After further investigation, it was determined that errno was not being set correctly due to the lack of a -mt option when building libti. I have updated the webrev, taking into account the changes made by 14734. No changes have been made to instantiate.c since the last webrev.
Updated webrev is at: http://cr.opensolaris.org/~sohn/14295c If the original reviewers (cc'd) could check the updated webrev, that would be appreciated. Thanks, Sue On 02/18/10 12:20 PM, sanjay nadkarni wrote: > > Sue and I have had an ongoing discussion about this code review. The > code addresses handling GPT labeled disks on sparc systems . > > The problem is that change as made is hiding a bizzare ioctl/errno > behavior. It is unclear if this is systemic issue for TI or ON or > specific to the text installer. A scan of ON src revealed that the cut'n > paste code has gone viral in ON and could also be potentially hiding > elsewhere too. > > I am recommending that for now, this bugfix should be tabled and that > the team should be allowed to push the changes into slim_source with > this bug open. Until the problem is completely understood and it would > okay to document the work around for sparc. > > > -Sanjay > > > > On 02/15/10 03:22 PM, Susan Sohn wrote: >> I'd like to get a review of my changes for: >> >> 14295 GPT labeled disks don't show up on SPARC with fix to bug 14144 >> http://defect.opensolaris.org/bz/show_bug.cgi?id=14295 >> >> Webrev: >> http://cr.opensolaris.org/~sohn/14295/ >> >> These changes implement the second part of the bug fix, to provide a >> warning >> popup to the user and to fix libti to recognize the GPT disk so that >> an SMI >> label can be applied. >> >> Thanks, >> Sue >> >> >> >> _______________________________________________ >> caiman-discuss mailing list >> caiman-discuss at opensolaris.org >> http://mail.opensolaris.org/mailman/listinfo/caiman-discuss >
