Re: Public repro case! Re: [PATCH/RFC] Allow writing loose objects that are corrupted in a pack file

From: Boyd Stephen Smith Jr. <bss@iguanasuicide.net>
Date: Jan 8, 2009, 02:52 UTC
Message-ID: <200901072053.00492.bss@iguanasuicide.net>
In-Reply-To: <885649360901071821t2ea481b5k83ab800f6aeb897@mail.gmail.com>

On Wednesday 2009 January 07 20:21:18 James Pickens wrote:

Show 8 quoted lines

>On Wed, Jan 7, 2009, Linus Torvalds <torvalds@linux-foundation.org> wrote:
>> Can you cnfirm that your "reproducible" case starts working with that
>> addition to your ~/.gitconfig? If so, the solution is pretty simple: we
>> should just lower the default pack windowsize.
>
>Umm... isn't that more of a workaround than a solution?  I.e., if you lower
>the default pack windowsize, couldn't the corruption still happen under the
>right conditions?

IMHO: I agree, somewhat. I'm fine with a "die()" message when there's not enough memory, but either corruption or just a spurious, but scary "<SHA> is corrupt" messages should be fixed.

-- 
Boyd Stephen Smith Jr.                     ,= ,-_-. =. 
bss@iguanasuicide.net                     ((_/)o o(\_))
ICQ: 514984 YM/AIM: DaTwinkDaddy           `-'(. .)`-' 
http://iguanasuicide.net/                      \_/