Alexander R. Eremin wrote:
> On Tue, 2009-12-08 at 14:11 +0100, Jan Damborsky wrote:
>> Alexander Eremin wrote:
>>> Done
>>>
>> Hi Alex,
>>
>> I built all images (x86&Sparc AI, x86 LiveCD) based on build 129
>> with DC containing the fix and all builds finished successfully.
>>
>> I checked log files and AI boot archives were created with
>> nbpi 16384, LiveCD boot archives had nbpi calculated as 8192
>> which are expected results.
>>
>> All installations using built images succeeded.
>>
>> For reference, I compared free space in boot archives without
>> and with the fix (size of boot archive itself didn't change):
>>
>> boot archive        gained free space (MiB)
>> -------------------------------------------------
>> x86 LiveCD 32bit    7
>> x86 LiveCD 64bit    8
>> x86 AI 32bit        6
>> x86 AI 64bit        7
>> Sparc AI            9
>>
>> Those results seem sane to me. Could you please check and
>> confirm if the numbers look fine ?
> I have for Sparc AI and x86 LiveCD same numbers

ok. That is good :-)

>> Also, could you please update bug report with suggested fix ?
> Done

Thank you for updating the bug with pointer to webrev and test
results. This is fine, however I had something slightly different
in mind.
I think it would be useful to capture the changes from high level
point of view, e.g. how particular pieces of DC were modified -
e.g.

* new nbpi tag introduced in DC manifest
* how algorithm for automatic nbpi calculation works and where it was 
implemented
* how DC deals with situation when nbpi is specified in AI manifest
* ...

That if somebody takes a look at bug report, it will be clear what
new functionality was added and how existing behavior of DC was modified
by the bug fix.

>> Thank you very much,
>> Jan
> Thanks Jan, perhaps we can now think about non-compression tests..

I agree. We could take a look how things look now.

Thank you,
Jan


Reply via email to