Re: Cortex-A53 GCC erratum compiler options

[Sad Clouds <cryintothebluesky%gmail.com@localhost>]

On Fri, 27 Sep 2019 18:54:55 +0900
Ryo Shimizu <ryo%nerv.org@localhost> wrote:

> 
> >> Hello, GCC for AArch64 has the following compiler options to work
> >> around Cortex-A53 CPU bugs. I assume they are not enabled by
> >> default by GCC.
> >> 
> >> Anyone knows if these options should always be enabled when
> >> building software for this CPU? If yes, does NetBSD already enable
> >> them for kernel and userland programs? If it doesn't, then should
> >> it?
> >> 
> >> 
> >> -mfix-cortex-a53-835769
> >> -mno-fix-cortex-a53-835769
> >> Enable or disable the workaround for the ARM Cortex-A53 erratum
> >> number
> >> 835769. This involves inserting a NOP instruction between memory
> >> instructions and 64-bit integer multiply-accumulate instructions.
> >> 
> >> -mfix-cortex-a53-843419
> >> -mno-fix-cortex-a53-843419
> >> Enable or disable the workaround for the ARM Cortex-A53 erratum
> >> number
> >> 843419. This erratum workaround is made at link time and this will
> >> only pass the corresponding flag to the linker.
> >
> >
> >OK found them, looks like they are enabled by default in NetBSD GCC
> >compilers for AArch64, as can be seen with "gcc -v" output.
> 
> for errata 843419, not only compiler options but also ld.elf_so and
> kobj_machdep.c should be fixed, but they are currently not.
> 
> There're still some errata left around aarch64 core and SoC,
> but they have not been addressed sufficiently.
> We need to do this, including Spectre.
> 
> Unfortunately, that work still hasn't gone underway...
> 
> -- 
> ryo shimizu

OK, thanks for the info