[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <d0aa29e9-54c6-bc8b-4827-0562b277265f@arm.com>
Date: Fri, 22 Mar 2019 16:26:29 +0000
From: Steven Price <steven.price@....com>
To: Kangjie Lu <kjlu@....edu>
Cc: Lorenzo Pieralisi <lorenzo.pieralisi@....com>,
linux-pci@...r.kernel.org, Michal Simek <michal.simek@...inx.com>,
linux-kernel@...r.kernel.org, pakki001@....edu,
Bjorn Helgaas <bhelgaas@...gle.com>,
linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH] pci: pcie-xilinx: fix a missing-check bug for
__get_free_pages
On 14/03/2019 05:58, Kangjie Lu wrote:
> In case __get_free_pages fail, the fix returns to avoid NULL
> pointer dereference.
>
> Signed-off-by: Kangjie Lu <kjlu@....edu>
> ---
> drivers/pci/controller/pcie-xilinx.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/pci/controller/pcie-xilinx.c b/drivers/pci/controller/pcie-xilinx.c
> index 9bd1a35cd5d8..b7083e995c45 100644
> --- a/drivers/pci/controller/pcie-xilinx.c
> +++ b/drivers/pci/controller/pcie-xilinx.c
> @@ -341,6 +341,9 @@ static void xilinx_pcie_enable_msi(struct xilinx_pcie_port *port)
> phys_addr_t msg_addr;
>
> port->msi_pages = __get_free_pages(GFP_KERNEL, 0);
> + if (unlikely(!port->msi_pages))
> + return;
Shouldn't the function return an error code for the calling function to
see? With this change the caller (xilinx_pcie_init_irq_domain) will
think that MSIs were enabled, but actually they weren't.
Steve
> +
> msg_addr = virt_to_phys((void *)port->msi_pages);
> pcie_write(port, 0x0, XILINX_PCIE_REG_MSIBASE1);
> pcie_write(port, msg_addr, XILINX_PCIE_REG_MSIBASE2);
>
Powered by blists - more mailing lists