-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
After upgrading from FC6 to F7, I see the following as I boot
Jul 3 22:04:43 sds-desk kernel: serial8250: too much work for irq17 Jul 3 22:04:43 sds-desk kernel: serial8250: too much work for irq17
Is this serious? - --
Steve
Steven Stern wrote:
After upgrading from FC6 to F7, I see the following as I boot
Jul 3 22:04:43 sds-desk kernel: serial8250: too much work for irq17 Jul 3 22:04:43 sds-desk kernel: serial8250: too much work for irq17
Is this serious?
It is definitely strange. Where did you get a card using a 8250 UART that is using IRQ 17? I didn't think anybody used them any more, as they are a UART without a transmit/receive buffer. Considering that it would be generating an interrupt for each character, you would not want it on a high IRQ. On the other hand, the device is probably not reporting itself correctly. If it does have a 8250 UART on it, see if you can change the IRQ to 3 or 4.
Mikkel
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Mikkel L. Ellertson wrote:
Steven Stern wrote:
After upgrading from FC6 to F7, I see the following as I boot
Jul 3 22:04:43 sds-desk kernel: serial8250: too much work for irq17 Jul 3 22:04:43 sds-desk kernel: serial8250: too much work for irq17
Is this serious?
It is definitely strange. Where did you get a card using a 8250 UART that is using IRQ 17? I didn't think anybody used them any more, as they are a UART without a transmit/receive buffer. Considering that it would be generating an interrupt for each character, you would not want it on a high IRQ. On the other hand, the device is probably not reporting itself correctly. If it does have a 8250 UART on it, see if you can change the IRQ to 3 or 4.
Mikkel
Here's some possible hardware (from /etc/sysconfig/hwconf)
class: OTHER bus: PCI detached: 0 desc: "Intel Corporation FA82537EP 56K V.92 Data/Fax Modem PCI" vendorId: 8086 deviceId: 1080 subVendorId: 1028 subDeviceId: 1000 pciType: 1 pcidom: 0 pcibus: 4 pcidev: 1 pcifn: 0
Here's /proc/interrupts
$ cat /proc/interrupts CPU0 CPU1 0: 294 0 IO-APIC-edge timer 1: 2274 2195 IO-APIC-edge i8042 7: 0 0 IO-APIC-edge parport0 8: 3 0 IO-APIC-edge rtc 9: 0 0 IO-APIC-fasteoi acpi 12: 70096 46743 IO-APIC-edge i8042 14: 56827 5962 IO-APIC-edge libata 15: 0 0 IO-APIC-edge libata 16: 48178 504269 IO-APIC-fasteoi wifi0, radeon@pci:0000:01:00.0 18: 36 17 IO-APIC-fasteoi uhci_hcd:usb1, ehci_hcd:usb5 19: 94 0 IO-APIC-fasteoi uhci_hcd:usb2 20: 86179 232010 IO-APIC-fasteoi uhci_hcd:usb3, snd_ca0106 21: 0 0 IO-APIC-fasteoi uhci_hcd:usb4 22: 57133 25898 IO-APIC-fasteoi libata NMI: 0 0 LOC: 5062797 5198387 ERR: 0 MIS: 0
- --
Steve
Steven Stern wrote:
Mikkel L. Ellertson wrote:
Steven Stern wrote:
After upgrading from FC6 to F7, I see the following as I boot
Jul 3 22:04:43 sds-desk kernel: serial8250: too much work for irq17 Jul 3 22:04:43 sds-desk kernel: serial8250: too much work for irq17
Is this serious?
It is definitely strange. Where did you get a card using a 8250 UART that is using IRQ 17? I didn't think anybody used them any more, as they are a UART without a transmit/receive buffer. Considering that it would be generating an interrupt for each character, you would not want it on a high IRQ. On the other hand, the device is probably not reporting itself correctly. If it does have a 8250 UART on it, see if you can change the IRQ to 3 or 4.
Mikkel
Here's some possible hardware (from /etc/sysconfig/hwconf)
class: OTHER bus: PCI detached: 0 desc: "Intel Corporation FA82537EP 56K V.92 Data/Fax Modem PCI" vendorId: 8086 deviceId: 1080 subVendorId: 1028 subDeviceId: 1000 pciType: 1 pcidom: 0 pcibus: 4 pcidev: 1 pcifn: 0
Yes, that is probably the problem hardware. For some reason, some modems will report that they are a 8250 UART, but they behave more like at 16550, with both send and receive FIFO buffers. They may or may not support the 16550 registers for controlling them. I don't remember all the details any more - I have not had to deal with them for a long time.
Mikkel