I reworked the patchkit to speed up perf backtracing based on the feedback. It now uses a new method to determine the kernel addresses using the page tables, and also avoids using copy_*_user in the backtrace at all.
The goal is to avoid perf exceeding the NMI handler CPU time quota. In addition this patchkit also optimizes the generic fault handling for one case in copy_*_user (running on systems with enhanced copy string) Hopefully this version is more acceptable. -Andi -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
