Bug#404148: i'm not convinced release notes are enough
Steve Langasek wrote:
>> In all doing respect, I think that it's a much greater risk to not use
>> iommu=soft per default than doing so. Even if we imagine that there
>> would by systems that don't work with the sw-iommu.... it's likely that
>> they simply break (at boot time). And then the affected user at least
>> knows that something is happening to him, while with no iommu=soft he
>> would probably never realize that he has problems.
>>
> That doesn't address how to set iommu=soft as a default, though. The only
> practical way that I see to accomplish this is in the kernel package itself,
> and there was doubt that there would be an opportunity to update the kernel
> again before release.
>
One possibility would be too add it per default to the grub and lilo
configs, perhaps with a pointer to this bug.
This way every users could simply decide wheter to remove this or not.
>> So such a patch would have to whitelist systems that are known to work,
>> instead of blacklist the others.
>>
> But AIUI the problem has so far only been reported on systems using an
> nvidia chipset, right?
Yes (and perhaps no).
First of all there were some people who had and Opteron and an Nvidia
chipset and that much main memory etc etc. (I mean a system
configuration where "we" would have supposed that they would suffer from
the error).... but they at least claimed to not suffer from it.
Perhaps some BIOSes might somewho workaround (not solve) it.
Another issue is the following: Just today someone added to the
bugzilla.kernel.org bug that he probably has the same error but his
system doesn't have an Nvidia chipset (see
http://bugzilla.kernel.org/show_bug.cgi?id=7768).
Anyway I don't believe that this is actually the same error.
> I'm not going to hold up as release-critical a
> bugfix for other systems where the problem hasn't been reported yet. If
> more information becomes available showing that the bug exists on non-nvidia
> systems, we should of course revisit it at that point.
>
see above
> In the meantime, I don't see any reason why we shouldn't patch the kernel to
> disable hw iommu on nvidia systems only. I believe the attached patch
> should do this. Are you in a position to confirm that this does disable hw
> iommu for you?
>
Unfortunately I'm not currently able to validate it, but I will forward
it to the thread at lkml and to the bug at kernel.org.
begin:vcard
fn:Mitterer, Christoph Anton
n:Mitterer;Christoph Anton
email;internet:calestyo@scientia.net
x-mozilla-html:TRUE
version:2.1
end:vcard
Reply to: