From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Cc: linuxppc-dev@lists.ozlabs.org, Paul Mackerras <paulus@samba.org>,
linux-mm@kvack.org
Subject: Re: [PATCH -V1 06/24] powerpc: Reduce PTE table memory wastage
Date: Tue, 05 Mar 2013 10:36:44 +1100 [thread overview]
Message-ID: <1362440204.21357.20.camel@pasglop> (raw)
In-Reply-To: <874ngr2zz1.fsf@linux.vnet.ibm.com>
On Mon, 2013-03-04 at 16:28 +0530, Aneesh Kumar K.V wrote:
> I added the below comment when initializing the list.
>
> +#ifdef CONFIG_PPC_64K_PAGES
> + /*
> + * Used to support 4K PTE fragment. The pages are added to list,
> + * when we have free framents in the page. We track the whether
> + * a page frament is available using page._mapcount. A value of
> + * zero indicate none of the fragments are used and page can be
> + * freed. A value of FRAG_MASK indicate all the fragments are used
> + * and hence the page will be removed from the below list.
> + */
> + INIT_LIST_HEAD(&init_mm.context.pgtable_list);
> +#endif
>
> I am not sure about why you say there is no consistent rule. Can you
> elaborate on that ?
Do you really need that list ? I assume it's meant to allow you to find
free frags when allocating but my worry is that you'll end up losing
quite a bit of node locality of PTE pages....
It may or may not work but can you investigate doing things differently
here ? The idea I want you to consider is to always allocate a full
page, but make the relationship of the fragments to PTE pages fixed. IE.
the fragment in the page is a function of the VA.
Basically, the algorithm for allocation is roughly:
- Walk the tree down to the PMD ptr (* that can be improved with a
generic change, see below)
- Check if any of the neighbouring PMDs is populated. If yes, you have
your page and pick the appropriate fragment based on the VA
- If not, allocate and populate
On free, similarly, you checked if all neighbouring PMDs have been
cleared, in which case you can fire off the page for RCU freeing.
(*) By changing pte_alloc_one to take the PMD ptr (which the call side
has right at hand) you can avoid the tree lookup.
Cheers,
Ben.
next prev parent reply other threads:[~2013-03-04 23:36 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-02-26 8:04 [PATCH -V1 00/24] THP support for PPC64 Aneesh Kumar K.V
2013-02-26 8:04 ` [PATCH -V1 01/24] powerpc: Use signed formatting when printing error Aneesh Kumar K.V
2013-02-26 8:04 ` [PATCH -V1 02/24] powerpc: Save DAR and DSISR in pt_regs on MCE Aneesh Kumar K.V
2013-02-26 8:04 ` [PATCH -V1 03/24] powerpc: Don't hard code the size of pte page Aneesh Kumar K.V
2013-02-27 23:09 ` Paul Mackerras
2013-02-26 8:04 ` [PATCH -V1 04/24] powerpc: Reduce the PTE_INDEX_SIZE Aneesh Kumar K.V
2013-02-26 8:04 ` [PATCH -V1 05/24] powerpc: Move the pte free routines from common header Aneesh Kumar K.V
2013-02-28 8:36 ` Paul Mackerras
2013-02-26 8:04 ` [PATCH -V1 06/24] powerpc: Reduce PTE table memory wastage Aneesh Kumar K.V
2013-03-04 4:58 ` Paul Mackerras
2013-03-04 10:58 ` Aneesh Kumar K.V
2013-03-04 23:36 ` Benjamin Herrenschmidt [this message]
2013-03-06 4:01 ` Aneesh Kumar K.V
2013-03-05 2:12 ` Paul Mackerras
2013-03-06 4:08 ` Aneesh Kumar K.V
2013-03-06 5:03 ` Aneesh Kumar K.V
2013-02-26 8:04 ` [PATCH -V1 07/24] powerpc: Add size argument to pgtable_cache_add Aneesh Kumar K.V
2013-03-04 5:13 ` Paul Mackerras
2013-03-04 11:02 ` Aneesh Kumar K.V
2013-03-05 1:50 ` Paul Mackerras
2013-03-06 4:23 ` Aneesh Kumar K.V
2013-02-26 8:04 ` [PATCH -V1 08/24] powerpc: Use encode avpn where we need only avpn values Aneesh Kumar K.V
2013-03-04 5:15 ` Paul Mackerras
2013-02-26 8:04 ` [PATCH -V1 09/24] powerpc: Decode the pte-lp-encoding bits correctly Aneesh Kumar K.V
2013-03-04 5:48 ` Paul Mackerras
2013-03-04 11:41 ` Aneesh Kumar K.V
2013-03-05 2:02 ` Paul Mackerras
2013-03-06 4:30 ` Aneesh Kumar K.V
2013-03-04 11:52 ` Aneesh Kumar K.V
2013-03-04 18:11 ` Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 10/24] powerpc: Return all the valid pte ecndoing in KVM_PPC_GET_SMMU_INFO ioctl Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 11/24] powerpc: Update tlbie/tlbiel as per ISA doc Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 12/24] powerpc: print both base and actual page size on hash failure Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 13/24] powerpc: Print page size info during boot Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 14/24] powerpc: Fix hpte_decode to use the correct decoding for page sizes Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 15/24] mm/THP: HPAGE_SHIFT is not a #define on some arch Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 16/24] mm/THP: Add pmd args to pgtable deposit and withdraw APIs Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 17/24] mm/THP: withdraw the pgtable after pmdp related operations Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 18/24] powerpc/THP: Implement transparent huge pages for ppc64 Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 19/24] powerpc/THP: Differentiate THP PMD entries from HUGETLB PMD entries Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 20/24] powerpc/THP: Add code to handle HPTE faults for large pages Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 21/24] powerpc: Handle huge page in perf callchain Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 22/24] powerpc/THP: hypervisor require few WIMG bit set Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 23/24] powerpc/THP: get_user_pages_fast changes Aneesh Kumar K.V
2013-02-26 8:05 ` [PATCH -V1 24/24] powerpc/THP: Enable THP on PPC64 Aneesh Kumar K.V
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=1362440204.21357.20.camel@pasglop \
--to=benh@kernel.crashing.org \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=linux-mm@kvack.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=paulus@samba.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).