From: Russell King - ARM Linux <linux@arm.linux.org.uk>
To: Catalin Marinas <catalin.marinas@arm.com>
Cc: linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 07/20] ARM: LPAE: Page table maintenance for the 3-level format
Date: Mon, 22 Nov 2010 12:58:13 +0000 [thread overview]
Message-ID: <20101122125813.GC31227@n2100.arm.linux.org.uk> (raw)
In-Reply-To: <1289584840-18097-8-git-send-email-catalin.marinas@arm.com>
On Fri, Nov 12, 2010 at 06:00:27PM +0000, Catalin Marinas wrote:
> diff --git a/arch/arm/include/asm/pgtable.h b/arch/arm/include/asm/pgtable.h
> index 97a5de3..41236f0 100644
> --- a/arch/arm/include/asm/pgtable.h
> +++ b/arch/arm/include/asm/pgtable.h
> @@ -124,7 +124,12 @@ extern pgprot_t pgprot_kernel;
> extern struct page *empty_zero_page;
> #define ZERO_PAGE(vaddr) (empty_zero_page)
>
> +#ifdef CONFIG_ARM_LPAE
> +#define pte_pfn(pte) ((pte_val(pte) & PTE_PFN_MASK) >> PAGE_SHIFT)
> +#else
> #define pte_pfn(pte) (pte_val(pte) >> PAGE_SHIFT)
> +#endif
Just make LPAE and non-LPAE both provide PTE_PFN_MASK - for non-LPAE
this can be defined as ~0UL to optimize it away. However, PTE_PFN_MASK
is the wrong name for this - you're not masking out the PFN, but the
physical address. It only becomes a PFN when you shift.
This is important because...
> +static inline pte_t *pmd_page_vaddr(pmd_t pmd)
> +{
> + return __va(pmd_val(pmd) & PTE_PFN_MASK);
... here it becomes much more confusing - it suggests that
"pmd_val(pmd) & PTE_PFN_MASK" gives you a PFN, which you then pass to
a function which takes a physical address.
Also, pmd_page_vaddr() in my patches ends up as:
static inline pte_t *pmd_page_vaddr(pmd_t pmd)
{
+ return __va(pmd_val(pmd) & PAGE_MASK);
}
which is almost the same. I'd suggest that this becomes for both:
static inline pte_t *pmd_page_vaddr(pmd_t pmd)
{
return __va(pmd_val(pmd) & PTE_PFN_MASK & PAGE_MASK);
}
but with PTE_PFN_MASK more appropriately named.
> +}
> +
> +#else /* !CONFIG_ARM_LPAE */
> +
> #define pmd_bad(pmd) (pmd_val(pmd) & 2)
>
> #define copy_pmd(pmdpd,pmdps) \
> @@ -252,7 +285,13 @@ static inline pte_t *pmd_page_vaddr(pmd_t pmd)
> return __va(ptr);
> }
>
> +#endif /* CONFIG_ARM_LPAE */
> +
> +#ifdef CONFIG_ARM_LPAE
> +#define pmd_page(pmd) pfn_to_page(__phys_to_pfn(pmd_val(pmd) & PTE_PFN_MASK))
> +#else
> #define pmd_page(pmd) pfn_to_page(__phys_to_pfn(pmd_val(pmd)))
> +#endif
Ditto.
> diff --git a/arch/arm/include/asm/proc-fns.h b/arch/arm/include/asm/proc-fns.h
> index 8fdae9b..f00ae99 100644
> --- a/arch/arm/include/asm/proc-fns.h
> +++ b/arch/arm/include/asm/proc-fns.h
> @@ -263,6 +263,18 @@
>
> #define cpu_switch_mm(pgd,mm) cpu_do_switch_mm(virt_to_phys(pgd),mm)
>
> +#ifdef CONFIG_ARM_LPAE
> +#define cpu_get_pgd() \
> + ({ \
> + unsigned long pg, pg2; \
> + __asm__("mrrc p15, 0, %0, %1, c2" \
> + : "=r" (pg), "=r" (pg2) \
> + : \
> + : "cc"); \
> + pg &= ~(PTRS_PER_PGD*sizeof(pgd_t)-1); \
> + (pgd_t *)phys_to_virt(pg); \
> + })
> +#else
> #define cpu_get_pgd() \
> ({ \
> unsigned long pg; \
> @@ -271,6 +283,7 @@
> pg &= ~0x3fff; \
I think this wants updating to use similar math to the one above.
