Hi Sarah - Just an additional nit To Jan's feedback:
1855 - there is a space that was added that doesn't appear necessary. The only other thing I was wondering....if debugging is turned on, then at line 855, it looks like the function will return and not complete the rest of the code path. Is that correct? thx, ginnie On 01/14/10 18:16, Sarah Jelinek wrote: > Hi Jan and Ginnie, > > Could you do a code review for: > > 6795293 libdiskmgt coredump on x86 with devices under mpxio > control/SUN-Universal Xport inbound mangement > > Webrev is here: > > http://cr.opensolaris.org/~sjelinek/6795293/ > > ****Building and Testing done: > -Both sparc and x86 built clean > -Tested on both x86 and sparc: > x86 testing reproduced bug and also shows bug fix > sparc testing for regression testing. Bug does not present > itself on sparc. > > -Reproduced bug and tested fix on oaf627.ireland. This is an x86 > box > with Qlogic FC controllers. > -Tested both mpxio and non-mpxio > -ran ::findleaks. No leaks detected. > > Test output is located here: > > /net/irperf.ireland/export/work/bfu/sjelinek > > The files in this directory are specifically: > ***Bad run(bug is present): > -core.test_td.2664.1263495933.gz-core showing original bug reproduced > -libdiskmgt_bad.txt > -test_td_bad.txt > > ***Good run(bug fixed): > -libdiskmgt_good.txt > -test_td_good.txt > -format.txt - shows output of format to show match with > test_td_good.txt data. > > ***Good run, mpxio disabled(bug fixed): > -libdiskmgt_nonmpxio.txt > -test_td_nonmpxio.txt > -format_nonmpxio.txt - shows output of format to show match with > -test_td_nonmpxio.txt data. > > > thanks, > sarah >
