linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Alexander Gordeev <agordeev@redhat.com>
To: linux-kernel@vger.kernel.org
Cc: Alexander Gordeev <agordeev@redhat.com>,
	Bjorn Helgaas <bhelgaas@google.com>,
	Michael Ellerman <michael@ellerman.id.au>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Tejun Heo <tj@kernel.org>,
	Ben Hutchings <bhutchings@solarflare.com>,
	David Laight <David.Laight@ACULAB.COM>,
	Mark Lord <kernel@start.ca>, "H. Peter Anvin" <hpa@zytor.com>,
	linux-pci@vger.kernel.org
Subject: [PATCH RFC v2 00/29] Introduce pcim_enable_msi*() family helpers
Date: Fri, 18 Oct 2013 19:12:00 +0200	[thread overview]
Message-ID: <cover.1382103786.git.agordeev@redhat.com> (raw)

This series is against "next" branch in Bjorn's repo:
git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci.git

Currently many device drivers need contiguously call functions
pci_enable_msix() for MSI-X or pci_enable_msi_block() for MSI
in a loop until success or failure. This update generalizes
this usage pattern and introduces pcim_enable_msi*() family
helpers.

As result, device drivers do not have to deal with tri-state
return values from pci_enable_msix() and pci_enable_msi_block()
functions directly and expected to have more clearer and straight
code.

So i.e. the request loop described in the documentation...

	int foo_driver_enable_msix(struct foo_adapter *adapter,
				   int nvec)
	{
		while (nvec >= FOO_DRIVER_MINIMUM_NVEC) {
			rc = pci_enable_msix(adapter->pdev,
					     adapter->msix_entries,
					     nvec);
			if (rc > 0)
				nvec = rc;
			else
				return rc;
		}

		return -ENOSPC;
	}

...would turn into a single helper call....

	rc = pcim_enable_msix_range(adapter->pdev,
				    adapter->msix_entries,
				    nvec,
				    FOO_DRIVER_MINIMUM_NVEC);


Device drivers with more specific requirements (i.e. a number of
MSI-Xs which is a multiple of a certain number within a specified
range) would still need to implement the loop using the two old
functions.

It is fair to say the proposal in v1 ("Re-design MSI/MSI-X interrupts
enablement pattern")is scrapped. Thus, I do not clarify what are the
changes since v1  - the whole v2 series could be
considered as completely new.

Device driver updates (patches 12-29) are provided to clarify the
change and expected to get reviews in follow-up post(s) when/if
patch 12 ("PCI/MSI: Introduce pcim_enable_msi*() family helpers")
is accepted.

Patches 1,2	- ACK'ed tweaks for s390 architecture
Patches 3,4	- fixes for PowerPC pSeries platform
Patches 5-12	- fixes, tweaks and changes of the generic MSI code
Patches 12-29	- example updates of few device drivers

The tree could be found in "pci-next-msi-v2" branch in repo:
https://github.com/a-gordeev/linux.git

Alexander Gordeev (29):
  PCI/MSI/s390: Fix single MSI only check
  PCI/MSI/s390: Remove superfluous check of MSI type
  PCI/MSI/pSeries: Fix wrong error code reporting
  PCI/MSI/pSeries: Make quota traversing and requesting race-safe
  PCI/MSI: Fix return value when populate_msi_sysfs() failed
  PCI/MSI: Get rid of useless count of msi_desc leftovers
  PCI/MSI: Return -ENOSYS for unimplemented interfaces, not -1
  PCI/MSI: Make pci_enable_msix() 'nvec' argument unsigned int
  PCI/MSI: Factor out pci_get_msi_cap() interface
  PCI/MSI: Get rid of pci_enable_msi_block_auto() interface
  PCI/MSI: Convert pci_msix_table_size() to a public interface
  PCI/MSI: Introduce pcim_enable_msi*() family helpers
  ipr: Do not call pci_disable_msi/msix() if pci_enable_msi/msix()
    failed
  ipr: Make use of pcim_enable_msi/msix() interfaces
  ixgbe: Make use of pcim_enable_msix_range() interface
  ixgbevf: Make use of pcim_enable_msix_range() interface
  megaraid: Make use of pcim_enable_msix() interface
  ntb: Fix missed call to pci_enable_msix()
  ntb: Make use of pcim_enable_msix/msix_exact() interfaces
  qib: Make use of pcim_enable_msix() and pci_msix_table_size()
    interfaces
  qla2xxx: Make use of pcim_enable_msix_range() interface
  qlge: Get rid of an redundant assignment
  qlge: Make use of pcim_enable_msix() interface
  tg3: Make use of pcim_enable_msix() interface
  vmxnet3: Return -EINVAL if number of requested MSI-Xs is not enough
  vmxnet3: Fixup a weird loop exit
  vmxnet3: Return -ENOSPC when not enough MSI-X vectors available
  vmxnet3: Limit number of rx queues to 1 if per-queue MSI-Xs failed
  vmxnet3: Make use of pcim_enable_msix_range() interface

 Documentation/PCI/MSI-HOWTO.txt                   |  169 ++++++++++++++++++---
 arch/powerpc/platforms/pseries/msi.c              |   26 +++-
 arch/s390/pci/pci.c                               |    4 +-
 drivers/ata/ahci.c                                |   56 +++++---
 drivers/infiniband/hw/qib/qib_pcie.c              |   48 +++---
 drivers/net/ethernet/broadcom/tg3.c               |    6 +-
 drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c      |   16 +--
 drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c |   33 ++---
 drivers/net/ethernet/qlogic/qlge/qlge_main.c      |   16 +--
 drivers/net/vmxnet3/vmxnet3_drv.c                 |   58 +++-----
 drivers/ntb/ntb_hw.c                              |   34 ++---
 drivers/pci/msi.c                                 |  129 +++++++++-------
 drivers/pci/pcie/portdrv_core.c                   |    5 +-
 drivers/scsi/ipr.c                                |   51 +++----
 drivers/scsi/megaraid/megaraid_sas_base.c         |   16 +--
 drivers/scsi/qla2xxx/qla_isr.c                    |   28 ++--
 include/linux/pci.h                               |   79 +++++++++--
 17 files changed, 466 insertions(+), 308 deletions(-)

