From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2.1) Gecko/20021218 Description of problem: Bad: IBM ThinkPad A30p LCD blanks on graphical install. Bad: IBM ThinkPad A30p LCD blanks on text install Bad: IBM ThinkPad A30p LCD blanks following "linux acpi=off" Good: IBM ThinkPad A30p Toggling between external and internal display (using Fn-F7) resets display. Good: IBM ThinkPad A30p works well with psyche. Bad: IBM ThinkPad A30p reports failed mediacheck for phoebe*.iso Good: Failed media on IBM ThinkPad A30p passes on Compaq Presario 5838. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: Download phoebe*.iso: Verify MD5 Create CD set using command-line: "cdrecord -v -eject speed=1 dev=0,0,0 <filename.iso>" 1.Select CD-Rom as Boot Device on IBM ThinkPad A30p during POST. 2.Select any mode of installation (Graphical,Text...) 3.Observe boot-loader output..followed by blank LCD Display. 4.Toggle internal=>external=>internal (Fn-F7...Fn-F7). 5.Observe "mediacheck" for boot CD-Rom. 6.Perform "mediacheck" for boot CD-Rom. 7.Observe "failed" media results. 8.(Power-Down) 9.Transfer CD-Rom to Compaq Presario 5838. 10. Boot CD-Rom 11. Perform "mediacheck" 12. Observe "PASS" media results. 13. Three-disk-set (phoebe) successfully installed on Compaq. Actual Results: Did NOT install phoebe due to "FAILED" media check on IBM ThinkPad A30p. Did install phoebe on Compaq Presario 5838 with "PASSED" media. Additional Note: RedHat "psyche" was installed via FTP on (same) IBM ThinkPad A30p but only after disk1 "PASSED" but disk2 and disk3 failed. Additional info:
I encountered the same screen blanking problem on my ThinkPad T30 during installation. In addition I have the same problem with RH8.0 during normal operation. I can 'fix' it by putting the machine into suspend by pressing the suspend key or closing the lid and bringing it back again (display cycle with Fn-F7 does not work on the T30). But this does not work during phoebe installation. Instead I had to keep moving the mouse during installation to prevent the screen blanking from occuring. The screen blanking is not a problem with normal operation after install on phoebe. I suspect it only happens with APM enabled, and phoebe uses ACPI. So I guess that the phoebe redhat-boot kernel used during install has APM.
Some more information. I have encountered this screen blanking problem with several generations ThinkPad's (at least the T20, T21 and T30) and a host of Red Hat Linux releases. But generally it has not been a problem during installation, but rather after install is done in either text or graphical mode. What happens is that if your not in ACPI mode that if you don't use the keyboard/mouse for a while is that the machine blanks the LCD (black screen) and turns off the LCD backlight. When you then press a key/move the mouse it turns the LCD backlight back on, but the display remains black. In most cases forcing a display refresh by putting the machine into suspend and bringing it back clears things up, but not always. You might try to start the install with apm=off as a kernel boot option.
Tried "linux apm=off" (No Joy)...I too have experienced 'screen flicker' and 'loss-of-backlight' on the A30p but with BIOS upgrade from IBM site and move to newer XFree86 (psyche) the effects were significantly reduced. The A30p sometimes fails to respond if left on overnight. F.W.I.W: Hardware accelerated X works well on the ATI Mobility RADEON M6 with psyche..but I lost sound for 'TuxRacer' (worked on 7.3..Valhalla...but desktop sounds continue to function)
I have never had a 'loss of backlight', rather the contrary. backlight is the only thing that comes back on after the the screen blanks and I press a key. I have seen screen flicker, but that only happens when I use an AC adapter not meant for the ThinkPad (Amp rating to low) in combination with a drained battery, so it has problems getting enough power out of the AC adapter.
>Bad: IBM ThinkPad A30p LCD blanks on graphical install. >Bad: IBM ThinkPad A30p LCD blanks on text install >Bad: IBM ThinkPad A30p LCD blanks following "linux acpi=off" If this occurs with or without XFree86 involved, then I am inclined to believe it is not an XFree86 related problem at all. More likely it is APM/ACPI/weirdo hardware issue, or a kernel issue with APM or somesuch. >I have encountered this screen blanking problem with several generations >ThinkPad's (at least the T20, T21 and T30) and a host of Red Hat Linux >releases. But generally it has not been a problem during installation, but >rather after install is done in either text or graphical mode. Again, if it occurs when XFree86 is not even part of the picture, then that implies it is not an XFree86 problem, but is a hardware/kernel/apm/acpi issue of some kind. Reassigning to kernel component.
I just installed Phoebe2 (8.0.93) where ACPI is no longer enabled. It still has the same screen blanking problem as the first public beta during install. But now the problem is also back after install (like in RH8.0 and before) due to the fact that ACPI is disabled ;-((
I spoke to someone at EmperorLinux at LinuxWorld (they resell laptops with Linux installed, including the T30). I was told to go into the BIOS and disable Screen Blanking to resolve the issue. I found the option under 'Config' -> 'Power' and it did indeed solve the screen blanking problem!.
Ahh.. so I was right. ;o) It is an APM/ACPI/BIOS/weirdo problem that is hardware specific. ;o)
BIOS issue confirmed. Closing
Actually, I suspect (untested by me) that the following patch which was recently included in 2.5 bitkeeper fixes the issue http://linux.bkbits.net:8080/linux-2.5/cset@1.1243.52.9? nav=index.html|ChangeSet@-7d From: Samuel Thibault <Samuel.Thibault> Some fixes for console blanking: on some laptops, doing VESA blanking after the bios did an apm blanking because of a screen closure thrashes the recovery (the video board doesn't seem to have synchronisation registers correctly initialized, since the LCD panel progressively turns white, maybe damaging it ?). I hence moved the schedule for vesa powerdown after the apm blank hook call, so that if it succeeds, it won't be called. I also moved the apm unblank & palette restoration after the vesa unblank, to have a more lifo scheme (also required, or the screen remains black). Btw, why del_timer_sync was called twice in timer_do_blank_screen when vesa_off_interval==0 ? --- 1.45/drivers/char/vt.c Wed May 7 23:26:53 2003 +++ 1.46/drivers/char/vt.c Thu Jun 5 23:36:44 2003 @@ -2712,14 +2712,7 @@ hide_cursor(currcons); if (!from_timer_handler) del_timer_sync(&console_timer); - if (vesa_off_interval) { - console_timer.function = vesa_powerdown_screen; - mod_timer(&console_timer, jiffies + vesa_off_interval); - } else { - if (!from_timer_handler) - del_timer_sync(&console_timer); - console_timer.function = unblank_screen_t; - } + console_timer.function = unblank_screen_t; save_screen(currcons); /* In case we need to reset origin, blanking hook returns 1 */ @@ -2730,6 +2723,12 @@ if (console_blank_hook && console_blank_hook(1)) return; + + if (vesa_off_interval) { + console_timer.function = vesa_powerdown_screen; + mod_timer(&console_timer, jiffies + vesa_off_interval); + } + if (vesa_blank_mode) sw->con_blank(vc_cons[currcons].d, vesa_blank_mode + 1); } @@ -2771,12 +2770,12 @@ } console_blanked = 0; - if (console_blank_hook) - console_blank_hook(0); - set_palette(currcons); if (sw->con_blank(vc_cons[currcons].d, 0)) /* Low-level driver cannot restore -> do it ourselves */ update_screen(fg_console); + if (console_blank_hook) + console_blank_hook(0); + set_palette(currcons); set_cursor(fg_console); }