[PATCH v7 0/8] Raid: enable talitos xor offload for improving performance

[PATCH v7 0/8] Raid: enable talitos xor offload for improving performance

[qiang.liu]

Hi all,

The following 8 patches enabling fsl-dma and talitos offload raid
operations for improving raid performance and balancing CPU load.

These patches include talitos, fsl-dma and carma module (caram uses
some features of fsl-dma).

Write performance will be improved by 25-30% tested by iozone.
Write performance is improved about 2% after using spin_lock_bh replace
spin_lock_irqsave.
CPU load will be reduced by 8%.

"fwiw, I gave v5 a test-drive, setting up a RAID5 array on ramdisks
[1], and this patchseries, along with FSL_DMA && NET_DMA set seems
to be holding water, so this series gets my:"

Tested-by: Kim Phillips <kim.phillips@freescale.com>

[1] mdadm --create --verbose --force /dev/md0 --level=raid5 --raid-devices=4 \
	/dev/ram[0123]

Changes in v7:
	- add test result which is provided by Kim Phillips;
	- correct one coding style issue in patch 5/8;
	- add comments by Arnd Bergmann in patch 6/8;

Changes in v6:
	- swap the order of original patch 3/6 and 4/6;
	- merge Ira's patch to reduce the size of original patch;
	- merge Ira's patch of carma in 8/8;
	- update documents and descriptions according to Ira's advice;

Changes in v5:
	- add detail description in patch 3/6 about the process of completed
	descriptor, the process is in align with fsl-dma Reference Manual,
	illustrate the potential risk and how to reproduce it;
	- drop the patch 7/7 in v4 according to Timur's comments;

Changes in v4:
	- fix an error in talitos when dest addr is same with src addr, dest
	should be freed only one time if src is same with dest addr;
	- correct coding style in fsl-dma according to Ira's comments;
	- fix a race condition in fsl-dma fsl_tx_status(), remove the interface
	which is used to free descriptors in queue ld_completed, this interface
	has been included in fsldma_cleanup_descriptor(), in v3, there is one
	place missed spin_lock protect;
	- split the original patch 3/4 up to 2 patches 3/7 and 4/7 according to
	Li Yang's comments;
	- fix a warning of unitialized cookie;
	- add memory copy self test in fsl-dma;
	- add more detail description about use spin_lock_bh() to instead of
	spin_lock_irqsave() according to Timur's comments.

Changes in v3:
	- change release process of fsl-dma descriptor for resolve the
	potential race condition;
	- add test result when use spin_lock_bh replace spin_lock_irqsave;
	- modify the benchmark results according to the latest patch.

Changes in v2:
	- rebase onto cryptodev tree;
	- split the patch 3/4 up to 3 independent patches;
	- remove the patch 4/4, the fix is not for cryptodev tree;

Qiang Liu (8):
      Talitos: Support for async_tx XOR offload
      fsl-dma: remove attribute DMA_INTERRUPT of dmaengine
      fsl-dma: add fsl_dma_free_descriptor() to reduce code duplication
      fsl-dma: move functions to avoid forward declarations
      fsl-dma: change release process of dma descriptor for supporting async_tx
      fsl-dma: use spin_lock_bh to instead of spin_lock_irqsave
      fsl-dma: fix a warning of unitialized cookie
      carma: remove unnecessary DMA_INTERRUPT capability

 drivers/crypto/Kconfig                  |    9 +
 drivers/crypto/talitos.c                |  413 ++++++++++++++++++++++++++
 drivers/crypto/talitos.h                |   53 ++++
 drivers/dma/fsldma.c                    |  488 +++++++++++++++++--------------
 drivers/dma/fsldma.h                    |   17 +-
 drivers/misc/carma/carma-fpga-program.c |    1 -
 drivers/misc/carma/carma-fpga.c         |    2 +-
 7 files changed, 761 insertions(+), 222 deletions(-)

Re: [PATCH v7 0/8] Raid: enable talitos xor offload for improving performance

[Ira W. Snyder]

On Thu, Aug 09, 2012 at 04:19:35PM +0800, qiang.liu@freescale.com wrote:
> Hi all,
> 
> The following 8 patches enabling fsl-dma and talitos offload raid
> operations for improving raid performance and balancing CPU load.
> 
> These patches include talitos, fsl-dma and carma module (caram uses
> some features of fsl-dma).
> 
> Write performance will be improved by 25-30% tested by iozone.
> Write performance is improved about 2% after using spin_lock_bh replace
> spin_lock_irqsave.
> CPU load will be reduced by 8%.
> 
> "fwiw, I gave v5 a test-drive, setting up a RAID5 array on ramdisks
> [1], and this patchseries, along with FSL_DMA && NET_DMA set seems
> to be holding water, so this series gets my:"
> 
> Tested-by: Kim Phillips <kim.phillips@freescale.com>
> 

The fsldma parts of the series all look great to me.

Thanks,
Ira

> [1] mdadm --create --verbose --force /dev/md0 --level=raid5 --raid-devices=4 \
> 	/dev/ram[0123]
> 
> Changes in v7:
> 	- add test result which is provided by Kim Phillips;
> 	- correct one coding style issue in patch 5/8;
> 	- add comments by Arnd Bergmann in patch 6/8;
> 
> Changes in v6:
> 	- swap the order of original patch 3/6 and 4/6;
> 	- merge Ira's patch to reduce the size of original patch;
> 	- merge Ira's patch of carma in 8/8;
> 	- update documents and descriptions according to Ira's advice;
> 
> Changes in v5:
> 	- add detail description in patch 3/6 about the process of completed
> 	descriptor, the process is in align with fsl-dma Reference Manual,
> 	illustrate the potential risk and how to reproduce it;
> 	- drop the patch 7/7 in v4 according to Timur's comments;
> 
> Changes in v4:
> 	- fix an error in talitos when dest addr is same with src addr, dest
> 	should be freed only one time if src is same with dest addr;
> 	- correct coding style in fsl-dma according to Ira's comments;
> 	- fix a race condition in fsl-dma fsl_tx_status(), remove the interface
> 	which is used to free descriptors in queue ld_completed, this interface
> 	has been included in fsldma_cleanup_descriptor(), in v3, there is one
> 	place missed spin_lock protect;
> 	- split the original patch 3/4 up to 2 patches 3/7 and 4/7 according to
> 	Li Yang's comments;
> 	- fix a warning of unitialized cookie;
> 	- add memory copy self test in fsl-dma;
> 	- add more detail description about use spin_lock_bh() to instead of
> 	spin_lock_irqsave() according to Timur's comments.
> 
> Changes in v3:
> 	- change release process of fsl-dma descriptor for resolve the
> 	potential race condition;
> 	- add test result when use spin_lock_bh replace spin_lock_irqsave;
> 	- modify the benchmark results according to the latest patch.
> 
> Changes in v2:
> 	- rebase onto cryptodev tree;
> 	- split the patch 3/4 up to 3 independent patches;
> 	- remove the patch 4/4, the fix is not for cryptodev tree;
> 
> Qiang Liu (8):
>       Talitos: Support for async_tx XOR offload
>       fsl-dma: remove attribute DMA_INTERRUPT of dmaengine
>       fsl-dma: add fsl_dma_free_descriptor() to reduce code duplication
>       fsl-dma: move functions to avoid forward declarations
>       fsl-dma: change release process of dma descriptor for supporting async_tx
>       fsl-dma: use spin_lock_bh to instead of spin_lock_irqsave
>       fsl-dma: fix a warning of unitialized cookie
>       carma: remove unnecessary DMA_INTERRUPT capability
> 
>  drivers/crypto/Kconfig                  |    9 +
>  drivers/crypto/talitos.c                |  413 ++++++++++++++++++++++++++
>  drivers/crypto/talitos.h                |   53 ++++
>  drivers/dma/fsldma.c                    |  488 +++++++++++++++++--------------
>  drivers/dma/fsldma.h                    |   17 +-
>  drivers/misc/carma/carma-fpga-program.c |    1 -
>  drivers/misc/carma/carma-fpga.c         |    2 +-
>  7 files changed, 761 insertions(+), 222 deletions(-)
> 
> _______________________________________________
> Linuxppc-dev mailing list
> Linuxppc-dev@lists.ozlabs.org
> https://lists.ozlabs.org/listinfo/linuxppc-dev

RE: [PATCH v7 0/8] Raid: enable talitos xor offload for improving performance

[Liu Qiang-B32616]

Hi Vinod,

Would you like to apply this series from patch 2/8 to 7/8) in your tree?
The link as below,
http://patchwork.ozlabs.org/patch/176023/
http://patchwork.ozlabs.org/patch/176024/
http://patchwork.ozlabs.org/patch/176025/
http://patchwork.ozlabs.org/patch/176026/
http://patchwork.ozlabs.org/patch/176027/
http://patchwork.ozlabs.org/patch/176028/

After that, Herbert will merge patch 1/8, http://patchwork.ozlabs.org/patch/176022/
and Greg apply patch 8/8, http://patchwork.ozlabs.org/patch/176029/

Thanks.
 

> -----Original Message-----
> From: Ira W. Snyder [mailto:iws@ovro.caltech.edu]
> Sent: Friday, August 10, 2012 1:03 AM
> To: Liu Qiang-B32616
> Cc: linux-crypto@vger.kernel.org; vinod.koul@intel.com;
> dan.j.williams@intel.com; herbert@gondor.hengli.com.au; arnd@arndb.de;
> gregkh@linuxfoundation.org; linuxppc-dev@lists.ozlabs.org; linux-
> kernel@vger.kernel.org; dan.j.williams@gmail.com
> Subject: Re: [PATCH v7 0/8] Raid: enable talitos xor offload for
> improving performance
> 
> On Thu, Aug 09, 2012 at 04:19:35PM +0800, qiang.liu@freescale.com wrote:
> > Hi all,
> >
> > The following 8 patches enabling fsl-dma and talitos offload raid
> > operations for improving raid performance and balancing CPU load.
> >
> > These patches include talitos, fsl-dma and carma module (caram uses
> > some features of fsl-dma).
> >
> > Write performance will be improved by 25-30% tested by iozone.
> > Write performance is improved about 2% after using spin_lock_bh
> > replace spin_lock_irqsave.
> > CPU load will be reduced by 8%.
> >
> > "fwiw, I gave v5 a test-drive, setting up a RAID5 array on ramdisks
> > [1], and this patchseries, along with FSL_DMA && NET_DMA set seems to
> > be holding water, so this series gets my:"
> >
> > Tested-by: Kim Phillips <kim.phillips@freescale.com>
> >
> 
> The fsldma parts of the series all look great to me.
> 
> Thanks,
> Ira
> 
> > [1] mdadm --create --verbose --force /dev/md0 --level=raid5 --raid-
> devices=4 \
> > 	/dev/ram[0123]
> >
> > Changes in v7:
> > 	- add test result which is provided by Kim Phillips;
> > 	- correct one coding style issue in patch 5/8;
> > 	- add comments by Arnd Bergmann in patch 6/8;
> >
> > Changes in v6:
> > 	- swap the order of original patch 3/6 and 4/6;
> > 	- merge Ira's patch to reduce the size of original patch;
> > 	- merge Ira's patch of carma in 8/8;
> > 	- update documents and descriptions according to Ira's advice;
> >
> > Changes in v5:
> > 	- add detail description in patch 3/6 about the process of
> completed
> > 	descriptor, the process is in align with fsl-dma Reference Manual,
> > 	illustrate the potential risk and how to reproduce it;
> > 	- drop the patch 7/7 in v4 according to Timur's comments;
> >
> > Changes in v4:
> > 	- fix an error in talitos when dest addr is same with src addr,
> dest
> > 	should be freed only one time if src is same with dest addr;
> > 	- correct coding style in fsl-dma according to Ira's comments;
> > 	- fix a race condition in fsl-dma fsl_tx_status(), remove the
> interface
> > 	which is used to free descriptors in queue ld_completed, this
> interface
> > 	has been included in fsldma_cleanup_descriptor(), in v3, there is
> one
> > 	place missed spin_lock protect;
> > 	- split the original patch 3/4 up to 2 patches 3/7 and 4/7
> according to
> > 	Li Yang's comments;
> > 	- fix a warning of unitialized cookie;
> > 	- add memory copy self test in fsl-dma;
> > 	- add more detail description about use spin_lock_bh() to instead
> of
> > 	spin_lock_irqsave() according to Timur's comments.
> >
> > Changes in v3:
> > 	- change release process of fsl-dma descriptor for resolve the
> > 	potential race condition;
> > 	- add test result when use spin_lock_bh replace spin_lock_irqsave;
> > 	- modify the benchmark results according to the latest patch.
> >
> > Changes in v2:
> > 	- rebase onto cryptodev tree;
> > 	- split the patch 3/4 up to 3 independent patches;
> > 	- remove the patch 4/4, the fix is not for cryptodev tree;
> >
> > Qiang Liu (8):
> >       Talitos: Support for async_tx XOR offload
> >       fsl-dma: remove attribute DMA_INTERRUPT of dmaengine
> >       fsl-dma: add fsl_dma_free_descriptor() to reduce code duplication
> >       fsl-dma: move functions to avoid forward declarations
> >       fsl-dma: change release process of dma descriptor for supporting
> async_tx
> >       fsl-dma: use spin_lock_bh to instead of spin_lock_irqsave
> >       fsl-dma: fix a warning of unitialized cookie
> >       carma: remove unnecessary DMA_INTERRUPT capability
> >
> >  drivers/crypto/Kconfig                  |    9 +
> >  drivers/crypto/talitos.c                |  413
> ++++++++++++++++++++++++++
> >  drivers/crypto/talitos.h                |   53 ++++
> >  drivers/dma/fsldma.c                    |  488 +++++++++++++++++------
> --------
> >  drivers/dma/fsldma.h                    |   17 +-
> >  drivers/misc/carma/carma-fpga-program.c |    1 -
> >  drivers/misc/carma/carma-fpga.c         |    2 +-
> >  7 files changed, 761 insertions(+), 222 deletions(-)
> >
> > _______________________________________________
> > Linuxppc-dev mailing list
> > Linuxppc-dev@lists.ozlabs.org
> > https://lists.ozlabs.org/listinfo/linuxppc-dev

