ide interrupts

ide interrupts

[Trevor Woerner]

There seems to be a lot of code in the IDE subsystem relating to
interrupts. Is there a way to set a device in polled (i.e.
non-interrupt) mode? Maybe some way to run the tasklet/bh periodically?

Does the IDE system want to monitor my ATA device's Ready_NotBusy
interrupt?

	Trevor

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide interrupts

[Benjamin Herrenschmidt]

On Fri, 2003-05-23 at 17:36, Trevor Woerner wrote:
> There seems to be a lot of code in the IDE subsystem relating to
> interrupts. Is there a way to set a device in polled (i.e.
> non-interrupt) mode? Maybe some way to run the tasklet/bh periodically?
>
> Does the IDE system want to monitor my ATA device's Ready_NotBusy
> interrupt?

No, linux IDE layer can't do that in a sane way, you are asking for
a _lot_ of trouble... (especially if doing DMA)

Ben.


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide interrupts

[linas]

On Fri, May 23, 2003 at 11:36:51AM -0400, Trevor Woerner wrote:
>
> There seems to be a lot of code in the IDE subsystem relating to
> interrupts. Is there a way to set a device in polled (i.e.
> non-interrupt) mode? Maybe some way to run the tasklet/bh periodically?

Why?

> Does the IDE system want to monitor my ATA device's Ready_NotBusy
> interrupt?

Why?

--linas

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide interrupts

[Trevor Woerner]

On May 23, 2003 12:09 pm, Benjamin Herrenschmidt wrote:
> On Fri, 2003-05-23 at 17:36, Trevor Woerner wrote:
> > There seems to be a lot of code in the IDE subsystem relating to
> > interrupts. Is there a way to set a device in polled (i.e.
> > non-interrupt) mode? Maybe some way to run the tasklet/bh
> > periodically?
> >
> > Does the IDE system want to monitor my ATA device's Ready_NotBusy
> > interrupt?
>
> No, linux IDE layer can't do that in a sane way, you are asking for
> a _lot_ of trouble... (especially if doing DMA)

"No" there isn't a way to set the device into polled mode, or "No" it
doesn't monitor the Ready_NotBusy interrupt?

	Trevor

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: ide interrupts

[Geert Uytterhoeven]

On Sun, 25 May 2003, Trevor Woerner wrote:
> On May 23, 2003 12:09 pm, Benjamin Herrenschmidt wrote:
> > On Fri, 2003-05-23 at 17:36, Trevor Woerner wrote:
> > > There seems to be a lot of code in the IDE subsystem relating to
> > > interrupts. Is there a way to set a device in polled (i.e.
> > > non-interrupt) mode? Maybe some way to run the tasklet/bh
> > > periodically?
> > >
> > > Does the IDE system want to monitor my ATA device's Ready_NotBusy
> > > interrupt?
> >
> > No, linux IDE layer can't do that in a sane way, you are asking for
> > a _lot_ of trouble... (especially if doing DMA)
>
> "No" there isn't a way to set the device into polled mode, or "No" it
> doesn't monitor the Ready_NotBusy interrupt?

Can't you use the Ready_NotBusy interrupt as IDE interrupt?

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

ide interrupts

[Kathy Frazier]

I am currently debugging a problem with an in-house developed PCI Bus Master
card on a Pentium 4 machine.  I saw something that surprised me that I
thought you might be able to shed some light on.  I noticed that my ide1
device (cdrom) is constantly generating interrupts (as per
/proc/interrupts).  The device is mounted, but there is no CD in the drive,
nor am I directly accessing it.  What is causing these interrupts?  I
thought it might be because  APM was enabled, so I added apm=off on the boot
line an re-booted.  However, it still is generating lots of interrupts.  The
section of boot up messages related to the ide stuff is below.  I also
included info resulting from hdparm -i /dev/hdc.

Thanks in advance for your help.

Kathy Frazier
Senior Software Engineer
Max Daetwyler Corporation-Dayton Division
2133 Lyons Road
Miamisburg, OH 45342
Tel #: 937.439-1582 ext 6158
Fax #: 937.439-1592
Email: kfrazier@daetwyler.com
http://www.daetwyler.com


Dec  1 13:29:42 rti10 kernel: Uniform Multi-Platform E-IDE driver Revision:
7.00beta-2.4
Dec  1 13:29:42 rti10 kernel: ide: Assuming 33MHz system bus speed for PIO
modes; override with idebus=xx
Dec  1 13:29:42 rti10 kernel: ICH4: IDE controller at PCI slot 00:1f.1
Dec  1 13:29:42 rti10 kernel: PCI: Found IRQ 10 for device 00:1f.1
Dec  1 13:29:42 rti10 kernel: PCI: Sharing IRQ 10 with 00:1d.2
Dec  1 13:29:42 rti10 aksparlnx: Warning: kernel-module version mismatch
Dec  1 13:29:42 rti10 kernel: ICH4: chipset revision 2
Dec  1 13:29:42 rti10 aksparlnx: ^I/opt/aksparlnx/drv/2.4.19/aksparlnx.o was
compiled for kernel version 2.4.19
Dec  1 13:29:42 rti10 kernel: ICH4: not 100%% native mode: will probe irqs
later
Dec  1 13:29:42 rti10 aksparlnx: ^Iwhile this kernel is version
2.4.20-8custom
Dec  1 13:29:42 rti10 kernel:     ide0: BM-DMA at 0xf000-0xf007, BIOS
settings: hda:DMA, hdb:pio
Dec  1 13:29:42 rti10 aksparlnx: Warning: loading
/opt/aksparlnx/drv/2.4.19/aksparlnx.o will taint the kernel: non-GPL
license - Copyright 1999-2002 Aladdin Knowledge Systems.
Dec  1 13:29:42 rti10 kernel:     ide1: BM-DMA at 0xf008-0xf00f, BIOS
settings: hdc:DMA, hdd:pio
Dec  1 13:29:42 rti10 aksparlnx:   See
http://www.tux.org/lkml/#export-tainted for information about tainted
modules
Dec  1 13:29:42 rti10 kernel: hda: ST340016A, ATA DISK drive
Dec  1 13:29:42 rti10 aksparlnx: Warning: loading
/opt/aksparlnx/drv/2.4.19/aksparlnx.o will taint the kernel: forced load
Dec  1 13:29:42 rti10 kernel: blk: queue c053ab40, I/O limit 4095Mb (mask
0xffffffff)
Dec  1 13:29:42 rti10 aksparlnx: Module aksparlnx loaded, with warnings
Dec  1 13:29:42 rti10 kernel: hdc: FX54++W, ATAPI CD/DVD-ROM drive
Dec  1 13:29:42 rti10 aksparlnx:
Dec  1 13:29:42 rti10 kernel: ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
Dec  1 13:29:42 rti10 rc: Starting aksparlnx:  succeeded
Dec  1 13:29:42 rti10 kernel: ide1 at 0x170-0x177,0x376 on irq 15
Dec  1 13:29:42 rti10 kernel: hda: host protected area => 1
Dec  1 13:29:42 rti10 kernel: hda: 78165360 sectors (40021 MB) w/2048KiB
Cache, CHS=4865/255/63, UDMA(100)
Dec  1 13:29:42 rti10 kernel: hdc: ATAPI 54X CD-ROM drive, 128kB Cache,
UDMA(33)
Dec  1 13:29:42 rti10 apmd[1060]: Version 3.0.2 (APM BIOS 1.2, Linux driver
1.16)
Dec  1 13:29:42 rti10 kernel: Uniform CD-ROM driver Revision: 3.12
Dec  1 13:29:42 rti10 apmd: apmd startup succeeded
Dec  1 13:29:42 rti10 kernel: ide-floppy driver 0.99.newide
Dec  1 13:29:42 rti10 kernel: Partition check:
Dec  1 13:29:42 rti10 aksusbd[1071]: loaded, daemon version: 1.5, key API
(USB) version: 3.81, key API (parallel) version: 3.81
Dec  1 13:29:42 rti10 aksusbd: aksusbd startup succeeded
Dec  1 13:29:42 rti10 kernel:  hda: hda1 hda2 hda3
Dec  1 13:29:42 rti10 kernel: ide-floppy driver 0.99.newide
Dec  1 13:29:38 rti10 kudzu:  succeeded
Dec  1 13:29:42 rti10 kernel: ohci1394: $Rev: 693 $ Ben Collins
<bcollins@debian.org>
Dec  1 13:29:38 rti10 kudzu: Updating /etc/fstab succeeded
Dec  1 13:29:42 rti10 kernel: PCI: Found IRQ 10 for device 02:03.0
Dec  1 13:29:38 rti10 sysctl: net.ipv4.ip_forward = 0
Dec  1 13:29:42 rti10 kernel: ohci1394_0: OHCI-1394 1.0 (PCI): IRQ=[10]
MMIO=[f2000000-f20007ff]  Max Packet=[2048]
Dec  1 13:29:42 rti10 automount[1124]: starting automounter version 3.1.7,
path = /home, maptype = file, mapname = /etc/auto.home
Dec  1 13:29:42 rti10 autofs: automount startup succeeded
Dec  1 13:29:38 rti10 sysctl: net.ipv4.conf.default.rp_filter = 1
Dec  1 13:29:42 rti10 kernel: ieee1394: SelfID completion called outside of
bus reset!
Dec  1 13:29:38 rti10 sysctl: kernel.core_uses_pid = 1
Dec  1 13:29:43 rti10 kernel: raw1394: /dev/raw1394 device initialized
Dec  1 13:29:38 rti10 network: Setting network parameters:  succeeded
Dec  1 13:29:43 rti10 kernel: usb.c: registered new driver usbdevfs
Dec  1 13:29:38 rti10 network: Bringing up loopback interface:  succeeded


/dev/hdc:

 Model=FX54++W, FwRev=U01L, SerialNo=
 Config={ Fixed Removeable DTR<=5Mbs DTR>10Mbs nonMagnetic }
 RawCHS=0/0/0, TrkSize=0, SectSize=0, ECCbytes=0
 BuffType=unknown, BuffSize=128kB, MaxMultSect=0
 (maybe): CurCHS=0/0/0, CurSects=0, LBA=yes, LBAsects=0
 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
 PIO modes:  pio0 pio1 pio2 pio3 pio4
 DMA modes:  sdma0 sdma1 sdma2 mdma0 mdma1 mdma2
 UDMA modes: udma0 udma1 *udma2
 AdvancedPM=no



Kathy Frazier
Senior Software Engineer
Max Daetwyler Corporation-Dayton Division
2133 Lyons Road
Miamisburg, OH 45342
Tel #: 937.439-1582 ext 6158
Fax #: 937.439-1592
Email: kfrazier@daetwyler.com
http://www.daetwyler.com

Re: ide interrupts

[Randy.Dunlap]

On Fri, 5 Dec 2003 11:25:30 -0500 "Kathy Frazier" <kfrazier@mdc-dayton.com> wrote:

| I am currently debugging a problem with an in-house developed PCI Bus Master
| card on a Pentium 4 machine.  I saw something that surprised me that I
| thought you might be able to shed some light on.  I noticed that my ide1
| device (cdrom) is constantly generating interrupts (as per
| /proc/interrupts).  The device is mounted, but there is no CD in the drive,
| nor am I directly accessing it.  What is causing these interrupts?  I
| thought it might be because  APM was enabled, so I added apm=off on the boot
| line an re-booted.  However, it still is generating lots of interrupts.  The
| section of boot up messages related to the ide stuff is below.  I also
| included info resulting from hdparm -i /dev/hdc.
| 
| Thanks in advance for your help.

Hi,

I'm not sure about this, but I suspect it to be related to
automount/autofs or some other daemon-like process that's trying to be
too helpful.  Find out how to disable that (listed in logs below)
and see if it stops.


