All of lore.kernel.org
 help / color / mirror / Atom feed
From: Marek Szyprowski <m.szyprowski@samsung.com>
To: "'Arnd Bergmann'" <arnd@arndb.de>,
	"'Michal Nazarewicz'" <mina86@mina86.com>,
	Marek Szyprowski <m.szyprowski@samsung.com>
Cc: linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	linux-media@vger.kernel.org, linux-mm@kvack.org,
	linaro-mm-sig@lists.linaro.org,
	"'Kyungmin Park'" <kyungmin.park@samsung.com>,
	"'Andrew Morton'" <akpm@linux-foundation.org>,
	"'KAMEZAWA Hiroyuki'" <kamezawa.hiroyu@jp.fujitsu.com>,
	"'Ankita Garg'" <ankita@in.ibm.com>,
	"'Daniel Walker'" <dwalker@codeaurora.org>,
	"'Mel Gorman'" <mel@csn.ul.ie>,
	"'Jesse Barker'" <jesse.barker@linaro.org>
Subject: RE: [PATCH 08/10] mm: cma: Contiguous Memory Allocator added
Date: Wed, 15 Jun 2011 09:11:39 +0200	[thread overview]
Message-ID: <000501cc2b2b$789a54b0$69cefe10$%szyprowski@samsung.com> (raw)
In-Reply-To: <201106142030.07549.arnd@arndb.de>

Hello,

On Tuesday, June 14, 2011 8:30 PM Arnd Bergmann wrote:

> On Tuesday 14 June 2011 18:58:35 Michal Nazarewicz wrote:
> > On Tue, 14 Jun 2011 18:03:00 +0200, Arnd Bergmann wrote:
> > > For all I know, that is something that is only true for a few very
> > > special Samsung devices,
> >
> > Maybe.  I'm just answering your question. :)
> >
> > Ah yes, I forgot that separate regions for different purposes could
> > decrease fragmentation.
> 
> That is indeed a good point, but having a good allocator algorithm
> could also solve this. I don't know too much about these allocation
> algorithms, but there are probably multiple working approaches to this.
> 
> > > I would suggest going forward without having multiple regions:
> >
> > Is having support for multiple regions a bad thing?  Frankly,
> > removing this support will change code from reading context passed
> > as argument to code reading context from global variable.  Nothing
> > is gained; functionality is lost.
> 
> What is bad IMHO is making them the default, which forces the board
> code to care about memory management details. I would much prefer
> to have contiguous allocation parameters tuned automatically to just
> work on most boards before we add ways to do board-specific hacks.

I see your concerns, but I really wonder how to determine the properties
of the global/default cma pool. You definitely don't want to give all
available memory o CMA, because it will have negative impact on kernel
operation (kernel really needs to allocate unmovable pages from time to
time). 

The only solution I see now is to provide Kconfig entry to determine
the size of the global CMA pool, but this still have some issues,
especially for multi-board kernels (each board probably will have
different amount of RAM and different memory-consuming devices
available). It looks that each board startup code still might need to
tweak the size of CMA pool. I can add a kernel command line option for
it, but such solution also will not solve all the cases (afair there
was a discussion about kernel command line parameters for memory 
configuration and the conclusion was that it should be avoided).

Best regards
-- 
Marek Szyprowski
Samsung Poland R&D Center




WARNING: multiple messages have this Message-ID (diff)
From: Marek Szyprowski <m.szyprowski@samsung.com>
To: 'Arnd Bergmann' <arnd@arndb.de>,
	'Michal Nazarewicz' <mina86@mina86.com>,
	Marek Szyprowski <m.szyprowski@samsung.com>
Cc: linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	linux-media@vger.kernel.org, linux-mm@kvack.org,
	linaro-mm-sig@lists.linaro.org,
	'Kyungmin Park' <kyungmin.park@samsung.com>,
	'Andrew Morton' <akpm@linux-foundation.org>,
	'KAMEZAWA Hiroyuki' <kamezawa.hiroyu@jp.fujitsu.com>,
	'Ankita Garg' <ankita@in.ibm.com>,
	'Daniel Walker' <dwalker@codeaurora.org>,
	'Mel Gorman' <mel@csn.ul.ie>,
	'Jesse Barker' <jesse.barker@linaro.org>
Subject: RE: [PATCH 08/10] mm: cma: Contiguous Memory Allocator added
Date: Wed, 15 Jun 2011 09:11:39 +0200	[thread overview]
Message-ID: <000501cc2b2b$789a54b0$69cefe10$%szyprowski@samsung.com> (raw)
In-Reply-To: <201106142030.07549.arnd@arndb.de>

Hello,

On Tuesday, June 14, 2011 8:30 PM Arnd Bergmann wrote:

> On Tuesday 14 June 2011 18:58:35 Michal Nazarewicz wrote:
> > On Tue, 14 Jun 2011 18:03:00 +0200, Arnd Bergmann wrote:
> > > For all I know, that is something that is only true for a few very
> > > special Samsung devices,
> >
> > Maybe.  I'm just answering your question. :)
> >
> > Ah yes, I forgot that separate regions for different purposes could
> > decrease fragmentation.
> 
> That is indeed a good point, but having a good allocator algorithm
> could also solve this. I don't know too much about these allocation
> algorithms, but there are probably multiple working approaches to this.
> 
> > > I would suggest going forward without having multiple regions:
> >
> > Is having support for multiple regions a bad thing?  Frankly,
> > removing this support will change code from reading context passed
> > as argument to code reading context from global variable.  Nothing
> > is gained; functionality is lost.
> 
> What is bad IMHO is making them the default, which forces the board
> code to care about memory management details. I would much prefer
> to have contiguous allocation parameters tuned automatically to just
> work on most boards before we add ways to do board-specific hacks.

I see your concerns, but I really wonder how to determine the properties
of the global/default cma pool. You definitely don't want to give all
available memory o CMA, because it will have negative impact on kernel
operation (kernel really needs to allocate unmovable pages from time to
time). 

The only solution I see now is to provide Kconfig entry to determine
the size of the global CMA pool, but this still have some issues,
especially for multi-board kernels (each board probably will have
different amount of RAM and different memory-consuming devices
available). It looks that each board startup code still might need to
tweak the size of CMA pool. I can add a kernel command line option for
it, but such solution also will not solve all the cases (afair there
was a discussion about kernel command line parameters for memory 
configuration and the conclusion was that it should be avoided).

Best regards
-- 
Marek Szyprowski
Samsung Poland R&D Center



--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

WARNING: multiple messages have this Message-ID (diff)
From: m.szyprowski@samsung.com (Marek Szyprowski)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 08/10] mm: cma: Contiguous Memory Allocator added
Date: Wed, 15 Jun 2011 09:11:39 +0200	[thread overview]
Message-ID: <000501cc2b2b$789a54b0$69cefe10$%szyprowski@samsung.com> (raw)
In-Reply-To: <201106142030.07549.arnd@arndb.de>

Hello,

On Tuesday, June 14, 2011 8:30 PM Arnd Bergmann wrote:

> On Tuesday 14 June 2011 18:58:35 Michal Nazarewicz wrote:
> > On Tue, 14 Jun 2011 18:03:00 +0200, Arnd Bergmann wrote:
> > > For all I know, that is something that is only true for a few very
> > > special Samsung devices,
> >
> > Maybe.  I'm just answering your question. :)
> >
> > Ah yes, I forgot that separate regions for different purposes could
> > decrease fragmentation.
> 
> That is indeed a good point, but having a good allocator algorithm
> could also solve this. I don't know too much about these allocation
> algorithms, but there are probably multiple working approaches to this.
> 
> > > I would suggest going forward without having multiple regions:
> >
> > Is having support for multiple regions a bad thing?  Frankly,
> > removing this support will change code from reading context passed
> > as argument to code reading context from global variable.  Nothing
> > is gained; functionality is lost.
> 
> What is bad IMHO is making them the default, which forces the board
> code to care about memory management details. I would much prefer
> to have contiguous allocation parameters tuned automatically to just
> work on most boards before we add ways to do board-specific hacks.

I see your concerns, but I really wonder how to determine the properties
of the global/default cma pool. You definitely don't want to give all
available memory o CMA, because it will have negative impact on kernel
operation (kernel really needs to allocate unmovable pages from time to
time). 

The only solution I see now is to provide Kconfig entry to determine
the size of the global CMA pool, but this still have some issues,
especially for multi-board kernels (each board probably will have
different amount of RAM and different memory-consuming devices
available). It looks that each board startup code still might need to
tweak the size of CMA pool. I can add a kernel command line option for
it, but such solution also will not solve all the cases (afair there
was a discussion about kernel command line parameters for memory 
configuration and the conclusion was that it should be avoided).

Best regards
-- 
Marek Szyprowski
Samsung Poland R&D Center

  parent reply	other threads:[~2011-06-15  7:12 UTC|newest]

Thread overview: 178+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-10  9:54 [PATCHv10 0/10] Contiguous Memory Allocator Marek Szyprowski
2011-06-10  9:54 ` Marek Szyprowski
2011-06-10  9:54 ` Marek Szyprowski
2011-06-10  9:54 ` [PATCH 01/10] lib: bitmap: Added alignment offset for bitmap_find_next_zero_area() Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54 ` [PATCH 02/10] lib: genalloc: Generic allocator improvements Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10 11:24   ` Alan Cox
2011-06-10 11:24     ` Alan Cox
2011-06-10 11:24     ` Alan Cox
2011-06-10 12:22     ` Marek Szyprowski
2011-06-10 12:22       ` Marek Szyprowski
2011-06-10 12:22       ` Marek Szyprowski
2011-06-10 12:52       ` Alan Cox
2011-06-10 12:52         ` Alan Cox
2011-06-10 12:52         ` Alan Cox
2011-06-10 17:16         ` Michal Nazarewicz
2011-06-10 17:16           ` Michal Nazarewicz
2011-06-10 17:16           ` Michal Nazarewicz
2011-06-14 15:49         ` [Linaro-mm-sig] " Jordan Crouse
2011-06-14 15:49           ` Jordan Crouse
2011-06-14 15:49           ` Jordan Crouse
2011-06-10  9:54 ` [PATCH 03/10] mm: move some functions from memory_hotplug.c to page_isolation.c Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54 ` [PATCH 04/10] mm: alloc_contig_freed_pages() added Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54 ` [PATCH 05/10] mm: alloc_contig_range() added Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54 ` [PATCH 06/10] mm: MIGRATE_CMA migration type added Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54 ` [PATCH 07/10] mm: MIGRATE_CMA isolation functions added Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54 ` [PATCH 08/10] mm: cma: Contiguous Memory Allocator added Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10 16:21   ` Arnd Bergmann
2011-06-10 16:21     ` Arnd Bergmann
2011-06-10 16:21     ` Arnd Bergmann
2011-06-13  9:05     ` Marek Szyprowski
2011-06-13  9:05       ` Marek Szyprowski
2011-06-13  9:05       ` Marek Szyprowski
2011-06-14 13:49       ` Arnd Bergmann
2011-06-14 13:49         ` Arnd Bergmann
2011-06-14 13:49         ` Arnd Bergmann
2011-06-14 13:55         ` Michal Nazarewicz
2011-06-14 13:55           ` Michal Nazarewicz
2011-06-14 13:55           ` Michal Nazarewicz
2011-06-14 16:03           ` Arnd Bergmann
2011-06-14 16:03             ` Arnd Bergmann
2011-06-14 16:03             ` Arnd Bergmann
2011-06-14 16:58             ` Michal Nazarewicz
2011-06-14 16:58               ` Michal Nazarewicz
2011-06-14 16:58               ` Michal Nazarewicz
2011-06-14 18:30               ` Arnd Bergmann
2011-06-14 18:30                 ` Arnd Bergmann
2011-06-14 18:30                 ` Arnd Bergmann
2011-06-14 18:40                 ` Michal Nazarewicz
2011-06-14 18:40                   ` Michal Nazarewicz
2011-06-14 18:40                   ` Michal Nazarewicz
2011-06-15  7:11                 ` Marek Szyprowski [this message]
2011-06-15  7:11                   ` Marek Szyprowski
2011-06-15  7:11                   ` Marek Szyprowski
2011-06-15  7:37                   ` Arnd Bergmann
2011-06-15  7:37                     ` Arnd Bergmann
2011-06-15  7:37                     ` Arnd Bergmann
2011-06-15  8:14                     ` Marek Szyprowski
2011-06-15  8:14                       ` Marek Szyprowski
2011-06-15  8:14                       ` Marek Szyprowski
2011-06-16  0:48                     ` [Linaro-mm-sig] " Philip Balister
2011-06-16  0:48                       ` Philip Balister
2011-06-16  0:48                       ` Philip Balister
2011-06-16  7:03                       ` Arnd Bergmann
2011-06-16  7:03                         ` Arnd Bergmann
2011-06-16  7:03                         ` Arnd Bergmann
2011-06-16  7:03                         ` Arnd Bergmann
2011-06-22  7:03                     ` Hans Verkuil
2011-06-22  7:03                       ` Hans Verkuil
2011-06-22  7:03                       ` Hans Verkuil
2011-06-22  7:32                       ` Michal Nazarewicz
2011-06-22  7:32                         ` Michal Nazarewicz
2011-06-22  7:32                         ` Michal Nazarewicz
2011-06-22 12:42                       ` Arnd Bergmann
2011-06-22 12:42                         ` Arnd Bergmann
2011-06-22 12:42                         ` Arnd Bergmann
2011-06-22 13:15                         ` Marek Szyprowski
2011-06-22 13:15                           ` Marek Szyprowski
2011-06-22 13:15                           ` Marek Szyprowski
2011-06-22 13:39                           ` Arnd Bergmann
2011-06-22 13:39                             ` Arnd Bergmann
2011-06-22 13:39                             ` Arnd Bergmann
2011-06-22 13:39                             ` Arnd Bergmann
2011-06-22 16:04                             ` Michal Nazarewicz
2011-06-22 16:04                               ` Michal Nazarewicz
2011-06-22 16:04                               ` Michal Nazarewicz
2011-06-22 15:54                           ` Michal Nazarewicz
2011-06-22 15:54                             ` Michal Nazarewicz
2011-06-22 15:54                             ` Michal Nazarewicz
2011-06-15 11:53                 ` Daniel Vetter
2011-06-15 11:53                   ` Daniel Vetter
2011-06-15 11:53                   ` Daniel Vetter
2011-06-15 13:12                   ` Thomas Hellstrom
2011-06-15 13:12                     ` Thomas Hellstrom
2011-06-15 13:12                     ` Thomas Hellstrom
2011-06-17 16:08                   ` Arnd Bergmann
2011-06-17 16:08                     ` Arnd Bergmann
2011-06-17 16:08                     ` Arnd Bergmann
2011-06-14 17:01             ` Daniel Stone
2011-06-14 17:01               ` Daniel Stone
2011-06-14 17:01               ` Daniel Stone
2011-06-14 18:58               ` Zach Pfeffer
2011-06-14 18:58                 ` Zach Pfeffer
2011-06-14 18:58                 ` Zach Pfeffer
2011-06-14 20:42                 ` Arnd Bergmann
2011-06-14 20:42                   ` Arnd Bergmann
2011-06-14 20:42                   ` Arnd Bergmann
2011-06-14 21:01                   ` Jordan Crouse
2011-06-14 21:01                     ` Jordan Crouse
2011-06-14 21:01                     ` Jordan Crouse
2011-06-15 11:27                     ` Arnd Bergmann
2011-06-15 11:27                       ` Arnd Bergmann
2011-06-15 11:27                       ` Arnd Bergmann
2011-06-15 11:27                       ` Arnd Bergmann
2011-06-15  6:29                   ` Subash Patel
2011-06-15  8:36                   ` Marek Szyprowski
2011-06-15  8:36                     ` Marek Szyprowski
2011-06-15  8:36                     ` Marek Szyprowski
2011-06-15 21:39                     ` Larry Bassel
2011-06-15 21:39                       ` Larry Bassel
2011-06-15 21:39                       ` Larry Bassel
2011-06-15 22:06                       ` Arnd Bergmann
2011-06-15 22:06                         ` Arnd Bergmann
2011-06-15 22:06                         ` Arnd Bergmann
2011-06-16 17:01                         ` Larry Bassel
2011-06-16 17:01                           ` Larry Bassel
2011-06-16 17:01                           ` Larry Bassel
2011-06-17 12:45                           ` Arnd Bergmann
2011-06-17 12:45                             ` Arnd Bergmann
2011-06-17 12:45                             ` Arnd Bergmann
2011-07-04  5:25                         ` Ankita Garg
2011-07-04  5:25                           ` Ankita Garg
2011-07-04  5:25                           ` Ankita Garg
2011-07-04 14:45                           ` Arnd Bergmann
2011-07-04 14:45                             ` Arnd Bergmann
2011-07-04 14:45                             ` Arnd Bergmann
2011-06-16  3:20                       ` Zach Pfeffer
2011-06-16  3:20                         ` Zach Pfeffer
2011-06-16  3:20                         ` Zach Pfeffer
2011-06-15  9:26                   ` Michal Nazarewicz
2011-06-15  9:26                     ` Michal Nazarewicz
2011-06-15  9:26                     ` Michal Nazarewicz
2011-06-15 11:20                     ` Arnd Bergmann
2011-06-15 11:20                       ` Arnd Bergmann
2011-06-15 11:20                       ` Arnd Bergmann
2011-06-15 11:30                       ` Michal Nazarewicz
2011-06-15 11:30                         ` Michal Nazarewicz
2011-06-15 11:30                         ` Michal Nazarewicz
2011-06-15  6:01             ` Subash Patel
2011-06-15  6:01               ` Subash Patel
2011-06-15  6:01               ` Subash Patel
2011-06-15  8:02         ` Marek Szyprowski
2011-06-15  8:02           ` Marek Szyprowski
2011-06-15  8:02           ` Marek Szyprowski
2011-06-15 11:14           ` Arnd Bergmann
2011-06-15 11:14             ` Arnd Bergmann
2011-06-15 11:14             ` Arnd Bergmann
2011-06-10  9:54 ` [PATCH 09/10] ARM: integrate CMA with dma-mapping subsystem Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54 ` [PATCH 10/10] ARM: S5PV210: add CMA support for FIMC devices on Aquila board Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski
2011-06-10  9:54   ` Marek Szyprowski

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='000501cc2b2b$789a54b0$69cefe10$%szyprowski@samsung.com' \
    --to=m.szyprowski@samsung.com \
    --cc=akpm@linux-foundation.org \
    --cc=ankita@in.ibm.com \
    --cc=arnd@arndb.de \
    --cc=dwalker@codeaurora.org \
    --cc=jesse.barker@linaro.org \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=kyungmin.park@samsung.com \
    --cc=linaro-mm-sig@lists.linaro.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mel@csn.ul.ie \
    --cc=mina86@mina86.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.