Re: [Qemu-ppc] [PULL 41/47] ppc/pnv: populate device tree for IPMI BT devices

[Peter Maydell]

On 24 April 2017 at 02:59, David Gibson <address@hidden> wrote:
> From: Cédric Le Goater <address@hidden>
>
> When an ipmi-bt device [1] is defined on the ISA bus, we need to
> populate the device tree with the object properties. Such devices are
> created with the command line options :
>
>    -device ipmi-bmc-sim,id=bmc0 -device isa-ipmi-bt,bmc=bmc0,irq=10
>
> [1] https://lists.gnu.org/archive/html/qemu-devel/2015-11/msg03168.html
>
> Signed-off-by: Cédric Le Goater <address@hidden>
> Signed-off-by: David Gibson <address@hidden>
> ---
>  hw/ppc/pnv.c | 35 +++++++++++++++++++++++++++++++++++
>  1 file changed, 35 insertions(+)
>
> diff --git a/hw/ppc/pnv.c b/hw/ppc/pnv.c
> index dfa21e4..977e126 100644
> --- a/hw/ppc/pnv.c
> +++ b/hw/ppc/pnv.c
> @@ -354,6 +354,39 @@ static void powernv_populate_serial(ISADevice *d, void 
> *fdt, int lpc_off)
>      _FDT((fdt_setprop_string(fdt, node, "device_type", "serial")));
>  }
>
> +static void powernv_populate_ipmi_bt(ISADevice *d, void *fdt, int lpc_off)
> +{
> +    const char compatible[] = "bt\0ipmi-bt";
> +    uint32_t io_base;
> +    uint32_t io_regs[] = {
> +        cpu_to_be32(1),
> +        0, /* 'io_base' retrieved from the 'ioport' property of 
> 'isa-ipmi-bt' */
> +        cpu_to_be32(3)
> +    };
> +    uint32_t irq;
> +    char *name;
> +    int node;
> +
> +    io_base = object_property_get_int(OBJECT(d), "ioport", &error_fatal);
> +    io_regs[1] = cpu_to_be32(io_base);
> +
> +    irq = object_property_get_int(OBJECT(d), "irq", &error_fatal);
> +
> +    name = g_strdup_printf("address@hidden", qdev_fw_name(DEVICE(d)), 
> io_base);
> +    node = fdt_add_subnode(fdt, lpc_off, name);
> +    _FDT(node);
> +    g_free(name);
> +
> +    fdt_setprop(fdt, node, "reg", io_regs, sizeof(io_regs));
> +    fdt_setprop(fdt, node, "compatible", compatible, sizeof(compatible));
> +
> +    /* Mark it as reserved to avoid Linux trying to claim it */
> +    _FDT((fdt_setprop_string(fdt, node, "status", "reserved")));
> +    _FDT((fdt_setprop_cell(fdt, node, "interrupts", irq)));
> +    _FDT((fdt_setprop_cell(fdt, node, "interrupt-parent",
> +                           fdt_get_phandle(fdt, lpc_off))));
> +}

Hi -- Coverity points out (CID 1374832) that the fdt_setprop() calls
in this function to set "reg" and "compatible" don't check whether the
function returns an error. Is there any reason why these calls
haven't been wrapped with the barf-on-errors _FDT() macro like the
other fdt_setprop* calls in the function?

thanks
-- PMM