Jan Damborsky wrote:
> Dave,
> 
> 
> Dave Miner wrote:
>> jan damborsky wrote:
>>> Hi Dave,
>>>
>>> based on the yesterday's discussion during Caiman meeting,
>>> I am assuming that following bug is to be qualified as
>>> 2008.11 stopper. This fix addresses scenario Rafal
>>> reported in bug 4872.
>>>
>>> Could you please let me know, if my understanding is correct ?
>>>
>> Yes, I've approved it.
> 
> Thank you
> 
>>> If this is the case, could I please ask two people to review
>>> those changes ?
>>>
>>> 4980 Installer should convert legacy Solaris partition (0x82) to 
>>> Solaris2 one (0xbf)
>>> http://defect.opensolaris.org/bz/show_bug.cgi?id=4980
>>>
>>> The webrev is available at:
>>> http://cr.opensolaris.org/~dambi/bug-4980
>>>
>> 131: s/SUNW/SUN/
> 
> Done.
> 
>>
>> 136: s/one// 
> 
> Done.
> 
>>
>> 153: it's not at all obvious to me why "0" is the right value to use 
>> here for the handle; perhaps a comment, at least. 
> 
> 'handle' parameter is currently unused
> in this function - I have put appropriate
> comment there explaining this.
> 
>>   I also would have expected to check the return value on this call, 
>> not just whether committed_disk_target == NULL; that's not explicitly 
>> guaranteed to be its state in error cases, though I agree that it is 
>> in the current implementation.
> 
> This is a good suggestion - I have modified
> the code, so that it checks for return value.
> Since om_set_disk_partition_info() returns
> OM_FAILURE in all cases when committed_disk_target
> is set to NULL, 'committed_disk_target == NULL'
> check became redundant and was removed.
> 
> Updated webrev with all changes has been published
> at the same location.
> 

Thanks, looks good.

Dave


Reply via email to