> @@ -81,7 +90,8 @@ void free_pgd_slow(struct mm_struct *mm, pgd_t *pgd)
> if (!pgd)
> return;
>
> - /* pgd is always present and good */
> + if (pgd_none(*pgd))
> + goto free;
This actually wants to become something more like:
+ pgd = pgd_base + pgd_index(0);
+ if (pgd_none_or_clear_bad(pgd))
+ goto no_pgd;
+ pmd = pmd_offset(pgd, 0);
+ if (pmd_none_or_clear_bad(pmd))
+ goto no_pmd;
pte = pmd_pgtable(*pmd);
pmd_clear(pmd);
pte_free(mm, pte);
+no_pmd:
+ pgd_clear(pgd);
pmd_free(mm, pmd);
+no_pgd:
free_pgd(pgd_base);
next prev parent reply other threads:[~2010-11-22 12:58 UTC|newest]
Thread overview: 77+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-11-12 18:00 [PATCH v2 00/20] ARM: Add support for the Large Physical Address Extensions Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 01/20] ARM: LPAE: Use PMD_(SHIFT|SIZE|MASK) instead of PGDIR_* Catalin Marinas
2010-11-22 12:43 ` Russell King - ARM Linux
2010-11-22 13:00 ` Catalin Marinas
2010-11-22 13:28 ` Russell King - ARM Linux
2010-11-12 18:00 ` [PATCH v2 02/20] ARM: LPAE: Factor out 2-level page table definitions into separate files Catalin Marinas
2010-11-15 23:31 ` Russell King - ARM Linux
2010-11-16 9:14 ` Catalin Marinas
2010-11-16 9:59 ` Russell King - ARM Linux
2010-11-16 10:02 ` Catalin Marinas
2010-11-16 10:04 ` Russell King - ARM Linux
2010-11-16 10:11 ` Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 03/20] ARM: LPAE: use u32 instead of unsigned long for 32-bit ptes Catalin Marinas
2010-11-14 13:19 ` Russell King - ARM Linux
2010-11-14 14:09 ` Catalin Marinas
2010-11-14 14:13 ` Catalin Marinas
2010-11-14 15:14 ` Russell King - ARM Linux
2010-11-15 9:39 ` Catalin Marinas
2010-11-15 9:47 ` Arnd Bergmann
2010-11-15 9:51 ` Catalin Marinas
2010-11-15 22:11 ` Nicolas Pitre
2010-11-15 23:35 ` Russell King - ARM Linux
2010-11-16 9:19 ` Catalin Marinas
2010-11-15 22:07 ` Nicolas Pitre
2010-11-15 17:36 ` Russell King - ARM Linux
2010-11-15 17:39 ` Catalin Marinas
2010-11-16 19:34 ` Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 04/20] ARM: LPAE: Do not assume Linux PTEs are always at PTRS_PER_PTE offset Catalin Marinas
2010-11-15 17:42 ` Russell King - ARM Linux
2010-11-15 21:46 ` Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 05/20] ARM: LPAE: Introduce L_PTE_NOEXEC and L_PTE_NOWRITE Catalin Marinas
2010-11-15 18:30 ` Russell King - ARM Linux
2010-11-16 10:07 ` Catalin Marinas
2010-11-16 15:18 ` Catalin Marinas
2010-11-16 15:32 ` Catalin Marinas
2010-11-16 18:19 ` Russell King - ARM Linux
2010-11-17 17:02 ` Catalin Marinas
2010-11-17 17:16 ` Russell King - ARM Linux
2010-11-17 17:22 ` Catalin Marinas
2010-11-17 17:24 ` Russell King - ARM Linux
2010-11-17 17:30 ` Catalin Marinas
2010-11-17 17:32 ` Russell King - ARM Linux
2010-11-17 17:34 ` Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 06/20] ARM: LPAE: Introduce the 3-level page table format definitions Catalin Marinas
2010-11-15 18:34 ` Russell King - ARM Linux
2010-11-16 9:57 ` Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 07/20] ARM: LPAE: Page table maintenance for the 3-level format Catalin Marinas
2010-11-22 12:58 ` Russell King - ARM Linux [this message]
2010-11-12 18:00 ` [PATCH v2 08/20] ARM: LPAE: MMU setup for the 3-level page table format Catalin Marinas
2010-11-14 10:13 ` Catalin Marinas
2010-11-22 13:10 ` Russell King - ARM Linux
2010-11-23 11:38 ` Catalin Marinas
2010-11-23 17:33 ` Russell King - ARM Linux
2010-11-23 17:35 ` Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 09/20] ARM: LPAE: Change setup_mm_for_reboot() to work with LPAE Catalin Marinas
2010-11-22 13:11 ` Russell King - ARM Linux
2010-11-12 18:00 ` [PATCH v2 10/20] ARM: LPAE: Remove the FIRST_USER_PGD_NR and USER_PTRS_PER_PGD definitions Catalin Marinas
2010-11-22 13:11 ` Russell King - ARM Linux
2010-11-12 18:00 ` [PATCH v2 11/20] ARM: LPAE: Add fault handling support Catalin Marinas
2010-11-22 13:15 ` Russell King - ARM Linux
2010-11-22 13:19 ` Catalin Marinas
2010-11-22 13:32 ` Russell King - ARM Linux
2010-11-22 13:38 ` Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 12/20] ARM: LPAE: Add context switching support Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 13/20] ARM: LPAE: Add SMP support for the 3-level page table format Catalin Marinas
2010-11-22 13:37 ` Russell King - ARM Linux
2010-11-12 18:00 ` [PATCH v2 14/20] ARM: LPAE: use phys_addr_t instead of unsigned long for physical addresses Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 15/20] ARM: LPAE: Use generic dma_addr_t type definition Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 16/20] ARM: LPAE: mark memory banks with start > ULONG_MAX as highmem Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 17/20] ARM: LPAE: use phys_addr_t for physical start address in early_mem Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 18/20] ARM: LPAE: add support for ATAG_MEM64 Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 19/20] ARM: LPAE: define printk format for physical addresses and page table entries Catalin Marinas
2010-11-22 13:43 ` Russell King - ARM Linux
2010-11-22 13:49 ` Catalin Marinas
2010-11-12 18:00 ` [PATCH v2 20/20] ARM: LPAE: Add the Kconfig entries Catalin Marinas
2010-11-13 12:38 ` Sergei Shtylyov
2010-11-14 10:11 ` 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=20101122125813.GC31227@n2100.arm.linux.org.uk \
--to=linux@arm.linux.org.uk \
--cc=catalin.marinas@arm.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
/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).