Various changes + helper utils #35
Conversation
For YUV420/422 formats, it may be desirable to preserve the subsampling factor when computing Y and U/V strides. In such cases, double the U/V stride to get the Y stride . Signed-off-by: Naushir Patuck <[email protected]>
|
Do we ever want to not preserve the ratio? At least for the default stride calculation. Can you explain the purpose of the stringy formats? I assume it is to avoid a dependency on V4L2. Does V4L2 or libcamera provide a way to get the identical strings? |
It's a bit more wasteful on memory so perhaps useful in the right circumstances. I think all the regression tests also expect this to be false.
It's just for convenience, see use in src/examples/convert.cpp. Completely separate from v4l2/libcamera enums. I did consider enums, but having free form strings is much easer to manage. |
|
LGTM |
Add helpers to convert from the pisp format integer description to a string and vice-versa. Note that for the latter, this may result in a one to many mapping, and the first matching string is returned. Signed-off-by: Naushir Patuck <[email protected]>
Ensure we set either the Bayer or RGB input flag depending on the input format used. Signed-off-by: Naushir Patuck <[email protected]>
naushir
force-pushed
the
utils
branch
2 times, most recently
from
2c8f14a to
ae8e876
Compare
|
Updated with support for mplanar formats. |
naushir
force-pushed
the
utils
branch
4 times, most recently
from
9091efa to
4965a1e
Compare
Signed-off-by: Naushir Patuck <[email protected]>
Signed-off-by: Naushir Patuck <[email protected]>
No description provided.