From: Yuri Tikhonov <yur@emcraft.com>
To: Milton Miller <miltonm@bga.com>
Cc: linux-ppc <linuxppc-dev@ozlabs.org>,
dzu@denx.de, Vladimir Panfilov <pvr@emcraft.com>,
Ilya Yanok <yanok@emcraft.com>, Wolfgang Denk <wd@denx.de>
Subject: Re[2]: [2/2] powerpc: support for 256K pages on PPC 44x
Date: Fri, 14 Nov 2008 07:32:49 +0300 [thread overview]
Message-ID: <828133638.20081114073249@emcraft.com> (raw)
In-Reply-To: <f6812adbe26a68b0545f5fd25b6d9d6a@bga.com>
Hello Milton,
On Tuesday, November 11, 2008 Milton Miller wrote:
[snip]
>>>>
>>>> #ifdef CONFIG_PTE_64BIT
>>>> typedef unsigned long long pte_basic_t;
>>>> +#ifdef CONFIG_PPC_256K_PAGES
>>>> +#define PTE_SHIFT (PAGE_SHIFT - 7)
>>>
>>> This seems to be missing the comment on how many ptes are actually in
>>> the page that are in the other if and else cases.
>>
>> Ok. I'll fix this. Actually it's another hack: we don't use full page
>> for PTE table because we need to reserve something for PGD
> I don't understand "we need to reserve something for PGD". Do you=20
> mean that you would not require a second page for the PGD because the=20
> full pagetable could fit in one page? My first reaction was to say=20
> then create pgtable-nopgd.h like the other two. The page walkers=20
> support this with the advent of gigantic pages. Then I realized that=20
> might not be optimal: while the page table might fit in one page, it=20
> would mean you always allocate the pte space to cover the full address
> space. Even if your processes spread out over the 3G of address space
> allocated to them (32 bit kernel), you will allocate space for 4G,=20
> wasting 1/4 of the pte space.
> That does imply you want to allocate the pte page from a slab instead=20
> of pgalloc. Is that covered?
Well, in case of 256K PAGE_SIZE we do not need the PGD level indeed
(18 bits are used for offset, and remaining 14 bits are for PTE index=20
inside the PTE table). Even the full 256K PTE page isn't necessary to=20
cover the full range: only half of it would be enough (with 14 bits we=20
can address only 16K PTEs).
But the head_44x.S code is essentially based on the assumption of=20
2-level page addressing. Also, I may guess that eliminating of the
PGD level won't be as easy as just a re-implementation of the TLB-miss=20
handlers in head_44x.S. So, the current approach for 256K-pages=20
support was just a compromise between the required for the project=20
functionality, and the effort necessary to achieve it.
Regards, Yuri
--
Yuri Tikhonov, Senior Software Engineer
Emcraft Systems, www.emcraft.com
next prev parent reply other threads:[~2008-11-14 4:33 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-10-16 2:22 [RFC PATCH] Support for big page sizes on 44x (Updated) Ilya Yanok
2008-10-16 2:22 ` [PATCH 1/2] powerpc: add 16K/64K pages support for the 44x PPC32 architectures Ilya Yanok
2008-10-17 15:54 ` prodyut hazarika
2008-10-18 12:58 ` Josh Boyer
2008-10-18 20:36 ` prodyut hazarika
2008-10-22 14:28 ` Christian Ehrhardt
2008-10-22 17:54 ` Christian Ehrhardt
2008-10-31 23:23 ` Hollis Blanchard
2008-11-01 11:30 ` Josh Boyer
2008-11-01 21:55 ` Benjamin Herrenschmidt
2008-11-02 13:41 ` Josh Boyer
2008-11-02 21:33 ` Benjamin Herrenschmidt
2008-11-03 0:33 ` Josh Boyer
2008-11-03 0:43 ` Benjamin Herrenschmidt
2008-11-03 11:26 ` Josh Boyer
2008-11-03 20:17 ` Benjamin Herrenschmidt
2008-11-03 19:55 ` Hollis Blanchard
2008-11-03 20:00 ` Josh Boyer
2008-11-05 17:33 ` Hollis Blanchard
2008-11-06 1:48 ` David Gibson
2008-11-11 13:19 ` Josh Boyer
2008-11-11 15:00 ` Hollis Blanchard
2008-11-10 15:09 ` [1/2] " Milton Miller
2008-11-10 16:50 ` Ilya Yanok
2008-10-16 2:22 ` [PATCH 2/2] powerpc: support for 256K pages on PPC 44x Ilya Yanok
2008-11-10 15:09 ` [2/2] " Milton Miller
2008-11-10 16:24 ` Ilya Yanok
2008-11-11 14:59 ` Milton Miller
2008-11-14 4:32 ` Yuri Tikhonov [this message]
2008-11-14 15:41 ` Re[2]: " Milton Miller
2008-11-27 0:30 ` Re[4]: " Yuri Tikhonov
2008-11-11 2:17 ` [RFC PATCH] Support for big page sizes on 44x (Updated) Benjamin Herrenschmidt
2008-11-11 2:22 ` Benjamin Herrenschmidt
2008-11-24 20:32 ` Hollis Blanchard
2008-11-24 23:06 ` Wolfgang Denk
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=828133638.20081114073249@emcraft.com \
--to=yur@emcraft.com \
--cc=dzu@denx.de \
--cc=linuxppc-dev@ozlabs.org \
--cc=miltonm@bga.com \
--cc=pvr@emcraft.com \
--cc=wd@denx.de \
--cc=yanok@emcraft.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).