From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <4797594B.7060504@domain.hid> Date: Wed, 23 Jan 2008 16:12:11 +0100 From: Jan Kiszka MIME-Version: 1.0 References: <47965E88.5000106@domain.hid> In-Reply-To: <47965E88.5000106@domain.hid> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: Re: [Xenomai-help] x86_64 user/system process accounting List-Id: Help regarding installation and common use of Xenomai List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Kyle Howell Cc: xenomai@xenomai.org Jan Kiszka wrote: > Kyle Howell wrote: >>> Kyle Howell wrote: >>>> I'm running Xenomai 2.4.1 against Linux 2.6.23.12 on an >>> x86_64 (Core2) >>>> system. I'm finding that all of my CPU cycles are being accounted as >>>> kernel time rather than user time. The correct processes are still >>>> billed, but the system is always 0% user. I'm wondering if >>> this problem >>>> is specific to my setup, or if it is a real bug. Can anyone else out >>>> there running x86_64 confirm or deny this behavior? >>> That bug should have been fixed by [1]. Already updated your ipipe >>> patch? This one [2] also includes your recently posted fix. >> Thanks, Jan. >> >> I'm basing off of ipipe-2.6.23-x86_64-1.4-01, which appears to include >> that patch. I don't think I saw this problem back on 2.4-rc3, so I may > > /me would be surprised if the issue was not present in that version or > any earlier one. > >> try rolling back some of the I-pipe patches since then. Also, I just >> tried a kernel with CONFIG_IPIPE turned on but CONFIG_XENOMAI turned >> off, and it didn't show this problem. I don't know if that casts doubt >> on Xenomai itself, or if I just made the pipeline too simple. >> >> Do you think any of your recent fixes against 2.6.24-x86 would be >> related to this? > > Nope, they are not related. But you could try if 2.6.24 with the latest > patch shows that effect, too. Maybe something slipped into the backport > of the original fix for 2.6.24-rcX. I will check with 2.6.23 tomorrow at > work and keep you posted. I currently have my fingers on a box with 2.6.23.14 with Xenomai 2.4.1 and I-pipe 1.4-01 (all standard), but I'm not able to reproduce your effect. Some simple shell loop nicely loads one CPU at 100%, and /proc/stat looks like this: cpu 38794 0 6325 411304 2882 0 3265 0 cpu0 34786 0 3094 189892 413 0 2716 0 cpu1 4007 0 3231 221411 2468 0 548 0 Could you send me your full .config? Jan -- Siemens AG, Corporate Technology, CT SE 2 Corporate Competence Center Embedded Linux