search for: pr17982

I request that we hold off until the final review is done. -eric On Tue, Dec 3, 2013 at 12:06 PM, Manman Ren <manman.ren at gmail.com> wrote: > > This set of patches add support for debug info metadata version and debug > info metadata will be dropped if a loaded bit code has no version or an > outdated version. > > 195494: Debug Info: add a constant for debug info

These don’t apply cleanly to the 3.4 branch. Could you add them in yourself? -bw On Dec 5, 2013, at 2:08 PM, Eric Christopher <echristo at gmail.com> wrote: > Final review is done. Feel free to merge :) > > Thanks for the wait. > > -eric > > On Tue, Dec 3, 2013 at 1:27 PM, Eric Christopher <echristo at gmail.com> wrote: >> I request that we hold off

This set of patches add support for debug info metadata version and debug info metadata will be dropped if a loaded bit code has no version or an outdated version. 195494: Debug Info: add a constant for debug info version number. 195495: Debug Info: add a "Debug Info Version" module flag to output the current debug info version number. 195504: Debug Info: update testing cases to specify

Final review is done. Feel free to merge :) Thanks for the wait. -eric On Tue, Dec 3, 2013 at 1:27 PM, Eric Christopher <echristo at gmail.com> wrote: > I request that we hold off until the final review is done. > > -eric > > On Tue, Dec 3, 2013 at 12:06 PM, Manman Ren <manman.ren at gmail.com> wrote: >> >> This set of patches add support for debug info

I just filed this as pr17982, but I'd like to call attention to it, since it seems like something we should try to fix in the 3.4 release. Here's the description from the PR: Sometime over the last year, and I'm fairly certain it has been since we released 3.3, the debug info metadata format has changed in incomp...

...tead of just reopening it. > > I'm not really sure the bug is needed, but if you want to open a bug > then that's fine. I think there are some holes in what's going to > specified where, but that can be discussed when someone starts working > on it. OK. I have reopened pr17982 to cover this issue. I don't intend to specify much in bugzilla, but we should have something to track the overall problem. > >> >> 2) Who is going to do this work? > > No idea. > > -eric

I don't have a strong opinion on how we fix it, but I think it's really important to do something and get it done for the 3.4 release. Here's what I would do: - implement Quentin's previous proposal for back-end diagnostics (and add a diagnostic handler to Apple's linker), so that we can warn about out-of-date debug info - add a version number to the debug info metadata -

...tead of just reopening it. > > I'm not really sure the bug is needed, but if you want to open a bug > then that's fine. I think there are some holes in what's going to > specified where, but that can be discussed when someone starts working > on it. OK. I have reopened pr17982 to cover this issue. I don't intend to specify much in bugzilla, but we should have something to track the overall problem. > >> >> 2) Who is going to do this work? > > No idea. > > -eric ------------------------------ Message: 5 Date: Tue, 19 Nov 2013 21:44...