i think i have found a problem with PG 3.1.. when i open up the control panel for the ati radeon drivers and click the "options" panel, i get a bsod.. i do not get the bsod if i disable pg's "protection".. this did not occur with the previous releases of PG 3.x.. it is not a major issue for me since i hardly ever use the "options" panel in the ati radeon control panel..
i think you have to give rundll32 access to global hooks or drivers... this has been covered before try to do a search for ATI ....
siliconman, i have a problem with updating the radeon drivers.. win xp sp2 includes radeon drivers, and when i uninstall the radeon drivers so that i can then install the new ones, windows automatically reinstalls the radeon drivers that are built into win xp sp2 when i reboot the computer.. how do you work around this problem? imo, microsoft really screwed up by including the radeon drivers in win xp sp2..
Here's how I replace the driver. 1. Disable ProcessGuard protection (uncheck "Protection enabled"). 2. Using msconfig, turn off all startup programs and reboot with all programs turned off. This is all programs listed in the STARTUP tab. 3. Using Add/Remove Programs in the Control Panel, remove the ATI Control Panel. 3a. Using Add/Remove Programs in the Control Panel, remove the ATI Drivers. If you have multiple display adapters installed, a dialog box will appear instructing you to choose which ATI Display driver to uninstall. Click the Select All button. 4. Reboot the computer in SAFE MODE. 5. When SAFE MODE is entered, the New Hardware wizard will pop up to install the ATI drivers. Cancel it immediately. You may get the Wizard 2-3 times. Cancel it immediately each time. 6. Using Driver Cleaner V3.3, clean the cabs under the Tools menu followed by cleaning ATI drivers. 7. Reboot the computer. The New Hardware wizard will again pop up. Cancel it immediately. It may pop up more than once. Cancel it each time. 8. Run the new ATI setup that you downloaded from the ATI site to install the new drivers, control panel, etc. 9. Reboot the computer with the new drivers active. 10. Using msconfig, turn on all Startup programs and reboot. 11. Re-enable ProcessGuard protection. I have ATI2evxx.exe from the System32 folder in Protection of PG with Install Hooks allowed. I have ATI2sgag.exe from the System32 folder in Protection of PG with no additional allows. Here's how ATI says you can uninstall/install it too. They do not use the Driver Cleaner program under SAFE MODE. The Dell folks on the Dell forum have always recommended using Driver Cleaner. http://www2.ati.com/drivers/Catalyst_412_release_notes.html#170177
i managed to install the new ati radeon drivers (using "driver cleaner"), and i reinstalled PG 3.1, but the problem still persist where when i click on the options tab of the ati radeon control panel, i get a bsod.. the only thing that is different on my computer besides PG 3.1 is that i am now using NOD32 instead of etrust ez av.. i still get the same bsod's with nod32's amon, imon, and dmon disabled.. the error code is (two different descriptions for the same error): 1. BCCode : 1000008e BCP1 : C0000005 BCP2 : 804DA1F4 BCP3 : B2A36B94 BCP4 : 00000000 OSVer : 5_1_2600 SP : 2_0 Product : 768_1 2. Error code 1000008e, parameter1 c0000005, parameter2 804da1f4, parameter3 b2a36b94, parameter4 00000000 thanks for trying to help
Sounds like "Jason time". Just curious- Are you running dual monitors? Does the BSOD occur if PG is disabled? Does the BSOD occur if you turn off PG GUI but leave Protection Enabled?
You need to delete the ATI folder in C:\ Disconnect from the internet, (disconnect the cable or disable the modem) delete the control panel first (dont restart) delete the driver using the add/remove in the control panel. After deleting the drivers when prompted to, restart the computer. -wayne
does the BSOD still occur if PG is first put into Learning Mode? Please test it that way and see what, if anything, gets added to your protection list after clicking the Options tab.
when pg3 is in learning mode, i still get the same bsod when i click the options tab in the radeon control panel.. i do not get the bsod if pg's protection is disabled..
Just to let you know this is not an isolated incident, I am experiencing the exact same issue with PG and ATI’s latest drivers. Disabling PG is the only way I’ve found to stop the BSOD when I hit the options tab. XP pro SP 2 Zone Alarm 5.5 Prevx pro NOD32 2.13 TDS3 Process Guard 3.050
It certainly seems this is an ATI driver issue where they don't handle something failing correctly in their code. Since it bluescreens immediately it is too quick for ProcessGuard to catch it in learning mode and hence the next time you do it, the same thing occurs. You guys might want to email ATI about this issue.
It was/is as far as I am aware. You can fix it I am told if you allow RUNDLL32.exe to install drivers?
