[PATCH v2 0/6] BCMA Cumulative patches from OpenWRT

[PATCH v2 0/6] BCMA Cumulative patches from OpenWRT

[Nathan Hintz]

Nathan Hintz (6):
  bcma: Find names of non BCM cores
  bcma: Move initialization of SPROM to prevent overwrite
  bcma: Account for variable PCI memory base/size
  bcma: reads/writes are always 4 bytes, so always map 4 bytes
  bcma: Add __devexit to bcma_host_pci_remove
  bcma: Add flush for BCMA_RESET_CTL write

 arch/mips/bcm47xx/setup.c      |    4 +++
 arch/mips/bcm47xx/sprom.c      |    2 -
 drivers/bcma/core.c            |    1 +
 drivers/bcma/driver_pci_host.c |   10 +++---
 drivers/bcma/host_pci.c        |    4 +-
 drivers/bcma/scan.c            |   54 +++++++++++++++++++++++++++++----------
 6 files changed, 52 insertions(+), 23 deletions(-)

-- 
1.7.7.6

Re: [PATCH v2 0/6] BCMA Cumulative patches from OpenWRT

[Arend van Spriel]

On 04/26/2012 06:23 AM, Nathan Hintz wrote:
> Nathan Hintz (6):
>    bcma: Find names of non BCM cores
>    bcma: Move initialization of SPROM to prevent overwrite
>    bcma: Account for variable PCI memory base/size
>    bcma: reads/writes are always 4 bytes, so always map 4 bytes
>    bcma: Add __devexit to bcma_host_pci_remove
>    bcma: Add flush for BCMA_RESET_CTL write
>
>   arch/mips/bcm47xx/setup.c      |    4 +++
>   arch/mips/bcm47xx/sprom.c      |    2 -
>   drivers/bcma/core.c            |    1 +
>   drivers/bcma/driver_pci_host.c |   10 +++---
>   drivers/bcma/host_pci.c        |    4 +-
>   drivers/bcma/scan.c            |   54 +++++++++++++++++++++++++++++----------
>   6 files changed, 52 insertions(+), 23 deletions(-)
>

None of these patches have any description. Some motivation for the 
changes would be appreciated like 'reported problem', 'new target 
support', etc.

Gr. AvS

Re: [PATCH v2 0/6] BCMA Cumulative patches from OpenWRT

[Kalle Valo]

"Arend van Spriel" <arend@broadcom.com> writes:

> On 04/26/2012 06:23 AM, Nathan Hintz wrote:
>> Nathan Hintz (6):
>>    bcma: Find names of non BCM cores
>>    bcma: Move initialization of SPROM to prevent overwrite
>>    bcma: Account for variable PCI memory base/size
>>    bcma: reads/writes are always 4 bytes, so always map 4 bytes
>>    bcma: Add __devexit to bcma_host_pci_remove
>>    bcma: Add flush for BCMA_RESET_CTL write
>>
>>   arch/mips/bcm47xx/setup.c      |    4 +++
>>   arch/mips/bcm47xx/sprom.c      |    2 -
>>   drivers/bcma/core.c            |    1 +
>>   drivers/bcma/driver_pci_host.c |   10 +++---
>>   drivers/bcma/host_pci.c        |    4 +-
>>   drivers/bcma/scan.c            |   54 +++++++++++++++++++++++++++++----------
>>   6 files changed, 52 insertions(+), 23 deletions(-)
>>
>
> None of these patches have any description. Some motivation for the
> changes would be appreciated like 'reported problem', 'new target
> support', etc.

I would say that it's required. Anyone can see the diff and see what was
changed, but nobody will figure out why the change was made unless it's
described in the commit log. So it's important to answer the question
"Why?".

-- 
Kalle Valo

Re: [PATCH v2 0/6] BCMA Cumulative patches from OpenWRT

[John W. Linville]

On Thu, Apr 26, 2012 at 02:50:54PM +0300, Kalle Valo wrote:
> "Arend van Spriel" <arend@broadcom.com> writes:
> 
> > On 04/26/2012 06:23 AM, Nathan Hintz wrote:
> >> Nathan Hintz (6):
> >>    bcma: Find names of non BCM cores
> >>    bcma: Move initialization of SPROM to prevent overwrite
> >>    bcma: Account for variable PCI memory base/size
> >>    bcma: reads/writes are always 4 bytes, so always map 4 bytes
> >>    bcma: Add __devexit to bcma_host_pci_remove
> >>    bcma: Add flush for BCMA_RESET_CTL write
> >>
> >>   arch/mips/bcm47xx/setup.c      |    4 +++
> >>   arch/mips/bcm47xx/sprom.c      |    2 -
> >>   drivers/bcma/core.c            |    1 +
> >>   drivers/bcma/driver_pci_host.c |   10 +++---
> >>   drivers/bcma/host_pci.c        |    4 +-
> >>   drivers/bcma/scan.c            |   54 +++++++++++++++++++++++++++++----------
> >>   6 files changed, 52 insertions(+), 23 deletions(-)
> >>
> >
> > None of these patches have any description. Some motivation for the
> > changes would be appreciated like 'reported problem', 'new target
> > support', etc.
> 
> I would say that it's required. Anyone can see the diff and see what was
> changed, but nobody will figure out why the change was made unless it's
> described in the commit log. So it's important to answer the question
> "Why?".

Will we be seeing this patchset reposted with better changelogs?

John
-- 
John W. Linville		Someday the world will need a hero, and you
linville@tuxdriver.com			might be all we have.  Be ready.

Re: [PATCH v2 0/6] BCMA Cumulative patches from OpenWRT

[Arend van Spriel]

On 05/02/2012 08:11 PM, John W. Linville wrote:
> On Thu, Apr 26, 2012 at 02:50:54PM +0300, Kalle Valo wrote:
>> "Arend van Spriel" <arend@broadcom.com> writes:
>>
>>> On 04/26/2012 06:23 AM, Nathan Hintz wrote:
>>>> Nathan Hintz (6):
>>>>    bcma: Find names of non BCM cores
>>>>    bcma: Move initialization of SPROM to prevent overwrite
>>>>    bcma: Account for variable PCI memory base/size
>>>>    bcma: reads/writes are always 4 bytes, so always map 4 bytes
>>>>    bcma: Add __devexit to bcma_host_pci_remove
>>>>    bcma: Add flush for BCMA_RESET_CTL write
>>>>
>>>>   arch/mips/bcm47xx/setup.c      |    4 +++
>>>>   arch/mips/bcm47xx/sprom.c      |    2 -
>>>>   drivers/bcma/core.c            |    1 +
>>>>   drivers/bcma/driver_pci_host.c |   10 +++---
>>>>   drivers/bcma/host_pci.c        |    4 +-
>>>>   drivers/bcma/scan.c            |   54 +++++++++++++++++++++++++++++----------
>>>>   6 files changed, 52 insertions(+), 23 deletions(-)
>>>>
>>>
>>> None of these patches have any description. Some motivation for the
>>> changes would be appreciated like 'reported problem', 'new target
>>> support', etc.
>>
>> I would say that it's required. Anyone can see the diff and see what was
>> changed, but nobody will figure out why the change was made unless it's
>> described in the commit log. So it's important to answer the question
>> "Why?".
> 
> Will we be seeing this patchset reposted with better changelogs?
> 
> John

Although not strongly stated by me I was certainly expecting a re-post.

Gr. AvS

Re: [PATCH v2 0/6] BCMA Cumulative patches from OpenWRT

[Nathan Hintz]

Hi:
On Wed, 2012-05-02 at 20:53 +0200, Arend van Spriel wrote:
> On 05/02/2012 08:11 PM, John W. Linville wrote:
> > On Thu, Apr 26, 2012 at 02:50:54PM +0300, Kalle Valo wrote:
> >> "Arend van Spriel" <arend@broadcom.com> writes:
> >>
> >>> On 04/26/2012 06:23 AM, Nathan Hintz wrote:
> >>>> Nathan Hintz (6):
> >>>>    bcma: Find names of non BCM cores
> >>>>    bcma: Move initialization of SPROM to prevent overwrite
> >>>>    bcma: Account for variable PCI memory base/size
> >>>>    bcma: reads/writes are always 4 bytes, so always map 4 bytes
> >>>>    bcma: Add __devexit to bcma_host_pci_remove
> >>>>    bcma: Add flush for BCMA_RESET_CTL write
> >>>>
> >>>>   arch/mips/bcm47xx/setup.c      |    4 +++
> >>>>   arch/mips/bcm47xx/sprom.c      |    2 -
> >>>>   drivers/bcma/core.c            |    1 +
> >>>>   drivers/bcma/driver_pci_host.c |   10 +++---
> >>>>   drivers/bcma/host_pci.c        |    4 +-
> >>>>   drivers/bcma/scan.c            |   54 +++++++++++++++++++++++++++++----------
> >>>>   6 files changed, 52 insertions(+), 23 deletions(-)
> >>>>
> >>>
> >>> None of these patches have any description. Some motivation for the
> >>> changes would be appreciated like 'reported problem', 'new target
> >>> support', etc.
> >>
> >> I would say that it's required. Anyone can see the diff and see what was
> >> changed, but nobody will figure out why the change was made unless it's
> >> described in the commit log. So it's important to answer the question
> >> "Why?".
> > 
> > Will we be seeing this patchset reposted with better changelogs?
> > 
> > John
> 
> Although not strongly stated by me I was certainly expecting a re-post.
> 
> Gr. AvS
> 
> 
I will repost; it will probably be a few days though.  I haven't used
GIT before other than cloning some repo's, so I have a learning curve to
overcome.

Any hints are welcome.

Thanks,
Nathan

Re: [PATCH v2 0/6] BCMA Cumulative patches from OpenWRT

[Arend van Spriel]

On 05/03/2012 07:04 AM, Nathan Hintz wrote:
> I will repost; it will probably be a few days though.  I haven't used
> GIT before other than cloning some repo's, so I have a learning curve to
> overcome.
> 
> Any hints are welcome.
> 
> Thanks,
> Nathan

Adding a description to a change is not a git specific thing, but a scm
best-practice. To easily add a description to each of your patches is by
doing an interactive rebase (details in [1]).

$ git rebase -i HEAD~6
* this will pop-up your favorite editor
* each line starts with word 'pick'
* replace it with letter 'r'
* close the editor

git will open the editor for each change and let you enter a
description. After that you can recreate your patches.

$ git format-patch --subject-prefix="PATCH v3" --cover-letter -s -M -6

I always give the -M to catch renames and -s to add signed-off-by.

Hope this helps.

Gr. AvS

[1] http://book.git-scm.com/4_interactive_rebasing.html