Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Thu, 26 Mar 2020 10:28:18 -0500
From:   Bjorn Helgaas <helgaas@kernel.org>
To:     Srinath Mannam <srinath.mannam@broadcom.com>
Cc:     Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>,
        Florian Fainelli <f.fainelli@gmail.com>,
        Ray Jui <rjui@broadcom.com>,
        Andrew Murray <amurray@thegoodpenguin.co.uk>,
        bcm-kernel-feedback-list@broadcom.com, linux-pci@vger.kernel.org,
        linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org,
        Roman Bacik <roman.bacik@broadcom.com>
Subject: Re: [PATCH 2/3] PCI: iproc: fix invalidating PAXB address mapping
Message-ID: <20200326152818.GA7634@google.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1585206447-1363-3-git-send-email-srinath.mannam@broadcom.com>
User-Agent: Mutt/1.12.2 (2019-09-21)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

[fixed Andrew's email addr]

Change subject to match convention, e.g.,

  PCI: iproc: Invalidate correct PAXB inbound windows

On Thu, Mar 26, 2020 at 12:37:26PM +0530, Srinath Mannam wrote:
> From: Roman Bacik <roman.bacik@broadcom.com>
> 
> Second stage bootloader prior to Linux boot may use all inbound windows
> including IARR1/IMAP1. We need to ensure all previous configuration of
> inbound windows are invalidated during the initialization stage of the
> Linux iProc PCIe driver. Add fix to invalidate IARR1/IMAP1 because it was
> missed in previous patch.
> 
> Fixes: 9415743e4c8a ("PCI: iproc: Invalidate PAXB address mapping")
> Signed-off-by: Roman Bacik <roman.bacik@broadcom.com>
> ---
>  drivers/pci/controller/pcie-iproc.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/pci/controller/pcie-iproc.c b/drivers/pci/controller/pcie-iproc.c
> index 6972ca4..e7f0d58 100644
> --- a/drivers/pci/controller/pcie-iproc.c
> +++ b/drivers/pci/controller/pcie-iproc.c
> @@ -351,6 +351,8 @@ static const u16 iproc_pcie_reg_paxb_v2[IPROC_PCIE_MAX_NUM_REG] = {
>  	[IPROC_PCIE_OMAP3]		= 0xdf8,
>  	[IPROC_PCIE_IARR0]		= 0xd00,
>  	[IPROC_PCIE_IMAP0]		= 0xc00,
> +	[IPROC_PCIE_IARR1]		= 0xd08,
> +	[IPROC_PCIE_IMAP1]		= 0xd70,

Wow, this is a little too subtle for my taste.

9415743e4c8a added this loop:

  +   for (idx = 0; idx < ib->nr_regions; idx++) {
  +       iproc_pcie_write_reg(pcie,
  +                            MAP_REG(IPROC_PCIE_IARR0, idx), 0);

This patch doesn't change the loop or the limit, so I *guess*
previously we invalidated IPROC_PCIE_IARR0, IPROC_PCIE_IARR2, ...,
(skipping IPROC_PCIE_IARR1), and now we will invalidate
IPROC_PCIE_IARR0, IPROC_PCIE_IARR1, ... instead?

>  	[IPROC_PCIE_IARR2]		= 0xd10,
>  	[IPROC_PCIE_IMAP2]		= 0xcc0,
>  	[IPROC_PCIE_IARR3]		= 0xe00,