[petsc-dev] petsc-dev: ASM with sbaij matrix and block size=1 broken?

[Brad Aagaard]

Barry,

Sorry for the hg goof. Here is the changeset that I think is related to 
the problem.

http://petsc.cs.iit.edu/petsc/petsc-dev/rev/7782710e3f96

Brad


On 11/4/11 8:15 PM, Barry Smith wrote:
>
> On Nov 4, 2011, at 10:03 PM, Matthew Knepley wrote:
>
>> On Sat, Nov 5, 2011 at 3:01 AM, Barry Smith<bsmith at mcs.anl.gov>  wrote:
>>
>> On Nov 4, 2011, at 8:43 PM, Jed Brown wrote:
>>
>>> On Fri, Nov 4, 2011 at 18:27, Matthew Knepley<knepley at gmail.com>  wrote:
>>> I am trying to understand, but having problems.
>>>
>>> 1) 20985 does not change anything here. Do you mean 20989?
>>>
>>> Please, please, please use the SHA1. The number is meaningless since it's different for everyone. I cannot forgive Mercurial for printing these numbers, they cause nothing but trouble.
>>
>>    Bug report to Mercurial folks?
>>
>> They also show the SHA1 at all times.
>
>     Yes, but as Jed observes most users don't understand that the 20989 number is meaningless while the SHA1 number is useful hence lots of time is wasting when people send the wrong number.  This has already happened numerous times on petsc-maint and petsc-dev (I am a guilty party).
>
>      If the door is a push door then why the hell would you put a pull handle on it? So why list a useless number right before the useful number (that is a sign of bad design) just skip the useless number.
>
>       Bad designs deserve bug reports just like errors deserve bug reports.
>
>     Barry
>
>>
>>     Matt
>>
>>    Barry
>>
>>
>>>
>>>
>>
>>
>>
>>
>> --
>> What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.
>> -- Norbert Wiener
>
>