From: Julien Grall <julien.grall@citrix.com>
To: Vijay Kilari <vijay.kilari@gmail.com>
Cc: Ian Campbell <ian.campbell@citrix.com>,
Stefano Stabellini <stefano.stabellini@eu.citrix.com>,
Prasun Kapoor <Prasun.Kapoor@caviumnetworks.com>,
Vijaya Kumar K <vijaya.kumar@caviumnetworks.com>,
Julien Grall <julien.grall@linaro.org>, Tim Deegan <tim@xen.org>,
"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>,
Stefano Stabellini <stefano.stabellini@citrix.com>,
manish.jaggi@caviumnetworks.com
Subject: Re: [RFC PATCH v2 13/22] xen/arm: its: Add virtual ITS command support
Date: Mon, 4 May 2015 14:44:46 +0100 [thread overview]
Message-ID: <554777CE.6060705@citrix.com> (raw)
In-Reply-To: <CALicx6uj_n91iUN5ZEC=WdqFF6PwAMS6k_t_aH1Y2hy=O4-9wQ@mail.gmail.com>
Hi Vijay,
On 04/05/2015 14:27, Vijay Kilari wrote:
> On Mon, May 4, 2015 at 6:34 PM, Julien Grall <julien.grall@citrix.com> wrote:
>>
>>
>> On 04/05/2015 13:58, Vijay Kilari wrote:
>>>
>>> On Thu, Apr 30, 2015 at 7:59 PM, Julien Grall <julien.grall@citrix.com>
>>> wrote:
>>>>
>>>> Hi,
>>>>
>>>> On 30/04/15 14:47, Stefano Stabellini wrote:
>>>>>>>>
>>>>>>>>
>>>>>>>> If the devid is not within this range, the ITS won't recognize the
>>>>>>>> value and
>>>>>>>> won't be able to send the interrupt.
>>>>>>>>
>>>>>>>> So this is clearly not the right value.
>>>>>>>
>>>>>>>
>>>>>>> Sure, in that case the maximum value allowed by GITS_TYPER.Devbits.
>>>>>>> Vijay, what is the value of GITS_TYPER.Devbits on your platform?
>>>>>>
>>>>>>
>>>>>> It is 21 bits
>>>>>
>>>>>
>>>>> I would imagine that 21 bits would be plenty to find an unused devid.
>>>>>
>>>>> Alternatively we could use an inexistent function of a real device, such
>>>>> as 00:00.1 (function 1 of the host bridge).
>>>>
>>>>
>>>> As discussed IRL, this idea sounds good to me.
>>>>
>>>> Although I would be happy with any other way which ensure the devid is
>>>> free.
>>>
>>>
>>> Has prototyped with 00.00.1 as device id. But I see that Dom0 boot is
>>> slow compared to polling mode. This could be because Dom0 has to keep
>>> trapping on creader to check if creader is updated or not.
>>
>>
>> How did you implement the interrupt mode? Could it be improve?
>
> 1) In physical ITS driver its_device is created with devID 00:00.1 with
> 256 MSI-x are reserved and is named as completion_dev, which is global.
That's a lot of MSI-x reserved... Can't you use only one per domain?
> 2) In Domain init,
> - one irq (called completion_irq) is allocated per domain for
> this device
So only 256 domain can run on your platform at the same time?
> and irq desc is allocated to this domain. This way we can get vITS
> context when interrupt is received.
> - An array of 32 requests(its_requests) is allocated which stores all
> the information about the ITS commands that are converted and written
> to physical ITS queue and each request info contains
> [CReader vITS] [CWriter vITS] [Physical Q Index] [Number of commands]
> [Completion irq] [ status ]
Why 32 requests?
Also some of the fields don't make much sense to me such as "Number of
Commands" , "Completion IRQ" and "status" I guess I will find out when
you will send the new series.
> 3) When vITS received ITS command. This command is converted to physical
> command and written to physical queue, INT command is appended with
> completion_irq and entry is made in its_requests.
> 4) On receiving completion_irq, vITS structure and its_requests
> info is parsed
> and Creader of vITS is upstated with [ Cwriter vITS ] stored
> for this request
> and this request is removed from its_requests.
You complicate the code by allowing 32 batch of command per domain
(looping is very slow). You should only allow one batch of command per
domain. When the batch is finished you can send another one.
