On Thu, Aug 03, 2006 at 04:03:19PM -0700, Nate Diller wrote:
>...
> Tested and benchmarked extensively on several platforms with several
> disk controllers and multiple brands of disk. Subsequently ported
> forward from 2.6.14, compile and boot tested on 2.6.18-rc1-mm2.
>...
> --- linux-2.6.18-rc1-mm2/block/Kconfig.iosched 2006-07-18
> 14:52:29.000000000 -0700
> +++ linux-dput/block/Kconfig.iosched 2006-08-02 17:40:01.000000000 -0700
>...
> +config IOSCHED_EL_SSTF
> + bool "Alternate Heuristic: Shortest Seek Time First" if
> IOSCHED_ELEVATOR
> + default n
> + ---help---
> + Elevator normally uses the C-SCAN one-way elevator algorithm,
> + which is useful in most situations to avoid queue congestion and
> + request starvation. In some cases, SSTF might be higher
> + performance, particularly with certain localized workloads.
> +
> + If you don't know that you want this, you don't.
If you want to make your scheduler tunable, this should be a run-time
option.
>...
> choice
> prompt "Default I/O scheduler"
> - default DEFAULT_CFQ
> + default DEFAULT_AS
> help
> Select the I/O scheduler which will be used by default for all
> block devices.
>
> config DEFAULT_AS
> - bool "Anticipatory" if IOSCHED_AS=y
> + bool "Anticipatory" if IOSCHED_AS
>
> config DEFAULT_DEADLINE
> - bool "Deadline" if IOSCHED_DEADLINE=y
> + bool "Deadline" if IOSCHED_DEADLINE
>
> config DEFAULT_CFQ
> - bool "CFQ" if IOSCHED_CFQ=y
> + bool "CFQ" if IOSCHED_CFQ
> +
> + config DEFAULT_ELEVATOR
> + bool "Elevator" if IOSCHED_ELEVATOR
>
> config DEFAULT_NOOP
> bool "No-op"
>...
It seems something went wrong when you forward ported your patch.
> --- linux-2.6.18-rc1-mm2/block/ll_rw_blk.c 2006-07-18
> 15:00:44.000000000 -0700
> +++ linux-dput/block/ll_rw_blk.c 2006-08-01 13:56:34.000000000 -0700
>...
> @@ -2951,7 +2953,7 @@ get_rq:
> blk_plug_device(q);
> add_request(q, req);
> out:
> - if (sync)
> +/* if (sync)*/
> __generic_unplug_device(q);
>
> spin_unlock_irq(q->queue_lock);
>...
Why?
cu
Adrian
--
Gentoo kernels are 42 times more popular than SUSE kernels among
KLive users (a service by SUSE contractor Andrea Arcangeli that
gathers data about kernels from many users worldwide).
There are three kinds of lies: Lies, Damn Lies, and Statistics.
Benjamin Disraeli
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
[Index of Archives]
[Kernel Newbies]
[Netfilter]
[Bugtraq]
[Photo]
[Stuff]
[Gimp]
[Yosemite News]
[MIPS Linux]
[ARM Linux]
[Linux Security]
[Linux RAID]
[Video 4 Linux]
[Linux for the blind]
[Linux Resources]