* [PATCH] cx231xx-video: fix WARN about mismatched capabilities
@ 2019-08-15 19:53 Brad Love
2019-08-16 6:34 ` Hans Verkuil
0 siblings, 1 reply; 2+ messages in thread
From: Brad Love @ 2019-08-15 19:53 UTC (permalink / raw)
To: linux-media, hverkuil; +Cc: Brad Love
Since 049e684f2d ("media: v4l2-dev: fix WARN_ON(!vdev->device_caps)"),
if device_caps != capabilities a WARNING is dumped:
[ 6.887391] ------------[ cut here ]------------
[ 6.887422] WARNING: CPU: 0 PID: 988 at drivers/media/v4l2-core/v4l2-ioctl.c:1071 v4l_querycap+0xca/0xe0 [videodev]
[ 6.887423] Modules linked in: si2157 lgdt3306a cx231xx_dvb dvb_core videobuf2_vmalloc videobuf2_memops videobuf2_common cx231xx_alsa cx25840 edac_mce_amd cx231xx kvm videobuf_vmalloc tveeprom cx2341x videobuf_core crct10dif_pclmul v4l2_common i2c_mux videodev mc crc32_pclmul ghash_clmulni_intel aesni_intel aes_x86_64 crypto_simd snd_hda_codec_hdmi cryptd glue_helper wmi_bmof k10temp nls_iso8859_1 ir_rc5_decoder snd_hda_codec_realtek snd_hda_codec_generic ledtrig_audio rc_hauppauge mceusb asix rc_core usbnet mii snd_hda_intel snd_hda_codec snd_hda_core snd_hwdep snd_pcm snd_seq_midi snd_seq_midi_event snd_rawmidi snd_seq snd_seq_device snd_timer snd ccp soundcore sch_fq_codel mac_hid parport_pc ppdev lp parport ip_tables x_tables autofs4 vfio_pci vfio_virqfd irqbypass vfio_iommu_type1 vfio hid_generic usbhid hid nouveau mxm_wmi video i2c_algo_bit ttm drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops drm i2c_piix4 ahci libahci wmi gpio_amdpt gpio_generic
[ 6.887470] CPU: 0 PID: 988 Comm: v4l_id Not tainted 5.3.0-rc4+ #58
[ 6.887471] Hardware name: MSI MS-7A39/B350M GAMING PRO (MS-7A39), BIOS 2.G0 04/27/2018
[ 6.887478] RIP: 0010:v4l_querycap+0xca/0xe0 [videodev]
[ 6.887481] Code: af 6f c4 da 0f 0b 41 8b 84 24 b0 01 00 00 8b 53 54 0d 00 00 00 80 89 d1 21 c1 39 c8 74 bc 48 c7 c7 c0 f8 ac c0 e8 89 6f c4 da <0f> 0b 8b 53 54 eb a9 0f 1f 44 00 00 66 2e 0f 1f 84 00 00 00 00 00
[ 6.887483] RSP: 0018:ffffb694c168bc78 EFLAGS: 00010286
[ 6.887485] RAX: 0000000000000024 RBX: ffffb694c168bd88 RCX: 0000000000000000
[ 6.887486] RDX: 0000000000000000 RSI: ffff9844cea17388 RDI: ffff9844cea17388
[ 6.887487] RBP: ffffb694c168bca0 R08: 0000000000000001 R09: 000000000000039e
[ 6.887488] R10: 0000000000000020 R11: 000000000000039e R12: ffff9844ca685248
[ 6.887489] R13: 0000000000000000 R14: ffff9844c7912000 R15: ffffffffc0bdcbe0
[ 6.887491] FS: 00007f16beef0540(0000) GS:ffff9844cea00000(0000) knlGS:0000000000000000
[ 6.887492] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 6.887493] CR2: 00007f10a2372128 CR3: 0000000409a10000 CR4: 00000000003406f0
[ 6.887494] Call Trace:
[ 6.887503] __video_do_ioctl+0x24c/0x480 [videodev]
[ 6.887509] ? tomoyo_init_request_info+0x80/0x90
[ 6.887516] video_usercopy+0x280/0x5f0 [videodev]
[ 6.887522] ? v4l_s_fmt+0x660/0x660 [videodev]
[ 6.887530] video_ioctl2+0x15/0x20 [videodev]
[ 6.887535] v4l2_ioctl+0x4d/0x60 [videodev]
[ 6.887539] do_vfs_ioctl+0xa6/0x600
