[PATCH chao/erofs-dev 1/2] staging: erofs: separate into init_once / always

Chao Yu chao at kernel.org
Mon Nov 5 02:27:34 AEDT 2018


Hi Xiang,

On 2018-11-4 18:04, Gao Xiang via Linux-erofs wrote:
> Hi Chao,
> 
> On 2018/11/2 18:26, Chao Yu wrote:
>> Hi Xiang,
>>
>> On 2018/10/31 10:39, Gao Xiang wrote:
>>> Hi Chao,
>>>
>>> Could you please update chao/erofs-dev branch based on the linus tree and add the following patch?
>>>
>>> =====patchset======
>>> staging: erofs: fix `trace_erofs_readpage' position    <-- trivial
>>> [v2] staging: erofs: fix race when the managed cache is enabled
>>> staging: erofs: fix refcount assertion in erofs_register_workgroup
>>>
>>> staging: erofs: atomic_cond_read_relaxed on ref-locked workgroup
>>> staging: erofs: fix `erofs_workgroup_{try_to_freeze, unfreeze}'
>>> staging: erofs: add a full barrier in erofs_workgroup_unfreeze
>>>
>>> staging: erofs: separate into init_once / always
>>> staging: erofs: locked before registering for all new workgroups

There is merge conflict when applying above two patches, could you rebase them
on last udpated erofs-dev branch? :)

Thanks,

>>>
>>> [v2 RESEND] staging: erofs: decompress asynchronously if PG_readahead page at first
>>> =====end patchset======
>>>
>>>
>>> and
>>>
>>> "staging: erofs: managed pages could be locked at the time of decompression"
>>> could be dropped now since I will refactor the z_erofs_vle_unzip
>>> (currently, A serious bug in it, I have fixed it internally).
>>>
>>> I am working on the reset stability patches reported by the internal beta test,
>>> expected to finish this week. :)
>> No problem, let me update them at this weekend...
> 
> Thanks...could take some time to update this?...
> I need find a code-base to work on :)
> 
> Thanks,
> Gao Xiang
> 
>>
>> Thanks,
>>


More information about the Linux-erofs mailing list