I have implented an I/O scheduler in Linux 2.6.But when I use fio to
test it,the kernel message often print some words,like this
"deadline_rb: forced dispatching is broken (nr_sorted=1),please report
this"
I have no idea about it,maybe my scheduler has some problems.But I
want to know why!
Thank y
Is In IO scheduler add_request and dispatch_request asynchronous or
synchronous??
In kernel,add request and dispatch request is asynchronous or synchronous?
" __make_request" first do add,then do dispatch,is right?
Does Add request and dispatch request have each other's thread???
Thank you!
-
-- Forwarded message --
From: 韩磊
Date: 2013/11/17
Subject: How to inital FIFO ticket spinlock in kernel 2.6.34.14?
To: Linux Kernel Mailing List
How to inital FIFO ticket spinlock in kernel 2.6.34.14?
Thank you!
--
To unsubscribe from this list: send the line "unsubscribe
What's that mean??
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
when we delete,add,search,amend the list_head,should we use spinlock
or rcu in case of conflicit to list_head???
Thank you!
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.or
-- Forwarded message --
From: 韩磊
Date: 2013/11/16
Subject: About deadline IO scheduler in kernel
To: Linux Kernel Mailing List
In deadline scheduler, in a FIFO list when a request insert it and
other request remove it in the same time,whether it make conflict???
Whether it
-- Forwarded message --
From: 韩磊
Date: 2013/11/16
Subject: About deadline IO scheduler in kernel
To: Linux Kernel Mailing List
In deadline scheduler, in a FIFO list when a request insert it and
other request remove it in the same time,whether it make conflict???
Whether it
In deadline scheduler, in a FIFO list when a request insert it and
other request remove it in the same time,whether it make conflict???
Whether it need lock??
Thank you!
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
-- Forwarded message --
From: 韩磊
Date: 2013/11/15
Subject: A problem about IO scheduler in kernel
To: Linux Kernel Mailing List
These days I was programming about IO scheduler called
"Simple_Deadline" in kernel.
"Simple_Deadline" is based on "deadline-
-- Forwarded message --
From: 韩磊
Date: 2013/11/15
Subject: A problem about IO scheduler in kernel
To: Linux Kernel Mailing List
These days I was programming about IO scheduler called
"Simple_Deadline" in kernel.
"Simple_Deadline" is based on "deadline-
These days I was programming about IO scheduler called
"Simple_Deadline" in kernel.
"Simple_Deadline" is based on "deadline-iosched".The algorithm is very
simple.It have two lists,one is read,the other is write.A request
enter lists based its weight which count accord to the request's
size,read or
I need a API in linux kernel about the disk's total count of
sector,not the sector size.
thank you!
2013/11/11 韩磊 :
> Thank you! I need disk's total count of sector,not the sector size.
>
> 2013/11/11 Mihai Donțu :
>> On Mon, 11 Nov 2013 20:31:14 +0800 韩磊 wrote:
&g
Thank you very much!
2013/11/12 Jeff Moyer :
> 韩磊 writes:
>
>> In IO scheduler level, whether all the read requests are synchronous?
>
> Generically, yes, I/O schedulers (and the whole block layer, in fact)
> consider READs synchronous:
>
> /*
> * We regard a req
-- Forwarded message --
From: 韩磊
Date: 2013/11/10
Subject: About read/write requests in IO scheduler
To: Linux Kernel Mailing List
In IO scheduler level, whether all the read requests are synchronous?
--
To unsubscribe from this list: send the line "unsubscribe linux-kerne
Thank you! I need disk's total count of sector,not the sector size.
2013/11/11 Mihai Donțu :
> On Mon, 11 Nov 2013 20:31:14 +0800 韩磊 wrote:
>> Can you tell me the API about the count of sector in linux kernel?
>>
>> which .h file and function?
>>
>
> L
Can you tell me the API about the count of sector in linux kernel?
which .h file and function?
Thank you!
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info
The disk sector
2013/11/10 韩磊 :
> I need the count of sector,but I don't know how to get it.
> Could you give me some favour?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More ma
In IO scheduler level, whether all the read requests are synchronous?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://ww
I need the count of sector,but I don't know how to get it.
Could you give me some favour?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read t
Thank you! I still have a long way to go in kernel.
2013/11/1 Andreas Mohr :
> Hi,
>
>> May I can use the ''double" as a basic variables?
>
> [see other replies]
>
>
> Look into terminus technicus "scaling math" as a replacement technique
> (in-kernel this is being used e.g. by clocksources/clo
May I can use the ''double" as a basic variables?
When I define the "double" as a function return type in linux kernel,the error
is "error: SSE register return with SSE disabled".
How to fix it??
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message
Thanks for your thought! It may not make much sense. Because I think
the probability of two bios have the same start sector and the
situation mentioned by Ming Lei is too low.
Thanks ,
Bonben
2013/10/25 Ming Lei :
> On Wed, Oct 23, 2013 at 6:59 PM, Jan Kara wrote:
>> On Wed 23-10-13 08
2013/10/23 Jan Kara :
> On Wed 23-10-13 08:47:44, 韩磊 wrote:
>> Nowadays,the IO schedulers in linux kernel have four types:
>>
>> deadline,noop,Anticiptory and CFQ.CFQ is the default scheduler.But CFQ is
>> not a good scheduler for SSD,dealine may be a good choice.
&
Nowadays,the IO schedulers in linux kernel have four types:
deadline,noop,Anticiptory and CFQ.CFQ is the default scheduler.But CFQ
is not a good scheduler for SSD,dealine may be a good choice.
When deadline runs,it has a mount of computation about merging and
sorting.Merge has three types: front_me
Nowadays,the IO schedulers in linux kernel have four types:
deadline,noop,Anticiptory and CFQ.CFQ is the default scheduler.But CFQ is
not a good scheduler for SSD,dealine may be a good choice.
When deadline runs,it has a mount of computation about merging and
sorting.Merge has three types: front
Nowadays,the IO schedulers in linux kernel have four types:
deadline,noop,Anticiptory and CFQ.CFQ is the default scheduler.But CFQ is
not a good scheduler for SSD,dealine may be a good choice.
When deadline runs,it has a mount of computation about merging and
sorting.Merge has three types: front
Nowadays,the IO schedulers in linux kernel have four types:
deadline,noop,Anticiptory and CFQ.CFQ is the default scheduler.But CFQ
is not a good scheduler for SSD,dealine may be a good choice.
When deadline runs,it has a mount of computation about merging and
sorting.Merge has three types: front_me
27 matches
Mail list logo
