On 05/01/2023 11:17, Alice Guo (OSS) wrote:
From: Alice Guo <alice.guo@xxxxxxx>
The procedure to program the calibration table of i.MX93 TMU:
1. disable monitoring mode
2. configure TCMCFG
3. write TTCFGR with TTCFGR[CAL_PT] = n
4. write TSCFGR with the sensor value of the calibration point n
5. write TTRCRn associated with the calibration point n
Need to repeat steps 3—5 for all calibration points from the calibration
table provided by the TMU device node.
When TRITSR[V] = 1 and TRITSR[TP5] = 1, need to add 0.5K to
TRITSR[TEMP].
IPBRR0s of lx2160ardb, lx2162aqds and i.MX93 have the same value
0x01900201 so that use the compatible to determine whether the current
TMU belongs to i.MX93 or not.
Signed-off-by: Alice Guo <alice.guo@xxxxxxx>
Reviewed-by: Jacky Bai <ping.bai@xxxxxxx>
Reviewed-by: Ye Li <ye.li@xxxxxxx>
Acked-by: Jason Liu <jason.hui.liu@xxxxxxx>
---
[ ... ]
@@ -369,6 +416,7 @@ static SIMPLE_DEV_PM_OPS(qoriq_tmu_pm_ops,
static const struct of_device_id qoriq_tmu_match[] = {
{ .compatible = "fsl,qoriq-tmu", },
{ .compatible = "fsl,imx8mq-tmu", },
+ { .compatible = "fsl,imx93-tmu", },
{},
Now that you have more than 2, I suggest the pass a ops structure to
.data, so you can call them directly instead of checking the TMU version
in the calibration, the initialization and get the temperature.
};
MODULE_DEVICE_TABLE(of, qoriq_tmu_match);
--
<http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs
Follow Linaro: <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog
