diff options
| author | Samuel Williams <sam8641@gmail.com> | 2018-04-01 11:00:27 -0400 |
|---|---|---|
| committer | Mauro Carvalho Chehab <mchehab+samsung@kernel.org> | 2018-05-04 15:27:31 -0400 |
| commit | 6f7e780b7e93aed10fe5d32f386add042644cb0b (patch) | |
| tree | 92852828b7428ff680a0ed0f378561a95f4f930a /drivers | |
| parent | a03e457bb1eef73c811ea2b3d5b57cef559ae8e0 (diff) | |
media: bttv: Fixed oops error when capturing at yuv410p
When capturing at yuv410p, sg_next was called too many times when chroma is
false, eventually returning NULL. This patch does fix this for my hardware.
Signed-off-by: Samuel Williams <sam8641@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Diffstat (limited to 'drivers')
| -rw-r--r-- | drivers/media/pci/bt8xx/bttv-risc.c | 17 |
1 files changed, 9 insertions, 8 deletions
diff --git a/drivers/media/pci/bt8xx/bttv-risc.c b/drivers/media/pci/bt8xx/bttv-risc.c index 3859dde98be2..6a6be0b49f70 100644 --- a/drivers/media/pci/bt8xx/bttv-risc.c +++ b/drivers/media/pci/bt8xx/bttv-risc.c @@ -189,20 +189,21 @@ bttv_risc_planar(struct bttv *btv, struct btcx_riscmem *risc, yoffset -= sg_dma_len(ysg); ysg = sg_next(ysg); } - while (uoffset && uoffset >= sg_dma_len(usg)) { - uoffset -= sg_dma_len(usg); - usg = sg_next(usg); - } - while (voffset && voffset >= sg_dma_len(vsg)) { - voffset -= sg_dma_len(vsg); - vsg = sg_next(vsg); - } /* calculate max number of bytes we can write */ ylen = todo; if (yoffset + ylen > sg_dma_len(ysg)) ylen = sg_dma_len(ysg) - yoffset; if (chroma) { + while (uoffset && uoffset >= sg_dma_len(usg)) { + uoffset -= sg_dma_len(usg); + usg = sg_next(usg); + } + while (voffset && voffset >= sg_dma_len(vsg)) { + voffset -= sg_dma_len(vsg); + vsg = sg_next(vsg); + } + if (uoffset + (ylen>>hshift) > sg_dma_len(usg)) ylen = (sg_dma_len(usg) - uoffset) << hshift; if (voffset + (ylen>>hshift) > sg_dma_len(vsg)) |