[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <20160520.214515.1319667674362252733.davem@davemloft.net>
Date: Fri, 20 May 2016 21:45:15 -0400 (EDT)
From: David Miller <davem@...emloft.net>
To: nitin.m.gupta@...cle.com
Cc: akpm@...ux-foundation.org, kirill.shutemov@...ux.intel.com,
khalid.aziz@...cle.com, aneesh.kumar@...ux.vnet.ibm.com,
adam.buchbinder@...il.com, minchan@...nel.org,
sfr@...b.auug.org.au, sudeep.holla@....com, mingo@...nel.org,
chris.hyser@...cle.com, dahi@...ux.vnet.ibm.com,
toshi.kani@....com, tony.luck@...el.com,
sparclinux@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v4] sparc64: Reduce TLB flushes during hugepte changes
From: Nitin Gupta <nitin.m.gupta@...cle.com>
Date: Wed, 30 Mar 2016 11:17:13 -0700
> During hugepage map/unmap, TSB and TLB flushes are currently
> issued at every PAGE_SIZE'd boundary which is unnecessary.
> We now issue the flush at REAL_HPAGE_SIZE boundaries only.
>
> Without this patch workloads which unmap a large hugepage
> backed VMA region get CPU lockups due to excessive TLB
> flush calls.
>
> Orabug: 22365539, 22643230, 22995196
>
> Signed-off-by: Nitin Gupta <nitin.m.gupta@...cle.com>
Looks good, applied and queued up for -stable.
Thanks for being so patient.
Powered by blists - more mailing lists