| Dec  1 13:29:42 rti10 kernel: Uniform Multi-Platform E-IDE driver Revision:
| 7.00beta-2.4
| Dec  1 13:29:42 rti10 kernel: ide: Assuming 33MHz system bus speed for PIO
| modes; override with idebus=xx
| Dec  1 13:29:42 rti10 kernel: ICH4: IDE controller at PCI slot 00:1f.1
| Dec  1 13:29:42 rti10 kernel: PCI: Found IRQ 10 for device 00:1f.1
| Dec  1 13:29:42 rti10 kernel: PCI: Sharing IRQ 10 with 00:1d.2
| Dec  1 13:29:42 rti10 aksparlnx: Warning: kernel-module version mismatch
| Dec  1 13:29:42 rti10 kernel: ICH4: chipset revision 2
| Dec  1 13:29:42 rti10 aksparlnx: ^I/opt/aksparlnx/drv/2.4.19/aksparlnx.o was
| compiled for kernel version 2.4.19
| Dec  1 13:29:42 rti10 kernel: ICH4: not 100%% native mode: will probe irqs
| later
| Dec  1 13:29:42 rti10 aksparlnx: ^Iwhile this kernel is version
| 2.4.20-8custom
| Dec  1 13:29:42 rti10 kernel:     ide0: BM-DMA at 0xf000-0xf007, BIOS
| settings: hda:DMA, hdb:pio
| Dec  1 13:29:42 rti10 aksparlnx: Warning: loading
| /opt/aksparlnx/drv/2.4.19/aksparlnx.o will taint the kernel: non-GPL
| license - Copyright 1999-2002 Aladdin Knowledge Systems.
| Dec  1 13:29:42 rti10 kernel:     ide1: BM-DMA at 0xf008-0xf00f, BIOS
| settings: hdc:DMA, hdd:pio
| Dec  1 13:29:42 rti10 aksparlnx:   See
| http://www.tux.org/lkml/#export-tainted for information about tainted
| modules
| Dec  1 13:29:42 rti10 kernel: hda: ST340016A, ATA DISK drive
| Dec  1 13:29:42 rti10 aksparlnx: Warning: loading
| /opt/aksparlnx/drv/2.4.19/aksparlnx.o will taint the kernel: forced load
| Dec  1 13:29:42 rti10 kernel: blk: queue c053ab40, I/O limit 4095Mb (mask
| 0xffffffff)
| Dec  1 13:29:42 rti10 aksparlnx: Module aksparlnx loaded, with warnings
| Dec  1 13:29:42 rti10 kernel: hdc: FX54++W, ATAPI CD/DVD-ROM drive
| Dec  1 13:29:42 rti10 aksparlnx:
| Dec  1 13:29:42 rti10 kernel: ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
| Dec  1 13:29:42 rti10 rc: Starting aksparlnx:  succeeded
| Dec  1 13:29:42 rti10 kernel: ide1 at 0x170-0x177,0x376 on irq 15
| Dec  1 13:29:42 rti10 kernel: hda: host protected area => 1
| Dec  1 13:29:42 rti10 kernel: hda: 78165360 sectors (40021 MB) w/2048KiB
| Cache, CHS=4865/255/63, UDMA(100)
| Dec  1 13:29:42 rti10 kernel: hdc: ATAPI 54X CD-ROM drive, 128kB Cache,
| UDMA(33)
| Dec  1 13:29:42 rti10 apmd[1060]: Version 3.0.2 (APM BIOS 1.2, Linux driver
| 1.16)
| Dec  1 13:29:42 rti10 kernel: Uniform CD-ROM driver Revision: 3.12
| Dec  1 13:29:42 rti10 apmd: apmd startup succeeded
| Dec  1 13:29:42 rti10 kernel: ide-floppy driver 0.99.newide
| Dec  1 13:29:42 rti10 kernel: Partition check:
| Dec  1 13:29:42 rti10 aksusbd[1071]: loaded, daemon version: 1.5, key API
| (USB) version: 3.81, key API (parallel) version: 3.81
| Dec  1 13:29:42 rti10 aksusbd: aksusbd startup succeeded
| Dec  1 13:29:42 rti10 kernel:  hda: hda1 hda2 hda3
| Dec  1 13:29:42 rti10 kernel: ide-floppy driver 0.99.newide
| Dec  1 13:29:38 rti10 kudzu:  succeeded
| Dec  1 13:29:42 rti10 kernel: ohci1394: $Rev: 693 $ Ben Collins
| <bcollins@debian.org>
| Dec  1 13:29:38 rti10 kudzu: Updating /etc/fstab succeeded
| Dec  1 13:29:42 rti10 kernel: PCI: Found IRQ 10 for device 02:03.0
| Dec  1 13:29:38 rti10 sysctl: net.ipv4.ip_forward = 0
| Dec  1 13:29:42 rti10 kernel: ohci1394_0: OHCI-1394 1.0 (PCI): IRQ=[10]
| MMIO=[f2000000-f20007ff]  Max Packet=[2048]
| Dec  1 13:29:42 rti10 automount[1124]: starting automounter version 3.1.7,
| path = /home, maptype = file, mapname = /etc/auto.home
| Dec  1 13:29:42 rti10 autofs: automount startup succeeded
| Dec  1 13:29:38 rti10 sysctl: net.ipv4.conf.default.rp_filter = 1
| Dec  1 13:29:42 rti10 kernel: ieee1394: SelfID completion called outside of
| bus reset!
| Dec  1 13:29:38 rti10 sysctl: kernel.core_uses_pid = 1
| Dec  1 13:29:43 rti10 kernel: raw1394: /dev/raw1394 device initialized
| Dec  1 13:29:38 rti10 network: Setting network parameters:  succeeded
| Dec  1 13:29:43 rti10 kernel: usb.c: registered new driver usbdevfs
| Dec  1 13:29:38 rti10 network: Bringing up loopback interface:  succeeded
| 
| 
| /dev/hdc:
| 
|  Model=FX54++W, FwRev=U01L, SerialNo=
|  Config={ Fixed Removeable DTR<=5Mbs DTR>10Mbs nonMagnetic }
|  RawCHS=0/0/0, TrkSize=0, SectSize=0, ECCbytes=0
|  BuffType=unknown, BuffSize=128kB, MaxMultSect=0
|  (maybe): CurCHS=0/0/0, CurSects=0, LBA=yes, LBAsects=0
|  IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
|  PIO modes:  pio0 pio1 pio2 pio3 pio4
|  DMA modes:  sdma0 sdma1 sdma2 mdma0 mdma1 mdma2
|  UDMA modes: udma0 udma1 *udma2
|  AdvancedPM=no

--
~Randy
MOTD:  Always include version info.