* [ubi][wl] the W/R performance of my wear-leveling unit 2
[not found] <1215763483.2784.85.camel@localhost.localdomain>
@ 2008-07-11 8:04 ` xiaochuan-xu
2008-07-14 10:36 ` Artem Bityutskiy
2008-07-16 5:06 ` Artem Bityutskiy
2 siblings, 0 replies; 3+ messages in thread
From: xiaochuan-xu @ 2008-07-11 8:04 UTC (permalink / raw)
To: linux-mtd
Hi,
one other thing forgeted,
the threshold value is 4096 for current UBI, and 100 for mine
current gives: | mine
--------------------+------------
real 0m2.279s 0m2.213s
user 0m0.120s 0m0.176s
sys 0m0.712s 0m0.660s
real 0m2.345s 0m2.213s
user 0m0.160s 0m0.176s
sys 0m0.724s 0m0.660s
real 0m2.699s 0m2.188s
user 0m0.152s 0m0.124s
sys 0m0.680s 0m0.744s
real 0m2.429s 0m2.156s
user 0m0.148s 0m0.136s
sys 0m0.744s 0m0.704s
real 0m2.323s 0m2.176s
user 0m0.144s 0m0.196s
sys 0m0.736s 0m0.644s
real 0m2.675s 0m2.386s
user 0m0.148s 0m0.168s
sys 0m0.740s 0m0.704s
real 0m2.433s 0m2.363s
user 0m0.140s 0m0.132s
sys 0m0.772s 0m0.720s
real 0m2.282s 0m2.309s
user 0m0.152s 0m0.128s
sys 0m0.760s 0m0.736s
real 0m2.635s 0m2.240s
user 0m0.120s 0m0.136s
sys 0m0.732s 0m0.760s
real 0m2.252s 0m2.170s
user 0m0.140s 0m0.148s
sys 0m0.732s 0m0.684s
real 0m2.950s 0m2.328s
user 0m0.128s 0m0.148s
sys 0m0.772s 0m0.724s
real 0m2.300s 0m2.122s
user 0m0.140s 0m0.156s
sys 0m0.740s 0m0.664s
--
yours Sincerely,
xiaochuan-xu (许小川)
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [ubi][wl] the W/R performance of my wear-leveling unit 2
[not found] <1215763483.2784.85.camel@localhost.localdomain>
2008-07-11 8:04 ` [ubi][wl] the W/R performance of my wear-leveling unit 2 xiaochuan-xu
@ 2008-07-14 10:36 ` Artem Bityutskiy
2008-07-16 5:06 ` Artem Bityutskiy
2 siblings, 0 replies; 3+ messages in thread
From: Artem Bityutskiy @ 2008-07-14 10:36 UTC (permalink / raw)
To: xiaochuan-xu; +Cc: linux-mtd
Hi.
On Fri, 2008-07-11 at 16:04 +0800, xiaochuan-xu wrote:
> Hi,
>
> one other thing forgeted,
> the threshold value is 4096 for current UBI, and 100 for mine
>
>
> current gives: | mine
> --------------------+------------
> real 0m2.279s 0m2.213s
> user 0m0.120s 0m0.176s
> sys 0m0.712s 0m0.660s
Will you please send the new patch? Did you learn how to send
patches inlined? I sent you the description in one of the previous
mails.
--
Best regards,
Artem Bityutskiy (Битюцкий Артём)
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [ubi][wl] the W/R performance of my wear-leveling unit 2
[not found] <1215763483.2784.85.camel@localhost.localdomain>
2008-07-11 8:04 ` [ubi][wl] the W/R performance of my wear-leveling unit 2 xiaochuan-xu
2008-07-14 10:36 ` Artem Bityutskiy
@ 2008-07-16 5:06 ` Artem Bityutskiy
2 siblings, 0 replies; 3+ messages in thread
From: Artem Bityutskiy @ 2008-07-16 5:06 UTC (permalink / raw)
To: xiaochuan-xu; +Cc: linux-mtd
On Fri, 2008-07-11 at 16:04 +0800, xiaochuan-xu wrote:
> one other thing forgeted,
> the threshold value is 4096 for current UBI, and 100 for mine
Could you please first send a patch which only removes the protection
trees. It should not contain any other unnecessary change, just
protection tree removal.
Everything should work fine without the protection tree, because it is
just an optimization. The performance should drop I guess.
After you have succeeded providing a small and clean prot. tree removal
patch, we may continue with further WL re-working.
Please, test your patch as well. Run the UBI tests which you may find
in mtd-utils.git/tests/ubi-tests/.
Thanks.
--
Best regards,
Artem Bityutskiy (Битюцкий Артём)
^ permalink raw reply [flat|nested] 3+ messages in thread