Greetings! So this is not a kernel error? These messages are debugging code only for userland segfaults? My apologies if so for misunderstanding. Looked like an oops.
I'll try to isolate, but my guess is that this is dependent on the memory layout of the parent program. Could some fork be running into already used memory? I have a smaller executable from a different gcl version which does not trigger this on the same machine. Why can't I get a segfault in gdb with an offending address in the normal fashion if this is either libc or my program? The sigkill is what made me suspect the kernel. Can I decode the instruction on paer? If so how? Take care, "Carlos O'Donell" <[email protected]> writes: > On Tue, Feb 2, 2010 at 1:42 PM, Camm Maguire <[email protected]> wrote: >>> Does system() fail *only* when running your specific application? >>> >> >> Are you asking if I've compiled a separate wrapper around system as a >> check? No not yet. But gcl, as opposed to gclcvs, executes a similar >> command without issue in the same environment. And since writing, the >> following has been pointed out to me from the kernel log: > > Do you have a tiny testcase showing the use of system() and does it work? > >> [4252735.880000] >> [4252735.880000] do_page_fault() pid=24223 command='raw_pre_gcl' type=15 >> address=0x00000000 > > Your process tried to access 0x00000000 and faulted. > >> [4252735.880000] >> [4252735.880000] YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI >> [4252735.880000] PSW: 00000000000011001111111100001111 Not tainted >> [4252735.880000] r00-03 000000ff000cff0f 00000000001283fc 000000000002ee2f >> 00000000faf01980 >> [4252735.880000] r04-07 00000000001223fc 0000000000000000 0000000000328758 >> 00000000faf01034 >> [4252735.880000] r08-11 000000000012873c 000000000036e7b0 0000000000479f68 >> 0000000000000001 >> [4252735.880000] r12-15 000000000032b87c 0000000000625708 0000000000620cc8 >> 00000000004ba044 >> [4252735.880000] r16-19 000000000032e6ec 0000000000329758 00000000006256e0 >> 00000000001223fc >> [4252735.880000] r20-23 00000000006255d8 000000000000006c 000000000032877b >> 0000000025252525 >> [4252735.880000] r24-27 00000000faf01ac8 00000000faf01ac8 0000000000328758 >> 00000000001223fc >> [4252735.880000] r28-31 0000000000328758 0000000000000000 00000000faf01ac0 >> 0000000040bc928b >> [4252735.880000] sr00-03 000000000000e000 0000000000000000 0000000000000000 >> 000000000000e000 >> [4252735.880000] sr04-07 000000000000e000 000000000000e000 000000000000e000 >> 000000000000e000 >> [4252735.880000] >> [4252735.880000] VZOUICununcqcqcqcqcqcrmunTDVZOUI >> [4252735.880000] FPSR: 00001100001111111111100000000000 >> [4252735.880000] FPER1: 00000000 >> [4252735.880000] fr00-03 0c3ff80000000000 0000000000000000 0000000000000000 >> 0000000000000000 >> [4252735.880000] fr04-07 0000000000000000 4a89174e00000000 12848d8400004e37 >> 024f67e800000000 >> [4252735.880000] fr08-11 f3f419bff0094230 00004a88c949ba09 000049566427d850 >> 000000012f83c3c0 >> [4252735.880000] fr12-15 0000000000000002 000000012f840010 000000012f840000 >> 0000000040622e70 >> [4252735.880000] fr16-19 0000000000000002 000000012f840010 0000000000000081 >> ffffffffc0000000 >> [4252735.880000] fr20-23 000000004010ae2c 0000000040641670 000000019999999a >> 0000000200000001 >> [4252735.880000] fr24-27 3ca0000000000000 3ff0000000000000 3ff0000000000000 >> 3ff0000000000000 >> [4252735.880000] fr28-31 0000000000000000 0000000040669598 0000000000000000 >> ffffffff00004e37 >> [4252735.880000] >> [4252735.880000] IASQ: 000000000000e000 000000000000e000 IAOQ: >> 000000000002ee2b 000000000002c013 >> [4252735.880000] IIR: 0cbc1280 ISR: 000000000000e000 IOR: >> 0000000000000000 > > I was executing instruction 0x0cbc1280 (away from my box right now or > I'd decode this). > >> [4252735.880000] CPU: 0 CR30: 0000000144d7c000 CR31: >> f73bff7ebebfbfe7 >> [4252735.880000] ORIG_R28: 000000000032877b >> [4252735.880000] IAOQ[0]: 000000000002ee2b >> [4252735.880000] IAOQ[1]: 000000000002c013 >> [4252735.880000] RP(r2): 000000000002ee2f >> >> Advice? > > Anything could be wrong. You have to debug it. > > Cheers, > Carlos. > > > > -- Camm Maguire [email protected] ========================================================================== "The earth is but one country, and mankind its citizens." -- Baha'u'llah -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
