Time to Break Out the Virtual Screwdrivers?

[Orethrius]

I recently purchased an HP dv8210us Pavilion notebook (Leviathan, per my sig) to act as a replacement for Frankystein after its screen failed.  I got most of the way through X config, loading Fluxbox, and so forth - but I'm having the strangest issues with X (though this in no way impacts the console, so I'm convinced it's a misconfig in xorg.conf).  Under any WM, when I use xterm or lynx or anything that uses text-display for that matter, entire vertical portions of text are missing.  I'm thinking this is a modeline issue, but I've found a number of conflicting "solutions" to this issue and would like a rundown of what REALLY needs to be done.  Additionally, the color depth on my 200M won't go higher than 4 (16-bit IIRC), but I suspect that's due to using the generic VGA drivers (X doesn't recognise ati or radeon for some reason, probably my fault).  Here's what I have thus far:

Code: [Select]

Section "ServerLayout"
Identifier "X.org Configured"
Screen  0  "Screen0" 0 0
InputDevice "Mouse0" "CorePointer"
InputDevice "Keyboard0" "CoreKeyboard"
EndSection

Section "Files"
RgbPath  "/usr/share/X11/rgb"
ModulePath   "/usr/lib64/xorg/modules"
FontPath "/usr/share/fonts/misc"
FontPath "/usr/share/fonts/75dpi"
FontPath "/usr/share/fonts/100dpi"
FontPath "/usr/share/fonts/TTF"
FontPath "/usr/share/fonts/Type1"
EndSection

Section "Module"
Load  "glx"
Load  "extmod"
Load  "xtrap"
Load  "record"
Load  "dbe"
Load  "dri"
Load  "freetype"
Load  "type1"
EndSection

Section "InputDevice"
Identifier  "Keyboard0"
Driver  "kbd"
EndSection

Section "InputDevice"
Identifier  "Mouse0"
Driver  "mouse"
Option "Protocol" "auto"
Option "Device" "/dev/input/mice"
Option "ZAxisMapping" "4 5 6 7"
EndSection

Section "Monitor"
Identifier   "Monitor0"
VendorName   "Monitor Vendor"
ModelName "Monitor Model"
Modeline "1440x900" 119.12  1440 1728 1760 1840  900 1052 1058 1080
EndSection

Section "Device"
### Available Driver options are:-
### Values: : integer, : float, : "True"/"False",
### : "String", : " Hz/kHz/MHz"
### [arg]: arg optional
#Option "ShadowFB"   # []
#Option "KGAUniversal"   # []
Identifier  "Card0"
Driver  "vga"
VendorName  "ATI Technologies Inc"
BoardName   "ATI Radeon XPRESS 200M 5955 (PCIE)"
BusID   "PCI:1:5:0"
EndSection

Section "Screen"
Identifier "Screen0"
Device "Card0"
Monitor "Monitor0"
SubSection "Display"
Viewport   0 0
Depth 1
EndSubSection
SubSection "Display"
Viewport   0 0
Depth 4
EndSubSection
SubSection "Display"
Viewport   0 0
Depth 8
EndSubSection
SubSection "Display"
Viewport   0 0
Depth 15
EndSubSection
SubSection "Display"
Viewport   0 0
Depth 16
EndSubSection
SubSection "Display"
Viewport   0 0
Depth 24
EndSubSection
EndSection

Any help would be greatly appreciated (btw this is an x86_64 build, the card - as indicated above - is a Radeon 200M, 9000 is the internal ref IIRC).

[piratePenguin]

I never have a Modeline in my xorg.conf, I always specify the HorizSync and VertRefresh in the Monitor section and then have a Modes "1024x768" line in the Display subsection of the Screen section, I wonder is there a difference?..

[H_TeXMeX_H]

What distro are you running ? Is there any way to auto configure it ?

I suppose I can't help much, I'm horrible at configuring things myself.

[_kill__bill]

1) Wouldn't this belong under tech support?

2) Try a different set of drivers or a different X server.

[Orethrius]

I never have a Modeline in my xorg.conf, I always specify the HorizSync and VertRefresh in the Monitor section and then have a Modes "1024x768" line in the Display subsection of the Screen section, I wonder is there a difference?..

That's fine and good, but I'm dealing with a 17" widescreen here.  It doesn't display 4:3 ratios right, so something needs to be done to switch it to 16:9 and I'm not sure what.

