On Mon, Nov 02, 2015 at 03:49:17PM +0000, Hunt, David wrote:
> On 02/11/2015 15:36, Jan Viktorin wrote:
> >On Mon, 2 Nov 2015 15:26:19 +0000
> --snip--
> >It was looking like we can share a lot of common code for both
> >architectures. I didn't know how much different are the cpuflags.
>
> CPU flags for ARMv8 are looking like this now. Quite different to the ARMv7
> ones.
>
> static const struct feature_entry cpu_feature_table[] = {
> FEAT_DEF(FP, 0x00000001, 0, REG_HWCAP, 0)
> FEAT_DEF(ASIMD, 0x00000001, 0, REG_HWCAP, 1)
> FEAT_DEF(EVTSTRM, 0x00000001, 0, REG_HWCAP, 2)
> FEAT_DEF(AES, 0x00000001, 0, REG_HWCAP, 3)
> FEAT_DEF(PMULL, 0x00000001, 0, REG_HWCAP, 4)
> FEAT_DEF(SHA1, 0x00000001, 0, REG_HWCAP, 5)
> FEAT_DEF(SHA2, 0x00000001, 0, REG_HWCAP, 6)
> FEAT_DEF(CRC32, 0x00000001, 0, REG_HWCAP, 7)
> FEAT_DEF(AARCH32, 0x00000001, 0, REG_PLATFORM, 0)
> FEAT_DEF(AARCH64, 0x00000001, 0, REG_PLATFORM, 1)
> };
>
> >IMHO, it'd be better to have two directories arm and arm64. I thought
> >to refer from arm64 to arm where possible. But I don't know whether is
> >this possible with the DPDK build system.
>
> I think both methodologies have their pros and cons. However, I'd lean
> towards the common directory with the "filename_32/64.h" scheme, as that
> similar to the x86 methodology, and we don't need to tweak the include paths
> to pull files from multiple directories.
>
I agree. Jan, could you please send the next version with
filename_32/64.h for atomic and cpuflags(ie for all header files).
I can re-base and send the complete arm64 patch based on your version.
Thanks,
Jerin
> Dave
>
