Re: [PATCH 0/8][RFC] IO latency/throughput fixes

[Geert Uytterhoeven]

On Tue, Apr 7, 2009 at 01:10, Linus Torvalds
<torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
> On Mon, 6 Apr 2009, Linus Torvalds wrote:
>> thing that we think people would be happiest with.
>>
>> I think "ordered" was a reasonable default, but that was at least partly
>> because _both_ ordered and writeback sucked (partly in different ways).
>>
>> I do think we could make it a config option.
>
> A patch _something_ like this.
>
> A few notes:
>
> Â- This is UNTESTED (of course)
>
> Â- If I did this right, this _only_ overrides the data mode if it's not
> Â explicitly specified on disk in the superblock mount options.
>
> IOW, if you have done a
>
> Â Â Â Âtune2fs -o journal_data_ordered
>
> then this will _not_ override that. Only in the absense of any explicit
> flags should this trigger and then make the choice be 'writeback'.
>
> And just to be _extra_ backwards compatible, if you really want the old
> behavior, and don't want to set the ordering flag explicitly, just answer
> 'y' to the EXT3_DEFAULTS_TO_ORDERED Kconfig question.
>
> What do people think? Anybody want to test?
>
> Â Â Â Â Â Â Â ÂLinus
>
> ---
> Âfs/ext3/Kconfig | Â 19 +++++++++++++++++++
> Âfs/ext3/super.c | Â Â8 +++++++-
> Â2 files changed, 26 insertions(+), 1 deletions(-)
>
> diff --git a/fs/ext3/Kconfig b/fs/ext3/Kconfig
> index 8e0cfe4..fb3c1a2 100644
> --- a/fs/ext3/Kconfig
> +++ b/fs/ext3/Kconfig
> @@ -28,6 +28,25 @@ config EXT3_FS
> Â Â Â Â ÂTo compile this file system support as a module, choose M here: the
> Â Â Â Â Âmodule will be called ext3.
>
> +config EXT3_DEFAULTS_TO_ORDERED
> + Â Â Â bool "Default to 'data=ordered' in ext3 (legacy option)"
> + Â Â Â depends on EXT3_FS
> + Â Â Â help
> + Â Â Â Â If a filesystem does not explicitly specify a data ordering
> + Â Â Â Â mode, and the journal capability allowed it, ext3 used to
> + Â Â Â Â historically default to 'data=ordered'.
> +
> + Â Â Â Â That was a rather unfortunate choice, because it leads to all
> + Â Â Â Â kinds of latency problems, and the 'data=writeback' mode is more
> + Â Â Â Â appropriate these days.
> +
> + Â Â Â Â You should probably always answer 'n' here, and if you really
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> + Â Â Â Â want to use 'data=ordered' mode, set it in the filesystem itself
> + Â Â Â Â with 'tune2fs -o journal_data_ordered'.
> +
> + Â Â Â Â But if you really want to enable the legacy default, you can do
> + Â Â Â Â so by answering 'y' to this question.
> +

So `allmodconfig' will enable it? Is that the right thing to do, or
should it be inverted?

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/