[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#554349: rred and stack overflows



Robert Lemmen <robertle@semistable.com> writes:

> On Wed, Nov 25, 2009 at 08:08:36PM +0100, David Kalnischkies wrote:
>> Really, i don't think it is a stackoverflow as it fails at a to prominent
>> cmd count... I have rewritten a few parts of rred to use long and more
>> sane return values and it patches the file now just fine. :)
>
> ok, excellent. but didn't someone say ulimit -s fixes it? 
>
>> While on it i have also integrated #463354 which will speed up the
>> patch-process a bit (but don't expect a miracle, the biggest timesucker
>> still is the decompression of the patches and applying them patch by
>> patch instead of concatenate the patches - but this has again a few
>> problems as this is far from a simple cat operation as you have to
>> recalculate line numbers)
>
> #463354 looks like it is also getting rid of the recursion, that is very 
> good!
>
>> I will tell you a secret: For most users with fast connections pdiffs
>> are (always) slower. Pdiffs fight against download size and
>> ...
>
> heh, i know! 
>
>> Michael has already a patch for apt to don't download the patch
>> if the patch is bigger than the actual file, so no need to implement
>> some logic in the generator as the file was itself correct, just to
>> big to be useful.
>
> cool, but i thought there is already logic to do that in the generator,
> plus changing the generator not to produce uselessly large files would
> also autmagically fix the whole problems for systems with an older apt
> that still contains the broken rred...
>
> cu  robert
>
> -- 
> Robert Lemmen                               http://www.semistable.com 

I would like to see a config option to limit the number of patch files
to download. As you said the loading/saving of the index files and
decompressing pdiff files uses up time. Even with a moderate download
speed anything over 4 to 8 pdiff files is slower than getting the full
file. But it depends on the download speed.

So maybe an option to limit the amount of pdiff files eigther strictly
to X pdiffs or to |pdiff size| * Y < |index size|.

MfG
        Goswin



Reply to: