On Fri, Dec 9, 2016 at 11:10 AM, Florian Weimer <[email protected]> wrote: > We would like to enable hardware-assisted lock optimizations in glibc on > multiple architectures. In general, this feature works only on production > hardware with current firmware, and not on pre-production machines some > vendors provide for architecture bringup.
Have you got an overview of the required instructions/generations of CPU architecture needed for each different architecture by chance? From there it might be easier to give details of where all the build HW is at. In the case of build VMs (x86/aarch64/Power64) does this need to be emulated by qemu/kvm or is it passed through from the host CPU? > Are the Fedora builders using hardware with support contracts, and are > vendor firmware updates applied occasionally, so that we can rely on what > the CPU/firmware reports about CPU features? (Some vendors had to disable > lock optimizations through firmware updates because the CPU implementation > was buggy.) > > Do we at least apply microcode updates early during Linux boot on the > builders? Only for x86, the cpu microcode update utility is only supported on that architecture. I believe all the other architectures apply any CPU firmware via the OS firmware/BIOS/what ever it's called on the various platforms. > What about default Fedora installations? Do they come with early boot > microcode updates? See the microcode_ctl package. > If all this is covered on the Fedora side, we probably can just enable lock > optimizations, and we do not need to implement complicated hardware > blacklisting. All builders run Fedora, some are physical hardware, some are VMs running on a RHEL hypervisor. As I mentioned above what does the virtual hardware need to support in this regard? Have you got some upstream docs about it for the various architectures? Peter _______________________________________________ devel mailing list -- [email protected] To unsubscribe send an email to [email protected]
