Differences between revisions 70 and 71
Revision 70 as of 2006-11-04 19:45:30
Size: 8300
Editor: EddyPetrisor
Comment: that problem is gone for quite a long time now
Revision 71 as of 2006-12-05 13:57:05
Size: 9692
Editor: ?fiandro
Comment: Updated page accordingly to what will be presumably left undone in Etch
Deletions are marked like this. Additions are marked like this.
Line 3: Line 3:
Overview of open issues and things to be done for the ["DebianInstaller/GUI"]. This is an overview of some of the bugs and issues that presumably will be left open or workarounded only in Etch and that i hope will be fixed later in a point release.
Line 5: Line 5:
A complete list of bugs affecting the g-i can be found in the [http://bugs.debian.org/cdebconf-gtk-udeb bug reports against cdebconf-gtk-udeb] page. This is no way meant to be an exaustive list of every bug somehow related to the graphical installer, for a complete buglist see packages [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=rootskel-gtk rootskel-gtk], [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=directfb directfb], [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=libdirectfb-0.9-25-udeb libdirectfb-0.9-25-udeb], [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=libgtk-directfb-2.0-0 libgtk-directfb-2.0-0], [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=libgtk-directfb-2.0-0 libgtk-directfb-2.0-0-udeb], [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=rootskel rootskel], [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=rootskel-gtk rootskel-gtk]
Line 8: Line 8:

= Architecture indipendent issues =

* Input is not handled in UTF-8 mode, some letteres cannot be correctly typed. [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=401296 #401296]

* Support dor dead keys was added in DFB post 0.9.25, but GTKDFB does not handle deadkeys anyway, so backporting the DrectFB patch is useless ATM [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=394871 #394871]

* Touchpad support in linux_input was added to DFB in CVS and then backported to 0.9.25-5 via a patch i wrote; Ville Sryala produced an updated, better, patch which unluckily crashes on some systems and hence we weren't able to include in our sources.
As we want to give best support possible to laptop users Ville's patch should be fixed and bacported in our sources. [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=400579 #400579]

* Diagnostic tool dfbinfo is now provided by libdirectfb-udeb, later we should move it to a specific udeb, possibly with the df_input tool [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=390437 #390437]

* A cdebconf-gtk-entropy plugin is needed if we want to be able to install on encrypted volumes with the the GTK frontend too.
Line 14: Line 27:
 * No known bugs affecting i386 is known ATM :)  * For historical reasons the vga16fb module is still loaded even if a framebuffer device is already provided by vesafb: this forced us to check that DirectFB actually accesses the framebuffer device provided by vesafb , better would be not starting at all two fb devices concurrently (this also wastes little memory). [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=401685 #401685]
Line 17: Line 30:
 * solve the console switching problem on AMD64 [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=373253 #373253]  * The console switching problem on AMD64 [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=373253 #373253] was workarounded by adding libgcc1.so.1 via the EXTRAFILES option available in the installer.
Proper fix requires adding that library via an udeb, or removing pthread_cancel() calls from DirectFB source code.
Line 20: Line 34:
* On PPC the framebuffer device has a fixed size that cannot be set at boot like it's done on i386, so the GTK frontend should adapt itself to go "fullscreen". [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=401693 #401693]
Line 21: Line 36:
 * Builds with the tentative fix from Attilio are available at [http://people.debian.org/~luther/g-i], please [wiki:Self:DebianInstaller/GUIPowerPCTesting test them]. * DirectFB's linux_input input module seems particulary willing to crash on this architecture, and as a result we had to selectively disable such module basing on experiments on on various PPC boxes.
Line 23: Line 38:
PPC has been for a long time the toughest architecture of the three to support because DirectFrameBuffer seems particulary willing to mess up with input devices on PPCs.
Right now HW acceleration provided by DFB seems to be broken for almost every suported video chipset on PPCs, so it was forced off by default.
Line 47: Line 60:
 * The list selection is invisible bug seems to be gone on both radeon 9800 and 9200 on pegasos, when using disable-module=radeon.

 * Going to the console on pegasos/radeon9200 leaves a garbagy font. Using disable-modules=radeon fixes this issue.

= TODO =
= Ideas for future development =
Line 61: Line 70:
 * Screen resolution.
   * Currently fixed at 800x600. Should be made flexible (adapt to resolution set by system or user).
  
* Interface should adapt itself so that text blocks are never too wide for optimal readability.
 * Interface should adapt itself so that text blocks are never too wide for optimal readability.
Line 65: Line 72:
 * Progress bar in separate window.
Line 72: Line 78:


== Reduce size of initrds ==
The initrds are currently very large which also results in fairly high memory requirements to run the graphical installer. Size has thus far not been a major concern, our focus has been on getting things working.

Options to reduce initrd size.
 * Library reduction for graphical libraries.
 * Remove unused files from udebs.
 * Reduction of space used by fonts (see above).

