> The compiler team has now confirmed this is a bug in the inlined > versions of the mem*() functions used with -xbuiltin and is working > on a fix. Once this is in a patch, we can ask the Nevada CBE team > to include it in the standard Nevada CBE - but I don't know how long > either stage of that will take. > > They've also provided a workaround flag that disables individual > inlined functions (-Wc,-Qinline-memcpy=0) - is this causing serious > problems that we need to apply the workaround until the CBE is patched?
Inlining these functions has the negative side effect that you can't interpose on these function calls any more. And the highly optimized platform specific implementations from libc_psr.so.1 are bypassed... So, does avoiding the function call and inlining these functions have a significant performance advantage? > (When I originally did the work, I remember breaking Ada programs as > being the biggest source of pain from using the application registers, > but don't know how bad the effects are on qemu, or what else may be > breaking from it.) qemu segfaults with the nevada X11 libs. Workaround would be to disable qemu's global register usage, but that has a significant performance impact on qemu's emulation speed. This message posted from opensolaris.org
