Date: Thu, 27 Apr 2000 20:00:22 +0100 (BST)
From: Nick Hibma <n_hibma@calcaphon.com>
To: m.butkus@tu-bs.de
Cc: freebsd-gnats-submit@freebsd.org, jin@gracie.lbl.gov, freebsd-bugs@freebsd.org
Subject: Re: kern/16767: open("/dev/ppi0") will crash machine without sync disk
Message-ID: <Pine.BSF.4.20.0004271959270.14642-100000@localhost>
In-Reply-To: <200004271832.UAA26106@rzbcosv0.rz.tu-bs.de>
next in thread | previous in thread | raw e-mail | index | archive | help
The first part is ok, the second patch should be including the sys/bus.h
file I think.
If someone could assign the PR to me, I'll have a look soonish.
Nick
> ===================================================================
> RCS file: /home/ncvs/src/sys/dev/ppbus/ppi.c,v
> retrieving revision 1.21
> retrieving revision 1.22
> diff -p -u -r1.21 -r1.22
> - --- src/sys/dev/ppbus/ppi.c 2000/01/23 14:41:04 1.21
> +++ /home/ncvs/src/sys/dev/ppbus/ppi.c 2000/04/07 00:09:24 1.22
> @@ -23,7 +23,7 @@
> * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
> * SUCH DAMAGE.
> *
> - - * $FreeBSD: src/sys/dev/ppbus/ppi.c,v 1.21 2000/01/23 14:41:04 peter Exp $
> + * $FreeBSD: src/sys/dev/ppbus/ppi.c,v 1.22 2000/04/07 00:09:24 jlemon Exp $
> *
> */
> #include "opt_ppb_1284.h"
> @@ -173,6 +173,10 @@ ppi_attach(device_t dev)
> /* declare our interrupt handler */
> ppi->intr_resource = bus_alloc_resource(dev, SYS_RES_IRQ,
> &zero, irq, irq, 1, RF_ACTIVE);
> + if (ppi->intr_resource == NULL) {
> + device_printf(dev, "can't allocate irq\n");
> + return (ENOMEM);
> + }
>
> make_dev(&ppi_cdevsw, device_get_unit(dev), /* XXX cleanup */
> UID_ROOT, GID_WHEEL,
> >>>> snap >>>>
>
> But this does not really solve the problem, IMHO, since the interrupt
> allocation should not have failed in the first place. This way,
> the ppi device fails to attach completely. From dmesg(8):
>
> > ppc0: <Parallel port> at port 0x278-0x27f irq 5 on isa0
> > ppc0: SMC-like chipset (ECP/EPP/PS2/NIBBLE) in COMPATIBLE mode
> > ppc0: FIFO with 16/16/16 bytes threshold
> > ppi0: <Parallel I/O> on ppbus0
> > ppi0: can't allocate irq
> > device_probe_and_attach: ppi0 attach returned 12
>
> I have then tried changing RF_ACTIVE to RF_SHAREABLE.
> This flag is used in lpt.c for interrupt allocation
> (see sys/dev/ppbus/lpt.c, rev 1.15.2.1, line 460ff):
>
> /* retrieve the ppbus irq */
> BUS_READ_IVAR(ppbus, dev, PPBUS_IVAR_IRQ, &irq);
>
> if (irq > 0) {
> /* declare our interrupt handler */
> sc->intr_resource = bus_alloc_resource(dev, SYS_RES_IRQ,
> &zero, irq, irq, 1, RF_SHAREABLE);
> }
>
> I don't know whether this is the "right" thing to do because I
> don't quite understand the meaning of these flags. However, things
> seem to work now. My friend has tested the modified ppi driver
> with the example from /usr/share/examples/ppi/ppilcd.c (he actually
> has the LCD hardware).
>
> This is the diff of my kernel changes, relative to ppi.c rev 1.21.
> It also contains the NULL pointer check from rev 1.22.
