Date: Tue, 10 Feb 2009 16:12:57 -0500 From: John Baldwin <jhb@freebsd.org> To: "Alexey Shuvaev" <shuvaev@physik.uni-wuerzburg.de> Cc: freebsd-current@freebsd.org Subject: Re: lpt stopped working Message-ID: <200902101612.57922.jhb@freebsd.org> In-Reply-To: <20090210201447.GA1664@wep4035.physik.uni-wuerzburg.de> References: <200902021643.39862.c47g@gmx.at> <200902091458.41637.jhb@freebsd.org> <20090210201447.GA1664@wep4035.physik.uni-wuerzburg.de>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tuesday 10 February 2009 3:14:47 pm Alexey Shuvaev wrote:
> On Mon, Feb 09, 2009 at 02:58:41PM -0500, John Baldwin wrote:
> > On Monday 09 February 2009 1:58:39 pm Christian Gusenbauer wrote:
> > > On Monday 09 February 2009, John Baldwin wrote:
> > > >
> > > > Please help to debug this so we can have working lpt0 in 8.0. No one
> > > > tested the patches months ago when I first posted them, and if folks
do
> > not
> > > > test them now I will simply remove the driver before 8.0 ships. I no
> > >
> > > Mea culpa, too. As you sent your patches, I thought someone else will do
the
> > > tests surely ... :-(.
> > >
> > > > longer have any hardware such that I can test this directly, so I am
> > > > depending on folks to test things I have asked for and report back. I
> > > > believe the last thing I asked for was for someone to do this when
they
> > lpt
> > > > was hung:
> > > >
> > > > Ok, can you run kgdb against your running kernel (Just run 'kgdb'
without
> > > > any arguments) and do the following:
> > > >
> > > > (kgdb) p *(struct ppb_data *)ppbus_devclass->devices[0]->softc
> > > >
> > > > Assuming the ppb_owner is not 0, can you then do this:
> > > >
> > > > (kgdb) p *(device_t)((struct ppb_data
> > > > *)ppbus_devclass->devices[0]->softc)->ppb_owner
> > >
> > > This is the output (unfortunately ppb_owner IS 0):
> > >
> > > (kgdb) p *(struct ppb_data *)ppbus_devclass->devices[0]->softc
> > > $1 = {class_id = 10, state = 1, error = 0, mode = 0, ppb_owner = 0x0,
> > > ppc_lock = 0xc56bfe7c, ppc_irq_res = 0xc573d5c0}
> >
> > And this is while lpd or the like is hung trying to write to /dev/lpt0?
> >
> Hello all!
>
> Ok, here we go. 1st, the system:
> ~> uname -a
> FreeBSD wep4035 8.0-CURRENT FreeBSD 8.0-CURRENT #0: Sat Feb 7 20:25:10 CET
2009 root@wep4035:/usr/obj/usr/src/sys/NOUSB amd64
>
> Parallel port (from dmesg):
> ppc0: <Parallel port> port 0x378-0x37f,0x778-0x77f irq 7 drq 3 on acpi0
> ppc0: SMC-like chipset (ECP/EPP/PS2/NIBBLE) in COMPATIBLE mode
> ppc0: FIFO with 16/16/9 bytes threshold
> ppc0: [ITHREAD]
> ppbus0: <Parallel port bus> on ppc0
> ppbus0: IEEE1284 device found /NIBBLE/ECP
> ppbus0: Probing for PnP devices:
> ppbus0: <HEWLETT-PACKARD OFFICEJET R60> PRINTER MLC,PCL,PML,SCL
> plip0: <PLIP network interface> on ppbus0
> plip0: [ITHREAD]
> lpt0: <Printer> on ppbus0
> lpt0: [ITHREAD]
> lpt0: Interrupt-driven port
> ppi0: <Parallel I/O> on ppbus0
>
> I have simplified things and do not run lpd. The command is:
> 'ktrace -id cat monitor_info > /dev/lpt0'
> (I think it should work?)
> kdump -E > ktrace.dump does not show anything interesting:
> [snip]
> 1564 cat 0.001120 CALL open(0x7fffffffee1c,O_RDONLY,<unused>0x6d)
> 1564 cat 0.001127 NAMI "monitor_info"
> 1564 cat 0.001143 RET open 3
> 1564 cat 0.001149 CALL fstat(0x1,0x7fffffffeac0)
> 1564 cat 0.001155 STRU struct stat {dev=83951360, ino=53,
mode=crw-rw---
> - , nlink=1, uid=0, gid=0, rdev=53, atime=1234294743.332164000,
stime=1234294743
> .332164000, ctime=1234294974, birthtime=-1, size=0, blksize=4096, blocks=0,
flag
> s=0x0 }
> 1564 cat 0.001160 RET fstat 0
> 1564 cat 0.001180 CALL
__sysctl(0x7fffffffea20,0x2,0x80085eeb8,0x7ffffff
> fea18,0,0)
> 1564 cat 0.001187 RET __sysctl 0
> 1564 cat 0.001191 CALL
__sysctl(0x7fffffffea60,0x2,0x7fffffffea7c,0x7fff
> ffffea70,0,0)
> 1564 cat 0.001198 RET __sysctl 0
> 1564 cat 0.001203 CALL
__sysctl(0x7fffffffea60,0x2,0x7fffffffea7c,0x7fffffffea70,0,0)
> 1564 cat 0.001208 RET __sysctl 0
> 1564 cat 0.001230 CALL
__sysctl(0x7fffffffe5f0,0x2,0x8008509e8,0x7fffffffe5e8,0,0)
> 1564 cat 0.001236 RET __sysctl 0
> 1564 cat 0.001242 CALL readlink(0x800722639,0x7fffffffe610,0x400)
> 1564 cat 0.001248 NAMI "/etc/malloc.conf"
> 1564 cat 0.001264 RET readlink -1 errno 2 No such file or directory
> 1564 cat 0.001270 CALL issetugid
> 1564 cat 0.001275 RET issetugid 0
> 1564 cat 0.001296 CALL break(0x600000)
> 1564 cat 0.001302 RET break 0
> 1564 cat 0.001317 CALL
__sysctl(0x7fffffffe850,0x2,0x7fffffffe86c,0x7fffffffe860,0,0)
> 1564 cat 0.001324 RET __sysctl 0
> 1564 cat 0.001329 CALL mmap(0,0x100000,PROT_READ|
PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0)
> 1564 cat 0.001335 RET mmap 8790016/0x800862000
> 1564 cat 0.001340 CALL mmap(0x800962000,0x9e000,PROT_READ|
PROT_WRITE,MAP_PRIVATE|MAP_ANON,0xffffffff,0)
> 1564 cat 0.001346 RET mmap 9838592/0x800962000
> 1564 cat 0.001351 CALL munmap(0x800862000,0x9e000)
> 1564 cat 0.001359 RET munmap 0
> 1564 cat 0.001378 CALL read(0x3,0x800902000,0x1000)
> 1564 cat 0.006845 GIO fd 3 read 4096 bytes
> "(II) VESA(0): VESA VBE DDC supported
> (II) VESA(0): VESA VBE DDC Level 2
> (II) VESA(0): VESA VBE DDC transfer in appr. 1 sec.
> (==) VESA(0): Write-combining range (0x0,0x1000) was already clear
> (==) VESA(0): Write-combining range (0x0,0x1000) was already clear
> [snip]
> 1 604 625 +hsync +vsync (46.9 kHz)
> (II) VESA(0): Modeline "800x600"x0.0 50.00 800 856 976 1040 600
63\
> 7 643 666 +hsync +vsync (48.1 kHz)
> (II) VESA(0): Modeline "1280x1024"x60.0 108.88 1280 1360 1496 1712
\
> 1024 1025 1028 1060 -hsync +vsync (63.6 kHz)
> (II) VESA(0): Modeline "1680x1050"x60.0 147.14 16"
> 1564 cat 0.006879 RET read 4096/0x1000
> 1564 cat 0.006888 CALL write(0x1,0x800902000,0x1000)
> 1564 cat 114.695563 RET write RESTART
> 1564 cat 114.695637 PSIG SIGINT SIG_DFL
>
> After 114 seconds I have hit Ctrl-C.
>
> And this is from kgdb:
> (kgdb) p *(struct ppb_data *)ppbus_devclass->devices[0]->softc
> $1 = {class_id = 0, state = 1, error = 0, mode = 0,
> ppb_owner = 0xffffff0004668700, ppc_lock = 0xffffff0004668eb8,
> ppc_irq_res = 0xffffff0004677900}
> (kgdb) p *(device_t)((struct
ppb_data*)ppbus_devclass->devices[0]->softc)->ppb
> _owner
> $2 = {ops = 0xffffff0001520000, link = {tqe_next = 0xffffff0004668500,
> tqe_prev = 0xffffff0004668908}, devlink = {tqe_next =
0xffffff0004668500,
> tqe_prev = 0xffffff0004668918}, parent = 0xffffff0004669000, children =
{
> tqh_first = 0x0, tqh_last = 0xffffff0004668730},
> driver = 0xffffffff806961c0, devclass = 0xffffff00014f4900, unit = 0,
> nameunit = 0xffffff0004666940 "lpt0", desc = 0xffffffff804fc110 "Printer",
> busy = 0, state = DS_ATTACHED, devflags = 0, flags = 7, order = 0 '\0',
> pad = 0 '\0', ivars = 0xffffff0004668800, softc = 0xffffff0004668400,
> sysctl_ctx = {tqh_first = 0xffffff0004674180,
> tqh_last = 0xffffff0004674288}, sysctl_tree = 0xffffff000467d480}
> (kgdb)
>
> The driver stays in this state even after exit of 'cat' process
> (I think this was already reported):
> ~> cat monitor_info > /dev/lpt0
> /dev/lpt0: Device busy.
Ok, so the first cat works, the second one gets EBUSY? Can you see if the
first 'cat' process is still around?
Hmm, I think I've found it. Due to a bug, lptclose() wasn't releasing the
bus.
--- //depot/user/jhb/acpipci/dev/ppbus/lpt.c
+++ /home/jhb/work/p4/acpipci/dev/ppbus/lpt.c
@@ -611,11 +611,8 @@
int err;
ppb_lock(ppbus);
- if (sc->sc_flags & LP_BYPASS) {
- sc->sc_state = 0;
- ppb_unlock(ppbus);
+ if (sc->sc_flags & LP_BYPASS)
goto end_close;
- }
if ((err = lpt_request_ppbus(lptdev, PPB_WAIT|PPB_INTR)) != 0) {
ppb_unlock(ppbus);
@@ -635,16 +632,16 @@
sc->sc_state &= ~OPEN;
callout_stop(&sc->sc_timer);
ppb_wctr(ppbus, LPC_NINIT);
- sc->sc_state = 0;
- sc->sc_xfercnt = 0;
/*
* unregistration of interrupt forced by release
*/
lpt_release_ppbus(lptdev);
- ppb_unlock(ppbus);
end_close:
+ sc->sc_state = 0;
+ sc->sc_xfercnt = 0;
+ ppb_unlock(ppbus);
lprintf(("closed.\n"));
return(0);
}
--
John Baldwin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200902101612.57922.jhb>