'''Update'''
The size of the initrds is now down to around 10MB. Further reduction would be nice, but it is no longer a major issue.

Introduction

This is an overview of some of the bugs and issues that presumably will be left open or workarounded only in Etch and that i hope will be fixed later in a point release.

This is no way meant to be an exaustive list of every bug somehow related to the graphical installer, for a complete buglist see packages [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=rootskel-gtk rootskel-gtk], [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=directfb directfb], [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=libdirectfb-0.9-25-udeb libdirectfb-0.9-25-udeb], [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=libgtk-directfb-2.0-0 libgtk-directfb-2.0-0], [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=libgtk-directfb-2.0-0 libgtk-directfb-2.0-0-udeb], [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=rootskel rootskel], [http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=rootskel-gtk rootskel-gtk]

?TableOfContents([2])

Architecture indipendent issues

* Input is not handled in UTF-8 mode, some letteres cannot be correctly typed. [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=401296 #401296]

* Support dor dead keys was added in DFB post 0.9.25, but GTKDFB does not handle deadkeys anyway, so backporting the DrectFB patch is useless ATM [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=394871 #394871]

* Touchpad support in linux_input was added to DFB in CVS and then backported to 0.9.25-5 via a patch i wrote; Ville Sryala produced an updated, better, patch which unluckily crashes on some systems and hence we weren't able to include in our sources. As we want to give best support possible to laptop users Ville's patch should be fixed and bacported in our sources. [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=400579 #400579]

* Diagnostic tool dfbinfo is now provided by libdirectfb-udeb, later we should move it to a specific udeb, possibly with the df_input tool [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=390437 #390437]

* A cdebconf-gtk-entropy plugin is needed if we want to be able to install on encrypted volumes with the the GTK frontend too.

Per-architecture issues

We're trying to schematize bugs affecting the g-i on a per-architecture base, architectures considered here are i386, PPC, AMD64 .

i386 specific

  • For historical reasons the vga16fb module is still loaded even if a framebuffer device is already provided by vesafb: this forced us to check that DirectFB actually accesses the framebuffer device provided by vesafb , better would be not starting at all two fb devices concurrently (this also wastes little memory). [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=401685 #401685]

AMD64 specific

Proper fix requires adding that library via an udeb, or removing pthread_cancel() calls from DirectFB source code.

PPC specific

* On PPC the framebuffer device has a fixed size that cannot be set at boot like it's done on i386, so the GTK frontend should adapt itself to go "fullscreen". [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=401693 #401693]

* DirectFB's linux_input input module seems particulary willing to crash on this architecture, and as a result we had to selectively disable such module basing on experiments on on various PPC boxes.

Read also this long [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=342053 thread] about the g-i on PPC machines.

Tester

Test date

Model

cat /proc/cpuinfo | grep board

cat /proc/fb

lspci | grep VGA

Magic directfbrc

Bad directfbrc lines

Notes

Sven Luther luther@debian.org

22/09/2006

Pegasos

CHRP Pegasos

0 ATI Radeon Yd

ATI Technologies Inc RV280 [Radeon 9200 SE] (rev 01)

no-hardware

Sven Luther luther@debian.org

22/09/2006

Pegasos

CHRP Pegasos

0 ATI Radeon NH

ATI Technologies Inc Radeon R350 [Radeon 9800 Pro]

no-hardware disable-module=radeon

alessandro de zorzi lota@klez.it

27/09/2006

ibook G4

?PowerBook6,7 MacRISC3 Power Macintosh

0 ATI Radeon NV

ATI Technologies Inc M11 NV [FireGL Mobility T2e] (rev 80)

no-hardware

disable-module=linux_input

Jack Malmostoso jackmalmostoso@freesurf.ch

26/09/2006

ibook G4

?PowerBook6,3 MacRISC3 Power Macintosh

0 ATI Radeon /c

ATI Technologies INC M9+ 5C63 [Radeon Mobility 9200 (AGP)] (rev 01)

no-hardware disable-module=linux_input

Elena Grassi grassi.e@gmail.com

05/10/2006

ibook G3

?PowerBook4,3 MacRISC2 MacRISC Power Macintosh

0 ATI Radeon LW

ATI Technologies Inc Radeon Mobility M7 LW [Radeon Mobility 7500]

no-hardware disable-module=linux_input

It needs video=radeonfb to be specified at boot time, otherwise even the textual console is corrupted

Luca Suriano luka@linux.it

01/10/2006

ibook G3

?PowerBook4,3 MacRISC2 MacRISC Power Macintosh

0 ATI Radeon LW

ATI Technologies Inc Radeon Mobility M7 LW [Radeon Mobility 7500]

no-hardware disable-module=linux_input

Marco Bertorello marco@nosgoth.homelinux.org

26/09/2006

ibook G4

?PowerBook 6,5 ?MacRisc3 Power Macintosh

0 ATI Radeon \c

ATI TECHNOLOGIES Inc. M9+ 5C63 [Radeon Mobility 9200 (AGP)] (rev 01)

no-hardware disable-module=linux_input

Attilio Fiandrotti

26/09/2006

ibook G4

?PowerBook6,7 MacRISC3 Power Macintosh

0 ATI Radeon NV

ATI Technologies Inc M11 NV [FireGL Mobility T2e] (rev 80)

no-hardware

disable-module=linux_input

Rick Thomas rbthomas55@pobox.com

26/09/2006

PowerMac G4

?PowerMac3,5 MacRISC2 MacRISC Power Macintosh

0 ATI Radeon QW

ATI Technologies Inc Radeon RV200 QW [Radeon 7500]

Unsupported video format dtected (!) DirectFB/FBDev: Current mode's pixelformat: rgba 8/0, 8/0, 8/0, 0/0 (8bit)

Rick Thomas rbthomas55@pobox.com

01/10/2006

PowerMac G3 [Blue&White]

?PowerMac1,1

0 OFfb ATY,XCLAIM

ATI Technologies Inc 3D Rage Pro (rev 5c)

no way to make this work

requires boot arg "video=ofonly" to even do text mode install. Details in Bug#342053

Rick Thomas rbthomas55@pobox.com

01/10/2006

PowerMac G4 [?QuickSilver]

?PowerMac3,4

0 ATY Rage128

ATI Technologies Inc Rage 128 PF/PRO AGP 4x TMDS

independant of enable or disable "linux_input"

Mirko Grava mgrava@rhx.it

27/09/2006

ibook G4

?PowerBook5,4 MacRISC3 Power Macintosh

0 ATI Radeon NP

VGA compatible controller: ATI Technologies Inc RV350 [Mobility Radeon 9600 M10] (prog-if 00 [VGA])

no-hardware

Eddy Petrișor eddy.petrisor@gmail.com

27/09/2006

powerbook G4

?PowerBook5,2 MacRISC3 Power Macintosh

0 ATI Radeon NP

ATI Technologies Inc RV350 [Mobility Radeon 9600 M10]

no-hardware disable-module=linux_input

disable-module=radeon has no ill effects

Attilio Fiandrotti

29/09/2006

MiniMAC

?PowerBook10,2 MacRISC3 Power Macintosh

ATI Radeon Yb

ATI Technologies Inc RV280 [Raeon9200] (rev 01)

no-hardware disable-module=linux_input

Keymap was messed up in DFB after language change

Sven Luther luther@debian.org

30/09/2006

XServe G5

?RackMac3,1 MacRISC4 Power Macintosh

0 ATI Radeon QY

ATI Technologies Inc Radeon RV100 QY [Radeon 7000/VE]

disable-module=radeon disable-module=linux_input

Jan Schukat shookie@email.de

02/10/2006

PowerMac G5

?PowerMac7,3 MacRISC4 Power Macintosh

0 NV32

nVidia Corporation NV34 [?GeForce FX 5200 Ultra] (rev a1)

no way to make this work

Diego Biurrn diego AT biurrun DOT de

05/10/2006

?PowerBook G4

?PowerBook6,8 MacRISC3 Power Macintosh

0 NV32

nVidia Corporation NV34M [?GeForce FX Go5200] (rev a1)

[http://lists.debian.org/debian-powerpc/2006/10/msg00066.html no way to make this work]

Holger Levsen debian@layer-acht.org

03/11/2006

imac g3 233mhz

motherboard : iMac MacRISC Power Macintosh

0 ATY Mach64

00:12.0 VGA compatible controller: ATI Technologies Inc 3D Rage IIC 215IIC [Mach64 GT IIC] (rev 3a)

just works fine :-)

Ideas for future development

Porting

The graphical version of the installer is currently available for Intel x86, AMD64 and PowerPC. The PowerPC port [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=341597 needs work] to get different types of system correctly supported. Other architectures the graphical installer could be ported to include Sparc, Alpha and HPPA.

Fonts

For font information, please see ["DebianInstaller/GUIFonts"]

Usability

  • Interface should adapt itself so that text blocks are never too wide for optimal readability.
  • Improve accessability (for the visually handicapped). One way to do this is to [http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=339735 support different themes] (color schemes). (Note: should not lead to a major increase in initrd size.)

  • Alternative (graphical) ways to start a shell or browse log files.
  • Interface design.
  • Create custom plug-ins for cdebconf to improve handling of some dialogs and make optimal use of possibilities offered by the graphical environment.

Interface design

Most of the work on the graphical side has been done by Eduardo Silva. He has set up a [http://www.geocities.com/jobezone/d-i_gtk.html webpage] with images he designed and some comments about possible future changes.