The layer enumeration start with 0 (0-15 for LS1021a and 0-63 for Vybrid) whereas the register enumeration start from 1 (1-10 for LS1021a and 1-9 for Vybrid). The loop started off from 0 for both iterations and initialized the number of layers inclusive, which is one layer too many.
All extensively written registers seem to be unassigned, it seems that the write to those registers did not do any harm in practice.
Signed-off-by: Stefan Agner stefan@agner.ch --- This two patches apply cleanly on top of my earlier DCU fixes patchset: https://lkml.org/lkml/2015/11/18/953
Changes since v1: - (none)
drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c index db69725..3c92889 100644 --- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c +++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_crtc.c @@ -169,8 +169,8 @@ int fsl_dcu_drm_crtc_create(struct fsl_dcu_drm_device *fsl_dev) reg_num = LS1021A_LAYER_REG_NUM; else reg_num = VF610_LAYER_REG_NUM; - for (i = 0; i <= fsl_dev->soc->total_layer; i++) { - for (j = 0; j < reg_num; j++) + for (i = 0; i < fsl_dev->soc->total_layer; i++) { + for (j = 1; j <= reg_num; j++) regmap_write(fsl_dev->regmap, DCU_CTRLDESCLN(i, j), 0); } regmap_update_bits(fsl_dev->regmap, DCU_DCU_MODE,