Re: [Xen-devel] [PATCH v2] x86/ctxt-switch: Document and improve GDT handling

[Jan Beulich <JBeulich@xxxxxxxx>]

On 09.07.2019 18:51, Andrew Cooper wrote:
> write_full_gdt_ptes() looks suspicious - it is not generally safe to
> iterate over (mfn + i).

I'm still not happy about this (misleading) statement: The safety of
this depends, as said, on how the allocation gets done. And the way
that is done at the moment, the addition is safe. I realize that
"generally" may be meant to cover that case, but personally I find
this not clear enough.

>  However, the loop is entirely unnecessary, as
> NR_RESERVED_GDT_PAGES is 1.  Dropping it makes the code substantially
> more clear, and with it dropped, write_full_gdt_ptes() becomes more
> obviously a poor name, so rename it to update_xen_slot_in_full_gdt().
> 
> Furthermore, calling virt_to_mfn() in the context switch path is a lot
> of wasted cycles for a result which is constant after boot.
> 
> Begin by documenting how Xen handles the GDTs across context switch.
> 
>  From this, we observe that load_full_gdt() is completely independent of
> the current CPU, and load_default_gdt() only needs the current CPU's
> regular GDT.  (This is a change in behaviour, as previously it may have
> used the compat GDT, but either will do.)
> 
> Add two extra per-cpu variables which cache the L1e for the regular and compat
> GDT, calculated in cpu_smpboot_alloc()/trap_init() as appropriate, so
> update_xen_slot_in_full_gdt() doesn't need to waste time performing the same
> calculation on every context switch.
> 
> One performance scenario of Juergen's (time to build the hypervisor on
> an 8 CPU system, with two single-vCPU MiniOS VMs constantly interrupting
> dom0 with events) shows the following, average over 5 measurements:
> 
>              elapsed  user   system
>    Unpatched  66.51  232.93  109.21
>    Patched    57.00  225.47  105.47
> 
> which is a substantial improvement.
> 
> Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
> Tested-by: Juergen Gross <jgross@xxxxxxxx>
> Reviewed-by: Juergen Gross <jgross@xxxxxxxx>

Preferably with the first sentence above changed/amended:
Reviewed-by: Jan Beulich <jbeulich@xxxxxxxx>

Jan
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel