On 13.08.2013, at 13:09, Efimov Vasily wrote:
>
> Signed-off-by: Efimov Vasily <r...@ispras.ru>
Please provide a patch description :).
> ---
> hw/ppc/virtex_ml507.c | 13 ++++++++++---
> 1 file changed, 10 insertions(+), 3 deletions(-)
>
> diff --git a/hw/ppc/virtex_ml507.c b/hw/ppc/virtex_ml507.c
> index 08e77fb..a00f709 100644
> --- a/hw/ppc/virtex_ml507.c
> +++ b/hw/ppc/virtex_ml507.c
> @@ -141,11 +141,18 @@ static int xilinx_load_device_tree(hwaddr addr,
> {
> char *path;
> int fdt_size;
> - void *fdt;
> + void *fdt = 0;
This should be NULL. NULL doesn't have to be 0 according to C IIRC.
> int r;
> + const char *dtb_filename;
>
> - /* Try the local "ppc.dtb" override. */
> - fdt = load_device_tree("ppc.dtb", &fdt_size);
> + dtb_filename = qemu_opt_get(qemu_get_machine_opts(), "dtb");
> + if (dtb_filename) {
> + fdt = load_device_tree(dtb_filename, &fdt_size);
> + }
> + if (!fdt) {
> + /* Try the local "ppc.dtb" override. */
> + fdt = load_device_tree("ppc.dtb", &fdt_size);
> + }
Could you please just remove the ppc.dtb override option? It's superfluous once
we have proper -dtb support.
Edgar, any objections?
Also, you should abort if the user specified a dtb through -dtb and that file
could not get loaded.
Alex
> if (!fdt) {
> path = qemu_find_file(QEMU_FILE_TYPE_BIOS, BINARY_DEVICE_TREE_FILE);
> if (path) {
> --
> 1.7.10.4
>
