I agree too... I tested out both patches Mauro and trents patches. I did not get any ops, and tested the patches for about three hours. I did not have the device removed, so I would assume it the dummy device definition that's causing the problem.
I vote that we should get rid of the dummy device. It's always caused a great deal of headaches. We should probably set it up like the saa7134 devices.
I vote that we should get rid of the dummy device. It's always caused a great deal of headaches. We should probably set it up like the saa7134 devices.
----- Original Message ----
From: Thierry MERLE <thierry.merle@xxxxxxx>
To: Mauro Carvalho Chehab <mchehab@xxxxxxxxxxxxx>
Cc: Dwaine Garden <dwainegarden@xxxxxxxxxx>; Linux and Kernel Video <video4linux-list@xxxxxxxxxx>; Linux DVB <linux-dvb@xxxxxxxxxxx>
Sent: Saturday, April 14, 2007 1:26:19 PM
Subject: Re: [Patch] USBVision - Fix NT1005 Bridge detection
Mauro Carvalho Chehab a écrit :
> Hi Dwaine and Thierry,
>
>
> The patch seems to be ok. However, it is not nice to do this:
>
>
>> + if ((!strcmp(usbvision_device_data[model].ModelString, "Dazzle
>> Fusion Model DVC-90 Rev 1 (SECAM)")) &&
>>
>
> Since, if for some reason, people decide to change the comment at
> usbvision-cards, it will break the above code. Instead, it is better to
> create a code for each board and for its value.
>
> I've did such change on:
> http://linuxtv.org/hg/~mchehab/usbvision
>
> It also honors CodingStyle max line size.
>
> I've also removed some duplicated stuff (ID Vendor/Product) and changed
> some flags to use bitmap. As result, It saved about 1K of kernelspace
> area.
>
> I also noticed that the same name string were used for more than one
> board. So, I've added a number for the duplicated names to avoid having
> the same name for two different boards. It would be better, however, to
> replace the duplicated definitions to a proper one, or to merge the
> entries, if they are just referring to the same board.
>
> May you test it and comment?
>
>
Acked-by: Thierry MERLE <thierry.merle@xxxxxxx>
I agree totally with this simplification.
I am currently debugging the Trent's patch that makes a kernel oops and
I don't know why.
The usbvision_probe is called event if the device is not plugged, and
the kernel oops occurs on this piece of code:
model = devid->driver_info;
printk(KERN_INFO "%s: %s found\n", __FUNCTION__,
usbvision_device_data[model].ModelString);
Memory alignment problem ? Why this probe is called even if there is no
device plugged-in ?
By putting something like:
model = devid->driver_info;
if(model<0 || model >HPG_WINTV) {
printk(KERN_INFO "model out of bounds %d\n",model);
return -ENODEV;
} else printk(KERN_INFO "model is %d\n",model);
The problem disappeared...
Regards,
Thierry
Apr 14 18:17:10 localhost kernel: BUG: unable to handle kernel paging
request at virtual address 0917ab30
Apr 14 18:17:10 localhost kernel: printing eip:
Apr 14 18:17:10 localhost kernel: c01d837b
Apr 14 18:17:10 localhost kernel: *pde = 00000000
Apr 14 18:17:10 localhost kernel: Oops: 0000 [#1]
Apr 14 18:17:10 localhost kernel: Modules linked in: usbvision videodev
tuner v4l2_common snd_pcm_oss snd_mixer_oss cpufreq_conservative
cinergyT2 dvb_core
mga drm ns558 analog joydev nvidia(P) snd_hda_intel snd_hda_codec
snd_pcm snd_timer snd snd_page_alloc
Apr 14 18:17:10 localhost kernel: CPU: 0
Apr 14 18:17:10 localhost kernel: EIP: 0060:[<c01d837b>] Tainted:
P VLI
Apr 14 18:17:10 localhost kernel: EFLAGS: 00210097 (2.6.21-rc6 #7)
Apr 14 18:17:10 localhost kernel: EIP is at vsnprintf+0x2d1/0x4d3
