Re: [x86 pmap changes] CVS commit: src/sys/arch

[Maxime Villard <max%m00nbsd.net@localhost>]

Le 07/01/2020 à 09:39, Maxime Villard a écrit :
> > Module Name:    src
> > Committed By:   ad
> > Date:           Sat Jan  4 22:49:20 UTC 2020
> >
> > Modified Files:
> >         src/sys/arch/x86/include: pmap.h pmap_pv.h
> >         src/sys/arch/x86/x86: pmap.c
> >         src/sys/arch/xen/x86: xen_pmap.c
> >
> > Log Message:
> > x86 pmap improvements, reducing system time during a build by about 15% on
> > my test machine:
>
> This breaks nvmm-intel. I have only given a quick glance, but this change
> already is wrong:
>
> -        old_pp->pp_attrs |= pmap_ept_to_pp_attrs(opte);
> +        old_pp->pp_attrs |= pmap_pte_to_pp_attrs(opte);
>
> This is an EPT function handling EPT PTEs, so "ept" was correct. Fixing
> this bug is not sufficient, so it seems that there are more bugs.
>
> Reverting the whole change puts nvmm-intel back in a functional state.
>
> You can test with this on an Intel CPU:
>
>      # modload nvmm
>      # /usr/tests/lib/libnvmm/./h_mem_assist
>
> This currently gives random crashes.
>
> Maxime

Also unrelated remark:

+	kmutex_t pm_lock		/* locks for pm_objs */
+	    __aligned(64);		/* give lock own cache line */

x86 CPUs will soon have 128-byte cache lines, and to ease the upgrade,
it is better to use COHERENCY_UNIT rather than hardcoded values.