From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ob0-f176.google.com (mail-ob0-f176.google.com [209.85.214.176]) by kanga.kvack.org (Postfix) with ESMTP id 73F596B0038 for ; Tue, 16 Sep 2014 18:23:59 -0400 (EDT) Received: by mail-ob0-f176.google.com with SMTP id wo20so441469obc.21 for ; Tue, 16 Sep 2014 15:23:59 -0700 (PDT) Received: from g5t1626.atlanta.hp.com (g5t1626.atlanta.hp.com. [15.192.137.9]) by mx.google.com with ESMTPS id u5si25381074obs.31.2014.09.16.15.23.57 for (version=TLSv1 cipher=RC4-SHA bits=128/128); Tue, 16 Sep 2014 15:23:58 -0700 (PDT) Message-ID: <1410905596.28990.414.camel@misato.fc.hp.com> Subject: Re: [PATCH v2 6/6] x86, pat: Update documentation for WT changes From: Toshi Kani Date: Tue, 16 Sep 2014 16:13:16 -0600 In-Reply-To: References: <1410367910-6026-1-git-send-email-toshi.kani@hp.com> <1410367910-6026-7-git-send-email-toshi.kani@hp.com> <1410379933.28990.287.camel@misato.fc.hp.com> <1410384895.28990.312.camel@misato.fc.hp.com> <1410815951.28990.384.camel@misato.fc.hp.com> <1410886335.28990.393.camel@misato.fc.hp.com> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org List-ID: To: Yigal Korman Cc: Andy Lutomirski , "H. Peter Anvin" , Thomas Gleixner , Ingo Molnar , Andrew Morton , Arnd Bergmann , "linux-mm@kvack.org" , "linux-kernel@vger.kernel.org" , Juergen Gross , Stefan Bader , Henrique de Moraes Holschuh , Konrad Rzeszutek Wilk On Wed, 2014-09-17 at 00:45 +0300, Yigal Korman wrote: > Perhaps your patch is still valid in the context of this patch: > https://lkml.org/lkml/2014/9/9/612 > Part of the reason for creating it was the fact that ioremap is using > a fake virtual address mapping. > So I think we can still use set_memory_wt for memory created with > add_persistent_memory. > What do you think? Yes, I am aware of the work. I agree that it will address issue 1), but we still have issue 2). Since this work will create page tables, we may be able to treat NV-DIMM ranges as RAM for reservation. However, we also have an issue with RAM that set_page_memtype() needs additional bit to track WT type. I am hoping that _PGMT_WB can be redefined for WT, but I need to check it more carefully. Considering the risk of such changes, this will be a separate item. > Also, a thought: maybe the driver that will be managing the memory > added by add_persistent_memory should be responsible for resolving > aliasing issues. Yes, that's one way to do it. Under the current design, though, reserve_memtype() should work to detect aliasing for supported use-cases. Thanks, -Toshi > > Thanks, > Yigal > > On Tue, Sep 16, 2014 at 7:52 PM, Toshi Kani wrote: > > On Mon, 2014-09-15 at 18:22 -0700, Andy Lutomirski wrote: > >> On Mon, Sep 15, 2014 at 2:19 PM, Toshi Kani wrote: > >> > On Wed, 2014-09-10 at 15:34 -0600, Toshi Kani wrote: > >> >> On Wed, 2014-09-10 at 13:29 -0700, Andy Lutomirski wrote: > >> >> > On Wed, Sep 10, 2014 at 1:12 PM, Toshi Kani wrote: > >> >> > > On Wed, 2014-09-10 at 11:30 -0700, Andy Lutomirski wrote: > >> >> > >> On Wed, Sep 10, 2014 at 9:51 AM, Toshi Kani wrote: > >> >> > >> > +Drivers may map the entire NV-DIMM range with ioremap_cache and then change > >> >> > >> > +a specific range to wt with set_memory_wt. > >> >> > >> > >> >> > >> That's mighty specific :) > >> >> > > > >> >> > > How about below? > >> >> > > > >> >> > > Drivers may use set_memory_wt to set WT type for cached reserve ranges. > >> >> > > >> >> > Do they have to be cached? > >> >> > >> >> Yes, set_memory_xyz only supports WB->type->WB transition. > >> >> > >> >> > How about: > >> >> > > >> >> > Drivers may call set_memory_wt on ioremapped ranges. In this case, > >> >> > there is no need to change the memory type back before calling > >> >> > iounmap. > >> >> > > >> >> > (Or only on cached ioremapped ranges if that is, in fact, the case.) > >> >> > >> >> Sounds good. Yes, I will use cashed ioremapped ranges. > >> > > >> > Well, testing "no need to change the memory type back before calling > >> > iounmap" turns out to be a good test case. I realized that > >> > set_memory_xyz only works properly for RAM. There are two problems for > >> > using this interface for ioremapped ranges. > >> > > >> > 1) set_memory_xyz calls reserve_memtype() with __pa(addr). However, > >> > __pa() translates the addr into a fake physical address when it is an > >> > ioremapped address. > >> > > >> > 2) reserve_memtype() does not work for set_memory_xyz. For RAM, the WB > >> > state is managed untracked. Hence, WB->new->WB is not considered as a > >> > conflict. For ioremapped ranges, WB is tracked in the same way as other > >> > cache types. Hence, WB->new is considered as a conflict. > >> > > >> > In my previous testing, 2) was undetected since 1) led using a fake > >> > physical address which was not tracked for WB. This made ioremapped > >> > ranges worked just like RAM. :-( > >> > > >> > Anyway, 1) can be fixed by using slow_virt_to_phys() instead of __pa(). > >> > set_memory_xyz is already slow, but this makes it even slower, though. > >> > > >> > For 2), WB has to be continuously tracked in order to detect aliasing, > >> > ex. ioremap_cache and ioremap to a same address. So, I think > >> > reserve_memtype() needs the following changes: > >> > - Add a new arg to see if an operation is to create a new mapping or to > >> > change cache attribute. > >> > - Track overlapping maps so that cache type change to an overlapping > >> > range can be detected and failed. > >> > > >> > This level of changes requires a separate set of patches if we pursue to > >> > support ioremapped ranges. So, I am considering to take one of the two > >> > options below. > >> > > >> > A) Drop the patch for set_memory_wt. > >> > > >> > B) Keep the patch for set_memory_wt, but document that it fails with > >> > -EINVAL and its use is for RAM only. > >> > > >> > >> I vote A. I see no great reason to add code that can't be used. Once > >> someone needs this ability, they can add it :) > > > > Agreed. I will drop the patch for now. Since _PGMT_WB does not seem to > > be used for tracking WB, we might be able to use this bit for WT. But I > > need to look at the code more carefully for sure. > > > >> It's too bad that ioremap is called ioremap and not iomap. Otherwise > >> the natural solution would be to add a different function call > >> ioremap_wt that's like set_memory_wt but for ioremap ranges. Calling > >> it ioreremap_wt sounds kind of disgusting :) > > > > :) > > > > Thanks, > > -Toshi > > -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org