On 1 Feb 2016 06:07, "Dirk Hohndel" <[email protected]> wrote: > > > > On Jan 31, 2016, at 11:05 AM, Willem Ferguson < [email protected]> wrote: > > > > On 31/01/2016 20:49, Dirk Hohndel wrote: > >> 1030cb2 is -676 > >> 464b88f is -675 > >> > >> But as Lubomir points out, between different Qt and potentially different mobile-components, there is some risk that your builds aren't consistent with mine. Especially when it comes to such odd crashes. > >> > >> I have actually taken a day off from Subsurface yesterday so I'm not 100% following what you and Willem are tracking down. I can certainly create specific builds if that would help. Which ones do we need? > >> > >> /D > >> > > Rick's rationale is to track down the specific patch that causes the crash. It is later than build 675 and earlier than 689. So I have been running various builds of Subsurface-mobile, checking which builds run as expected and which crash. > > OK, I'm building all the revisions between 675 and 758 one by one and am putting them all into daily. This will take a while but so will your testing, so I'm pretty sure I'll be faster than you :-) > > 675 - 679 are already there, a new one is added approximately every minute :-) >
If our build environments are similar enough. The offending commit will be mine, 1030cb2. Oops. That build (of mine) failed for Willem. The thing is I thought it was the same as 675, which worked for him. Can you create a test from the latest master, with that commit reverted? I think gridlayout is problematic because it isn't supposed to have a set number of columns. The items should fit in a grid, with as many as fit on each row. We might be better off with individual row items. Rick
_______________________________________________ subsurface mailing list [email protected] http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface
