From: 嵌入式工程师 <ericvic@163.com>
To: Gilles Chanteperdrix <gilles.chanteperdrix@xenomai.org>,
"xenomai@xenomai.org" <xenomai@xenomai.org>
Subject: [Xenomai] On mx6q specifically, it seems the framebuffer is a latency killer
Date: Tue, 25 Mar 2014 16:41:39 +0800 (CST) [thread overview]
Message-ID: <453fa613.d9b0.144f866f11d.Coremail.ericvic@163.com> (raw)
I run qt on my imx6q board and patch with 2.6.2.1 , when i use the latency without qt and over load mode the cpu status is Mem: 48856K used, 852880K free, 0K shrd, 1444K buff, 27540K cached
CPU0: 100% usr 0.0% sys 0.0% nic 0.0% idle 0.0% io 0.0% irq 0.0% sirq
CPU1: 100% usr 0.0% sys 0.0% nic 0.0% idle 0.0% io 0.0% irq 0.0% sirq
CPU2: 99.0% usr 0.9% sys 0.0% nic 0.0% idle 0.0% io 0.0% irq 0.0% sirq
CPU3: 100% usr 0.0% sys 0.0% nic 0.0% idle 0.0% io 0.0% irq 0.0% sirq
the latency
RTH|----lat min|----lat avg|----lat max|-overrun|---msw|---lat best|--lat worst
RTD| 0.032| 0.247| 3.241| 0| 0| -0.077| 9.492
RTD| 0.032| 0.234| 2.763| 0| 0| -0.077| 9.492
RTD| 0.026| 0.239| 3.171| 0| 0| -0.077| 9.492
RTD| 0.028| 0.241| 2.793| 0| 0| -0.077| 9.492
RTD| 0.024| 0.241| 2.811| 0| 0| -0.077| 9.492
RTD| 0.034| 0.238| 2.793| 0| 0| -0.077| 9.492
RTD| 0.027| 0.237| 2.780| 0| 0| -0.077| 9.492
RTD| 0.027| 0.242| 2.748| 0| 0| -0.077| 9.492
RTD| 0.023| 0.266| 3.326| 0| 0| -0.077| 9.492
RTD| 0.033| 0.249| 3.551| 0| 0| -0.077| 9.492
RTD| 0.041| 0.249| 2.973| 0| 0| -0.077| 9.492
RTD| 0.041| 0.250| 2.926| 0| 0| -0.077| 9.492
RTD| 0.039| 0.255| 2.745| 0| 0| -0.077| 9.492
RTD| 0.030| 0.247| 2.631| 0| 0| -0.077| 9.492
RTD| 0.036| 0.258| 2.693| 0| 0| -0.077| 9.492
RTD| 0.032| 0.236| 2.829| 0| 0| -0.077| 9.492
RTD| 0.038| 0.251| 2.695| 0| 0| -0.077| 9.492
RTD| 0.034| 0.257| 2.664| 0| 0| -0.077| 9.492
RTD| 0.036| 0.252| 2.735| 0| 0| -0.077| 9.492
RTD| 0.032| 0.248| 2.819| 0| 0| -0.077| 9.492
RTD| 0.021| 0.258| 2.760| 0| 0| -0.077| 9.492
RTT| 00:01:25 (in-kernel periodic task, 1000 us period, priority 99)
RTH|----lat min|----lat avg|----lat max|-overrun|---msw|---lat best|--lat worst
RTD| 0.043| 0.262| 3.087| 0| 0| -0.077| 9.492
RTD| 0.037| 0.266| 2.625| 0| 0| -0.077| 9.492
RTD| 0.033| 0.262| 2.666| 0| 0| -0.077| 9.492
RTD| 0.035| 0.281| 2.792| 0| 0| -0.077| 9.492
RTD| 0.037| 0.255| 2.856| 0| 0| -0.077| 9.492
RTD| 0.027| 0.254| 2.786| 0| 0| -0.077| 9.492
RTD| 0.026| 0.268| 2.996| 0| 0| -0.077| 9.492
RTD| 0.038| 0.269| 3.177| 0| 0| -0.077| 9.492
RTD| 0.042| 0.268| 2.771| 0| 0| -0.077| 9.492
RTD| 0.028| 0.275| 2.636| 0| 0| -0.077| 9.492
RTD| 0.018| 0.255| 2.911| 0| 0| -0.077| 9.492
^C---|-----------|-----------|-----------|--------|------|-------------------------
but when I start the qt demos with deform and not in over load mode the cpu status
Mem: 66288K used, 835448K free, 0K shrd, 1452K buff, 31316K cached
CPU0: 85.2% usr 0.9% sys 0.0% nic 2.9% idle 0.0% io 0.0% irq 10.7% sirq
CPU1: 0.0% usr 1.9% sys 0.0% nic 98.0% idle 0.0% io 0.0% irq 0.0% sirq
CPU2: 42.1% usr 4.9% sys 0.0% nic 52.9% idle 0.0% io 0.0% irq 0.0% sirq
CPU3: 49.0% usr 0.0% sys 0.0% nic 50.9% idle 0.0% io 0.0% irq 0.0% sirq
and the latency
RTH|----lat min|----lat avg|----lat max|-overrun|---msw|---lat best|--lat worst
RTD| 2.524| 7.451| 26.214| 0| 0| -0.054| 44.532
RTD| 2.712| 7.461| 26.547| 0| 0| -0.054| 44.532
RTD| 2.716| 7.547| 24.911| 0| 0| -0.054| 44.532
RTD| 0.023| 7.071| 30.170| 0| 0| -0.054| 44.532
RTD| 0.045| 7.034| 23.898| 0| 0| -0.054| 44.532
RTD| 2.428| 7.632| 26.832| 0| 0| -0.054| 44.532
RTD| 0.008| 7.142| 28.521| 0| 0| -0.054| 44.532
RTD| -0.032| 5.846| 37.555| 0| 0| -0.054| 44.532
RTD| -0.012| 6.960| 34.334| 0| 0| -0.054| 44.532
RTD| -0.030| 5.557| 34.056| 0| 0| -0.054| 44.532
RTD| -0.016| 5.520| 31.408| 0| 0| -0.054| 44.532
RTD| -0.024| 5.889| 37.510| 0| 0| -0.054| 44.532
RTD| -0.020| 5.795| 34.795| 0| 0| -0.054| 44.532
RTD| -0.033| 5.369| 25.680| 0| 0| -0.054| 44.532
RTD| -0.019| 4.367| 49.594| 0| 0| -0.054| 49.594
RTD| -0.013| 4.021| 31.330| 0| 0| -0.054| 49.594
RTD| -0.005| 5.856| 39.871| 0| 0| -0.054| 49.594
RTD| 0.011| 6.889| 37.581| 0| 0| -0.054| 49.594
RTD| -0.019| 5.912| 32.818| 0| 0| -0.054| 49.594
RTD| -0.023| 3.746| 32.790| 0| 0| -0.054| 49.594
RTD| -0.012| 4.137| 45.545| 0| 0| -0.054| 49.594
RTT| 00:04:55 (in-kernel periodic task, 1000 us period, priority 99)
RTH|----lat min|----lat avg|----lat max|-overrun|---msw|---lat best|--lat worst
RTD| -0.024| 3.580| 29.679| 0| 0| -0.054| 49.594
RTD| 0.010| 5.488| 26.422| 0| 0| -0.054| 49.594
RTD| 2.661| 7.522| 26.701| 0| 0| -0.054| 49.594
RTD| 2.646| 7.722| 30.871| 0| 0| -0.054| 49.594
RTD| 2.867| 7.556| 30.040| 0| 0| -0.054| 49.594
RTD| 0.032| 7.417| 28.622| 0| 0| -0.054| 49.594
RTD| -0.001| 5.395| 25.174| 0| 0| -0.054| 49.594
RTD| -0.001| 6.977| 28.238| 0| 0| -0.054| 49.594
RTD| 0.316| 6.906| 32.367| 0| 0| -0.054| 49.594
RTD| -0.009| 5.827| 24.501| 0| 0| -0.054| 49.594
RTD| -0.007| 6.329| 24.979| 0| 0| -0.054| 49.594
RTD| -0.021| 5.515| 30.055| 0| 0| -0.054| 49.594
RTD| -0.003| 5.541| 26.237| 0| 0| -0.054| 49.594
RTD| -0.006| 7.325| 23.824| 0| 0| -0.054| 49.594
RTD| 0.010| 6.627| 22.536| 0| 0| -0.054| 49.594
RTD| 0.406| 7.061| 23.982| 0| 0| -0.054| 49.594
RTD| 0.156| 6.987| 23.474| 0| 0| -0.054| 49.594
RTD| 0.092| 6.806| 22.813| 0| 0| -0.054| 49.594
^C---|-----------|-----------|-----------|--------|------|-------------------------
where I am wrong ?thanks!
The problem still have
RTT| 00:12:37 (in-kernel periodic task, 1000 us period, priority 99)
RTH|----lat min|----lat avg|----lat max|-overrun|---msw|---lat best|--lat worst
RTD| 0.688| 1.159| 3.580| 0| 0| 0.556| 394.432
RTD| 0.704| 1.183| 3.502| 0| 0| 0.556| 394.432
RTD| 0.662| 1.167| 3.985| 0| 0| 0.556| 394.432
RTD| 0.732| 1.186| 3.425| 0| 0| 0.556| 394.432
RTD| 0.728| 1.183| 4.244| 0| 0| 0.556| 394.432
RTD| 0.678| 1.186| 3.744| 0| 0| 0.556| 394.432
RTD| 0.688| 1.144| 3.461| 0| 0| 0.556| 394.432
RTD| 0.701| 1.190| 3.882| 0| 0| 0.556| 394.432
RTD| 0.701| 1.147| 3.533| 0| 0| 0.556| 394.432
RTD| 0.693| 1.190| 3.783| 0| 0| 0.556| 394.432
RTD| 0.683| 1.155| 4.593| 0| 0| 0.556| 394.432
RTD| 0.669| 1.170| 3.697| 0| 0| 0.556| 394.432
RTD| 0.701| 1.162| 3.755| 0| 0| 0.556| 394.432
RTD| 0.688| 1.188| 3.305| 0| 0| 0.556| 394.432
RTD| 0.718| 1.180| 3.564| 0| 0| 0.556| 394.432
RTD| 0.686| 1.190| 4.303| 0| 0| 0.556| 394.432
RTD| 0.706| 1.171| 3.182| 0| 0| 0.556| 394.432
RTD| 0.666| 1.161| 3.676| 0| 0| 0.556| 394.432
RTD| 0.704| 1.167| 4.214| 0| 0| 0.556| 394.432
RTD| 0.708| 1.165| 3.776| 0| 0| 0.556| 394.432
RTD| 0.714| 1.158| 3.724| 0| 0| 0.556| 394.432
RTT| 00:12:58 (in-kernel periodic task, 1000 us period, priority 99)
RTH|----lat min|----lat avg|----lat max|-overrun|---msw|---lat best|--lat worst
RTD| 0.667| 1.168| 7.605| 0| 0| 0.556| 394.432
RTD| 0.701| 1.152| 3.488| 0| 0| 0.556| 394.432
RTD| 0.687| 1.171| 4.825| 0| 0| 0.556| 394.432
RTD| 0.691| 1.163| 3.366| 0| 0| 0.556| 394.432
RTD| 0.624| 1.201| 3.384| 0| 0| 0.556| 394.432
RTD| 0.662| 1.177| 15.516| 0| 0| 0.556| 394.432
RTD| 0.692| 1.188| 4.120| 0| 0| 0.556| 394.432
RTD| 0.680| 1.179| 3.724| 0| 0| 0.556| 394.432
RTD| 0.700| 1.194| 3.471| 0| 0| 0.556| 394.432
RTD| 0.662| 1.198| 3.469| 0| 0| 0.556| 394.432
RTD| 0.710| 1.207| 3.339| 0| 0| 0.556| 394.432
RTD| 0.712| 1.213| 3.838| 0| 0| 0.556| 394.432
RTD| 0.694| 1.196| 3.543| 0| 0| 0.556| 394.432
RTD| 0.726| 1.185| 7.985| 0| 0| 0.556| 394.432
RTD| 0.639| 1.196| 3.165| 0| 0| 0.556| 394.432
next reply other threads:[~2014-03-25 8:41 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-03-25 8:41 嵌入式工程师 [this message]
-- strict thread matches above, loose matches on Subject: below --
2014-03-25 2:07 [Xenomai] On mx6q specifically, it seems the framebuffer is a latency killer 嵌入式工程师
2014-03-25 7:12 ` 嵌入式工程师
2014-03-25 11:30 ` Gilles Chanteperdrix
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=453fa613.d9b0.144f866f11d.Coremail.ericvic@163.com \
--to=ericvic@163.com \
--cc=gilles.chanteperdrix@xenomai.org \
--cc=xenomai@xenomai.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.