diff options
| author | Hans Verkuil <hverkuil@xs4all.nl> | 2014-11-21 07:57:12 -0300 |
|---|---|---|
| committer | Mauro Carvalho Chehab <mchehab@osg.samsung.com> | 2014-11-25 08:12:53 -0200 |
| commit | 872dfcfe82438f5211f5fa1dc6406d033a8a2448 (patch) | |
| tree | d3a736a51219e493ed94c3ecc203b36168d887fc /drivers/media/pci/ivtv | |
| parent | 6dfa5131804e8cc0866e2f21ed79ce4b08976cff (diff) | |
[media] bttv/cx25821/cx88/ivtv: use sg_next instead of sg++
Never use sg++, always use sg = sg_next(sg). Scatterlist entries can
be combined if the memory is contiguous but sg++ won't know about that.
As far as I can tell cx88 and ivtv are really broken because of this,
and bttv and cx25821 are OK because vb1 doesn't combine scatterlist
entries.
But regardless, sg++ should never be used, only sg_next is safe.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Diffstat (limited to 'drivers/media/pci/ivtv')
| -rw-r--r-- | drivers/media/pci/ivtv/ivtv-udma.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/drivers/media/pci/ivtv/ivtv-udma.c b/drivers/media/pci/ivtv/ivtv-udma.c index 7338cb2d0a38..bee2329e0b2e 100644 --- a/drivers/media/pci/ivtv/ivtv-udma.c +++ b/drivers/media/pci/ivtv/ivtv-udma.c @@ -76,7 +76,7 @@ void ivtv_udma_fill_sg_array (struct ivtv_user_dma *dma, u32 buffer_offset, u32 int i; struct scatterlist *sg; - for (i = 0, sg = dma->SGlist; i < dma->SG_length; i++, sg++) { + for (i = 0, sg = dma->SGlist; i < dma->SG_length; i++, sg = sg_next(sg)) { dma->SGarray[i].size = cpu_to_le32(sg_dma_len(sg)); dma->SGarray[i].src = cpu_to_le32(sg_dma_address(sg)); dma->SGarray[i].dst = cpu_to_le32(buffer_offset); |