On 06/07/2012 10:46 PM, WeiFeng Liu wrote:
> Help two questions about update_global_block_rsv()
>
> update_global_block_rsv(fs_info) {
>
> ...
> block_rsv = &fs_info->global_block_rsv;
> sinfo = block_rsv->space_info;
>
> num_bytes = sinfo->bytes_used + sinfo->bytes_pinned +
> sinfo->bytes_reserved + sinfo->bytes_readonly +
> sinfo->bytes_may_use;
>
> if (sinfo->total_bytes > num_bytes) {
> num_bytes = sinfo->total_bytes - num_bytes;
> block_rsv->reserved += num_bytes;
> sinfo->bytes_may_use += num_bytes;
> }
> ...
> }
>
> 1. I don't quite understand the lines:
> block_rsv->reserved += num_bytes;
> sinfo->bytes_may_use += num_bytes;
>
> why aren't they like these?
> block_rsv->reserved = num_bytes;
> sinfo->bytes_may_use = num_bytes;
>
> assuming that nothing else is changed, and we just call
> update_global_block_rsv() for several times, then block_rsv->reserved will
> increase constantly, and so sinfo->bytes_may_use; can anyone tell me what it
> means? I mean if nothing else is changed the two variables should keep their
> values unchanged also, is that right? a bit puzzlement to me.
>
the following if-statement tells the truth:
if (block_rsv->reserved >= block_rsv->size) {
...
}
> 2. Is the meaning of sinfo->bytes_may_use that bytes which is available for use
> or that bytes which will likely/unlikely be used?
>
You need to understand how a block_rsv works,
both block_rsv and space_info are a kind of _container_:
Introducing metadata reseravtion contexts has two major advantages.
First, it makes metadata reservation more traceable.
Second, it can reclaim freed space and re-add them to the itself after transaction committed.
here is a basic model:
+--+--->size +--+--->size +--------------+
| | | | | |
| | | | | |
+--+--->reserved +--+--->reserved +--------------+-->may_used
| | | | | |
| | | | | |
| | | | | |
+--+ +--+ +--------------+
block_rsv + block_rsv + ... = space_info(flag=4)
A B |
|
+--> reserved meta bytes
Hope that this helps.
thanks,
liubo
> thanks
>
> WeiFeng Liu
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
