On 2/8/21 5:57 AM, Peter Maydell wrote:
>> void unlock_user(void *host_ptr, abi_ulong guest_addr, size_t len);
....
>> - if (len > 0) {
>> - memcpy(g2h_untagged(guest_addr), host_ptr, len);
>> + if (len != 0) {
>> + memcpy(host_ptr_conv, host_ptr, len);
>> }
>
> Why the change from checking >0 to checking !=0 ? I'd rather not
> have to go through and audit all the callsites to confirm none
> of them pass a "length-or-negative-errno" value here...
The comparison change should have happened with the type change to unsigned in
the previous patch.
r~
