Re: jffs2: Excess summary entries

[Wei Fang <fangwei1@huawei.com>]

Hi Thomas,

On 2015/11/6 2:36, Thomas.Betker@rohde-schwarz.com wrote:
> We ran into a problem with jffs2 where the filesystem became unusable 
> after some specific MTD failures; summary was enabled, write buffering was 
> disabled:

[snip...]

> Here are the solutions I have considered so far:
> 
> The minimal solution would be to check *retlen == 0 in 
> jffs2_flash_direct_writev() and jffs2_flash_direct_write() before running 
> jffs2_sum_add_kvec().
> 
>         if (jffs2_sum_active() && *retlen) {
>                 ...
>                 res = jffs2_sum_add_kvec(...)
>                 ...
>         }
> 
> The general failure case, though, is (ret != 0 || *retlen != len), where 
> 'ret' is the return code of mtd_writev(), and 'len' is the data size to be 
> written. When write buffering is enabled, jffs2_flash_writev() in wbuf.c 
> skips the summary entry in this case; perhaps we should do this in 
> writev.c as well?
> 
>         if (jffs2_sum_active() && !ret && *retlen == len) {
>                 ...
>                 res = jffs2_sum_add_kvec(...)
>                 ...
>         }

I prefer:
	if (jffs2_sum_active() && *retlen == len) {
		...
		res = jffs2_sum_add_kvec(...)
		...
	}

In the case that part of node has been written to flash, this whole node
will be marked as dirty node, only in memory, not marked without
JFFS2_NODE_ACCURATE on flash.

If the summary is stored when *retlen != len, there are two cases:

* In most case, another write with the same node info performs
  successfully later, the node written partially before will be marked
  as obsolete node when scan, and we won't read from it
* This node is the newest node about this region, it will be treated
  as normal node when scan, and we may read the data already corrupted.
  Yet it won't break any rules of JFFS2 and lead to a muddle

The node written partially will be treated as normal node in a full scan
routine too, so I think we should mark this node as dirty on flash in
the case that *retlen != len.

Thanks,
Wei

> I ran some quick tests, simulating write failures, and it seems that 
> adding the summary entry doesn't harm when *retlen != 0 [so the minimal 
> solution would suffice]. This is because the calling function will reserve 
> the node space, marking it as dirty, and there is no confusion about 
> unchecked space.
>
> On the other hand, running the same quick tests _without_ adding the 
> summary entry didn't seem to harm either [so the general solution would 
> work as well]. It is entirely possible that I have overlooked something, 
> though.
> 
> Any opinions on that? When in doubt, I would provide a patch for the 
> minimal solution, changing as little as possible. However, it may make 
> sense to go for the general solution to be consistent with write 
> buffering.
> 
> Best regards,
> Thomas Betker
> 
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
> 
>