[GIT PULL] PCI updates for v3.14

[GIT PULL] PCI updates for v3.14

[Bjorn Helgaas]

Hi Linus,

This is a fix for an AGP regression exposed by e501b3d87f00 ("agp: Support
64-bit APBASE"), which we merged in v3.14-rc1.  We've warned about the
conflict between the GART and PCI resources and cleared out the PCI
resource for a long time, but after e501b3d87f00, we still *use* that
cleared-out PCI resource.  I think the GART resource is incorrect, so this
patch removes it.


The following changes since commit ac93ac7403493f8707b7734de9f40d5cb5db9045:

  PCI: Don't check resource_size() in pci_bus_alloc_resource() (2014-03-12 11:19:20 -0600)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git tags/pci-v3.14-fixes-3

for you to fetch changes up to 707d4eefbdb31f8e588277157056b0ce637d6c68:

  Revert "[PATCH] Insert GART region into resource map" (2014-03-18 14:26:12 -0600)

----------------------------------------------------------------
PCI updates for v3.14:

  Resource management
    - Revert "Insert GART region into resource map"

----------------------------------------------------------------
Bjorn Helgaas (1):
      Revert "[PATCH] Insert GART region into resource map"

 arch/x86/kernel/aperture_64.c | 20 +-------------------
 1 file changed, 1 insertion(+), 19 deletions(-)

[GIT PULL] PCI updates for v3.14

[Bjorn Helgaas]

Hi Linus,

This is a fix for an AGP regression exposed by e501b3d87f00 ("agp: Support
64-bit APBASE"), which we merged in v3.14-rc1.  We've warned about the
conflict between the GART and PCI resources and cleared out the PCI
resource for a long time, but after e501b3d87f00, we still *use* that
cleared-out PCI resource.  I think the GART resource is incorrect, so this
patch removes it.


The following changes since commit ac93ac7403493f8707b7734de9f40d5cb5db9045:

  PCI: Don't check resource_size() in pci_bus_alloc_resource() (2014-03-12 11:19:20 -0600)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git tags/pci-v3.14-fixes-3

for you to fetch changes up to 707d4eefbdb31f8e588277157056b0ce637d6c68:

  Revert "[PATCH] Insert GART region into resource map" (2014-03-18 14:26:12 -0600)

----------------------------------------------------------------
PCI updates for v3.14:

  Resource management
    - Revert "Insert GART region into resource map"

----------------------------------------------------------------
Bjorn Helgaas (1):
      Revert "[PATCH] Insert GART region into resource map"

 arch/x86/kernel/aperture_64.c | 20 +-------------------
 1 file changed, 1 insertion(+), 19 deletions(-)

[GIT PULL] PCI updates for v3.14

[Bjorn Helgaas]

Hi Linus,

These are two important regression fixes for bugs we've introduced so far
in v3.14.

One of the resource allocation changes from the merge window is broken for
32-bit kernels where we don't use _CRS for PCI host bridges (mostly
pre-2008 machines), so there's a fix for that.

The INTx enable change we put in after the merge window turned out to break
pciehp because we re-enable INTx on the hotplug bridge, which apparently
breaks MSI for future hotplug events.

The following changes since commit fa389e220254c69ffae0d403eac4146171062d08:

  Linux 3.14-rc6 (2014-03-09 19:41:57 -0700)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git tags/pci-v3.14-fixes-2

for you to fetch changes up to ac93ac7403493f8707b7734de9f40d5cb5db9045:

  PCI: Don't check resource_size() in pci_bus_alloc_resource() (2014-03-12 11:19:20 -0600)

----------------------------------------------------------------
PCI updates for v3.14:

  Resource management
    - Fix allocation from 0-4GB regions (Bjorn Helgaas)

  Miscellaneous
    - Enable INTx only when MSI/MSI-X not enabled (Bjorn Helgaas)

----------------------------------------------------------------
Bjorn Helgaas (2):
      PCI: Enable INTx in pci_reenable_device() only when MSI/MSI-X not enabled
      PCI: Don't check resource_size() in pci_bus_alloc_resource()

 drivers/pci/bus.c | 2 --
 drivers/pci/pci.c | 3 +++
 2 files changed, 3 insertions(+), 2 deletions(-)

[GIT PULL] PCI updates for v3.14

[Bjorn Helgaas]

Hi Linus,

The most interesting thing here is the change to enable INTx (by clearing
PCI_COMMAND_INTX_DISABLE) if the BIOS left INTx disabled.  Apparently the
Baytrail BIOS does this, which means EHCI doesn't work.

Also, fix an AHCI MSI regression and other issues with the recent MSI
changes. This also adds pci_enable_msi_exact() and pci_enable_msix_exact(),
which aren't regression fixes, but will keep us from touching drivers twice
(once to stop using the deprecated pci_enable_msi(), etc., and again to use
the *_exact() variants).

There's also a minor MVEBU fix.

Bjorn


The following changes since commit 38dbfb59d1175ef458d006556061adeaa8751b72:

  Linus 3.14-rc1 (2014-02-02 16:42:13 -0800)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git tags/pci-v3.14-fixes-1

for you to fetch changes up to fc40363b2140f5777c88e67814fac9327bf1ee68:

  ahci: Fix broken fallback to single MSI mode (2014-02-14 14:27:19 -0700)

----------------------------------------------------------------
PCI updates for v3.14:

  MSI
    - Fix AHCI single-MSI fallback (Alexander Gordeev)
    - Fix populate_msi_sysfs() error paths (Greg Kroah-Hartman)
    - Fix htmldocs problem (Masanari Iida)
    - Add pci_enable_msi_exact() and pci_enable_msix_exact() (Alexander Gordeev)
    - Update documentation (Alexander Gordeev)

  Miscellaneous
    - mvebu: expose device ID & revision via lspci (Andrew Lunn)
    - Enable INTx if the BIOS left them disabled (Bjorn Helgaas)

----------------------------------------------------------------
Alexander Gordeev (4):
      PCI/MSI: Add pci_enable_msi() documentation back
      PCI/MSI: Fix cut-and-paste errors in documentation
      PCI/MSI: Add pci_enable_msi_exact() and pci_enable_msix_exact()
      ahci: Fix broken fallback to single MSI mode

Andrew Lunn (1):
      PCI: mvebu: Use Device ID and revision from underlying endpoint

Bjorn Helgaas (1):
      PCI: Enable INTx if BIOS left them disabled

Greg Kroah-Hartman (2):
      PCI/MSI: Check kmalloc() return value, fix leak of name
      PCI/MSI: Fix leak of msi_attrs

Masanari Iida (1):
      PCI/MSI: Fix pci_msix_vec_count() htmldocs failure

 Documentation/PCI/MSI-HOWTO.txt | 119 ++++++++++++++++++++++++++++++++++++----
 drivers/ata/ahci.c              |   4 +-
 drivers/pci/host/pci-mvebu.c    |  11 +---
 drivers/pci/msi.c               |  10 +++-
 drivers/pci/pci.c               |  10 ++++
 include/linux/pci.h             |  20 +++++++
 6 files changed, 152 insertions(+), 22 deletions(-)