Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1468949 > unrolled thread

[PATCH v4 0/4] i2c: designware: improve performance for transfers

Started byLucas De Marchi <lucas.demarchi@intel.com>
First post2016-08-24 00:20 +0200
Last post2016-08-25 22:40 +0200
Articles 4 — 4 participants

Back to article view | Back to linux.kernel

Contents

  [PATCH v4 0/4] i2c: designware: improve performance for transfers Lucas De Marchi <lucas.demarchi@intel.com> - 2016-08-24 00:20 +0200
    Re: [PATCH v4 0/4] i2c: designware: improve performance for transfers Jarkko Nikula <jarkko.nikula@linux.intel.com> - 2016-08-24 13:00 +0200
    Re: [PATCH v4 0/4] i2c: designware: improve performance for transfers Christian Ruppert <christian.ruppert@alitech.com> - 2016-08-25 18:30 +0200
    Re: [PATCH v4 0/4] i2c: designware: improve performance for transfers Wolfram Sang <wsa@the-dreams.de> - 2016-08-25 22:40 +0200

#1468949 — [PATCH v4 0/4] i2c: designware: improve performance for transfers

FromLucas De Marchi <lucas.demarchi@intel.com>
Date2016-08-24 00:20 +0200
Subject[PATCH v4 0/4] i2c: designware: improve performance for transfers
Message-ID<s9sd3-1FT-3@gated-at.bofh.it>
Diff from v3:

    - Fix over 80chars in one place

    - Move check for adapter being able to dynamically update TAR to be done
      on probe time rather than init as requested by Jarkko

For the previous version, Christian had added:

Tested-by: Christian Ruppert <christian.ruppert@alitech.com>
on TB101 with Linux-4.7

And Jarkko added his Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
on patches 1 and 3 (now 4).

There's a new patch #2 as a preparatory work to move the check
mentioned above to i2c_dw_probe().

v3 of "i2c: designware: do not disable adapter after transfer". Differences
are:

    - Now there's a first patch that does not depend on IC_TAR being dynamically
      enabled/disabled: it just doesn't wait for the state change when not needed.

    - We added a patch that allows detecting if HW supports the dynamic TAR updates

    - In the last patch the bits were changed as suggested by Jarkko.

    - This is tested on BayTrail and CherryTrail, both of them returning true for
      "dynamically update TAR"

José Roberto de Souza (1):
  i2c: designware: wait for disable/enable only if necessary

Lucas De Marchi (3):
  i2c: designware: add common functions for locking
  i2c: designware: detect when dynamic tar update is possible
  i2c: designware: do not disable adapter after transfer

 drivers/i2c/busses/i2c-designware-core.c | 162 +++++++++++++++++++++----------
 drivers/i2c/busses/i2c-designware-core.h |   1 +
 2 files changed, 111 insertions(+), 52 deletions(-)

-- 
2.7.4

[toc] | [next] | [standalone]

#1469337

FromJarkko Nikula <jarkko.nikula@linux.intel.com>
Date2016-08-24 13:00 +0200
Message-ID<s9E4y-17j-41@gated-at.bofh.it>
In reply to#1468949
On 08/24/2016 01:18 AM, Lucas De Marchi wrote:
> Diff from v3:
>
>     - Fix over 80chars in one place
>
>     - Move check for adapter being able to dynamically update TAR to be done
>       on probe time rather than init as requested by Jarkko
>
> For the previous version, Christian had added:
>
> Tested-by: Christian Ruppert <christian.ruppert@alitech.com>
> on TB101 with Linux-4.7
>
> And Jarkko added his Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
> on patches 1 and 3 (now 4).
>
Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>

[toc] | [prev] | [next] | [standalone]

#1470256

FromChristian Ruppert <christian.ruppert@alitech.com>
Date2016-08-25 18:30 +0200
Message-ID<sa5Hr-3mY-3@gated-at.bofh.it>
In reply to#1468949
On 24.08.2016 00:18, Lucas De Marchi wrote:
> Diff from v3:
> 
>     - Fix over 80chars in one place
> 
>     - Move check for adapter being able to dynamically update TAR to be done
>       on probe time rather than init as requested by Jarkko
> 
> For the previous version, Christian had added:
> 
> Tested-by: Christian Ruppert <christian.ruppert@alitech.com>
> on TB101 with Linux-4.7

Tested-by: Christian Ruppert <christian.ruppert@alitech.com>
Re-tested this iteration in the same configuration (TB101, Linux-4.7).
Still works perfectly.

> And Jarkko added his Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
> on patches 1 and 3 (now 4).
> 
> There's a new patch #2 as a preparatory work to move the check
> mentioned above to i2c_dw_probe().
> 
> v3 of "i2c: designware: do not disable adapter after transfer". Differences
> are:
> 
>     - Now there's a first patch that does not depend on IC_TAR being dynamically
>       enabled/disabled: it just doesn't wait for the state change when not needed.
> 
>     - We added a patch that allows detecting if HW supports the dynamic TAR updates
> 
>     - In the last patch the bits were changed as suggested by Jarkko.
> 
>     - This is tested on BayTrail and CherryTrail, both of them returning true for
>       "dynamically update TAR"
> 
> José Roberto de Souza (1):
>   i2c: designware: wait for disable/enable only if necessary
> 
> Lucas De Marchi (3):
>   i2c: designware: add common functions for locking
>   i2c: designware: detect when dynamic tar update is possible
>   i2c: designware: do not disable adapter after transfer
> 
>  drivers/i2c/busses/i2c-designware-core.c | 162 +++++++++++++++++++++----------
>  drivers/i2c/busses/i2c-designware-core.h |   1 +
>  2 files changed, 111 insertions(+), 52 deletions(-)
>

[toc] | [prev] | [next] | [standalone]

#1470403

FromWolfram Sang <wsa@the-dreams.de>
Date2016-08-25 22:40 +0200
Message-ID<sa9Bn-5Uq-5@gated-at.bofh.it>
In reply to#1468949

[Multipart message — attachments visible in raw view] — view raw

On Tue, Aug 23, 2016 at 07:18:52PM -0300, Lucas De Marchi wrote:
> Diff from v3:
> 
>     - Fix over 80chars in one place
> 
>     - Move check for adapter being able to dynamically update TAR to be done
>       on probe time rather than init as requested by Jarkko
> 
> For the previous version, Christian had added:
> 
> Tested-by: Christian Ruppert <christian.ruppert@alitech.com>
> on TB101 with Linux-4.7
> 
> And Jarkko added his Acked-by: Jarkko Nikula <jarkko.nikula@linux.intel.com>
> on patches 1 and 3 (now 4).
> 
> There's a new patch #2 as a preparatory work to move the check
> mentioned above to i2c_dw_probe().
> 
> v3 of "i2c: designware: do not disable adapter after transfer". Differences
> are:
> 
>     - Now there's a first patch that does not depend on IC_TAR being dynamically
>       enabled/disabled: it just doesn't wait for the state change when not needed.
> 
>     - We added a patch that allows detecting if HW supports the dynamic TAR updates
> 
>     - In the last patch the bits were changed as suggested by Jarkko.
> 
>     - This is tested on BayTrail and CherryTrail, both of them returning true for
>       "dynamically update TAR"
> 
> José Roberto de Souza (1):
>   i2c: designware: wait for disable/enable only if necessary
> 
> Lucas De Marchi (3):
>   i2c: designware: add common functions for locking
>   i2c: designware: detect when dynamic tar update is possible
>   i2c: designware: do not disable adapter after transfer

Applied to for-next, thanks! And thanks to all reviewers and testers,
much appreciated!

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web