Re: [PATCH 4/9] x86/mm/tlb: Flush remote and local TLBs concurrently

2019-06-25 Thread Andy Lutomirski
On Tue, Jun 25, 2019 at 8:48 PM Nadav Amit wrote: > > > On Jun 25, 2019, at 8:36 PM, Andy Lutomirski wrote: > > > > On Wed, Jun 12, 2019 at 11:49 PM Nadav Amit wrote: > >> To improve TLB shootdown performance, flush the remote and local TLBs > >> concurrently. Introduce flush_tlb_multi() that do

Re: [PATCH 4/9] x86/mm/tlb: Flush remote and local TLBs concurrently

2019-06-25 Thread Nadav Amit
> On Jun 25, 2019, at 8:36 PM, Andy Lutomirski wrote: > > On Wed, Jun 12, 2019 at 11:49 PM Nadav Amit wrote: >> To improve TLB shootdown performance, flush the remote and local TLBs >> concurrently. Introduce flush_tlb_multi() that does so. The current >> flush_tlb_others() interface is kept, si

Re: [PATCH 4/9] x86/mm/tlb: Flush remote and local TLBs concurrently

2019-06-25 Thread Andy Lutomirski
On Wed, Jun 12, 2019 at 11:49 PM Nadav Amit wrote: > > To improve TLB shootdown performance, flush the remote and local TLBs > concurrently. Introduce flush_tlb_multi() that does so. The current > flush_tlb_others() interface is kept, since paravirtual interfaces need > to be adapted first before

Re: [PATCH 4/9] x86/mm/tlb: Flush remote and local TLBs concurrently

2019-06-25 Thread Nadav Amit
> On Jun 25, 2019, at 8:00 PM, Dave Hansen wrote: > > On 6/25/19 7:35 PM, Nadav Amit wrote: const struct flush_tlb_info *f = info; + enum tlb_flush_reason reason; + + reason = (f->mm == NULL) ? TLB_LOCAL_SHOOTDOWN : TLB_LOCAL_MM_SHOOTDOWN; >>> >>> Should we just add the "re

Re: [PATCH 4/9] x86/mm/tlb: Flush remote and local TLBs concurrently

2019-06-25 Thread Dave Hansen
On 6/25/19 7:35 PM, Nadav Amit wrote: >>> const struct flush_tlb_info *f = info; >>> + enum tlb_flush_reason reason; >>> + >>> + reason = (f->mm == NULL) ? TLB_LOCAL_SHOOTDOWN : TLB_LOCAL_MM_SHOOTDOWN; >> >> Should we just add the "reason" to flush_tlb_info? It's OK-ish to imply >> it like

Re: [PATCH 4/9] x86/mm/tlb: Flush remote and local TLBs concurrently

2019-06-25 Thread Nadav Amit
> On Jun 25, 2019, at 2:29 PM, Dave Hansen wrote: > > On 6/12/19 11:48 PM, Nadav Amit wrote: >> To improve TLB shootdown performance, flush the remote and local TLBs >> concurrently. Introduce flush_tlb_multi() that does so. The current >> flush_tlb_others() interface is kept, since paravirtual i

Re: [PATCH 4/9] x86/mm/tlb: Flush remote and local TLBs concurrently

2019-06-25 Thread Dave Hansen
On 6/12/19 11:48 PM, Nadav Amit wrote: > To improve TLB shootdown performance, flush the remote and local TLBs > concurrently. Introduce flush_tlb_multi() that does so. The current > flush_tlb_others() interface is kept, since paravirtual interfaces need > to be adapted first before it can be remov

[PATCH 4/9] x86/mm/tlb: Flush remote and local TLBs concurrently

2019-06-13 Thread Nadav Amit
To improve TLB shootdown performance, flush the remote and local TLBs concurrently. Introduce flush_tlb_multi() that does so. The current flush_tlb_others() interface is kept, since paravirtual interfaces need to be adapted first before it can be removed. This is left for future work. In such PV en