>
> <<<< snip <<<<
> - --- sys/dev/ppbus/ppi.c.orig Wed Apr 26 22:08:37 2000
> +++ sys/dev/ppbus/ppi.c Wed Apr 26 22:14:31 2000
> @@ -172,7 +172,11 @@
>
> /* declare our interrupt handler */
> ppi->intr_resource = bus_alloc_resource(dev, SYS_RES_IRQ,
> - - &zero, irq, irq, 1, RF_ACTIVE);
> + &zero, irq, irq, 1, RF_SHAREABL
> E);
> + if (ppi->intr_resource == NULL) {
> + device_printf(dev, "can't allocate irq\n");
> + return (ENOMEM);
> + }
>
> make_dev(&ppi_cdevsw, device_get_unit(dev), /* XXX cleanup */
> UID_ROOT, GID_WHEEL,
> >>>> snap >>>>
>
> Also, I had to make the following changes to ppilcd to make it compile.
> This patch is relative to:
>
> $FreeBSD: src/share/examples/ppi/Makefile,v 1.2 1999/08/28 00:19:27 peter Exp
> $FreeBSD: src/share/examples/ppi/ppilcd.c,v 1.2 1999/08/28 00:19:28 peter Exp
>
>
> <<<< snip <<<<
> *** share/examples/ppi/Makefile.orig Thu Apr 27 11:15:36 2000
> - --- share/examples/ppi/Makefile Thu Apr 27 11:15:48 2000
> ***************
> *** 5,8 ****
> - --- 5,10 ----
> PROG= ppilcd
> NOMAN= YES
>
> + CFLAGS +=-I/sys/dev/ppbus
> +
> .include <bsd.prog.mk>
> *** share/examples/ppi/ppilcd.c.orig Thu Apr 27 11:15:28 2000
> - --- share/examples/ppi/ppilcd.c Thu Apr 27 11:17:08 2000
> ***************
> *** 14,19 ****
> - --- 14,21 ----
> #include <err.h>
> #include <sysexits.h>
>
> + typedef void *device_t;
> +
> /* XXX should be in <machine/> */
> #include "ppbconf.h"
> #include "ppi.h"
> ***************
> *** 89,95 ****
> int debuglevel = 0;
> int vflag = 0;
>
> ! void
> main(int argc, char *argv[])
> {
> extern char *optarg;
> - --- 91,97 ----
> int debuglevel = 0;
> int vflag = 0;
>
> ! int
> main(int argc, char *argv[])
> {
> extern char *optarg;
> >>>> snap >>>>
>
> In case it matters, this is the dmesg(8) output of my box:
>
> <<<< snip <<<<
> Copyright (c) 1992-2000 The FreeBSD Project.
> Copyright (c) 1982, 1986, 1989, 1991, 1993
> The Regents of the University of California. All rights reserved.
> FreeBSD 4.0-STABLE #9: Tue Apr 25 23:25:47 CEST 2000
> root@bagheera.ts.rz.tu-bs.de:/usr/src/sys/compile/BAGHEERA
> Timecounter "i8254" frequency 1193182 Hz
> CPU: Pentium/P55C (165.79-MHz 586-class CPU)
> Origin = "GenuineIntel" Id = 0x543 Stepping = 3
> Features=0x8001bf<FPU,VME,DE,PSE,TSC,MSR,MCE,CX8,MMX>
> real memory = 67108864 (65536K bytes)
> avail memory = 61960192 (60508K bytes)
> Preloaded elf kernel "kernel" at 0xc034b000.
> Preloaded userconfig_script "/boot/kernel.conf" at 0xc034b09c.