Apr 14 18:17:10 localhost kernel: eax: 0917ab30 ebx: c03f01b4 ecx:
0917ab30 edx: fffffffe
Apr 14 18:17:10 localhost kernel: esi: c75ebe7c edi: 0000000a ebp:
ffffffff esp: c75ebd98
Apr 14 18:17:10 localhost kernel: ds: 007b es: 007b fs: 00d8 gs:
0033 ss: 0068
Apr 14 18:17:10 localhost kernel: Process modprobe (pid: 4759,
ti=c75ea000 task=e24e6520 task.ti=c75ea000)
Apr 14 18:17:10 localhost kernel: Stack: e7dde000 ead775e0 584da500
ead775e0 c03ef3e0 ead775e0 ffffffff ffffffff
Apr 14 18:17:10 localhost kernel: 00000000 c03f05a0 00000400
c03f01a0 ed0edf6f 00000400 eb9c6400 c03f01a0
Apr 14 18:17:10 localhost kernel: ea412fa0 c01d858f c75ebe74
e49c9b70 c0117ec8 c75ebe74 ebc3fc18 00000001
Apr 14 18:17:10 localhost kernel: Call Trace:
Apr 14 18:17:10 localhost kernel: [<c01d858f>] vscnprintf+0x12/0x1d
Apr 14 18:17:10 localhost kernel: [<c0117ec8>] vprintk+0x4d/0x1f3
Apr 14 18:17:10 localhost kernel: [<c0177830>] __sysfs_new_dirent+0x12/0x40
Apr 14 18:17:10 localhost kernel: [<c017791a>]
__sysfs_make_dirent+0x11/0x6c
Apr 14 18:17:10 localhost kernel: [<c0117e78>] printk+0x16/0x19
Apr 14 18:17:10 localhost kernel: [<ed0eaa40>]
usbvision_probe+0x9a/0x3f6 [usbvision]
Apr 14 18:17:10 localhost kernel: [<c02665e2>]
usb_probe_interface+0x66/0x88
Apr 14 18:17:10 localhost kernel: [<c022fc9c>] really_probe+0x9c/0x12e
Apr 14 18:17:10 localhost kernel: [<c0266893>] usb_match_id+0x34/0x42
Apr 14 18:17:10 localhost kernel: [<c022fde0>]
driver_probe_device+0xa3/0xaf
Apr 14 18:17:10 localhost kernel: [<c02e78f6>] klist_next+0x48/0x6e
Apr 14 18:17:10 localhost kernel: [<c022fe52>] __driver_attach+0x0/0x75
Apr 14 18:17:10 localhost kernel: [<c022fe96>] __driver_attach+0x44/0x75
Apr 14 18:17:10 localhost kernel: [<c022f304>] bus_for_each_dev+0x35/0x59
Apr 14 18:17:10 localhost kernel: [<c022fedb>] driver_attach+0x14/0x16
Apr 14 18:17:10 localhost kernel: [<c022fe52>] __driver_attach+0x0/0x75
Apr 14 18:17:10 localhost kernel: [<c022f748>] bus_add_driver+0x5a/0xe0
Apr 14 18:17:10 localhost kernel: [<c0266cc1>]
usb_register_driver+0x65/0xb8
Apr 14 18:17:10 localhost kernel: [<eca5b11b>]
usbvision_init+0x11b/0x161 [usbvision]
Apr 14 18:17:10 localhost kernel: [<c012f0b7>] sys_init_module+0x89/0x133
Apr 14 18:17:10 localhost kernel: [<c0103ac0>] syscall_call+0x7/0xb
Apr 14 18:17:10 localhost kernel: =======================
Apr 14 18:17:10 localhost kernel: Code: 01 00 00 3b 5c 24 24 73 f0 c6 03
20 eb eb 89 f0 83 c6 04 8b 54 24 1c 8b 08 b8 c2 f7 34 c0 81 f9 ff 0f 00
00 0f 46 c8 89 c8 eb 06 <80> 38 00 74 07 40 4a 83 fa ff 75 f4 29 c8 f6
44 24 20 10 89 c7
Apr 14 18:17:10 localhost kernel: EIP: [<c01d837b>]
vsnprintf+0x2d1/0x4d3 SS:ESP 0068:c75ebd98
_______________________________________________
linux-dvb mailing list
linux-dvb@xxxxxxxxxxx
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
From: Thierry MERLE <thierry.merle@xxxxxxx>
To: Mauro Carvalho Chehab <mchehab@xxxxxxxxxxxxx>
Cc: Dwaine Garden <dwainegarden@xxxxxxxxxx>; Linux and Kernel Video <video4linux-list@xxxxxxxxxx>; Linux DVB <linux-dvb@xxxxxxxxxxx>
Sent: Saturday, April 14, 2007 1:26:19 PM
Subject: Re: [Patch] USBVision - Fix NT1005 Bridge detection
Mauro Carvalho Chehab a écrit :
> Hi Dwaine and Thierry,
>
>
> The patch seems to be ok. However, it is not nice to do this:
>
>
>> + if ((!strcmp(usbvision_device_data[model].ModelString, "Dazzle
>> Fusion Model DVC-90 Rev 1 (SECAM)")) &&
>>
>
> Since, if for some reason, people decide to change the comment at
> usbvision-cards, it will break the above code. Instead, it is better to
> create a code for each board and for its value.
>
> I've did such change on:
> http://linuxtv.org/hg/~mchehab/usbvision
>
> It also honors CodingStyle max line size.
>
> I've also removed some duplicated stuff (ID Vendor/Product) and changed
> some flags to use bitmap. As result, It saved about 1K of kernelspace
> area.
>
> I also noticed that the same name string were used for more than one
> board. So, I've added a number for the duplicated names to avoid having
> the same name for two different boards. It would be better, however, to
> replace the duplicated definitions to a proper one, or to merge the
> entries, if they are just referring to the same board.
>
> May you test it and comment?
>
>
Acked-by: Thierry MERLE <thierry.merle@xxxxxxx>
I agree totally with this simplification.
I am currently debugging the Trent's patch that makes a kernel oops and
I don't know why.
The usbvision_probe is called event if the device is not plugged, and
the kernel oops occurs on this piece of code:
model = devid->driver_info;
printk(KERN_INFO "%s: %s found\n", __FUNCTION__,
usbvision_device_data[model].ModelString);
Memory alignment problem ? Why this probe is called even if there is no
device plugged-in ?
By putting something like:
model = devid->driver_info;
if(model<0 || model >HPG_WINTV) {
printk(KERN_INFO "model out of bounds %d\n",model);
return -ENODEV;
} else printk(KERN_INFO "model is %d\n",model);
The problem disappeared...
Regards,
Thierry
Apr 14 18:17:10 localhost kernel: BUG: unable to handle kernel paging
request at virtual address 0917ab30
Apr 14 18:17:10 localhost kernel: printing eip:
Apr 14 18:17:10 localhost kernel: c01d837b
Apr 14 18:17:10 localhost kernel: *pde = 00000000
Apr 14 18:17:10 localhost kernel: Oops: 0000 [#1]
Apr 14 18:17:10 localhost kernel: Modules linked in: usbvision videodev
tuner v4l2_common snd_pcm_oss snd_mixer_oss cpufreq_conservative
cinergyT2 dvb_core
mga drm ns558 analog joydev nvidia(P) snd_hda_intel snd_hda_codec
snd_pcm snd_timer snd snd_page_alloc
Apr 14 18:17:10 localhost kernel: CPU: 0
Apr 14 18:17:10 localhost kernel: EIP: 0060:[<c01d837b>] Tainted:
P VLI
Apr 14 18:17:10 localhost kernel: EFLAGS: 00210097 (2.6.21-rc6 #7)
Apr 14 18:17:10 localhost kernel: EIP is at vsnprintf+0x2d1/0x4d3
Apr 14 18:17:10 localhost kernel: eax: 0917ab30 ebx: c03f01b4 ecx:
0917ab30 edx: fffffffe
Apr 14 18:17:10 localhost kernel: esi: c75ebe7c edi: 0000000a ebp:
ffffffff esp: c75ebd98
Apr 14 18:17:10 localhost kernel: ds: 007b es: 007b fs: 00d8 gs:
0033 ss: 0068
Apr 14 18:17:10 localhost kernel: Process modprobe (pid: 4759,
ti=c75ea000 task=e24e6520 task.ti=c75ea000)
Apr 14 18:17:10 localhost kernel: Stack: e7dde000 ead775e0 584da500
ead775e0 c03ef3e0 ead775e0 ffffffff ffffffff
Apr 14 18:17:10 localhost kernel: 00000000 c03f05a0 00000400
c03f01a0 ed0edf6f 00000400 eb9c6400 c03f01a0
Apr 14 18:17:10 localhost kernel: ea412fa0 c01d858f c75ebe74
e49c9b70 c0117ec8 c75ebe74 ebc3fc18 00000001
Apr 14 18:17:10 localhost kernel: Call Trace:
Apr 14 18:17:10 localhost kernel: [<c01d858f>] vscnprintf+0x12/0x1d
Apr 14 18:17:10 localhost kernel: [<c0117ec8>] vprintk+0x4d/0x1f3
Apr 14 18:17:10 localhost kernel: [<c0177830>] __sysfs_new_dirent+0x12/0x40
Apr 14 18:17:10 localhost kernel: [<c017791a>]
__sysfs_make_dirent+0x11/0x6c
Apr 14 18:17:10 localhost kernel: [<c0117e78>] printk+0x16/0x19
Apr 14 18:17:10 localhost kernel: [<ed0eaa40>]
usbvision_probe+0x9a/0x3f6 [usbvision]
Apr 14 18:17:10 localhost kernel: [<c02665e2>]
usb_probe_interface+0x66/0x88
Apr 14 18:17:10 localhost kernel: [<c022fc9c>] really_probe+0x9c/0x12e
Apr 14 18:17:10 localhost kernel: [<c0266893>] usb_match_id+0x34/0x42
Apr 14 18:17:10 localhost kernel: [<c022fde0>]
driver_probe_device+0xa3/0xaf
Apr 14 18:17:10 localhost kernel: [<c02e78f6>] klist_next+0x48/0x6e
Apr 14 18:17:10 localhost kernel: [<c022fe52>] __driver_attach+0x0/0x75
Apr 14 18:17:10 localhost kernel: [<c022fe96>] __driver_attach+0x44/0x75
Apr 14 18:17:10 localhost kernel: [<c022f304>] bus_for_each_dev+0x35/0x59
Apr 14 18:17:10 localhost kernel: [<c022fedb>] driver_attach+0x14/0x16
Apr 14 18:17:10 localhost kernel: [<c022fe52>] __driver_attach+0x0/0x75
Apr 14 18:17:10 localhost kernel: [<c022f748>] bus_add_driver+0x5a/0xe0
Apr 14 18:17:10 localhost kernel: [<c0266cc1>]
usb_register_driver+0x65/0xb8
Apr 14 18:17:10 localhost kernel: [<eca5b11b>]
usbvision_init+0x11b/0x161 [usbvision]
Apr 14 18:17:10 localhost kernel: [<c012f0b7>] sys_init_module+0x89/0x133
Apr 14 18:17:10 localhost kernel: [<c0103ac0>] syscall_call+0x7/0xb
Apr 14 18:17:10 localhost kernel: =======================
Apr 14 18:17:10 localhost kernel: Code: 01 00 00 3b 5c 24 24 73 f0 c6 03
20 eb eb 89 f0 83 c6 04 8b 54 24 1c 8b 08 b8 c2 f7 34 c0 81 f9 ff 0f 00
00 0f 46 c8 89 c8 eb 06 <80> 38 00 74 07 40 4a 83 fa ff 75 f4 29 c8 f6
44 24 20 10 89 c7
Apr 14 18:17:10 localhost kernel: EIP: [<c01d837b>]
vsnprintf+0x2d1/0x4d3 SS:ESP 0068:c75ebd98
_______________________________________________
linux-dvb mailing list
linux-dvb@xxxxxxxxxxx
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
_______________________________________________ linux-dvb mailing list linux-dvb@xxxxxxxxxxx http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
