[mythtv-users] Can't get LIRC to work with my Serial card - IRQ tied up?
Ryan Steffes
rbsteffes at gmail.com
Sun Sep 23 03:15:00 UTC 2007
On 9/22/07, Jay Mallar <jaymallar at gmail.com> wrote:
> First, I realize this isn't necessarily a good topic for this list, but
> I'm sort of stuck on what to do next. My problem isn't really with IRC,
> I don't think it's with my serial card, and I just don't know where to turn.
>
> My machine:
>
> Motherboard: Asus P5N32-E SLI Plus
> http://usa.asus.com/products.aspx?l1=3&l2=11&l3=495&l4=0&model=1567&modelmenu=1
>
> 4 PCI-e Serial Port Card:
> http://www.qualitycables.com/productdetails1.cfm?sku=CM-4SRS232&cats=99172&catid=495%2C538%2C472%2C538%2C882%2C909
>
> Homebrew Serial IR Receiver:
> http://www.irblaster.info/receiver.html
>
> Everything seems to be recognized by my machine during startup:
>
> ...
> Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing enabled
> ACPI: PCI Interrupt Link [APC6] enabled at IRQ 16
> ACPI: PCI Interrupt 0000:0a:00.0[A] -> Link [APC6] -> GSI 16 (level,
> low) -> IRQ 16
> ttyS0: detected caps 00000700 should be 00000100
> 0000:0a:00.0: ttyS0 at I/O 0x5f00 (irq = 16) is a 16C950/954
> ttyS1: detected caps 00000700 should be 00000100
> 0000:0a:00.0: ttyS1 at I/O 0x5f08 (irq = 16) is a 16C950/954
> ttyS2: detected caps 00000700 should be 00000100
> 0000:0a:00.0: ttyS2 at I/O 0x5f10 (irq = 16) is a 16C950/954
> ttyS3: detected caps 00000700 should be 00000100
> 0000:0a:00.0: ttyS3 at I/O 0x5f18 (irq = 16) is a 16C950/954
> ...
>
> lspci -v shows:
>
> 0a:00.0 Serial controller: Oxford Semiconductor Ltd OX16PCI954 (Quad
> 16950 UART) function 0 (Uart) (prog-if 06 [16950])
> Subsystem: Oxford Semiconductor Ltd Unknown device 0000
> Flags: medium devsel, IRQ 16
> I/O ports at 5f00 [size=32]
> Memory at ef1ff000 (32-bit, non-prefetchable) [size=4K]
> I/O ports at 5e00 [size=32]
> Memory at ef1fe000 (32-bit, non-prefetchable) [size=4K]
> Capabilities: [40] Power Management version 1
>
> 0a:00.1 Parallel controller: Oxford Semiconductor Ltd OX16PCI954 (Quad
> 16950 UART) function 1 (parallel port) (prog-if 01 [BiDir])
> Subsystem: Oxford Semiconductor Ltd Unknown device 0000
> Flags: medium devsel, IRQ 16
> I/O ports at 5d00 [size=8]
> I/O ports at 5c00 [size=8]
> I/O ports at 5b00 [size=32]
> Memory at ef1fd000 (32-bit, non-prefetchable) [size=4K]
> Capabilities: [40] Power Management version 1
>
> I have my IR receiver hooked to the first serial port, which I believe
> is at 0x5f00 IRQ 16.
>
> I'm using the latest CVS LIRC drivers (which appear to be 0.8.3).
>
> When running setup.sh, I told the tool I'm using a homebrew IR receiver
> configured at 0x5f00 IRQ 16 (I chose the Other option to configure my
> values rather than choose the COM1 option which defaulted to a different
> address):
>
> I did get the msg:
>
> You will have to use the lirc_serial kernel module.
>
> Ran 'make' and 'make install' which did install the lirc_serial.ko file.
>
> Now to load everything up. First the modules:
>
> 'modprobe lirc_serial' returned with no errors and the following in my log:
>
> Sep 22 10:35:46 localhost kernel: lirc_dev: IR Remote Control driver
> registered, at major 61
> Sep 22 10:35:46 localhost kernel: lirc_serial: auto-detected active low
> receiver
> Sep 22 10:35:46 localhost kernel: lirc_dev: lirc_register_plugin:
> sample_rate: 0
>
> The sample rate seemed weird to me - but I let that go for now.
>
> :lsmod | grep irc
> lirc_serial 17156 0
> lirc_dev 18676 1 lirc_serial
>
> Now I run lircd:
>
> :lircd
>
> It returns immediately without error. Logs show:
>
> Sep 22 10:38:33 localhost lircd-0.8.3-CVS[31147]: lircd(serial) ready
>
> Okay - should be good to go now, so I try 'irw' to test it:
>
> :irw
>
> This also returns immediately with no error, but this in the logs:
>
> Sep 22 10:39:25 localhost kernel: lirc_serial: IRQ 16 busy
> Sep 22 10:39:25 localhost lircd-0.8.3-CVS[31147]: default_init(): Device
> or resource busy
> Sep 22 10:39:25 localhost lircd-0.8.3-CVS[31147]: caught signal
>
> And that's where I get stuck. I do think my problem is IRQ 16 being
> busy, but just not sure how to fix it.
>
> Here's some other info:
>
> :cat /proc/interrupts
> CPU0 CPU1 CPU2 CPU3
> 0: 1843 0 0 0 IO-APIC-edge timer
> 1: 276 0 997 0 IO-APIC-edge i8042
> 8: 0 0 0 0 IO-APIC-edge rtc0
> 9: 0 0 0 0 IO-APIC-fasteoi acpi
> 12: 2838 0 2639 0 IO-APIC-edge i8042
> 14: 0 0 0 0 IO-APIC-edge libata
> 15: 0 0 0 0 IO-APIC-edge libata
> 16: 19731235 0 13793052 0 IO-APIC-fasteoi nvidia
> 17: 18967670 0 55609 0 IO-APIC-fasteoi
> ohci_hcd:usb1, sata_nv
> 18: 0 0 0 0 IO-APIC-fasteoi
> ehci_hcd:usb2
> 19: 81751 5388 110312 0 IO-APIC-fasteoi sata_nv
> 20: 7157 0 18186 0 IO-APIC-fasteoi
> sata_nv, HDA Intel
> 21: 2128 0 1768 0 IO-APIC-fasteoi
> firewire_ohci
> 22: 1732 0 210608 0 IO-APIC-fasteoi ivtv0
> 214: 1654 0 0 25535507 PCI-MSI-edge eth0
> NMI: 0 0 0 0
> LOC: 24847603 20553428 29031709 24752948
> ERR: 1
> MIS: 0
>
> :cat /proc/ioports
> 0000-001f : dma1
> 0020-0021 : pic1
> 0040-0043 : timer0
> 0050-0053 : timer1
> 0060-006f : keyboard
> 0070-0073 : rtc0
> 0080-008f : dma page reg
> 00a0-00a1 : pic2
> 00c0-00df : dma2
> 00f0-00ff : fpu
> 0170-0177 : 0000:00:0d.0
> 0170-0177 : libata
> 01f0-01f7 : 0000:00:0d.0
> 01f0-01f7 : libata
> 0376-0376 : 0000:00:0d.0
> 0376-0376 : libata
> 03c0-03df : vga+
> 03f6-03f6 : 0000:00:0d.0
> 03f6-03f6 : libata
> 0960-0967 : 0000:00:0e.1
> 0960-0967 : libata
> 0970-0977 : 0000:00:0e.0
> 0970-0977 : libata
> 09e0-09e7 : 0000:00:0e.1
> 09e0-09e7 : libata
> 09f0-09f7 : 0000:00:0e.0
> 09f0-09f7 : libata
> 0b60-0b63 : 0000:00:0e.1
> 0b60-0b63 : libata
> 0b70-0b73 : 0000:00:0e.0
> 0b70-0b73 : libata
> 0be0-0be3 : 0000:00:0e.1
> 0be0-0be3 : libata
> 0bf0-0bf3 : 0000:00:0e.0
> 0bf0-0bf3 : libata
> 0cf8-0cff : PCI conf1
> 1000-107f : pnp 00:01
> 1000-1003 : ACPI PM1a_EVT_BLK
> 1004-1005 : ACPI PM1a_CNT_BLK
> 1008-100b : ACPI PM_TMR
> 101c-101c : ACPI PM2_CNT_BLK
> 1020-1027 : ACPI GPE0_BLK
> 1080-10ff : pnp 00:01
> 1400-147f : pnp 00:01
> 1480-14ff : pnp 00:01
> 14a0-14af : ACPI GPE1_BLK
> 1800-187f : pnp 00:01
> 1880-18ff : pnp 00:01
> 1c00-1c3f : 0000:00:0a.1
> 1c00-1c3f : nForce2_smbus
> 1c80-1cbf : 0000:00:0a.1
> 1c80-1cbf : nForce2_smbus
> 4000-4fff : PCI Bus #0b
> 5000-5fff : PCI Bus #09
> 5000-5fff : PCI Bus #0a
> 5b00-5b1f : 0000:0a:00.1
> 5c00-5c07 : 0000:0a:00.1
> 5d00-5d07 : 0000:0a:00.1
> 5d00-5d02 : parport0
> 5e00-5e1f : 0000:0a:00.0
> 5f00-5f1f : 0000:0a:00.0
> 5f00-5f07 : lirc_serial
> 5f08-5f0f : serial
> 5f10-5f17 : serial
> 5f18-5f1f : serial
> 6000-6fff : PCI Bus #08
> 7000-7fff : PCI Bus #07
> 8000-8fff : PCI Bus #06
> 9000-9fff : PCI Bus #05
> a000-afff : PCI Bus #03
> b000-bfff : PCI Bus #02
> c000-cfff : PCI Bus #04
> cf00-cf7f : 0000:04:0b.0
> d000-dfff : PCI Bus #01
> eb00-eb07 : 0000:00:12.0
> eb00-eb07 : forcedeth
> ec00-ec07 : 0000:00:11.0
> ec00-ec07 : forcedeth
> ed00-ed0f : 0000:00:0e.2
> ed00-ed0f : libata
> ee00-ee03 : 0000:00:0e.2
> ee00-ee03 : libata
> ef00-ef07 : 0000:00:0e.2
> ef00-ef07 : libata
> f000-f003 : 0000:00:0e.2
> f000-f003 : libata
> f100-f107 : 0000:00:0e.2
> f100-f107 : libata
> f200-f20f : 0000:00:0e.1
> f200-f20f : libata
> f700-f70f : 0000:00:0e.0
> f700-f70f : libata
> fc00-fc0f : 0000:00:0d.0
> fc00-fc0f : libata
> ff00-ff3f : 0000:00:0a.1
>
> The IO port output looks good, but the IRQ output appears to have
> assigned IRQ 16 to nvidia. I do have aN NVidia GeForce 6200 installed
> as my video card (which works perfectly).
>
> So if I had to guess, I'm hitting an IRQ sharing issue, but not sure how
> to correct it. Neither my Video card not my Serial port card have any
> jumpers or any ability to change IRQs via the hardware.
>
> Suggestions?
>
>
Likely you're getting the default device bound to the serial port.
See if you can free it up with setserial:
# setserial /dev/ttyS0 uart none
There's a guide to setting up mythtv on the wiki
(http://www.mythtv.org/wiki/index.php/Installing_MythTV_on_an_Athlon64_using_Fedora_Core)
that mentions how to do it automatically in your modprobe.conf, or you
can put it in the lircd start up script like I did.
Ry
More information about the mythtv-users
mailing list