From: Mike Rapoport <rppt@linux.ibm.com>
To: Pavel Tatashin <pasha.tatashin@gmail.com>
Cc: Alexander Duyck <alexander.h.duyck@linux.intel.com>,
linux-mm@kvack.org, akpm@linux-foundation.org,
pavel.tatashin@microsoft.com, mhocko@suse.com,
dave.jiang@intel.com, linux-kernel@vger.kernel.org,
willy@infradead.org, davem@davemloft.net,
yi.z.zhang@linux.intel.com, khalid.aziz@oracle.com,
rppt@linux.vnet.ibm.com, vbabka@suse.cz,
sparclinux@vger.kernel.org, dan.j.williams@intel.com,
ldufour@linux.vnet.ibm.com, mgorman@techsingularity.net,
mingo@kernel.org, kirill.shutemov@linux.intel.com
Subject: Re: [mm PATCH v3 1/6] mm: Use mm_zero_struct_page from SPARC on all 64b architectures
Date: Wed, 17 Oct 2018 10:30:46 +0300 [thread overview]
Message-ID: <20181017073045.GA20004@rapoport-lnx> (raw)
In-Reply-To: <57c559f6-4858-7a52-7fbb-979caa08f240@gmail.com>
On Tue, Oct 16, 2018 at 03:01:11PM -0400, Pavel Tatashin wrote:
>
>
> On 10/15/18 4:26 PM, Alexander Duyck wrote:
> > This change makes it so that we use the same approach that was already in
> > use on Sparc on all the archtectures that support a 64b long.
> >
> > This is mostly motivated by the fact that 8 to 10 store/move instructions
> > are likely always going to be faster than having to call into a function
> > that is not specialized for handling page init.
> >
> > An added advantage to doing it this way is that the compiler can get away
> > with combining writes in the __init_single_page call. As a result the
> > memset call will be reduced to only about 4 write operations, or at least
> > that is what I am seeing with GCC 6.2 as the flags, LRU poitners, and
> > count/mapcount seem to be cancelling out at least 4 of the 8 assignments on
> > my system.
> >
> > One change I had to make to the function was to reduce the minimum page
> > size to 56 to support some powerpc64 configurations.
> >
> > Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com>
>
>
> I have tested on Broadcom's Stingray cpu with 48G RAM:
> __init_single_page() takes 19.30ns / 64-byte struct page
> Wit the change it takes 17.33ns / 64-byte struct page
I gave it a run on an OpenPower (S812LC 8348-21C) with Power8 processor and
with 128G of RAM. My results for 64-byte struct page were:
before: 4.6788ns
after: 4.5882ns
My two cents :)
> Please add this data and also the data from Intel to the description.
>
> Thank you,
> Pavel
>
> > ---
> > arch/sparc/include/asm/pgtable_64.h | 30 ------------------------------
> > include/linux/mm.h | 34 ++++++++++++++++++++++++++++++++++
> > 2 files changed, 34 insertions(+), 30 deletions(-)
> >
> > diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h
> > index 1393a8ac596b..22500c3be7a9 100644
> > --- a/arch/sparc/include/asm/pgtable_64.h
> > +++ b/arch/sparc/include/asm/pgtable_64.h
> > @@ -231,36 +231,6 @@
> > extern struct page *mem_map_zero;
> > #define ZERO_PAGE(vaddr) (mem_map_zero)
> >
> > -/* This macro must be updated when the size of struct page grows above 80
> > - * or reduces below 64.
> > - * The idea that compiler optimizes out switch() statement, and only
> > - * leaves clrx instructions
> > - */
> > -#define mm_zero_struct_page(pp) do { \
> > - unsigned long *_pp = (void *)(pp); \
> > - \
> > - /* Check that struct page is either 64, 72, or 80 bytes */ \
> > - BUILD_BUG_ON(sizeof(struct page) & 7); \
> > - BUILD_BUG_ON(sizeof(struct page) < 64); \
> > - BUILD_BUG_ON(sizeof(struct page) > 80); \
> > - \
> > - switch (sizeof(struct page)) { \
> > - case 80: \
> > - _pp[9] = 0; /* fallthrough */ \
> > - case 72: \
> > - _pp[8] = 0; /* fallthrough */ \
> > - default: \
> > - _pp[7] = 0; \
> > - _pp[6] = 0; \
> > - _pp[5] = 0; \
> > - _pp[4] = 0; \
> > - _pp[3] = 0; \
> > - _pp[2] = 0; \
> > - _pp[1] = 0; \
> > - _pp[0] = 0; \
> > - } \
> > -} while (0)
> > -
> > /* PFNs are real physical page numbers. However, mem_map only begins to record
> > * per-page information starting at pfn_base. This is to handle systems where
> > * the first physical page in the machine is at some huge physical address,
> > diff --git a/include/linux/mm.h b/include/linux/mm.h
> > index bb0de406f8e7..ec6e57a0c14e 100644
> > --- a/include/linux/mm.h
> > +++ b/include/linux/mm.h
> > @@ -102,8 +102,42 @@ static inline void set_max_mapnr(unsigned long limit) { }
> > * zeroing by defining this macro in <asm/pgtable.h>.