Just to let you know (sound familiar?) everytime I click on the options tab of the ATI Control Panel WITH PG ON OR IN LEARNING MODE I get a BSOD too!! With PG disabled this does NOT OCCUR. Only with PG enabled or in learning mode does this occur. It's manifestly obvious that this BSOD is only occuring when PG is active in some way so what's with blaming ATI drivers? If I never heard of PG and never ever installed it I would never ever get a BSOD when I clicked on the ATI Control Panel so where's the connection that it's ATI's fault? I have had an ATI card for a year and never once got this BSOD until the day I installed PG but yeah sure it's ATI's fault. Let's blame it all on ATI. PG could never ever have a bug or fault in it - that's unthinkable! Whatever anyone says PG causes a BSOD when the ATI Control Panel Options menu is clicked on - period. Disable, uninstall, destroy PG and you'll never se another 'ATI' BSOD'. I have had this BSOD occur with every set of ATI drivers WHENEVER I USE PG so let's contact ATI and tell them that all their updates are faulty - I'm sure they'll like that!! Dave
i found that if i check the box in pg's "protection" to allow "rundll32.exe" to "access physical memory", then i don't get the same bsod.. i did not get the same bsod's with previous releases of pg3.x..
Well Dave, I will try and simplify the issue so you can understand it :- Without ProcessGuard installed the ATI driver tries to get access to physical memory and succeeds, it doesn't check that it did get the access and simply does what it wants, no blue screens, no issues. With ProcessGuard installed the ATI driver tries to get access to physical memory and fails (unless rundll32.dll has been given access), since the ATI driver either assumes it got this access or does not know how to handle failure to get this access it ends up causing a blue screen. So who do you blame, the program which blocks the access in a safe way, or the program which cannot handle (when it can if programmed properly) access being blocked? If it was a bug in ProcessGuard causing the bluescreen I would say yes it is a PG issue, we will deal with it. In this case it is a bug in the ATI driver because they fail to handle an error which Windows sends them after failure to get the physical memory access. In other words there is nothing in PG to fix in regards to this blue screen so it is out of my hands. This bluescreen would occur without ProcessGuard if the ATI driver could not for some reason gain physical memory access, it just so happens that this event is rare unless you have something like ProcessGuard on your system.
That's funny I didn't get it this time but I checked and that dll is already ticked the way you said but I don't remember ticking it. Anyway if it happens again I'll report it. Dave
i still think something is screwy.. with the previous versions of pg, in learning mode, everything went along smoothly, but with build 3.1, even in learning mode, it still will block things, even though it will also make the allowance for the blocked process to run the next time.. like it will say "boclean was blocked from creating a global hook", but in pg's "protection", boclean will have a check to allow the global hook the next time.. in previous releases of pg3, i don't remember seeing any pg-alerts about processes being blocked while pg was in learning mode.. i think i can see jason's point though.. other programs don't crash when they are blocked from accessing memory..
It's no big problem Jason because how many times do I need to access the options menu - usually only once when I install new drivers but I was confused as to why this only happened when I had PG installed. I know PG is only doing it's job but if the Control Panel can't access it's own menu then it would have to give some sort of error wouldn't it and because this situation is probably not foreseen (acess denied) then a script was not written or programmed to deal with the situation so Windows had a fit and BSOD'D I suppose. But if allowing access to the dll fixes the situation then how does a newbie find this out? Because of the BSOD he/she wouldn't see the PG icon turned red but a blue screen instead and wouldn't be able to track down how to resolve the issue. Does PG have enough time before the BSOD to log the blocking of the dll? Because if it's in the log after reboot then the dll can be seen to have been denied access and the situation corrected but if PG cannot log the event then unless someone finds out by accident or on this forum they might think that PG is the problem. Regards Dave
Depends on a case by case basis whether ProcessGuard has enough time. I would think in most cases not because it usually goes something like this :- 1) GetPhysicalMemoryAccess 2) AccessPhysicalMemory So even though ProcessGuard blocks #1 since #2 is directly following it (in real time this would be a few microseconds if you are lucky) there isn't enough time for PG to do anything at all. If it was like this :- 1) GetPhysicalMemoryAccess 2) Do some various things here which take some time (milliseconds) 3) AccessPhysicalMemory Then it would probably be enough time in some instances, depending on how long #2 takes. Ideally the ATI driver would do something like this :- 1) GetPhysicalMemoryAccess 2) See if we really got the access, if we didn't then abort, else go to #3 3) AccessPhysicalMemory But it doesn't, which leads to the problem.
I suppose then when something like this happens it's best to contact DCS to try and help track down the culprit? Dave