> I am adding one INT per command. This can be improved to add one INT
> cmd for all
> the pending commands. Existing Linux driver sends 2 commands at a time.
You should not assume that other OS will send 2 commands at the same
time... It could be more or less.
Although, having a INT per command is rather slow. One INT command per
batch would improve the boot time.
Regards,
--
Julien Grall
next prev parent reply other threads:[~2015-05-04 13:44 UTC|newest]
Thread overview: 109+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-03-19 14:37 [RFC PATCH v2 00/22] xen/arm: Add ITS support vijay.kilari
2015-03-19 14:37 ` [RFC PATCH v2 01/22] add linked list apis vijay.kilari
2015-03-19 14:37 ` [RFC PATCH v2 02/22] Use linked list accessors for page_list helper function vijay.kilari
2015-03-19 14:37 ` [RFC PATCH v2 03/22] xen/arm: Add bitmap_find_next_zero_area " vijay.kilari
2015-03-20 13:35 ` Julien Grall
2015-03-19 14:37 ` [RFC PATCH v2 04/22] xen/arm: its: Import GICv3 ITS driver from linux vijay.kilari
2015-03-19 14:37 ` [RFC PATCH v2 05/22] xen/arm: gicv3: Refactor redistributor information vijay.kilari
2015-03-19 14:37 ` [RFC PATCH v2 06/22] xen/arm: its: Port ITS driver to xen vijay.kilari
2015-03-20 15:06 ` Julien Grall
2015-03-23 12:24 ` Vijay Kilari
2015-03-23 13:27 ` Julien Grall
2015-04-01 11:34 ` Ian Campbell
2015-04-02 8:25 ` Vijay Kilari
2015-04-02 9:25 ` Ian Campbell
2015-04-02 10:05 ` Vijay Kilari
2015-04-02 13:57 ` Julien Grall
2015-03-19 14:37 ` [RFC PATCH v2 07/22] xen/arm: its: Move ITS command encode helper functions vijay.kilari
2015-03-19 14:37 ` [RFC PATCH v2 08/22] xen/arm: its: Remove unused code in ITS driver vijay.kilari
2015-03-19 14:37 ` [RFC PATCH v2 09/22] xen/arm: its: Add helper functions to decode ITS Command vijay.kilari
2015-04-01 11:40 ` Ian Campbell
2015-05-11 14:14 ` Vijay Kilari
2015-05-11 14:25 ` Julien Grall
2015-05-11 14:25 ` Julien Grall
2015-05-11 14:36 ` Vijay Kilari
2015-05-11 22:06 ` Julien Grall
2015-03-19 14:37 ` [RFC PATCH v2 10/22] xen/arm: Add helper function to get domain page vijay.kilari
2015-03-20 16:39 ` Julien Grall
2015-03-19 14:37 ` [RFC PATCH v2 11/22] xen/arm: its: Move its_device structure to header file vijay.kilari
2015-03-19 14:37 ` [RFC PATCH v2 12/22] xen/arm: its: Update irq descriptor for LPIs support vijay.kilari
2015-03-20 16:44 ` Julien Grall
2015-03-30 14:32 ` Vijay Kilari
2015-03-30 15:29 ` Julien Grall
2015-03-19 14:38 ` [RFC PATCH v2 13/22] xen/arm: its: Add virtual ITS command support vijay.kilari
2015-03-21 0:28 ` Julien Grall
2015-03-23 15:52 ` Julien Grall
2015-03-24 11:48 ` Julien Grall
2015-03-30 15:02 ` Vijay Kilari
2015-03-30 15:47 ` Julien Grall
2015-04-01 11:46 ` Ian Campbell
2015-04-01 12:02 ` Julien Grall
2015-04-02 9:13 ` Ian Campbell
2015-04-02 11:06 ` Julien Grall
2015-04-02 11:18 ` Ian Campbell
2015-04-02 13:47 ` Julien Grall
2015-04-28 9:28 ` Vijay Kilari
2015-04-28 9:56 ` Stefano Stabellini
2015-04-28 10:35 ` Julien Grall
2015-04-28 11:36 ` Vijay Kilari
2015-04-28 16:15 ` Julien Grall
2015-04-29 1:44 ` Vijay Kilari
2015-04-29 11:56 ` Julien Grall
2015-04-29 12:12 ` Manish Jaggi
2015-04-29 12:21 ` Julien Grall
2015-04-29 12:33 ` Manish Jaggi
2015-04-29 13:01 ` Julien Grall
2015-04-29 13:08 ` Manish Jaggi
2015-04-29 13:16 ` Julien Grall
2015-04-29 13:35 ` Julien Grall
2015-04-29 16:26 ` Vijay Kilari
2015-04-29 16:30 ` Vijay Kilari
2015-04-29 18:04 ` Julien Grall
2015-04-30 10:02 ` Stefano Stabellini
2015-04-30 10:09 ` Julien Grall
2015-04-30 10:15 ` Stefano Stabellini
2015-04-30 10:20 ` Julien Grall
2015-04-30 10:50 ` Stefano Stabellini
2015-04-30 13:19 ` Vijay Kilari
2015-04-30 13:47 ` Stefano Stabellini
2015-04-30 14:29 ` Julien Grall
2015-05-04 12:58 ` Vijay Kilari
2015-05-04 13:04 ` Julien Grall
2015-05-04 13:27 ` Vijay Kilari
2015-05-04 13:44 ` Julien Grall [this message]
2015-05-04 13:54 ` Julien Grall
2015-05-04 15:19 ` Vijay Kilari
2015-05-04 17:00 ` Julien Grall
2015-05-05 10:28 ` Stefano Stabellini
2015-05-05 11:06 ` Vijay Kilari
2015-05-05 11:47 ` Julien Grall
2015-05-05 12:00 ` Vijay Kilari
2015-05-05 12:08 ` Julien Grall
2015-05-05 11:08 ` Julien Grall
2015-05-05 11:45 ` Vijay Kilari
2015-05-05 11:54 ` Stefano Stabellini
2015-05-05 10:39 ` Stefano Stabellini
2015-05-05 11:10 ` Julien Grall
2015-05-05 11:57 ` Stefano Stabellini
2015-05-05 12:03 ` Julien Grall
2015-03-19 14:38 ` [RFC PATCH v2 14/22] xen/arm: its: Add emulation of ITS control registers vijay.kilari
2015-03-24 17:12 ` Julien Grall
2015-03-19 14:38 ` [RFC PATCH v2 15/22] xen/arm: its: Add support to emulate GICR register for LPIs vijay.kilari
2015-03-27 15:46 ` Julien Grall
2015-03-19 14:38 ` [RFC PATCH v2 16/22] xen/arm: its: implement hw_irq_controller " vijay.kilari
2015-03-27 17:02 ` Julien Grall
2015-03-19 14:38 ` [RFC PATCH v2 17/22] xen/arm: its: Map ITS translation space vijay.kilari
2015-03-27 17:07 ` Julien Grall
2015-03-19 14:38 ` [RFC PATCH v2 18/22] xen/arm: its: Dynamic allocation of LPI descriptors vijay.kilari
2015-03-19 14:38 ` [RFC PATCH v2 19/22] xen/arm: its: Support ITS interrupt handling vijay.kilari
2015-03-19 14:38 ` [RFC PATCH v2 20/22] xen/arm: its: Generate ITS node for Dom0 vijay.kilari
2015-03-19 14:38 ` [RFC PATCH v2 21/22] xen/arm: its: Initialize virtual and physical ITS driver vijay.kilari
2015-03-19 14:38 ` [RFC PATCH v2 22/22] xen/arm: its: Generate ITS dt node for DomU vijay.kilari
2015-03-20 13:37 ` [RFC PATCH v2 00/22] xen/arm: Add ITS support Julien Grall
2015-03-20 16:23 ` Julien Grall
2015-03-23 12:37 ` Vijay Kilari
2015-03-23 13:11 ` Julien Grall
2015-03-23 15:18 ` Vijay Kilari
2015-03-23 15:30 ` Julien Grall
2015-03-23 16:09 ` Vijay Kilari
2015-03-23 16:18 ` Julien Grall
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=554777CE.6060705@citrix.com \
--to=julien.grall@citrix.com \
--cc=Prasun.Kapoor@caviumnetworks.com \
--cc=ian.campbell@citrix.com \
--cc=julien.grall@linaro.org \
--cc=manish.jaggi@caviumnetworks.com \
--cc=stefano.stabellini@citrix.com \
--cc=stefano.stabellini@eu.citrix.com \
--cc=tim@xen.org \
--cc=vijay.kilari@gmail.com \
--cc=vijaya.kumar@caviumnetworks.com \
--cc=xen-devel@lists.xen.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.