Re: [PATCH v7 0/8] Raid: enable talitos xor offload for improving performance

[Dan Williams]

On Tue, Aug 14, 2012 at 2:04 AM, Liu Qiang-B32616 <B32616@freescale.com> wrote:
> Hi Vinod,
>
> Would you like to apply this series from patch 2/8 to 7/8) in your tree?
> The link as below,
> http://patchwork.ozlabs.org/patch/176023/
> http://patchwork.ozlabs.org/patch/176024/
> http://patchwork.ozlabs.org/patch/176025/
> http://patchwork.ozlabs.org/patch/176026/
> http://patchwork.ozlabs.org/patch/176027/
> http://patchwork.ozlabs.org/patch/176028/
>

Hi, sorry for the recent silence I've been transitioning and am now
just catching up.  I'll take a look and then it's fine for these to go
through Vinod's tree.

--
Dan

RE: [PATCH v7 0/8] Raid: enable talitos xor offload for improving performance

[Liu Qiang-B32616]

> -----Original Message-----
> From: dan.j.williams@gmail.com [mailto:dan.j.williams@gmail.com] On
> Behalf Of Dan Williams
> Sent: Wednesday, August 15, 2012 4:02 AM
> To: Liu Qiang-B32616
> Cc: dan.j.williams@intel.com; vinod.koul@intel.com; arnd@arndb.de;
> herbert@gondor.apana.org.au; gregkh@linuxfoundation.org; linuxppc-
> dev@lists.ozlabs.org; linux-kernel@vger.kernel.org; linux-
> crypto@vger.kernel.org; Ira W. Snyder
> Subject: Re: [PATCH v7 0/8] Raid: enable talitos xor offload for
> improving performance
> 
> On Tue, Aug 14, 2012 at 2:04 AM, Liu Qiang-B32616 <B32616@freescale.com>
> wrote:
> > Hi Vinod,
> >
> > Would you like to apply this series from patch 2/8 to 7/8) in your tree?
> > The link as below,
> > http://patchwork.ozlabs.org/patch/176023/
> > http://patchwork.ozlabs.org/patch/176024/
> > http://patchwork.ozlabs.org/patch/176025/
> > http://patchwork.ozlabs.org/patch/176026/
> > http://patchwork.ozlabs.org/patch/176027/
> > http://patchwork.ozlabs.org/patch/176028/
> >
> 
> Hi, sorry for the recent silence I've been transitioning and am now
> just catching up.  I'll take a look and then it's fine for these to go
> through Vinod's tree.
Hello Dan,

Please review, this issue has been continued since many years. I hope we can fix
it this time. Thanks.

> 
> --
> Dan

RE: [PATCH v7 0/8] Raid: enable talitos xor offload for improving performance

[Liu Qiang-B32616]

Hi Dan,

Ping?
Can you apply these patches? Thanks.