[ 6.887542] ? tomoyo_file_ioctl+0x19/0x20
[ 6.887545] ksys_ioctl+0x6d/0x80
[ 6.887547] __x64_sys_ioctl+0x1a/0x20
[ 6.887551] do_syscall_64+0x5a/0x130
[ 6.887555] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[ 6.887557] RIP: 0033:0x7f16bea0a5d7
[ 6.887560] Code: b3 66 90 48 8b 05 b1 48 2d 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 81 48 2d 00 f7 d8 64 89 01 48
[ 6.887561] RSP: 002b:00007ffdad4ea2b8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
[ 6.887562] RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007f16bea0a5d7
[ 6.887563] RDX: 00007ffdad4ea2c0 RSI: 0000000080685600 RDI: 0000000000000003
[ 6.887564] RBP: 0000557718bd4e70 R08: 0000000000000000 R09: 0000000000000000
[ 6.887565] R10: 0000000000000000 R11: 0000000000000246 R12: 00007ffdad4ea2c0
[ 6.887566] R13: 00007ffdad4ea420 R14: 0000000000000000 R15: 0000000000000000
[ 6.887568] ---[ end trace f385f48cdd44f9e9 ]---
Fix this for cx231xx-video by only adding appropriate capabilities for
a specific v4l2 device, based on their file handle type.
Verified working for video device, analog tuner, VBI, and radio.
Signed-off-by: Brad Love <brad@nextdimension.cc>
---
drivers/media/usb/cx231xx/cx231xx-video.c | 12 ++++++++----
1 file changed, 8 insertions(+), 4 deletions(-)
diff --git a/drivers/media/usb/cx231xx/cx231xx-video.c b/drivers/media/usb/cx231xx/cx231xx-video.c
index 45eed56137e1..643da9b930c5 100644
--- a/drivers/media/usb/cx231xx/cx231xx-video.c
+++ b/drivers/media/usb/cx231xx/cx231xx-video.c
@@ -1576,10 +1576,14 @@ int cx231xx_querycap(struct file *file, void *priv,
strscpy(cap->driver, "cx231xx", sizeof(cap->driver));
strscpy(cap->card, cx231xx_boards[dev->model].name, sizeof(cap->card));
usb_make_path(dev->udev, cap->bus_info, sizeof(cap->bus_info));
- cap->capabilities = V4L2_CAP_READWRITE |
- V4L2_CAP_VBI_CAPTURE | V4L2_CAP_VIDEO_CAPTURE |
- V4L2_CAP_STREAMING | V4L2_CAP_DEVICE_CAPS;
- if (video_is_registered(&dev->radio_dev))
+ cap->capabilities = V4L2_CAP_READWRITE | V4L2_CAP_STREAMING |
+ V4L2_CAP_DEVICE_CAPS;
+
+ if (fh->type == V4L2_BUF_TYPE_VIDEO_CAPTURE)
+ cap->capabilities |= V4L2_CAP_VIDEO_CAPTURE;
+ else if (fh->type == V4L2_BUF_TYPE_VBI_CAPTURE)
+ cap->capabilities |= V4L2_CAP_VBI_CAPTURE;
+ else if (fh->type == 0 && video_is_registered(&dev->radio_dev))
cap->capabilities |= V4L2_CAP_RADIO;
if (dev->tuner_type != TUNER_ABSENT)
cap->capabilities |= V4L2_CAP_TUNER;
--
2.21.0
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] cx231xx-video: fix WARN about mismatched capabilities
2019-08-15 19:53 [PATCH] cx231xx-video: fix WARN about mismatched capabilities Brad Love
@ 2019-08-16 6:34 ` Hans Verkuil
0 siblings, 0 replies; 2+ messages in thread
From: Hans Verkuil @ 2019-08-16 6:34 UTC (permalink / raw)
To: Brad Love, linux-media
On 8/15/19 9:53 PM, Brad Love wrote:
> Since 049e684f2d ("media: v4l2-dev: fix WARN_ON(!vdev->device_caps)"),
> if device_caps != capabilities a WARNING is dumped:
That's not what it tests!
