Bug 137343
| Summary: | RH40-beta1, embedded IDE/PCI drivers not honoring Sub ID's/Class code | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | Red Hat Enterprise Linux 4 | Reporter: | Manoj Jose <manojj> | ||||||||
| Component: | kernel | Assignee: | Alan Cox <alan> | ||||||||
| Status: | CLOSED ERRATA | QA Contact: | |||||||||
| Severity: | medium | Docs Contact: | |||||||||
| Priority: | medium | ||||||||||
| Version: | 4.0 | CC: | alan, coughlan, dale.l.busacker, davej, dely.l.sy, dmair, fhirtz, jgarzik, jturner, levent.akyil, magdalena.glinkowski, mwesley, poelstra, tao, tburke, ykang | ||||||||
| Target Milestone: | --- | ||||||||||
| Target Release: | --- | ||||||||||
| Hardware: | i686 | ||||||||||
| OS: | Linux | ||||||||||
| Whiteboard: | |||||||||||
| Fixed In Version: | RHSA-2005-514 | Doc Type: | Bug Fix | ||||||||
| Doc Text: | Story Points: | --- | |||||||||
| Clone Of: | Environment: | ||||||||||
| Last Closed: | 2005-10-05 12:31:44 UTC | Type: | --- | ||||||||
| Regression: | --- | Mount Type: | --- | ||||||||
| Documentation: | --- | CRM: | |||||||||
| Verified Versions: | Category: | --- | |||||||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||||||
| Embargoed: | |||||||||||
| Bug Depends On: | |||||||||||
| Bug Blocks: | 156322 | ||||||||||
| Attachments: |
|
||||||||||
Created attachment 108266 [details]
noraid for ide, from the IT
Summary from rhkernel-list: This can already be accomplished with the built-in IDE driver using ideX=noprobe and/or hdX=noprobe. They should also check their module to ensure that it reserves its I/O ressources correctly. Finally, the proposed patch is not safe for any device that isn't storage class, but has other magic involved. Nor is it ever going to fly upstream. ideX=noprobe and hdX=noprobe will work only for IDE modules. But rememeber 2.6 kernel ship with Build in IDE drivers with kernel. So we need to have some other mechanism to make it work. Or make the IDE drivers back to modules. thanks Manoj From linux/Documentation/kernel-parameters.txt
Module parameters for modules that are built into the kernel image
are specified on the kernel command line with the module name plus
'.' plus parameter name, with '=' and value if appropriate, such as:
usbcore.blinkenlights=1
We have problems with following IDE drivers.. serverworks, siimage, cmdxxx. I did not see any module parameters supported by these drivers. r u suggesting to have the driver changed to have the module parameters or module parameters already present? -manoj Any updates on this issue? This is currently affecting our Blade Product (McCarran). The RHEL4 installation should correctly detect and allow md to be set up for raid 0 or raid 1 on these controllers without third party modules. No performance loss is involved as these are software raid devices. That's correct. However, the difference with RH 4.0 now is that it won't allow 3rd party RAID vendor drivers to be used. We are locked out from using those drivers. We'd like to be able to use 3rd party drivers for RAID setups. Well assuming they work (and that I guess would be an unsupported configuration) Boot with ide0=noprobe ide1=noprobe and the Linux IDE layer will ignore the first two IDE controllers. At that point you can load other drivers. Hi Here is our input regarding this. Our systems uses first two ide controlers for onboard IDE devices and CDROM devices. So we had to dissable the IDE RAID controller alone. We gave boot option ide2=noprobe ide3=noprobe ide4=noprobe ide5=noprobe (rh40-rc1) and it doesnt seem to stop the IDE driver from loading. But if we give ide0=noprobe ide1=noprobe we cannot even use CD-ROM devices to install RH40 RC1. We need to dissable a particular controller's driver from loading. Remember these IDE drivers are part of 2.6 kernel. thanks Manoj Moving this off of the "U1 kernel placeholder" since it's not going to make it. Created attachment 115830 [details]
The next try of a patch
Created attachment 117194 [details]
Patch from linked IT entry
Patch looks a good compromise to me An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on the solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHSA-2005-514.html |
From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 1.1.4322) Description of problem: In rh40-beta1 embedded IDE/PCI drivers are loading with out checking sub system ID's or class code. It will lead some of the drivers to load on our Sil-680, Sil-3112 and CSB6 based RAID controllers and will not permit LSI drivers to load! How can we prevent these IDE/PCI drivers from loading?. Version-Release number of selected component (if applicable): 2.6.8-1.528.2.10smp How reproducible: Always Steps to Reproduce: 1. Just install RH40-Beta1 on SIL-3112/CSB6/SIL-680 based LSI Raid controllers. You can see the embedded kernel drivers loading. 2. 3. Additional info: