Hi Peter, On 3/17/21 8:09 PM, Peter Maydell wrote:
On Wed, 17 Mar 2021 at 04:44, Gavin Shan <gs...@redhat.com> wrote:There is a added clock to trace buad rate change since v5.2.0 by commit aac63e0e6ea3 ("hw/char/pl011: add a clock input"). The added clock causes migration failure. For example, migration from v5.2.0 to v5.1.0 can fail with the following error messages: qemu-system-aarch64: error while loading state for instance 0x0 \ of device 'pl011' qemu-system-aarch64: load of migration failed: No such file or \ directory This fixes the issue by reporting the baud rate change at post load time so that the clock won't be migrated by sub-section to avoid the migration failure. Cc: qemu-sta...@nongnu.org Fixes: aac63e0e6ea3 ("hw/char/pl011: add a clock input") Signed-off-by: Gavin Shan <gs...@redhat.com> --- hw/char/pl011.c | 22 +++++++++------------- 1 file changed, 9 insertions(+), 13 deletions(-) diff --git a/hw/char/pl011.c b/hw/char/pl011.c index c5621a195f..401bd28536 100644 --- a/hw/char/pl011.c +++ b/hw/char/pl011.c @@ -322,20 +322,20 @@ static const MemoryRegionOps pl011_ops = { .endianness = DEVICE_NATIVE_ENDIAN, }; -static const VMStateDescription vmstate_pl011_clock = { - .name = "pl011/clock", - .version_id = 1, - .minimum_version_id = 1, - .fields = (VMStateField[]) { - VMSTATE_CLOCK(clk, PL011State), - VMSTATE_END_OF_LIST() - } -}; +static int pl011_post_load(void *opaque, int version_id) +{ + PL011State *s = PL011(opaque); + + pl011_trace_baudrate_change(s); + + return 0; +} static const VMStateDescription vmstate_pl011 = { .name = "pl011", .version_id = 2, .minimum_version_id = 2, + .post_load = pl011_post_load, .fields = (VMStateField[]) { VMSTATE_UINT32(readbuff, PL011State), VMSTATE_UINT32(flags, PL011State), @@ -355,10 +355,6 @@ static const VMStateDescription vmstate_pl011 = { VMSTATE_INT32(read_trigger, PL011State), VMSTATE_END_OF_LIST() }, - .subsections = (const VMStateDescription * []) { - &vmstate_pl011_clock, - NULL - } };Doesn't dropping the subsection break migration compat ?
It's why this patch needs to be backported to stable branches. In that way, we won't have migration compatible issue. Thanks, Gavin
