stm32 clock mux for CK48 definition #96353
Conversation
FRASTM
changed the title
FRASTM
force-pushed
the
issue94934
branch
2 times, most recently
from
d4d97b6 to
6c38115
Compare
FRASTM
requested review from
djiatsaf-st,
erwango,
etienne-lms,
gautierg-st and
mathieuchopstm
| &pll { | ||
| div-m = <4>; | ||
| mul-n = <192>; | ||
| div-p = <4>; | ||
| div-q = <8>; | ||
| clocks = <&clk_hse>; | ||
| status = "okay"; | ||
| }; | ||
|
|
||
| &rcc { | ||
| clocks = <&pll>; | ||
| clock-frequency = <DT_FREQ_M(96)>; | ||
| ahb-prescaler = <1>; | ||
| apb1-prescaler = <2>; | ||
| apb2-prescaler = <1>; | ||
| }; |
There was a problem hiding this comment.
Intentional change?
Would it be worth to add an alternate test with its own DTS overlay file?
There was a problem hiding this comment.
yes, when the sdmmc is clocked by the PLL_q, a freq = 48MHz is required.
This overlay file "f4_sdmmc48_pll.overlay"is for running this testcase by un-commenting.
The alternate test is "drivers.clock.stm32_clock_configuration.common_device.f4.sdmmc_48".
Two possible config are in this overlay (selection is by commenting/un-commenting). Is there any reason today to change with 2 different testcases/overlay files, if it was your question ?
There was a problem hiding this comment.
Ok, fine. Already having a test for SDMMC over CK48 seems enough.
I just found strange this added inline comments in this commit.
| &pll { | ||
| div-m = <4>; | ||
| mul-n = <192>; | ||
| div-p = <4>; | ||
| div-q = <8>; | ||
| clocks = <&clk_hse>; | ||
| status = "okay"; | ||
| }; | ||
|
|
||
| &rcc { | ||
| clocks = <&pll>; | ||
| clock-frequency = <DT_FREQ_M(96)>; | ||
| ahb-prescaler = <1>; | ||
| apb1-prescaler = <2>; | ||
| apb2-prescaler = <1>; | ||
| }; |
There was a problem hiding this comment.
Ok, fine. Already having a test for SDMMC over CK48 seems enough.
I just found strange this added inline comments in this commit.
FRASTM
dismissed stale reviews from etienne-lms and mathieuchopstm
via
efda26e
|
Add a commit to enable the clk48 |
zephyrbot
added
the
area: Tests
comment updated |
|
@FRASTM, I think you need to rebase to get the CI tests to all succeed. |
Define the STM32_CK48_ENABLED especially for the stm32F4 series when ck48 node is enabled to leverage its already implemented support. Signed-off-by: Francois Ramu <[email protected]>
|
rebase on 9c5325d |
There was a problem hiding this comment.
@erwango, could you update your review status? You #96353 (review) comment was addressed: see #94934 (comment).
|
This commit enables the clk48 clock mux if STM32_CK48_ENABLED is set by the device tree. Signed-off-by: Francois Ramu <[email protected]>
6 participants
Define the STM32_CK48_ENABLED especially for the stm32F4 series
FIxes #94934