public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* Re: Fw: Oops at bttv_risc_packed (2.6.8-rc1)
       [not found] <20040804212106.04f19bad.akpm@osdl.org>
@ 2004-08-17 14:22 ` Gerd Knorr
  2004-09-08  7:37   ` Duncan Sands
  0 siblings, 1 reply; 2+ messages in thread
From: Gerd Knorr @ 2004-08-17 14:22 UTC (permalink / raw)
  To: Andrew Morton; +Cc: Duncan Sands, linux-kernel

> DEBUG_PAGEALLOC

Hmm, amd64 hasn't this one?

>  [pg0+543078197/1069322240] bttv_buffer_risc+0x4b5/0x5b0 [bttv]

Looks like the buffer for the risc instructions overflows.  No idea why
through, the size calculation looks ok.  Can you please apply the debug
patch below, load bttv with "bttv_debug=1" insmod option and try again?

What app triggers this btw, and which capture parameters (size, format)?

  Gerd

Index: bttv-risc.c
===================================================================
RCS file: /home/cvsroot/video4linux/bttv-risc.c,v
retrieving revision 1.3
diff -u -p -r1.3 bttv-risc.c
--- bttv-risc.c	6 Jul 2004 07:52:17 -0000	1.3
+++ bttv-risc.c	17 Aug 2004 14:17:14 -0000
@@ -55,6 +55,8 @@ bttv_risc_packed(struct bttv *btv, struc
 	instructions += 2;
 	if ((rc = btcx_riscmem_alloc(btv->c.pci,risc,instructions*8)) < 0)
 		return rc;
+	dprintk("bttv%d: risc packed: bpl %d lines %d instr %d size %d ptr %p\n",
+		btv->c.nr, bpl, lines, instructions, risc->size, risc->cpu);
 
 	/* sync instruction */
 	rp = risc->cpu;
@@ -99,8 +101,10 @@ bttv_risc_packed(struct bttv *btv, struc
 			offset += todo;
 		}
 		offset += padding;
+		dprintk("bttv%d: risc packed:   line %d ptr %p\n",
+			btv->c.nr, line, rp);
 	}
-	dprintk("bttv%d: risc planar: %d sglist elems\n", btv->c.nr, (int)(sg-sglist));
+	dprintk("bttv%d: risc packed: %d sglist elems\n", btv->c.nr, (int)(sg-sglist));
 
 	/* save pointer to jmp instruction address */
 	risc->jmp = rp;

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: Fw: Oops at bttv_risc_packed (2.6.8-rc1)
  2004-08-17 14:22 ` Fw: Oops at bttv_risc_packed (2.6.8-rc1) Gerd Knorr
@ 2004-09-08  7:37   ` Duncan Sands
  0 siblings, 0 replies; 2+ messages in thread
From: Duncan Sands @ 2004-09-08  7:37 UTC (permalink / raw)
  To: Gerd Knorr; +Cc: Andrew Morton, linux-kernel

[-- Attachment #1: Type: text/plain, Size: 3331 bytes --]

On Tuesday 17 August 2004 16:22, Gerd Knorr wrote:
> > DEBUG_PAGEALLOC
> 
> Hmm, amd64 hasn't this one?
> 
> >  [pg0+543078197/1069322240] bttv_buffer_risc+0x4b5/0x5b0 [bttv]
> 
> Looks like the buffer for the risc instructions overflows.  No idea why
> through, the size calculation looks ok.  Can you please apply the debug
> patch below, load bttv with "bttv_debug=1" insmod option and try again?
> 
> What app triggers this btw, and which capture parameters (size, format)?

Hi Gerd, I haven't been able to capture a new oops - but I continue to get
the usual hard hangs.  Here's the output before a hang, captured using
netconsole.  The app is xawtv, using PAL composite video.  I've attached
the xawtv verbose output (for a different run).  The time between starting and
hanging was around one minute.  I don't see anything strange...

All the best,

Duncan.

bttv: driver version 0.9.15 loaded
bttv: using 8 buffers with 2080k (520 pages) each for capture
bttv: Bt8xx card found (0).
bttv0: Bt878 (rev 17) at 0000:00:08.0, irq: 19, latency: 32, mmio: 0xdfbfe000
bttv0: detected: Hauppauge WinTV [card=10], PCI subsystem ID is 0070:13eb
bttv0: using: Hauppauge (bt878) [card=10,autodetected]
bttv0: Hauppauge/Voodoo msp34xx: reset line init [5]
tuner: chip found at addr 0xc2 i2c-bus bt878 #0 [sw]
bttv0: i2c attach [client=(tuner unset)]
bttv0: Hauppauge eeprom: model=44806, tuner=Philips FI1216MF MK2 (3), radio=no
bttv0: using tuner=3
tuner: type set to 3 (Philips (SECAM+PAL_BG) (FI1216MF, FM1216MF, FR1216MF)) by bt878 #0 [sw]
bttv0: i2c: checking for MSP34xx @ 0x80... not found
bttv0: i2c: checking for TDA9875 @ 0xb0... not found
bttv0: i2c: checking for TDA7432 @ 0x8a... not found
bttv0: registered device video0
bttv0: registered device vbi0
bttv0: PLL: 28636363 => 35468950 . ok
bttv0: PLL: no change required
switch_overlay: enter [new=dddf6360]
switch_overlay: done
bttv0: next set: top=dddf6360 bottom=dddf6360 [screen=dddf6360,irq=0,0]
switch_overlay: enter [new=dddf6578]
switch_overlay: old=dddf6360 state is 3
bttv0: next set: top=dddf6578 bottom=dddf6578 [screen=dddf6578,irq=0,0]
switch_overlay: done
switch_overlay: enter [new=dddf6254]
switch_overlay: old=dddf6578 state is 3
bttv0: next set: top=dddf6254 bottom=dddf6254 [screen=dddf6254,irq=0,0]
switch_overlay: done
switch_overlay: enter [new=dddf6ab4]
switch_overlay: old=dddf6254 state is 3
bttv0: next set: top=dddf6ab4 bottom=dddf6ab4 [screen=dddf6ab4,irq=0,0]
switch_overlay: done
switch_overlay: enter [new=dddf69a8]
switch_overlay: old=dddf6ab4 state is 3
bttv0: next set: top=dddf69a8 bottom=dddf69a8 [screen=dddf69a8,irq=0,0]
switch_overlay: done
switch_overlay: enter [new=dddf689c]
switch_overlay: old=dddf69a8 state is 3
bttv0: next set: top=dddf689c bottom=dddf689c [screen=dddf689c,irq=0,0]
switch_overlay: done
switch_overlay: enter [new=dddf6ab4]
switch_overlay: old=dddf689c state is 3
bttv0: next set: top=dddf6ab4 bottom=dddf6ab4 [screen=dddf6ab4,irq=0,0]
switch_overlay: done
switch_overlay: enter [new=dddf6254]
switch_overlay: old=dddf6ab4 state is 3
bttv0: next set: top=dddf6254 bottom=dddf6254 [screen=dddf6254,irq=0,0]
switch_overlay: done
switch_overlay: enter [new=dddf689c]
switch_overlay: old=dddf6254 state is 3
bttv0: next set: top=dddf689c bottom=dddf689c [screen=dddf689c,irq=0,0]
switch_overlay: done

...hung...

[-- Attachment #2: xawtv.log.gz --]
[-- Type: application/x-gzip, Size: 3924 bytes --]

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2004-09-08  7:37 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
     [not found] <20040804212106.04f19bad.akpm@osdl.org>
2004-08-17 14:22 ` Fw: Oops at bttv_risc_packed (2.6.8-rc1) Gerd Knorr
2004-09-08  7:37   ` Duncan Sands

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox