linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: lorenzo.pieralisi@arm.com (Lorenzo Pieralisi)
To: linux-arm-kernel@lists.infradead.org
Subject: [RFT PATCH v2 29/42] ARM: PCI: Remove pci_fixup_irqs() call for bios32 host controllers
Date: Mon, 3 Jul 2017 11:26:19 +0100	[thread overview]
Message-ID: <20170703102619.GE19119@red-moon> (raw)
In-Reply-To: <20170702211950.GD18324@bhelgaas-glaptop.roam.corp.google.com>

On Sun, Jul 02, 2017 at 04:19:50PM -0500, Bjorn Helgaas wrote:
> On Sat, Jul 01, 2017 at 03:06:30PM +0100, Lorenzo Pieralisi wrote:
> > Hi Bjorn,
> > 
> > On Thu, Jun 08, 2017 at 03:13:29PM +0100, Lorenzo Pieralisi wrote:
> > > Legacy PCI host controllers (ie host controllers that set-up the PCI bus
> > > through the ARM pci_common_init() API) are currently relying on
> > > pci_fixup_irqs() to assign legacy PCI irqs to devices. This is not ideal
> > > in that pci_fixup_irqs() assign IRQs for all PCI devices present in a
> > > given system some of which may well be enabled by the time
> > > pci_fixup_irqs() is called (ie a system with multiple host controllers).
> > > With the introduction of struct pci_host_bridge.(*map_irq) pointer it is
> > > possible to assign IRQs for all devices originating from a PCI host
> > > bridge at probe time; this is implemented through pci_assign_irq() that
> > > relies on the struct pci_host_bridge.map_irq pointer to map IRQ for a
> > > given device.
> > > 
> > > The benefits this brings are twofold:
> > > 
> > > - the IRQ for a device is assigned once at probe time
> > > - the IRQ assignment works also for hotplugged devices
> > > 
> > > Remove pci_fixup_irqs() call from bios32 code and rely on
> > > pci_assign_irq() to carry out the IRQ mapping at device probe
> > > time.
> > > 
> > > The map_irq() and swizzle_irq() struct pci_host_bridge callbacks are
> > > set-up in the struct pci_host_bridge created in the bios32
> > > pcibios_init_hw() function and mach-* code paths (for PCI mach
> > > implementations that require a specific struct hw_pci.(*scan) function
> > > callback).
> > > 
> > > Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
> > > Cc: Jason Cooper <jason@lakedaemon.net>
> > > Cc: Bjorn Helgaas <bhelgaas@google.com>
> > > Cc: Russell King <linux@armlinux.org.uk>
> > > Cc: Andrew Lunn <andrew@lunn.ch>
> > > ---
> > >  arch/arm/kernel/bios32.c    |  4 ++--
> > >  arch/arm/mach-dove/pcie.c   | 18 ++++++++++--------
> > >  arch/arm/mach-iop13xx/pci.c |  2 ++
> > >  arch/arm/mach-orion5x/pci.c | 28 +++++++++++++++-------------
> > >  4 files changed, 29 insertions(+), 23 deletions(-)
> > 
> > As I mentioned to you, this patch is incomplete and needs an upgrade
> > (ie I misread how Orion and the gazillions board file that go with
> > it to map irqs - apologies but it is very hard to untangle).
> > 
> > I am inlining below an incremental patch to be folded in commit:
> > 
> > feaec00e757a ("ARM/PCI: Remove pci_fixup_irqs() call for bios32 host
> > controllers")
> > 
> > in your pci/irq-fixups branch, or if you prefer I can send you the
> > resulting squashed-in patch to replace this one, as you prefer.
> 
> Thanks, I folded this into pci/irq-fixups.  The resulting patch is
> merely this:

Yes that's what I expected, we remove pci_fixup_irqs() and add the
hooks into the host bridge structure that should cover all host
bridges initialized via pci_common_init_dev().

Thanks !
Lorenzo

> commit 16508469c0f3198a63923aaec08efa1c63f5fbcc
> Author: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
> Date:   Wed Jun 28 15:14:04 2017 -0500
> 
>     ARM/PCI: Remove pci_fixup_irqs() call for bios32 host controllers
>     
>     Legacy PCI host controllers (ie host controllers that set-up the PCI bus
>     through the ARM pci_common_init() API) are currently relying on
>     pci_fixup_irqs() to assign legacy PCI irqs to devices.  This is not ideal
>     in that pci_fixup_irqs() assigns IRQs for all PCI devices present in a given
>     system some of which may well be enabled by the time pci_fixup_irqs() is
>     called (ie a system with multiple host controllers).  With the introduction
>     of struct pci_host_bridge.(*map_irq) pointer it is possible to assign IRQs
>     for all devices originating from a PCI host bridge at probe time; this is
>     implemented through pci_assign_irq() that relies on the struct
>     pci_host_bridge.map_irq pointer to map IRQ for a given device.
>     
>     The benefits this brings are twofold:
>     
>       - the IRQ for a device is assigned once at probe time
>       - the IRQ assignment works also for hotplugged devices
>     
>     Remove pci_fixup_irqs() call from bios32 code and rely on pci_assign_irq()
>     to carry out the IRQ mapping at device probe time.
>     
>     The map_irq() and swizzle_irq() struct pci_host_bridge callbacks are set-up
>     in the struct pci_host_bridge created in the bios32 pcibios_init_hw()
>     function and mach-* code paths (for PCI mach implementations that require a
>     specific struct hw_pci.(*scan) function callback).
>     
>     Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
>     [bhelgaas: folded in fixes from Lorenzo:
>     http://lkml.kernel.org/r/20170701140629.GC8977 at red-moon]
>     Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
>     Cc: Jason Cooper <jason@lakedaemon.net>
>     Cc: Russell King <linux@armlinux.org.uk>
>     Cc: Andrew Lunn <andrew@lunn.ch>
> 
> diff --git a/arch/arm/kernel/bios32.c b/arch/arm/kernel/bios32.c
> index 4c7621ac389c..56dc1a3a33b4 100644
> --- a/arch/arm/kernel/bios32.c
> +++ b/arch/arm/kernel/bios32.c
> @@ -484,6 +484,9 @@ static void pcibios_init_hw(struct device *parent, struct hw_pci *hw,
>  				break;
>  			}
>  
> +			bridge->map_irq = pcibios_map_irq;
> +			bridge->swizzle_irq = pcibios_swizzle;
> +
>  			if (hw->scan)
>  				ret = hw->scan(nr, bridge);
>  			else {
> @@ -530,8 +533,6 @@ void pci_common_init_dev(struct device *parent, struct hw_pci *hw)
>  	if (hw->postinit)
>  		hw->postinit();
>  
> -	pci_fixup_irqs(pcibios_swizzle, pcibios_map_irq);
> -
>  	list_for_each_entry(sys, &head, node) {
>  		struct pci_bus *bus = sys->bus;
>  

  reply	other threads:[~2017-07-03 10:26 UTC|newest]

Thread overview: 73+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-06-08 14:13 [RFT PATCH v2 00/42] PCI: ARM/ARM64: remove pci_fixup_irqs() usage Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 01/42] PCI: xilinx-nwl: Remove nwl_pcie_enable_msi() unused bus parameter Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 02/42] drivers: pci: host: iproc: Convert link check to raw PCI config accessors Lorenzo Pieralisi
2017-06-08 15:56   ` Ray Jui
2017-06-08 16:36     ` Lorenzo Pieralisi
2017-06-11  4:12       ` Oza Oza
2017-06-12 16:13         ` Lorenzo Pieralisi
2017-06-12 17:40           ` Oza Oza
2017-06-12 18:52             ` Ray Jui
2017-06-13  8:22               ` Oza Oza
2017-06-13 17:18                 ` Ray Jui
2017-06-14 13:39                   ` Lorenzo Pieralisi
2017-06-21 14:39                     ` Oza Oza
2017-07-19 12:13                       ` Oza Oza
2017-07-19 17:48                         ` Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 03/42] drivers: pci: host: ftpci100: convert IRQ masking " Lorenzo Pieralisi
2017-06-16  9:05   ` Linus Walleij
2017-06-08 14:13 ` [RFT PATCH v2 04/42] PCI: Initialize bridge release function at bridge allocation Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 05/42] PCI: Add pci_free_host_bridge interface Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 06/42] PCI: Add devm_pci_alloc_host_bridge() interface Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 07/42] drivers: pci: host: ftpci100: Fix host bridge memory leakage Lorenzo Pieralisi
2017-06-16  9:02   ` Linus Walleij
2017-06-08 14:13 ` [RFT PATCH v2 08/42] drivers: pci: host: tegra: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 09/42] PCI: Introduce pci_scan_root_bus_bridge() Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 10/42] PCI: Make pci_register_host_bridge() PCI core internal Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 11/42] ARM: PCI: bios32: Convert PCI scan API to pci_scan_root_bus_bridge() Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 12/42] PCI: designware: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 13/42] PCI: aardvark: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 14/42] PCI: rcar: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 15/42] PCI: iproc: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 16/42] PCI: versatile: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 17/42] PCI: altera: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 18/42] PCI: xilinx: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 19/42] PCI: xgene: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 20/42] PCI: host-common: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 21/42] PCI: rockchip: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 22/42] PCI: xilinx-nwl: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 23/42] PCI: Remove pci_scan_root_bus_msi() Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 24/42] PCI: Build setup-irq.o on all arches Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 25/42] PCI: Add IRQ mapping function pointers to pci_host_bridge struct Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 26/42] PCI: Add pci_assign_irq() function and have pci_fixup_irqs() use it Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 27/42] OF/PCI: Update of_irq_parse_and_map_pci() comment Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 28/42] PCI: Add a call to pci_assign_irq() in pci_device_probe() Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 29/42] ARM: PCI: Remove pci_fixup_irqs() call for bios32 host controllers Lorenzo Pieralisi
2017-07-01 14:06   ` Lorenzo Pieralisi
2017-07-02 21:19     ` Bjorn Helgaas
2017-07-03 10:26       ` Lorenzo Pieralisi [this message]
2017-06-08 14:13 ` [RFT PATCH v2 30/42] PCI: tegra: Drop pci_fixup_irqs() Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 31/42] PCI: xilinx: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 32/42] PCI: rcar: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 33/42] PCI: iproc: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 34/42] PCI: designware-host: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 35/42] PCI: ftpci100: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 36/42] PCI: host-common: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 37/42] PCI: versatile: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 38/42] PCI: altera: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 39/42] PCI: xgene: Move to struct pci_host_bridge IRQ mapping functions Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 40/42] PCI: rockchip: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 41/42] PCI: xilinx-nwl: " Lorenzo Pieralisi
2017-06-08 14:13 ` [RFT PATCH v2 42/42] ARM64: PCI: Drop DT IRQ allocation from pcibios_alloc_irq() Lorenzo Pieralisi
2017-06-12 15:45 ` [RFT PATCH v2 00/42] PCI: ARM/ARM64: remove pci_fixup_irqs() usage Will Deacon
2017-06-12 16:20   ` Lorenzo Pieralisi
2017-06-12 23:58 ` Khuong Dinh
2017-06-19 23:12 ` Bjorn Helgaas
2017-06-20 14:44   ` Lorenzo Pieralisi
2017-06-21  8:39 ` Linus Walleij
2017-06-21  9:50   ` Lorenzo Pieralisi
2017-06-21 10:30   ` Lorenzo Pieralisi
2017-06-21 10:45     ` Lorenzo Pieralisi
2017-06-21 14:51       ` Linus Walleij
2017-06-21 15:14       ` Linus Walleij
2017-06-21 15:41         ` Lorenzo Pieralisi
2017-06-21 16:28           ` Linus Walleij

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=20170703102619.GE19119@red-moon \
    --to=lorenzo.pieralisi@arm.com \
    --cc=linux-arm-kernel@lists.infradead.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).