What distro are you running ? Is there any way to auto configure it ?

Gentoo, installed from the AMD64 2006.0 Universal CD.  Unfortunately, that IS the auto-configure output (X -configure) except that I manually changed /dev/mouse to /dev/input/mice to fix input support.  That Modeline is experimental, and apparently non-functional.

1) Wouldn't this belong under tech support?

Moved thread, thanks.

2) Try a different set of drivers or a different X server.

Additionally, the color depth on my 200M won't go higher than 4 (16-bit IIRC), but I suspect that's due to using the generic VGA drivers (X doesn't recognise ati or radeon for some reason, probably my fault).

The only other thing I can think of is to use another card's drivers, but I can't help feeling that's the WORST thing to do.  I'm not going to try XFree or XModular because, like I said, this problem directly affects neither the execution of programs nor the console (to say nothing of the fact that it took me twenty hours to persuade it to work in the first place).  It's really an asthetic problem, but it's bad enough to prevent me from making any headway in text-based X programs.  I don't believe there's no solution to this somewhere, it just seems that all the sites that provide one disagree on it.

[H_TeXMeX_H]

If the modeline is fucked try this site:

http://xtiming.sourceforge.net/cgi-bin/xtiming.pl

[Orethrius]

Okay, I managed to get X back to normal (it was the VGA drivers) but now I can't get the ATi driver module to compile.  I get the following error message from the 2.6.x Makefile:

Code: [Select]

Makefile:49: *** mixed implicit and normal rules.  Stop.
Here is the full Makefile listing:

Code: [Select]

GCC_VER_MAJ      = $(shell $(CC) --version | head -n 1 | sed -e 's/[^0-9. ]//g;s/^ *//;s/^\(.\)\..*$$/\1/')

obj-m           += fglrx.o
fglrx-libs      += libfglrx_ip.a.GCC$(GCC_VER_MAJ)
fglrx-c-objs    += firegl_public.o
fglrx-objs      += $(fglrx-c-objs) $(fglrx-libs)
fglrx-hdrs      += firegl_public.h
drm-hdrs        += drm.h drmP.h drm_os_linux.h drm_proc.h drm_compat.h

ifeq ($(PAGE_ATTR_FIX),)
PAGE_ATTR_FIX   =1
endif

EXTRA_CFLAGS    += \
                -I$(PWD) \
                -DFGL \
                -DFGL_LINUX \
                -DFGL_GART_RESERVED_SLOT \
                -DFGL_LINUX253P1_VMA_API \
                -DPAGE_ATTR_FIX=$(PAGE_ATTR_FIX) \

ifeq ($(KERNELRELEASE),)
# on first call from remote location we get into this path
# whilst on second call all is managed by the embedding kernel makefile

KVER            = $(shell uname -r)
# KVER            = 2.6.0-0.test1.1.26custom

ifeq (1,1)
KDIR            = /lib/modules/$(KVER)/build
else
KDIR            = /usr/src/linux-$(KVER)
endif
PWD             = $(shell pwd)

fglrx-cfiles    = ${fglrx-c-objs:.o=.c}

# default:: kmod_build

kmod_build:: $(fglrx-libs) $(fglrx-cfiles) $(fglrx-hdrs) $(drm-hdrs)
$(MAKE) -C $(KDIR) SUBDIRS=$(PWD) modules

%.c:
@ln -s ../$@

%.h:
@ln -s ../$@

[i]%.GCC$(GCC_VER_MAJ):[/i]
@ln -s ../$@

# distclean:: clean

clean::
rm -f *.c *.h *.o *.ko *.GCC* .??*

endif

# ### EOF ###

I'm at a complete loss as to what needs to be fixed here to move on.  Anyone?  

EDIT: Italicized line 49 for readability.

EDIT2: Manually changed each GCC_VER_MAJ to the current major version of 3, ATi must've FUBARed the shell script as now it (seems to have) MODPOSTed.

EDIT3: Finally got the ati-drivers to emerge.  Now when I try to use fglrx on my 64-build I get the Black Screen of Death (though oddly enough, I can still shell into it).  NICE ONE ATI.  If this isn't fixed in the next month, well - let's just say I've a blister-packed icepick and a VERY short list of demonstration venues. :mad: