Re: [PATCH] doc: add description to dirtytime_expire_seconds

[Yang Shi <yang.shi@linux.alibaba.com>]

ping


Ted,


Any comment is appreciated.


Regards,

Yang



On 5/30/18 4:56 PM, Yang Shi wrote:
> commit 1efff914afac8a965ad63817ecf8861a927c2ace ("fs: add
> dirtytime_expire_seconds sysctl") introduced dirtytime_expire_seconds
> knob, but there is not description about it in
> Documentation/sysctl/vm.txt.
>
> Add the description for it.
>
> Cc: Theodore Ts'o <tytso@mit.edu>
> Signed-off-by: Yang Shi <yang.shi@linux.alibaba.com>
> ---
> I didn't dig into the old review discussion about why the description
> was not added at the first place. I'm supposed every knob under /proc/sys
> should have a brief description.
>
>   Documentation/sysctl/vm.txt | 11 +++++++++++
>   1 file changed, 11 insertions(+)
>
> diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
> index 17256f2..f4f4f9c 100644
> --- a/Documentation/sysctl/vm.txt
> +++ b/Documentation/sysctl/vm.txt
> @@ -27,6 +27,7 @@ Currently, these files are in /proc/sys/vm:
>   - dirty_bytes
>   - dirty_expire_centisecs
>   - dirty_ratio
> +- dirtytime_expire_seconds
>   - dirty_writeback_centisecs
>   - drop_caches
>   - extfrag_threshold
> @@ -178,6 +179,16 @@ The total available memory is not equal to total system memory.
>   
>   ==============================================================
>   
> +dirtytime_expire_seconds
> +
> +When a lazytime inode is constantly having its pages dirtied, it with an
> +updated timestamp will never get chance to be written out.  This tunable
> +is used to define when dirty inode is old enough to be eligible for
> +writeback by the kernel flusher threads. And, it is also used as the
> +interval to wakeup dirtytime_writeback thread. It is expressed in seconds.
> +
> +==============================================================
> +
>   dirty_writeback_centisecs
>   
>   The kernel flusher threads will periodically wake up and write `old' data