On Tue, Aug 7, 2012 at 10:22 AM, Douglas Gregor <[email protected]> wrote: > > On Aug 6, 2012, at 11:31 PM, Chandler Carruth <[email protected]> wrote: > > On Mon, Aug 6, 2012 at 11:16 PM, Douglas Gregor <[email protected]> wrote: >> >> >> On Aug 6, 2012, at 1:11 PM, Sebastian Pop <[email protected]> wrote: >> >> > Ping^4. >> > >> > Hi, I'm waiting for a month now for an OK to commit. >> > It seems that really nobody except me cares for this change. >> > Does it really matter to somebody if I'm pushing this change out to svn? >> > >> > How should I proceed in timeout cases like this? >> >> Patch looks good, please go ahead and commit. If this happens again, >> please CC me. If I can't review the patch, I'll find someone who can. Sorry >> about that! > > > It's also largely my fault. I've been neglecting CMake patches. =/ Sebastian > pinged me on IRC and I still didn't get to it. > > FWIW, the first patch makes perfect sense.
Thanks, committed as r161504. > > The second patch I'm not as clear on... I feel like either: > > 1) We shouldn't build (or test) c-index-test and friends if they necessitate > a static library and that isn't supported in the CMake configuration, or > > 2) We should link c-index-test and friends dynamically when explicitly > requested to do so. > > > I think I prefer #2. What do you think? > > > I'd prefer #2. > I'm still struggling with this one: if I'm removing -llibclang for shared builds, then I still get a lot of undefined symbols. Any hints on which .so files should be added to the link command to replace the static libclang.a? Thanks, Sebastian -- Qualcomm Innovation Center, Inc is a member of Code Aurora Forum _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
