clk: qcom: Fix pre-divider usage for pixel RCG
The clk_rcg_pixel_set_rate clk_op sets up the pre-divider by reading its current value from the NS register. Using the pre-divider wasn't really intended when creating these ops. The pixel RCG was only intended to achieve fractional multiplication provided in the pixel_table array. Leaving the pre-divider to the existing register value results in a wrong pixel clock when the bootloader sets up the display. This was left unidentified because the IFC6410 Plus board on which this was verified didn't have a bootloader that configured the display. Don't set the RCG pre-divider in freq_tbl to the existing NS register value. Force it to 1 and only use the M/N counter to achieve the desired fractional multiplication. Cc: Vinay Simha <vinaysimha@inforcecomputing.com> Signed-off-by: Archit Taneja <architt@codeaurora.org> Tested-by: John Stultz <john.stultz@linaro.org> Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
This commit is contained in:
parent
d3622b5885
commit
811a498e5e
|
@ -638,7 +638,6 @@ static int clk_rcg_pixel_set_rate(struct clk_hw *hw, unsigned long rate,
|
|||
return ret;
|
||||
|
||||
src = ns_to_src(&rcg->s, ns);
|
||||
f.pre_div = ns_to_pre_div(&rcg->p, ns) + 1;
|
||||
|
||||
for (i = 0; i < num_parents; i++) {
|
||||
if (src == rcg->s.parent_map[i].cfg) {
|
||||
|
@ -647,6 +646,9 @@ static int clk_rcg_pixel_set_rate(struct clk_hw *hw, unsigned long rate,
|
|||
}
|
||||
}
|
||||
|
||||
/* bypass the pre divider */
|
||||
f.pre_div = 1;
|
||||
|
||||
/* let us find appropriate m/n values for this */
|
||||
for (; frac->num; frac++) {
|
||||
request = (rate * frac->den) / frac->num;
|
||||
|
|
Loading…
Reference in New Issue