robert,
I haven't seen any reports of that on the opensuse list and they are using 2.6.37 kernels for factory.
They did have issues like that back in 2.6.34 timeframe (ie. when os 11.3 was released), so one had to disable the new KMS driver if you had an intel graphics adapter.
So I'm not sure which kernel you're worried about.
i was getting that with 2.6.37-rc7, i'm just about to try with the current git pull to see if anything's changed. i'll let you know. rday -- ======================================================================== Robert P. J. Day Waterloo, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================
On Fri, 14 Jan 2011, Greg Freemyer wrote:
On Fri, Jan 14, 2011 at 11:15 AM, Robert P. J. Day <rpjday@crashcourse.ca> wrote:
i haven't been following closely enough -- has there been a solution checked in for the (common) problem of recent git trees producing a totally black screen shortly after boot with onboard intel i915 graphics?
i'd really like to start building kernels again for my ubuntu system but everything lately still gives me a black screen.
rday
robert,
I haven't seen any reports of that on the opensuse list and they are using 2.6.37 kernels for factory.
They did have issues like that back in 2.6.34 timeframe (ie. when os 11.3 was released), so one had to disable the new KMS driver if you had an intel graphics adapter.
So I'm not sure which kernel you're worried about.
here's a report of mine from back in december: http://www.spinics.net/lists/newbies/msg41792.html so this has been going on far more recently than 2.6.34. what did disabling the KMS driver involve? i can try that as a workaround. rday -- ======================================================================== Robert P. J. Day Waterloo, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================
On Fri, Jan 14, 2011 at 5:50 PM, Robert P. J. Day <rpjday@crashcourse.ca> wrote:
On Fri, 14 Jan 2011, Greg Freemyer wrote:
On Fri, Jan 14, 2011 at 11:15 AM, Robert P. J. Day <rpjday@crashcourse.ca> wrote:
i haven't been following closely enough -- has there been a solution checked in for the (common) problem of recent git trees producing a totally black screen shortly after boot with onboard intel i915 graphics?
i'd really like to start building kernels again for my ubuntu system but everything lately still gives me a black screen.
rday
robert,
I haven't seen any reports of that on the opensuse list and they are using 2.6.37 kernels for factory.
They did have issues like that back in 2.6.34 timeframe (ie. when os 11.3 was released), so one had to disable the new KMS driver if you had an intel graphics adapter.
So I'm not sure which kernel you're worried about.
here's a report of mine from back in december:
http://www.spinics.net/lists/newbies/msg41792.html
so this has been going on far more recently than 2.6.34. what did disabling the KMS driver involve? i can try that as a workaround.
rday
Add "i915.modeset=0" to your boot command line. At least in 2.6.34 timeframe it used an alternate driver. Greg
On Fri, 14 Jan 2011, Greg Freemyer wrote:
Add "i915.modeset=0" to your boot command line. At least in 2.6.34 timeframe it used an alternate driver.
do you know *which* alternate driver, so i make sure i compile it in? rday -- ======================================================================== Robert P. J. Day Waterloo, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================
On Fri, Jan 14, 2011 at 6:08 PM, Robert P. J. Day <rpjday@crashcourse.ca> wrote:
On Fri, 14 Jan 2011, Greg Freemyer wrote:
Add "i915.modeset=0" to your boot command line. At least in 2.6.34 timeframe it used an alternate driver.
do you know *which* alternate driver, so i make sure i compile it in?
UMS, but I was using pre-compiled kernels, so I don't know what the config option was. Greg
On Fri, 14 Jan 2011, Greg Freemyer wrote:
On Fri, Jan 14, 2011 at 6:08 PM, Robert P. J. Day <rpjday@crashcourse.ca> wrote:
On Fri, 14 Jan 2011, Greg Freemyer wrote:
Add "i915.modeset=0" to your boot command line. At least in 2.6.34 timeframe it used an alternate driver.
do you know *which* alternate driver, so i make sure i compile it in?
UMS, but I was using pre-compiled kernels, so I don't know what the config option was.
victory is mine! just built a new kernel from latest git pull, changed nothing, installed, booted and it works. no idea what's changed or what fixed it. rday -- ======================================================================== Robert P. J. Day Waterloo, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================
victory is mine! just built a new kernel from latest git pull, changed nothing, installed, booted and it works. no idea what's changed or what fixed it.
Confirmed! I had the same problem on my two intel machines. At first I thought it was a misconfiguration of mine (X started and virtual consoles worked from then on, but I had no boot messages after the initial kernel mode setting). Glad to see this fixed. Alex.
On Sat, 15 Jan 2011, Alexandre Courbot wrote:
victory is mine! just built a new kernel from latest git pull, changed nothing, installed, booted and it works. no idea what's changed or what fixed it.
Confirmed! I had the same problem on my two intel machines. At first I thought it was a misconfiguration of mine (X started and virtual consoles worked from then on, but I had no boot messages after the initial kernel mode setting). Glad to see this fixed.
are you confirming that you had the same problem, or that you can see it's fixed? the blackout clearly started early in the boot, when the kernel tried to switch to high-res mode to continue printing kernel messages. suddenly, with a kernel based on a pull from the latest merge window, that switch works just fine and booting is good. but i still have no idea which commit resolved that. rday -- ======================================================================== Robert P. J. Day Waterloo, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================
On Sat, 15 Jan 2011, Robert P. J. Day wrote:
On Sat, 15 Jan 2011, Alexandre Courbot wrote:
victory is mine! just built a new kernel from latest git pull, changed nothing, installed, booted and it works. no idea what's changed or what fixed it.
Confirmed! I had the same problem on my two intel machines. At first I thought it was a misconfiguration of mine (X started and virtual consoles worked from then on, but I had no boot messages after the initial kernel mode setting). Glad to see this fixed.
are you confirming that you had the same problem, or that you can see it's fixed? the blackout clearly started early in the boot, when the kernel tried to switch to high-res mode to continue printing kernel messages. suddenly, with a kernel based on a pull from the latest merge window, that switch works just fine and booting is good. but i still have no idea which commit resolved that.
a quick perusal of the git log shows a number of recent i915-related commits. still not sure what fixed things so i'll be keeping at least a couple earlier working kernels around. rday -- ======================================================================== Robert P. J. Day Waterloo, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================
On 15/01/11 10:37, Robert P. J. Day wrote:
a quick perusal of the git log shows a number of recent i915-related commits. still not sure what fixed things so i'll be keeping at least a couple earlier working kernels around.
There was a recent mention of what I think is the issue on LWN, I tried to find out what the 'blank screen' issue was when mentioned as I have several boxes with intel graphics that haven't had issues, but couldn't find any more info, anyway excerpt and link below... "While we were on our holiday break, 2.6.37-rc8 was released on December 28. "The -rc8 release shouldn't be all that exciting. The most noticeable is probably the fact that hopefully the 'blank screen' problem with intel graphics is fixed. But on the whole, it's all just a collection of random fixes all over." See the full changelog for all the details." -lwn [1] Phil [1] http://lwn.net/Articles/420887/
On Sat, 15 Jan 2011, Philip Downer wrote:
On 15/01/11 10:37, Robert P. J. Day wrote:
a quick perusal of the git log shows a number of recent i915-related commits. still not sure what fixed things so i'll be keeping at least a couple earlier working kernels around.
There was a recent mention of what I think is the issue on LWN, I tried to find out what the 'blank screen' issue was when mentioned as I have several boxes with intel graphics that haven't had issues, but couldn't find any more info, anyway excerpt and link below...
"While we were on our holiday break, 2.6.37-rc8 was released on December 28. "The -rc8 release shouldn't be all that exciting. The most noticeable is probably the fact that hopefully the 'blank screen' problem with intel graphics is fixed. But on the whole, it's all just a collection of random fixes all over." See the full changelog for all the details." -lwn [1]
Phil
i remember reading that, but i *tried* a kernel based on -rc8 and it still had the same issue. at this point, i'm going to count my blessings and just hope it's been resolved once and for all. rday -- ======================================================================== Robert P. J. Day Waterloo, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================
are you confirming that you had the same problem, or that you can see it's fixed?
Both. At first I thought I forgot to enable something in my kernel configuration and did not care that much about it, then I saw your message and tried again with the latest git, to see the problem was fixed.
the blackout clearly started early in the boot, when the kernel tried to switch to high-res mode to continue printing kernel messages. suddenly, with a kernel based on a pull from the latest merge window, that switch works just fine and booting is good.
Yes, this is exactly the behavior I had. Alex.
participants (4)
-
Alexandre Courbot -
Greg Freemyer -
Philip Downer -
Robert P. J. Day