Before the patch [1], the clock probe was done directly in the
clk-mt8365 driver. In this probe function, the array which stores the
data clocks is sized using the higher defined numbers (*_NR_CLOCK) in
the clock lists [2]. Currently, with the patch [1], the specific
clk-mt8365 probe function is replaced by the mtk generic one [3], which
size the clock data array by adding all the clock descriptor array size
provided by the clk-mt8365 driver.
Actually, all clock indexes come from the header file [2], that mean, if
there are more clock (then more index) in the header file [2] than the
number of clock declared in the clock descriptor arrays (which is the
case currently), the clock data array will be undersized and then the
generic probe function will overflow when it will try to write in
"clk_data[CLK_INDEX]". Actually, instead of crashing at boot, the probe
function returns an error in the log which looks like:
"of_clk_hw_onecell_get: invalid index 135", then this clock isn't
enabled.
Solve this issue by adding in the driver the missing clocks declared in
the header clock file [2].
[1]: Commit ffe91cb28f6a ("clk: mediatek: mt8365: Convert to
mtk_clk_simple_{probe,remove}()")
[2]: include/dt-bindings/clock/mediatek,mt8365-clk.h
[3]: drivers/clk/mediatek/clk-mtk.c
Fixes: ffe91cb28f6a ("clk: mediatek: mt8365: Convert to mtk_clk_simple_{probe,remove}()")
Signed-off-by: Alexandre Mergnat <[email protected]>
---
Fix a regression which prevent the initialization of the
latest indexed clocks. The regression is introduced with [1].
[1]: Commit ffe91cb28f6a ("clk: mediatek: mt8365: Convert to
mtk_clk_simple_{probe,remove}()")
---
Changes in v3:
- Remove all binding changes.
- Link to v2: https://lore.kernel.org/r/[email protected]
Changes in v2:
- Add the missing clocks in the mt8365 clock driver.
- Revert all change in binding file except for one clock.
- Link to v1: https://lore.kernel.org/r/[email protected]
---
drivers/clk/mediatek/clk-mt8365.c | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/drivers/clk/mediatek/clk-mt8365.c b/drivers/clk/mediatek/clk-mt8365.c
index 6b4e193f648d..c366ac1611e6 100644
--- a/drivers/clk/mediatek/clk-mt8365.c
+++ b/drivers/clk/mediatek/clk-mt8365.c
@@ -23,6 +23,7 @@
static DEFINE_SPINLOCK(mt8365_clk_lock);
static const struct mtk_fixed_clk top_fixed_clks[] = {
+ FIXED_CLK(CLK_TOP_CLK_NULL, "clk_null", NULL, 0),
FIXED_CLK(CLK_TOP_I2S0_BCK, "i2s0_bck", NULL, 26000000),
FIXED_CLK(CLK_TOP_DSI0_LNTC_DSICK, "dsi0_lntc_dsick", "clk26m",
75000000),
@@ -559,6 +560,14 @@ static const struct mtk_clk_divider top_adj_divs[] = {
0x324, 16, 8, CLK_DIVIDER_ROUND_CLOSEST),
DIV_ADJ_F(CLK_TOP_APLL12_CK_DIV3, "apll12_ck_div3", "apll_i2s3_sel",
0x324, 24, 8, CLK_DIVIDER_ROUND_CLOSEST),
+ DIV_ADJ_F(CLK_TOP_APLL12_CK_DIV4, "apll12_ck_div4", "apll_tdmout_sel",
+ 0x328, 0, 8, CLK_DIVIDER_ROUND_CLOSEST),
+ DIV_ADJ_F(CLK_TOP_APLL12_CK_DIV4B, "apll12_ck_div4b", "apll_tdmout_sel",
+ 0x328, 8, 8, CLK_DIVIDER_ROUND_CLOSEST),
+ DIV_ADJ_F(CLK_TOP_APLL12_CK_DIV5, "apll12_ck_div5", "apll_tdmin_sel",
+ 0x328, 16, 8, CLK_DIVIDER_ROUND_CLOSEST),
+ DIV_ADJ_F(CLK_TOP_APLL12_CK_DIV5B, "apll12_ck_div5b", "apll_tdmin_sel",
+ 0x328, 24, 8, CLK_DIVIDER_ROUND_CLOSEST),
DIV_ADJ_F(CLK_TOP_APLL12_CK_DIV6, "apll12_ck_div6", "apll_spdif_sel",
0x32c, 0, 8, CLK_DIVIDER_ROUND_CLOSEST),
};
@@ -696,6 +705,7 @@ static const struct mtk_gate ifr_clks[] = {
GATE_IFR3(CLK_IFR_GCPU, "ifr_gcpu", "axi_sel", 8),
GATE_IFR3(CLK_IFR_TRNG, "ifr_trng", "axi_sel", 9),
GATE_IFR3(CLK_IFR_AUXADC, "ifr_auxadc", "clk26m", 10),
+ GATE_IFR3(CLK_IFR_CPUM, "ifr_cpum", "clk26m", 11),
GATE_IFR3(CLK_IFR_AUXADC_MD, "ifr_auxadc_md", "clk26m", 14),
GATE_IFR3(CLK_IFR_AP_DMA, "ifr_ap_dma", "axi_sel", 18),
GATE_IFR3(CLK_IFR_DEBUGSYS, "ifr_debugsys", "axi_sel", 24),
@@ -717,6 +727,8 @@ static const struct mtk_gate ifr_clks[] = {
GATE_IFR5(CLK_IFR_PWRAP_TMR, "ifr_pwrap_tmr", "clk26m", 12),
