>>> On 09.03.18 at 15:40, <jgr...@suse.com> wrote:
> On 08/03/18 16:06, Jan Beulich wrote:
>>>>> On 02.03.18 at 09:14, <jgr...@suse.com> wrote:
>>> @@ -123,22 +142,14 @@ unsigned int flush_area_local(const void *va,
>>> unsigned int flags)
>>> u32 t = pre_flush();
>>>
>>> if ( !cpu_has_invpcid )
>>> - {
>>> - unsigned long cr4 = read_cr4();
>>> -
>>> - write_cr4(cr4 & ~X86_CR4_PGE);
>>> - barrier();
>>> - write_cr4(cr4);
>>> - }
>>> + do_flush_tlb(0);
>>> else
>>> - {
>>> /*
>>> * Using invpcid to flush all mappings works
>>> * regardless of whether PCID is enabled or not.
>>> * It is faster than read-modify-write CR4.
>>> */
>>> invpcid_flush_all();
>>> - }
>>
>> Btw, this is correct for FLUSH_TLB_GLOBAL, but goes too far for
>> FLUSH_TLB.
>
> You are aware that my patches didn't change anything in this regard?
Oh, indeed - I've just responded to Wei's patch.
Jan
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
