Linux 2.6.3-rc4 Massive strange corruption with new radeonfb
Upon bootup, radeonfb is obviously not initializing the hardware correctly. Massive amounts of random-looking garbage, plus a weird effect I've never seen before, like someone pouring milk _up_ the screen. (Yeah, it's the best I could come up with)
It's a Dell Inspiron 8600 with Mobile Radeon 9600 and 1920x1200 LCD.
Here's the relevant parts from dmesg:
radeonfb_pci_register BEGIN
radeonfb: probed SDR SGRAM 131072k videoram
radeonfb: mapped 16384k videoram
radeonfb: Found Intel x86 BIOS ROM Image
radeonfb: Retreived PLL infos from BIOS
radeonfb: Reference=27.00 MHz (RefDiv=6) Memory=337.00 Mhz, System=243.00 MHz 1 chips in connector info
- chip 1 has 2 connectors
* connector 0 of type 2 (CRT) : 2300
* connector 1 of type 4 (DVI-D) : 4210
Starting monitor auto detection...
radeonfb: I2C (port 1) ... not found
radeonfb: I2C (port 2) ... not found
radeonfb: I2C (port 3) ... not found
radeonfb: I2C (port 4) ... not found
radeonfb: Reversed DACs detected
radeonfb: Reversed TMDS detected
radeonfb: I2C (port 2) ... not found
radeonfb: I2C (port 4) ... not found
Non-DDC laptop panel detected
radeonfb: I2C (port 3) ... not found
radeonfb: I2C (port 4) ... not found
radeonfb: Monitor 1 type LCD found
radeonfb: Monitor 2 type no found
radeonfb: panel ID string:
radeonfb: detected LVDS panel size from BIOS: 0x0
BIOS provided panel power delay: 0
Scanning BIOS table ...
Didn't find panel in BIOS table !
Guessing panel info...
radeonfb: Assuming panel size 8x1
radeonfb: Power Management enabled for Mobility chipsets
radeonfb: ATI Radeon NP SDR SGRAM 128 MBradeonfb_pci_register END
hStart = 664, hEnd = 760, hTotal = 800
vStart = 409, vEnd = 411, vTotal = 450h_total_disp = 0x4f0063 hsync_strt_wid = 0x8c0292v_total_disp = 0x18f01c1 vsync_strt_wid = 0x820198
pixclock = 39729
freq = 2517
post div = 0x3fb_div = 0x2dppll_div_3 = 0x3002dlvds_gen_cntl: 003cffa5
Obviously, it isn't detecting the screen size properly. I even tried hard-coding the resolution in the detection code, to no avail. Any suggestions?
Thanks.
Charles
Quoted 6 lines · Charles Johnston
Upon bootup, radeonfb is obviously not initializing the hardware correctly. Massive amounts of random-looking garbage, plus a weird effect I've never seen before, like someone pouring milk _up_ the screen. (Yeah, it's the best I could come up with)
It's a Dell Inspiron 8600 with Mobile Radeon 9600 and 1920x1200 LCD.
Looks like the driver cannot find any info about your flat panel in the BIOS ROM image. I suppose we can thank DELL for hacking the BIOS in ways that aren't compatible with all others laptops...
Can you try commenting out the call to radeon_map_ROM() and let it look for the RAM based BIOS instead ? Let me know...
Quoted 63 lines
Here's the relevant parts from dmesg:
radeonfb_pci_registerBEGIN
radeonfb: probed SDR SGRAM 131072k videoram
radeonfb: mapped 16384k videoram
radeonfb: Found Intel x86 BIOS ROM Image
radeonfb: Retreived PLL infos from BIOS
radeonfb: Reference=27.00 MHz (RefDiv=6) Memory=337.00 Mhz, System=243.00 MHz
1 chips in connector info
- chip 1 has 2 connectors
* connector 0 of type 2 (CRT) : 2300
* connector 1 of type 4 (DVI-D) : 4210
Starting monitor auto detection...
radeonfb: I2C (port 1) ... not found
radeonfb: I2C (port 2) ... not found
radeonfb: I2C (port 3) ... not found
radeonfb: I2C (port 4) ... not found
radeonfb: Reversed DACs detected
radeonfb: Reversed TMDS detected
radeonfb: I2C (port 2) ... not found
radeonfb: I2C (port 4) ... not found
Non-DDC laptop panel detected
radeonfb: I2C (port 3) ... not found
radeonfb: I2C (port 4) ... not found
radeonfb: Monitor 1 type LCD found
radeonfb: Monitor 2 type no found
radeonfb: panel ID string:
radeonfb: detected LVDS panel size from BIOS:0x0BIOS provided panel power delay: 0
Scanning BIOS table ...
Didn't find panel in BIOS table !
Guessing panel info...
radeonfb: Assuming panel size 8x1
radeonfb: Power Management enabled for Mobility chipsets
radeonfb: ATI Radeon NP SDR SGRAM 128 MBradeonfb_pci_registerENDhStart = 664, hEnd = 760, hTotal = 800
vStart = 409, vEnd = 411, vTotal = 450h_total_disp=0x4f0063hsync_strt_wid=0x8c0292v_total_disp=0x18f01c1vsync_strt_wid=0x820198
pixclock = 39729
freq = 2517
post div =0x3fb_div=0x2dppll_div_3=0x3002dlvds_gen_cntl:003cffa5Obviously, it isn't detecting the screen size properly. I even tried hard-coding the resolution in the detection code, to no avail. Any suggestions?
Thanks.
Charles
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Signature · 1 line
Benjamin Herrenschmidt <[email protected]>
Quoted 8 lines · Benjamin Herrenschmidt
Quoted 4 lines · Charles Johnston
Upon bootup, radeonfb is obviously not initializing the hardware correctly. Massive amounts of random-looking garbage, plus a weird effect I've never seen before, like someone pouring milk _up_ the screen. (Yeah, it's the best I could come up with)
It's a Dell Inspiron 8600 with Mobile Radeon 9600 and 1920x1200 LCD.
Quoted 7 lines
Looks like the driver cannot find any info about your flat panel in the BIOS ROM image. I suppose we can thank DELL for hacking the BIOS in ways that aren't compatible with all others laptops...
Can you try commenting out the call to
radeon_map_ROM()and let it look for the RAM based BIOS instead ? Let me know...
Ok, it worked fine with that line commented out. I can switch vt's, be in X, etc. no problems.
The only issue I see is when I do a 'clear' on the vt, it doesn't clear the text, but blanks every nth row of pixels. Switching vt's and back clears the screen.
There are also a few rows of garbage pixels at the bottom that linger across vt switches.
Charles Johnston
[email protected]
Quoted 3 lines
Ok, it worked fine with that line commented out. I can switch vt's, be in X, etc. no problems.
Can you send me the dmesg log still please ?
Quoted 3 lines
The only issue I see is when I do a 'clear' on the vt, it doesn't clear the text, but blanks every nth row of pixels. Switching vt's and back clears the screen.
Does this happen even without using XFree ? There is a known problem with clears when switching _from_ XFree... I'm working on a fix.
Quoted 2 lines
There are also a few rows of garbage pixels at the bottom that linger across vt switches.
Yes, same as above afaik
Quoted 5 lines · Benjamin Herrenschmidt
Quoted 1 line
Ok, it worked fine with that line commented out. I can switch vt's, be in X, etc. no problems.
Can you send me the dmesg log still please ?
Here it is:
radeonfb_pci_register BEGIN
radeonfb: probed SDR SGRAM 131072k videoram
radeonfb: mapped 16384k videoram
radeonfb: Retreived PLL infos from BIOS
radeonfb: Reference=27.00 MHz (RefDiv=6) Memory=337.00 Mhz, System=243.00 MHz 1 chips in connector info
- chip 1 has 2 connectors
* connector 0 of type 2 (CRT) : 2300
* connector 1 of type 4 (DVI-D) : 4210
Starting monitor auto detection...
radeonfb: I2C (port 1) ... not found
radeonfb: I2C (port 2) ... not found
radeonfb: I2C (port 3) ... not found
radeonfb: I2C (port 4) ... not found
radeonfb: Reversed DACs detected
radeonfb: Reversed TMDS detected
radeonfb: I2C (port 2) ... not found
radeonfb: I2C (port 4) ... not found
Non-DDC laptop panel detected
radeonfb: I2C (port 3) ... not found
radeonfb: I2C (port 4) ... not found
radeonfb: Monitor 1 type LCD found
radeonfb: Monitor 2 type no found
radeonfb: panel ID string: U0674^B<9A>M1LW02
radeonfb: detected LVDS panel size from BIOS: 1920x1200
BIOS provided panel power delay: 1000
radeondb: BIOS provided dividers will be usedref_divider = 6post_divider = 1fbk_divider = 48
Scanning BIOS table ...
320 x 350
320 x 400
320 x 400
320 x 480
400 x 600
512 x 384
640 x 350
640 x 400
640 x 475
640 x 480
720 x 480
720 x 576
800 x 600
848 x 480
1024 x 768
1280 x 1024
1280 x 768
1280 x 800
1600 x 1200
1680 x 1050
1920 x 1200
Found panel in BIOS table:
hblank: 264
hOver_plus: 96
hSync_width: 32
vblank: 35
vOver_plus: 2
vSync_width: 6
clock: 16175
Setting up default mode based on panel info
radeonfb: Power Management enabled for Mobility chipsets
radeonfb: ATI Radeon NP SDR SGRAM 128 MBradeonfb_pci_register END
hStart = 2016, hEnd = 2048, hTotal = 2184
vStart = 1202, vEnd = 1208, vTotal = 1235h_total_disp = 0xef0110 hsync_strt_wid = 0x407dav_total_disp = 0x4af04d2 vsync_strt_wid = 0x604b1
pixclock = 6182
freq = 16175lvds_gen_cntl: 003cffa5
Quoted 7 lines
Quoted 2 lines
The only issue I see is when I do a 'clear' on the vt, it doesn't clear the text, but blanks every nth row of pixels. Switching vt's and back clears the screen.
Does this happen even without using XFree ? There is a known problem with clears when switching _from_ XFree... I'm working on a fix.
Yes, the blanking issue is still present without XFree.
Quoted 6 lines
Quoted 2 lines
There are also a few rows of garbage pixels at the bottom that linger across vt switches.
Yes, same as above afaik
The bottom garbage is not present without XFree.
Quoted 3 lines
Yes, the blanking issue is still present without XFree.
Ok, that would be the fillrect() operation not working properly.
I'll look at it.
Ben.