Patchwork [GIT,PULL] PCI updates for 3.5 merge window

login
register
mail settings
Submitter Bjorn Helgaas
Date May 21, 2012, 5:05 p.m.
Message ID <20120521170539.GA13635@google.com>
Download mbox
Permalink /patch/160396/
State Not Applicable
Headers show

Pull-request

git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git next

Comments

Bjorn Helgaas - May 21, 2012, 5:05 p.m.
Hi Linus,

Please pull the signed tag "pci-for-3.5" to get these PCI changes for the
v3.5 merge window.  I'm a newbie, so don't hesitate to help me improve my
pull requests :)

The changes include:
  - Host bridge cleanups from Yinghai
  - Disable Bus Master bit on PCI device shutdown (kexec-related)
  - Stratus ftServer fix
  - pci_dev_reset() locking fix
  - IvyBridge graphics erratum workaround

Bjorn


The following changes since commit 66f75a5d028beaf67c931435fdc3e7823125730c:

  Linux 3.4-rc4 (2012-04-21 14:47:52 -0700)

are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git next

Bjorn Helgaas (6):
      PCI: work around Stratus ftServer broken PCIe hierarchy
      Merge branch 'topic/yinghai-hostbridge-cleanup' into next
      MAINTAINERS: update PCI git tree and patchwork
      Merge branch 'topic/stratus' into next
      x86/PCI: only check for spinlock being held in SMP kernels
      microblaze/PCI: fix "io_offset undeclared" error

Chunhe Lan (1):
      PCI: fix uninitialized variable 'cap_mask'

Khalid Aziz (1):
      PCI: disable Bus Master on PCI device shutdown

Konrad Rzeszutek Wilk (1):
      PCI: move mutex locking out of pci_dev_reset function

Wei Yang (1):
      resources: add resource_overlaps()

Xudong Hao (1):
      PCI: work around IvyBridge internal graphics FLR erratum

Yinghai Lu (12):
      PCI: move host bridge-related code to host-bridge.c
      x86/PCI: fix memleak with get_current_resources()
      PCI: rename pci_host_bridge() to find_pci_root_bridge()
      PCI: add generic device into pci_host_bridge struct
      PCI: add host bridge release support
      x86/PCI: refactor get_current_resources()
      x86/PCI: add host bridge resource release for _CRS path
      x86/PCI: embed name into pci_root_info struct
      x86/PCI: embed pci_sysdata into pci_root_info on ACPI path
      x86/PCI: dynamically allocate pci_root_info for native host bridge drivers
      x86/PCI: merge pcibios_scan_root() and pci_scan_bus_on_node()
      x86/PCI: fix unused variable warning in amd_bus.c

 Documentation/kernel-parameters.txt |    3 +
 MAINTAINERS                         |   10 +--
 arch/microblaze/pci/pci-common.c    |    1 +
 arch/x86/pci/acpi.c                 |  128 ++++++++++++++++-------------
 arch/x86/pci/amd_bus.c              |   91 +++++++++------------
 arch/x86/pci/broadcom_bus.c         |   12 +--
 arch/x86/pci/bus_numa.c             |   69 +++++++++++-----
 arch/x86/pci/bus_numa.h             |   18 +++--
 arch/x86/pci/common.c               |   43 +++++-----
 arch/x86/pci/i386.c                 |    2 +-
 drivers/pci/Makefile                |    2 +-
 drivers/pci/host-bridge.c           |   96 ++++++++++++++++++++++
 drivers/pci/pci-driver.c            |    6 ++
 drivers/pci/pci.c                   |   30 +++++---
 drivers/pci/pcie/portdrv_core.c     |    2 +-
 drivers/pci/probe.c                 |  154 ++++++++++-------------------------
 drivers/pci/quirks.c                |   58 +++++++++++++
 include/asm-generic/pci-bridge.h    |    6 ++
 include/linux/ioport.h              |    7 ++
 include/linux/pci.h                 |    9 ++-
 20 files changed, 442 insertions(+), 305 deletions(-)
 create mode 100644 drivers/pci/host-bridge.c
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Linus Torvalds - May 21, 2012, 11:26 p.m.
On Mon, May 21, 2012 at 10:05 AM, Bjorn Helgaas <bhelgaas@google.com> wrote:
>
> Please pull the signed tag "pci-for-3.5" to get these PCI changes for the
> v3.5 merge window.  I'm a newbie, so don't hesitate to help me improve my
> pull requests :)

Looks fine, except please actually mention the tag itself on the repo
pull line. IOW, not like this:

>  git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git next

but simply

  git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git tags/pci-for-3.5

otherwise it's much too easy for my to just cut-and-paste the line you
supplied without noticing that there was a tag mentioned earlier..

              Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Tony Luck - May 23, 2012, 4:41 p.m.
On Mon, May 21, 2012 at 4:26 PM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
> Looks fine, except please actually mention the tag itself on the repo
> pull line. IOW, not like this:
>
>>  git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git next
>
> but simply
>
>  git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git tags/pci-for-3.5

git request-pull seems to look for branches before tags when generating
this text.  It just did exactly the same thing for me (since the tag I created
also pointed to my "next" branch).

This would seem to be the common case for people following the rules
and sending you stuff in the merge window that they have had cooking
in linux-next.  Can "request-pull" be smarter - or is their some magic I should
be using on the "git request-pull" command line that will tell it that I really
mean the tag and not the branch?

Or is "smarter" already in a newer version of git - I have:

$ git --version
git version 1.7.9.5

-Tony
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Linus Torvalds - May 23, 2012, 4:52 p.m.
On Wed, May 23, 2012 at 9:41 AM, Tony Luck <tony.luck@gmail.com> wrote:
>
> Or is "smarter" already in a newer version of git - I have:
>
> $ git --version
> git version 1.7.9.5

I think you want 1.7.10 for anything to do with tagged messages.

You can do everything with 1.7.9 (it can still handle the tags
correctly), but it doesn't DTRT by default: you have to explicitly
mention the "tag" part (ie you need to use "tags/xyz" instead of just
"xyz"), and git pull-request doesn't realize that pulling a signed tag
does more than just pulling the commit it points to etc etc. So with
1.7.9 you can make things work, but you would end up to work around
the fact that the whole signed tag pulling wasn't really a real
feature before that.

So 1.7.10 is when pulling a signed tag was really introduced: it's
just that earlier versions can still *create* those tags (so the only
one who *had* to have 1.7.10 was the person doing the pull).

I would strongly recommend just doing

   git clone git://github.com/gitster/git
   cd git
   make
   make install

as just a normal user (*NOT* as root), which will install git in your
~/bin directory. Then, just make sure that you have ~/bin in your PATH
before /usr/bin, and you're all set. No need to mess with or worry
about distro git binaries - leave them alone, but get the best and the
brightest for your own use.

                     Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Tony Luck - May 23, 2012, 5:35 p.m.
On Wed, May 23, 2012 at 9:52 AM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
> I would strongly recommend just doing
>
>   git clone git://github.com/gitster/git
>   cd git
>   make
>   make install

My distro needed some "curl" and "expat" bits added ... but that worked
and I now have shiny new:
$ git --version
git version 1.7.10.2.552.gaa3bb87

which does indeed have a smarter git request-pull.

Thanks

-Tony
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html