Re: [PATCH v4 2/7] system/physmem: poisoned memory discard on reboot

[William Roche <william.roche@oracle.com>]

On 1/14/25 15:00, David Hildenbrand wrote:
>> If we can get the current set of fixes integrated, I'll submit another
>> fix proposal to take the fd_offset into account in a second time. (Not
>> enlarging the current set)
>>
>> But here is what I'm thinking about. That we can discuss later if you 
>> want:
>>
>> @@ -3730,11 +3724,12 @@ int ram_block_discard_range(RAMBlock *rb,
>> uint64_t start, size_t length)
>>                }
>>
>>                ret = fallocate(rb->fd, FALLOC_FL_PUNCH_HOLE |
>> FALLOC_FL_KEEP_SIZE,
>> -                            start, length);
>> +                            start + rb->fd_offset, length);
>>                if (ret) {
>>                    ret = -errno;
>>                    error_report("%s: Failed to fallocate %s:%" PRIx64 "
>> +%zx (%d)",
>> -                             __func__, rb->idstr, start, length, ret);
>> +                             __func__, rb->idstr, start + rb->fd_offset,
>> +                            length, ret);
>>                    goto err;
>>                }
>>
>>
>> Or I can integrate that as an addition patch if you prefer.
> 
> Very good point! We missed to take fd_offset into account here.
> 
> Can you send that out as a separate fix?
> 
> Fixed: 4b870dc4d0c0 ("hostmem-file: add offset option"

Thanks to Peter Xu and to you for your reviews of my proposal for this 
separate fix that should be on track to be integrated soon.