Re: Cleanup patches to follow

Alex Rousskov wrote:
> On Wed, 2008-09-10 at 23:42 -0600, Alex Rousskov wrote:
>
>> I am going to post a few [MERGE] requests after this email.
>
> Done. I hope I did not forget anything. Separating changes into
> individual [MERGE] requests was rather painful -- I had to create a
> separate bzr branch for each submit (there is probably a better way to
> do that).
>
> If you want to test the submitted code, I recommend also applying Amos
> XFF patch if he has not committed it yet. Also, I have committed many
> minor fixes directly to trunk so please do not forget to 'bzr update'
> before applying these patches.

XFF committed.

>
> API documentation notes will be committed separately, once the format
> (.h or .dox) has been agreed on.
>
> Cheers,
>
> Alex.
>
>> In aggregate, they fix all critical and reported Comm and AsyncCall
>> segfault- and assertion-causing problems that I am aware of. The cleaned
>> up code works in my lab tests and on one live IRCache node. I think it
>> should be committed.
>>
>> The fixes are related and were tested together, but I want to submit
>> them for review in a few semi-independent chunks for clarity sake. The
>> fixes should be committed together but with different commit messages
>> (from email bodies).
>>
>> If the code is committed, I will go through affected bug reports (that I
>> know of) and update or close them. I will continue to monitor bug
>> reports. Other than bugs, I will work on preparing the last eCAP code
>> portion for merging.
>>
>>
>> Finally, I am aware of one unreported(?) problem related to the shutdown
>> sequence. I have added an XXX to main.cc already but will discuss that
>> problem and a fix separately, tomorrow. The bug may lead to assertions
>> at the very end of the shutdown sequence, but does not affect the
>> "runtime" code.
>>
>>
>> Thank you,
>>
>> Alex.
>>
>

Amos

-- 
Please use Squid 2.7.STABLE4 or 3.0.STABLE8