| 1 |
Hi, |
| 2 |
|
| 3 |
On Fri, 18 Nov 2005 00:56:18 +1000 |
| 4 |
"Alan E. Davis" <lngndvs@×××××.com> wrote: |
| 5 |
|
| 6 |
> I backed down to 1024xsomething: vertical lines were scalloped/wavy. Someone |
| 7 |
> mentioned this would be a timing issue, but I don't know what I'd do to |
| 8 |
> microadjust timing? xvidtune? I'll try it. |
| 9 |
|
| 10 |
That won't help. I'm currently using a mystique and am having the very |
| 11 |
same problem. It's _very_ easy to circumvent but requires a recompile |
| 12 |
of xorg. The fix still hasn't made it into xorg CVS. Maybe time to |
| 13 |
write another bug report. But i think I remember having written one for |
| 14 |
XFree86, which behaved the same way. I was told that it was merely a |
| 15 |
timing issue but AFAIK this isn't configurable und it was fixed in 4.5. |
| 16 |
But I never tried that out. |
| 17 |
|
| 18 |
To "fix" the issue, I usually did the following: |
| 19 |
1. $ emerge xorg-x11 |
| 20 |
waited after the moment when everything is unpacked and patched and hit CTRL-Z |
| 21 |
2. $ vi /var/tmp/portage/xorg-x11.../work/xc/programs/Xserver/hw/xfree86/drivers/mga/mga_storm.c |
| 22 |
|
| 23 |
search for the lines: |
| 24 |
|/* MGAISBUSY() reportedly causes a freeze for Mystique revision 2 and older */ |
| 25 |
| if (!(pMga->Chipset == PCI_CHIP_MGA1064 && (pMga->ChipRev >= 0 && pMga->ChipRev <= 2))) |
| 26 |
| while(MGAISBUSY()); |
| 27 |
|
| 28 |
Reading the comment exactly, it is easy to fix it (i.e. it couldn't work for Rev. 2): |
| 29 |
| if (!(pMga->Chipset == PCI_CHIP_MGA1064 && (pMga->ChipRev >= 0 && pMga->ChipRev < 2))) |
| 30 |
(i.e., changing "<= 2" to "< 2") |
| 31 |
|
| 32 |
3. $ fg |
| 33 |
to continue the suspended emerge job |
| 34 |
|
| 35 |
I'd suggest making a binary package and saving it somewhere... |
| 36 |
|
| 37 |
|
| 38 |
-hwh |
| 39 |
-- |
| 40 |
gentoo-user@g.o mailing list |
Archives