> > */
>
> The comment above becomes outdated. Please change, we use optimized
> mm_zero_struct_page on every 64-bit platform.
>
> > #ifndef mm_zero_struct_page
> > +#if BITS_PER_LONG == 64
> > +/* This function must be updated when the size of struct page grows above 80
> > + * or reduces below 64. The idea that compiler optimizes out switch()
> > + * statement, and only leaves move/store instructions
> > + */
> > +#define mm_zero_struct_page(pp) __mm_zero_struct_page(pp)
> > +static inline void __mm_zero_struct_page(struct page *page)
> > +{
> > + unsigned long *_pp = (void *)page;
> > +
> > + /* Check that struct page is either 56, 64, 72, or 80 bytes */
> > + BUILD_BUG_ON(sizeof(struct page) & 7);
> > + BUILD_BUG_ON(sizeof(struct page) < 56);
> > + BUILD_BUG_ON(sizeof(struct page) > 80);
> > +
> > + switch (sizeof(struct page)) {
> > + case 80:
> > + _pp[9] = 0; /* fallthrough */
> > + case 72:
> > + _pp[8] = 0; /* fallthrough */
> > + default:
> > + _pp[7] = 0; /* fallthrough */
> > + case 56:
> > + _pp[6] = 0;
> > + _pp[5] = 0;
> > + _pp[4] = 0;
> > + _pp[3] = 0;
> > + _pp[2] = 0;
> > + _pp[1] = 0;
> > + _pp[0] = 0;
> > + }
> > +}
> > +#else
> > #define mm_zero_struct_page(pp) ((void)memset((pp), 0, sizeof(struct page)))
> > #endif
> > +#endif
> >
> > /*
> > * Default maximum number of active map areas, this limits the number of vmas
> >
>
--
Sincerely yours,
Mike.
next prev parent reply other threads:[~2018-10-17 7:31 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-15 20:26 [mm PATCH v3 0/6] Deferred page init improvements Alexander Duyck
2018-10-15 20:26 ` [mm PATCH v3 1/6] mm: Use mm_zero_struct_page from SPARC on all 64b architectures Alexander Duyck
2018-10-16 19:01 ` Pavel Tatashin
2018-10-17 7:30 ` Mike Rapoport [this message]
2018-10-17 14:52 ` Alexander Duyck
2018-10-17 8:47 ` Michal Hocko
2018-10-17 15:07 ` Alexander Duyck
2018-10-17 15:12 ` Pavel Tatashin
2018-10-17 15:40 ` David Laight
2018-10-17 16:31 ` Alexander Duyck
2018-10-17 17:08 ` Pavel Tatashin
2018-10-17 16:34 ` Michal Hocko
2018-10-15 20:27 ` [mm PATCH v3 2/6] mm: Drop meminit_pfn_in_nid as it is redundant Alexander Duyck
2018-10-16 20:33 ` Pavel Tatashin
2018-10-16 20:49 ` Alexander Duyck
2018-10-16 21:06 ` Pavel Tatashin
2018-10-17 9:04 ` Michal Hocko
2018-10-15 20:27 ` [mm PATCH v3 3/6] mm: Use memblock/zone specific iterator for handling deferred page init Alexander Duyck
2018-10-17 9:11 ` Michal Hocko
2018-10-17 15:17 ` Alexander Duyck
2018-10-17 16:42 ` Mike Rapoport
2018-10-15 20:27 ` [mm PATCH v3 4/6] mm: Move hot-plug specific memory init into separate functions and optimize Alexander Duyck
2018-10-17 9:18 ` Michal Hocko
2018-10-17 15:26 ` Alexander Duyck
2018-10-24 12:36 ` Michal Hocko
2018-10-24 15:08 ` Alexander Duyck
2018-10-24 15:27 ` Michal Hocko
2018-10-24 17:35 ` Alexander Duyck
2018-10-25 12:41 ` Michal Hocko
2018-10-15 20:27 ` [mm PATCH v3 5/6] mm: Use common iterator for deferred_init_pages and deferred_free_pages Alexander Duyck
2018-10-15 20:27 ` [mm PATCH v3 6/6] mm: Add reserved flag setting to set_page_links Alexander Duyck
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=20181017073045.GA20004@rapoport-lnx \
--to=rppt@linux.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=alexander.h.duyck@linux.intel.com \
--cc=dan.j.williams@intel.com \
--cc=dave.jiang@intel.com \
--cc=davem@davemloft.net \
--cc=khalid.aziz@oracle.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=ldufour@linux.vnet.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@techsingularity.net \
--cc=mhocko@suse.com \
--cc=mingo@kernel.org \
--cc=pasha.tatashin@gmail.com \
--cc=pavel.tatashin@microsoft.com \
--cc=rppt@linux.vnet.ibm.com \
--cc=sparclinux@vger.kernel.org \
--cc=vbabka@suse.cz \
--cc=willy@infradead.org \
--cc=yi.z.zhang@linux.intel.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).