linux-kernel - Re: [PATCH v9 2/2] PCI: xilinx-cpm: Add Versal CPM Root Port driver

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20200805220326.GA550962@bjorn-Precision-5520>
Date:   Wed, 5 Aug 2020 17:03:26 -0500
From:   Bjorn Helgaas <helgaas@...nel.org>
To:     Lorenzo Pieralisi <lorenzo.pieralisi@....com>
Cc:     Bharat Kumar Gogada <bharat.kumar.gogada@...inx.com>,
        linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org,
        bhelgaas@...gle.com, robh@...nel.org, maz@...nel.org
Subject: Re: [PATCH v9 2/2] PCI: xilinx-cpm: Add Versal CPM Root Port driver

On Wed, Aug 05, 2020 at 10:39:28PM +0100, Lorenzo Pieralisi wrote:
> On Wed, Aug 05, 2020 at 03:43:58PM -0500, Bjorn Helgaas wrote:
> > On Tue, Jun 16, 2020 at 06:26:54PM +0530, Bharat Kumar Gogada wrote:
> > > - Add support for Versal CPM as Root Port.
> > > - The Versal ACAP devices include CCIX-PCIe Module (CPM). The integrated
> > >   block for CPM along with the integrated bridge can function
> > >   as PCIe Root Port.
> > > - Bridge error and legacy interrupts in Versal CPM are handled using
> > >   Versal CPM specific interrupt line.
> > 
> > > +static void xilinx_cpm_pcie_init_port(struct xilinx_cpm_pcie_port *port)
> > > +{
> > > +	if (cpm_pcie_link_up(port))
> > > +		dev_info(port->dev, "PCIe Link is UP\n");
> > > +	else
> > > +		dev_info(port->dev, "PCIe Link is DOWN\n");
> > > +
> > > +	/* Disable all interrupts */
> > > +	pcie_write(port, ~XILINX_CPM_PCIE_IDR_ALL_MASK,
> > > +		   XILINX_CPM_PCIE_REG_IMR);
> > > +
> > > +	/* Clear pending interrupts */
> > > +	pcie_write(port, pcie_read(port, XILINX_CPM_PCIE_REG_IDR) &
> > > +		   XILINX_CPM_PCIE_IMR_ALL_MASK,
> > > +		   XILINX_CPM_PCIE_REG_IDR);
> > > +
> > > +	/*
> > > +	 * XILINX_CPM_PCIE_MISC_IR_ENABLE register is mapped to
> > > +	 * CPM SLCR block.
> > > +	 */
> > > +	writel(XILINX_CPM_PCIE_MISC_IR_LOCAL,
> > > +	       port->cpm_base + XILINX_CPM_PCIE_MISC_IR_ENABLE);
> > > +	/* Enable the Bridge enable bit */
> > > +	pcie_write(port, pcie_read(port, XILINX_CPM_PCIE_REG_RPSC) |
> > > +		   XILINX_CPM_PCIE_REG_RPSC_BEN,
> > > +		   XILINX_CPM_PCIE_REG_RPSC);
> > > +}
> > > +
> > > +/**
> > > + * xilinx_cpm_pcie_parse_dt - Parse Device tree
> > > + * @port: PCIe port information
> > > + * @bus_range: Bus resource
> > > + *
> > > + * Return: '0' on success and error value on failure
> > > + */
> > > +static int xilinx_cpm_pcie_parse_dt(struct xilinx_cpm_pcie_port *port,
> > > +				    struct resource *bus_range)
> > > +{
> > > +	struct device *dev = port->dev;
> > > +	struct platform_device *pdev = to_platform_device(dev);
> > > +	struct resource *res;
> > > +
> > > +	port->cpm_base = devm_platform_ioremap_resource_byname(pdev,
> > > +							       "cpm_slcr");
> > > +	if (IS_ERR(port->cpm_base))
> > > +		return PTR_ERR(port->cpm_base);
> > > +
> > > +	res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "cfg");
> > > +	if (!res)
> > > +		return -ENXIO;
> > > +
> > > +	port->cfg = pci_ecam_create(dev, res, bus_range,
> > > +				    &pci_generic_ecam_ops);
> > 
> > Aren't we passing an uninitialized pointer (bus_range) here?  This
> > looks broken to me.
> > 
> > The kernelci build warns about it:
> > https://kernelci.org/build/next/branch/master/kernel/next-20200805/
> > 
> >   /scratch/linux/drivers/pci/controller/pcie-xilinx-cpm.c:557:39: warning: variable 'bus_range' is uninitialized when used here [-Wuninitialized]
> > 
> > I'm dropping this for now.  I can't believe this actually works.
> 
> It is caused by my rebase to fix -next after the rework in pci/misc
> (I had to drop the call to pci_parse_request_of_pci_ranges()).
> 
> I will look into this tomorrow if Rob does not beat me to it.
> 
> Apologies, it is a new driver that was based on an interface
> that is being reworked, for good reasons, in pci/misc.

Oh, yep, I think I see what happened.  I'll try to fix this in hopes
of making linux-next tonight.

Bjorn