Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 17 Jan 2002 16:39:50 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 17 Jan 2002 16:39:40 -0500 Received: from smtp.cogeco.net ([216.221.81.25]:37595 "EHLO fep3.cogeco.net") by vger.kernel.org with ESMTP id ; Thu, 17 Jan 2002 16:39:31 -0500 Subject: Re: hangs using opengl From: "Nix N. Nix" To: vda@port.imtp.ilyichevsk.odessa.ua Cc: linux-kernel@vger.kernel.org In-Reply-To: <200201171926.g0HJQUE19410@Port.imtp.ilyichevsk.odessa.ua> In-Reply-To: <3C4712DB.6090201@kabelfoon.nl> <1011292729.12873.27.camel@tux> <1011294257.13517.1.camel@tux> <200201171926.g0HJQUE19410@Port.imtp.ilyichevsk.odessa.ua> Content-Type: text/plain Content-Transfer-Encoding: 7bit X-Mailer: Evolution/1.0.1 Date: 17 Jan 2002 16:38:32 -0500 Message-Id: <1011303569.14984.3.camel@tux> Mime-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 2002-01-17 at 18:26, Denis Vlasenko wrote: > On 17 January 2002 17:03, Nix N. Nix wrote: > > > Via has recently released a patch to Via-chipset based boards that > > > addresses issues Windows XP users have been experiencing (BSODs and > > > friends) while playing OpenGL games, especially with NVidia chips. The > > > problem and the description of the fix (as taken from the Readme.txt > > > from Via's patch) are as follow: > > > > > > So what does it do? It closes the RX55 memory register in BIOS. The RX55 > > > register's official name and function is Memory Write Queue (MWQ) timer. > > > The MWQ timer is actually a timing device included in the memory host > > > controller to prevent write data being held in the memory queue too > > > long. After the data has been in the queue too long it times out. This > > > timed out data is then given a higher write request priority. Now that > > > might sound nice a bit of extra performance BUT the procedure fails when > > > overloaded. 3D games and Win XP put too much load on the memory queuing > > > timer procedure. The nVidia new driver exaggerates the problem even more > > > as the driver enables nVidia cards to use even more memory than previous > > > driver versions. > > [snip] > > > > In light of VIAs discoveries, and the fact that the patch that they now > > > have available for Windows is not available for Linux also, I was > > > wondering if somebody on this list may be kind enough to help us with > > > what may very well be the symptoms of the same problem, but on Linux. > > > Could the code that accomplishes the above (turn off the RX55 register) > > > be made into a patch that can be applied to the kernel, thus providing > > > an equivalent patch for Linux systems ? > > Athlon bug stomper is already in mainline. If latest kernel does not work for > you, check that register (man lspci) and 0x95 also. Try to disable them (man I did lspci -vv -xxx Each entry is followed by a table of hex. Is it entries 0x55 and 0x95 in that table that you're referring to ? If so, well, they're both 0: 01:00.0 VGA compatible controller: nVidia Corporation NV11 (rev a1) (prog-if 00 [VGA]) Subsystem: Asustek Computer, Inc.: Unknown device 4015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz+ UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- SERR- setpci). Report back - we may need to further improve stomper. > -- > vda > - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/