public inbox for linux-arm-kernel@lists.infradead.org 
 help / color / mirror / Atom feed
From: catalin.marinas@arm•com (Catalin Marinas)
To: linux-arm-kernel@lists•infradead.org
Subject: [PATCH 1/7] Add various hugetlb arm high level hooks
Date: Tue, 7 Feb 2012 11:54:48 +0000	[thread overview]
Message-ID: <20120207115448.GE3351@arm.com> (raw)
In-Reply-To: <4F3085B9.9070205@gmail.com>

On Tue, Feb 07, 2012 at 02:00:25AM +0000, bill4carson wrote:
> On 2012?02?07? 01:07, Catalin Marinas wrote:
> > On Mon, Jan 30, 2012 at 07:57:12AM +0000, bill4carson at gmail.com wrote:
> >> +static inline void set_hugepte_section(struct mm_struct *mm, unsigned long addr,
> >> +                                  pte_t *ptep, pte_t pte)
> >> +{
> >> +       pgd_t *pgd;
> >> +       pud_t *pud;
> >> +       pmd_t *pmd;
> >> +
> >> +       int col, row;
> >> +       pte_t **huge_linuxpte =&mm->huge_2m_pte[0];
> >> +
> >> +       row = HUGEPAGE_2M_PTE_ARRAY_ROW(addr);
> >> +       col = HUGEPAGE_2M_PTE_ARRAY_COL(addr);
> >> +
> >> +       /* an valid pte pointer is expected */
> >> +       BUG_ON(huge_linuxpte[row] == 0);
> >> +       BUG_ON(ptep !=&huge_linuxpte[row][col]);
> >> +
> >> +       /* set linux pte first */
> >> +       huge_linuxpte[row][col] = pte;
> >> +
> >> +       /* set hardware pte */
> >> +       pgd = pgd_offset(mm, addr);
> >> +       pud = pud_offset(pgd, addr);
> >> +       pmd = pmd_offset(pud, addr);
> >> +
> >> +       set_hugepte_at(mm, addr, pmd, pte);
> >> +}
> >
> > I haven't followed the whole structure of your patches but do we need to
> > walk the page tables here? Isn't the ptep the same as the pmd when
> > passed to this function (at least it was with my LPAE implementation).
> 
> Here, ptep is not the same as pmd, mm layer always manages linux pte.
> For normal page, linux pte and hardware pte is just an 2048 bytes offset
> away and cpu_v7_set_pte_ext can set both linux/hardware pte easily.
> 
> For huge page, linux pte is stored somewhere else,far away from hardware
> pte table, in set_hugepte_section, "ptep" is the huge page based linux
> pte address, first set linux pte value "pte" in there; and hardware pmd
> address could only be derived from "addr" by a page table walk, then
> setting hardware pmd value in set_hugepte_at, which
> cpu_v7_set_hugepte_ext does the whole job.

OK. I thought we already have some pointer in the Linux pmd for the
actual hardware pmd. Walking the page table isn't expensive in this
case, with only 2 levels. For LPAE we don't even need this since we
don't have a separate Linux pte/pmd.

-- 
Catalin

  reply	other threads:[~2012-02-07 11:54 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-01-30  7:57 [RFC] ARM hugetlb support bill4carson at gmail.com
2012-01-30  7:57 ` [PATCH 1/7] Add various hugetlb arm high level hooks bill4carson at gmail.com
2012-02-06 17:07   ` Catalin Marinas
2012-02-07  2:00     ` bill4carson
2012-02-07 11:54       ` Catalin Marinas [this message]
2012-02-07 12:15   ` Catalin Marinas
2012-02-07 12:57     ` carson bill
2012-01-30  7:57 ` [PATCH 2/7] Add various hugetlb page table fix bill4carson at gmail.com
2012-01-31  9:57   ` Catalin Marinas
2012-01-31  9:58   ` Russell King - ARM Linux
2012-01-31 12:25     ` Catalin Marinas
2012-02-01  3:10       ` bill4carson
2012-02-06 16:26         ` Catalin Marinas
2012-02-07  1:42           ` bill4carson
2012-02-07 11:50             ` Catalin Marinas
2012-02-07 13:24               ` carson bill
2012-02-07 14:11                 ` Catalin Marinas
2012-02-07 14:46                   ` carson bill
2012-02-07 15:09                     ` Catalin Marinas
2012-02-07 15:41                       ` carson bill
2012-01-30  7:57 ` [PATCH 3/7] Introduce set_hugepte_ext api for huge page hardware page table setup bill4carson at gmail.com
2012-01-30  7:57 ` [PATCH 4/7] Store huge page linux pte in mm_struct bill4carson at gmail.com
2012-01-31  9:37   ` Catalin Marinas
2012-01-31 10:01   ` Russell King - ARM Linux
2012-02-01  5:45     ` bill4carson
2012-02-06  2:04       ` bill4carson
2012-02-06 10:29         ` Catalin Marinas
2012-02-06 14:40           ` carson bill
2012-01-30  7:57 ` [PATCH 5/7] Using do_page_fault for section fault handling bill4carson at gmail.com
2012-01-30  7:57 ` [PATCH 6/7] Add hugetlb Kconfig option bill4carson at gmail.com
2012-01-30  7:57 ` [PATCH 7/7] Minor compiling fix bill4carson at gmail.com
2012-01-31  9:29 ` [RFC] ARM hugetlb support Catalin Marinas
2012-02-01  1:56   ` bill4carson
2012-02-02 14:38     ` Catalin Marinas
2012-02-03  1:41       ` bill4carson
2012-02-06 16:29         ` Catalin Marinas
  -- strict thread matches above, loose matches on Subject: below --
2012-02-13  9:44 [RFC-PATCH V2] " Bill Carson
2012-02-13  9:44 ` [PATCH 1/7] Add various hugetlb arm high level hooks Bill Carson
2012-02-28 17:30   ` Catalin Marinas
2012-02-29  2:34     ` bill4carson
2012-02-29  9:39       ` Catalin Marinas
2012-02-29 10:21         ` bill4carson
2012-02-29 10:23           ` Catalin Marinas
2012-02-29 10:32   ` Catalin Marinas
2012-02-29 11:28     ` bill4carson
2012-02-29 11:36       ` Catalin Marinas
2012-02-29 15:38       ` Catalin Marinas
2012-03-08  0:35         ` bill4carson
2012-03-08  9:21           ` Catalin Marinas
2012-02-29 12:31   ` Arnd Bergmann
2012-02-29 14:22     ` Catalin Marinas

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=20120207115448.GE3351@arm.com \
    --to=catalin.marinas@arm$(echo .)com \
    --cc=linux-arm-kernel@lists$(echo .)infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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