-- 
1.7.7.6


             reply	other threads:[~2013-10-19  7:04 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-10-18 17:12 Alexander Gordeev [this message]
2013-10-18 17:12 ` [PATCH RFC v2 01/29] PCI/MSI/s390: Fix single MSI only check Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 02/29] PCI/MSI/s390: Remove superfluous check of MSI type Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 03/29] PCI/MSI/pSeries: Fix wrong error code reporting Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 04/29] PCI/MSI/pSeries: Make quota traversing and requesting race-safe Alexander Gordeev
2013-11-20 15:43   ` Tejun Heo
2013-10-18 17:12 ` [PATCH RFC v2 05/29] PCI/MSI: Fix return value when populate_msi_sysfs() failed Alexander Gordeev
2013-11-20 15:50   ` Tejun Heo
2013-10-18 17:12 ` [PATCH RFC v2 06/29] PCI/MSI: Get rid of useless count of msi_desc leftovers Alexander Gordeev
2013-11-20 16:07   ` Tejun Heo
2013-11-22 18:27     ` Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 07/29] PCI/MSI: Return -ENOSYS for unimplemented interfaces, not -1 Alexander Gordeev
2013-11-20 16:07   ` Tejun Heo
2013-10-18 17:12 ` [PATCH RFC v2 08/29] PCI/MSI: Make pci_enable_msix() 'nvec' argument unsigned int Alexander Gordeev
2013-11-20 16:14   ` Tejun Heo
2013-11-22 18:34     ` Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 09/29] PCI/MSI: Factor out pci_get_msi_cap() interface Alexander Gordeev
2013-11-20 16:18   ` Tejun Heo
2013-11-25 13:49     ` Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 10/29] PCI/MSI: Get rid of pci_enable_msi_block_auto() interface Alexander Gordeev
2013-11-20 16:21   ` Tejun Heo
2013-10-18 17:12 ` [PATCH RFC v2 11/29] PCI/MSI: Convert pci_msix_table_size() to a public interface Alexander Gordeev
2013-11-20 16:23   ` Tejun Heo
2013-11-22 18:38     ` Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 12/29] PCI/MSI: Introduce pcim_enable_msi*() family helpers Alexander Gordeev
2013-10-24 10:51   ` Tejun Heo
2013-10-24 10:57     ` David Laight
2013-10-24 11:01       ` Tejun Heo
2013-10-24 11:41       ` Alexander Gordeev
2013-10-25  2:22         ` Mark Lord
2013-10-25  9:10           ` David Laight
2013-10-25 10:01             ` Alexander Gordeev
2013-10-25 14:52               ` intel e1000e pci driver issue on RT Linux kernel Karicheri, Muralidharan
2013-10-25 14:59                 ` Karicheri, Muralidharan
2013-10-27 22:27               ` [PATCH RFC v2 12/29] PCI/MSI: Introduce pcim_enable_msi*() family helpers Michael Ellerman
2013-10-28 16:30               ` Mark Lord
2013-10-24 14:31     ` Alexander Gordeev
2013-10-25  2:23       ` Mark Lord
2013-11-04  8:12     ` Alexander Gordeev
2013-11-20 17:15   ` Tejun Heo
2013-11-22 18:44     ` Alexander Gordeev
2013-11-22 18:44       ` Tejun Heo
2013-11-22 18:54         ` Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 13/29] ipr: Do not call pci_disable_msi/msix() if pci_enable_msi/msix() failed Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 14/29] ipr: Make use of pcim_enable_msi/msix() interfaces Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 15/29] ixgbe: Make use of pcim_enable_msix_range() interface Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 16/29] ixgbevf: " Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 17/29] megaraid: Make use of pcim_enable_msix() interface Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 18/29] ntb: Fix missed call to pci_enable_msix() Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 19/29] ntb: Make use of pcim_enable_msix/msix_exact() interfaces Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 20/29] qib: Make use of pcim_enable_msix() and pci_msix_table_size() interfaces Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 21/29] qla2xxx: Make use of pcim_enable_msix_range() interface Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 22/29] qlge: Get rid of an redundant assignment Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 23/29] qlge: Make use of pcim_enable_msix() interface Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 24/29] tg3: " Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 25/29] vmxnet3: Return -EINVAL if number of requested MSI-Xs is not enough Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 26/29] vmxnet3: Fixup a weird loop exit Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 27/29] vmxnet3: Return -ENOSPC when not enough MSI-X vectors available Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 28/29] vmxnet3: Limit number of rx queues to 1 if per-queue MSI-Xs failed Alexander Gordeev
2013-10-18 17:12 ` [PATCH RFC v2 29/29] vmxnet3: Make use of pcim_enable_msix_range() interface Alexander Gordeev
2013-10-21  8:51   ` David Laight
2013-10-21 10:19     ` Alexander Gordeev
2013-11-11 19:56 ` [PATCH RFC v2 00/29] Introduce pcim_enable_msi*() family helpers Alexander Gordeev
2013-11-20 17:16 ` Tejun Heo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=cover.1382103786.git.agordeev@redhat.com \
    --to=agordeev@redhat.com \
    --cc=David.Laight@ACULAB.COM \
    --cc=benh@kernel.crashing.org \
    --cc=bhelgaas@google.com \
    --cc=bhutchings@solarflare.com \
    --cc=hpa@zytor.com \
    --cc=kernel@start.ca \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=michael@ellerman.id.au \
    --cc=tj@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).