The full check is:
/*
* Check that capabilities is a superset of
* vfd->device_caps | V4L2_CAP_DEVICE_CAPS
*/
WARN_ON((cap->capabilities &
(vfd->device_caps | V4L2_CAP_DEVICE_CAPS)) !=
(vfd->device_caps | V4L2_CAP_DEVICE_CAPS));
So in this case cap->capabilities is missing a capability that is
set in cap->device_caps.
cap->capabilities is supposed to be the union of the device_caps of
all video_device structs created by the driver.
So apparently there is at least one cap missing. Can you take a
look at which cap exactly is missing?
Does this happen only when the driver is loaded, or also when you
run 'v4l2-ctl -D' for the various devices the driver created?
Regards,
Hans
>
>
> [ 6.887391] ------------[ cut here ]------------
> [ 6.887422] WARNING: CPU: 0 PID: 988 at drivers/media/v4l2-core/v4l2-ioctl.c:1071 v4l_querycap+0xca/0xe0 [videodev]
> [ 6.887423] Modules linked in: si2157 lgdt3306a cx231xx_dvb dvb_core videobuf2_vmalloc videobuf2_memops videobuf2_common cx231xx_alsa cx25840 edac_mce_amd cx231xx kvm videobuf_vmalloc tveeprom cx2341x videobuf_core crct10dif_pclmul v4l2_common i2c_mux videodev mc crc32_pclmul ghash_clmulni_intel aesni_intel aes_x86_64 crypto_simd snd_hda_codec_hdmi cryptd glue_helper wmi_bmof k10temp nls_iso8859_1 ir_rc5_decoder snd_hda_codec_realtek snd_hda_codec_generic ledtrig_audio rc_hauppauge mceusb asix rc_core usbnet mii snd_hda_intel snd_hda_codec snd_hda_core snd_hwdep snd_pcm snd_seq_midi snd_seq_midi_event snd_rawmidi snd_seq snd_seq_device snd_timer snd ccp soundcore sch_fq_codel mac_hid parport_pc ppdev lp parport ip_tables x_tables autofs4 vfio_pci vfio_virqfd irqbypass vfio_iommu_type1 vfio hid_generic usbhid hid nouveau mxm_wmi video i2c_algo_bit ttm drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops drm i2c_piix4 ahci libahci wmi gpio_amdpt gpio_generic
> [ 6.887470] CPU: 0 PID: 988 Comm: v4l_id Not tainted 5.3.0-rc4+ #58
> [ 6.887471] Hardware name: MSI MS-7A39/B350M GAMING PRO (MS-7A39), BIOS 2.G0 04/27/2018
> [ 6.887478] RIP: 0010:v4l_querycap+0xca/0xe0 [videodev]
> [ 6.887481] Code: af 6f c4 da 0f 0b 41 8b 84 24 b0 01 00 00 8b 53 54 0d 00 00 00 80 89 d1 21 c1 39 c8 74 bc 48 c7 c7 c0 f8 ac c0 e8 89 6f c4 da <0f> 0b 8b 53 54 eb a9 0f 1f 44 00 00 66 2e 0f 1f 84 00 00 00 00 00
> [ 6.887483] RSP: 0018:ffffb694c168bc78 EFLAGS: 00010286
> [ 6.887485] RAX: 0000000000000024 RBX: ffffb694c168bd88 RCX: 0000000000000000
> [ 6.887486] RDX: 0000000000000000 RSI: ffff9844cea17388 RDI: ffff9844cea17388
> [ 6.887487] RBP: ffffb694c168bca0 R08: 0000000000000001 R09: 000000000000039e
> [ 6.887488] R10: 0000000000000020 R11: 000000000000039e R12: ffff9844ca685248
> [ 6.887489] R13: 0000000000000000 R14: ffff9844c7912000 R15: ffffffffc0bdcbe0
> [ 6.887491] FS: 00007f16beef0540(0000) GS:ffff9844cea00000(0000) knlGS:0000000000000000
> [ 6.887492] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 6.887493] CR2: 00007f10a2372128 CR3: 0000000409a10000 CR4: 00000000003406f0
> [ 6.887494] Call Trace:
> [ 6.887503] __video_do_ioctl+0x24c/0x480 [videodev]
> [ 6.887509] ? tomoyo_init_request_info+0x80/0x90
> [ 6.887516] video_usercopy+0x280/0x5f0 [videodev]
> [ 6.887522] ? v4l_s_fmt+0x660/0x660 [videodev]
> [ 6.887530] video_ioctl2+0x15/0x20 [videodev]
> [ 6.887535] v4l2_ioctl+0x4d/0x60 [videodev]
> [ 6.887539] do_vfs_ioctl+0xa6/0x600
> [ 6.887542] ? tomoyo_file_ioctl+0x19/0x20
> [ 6.887545] ksys_ioctl+0x6d/0x80
> [ 6.887547] __x64_sys_ioctl+0x1a/0x20
> [ 6.887551] do_syscall_64+0x5a/0x130
> [ 6.887555] entry_SYSCALL_64_after_hwframe+0x44/0xa9
> [ 6.887557] RIP: 0033:0x7f16bea0a5d7
> [ 6.887560] Code: b3 66 90 48 8b 05 b1 48 2d 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 81 48 2d 00 f7 d8 64 89 01 48
> [ 6.887561] RSP: 002b:00007ffdad4ea2b8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
> [ 6.887562] RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007f16bea0a5d7
> [ 6.887563] RDX: 00007ffdad4ea2c0 RSI: 0000000080685600 RDI: 0000000000000003
> [ 6.887564] RBP: 0000557718bd4e70 R08: 0000000000000000 R09: 0000000000000000
> [ 6.887565] R10: 0000000000000000 R11: 0000000000000246 R12: 00007ffdad4ea2c0
> [ 6.887566] R13: 00007ffdad4ea420 R14: 0000000000000000 R15: 0000000000000000
> [ 6.887568] ---[ end trace f385f48cdd44f9e9 ]---
>
>
> Fix this for cx231xx-video by only adding appropriate capabilities for
> a specific v4l2 device, based on their file handle type.
>
> Verified working for video device, analog tuner, VBI, and radio.
>
>
> Signed-off-by: Brad Love <brad@nextdimension.cc>
> ---
> drivers/media/usb/cx231xx/cx231xx-video.c | 12 ++++++++----
> 1 file changed, 8 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/media/usb/cx231xx/cx231xx-video.c b/drivers/media/usb/cx231xx/cx231xx-video.c
> index 45eed56137e1..643da9b930c5 100644
> --- a/drivers/media/usb/cx231xx/cx231xx-video.c
> +++ b/drivers/media/usb/cx231xx/cx231xx-video.c
> @@ -1576,10 +1576,14 @@ int cx231xx_querycap(struct file *file, void *priv,
> strscpy(cap->driver, "cx231xx", sizeof(cap->driver));
> strscpy(cap->card, cx231xx_boards[dev->model].name, sizeof(cap->card));
> usb_make_path(dev->udev, cap->bus_info, sizeof(cap->bus_info));
> - cap->capabilities = V4L2_CAP_READWRITE |
> - V4L2_CAP_VBI_CAPTURE | V4L2_CAP_VIDEO_CAPTURE |
> - V4L2_CAP_STREAMING | V4L2_CAP_DEVICE_CAPS;
> - if (video_is_registered(&dev->radio_dev))
> + cap->capabilities = V4L2_CAP_READWRITE | V4L2_CAP_STREAMING |
> + V4L2_CAP_DEVICE_CAPS;
> +
> + if (fh->type == V4L2_BUF_TYPE_VIDEO_CAPTURE)
> + cap->capabilities |= V4L2_CAP_VIDEO_CAPTURE;
> + else if (fh->type == V4L2_BUF_TYPE_VBI_CAPTURE)
> + cap->capabilities |= V4L2_CAP_VBI_CAPTURE;
> + else if (fh->type == 0 && video_is_registered(&dev->radio_dev))
> cap->capabilities |= V4L2_CAP_RADIO;
> if (dev->tuner_type != TUNER_ABSENT)
> cap->capabilities |= V4L2_CAP_TUNER;
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2019-08-16 6:34 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-08-15 19:53 [PATCH] cx231xx-video: fix WARN about mismatched capabilities Brad Love
2019-08-16 6:34 ` Hans Verkuil
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).