From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <453BAD90.9020306@domain.hid> Date: Sun, 22 Oct 2006 19:42:40 +0200 From: Wolfgang Grandegger MIME-Version: 1.0 Subject: Re: [Xenomai-core] switchtest FPU register error on PPC 2.6.18 References: <453B4816.4010007@domain.hid> <17723.27846.168894.602494@domain.hid> In-Reply-To: <17723.27846.168894.602494@domain.hid> Content-Type: multipart/mixed; boundary="------------080507090204000906020704" List-Id: "Xenomai life and development \(bug reports, patches, discussions\)" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Gilles Chanteperdrix Cc: xenomai-core This is a multi-part message in MIME format. --------------080507090204000906020704 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Gilles Chanteperdrix wrote: > Wolfgang Grandegger wrote: > > Hello, > > > > I just realized the error message: > > > > Error after context switch from task 8(rtup-8) to task > > 10(rtup_ufpp-10), FPU registers were set to 1000 (maybe task > > sleeper_ufps-0) > > > > when running "switchtest" on an PPC under Linux 2.6.18 with Xenomai head > > of SVN. I have attached the full output log. > > Are you sure you did ? Oh, well, ... see attachment. > > Can some tell me what went > > exactly wrong and where I should start looking at? > > It means that after a switch from task 8 to task 10, the FPU context is > the one of task 0. So, some context switches which should have switched > the FPU did not. If you are lucky, you should get this error > systematically. > > In order to investigate, you should try and reduce the number of tasks > created by switchtest. switchtest --help will tell you the list of tasks > created by switchtest by default, start from this list, and try removing > some tasks, keeping the failure. When you have the minimum set of tasks > that create the failure, try to trace the fpu switches, to see which one > does not do its job correctly. OK, thanks for the hint. I forgot to mention that it only happens once after reboot. Re-exectution does not show the error any more. Wolfgang. --------------080507090204000906020704 Content-Type: text/x-log; name="switchtest-error.log" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="switchtest-error.log" bash-3.00# cd xenomai/testsuite/switchtest/ bash-3.00# ./run * * * Type ^C to stop this application. * * == Testing FPU check routines... r0: 1 != 2 r1: 1 != 2 r2: 1 != 2 r3: 1 != 2 r4: 1 != 2 r5: 1 != 2 r6: 1 != 2 r7: 1 != 2 r8: 1 != 2 r9: 1 != 2 r10: 1 != 2 r11: 1 != 2 r12: 1 != 2 r13: 1 != 2 r14: 1 != 2 r15: 1 != 2 r16: 1 != 2 r17: 1 != 2 r18: 1 != 2 r19: 1 != 2 r20: 1 != 2 r21: 1 != 2 r22: 1 != 2 r23: 1 != 2 r24: 1 != 2 r25: 1 != 2 r26: 1 != 2 r27: 1 != 2 r28: 1 != 2 r29: 1 != 2 r30: 1 != 2 r31: 1 != 2 == FPU check routines: OK. == Threads: sleeper_ufps-0 rtk-1 rtk-2 rtk_fp-3 rtk_fp-4 rtk_fp_ufpp-5 rtk_fp_ufpp-6 rtup-7 rtup-8 rtup_ufpp-9 rtup_ufpp-10 rtus-11 rtus-12 rtus_ufps-13 rtus_ufps-14 rtuo-15 rtuo-16 rtuo_ufpp-17 rtuo_ufpp-18 rtuo_ufps-19 rtuo_ufps-20 rtuo_ufpp_ufps-21 rtuo_ufpp_ufps-22 r0: 1000 != 10 r1: 1000 != 10 r2: 1000 != 10 r3: 1000 != 10 r4: 1000 != 10 r5: 1000 != 10 r6: 1000 != 10 r7: 1000 != 10 r8: 1000 != 10 r9: 1000 != 10 r10: 1000 != 10 r11: 1000 != 10 r12: 1000 != 10 r13: 1000 != 10 r14: 1000 != 10 r15: 1000 != 10 r16: 1000 != 10 r17: 1000 != 10 r18: 1000 != 10 r19: 1000 != 10 r20: 1000 != 10 r21: 1000 != 10 r22: 1000 != 10 r23: 1000 != 10 r24: 1000 != 10 r25: 1000 != 10 r26: 1000 != 10 r27: 1000 != 10 r28: 1000 != 10 r29: 1000 != 10 r30: 1000 != 10 r31: 1000 != 10 Error after context switch from task 8(rtup-8) to task 10(rtup_ufpp-10), FPU registers were set to 1000 (maybe task sleeper_ufps-0) RTT| 00:00:00 RTH|ctx switches|-------total RTD| 28| 28 --------------080507090204000906020704--