I found this neat piece in the forums: https://forums.gentoo.org/viewtopic-t-1052716-start-0.html Of wich the webkit-gtk patch is really nice. As webkit is quite monstrous in size, that's a sweet spot to have. Could you include the patch below, (credits go to the original author) so we can have -flto switched on and webkit won't fail us upon linking? diff --git a/Source/JavaScriptCore/llint/LowLevelInterpreter.cpp b/Source/JavaScriptCore/llint/LowLevelInterpreter.cpp --- a/Source/JavaScriptCore/llint/LowLevelInterpreter.cpp 2016-01-14 21:20:42.304902905 +0100 +++ b/Source/JavaScriptCore/llint/LowLevelInterpreter.cpp 2016-01-14 21:25:52.000000000 +0100 @@ -23,6 +23,15 @@ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ +// If we are using gcc >= 5.0, make sure that this compilation unit is +// compiled with -fno-lto, because we may be using inline assembly +// included from LLIntAssembly.h +#ifdef __GNUC__ +#if __GNUC__ >= 5 +#pragma GCC optimize ("no-lto") +#endif +#endif + #include "config.h" #include "LowLevelInterpreter.h"
We won't be including it until it is at least included in webkit upstream (not necessarily the stable branch we track). This would ensure us some knowledgeable review on it has happened (for example off-hand I have no clue if it might not be more appropriate to have the pragma in the header file instead). In other words, someone needs to bring this upstream, or find it already in their git (I haven't checked - maybe their development versions already carry this). To speed this up, feel free to help with that, and point "See also" or "URL" field to the upstream bugzilla ticket if applicable. If we (webkit-gtk maintainer in gentoo) need to take care of it, then I'm not sure when we get to it eventually - might take a long while with other priorities.
(In reply to Mart Raudsepp from comment #1) > We won't be including it until it is at least included in webkit upstream > (not necessarily the stable branch we track). This would ensure us some > knowledgeable review on it has happened (for example off-hand I have no clue > if it might not be more appropriate to have the pragma in the header file > instead). > > In other words, someone needs to bring this upstream, or find it already in > their git (I haven't checked - maybe their development versions already > carry this). > To speed this up, feel free to help with that, and point "See also" or "URL" > field to the upstream bugzilla ticket if applicable. > If we (webkit-gtk maintainer in gentoo) need to take care of it, then I'm > not sure when we get to it eventually - might take a long while with other > priorities. Sorry for the late reply, it's filed on their bugtracker, but not assigned to anybody yet: https://bugs.webkit.org/show_bug.cgi?id=180625 Cheers, Walter
