Re: [PATCH 097/102] efivars: explicitly calculate length of VariableName

From: Lingzhu Xiang
Date: Wed Apr 17 2013 - 23:32:28 EST


On 04/17/2013 09:28 PM, Luis Henriques wrote:
> On Wed, Apr 17, 2013 at 08:13:57PM +0800, Lingzhu Xiang wrote:
>> On 04/17/2013 07:56 PM, Andy Whitcroft wrote:
>>> On Wed, Apr 17, 2013 at 12:37:43PM +0800, Lingzhu Xiang wrote:
>>>> On 04/16/2013 06:33 PM, Luis Henriques wrote:
>>>>> 68d929862e29a8b52a7f2f2f86a0600423b093cd efi: be more paranoid about available space when creating variables
>>>>
>>>> This prevents a bricking issue for some Samsung devices but causes
>>>> regression on many other machines.
>>>>
>>>> https://bugzilla.kernel.org/show_bug.cgi?id=55471
>>>> https://bugzilla.redhat.com/show_bug.cgi?id=947142
>>>> http://article.gmane.org/gmane.linux.kernel.efi/1078
>>>> http://article.gmane.org/gmane.linux.kernel.stable/47150
>>>>
>>>> This patchset will fix it again:
>>>>
>>>> http://thread.gmane.org/gmane.linux.kernel.efi/1081
>>>
>>> Thanks for the pointer. I note that in that thread you yourself imply
>>> there are still issues after applying them? Was that accurate, is there
>>> yet further patches needed?
>>
>> I just find that issue. Didn't see that when writing the above.
>>
>> Earlier I was testing on a different machine and the result was
>> good. I guess the patch still needs some more testing.
>
> Thanks Lingzhu.
>
> So, I guess you would recommend me to drop the whole series until we
> have this patchset accepted, tested and back-ported for 3.5, correct?
>
> (I tried to isolate the one you pointed out, but I'm afraid there are
> too many dependencies between them to drop a single patch.)

Unfortunately yes.

The whole series are mostly solving pstore bugs and pstore trashing
firmware. If you want to prevent more bricking from happening asap,
a temporary choice is to allow disabling efi pstore altogether with these:

ed9dc8ce7a1c8115dba9483a9b51df8b63a2e0ef efivars: Allow disabling use as a pstore backend
ec0971ba5372a4dfa753f232449d23a8fd98490e efivars: Add module parameter to disable use as a pstore backend
ca0ba26fbbd2d81c43085df49ce0abfe34535a90 efivars: Fix check for CONFIG_EFI_VARS_PSTORE_DEFAULT_DISABLE
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/