1. 01 Mar, 2016 7 commits
    • Mauro Carvalho Chehab's avatar
      [media] technisat-usb2: don't do DMA on the stack · 5d0f2df4
      Mauro Carvalho Chehab authored
      As warned by smatch:
      	drivers/media/usb/dvb-usb/technisat-usb2.c:263 technisat_usb2_set_led() error: doing dma on the stack (led)
      	drivers/media/usb/dvb-usb/technisat-usb2.c:280 technisat_usb2_set_led_timer() error: doing dma on the stack (&b)
      	drivers/media/usb/dvb-usb/technisat-usb2.c:341 technisat_usb2_identify_state() error: doing dma on the stack (version)
      	drivers/media/usb/dvb-usb/technisat-usb2.c:609 technisat_usb2_get_ir() error: doing dma on the stack (buf)
      	drivers/media/usb/dvb-usb/technisat-usb2.c:619 technisat_usb2_get_ir() error: doing dma on the stack (buf)
      
      Create a buffer at the device state and use it for all the DMA
      transfers.
      Signed-off-by: default avatarMauro Carvalho Chehab <mchehab@osg.samsung.com>
      5d0f2df4
    • Mauro Carvalho Chehab's avatar
      [media] dib0090: Do the right check for state->rf_ramp · 57bcbde9
      Mauro Carvalho Chehab authored
      Smatch with -pkernel --no-data keeps complaining about rf_ramp:
      	drivers/media/dvb-frontends/dib0090.c:1119 dib0090_pwm_gain_reset() error: we previously assumed 'state->rf_ramp' could be null (see line 1086)
      Signed-off-by: default avatarMauro Carvalho Chehab <mchehab@osg.samsung.com>
      57bcbde9
    • Mauro Carvalho Chehab's avatar
      [media] drxj: don't do math if not needed · b6554ea5
      Mauro Carvalho Chehab authored
      While there's no risk of divison by zero, the logic there is akward, as it
      does the calculus for the numerator and denominator before checking
      if this will be used.
      
      Change the order to check first if the denominator is zero, and only
      calculating the numerator/denominator if not.
      
      This should also avoid those smatch errors:
      	drivers/media/dvb-frontends/drx39xyj/drxj.c:9605 ctrl_get_qam_sig_quality() debug: sval_binop_unsigned: divide by zero
      	drivers/media/dvb-frontends/drx39xyj/drxj.c:9605 ctrl_get_qam_sig_quality() debug: sval_binop_unsigned: divide by zero
      	drivers/media/dvb-frontends/drx39xyj/drxj.c:9605 ctrl_get_qam_sig_quality() debug: sval_binop_unsigned: divide by zero
      	drivers/media/dvb-frontends/drx39xyj/drxj.c:9605 ctrl_get_qam_sig_quality() debug: sval_binop_unsigned: divide by zero
      	drivers/media/dvb-frontends/drx39xyj/drxj.c:9605 ctrl_get_qam_sig_quality() debug: sval_binop_unsigned: divide by zero
      Signed-off-by: default avatarMauro Carvalho Chehab <mchehab@osg.samsung.com>
      b6554ea5
    • Mauro Carvalho Chehab's avatar
      [media] airspy: fix bit set/clean mess on s->flags · f6f7b58e
      Mauro Carvalho Chehab authored
      As warned by smatch:
      	drivers/media/usb/airspy/airspy.c:541 airspy_start_streaming() warn: test_bit() takes a bit number
      	drivers/media/usb/airspy/airspy.c:569 airspy_start_streaming() warn: test_bit() takes a bit number
      	drivers/media/usb/airspy/airspy.c:605 airspy_stop_streaming() warn: test_bit() takes a bit number
      
      set_bit/clear_bit argument is the bit number, and not 1 << bit.
      
      Thankfully, one of the bits was not used (URB_BUF), with would
      otherwise cause a driver misfunctioning.
      
      Clean this mess by always using set_bit/clear_bit/test_bit and
      removing the unused bit.
      Signed-off-by: default avatarMauro Carvalho Chehab <mchehab@osg.samsung.com>
      f6f7b58e
    • Hans de Goede's avatar
      [media] gspca: Remove unused ovfx2_vga_mode/ovfx2_cif_mode arrays · b44d3669
      Hans de Goede authored
      Remove the unused ovfx2_vga_mode/ovfx2_cif_mode arrays from the ov519
      driver.
      Signed-off-by: default avatarHans de Goede <hdegoede@redhat.com>
      Signed-off-by: default avatarMauro Carvalho Chehab <mchehab@osg.samsung.com>
      b44d3669
    • Wesley Post's avatar
      [media] gspca: Fix ov519 i2c r/w not working when connected to a xhci host · f7c7ac48
      Wesley Post authored
      Fix the ov519 driver not working (unable to talk to the sensor) when
      plugged into a xhci host. The root cause here is that uhci/ohci/ehci
      hosts typically will send any pending async requests every milli-second
      and then go to sleep for the rest if the milli-second, where as xhci hosts
      send them immediately, causing things to go too fast for the ov519 bridge.
      
      This commit adds a few delays fixing this.
      Signed-off-by: default avatarWesley Post <pa4wdh@xs4all.nl>
      [hdegoede@redhat.com: Also add delays to w996Xcf.c, as that needs them too]
      Signed-off-by: default avatarHans de Goede <hdegoede@redhat.com>
      Signed-off-by: default avatarMauro Carvalho Chehab <mchehab@osg.samsung.com>
      f7c7ac48
    • Hans de Goede's avatar
      [media] bttv: Width must be a multiple of 16 when capturing planar formats · 5c915c68
      Hans de Goede authored
      On my bttv card "Hauppauge WinTV [card=10]" capturing in YV12 fmt at max
      size results in a solid green rectangle being captured (all colors 0 in
      YUV).
      
      This turns out to be caused by max-width (924) not being a multiple of 16.
      
      We've likely never hit this problem before since normally xawtv / tvtime,
      etc. will prefer packed pixel formats. But when using a video card which
      is using xf86-video-modesetting + glamor, only planar XVideo fmts are
      available, and xawtv will chose a matching capture format to avoid needing
      to do conversion, triggering the solid green window problem.
      
      Cc: stable@vger.kernel.org
      Signed-off-by: default avatarHans de Goede <hdegoede@redhat.com>
      Acked-by: default avatarHans Verkuil <hans.verkuil@cisco.com>
      Signed-off-by: default avatarMauro Carvalho Chehab <mchehab@osg.samsung.com>
      5c915c68
  2. 27 Feb, 2016 22 commits
  3. 23 Feb, 2016 11 commits