On Wednesday 30 March 2016 23:12:54 Jon Elson wrote: > Hello, > > I have a customer who bought one of my Universal Stepper > Controllers, that uses the EPP mode on the parallel port for > communication. He is not getting any response from the > board. I've gone through the usual tests with him, and so > far have not made any progress. One thing he mentioned is : > > "I tried both addresses e000 and e010 (e010 must be > correct), clearing out all the parallel port modules > (parport.ko is necessary, else there are undefined symbols > while loading hal_ppmc.ko)" > > I'm thinking that rmmod'ing the various parallel port > modules may prevent the parallel port from working at all. > I think I've seen this on other systems. Does anybody > recognize the undefined symbols problem when loading a > kernel module that uses the parallel port? (hal_ppmc uses > inb and outb instructions to manipulate the parport > registers directly. It uses something that Jeff Epler set > up to put the port into EPP mode, using the info left in the > BIOS data block by PCI enumeration to find the address of > the ECR.) > > The customer's parport is PCIe, but we have already verified > these seem to work just like the old PCI. > > Thanks in advance for any ideas. > > Jon > Does his dmesg output not contain that data? Something similar to this:
[ 16.411015] parport_pc 00:05: reported by Plug and Play ACPI [ 16.412766] parport0: PC-style at 0x378, irq 7 [PCSPP,TRISTATE,EPP] And the last I knew, we had to blacklist these modules from an lsmod listing: parport_pc 26287 1 parport 35743 3 lp,ppdev,parport_pc in order that they not preclaim the port, preventing the LCNC driver from accessing it. However I think that was a pre 2.6 problem as my toy mill shows those two modules in the lsmod output both before and after running LinuxCNC, and its doing sw stepping. IIRC the parports address s/b in the loadrt line at the top of the .hal file, and its listed in the dmesg output after having run linuxcnc -l [130464.592847] I-pipe: head domain RTAI registered. [130464.592864] RTAI[hal]: compiled with gcc version 4.7.2 (Debian 4.7.2-5) . [130464.592961] RTAI[hal]: mounted (IPIPE-NOTHREADS, IMMEDIATE (INTERNAL IRQs DISPATCHED), ISOL_CPUS_MASK: 0). [130464.592971] SYSINFO: CPUs 2, LINUX APIC IRQ 2312, TIM_FREQ 12499458, CLK_FREQ 1800087000, CPU_FREQ 1800087000 [130464.592978] RTAI_APIC_TIMER_IPI: RTAI DEFINED 2314, VECTOR 2314; LINUX_APIC_TIMER_IPI: RTAI DEFINED 2312, VECTOR 2312 [130464.592987] TIMER NAME: lapic; VARIOUSLY FOUND APIC FREQs: 12499458, 12499458, 12401750 [130464.628584] RTAI[malloc]: global heap size = 2097152 bytes, <BSD>. [130464.628666] , <uses LINUX SYSCALLs>, kstacks pool size = 524288 bytes. [130464.628679] RTAI[sched]: hard timer type/freq = APIC/12499458(Hz); default timing: oneshot; linear timed lists. [130464.628688] RTAI[sched]: Linux timer freq = 250 (Hz), TimeBase freq = 1800087000 hz. [130464.628695] RTAI[sched]: timer setup = 999 ns, resched latency = 2943 ns. [130464.661290] RTAI[math]: loaded. [130464.917953] config string '0x378 out ' [130483.557094] RTAI[math]: unloaded. [130483.561850] SCHED releases registered named ALIEN PEDV$D [130483.579087] RTAI[malloc]: unloaded. [130483.676037] RTAI[sched]: unloaded (forced hard/soft/hard transitions: traps 0, syscalls 0). [130483.679967] I-pipe: head domain RTAI unregistered. [130483.680433] RTAI[hal]: unmounted. No clue if there's anything useful above, Jon. The RTAI stuff above is from a D525MW board of a machine that is still using software stepping via the regular parport. Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Genes Web page <http://geneslinuxbox.net:6309/gene> ------------------------------------------------------------------------------ Transform Data into Opportunity. Accelerate data analysis in your applications with Intel Data Analytics Acceleration Library. Click to learn more. http://pubads.g.doubleclick.net/gampad/clk?id=278785471&iu=/4140 _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