- Qiang

> -----Original Message-----
> From: dan.j.williams@gmail.com [mailto:dan.j.williams@gmail.com] On
> Behalf Of Dan Williams
> Sent: Wednesday, August 15, 2012 4:02 AM
> To: Liu Qiang-B32616
> Cc: dan.j.williams@intel.com; vinod.koul@intel.com; arnd@arndb.de;
> herbert@gondor.apana.org.au; gregkh@linuxfoundation.org; linuxppc-
> dev@lists.ozlabs.org; linux-kernel@vger.kernel.org; linux-
> crypto@vger.kernel.org; Ira W. Snyder
> Subject: Re: [PATCH v7 0/8] Raid: enable talitos xor offload for
> improving performance
> 
> On Tue, Aug 14, 2012 at 2:04 AM, Liu Qiang-B32616 <B32616@freescale.com>
> wrote:
> > Hi Vinod,
> >
> > Would you like to apply this series from patch 2/8 to 7/8) in your tree?
> > The link as below,
> > http://patchwork.ozlabs.org/patch/176023/
> > http://patchwork.ozlabs.org/patch/176024/
> > http://patchwork.ozlabs.org/patch/176025/
> > http://patchwork.ozlabs.org/patch/176026/
> > http://patchwork.ozlabs.org/patch/176027/
> > http://patchwork.ozlabs.org/patch/176028/
> >
> 
> Hi, sorry for the recent silence I've been transitioning and am now
> just catching up.  I'll take a look and then it's fine for these to go
> through Vinod's tree.
> 
> --
> Dan

Re: [PATCH v7 0/8] Raid: enable talitos xor offload for improving performance

[Dan Williams]

On Wed, 2012-08-29 at 11:15 +0000, Liu Qiang-B32616 wrote:
> Hi Dan,
> 
> Ping?
> Can you apply these patches? Thanks.
> 

I'm working my way through them.

The first thing I notice is that xor_chan->desc_lock is taken
inconsistently.  I.e. spin_lock_irqsave() in talitos_process_pending()
and spin_lock_bh() everywhere else.  Have you run these patches with
lockdep?

--
Dan

RE: [PATCH v7 0/8] Raid: enable talitos xor offload for improving performance

[Liu Qiang-B32616]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="utf-8", Size: 1804 bytes --]

> -----Original Message-----
> From: Dan Williams [mailto:djbw@fb.com]
> Sent: Wednesday, August 29, 2012 10:53 PM
> To: Liu Qiang-B32616
> Cc: vinod.koul@intel.com; arnd@arndb.de; herbert@gondor.apana.org.au;
> gregkh@linuxfoundation.org; linuxppc-dev@lists.ozlabs.org; linux-
> kernel@vger.kernel.org; linux-crypto@vger.kernel.org; Ira W. Snyder
> Subject: Re: [PATCH v7 0/8] Raid: enable talitos xor offload for
> improving performance
> 
> On Wed, 2012-08-29 at 11:15 +0000, Liu Qiang-B32616 wrote:
> > Hi Dan,
> >
> > Ping?
> > Can you apply these patches? Thanks.
> >
> 
> I'm working my way through them.
> 
> The first thing I notice is that xor_chan->desc_lock is taken
> inconsistently.  I.e. spin_lock_irqsave() in talitos_process_pending()
> and spin_lock_bh() everywhere else.  Have you run these patches with
> lockdep?
Thanks for your reply.
LOCKDEP is enabled as you suggested, there is not any info about "inconsistent lock state" displayed.
I don't know whether it's enough.

I'm confused about the attribute of DMA_INTERRUPT, my understanding is this interface is only used to trigger an interrupt (make sure all former operations are finished before switching to other channels), but fsl-dma will trigger an interrupt by "Programmed Error". I'm wondering whether other hardware are same with fsl-dma (the interrupt is a normal interrupt, but not an error) i.e. xscale-iop?
If other hardware also trigger an interrupt by an abnormal error, maybe my patch 2/8 should be reverted because it violates the rules of this attribute.

BTW, could you please reply in the patch if you have any comments. Thanks.

> 
> --
> Dan
> 
> 

ÿôèº{.nÇ+‰·Ÿ®‰­†+%ŠËÿ±éݶ\x17¥Šwÿº{.nÇ+‰·¥Š{±þG«éÿŠ{ayº\x1dʇڙë,j\a­¢f£¢·hšïêÿ‘êçz_è®\x03(­éšŽŠÝ¢j"ú\x1a¶^[m§ÿÿ¾\a«þG«éÿ¢¸?™¨è­Ú&£ø§~á¶iO•æ¬z·švØ^\x14\x04\x1a¶^[m§ÿÿÃ\fÿ¶ìÿ¢¸?–I¥