On Fri, 3 Mar 2017, Peter Maydell wrote:
On 3 March 2017 at 01:03, BALATON Zoltan <bala...@eik.bme.hu> wrote:
Adding vmstate saving is not in this patch because the state structure
will be changed in further patches, then another patch will add
vmstate descriptor after those changes.
Signed-off-by: BALATON Zoltan <bala...@eik.bme.hu>
+static void sm501_init(SM501State *s, DeviceState *dev, uint32_t base,
+ uint32_t local_mem_bytes)
+{
+ s->base = base;
+ s->local_mem_size_index = get_local_mem_size_index(local_mem_bytes);
+ SM501_DPRINTF("sm501 local mem size=%x. index=%d\n", get_local_mem_size(s),
+ s->local_mem_size_index);
+ if (get_local_mem_size(s) != local_mem_bytes) {
+ error_report("Warning: sm501 VRAM size adjusted to %" PRIu32,
+ get_local_mem_size(s));
+ }
Just noticed this. I think reporting the error upwards by
failing device realize is better than adjusting the value.
It's what we tend to do for other devices. Management tools
like libvirt prefer to get hard errors if there's a config
file error that means they misconfigure a device.
Making this change would need rebasing the whole series again an fixing
conflicts. Do you insist on this or can we live with this warning which is
closer to the original behaviour of this device?
