On 1/23/21 11:40 AM, Bin Meng wrote:
> From: Bin Meng <bin.m...@windriver.com>
>
> With all these fixes and improvements, there is no way for the
> VMStateDescription to keep backward compatibility. We will have
> to bump up version ids.
Unfortunately this breaks bisectability (think about downstream
distributions cherry-picking patches individually).
I don't think there is a problem increasing 2 -> 3 -> 4 -> 5
(Cc'ed David in case). Could you respin increasing the version
on each VMState change?
>
> The s->mode check in the post_load() hook is also updated.
>
> Signed-off-by: Bin Meng <bin.m...@windriver.com>
>
> ---
>
> Changes in v2:
> - new patch: bump up version ids of VMStateDescription
>
> hw/sd/ssi-sd.c | 11 ++++++++---
> 1 file changed, 8 insertions(+), 3 deletions(-)
>
> diff --git a/hw/sd/ssi-sd.c b/hw/sd/ssi-sd.c
> index ee4fbc3dfe..0c507f3ec5 100644
> --- a/hw/sd/ssi-sd.c
> +++ b/hw/sd/ssi-sd.c
> @@ -4,6 +4,11 @@
> * Copyright (c) 2007-2009 CodeSourcery.
> * Written by Paul Brook
> *
> + * Copyright (c) 2021 Wind River Systems, Inc.
> + * Improved by Bin Meng <bin.m...@windriver.com>
> + *
> + * Validated with U-Boot v2021.01 and Linux v5.10 mmc_spi driver
> + *
> * This code is licensed under the GNU GPL v2.
> *
> * Contributions after 2012-01-13 are licensed under the terms of the
> @@ -319,7 +324,7 @@ static int ssi_sd_post_load(void *opaque, int version_id)
> {
> ssi_sd_state *s = (ssi_sd_state *)opaque;
>
> - if (s->mode > SSI_SD_DATA_READ) {
> + if (s->mode > SSI_SD_SKIP_CRC16) {
> return -EINVAL;
> }
> if (s->mode == SSI_SD_CMDARG &&
> @@ -337,8 +342,8 @@ static int ssi_sd_post_load(void *opaque, int version_id)
>
> static const VMStateDescription vmstate_ssi_sd = {
> .name = "ssi_sd",
> - .version_id = 2,
> - .minimum_version_id = 2,
> + .version_id = 3,
> + .minimum_version_id = 3,
> .post_load = ssi_sd_post_load,
> .fields = (VMStateField []) {
> VMSTATE_UINT32(mode, ssi_sd_state),
>