GATE_IFR5(CLK_IFR_PWRAP_SPI, "ifr_pwrap_spi", "clk26m", 13),
GATE_IFR5(CLK_IFR_PWRAP_SYS, "ifr_pwrap_sys", "clk26m", 14),
+ GATE_MTK_FLAGS(CLK_IFR_MCU_PM_BK, "ifr_mcu_pm_bk", NULL, &ifr5_cg_regs,
+ 17, &mtk_clk_gate_ops_setclr, CLK_IGNORE_UNUSED),
GATE_IFR5(CLK_IFR_IRRX_26M, "ifr_irrx_26m", "clk26m", 22),
GATE_IFR5(CLK_IFR_IRRX_32K, "ifr_irrx_32k", "clk32k", 23),
GATE_IFR5(CLK_IFR_I2C0_AXI, "ifr_i2c0_axi", "i2c_sel", 24),
---
base-commit: ac9a78681b921877518763ba0e89202254349d1b
change-id: 20230517-fix-clk-index-96043baf65be
Best regards,
--
Alexandre Mergnat <[email protected]>
On Fri, May 26, 2023 at 03:10:43PM +0200, Alexandre Mergnat wrote:
> Before the patch [1], the clock probe was done directly in the
> clk-mt8365 driver. In this probe function, the array which stores the
> data clocks is sized using the higher defined numbers (*_NR_CLOCK) in
> the clock lists [2]. Currently, with the patch [1], the specific
> clk-mt8365 probe function is replaced by the mtk generic one [3], which
> size the clock data array by adding all the clock descriptor array size
> provided by the clk-mt8365 driver.
>
> Actually, all clock indexes come from the header file [2], that mean, if
> there are more clock (then more index) in the header file [2] than the
> number of clock declared in the clock descriptor arrays (which is the
> case currently), the clock data array will be undersized and then the
> generic probe function will overflow when it will try to write in
> "clk_data[CLK_INDEX]". Actually, instead of crashing at boot, the probe
> function returns an error in the log which looks like:
> "of_clk_hw_onecell_get: invalid index 135", then this clock isn't
> enabled.
>
> Solve this issue by adding in the driver the missing clocks declared in
> the header clock file [2].
>
> [1]: Commit ffe91cb28f6a ("clk: mediatek: mt8365: Convert to
> mtk_clk_simple_{probe,remove}()")
> [2]: include/dt-bindings/clock/mediatek,mt8365-clk.h
> [3]: drivers/clk/mediatek/clk-mtk.c
>
> Fixes: ffe91cb28f6a ("clk: mediatek: mt8365: Convert to mtk_clk_simple_{probe,remove}()")
>
> Signed-off-by: Alexandre Mergnat <[email protected]>
Reviewed-by: Markus Schneider-Pargmann <[email protected]>
Tested-by: Markus Schneider-Pargmann <[email protected]>
Best,
Markus
Hi Steven,
Gentle reminder ping for this patch which fix a regression introduced in
the v6.4-rc1.
--
Regards,
Alexandre
Quoting Alexandre Mergnat (2023-05-26 06:10:43)
> Before the patch [1], the clock probe was done directly in the
> clk-mt8365 driver. In this probe function, the array which stores the
> data clocks is sized using the higher defined numbers (*_NR_CLOCK) in
> the clock lists [2]. Currently, with the patch [1], the specific
> clk-mt8365 probe function is replaced by the mtk generic one [3], which
> size the clock data array by adding all the clock descriptor array size
> provided by the clk-mt8365 driver.
>
> Actually, all clock indexes come from the header file [2], that mean, if
> there are more clock (then more index) in the header file [2] than the
> number of clock declared in the clock descriptor arrays (which is the
> case currently), the clock data array will be undersized and then the
> generic probe function will overflow when it will try to write in
> "clk_data[CLK_INDEX]". Actually, instead of crashing at boot, the probe
> function returns an error in the log which looks like:
> "of_clk_hw_onecell_get: invalid index 135", then this clock isn't
> enabled.
>
> Solve this issue by adding in the driver the missing clocks declared in
> the header clock file [2].
>
> [1]: Commit ffe91cb28f6a ("clk: mediatek: mt8365: Convert to
> mtk_clk_simple_{probe,remove}()")
> [2]: include/dt-bindings/clock/mediatek,mt8365-clk.h
> [3]: drivers/clk/mediatek/clk-mtk.c
>
> Fixes: ffe91cb28f6a ("clk: mediatek: mt8365: Convert to mtk_clk_simple_{probe,remove}()")
>
> Signed-off-by: Alexandre Mergnat <[email protected]>
> ---
Applied to clk-fixes