> Intel Pentium detected, installing workaround for F00F bug
> md0: Malloc disk
> npx0: <math processor> on motherboard
> npx0: INT 16 interface
> pcib0: <Host to PCI bridge> on motherboard
> pci0: <PCI bus> on pcib0
> isab0: <Intel 82371FB PCI to ISA bridge> at device 7.0 on pci0
> isa0: <ISA bus> on isab0
> atapci0: <Intel PIIX ATA controller> port 0xe800-0xe80f at device 7.1 on pci0
> ata0: at 0x1f0 irq 14 on atapci0
> ncr0: <ncr 53c810 fast10 scsi> port 0xe400-0xe4ff mem 0xe6000000-0xe60000ff irq 12 at device 9.0 on pci0
> ncr0: driver is using old-style compatability shims
> ed0: <NE2000 PCI Ethernet (RealTek 8029)> port 0xe000-0xe01f irq 10 at device 10.0 on pci0
> ed0: supplying EUI64: 00:00:21:ff:fe:48:36:b9
> ed0: address 00:00:21:48:36:b9, type NE2000 (16 bit)
> pci0: <ATI Mach64-GP graphics accelerator> at 12.0
> fdc0: <NEC 72065B or clone> at port 0x3f0-0x3f5,0x3f7 irq 6 drq 2 on isa0
> fdc0: FIFO enabled, 8 bytes threshold
> fd0: <1440-KB 3.5" drive> on fdc0 drive 0
> atkbdc0: <Keyboard controller (i8042)> at port 0x60,0x64 on isa0
> atkbd0: <AT Keyboard> irq 1 on atkbdc0
> vga0: <Generic ISA VGA> at port 0x3c0-0x3df iomem 0xa0000-0xbffff on isa0
> sc0: <System console> on isa0
> sc0: VGA <16 virtual consoles, flags=0x200>
> sio0 at port 0x3f8-0x3ff irq 4 flags 0x10 on isa0
> sio0: type 16550A
> sio1 at port 0x2f8-0x2ff irq 3 on isa0
> sio1: type 16550A
> ppc0: <Parallel port> at port 0x278-0x27f irq 5 on isa0
> ppc0: SMC-like chipset (ECP/EPP/PS2/NIBBLE) in COMPATIBLE mode
> ppc0: FIFO with 16/16/16 bytes threshold
> ppi0: <Parallel I/O> on ppbus0
> ppi0: can't allocate irq
> device_probe_and_attach: ppi0 attach returned 12
> lpt0: <Printer> on ppbus0
> lpt0: Interrupt-driven port
> plip0: <PLIP network interface> on ppbus0
> pcm0: <AD1816> at port 0x220-0x22f,0x388-0x38b,0x100-0x10f irq 7 drq 1,0 on isa0
> unknown0: <Analog Devices AD1815> at port 0x330-0x331 irq 9 on isa0
> unknown1: <Analog Devices AD1815> at port 0x200-0x207 on isa0
> ad0: 14664MB <IBM-DJNA-351520> [29795/16/63] at ata0-master using WDMA2
> Waiting 5 seconds for SCSI devices to settle
> sa0 at ncr0 bus 0 target 4 lun 0
> sa0: <HP HP35470A T603> Removable Sequential Access SCSI-2 device
> sa0: 5.000MB/s transfers (5.000MHz, offset 8)
> Mounting root from ufs:/dev/ad0s1a
> cd0 at ncr0 bus 0 target 2 lun 0
> cd0: <SONY CD-ROM CDU-8002 1.8g> Removable CD-ROM SCSI-CCS device
> cd0: 3.300MB/s transfers
> cd0: Attempt to query device size failed: NOT READY, Vendor Specific ASC
> IP packet filtering initialized, divert disabled, rule-based forwarding disabled, default to deny, logging disabled
> >>>> snap >>>>
>
> It would be nice if someone commited this fix/workaround (or a better one)
> to -CURRENT and -STABLE. Thanks!
>
> Yours,
> Martin
>
>
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-bugs" in the body of the message
>
--
n_hibma@webweaving.org
n_hibma@freebsd.org USB project
http://www.etla.net/~n_hibma/
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-bugs" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.20.0004271959270.14642-100000>