[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200508212321.GR8135@suse.de>
Date: Fri, 8 May 2020 23:23:21 +0200
From: Joerg Roedel <jroedel@...e.de>
To: Peter Zijlstra <peterz@...radead.org>
Cc: Joerg Roedel <joro@...tes.org>, x86@...nel.org, hpa@...or.com,
Dave Hansen <dave.hansen@...ux.intel.com>,
Andy Lutomirski <luto@...nel.org>, rjw@...ysocki.net,
Arnd Bergmann <arnd@...db.de>,
Andrew Morton <akpm@...ux-foundation.org>,
Steven Rostedt <rostedt@...dmis.org>,
Vlastimil Babka <vbabka@...e.cz>,
Michal Hocko <mhocko@...nel.org>, linux-kernel@...r.kernel.org,
linux-acpi@...r.kernel.org, linux-arch@...r.kernel.org,
linux-mm@...ck.org
Subject: Re: [RFC PATCH 2/7] mm/vmalloc: Track which page-table levels were
modified
On Fri, May 08, 2020 at 09:10:48PM +0200, Peter Zijlstra wrote:
> On Fri, May 08, 2020 at 04:40:38PM +0200, Joerg Roedel wrote:
> > + if (mask & ARCH_PAGE_TABLE_SYNC_MASK)
> > + arch_sync_kernel_mappings(start, end);
>
> So you're relying on the compiler DCE'ing the call in the 'normal' case.
>
> Works I suppose, but I went over the patches twice to look for a default
> implementation of it before I figured that out ...
Yes, works on all architectures I compile-tested this on, with gcc-9.3.
I should probably add a comment about that.
Joerg
Powered by blists - more mailing lists