On Wed, May 20, 2015 at 02:22:17PM -0700, Linus Torvalds wrote: > On Wed, May 20, 2015 at 11:32 AM, Linus Torvalds > <[email protected]> wrote: > > > > I'll try to figure out *why* those three dives fail to download. I > > suspect it's some issue where the length of the dive is just right to > > cause a packetization case that I don't handle right. I'll have to > > analyze those three cases more closely. > > Yes. > > The "receive_data()" function continues to try to read more data if > the previous packet was full - but in some cases the full packet > contained the final data, and receive_data() is just done and > shouldn't try to read any more. > > The attached patch should fix it. Knock wood. > > Dirk - can you make a nightly windows build with this, so that it can > get tested? The situation doesn't trigger with my dives, and while I > think I got it right...
Done Your patch has been pushed to ldc[Subsurface-testing] (which has been rebased on latest libdivecomputer master, just to be annoying) /D _______________________________________________ subsurface mailing list [email protected] http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface
