Bug 6001 - cannot enable fastwrite for via kt600/Radeon 9200
Summary: cannot enable fastwrite for via kt600/Radeon 9200
Status: RESOLVED WONTFIX
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/other (show other bugs)
Version: XOrg git
Hardware: x86 (IA32) Linux (All)
: high normal
Assignee: Default DRI bug account
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2006-02-22 23:22 UTC by Alex
Modified: 2007-11-09 07:26 UTC (History)
3 users (show)

See Also:
i915 platform:
i915 features:


Attachments

Description Alex 2006-02-22 23:22:52 UTC
I cannot enable fastwrite on via kt600 motheboard for Radeon 9200 card. Windows 
driver and fglrx do this without any problems and system is stable. Problem 
with radeon module: xorg/system freeze if this option is enabled.
Comment 1 Alex 2006-02-22 23:26:07 UTC
xorg-7.0, kernel 2.6.15. But with xorg-6.8.2 and kernel 2.6.14 was the same.
Comment 2 Benjamin Herrenschmidt 2006-02-23 20:57:37 UTC
Are you absolutely sure it works with fglrx and/or windows ? have you double
checked the actual config space settings to verify that FW is indeed enabled
with these ? There are all sort of strange issues with FW and it tends to be
blacklisted at the driver level in lots of situations....

It could also be possible that fglrx and/or windows tweaks some settings
somewhere related to the AGP interface that makes FW reliable while it's not by
default (clock skew, additional wait states, that sort of thing...). It might be
useful to compare the values of some of the AGP & AGP PLL related registers...
Comment 3 Alex 2006-02-24 05:22:03 UTC
Yes, It is enabled. I'm absolutely sure. 
 
I have utility for windows (it is like lspci): pci32. Below is pci32's output: 
 
 Bus 1 (AGP), Device Number 0, Device Function 0 
 Vendor 1002h ATI Technologies Inc 
 Device 5961h Radeon 9200 Series (RV280) 
 Command 0007h (I/O Access, Memory Access, BusMaster) 
 Status 02B0h (Has Capabilities List, Supports 66MHz, Supports Back-To-Back 
Trans., Medium Timing) 
 Revision 01h, Header Type 80h, Bus Latency Timer FFh 
 Minimum Bus Grant 08h, Maximum Bus Latency 00h 
 Self test 00h (Self test not supported) 
 Cache line size 32 Bytes (8 DWords) 
 PCI Class Display, type VGA 
 Subsystem ID 40181458h GIGABYTE Radeon 9200 
 Subsystem Vendor 1458h Giga-Byte Technology 
 Address 0 is a Memory Address (anywhere in 0-4Gb, Prefetchable) : D0000000h 
 Address 1 is an I/O Port : 0000D000h 
 Address 2 is a Memory Address (anywhere in 0-4Gb) : E8020000h 
 System IRQ 16, INT# A 
 Expansion ROM of 128Kb decoded by this card (Currently disabled) 
 New Capabilities List Present: 
   AGP Capability, Version 3.0 (AGP 8x and 4x, core register support) 
     AGP Speed(s) Supported : 4x 8x 
     FW Transfers Supported : Yes 
     >4Gb Address Space Supported : No 
     Sideband Addressing Supported : Yes 
     AGP v3.0 Operation Mode Available : Yes 
     Isosynchronous Transactions Supported : No 
     Maximum Command Queue Length : 256 bytes 
     AGP Speed Selected : 8x 
     FW Transfers Enabled : Yes 
     >4Gb Address Space Enabled : No 
     AGP Enabled : Yes 
     Sideband Addressing Enabled : Yes 
     AGP v3.0 Operation Mode : Enabled 
     Current Command Queue Length : 32 bytes 
   Power Management Capability, Version 1.1 
     Supports low power State D1 
     Supports low power State D2 
     Does not support PME# signalling 
     Current Power State : D0 (Device operational, no power saving) 
Comment 4 Luc Verhaegen 2007-11-09 07:26:59 UTC
AGPFastWrite should not enabled unless you want to go look for trouble anyway.


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.