From: Miquel Raynal <miquel.raynal@bootlin.com>
To: Jacky Huang <ychuang3@nuvoton.com>,
Shan-Chun Hung <schung@nuvoton.com>,
Michael Turquette <mturquette@baylibre.com>,
Stephen Boyd <sboyd@kernel.org>,
Richard Cochran <richardcochran@gmail.com>,
Arnd Bergmann <arnd@arndb.de>,
Brian Masney <bmasney+clk@redhat.com>,
Jerome Brunet <jbrunet+clk@baylibre.com>,
Rob Herring <robh@kernel.org>,
Krzysztof Kozlowski <krzk+dt@kernel.org>,
Conor Dooley <conor+dt@kernel.org>,
Andrew Jeffery <andrew@codeconstruct.com.au>,
Avi Fishman <avifishman70@gmail.com>,
Tomer Maimon <tmaimon77@gmail.com>,
Tali Perry <tali.perry1@gmail.com>,
Patrick Venture <venture@google.com>,
Nancy Yuen <yuenn@google.com>,
Benjamin Fair <benjaminfair@google.com>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>,
Steam Lin <STLin2@winbond.com>,
linux-arm-kernel@lists.infradead.org, linux-clk@vger.kernel.org,
linux-kernel@vger.kernel.org,
Krzysztof Kozlowski <krzk@kernel.org>,
devicetree@vger.kernel.org, openbmc@lists.ozlabs.org,
Miquel Raynal <miquel.raynal@bootlin.com>,
stable@vger.kernel.org
Subject: [PATCH v12 05/10] clk: nuvoton: ma35d1: Reparent the gates correctly
Date: Wed, 18 Nov 2026 17:59:14 +0100 [thread overview]
Message-ID: <20261118-perso-ma35d1-upstream-clk-v12-5-bee2c8dad4e1@bootlin.com> (raw)
In-Reply-To: <20261118-perso-ma35d1-upstream-clk-v12-0-bee2c8dad4e1@bootlin.com>
The clock tree is still incorrect. There are four main clocks in this
system:
- HXT: external high-speed crystal oscillator
- LXT: external low-speed crystal oscillator
- HIRC: internal high-speed RC oscillator
- LIRC: internal low-speed RC oscillator
Each clock has a 'gate' control to disable it. While in the TRM the
'gate' appears aside, in the CCF topology, the gates should be
below. This is because if we gate one of those clocks, none of the
children clocks continue ticking.
Reparent hxt_gate under hxt, and all hxt children under hxt_gate.
Do the same for the three other clock sources listed above.
After this change, the frequencies are identical, but the clock tree
looks much closer to the reality.
Fixes: 691521a367cf ("clk: nuvoton: Add clock driver for ma35d1 clock controller")
Cc: stable@vger.kernel.org
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
drivers/clk/nuvoton/clk-ma35d1.c | 78 +++++++++++++++++++++++-----------------
1 file changed, 45 insertions(+), 33 deletions(-)
diff --git a/drivers/clk/nuvoton/clk-ma35d1.c b/drivers/clk/nuvoton/clk-ma35d1.c
index 4019df22a74e..d69c9fc0f39e 100644
--- a/drivers/clk/nuvoton/clk-ma35d1.c
+++ b/drivers/clk/nuvoton/clk-ma35d1.c
@@ -74,34 +74,46 @@ static const struct clk_parent_data lxtg_parent_data[] = {
};
/* Mux parent selection, by index into the controller's clock table */
-static const int ca35clk_parent_idx[] = { HXT, CAPLL, DDRPLL };
+static const int ca35clk_parent_idx[] = { HXT_GATE, CAPLL, DDRPLL };
static const int sysclk0_parent_idx[] = { EPLL_DIV2, SYSPLL };
-static const int sysclk1_parent_idx[] = { HXT, SYSPLL };
+static const int sysclk1_parent_idx[] = { HXT_GATE, SYSPLL };
static const int axiclk_parent_idx[] = { AXICLK_DIV2, AXICLK_DIV4 };
-static const int ccap_parent_idx[] = { HXT, VPLL, APLL, SYSPLL };
+static const int ccap_parent_idx[] = { HXT_GATE, VPLL, APLL, SYSPLL };
static const int sdh_parent_idx[] = { SYSPLL, APLL };
static const int dcu_parent_idx[] = { EPLL_DIV2, SYSPLL };
static const int gfx_parent_idx[] = { EPLL, SYSPLL };
-static const int dbg_parent_idx[] = { HIRC, SYSPLL };
-static const int timer0_parent_idx[] = { HXT, LXT, PCLK0, -1, -1, LIRC, -1, HIRC };
-static const int timer1_parent_idx[] = { HXT, LXT, PCLK0, -1, -1, LIRC, -1, HIRC };
-static const int timer2_parent_idx[] = { HXT, LXT, PCLK1, -1, -1, LIRC, -1, HIRC };
-static const int timer3_parent_idx[] = { HXT, LXT, PCLK1, -1, -1, LIRC, -1, HIRC };
-static const int timer4_parent_idx[] = { HXT, LXT, PCLK2, -1, -1, LIRC, -1, HIRC };
-static const int timer5_parent_idx[] = { HXT, LXT, PCLK2, -1, -1, LIRC, -1, HIRC };
-static const int timer6_parent_idx[] = { HXT, LXT, PCLK0, -1, -1, LIRC, -1, HIRC };
-static const int timer7_parent_idx[] = { HXT, LXT, PCLK0, -1, -1, LIRC, -1, HIRC };
-static const int timer8_parent_idx[] = { HXT, LXT, PCLK1, -1, -1, LIRC, -1, HIRC };
-static const int timer9_parent_idx[] = { HXT, LXT, PCLK1, -1, -1, LIRC, -1, HIRC };
-static const int timer10_parent_idx[] = { HXT, LXT, PCLK2, -1, -1, LIRC, -1, HIRC };
-static const int timer11_parent_idx[] = { HXT, LXT, PCLK2, -1, -1, LIRC, -1, HIRC };
-static const int uart_parent_idx[] = { HXT, SYSCLK1_DIV2 };
-static const int wdt0_parent_idx[] = { -1, LXT, PCLK3_DIV4096, LIRC };
-static const int wdt1_parent_idx[] = { -1, LXT, PCLK3_DIV4096, LIRC };
-static const int wdt2_parent_idx[] = { -1, LXT, PCLK4_DIV4096, LIRC };
-static const int wwdt0_parent_idx[] = { -1, -1, PCLK3_DIV4096, LIRC };
-static const int wwdt1_parent_idx[] = { -1, -1, PCLK3_DIV4096, LIRC };
-static const int wwdt2_parent_idx[] = { -1, -1, PCLK4_DIV4096, LIRC };
+static const int dbg_parent_idx[] = { HIRC_GATE, SYSPLL };
+static const int timer0_parent_idx[] = { HXT_GATE, LXT_GATE, PCLK0, -1, -1,
+ LIRC_GATE, -1, HIRC_GATE };
+static const int timer1_parent_idx[] = { HXT_GATE, LXT_GATE, PCLK0, -1, -1,
+ LIRC_GATE, -1, HIRC_GATE };
+static const int timer2_parent_idx[] = { HXT_GATE, LXT_GATE, PCLK1, -1, -1,
+ LIRC_GATE, -1, HIRC_GATE };
+static const int timer3_parent_idx[] = { HXT_GATE, LXT_GATE, PCLK1, -1, -1,
+ LIRC_GATE, -1, HIRC_GATE };
+static const int timer4_parent_idx[] = { HXT_GATE, LXT_GATE, PCLK2, -1, -1,
+ LIRC_GATE, -1, HIRC_GATE };
+static const int timer5_parent_idx[] = { HXT_GATE, LXT_GATE, PCLK2, -1, -1,
+ LIRC_GATE, -1, HIRC_GATE };
+static const int timer6_parent_idx[] = { HXT_GATE, LXT_GATE, PCLK0, -1, -1,
+ LIRC_GATE, -1, HIRC_GATE };
+static const int timer7_parent_idx[] = { HXT_GATE, LXT_GATE, PCLK0, -1, -1,
+ LIRC_GATE, -1, HIRC_GATE };
+static const int timer8_parent_idx[] = { HXT_GATE, LXT_GATE, PCLK1, -1, -1,
+ LIRC_GATE, -1, HIRC_GATE };
+static const int timer9_parent_idx[] = { HXT_GATE, LXT_GATE, PCLK1, -1, -1,
+ LIRC_GATE, -1, HIRC_GATE };
+static const int timer10_parent_idx[] = { HXT_GATE, LXT_GATE, PCLK2, -1, -1,
+ LIRC_GATE, -1, HIRC_GATE };
+static const int timer11_parent_idx[] = { HXT_GATE, LXT_GATE, PCLK2, -1, -1,
+ LIRC_GATE, -1, HIRC_GATE };
+static const int uart_parent_idx[] = { HXT_GATE, SYSCLK1_DIV2 };
+static const int wdt0_parent_idx[] = { -1, LXT_GATE, PCLK3_DIV4096, LIRC_GATE };
+static const int wdt1_parent_idx[] = { -1, LXT_GATE, PCLK3_DIV4096, LIRC_GATE };
+static const int wdt2_parent_idx[] = { -1, LXT_GATE, PCLK4_DIV4096, LIRC_GATE };
+static const int wwdt0_parent_idx[] = { -1, -1, PCLK3_DIV4096, LIRC_GATE };
+static const int wwdt1_parent_idx[] = { -1, -1, PCLK3_DIV4096, LIRC_GATE };
+static const int wwdt2_parent_idx[] = { -1, -1, PCLK4_DIV4096, LIRC_GATE };
static const int spi0_parent_idx[] = { PCLK1, APLL };
static const int spi1_parent_idx[] = { PCLK2, APLL };
static const int spi2_parent_idx[] = { PCLK1, APLL };
@@ -109,11 +121,11 @@ static const int spi3_parent_idx[] = { PCLK2, APLL };
static const int qspi_parent_idx[] = { PCLK0, APLL };
static const int i2s_parent_idx[] = { APLL, SYSCLK1_DIV2 };
static const int can_parent_idx[] = { APLL, VPLL };
-static const int cko_parent_idx[] = { HXT, LXT, HIRC, LIRC,
+static const int cko_parent_idx[] = { HXT_GATE, LXT_GATE, HIRC_GATE, LIRC_GATE,
AXICLK_DIV4, SYSPLL, DDRPLL,
EPLL_DIV2, APLL, VPLL };
-static const int smc_parent_idx[] = { HXT, PCLK4 };
-static const int kpi_parent_idx[] = { HXT, LXT };
+static const int smc_parent_idx[] = { HXT_GATE, PCLK4 };
+static const int kpi_parent_idx[] = { HXT_GATE, LXT_GATE };
static const struct clk_div_table ip_div_table[] = {
{0, 2}, {1, 4}, {2, 6}, {3, 8}, {4, 10},
@@ -283,16 +295,16 @@ static int ma35d1_clocks_probe(struct platform_device *pdev)
clk_base + REG_CLK_PWRCTL, 3);
hws[CAPLL] = ma35d1_reg_clk_pll(dev, CAPLL, pllmode[0], "capll",
- hws[HXT], clk_base + REG_CLK_PLL0CTL0);
+ hws[HXT_GATE], clk_base + REG_CLK_PLL0CTL0);
hws[SYSPLL] = ma35d1_clk_fixed("syspll", 180000000);
hws[DDRPLL] = ma35d1_reg_clk_pll(dev, DDRPLL, pllmode[1], "ddrpll",
- hws[HXT], clk_base + REG_CLK_PLL2CTL0);
+ hws[HXT_GATE], clk_base + REG_CLK_PLL2CTL0);
hws[APLL] = ma35d1_reg_clk_pll(dev, APLL, pllmode[2], "apll",
- hws[HXT], clk_base + REG_CLK_PLL3CTL0);
+ hws[HXT_GATE], clk_base + REG_CLK_PLL3CTL0);
hws[EPLL] = ma35d1_reg_clk_pll(dev, EPLL, pllmode[3], "epll",
- hws[HXT], clk_base + REG_CLK_PLL4CTL0);
+ hws[HXT_GATE], clk_base + REG_CLK_PLL4CTL0);
hws[VPLL] = ma35d1_reg_clk_pll(dev, VPLL, pllmode[4], "vpll",
- hws[HXT], clk_base + REG_CLK_PLL5CTL0);
+ hws[HXT_GATE], clk_base + REG_CLK_PLL5CTL0);
hws[EPLL_DIV2] = ma35d1_clk_fixed_factor(dev, "epll_div2", "epll", 1, 2);
hws[EPLL_DIV4] = ma35d1_clk_fixed_factor(dev, "epll_div4", "epll", 1, 4);
@@ -460,7 +472,7 @@ static int ma35d1_clocks_probe(struct platform_device *pdev)
hws[CKO_GATE] = ma35d1_clk_gate(dev, "cko_gate", "cko_div",
clk_base + REG_CLK_SYSCLK1, 13);
- hws[GTMR_GATE] = ma35d1_clk_gate(dev, "gtmr_gate", "hirc",
+ hws[GTMR_GATE] = ma35d1_clk_gate(dev, "gtmr_gate", "hirc_gate",
clk_base + REG_CLK_SYSCLK1, 14);
hws[GPA_GATE] = ma35d1_clk_gate(dev, "gpa_gate", "hclk0",
@@ -661,7 +673,7 @@ static int ma35d1_clocks_probe(struct platform_device *pdev)
hws[UART16_GATE] = ma35d1_clk_gate(dev, "uart16_gate", "uart16_div",
clk_base + REG_CLK_APBCLK0, 28);
- hws[RTC_GATE] = ma35d1_clk_gate(dev, "rtc_gate", "lxt",
+ hws[RTC_GATE] = ma35d1_clk_gate(dev, "rtc_gate", "lxt_gate",
clk_base + REG_CLK_APBCLK0, 29);
hws[DDR_GATE] = ma35d1_clk_gate(dev, "ddr_gate", "ddrpll",
clk_base + REG_CLK_APBCLK0, 30);
--
2.56.0
next prev parent reply other threads:[~2026-11-18 17:06 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-18 16:59 [PATCH v12 00/10] clk: nuvoton: ma35d1: Fix mux parenting and peripheral clock rates Miquel Raynal
2026-11-18 16:59 ` [PATCH v12 01/10] dt-bindings: clock: ma35d1: Mark HXT and LXT legacy Miquel Raynal
2026-11-18 16:59 ` [PATCH v12 02/10] clk: nuvoton: ma35d1: Use clk_hw pointers as mux parents Miquel Raynal
2026-11-18 16:59 ` [PATCH v12 03/10] clk: nuvoton: ma35d1: Avoid possible error pointer dereferencing Miquel Raynal
2026-11-18 16:59 ` [PATCH v12 04/10] clk: nuvoton: ma35d1: Retrieve HXT/LXT from DT when possible Miquel Raynal
2026-11-18 16:59 ` Miquel Raynal [this message]
2026-11-18 16:59 ` [PATCH v12 06/10] clk: nuvoton: ma35d1: Use device managed helpers for fixed rate clocks Miquel Raynal
2026-11-18 16:59 ` [PATCH v12 07/10] clk: Create devm_clk_hw_register_fixed_rate_parent_hw() Miquel Raynal
2026-11-18 16:59 ` [PATCH v12 08/10] clk: nuvoton: ma35d1: Reparent SYSPLL correctly Miquel Raynal
2026-11-18 16:59 ` [PATCH v12 09/10] clk: nuvoton: ma35d1: Fix CCAP muxes bus width Miquel Raynal
2026-11-18 16:59 ` [PATCH v12 10/10] MAINTAINERS: Add myself as Nuvoton MA35D1 clock controller Miquel Raynal
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20261118-perso-ma35d1-upstream-clk-v12-5-bee2c8dad4e1@bootlin.com \
--to=miquel.raynal@bootlin.com \
--cc=STLin2@winbond.com \
--cc=andrew@codeconstruct.com.au \
--cc=arnd@arndb.de \
--cc=avifishman70@gmail.com \
--cc=benjaminfair@google.com \
--cc=bmasney+clk@redhat.com \
--cc=conor+dt@kernel.org \
--cc=devicetree@vger.kernel.org \
--cc=jbrunet+clk@baylibre.com \
--cc=krzk+dt@kernel.org \
--cc=krzk@kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-clk@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mturquette@baylibre.com \
--cc=openbmc@lists.ozlabs.org \
--cc=richardcochran@gmail.com \
--cc=robh@kernel.org \
--cc=sboyd@kernel.org \
--cc=schung@nuvoton.com \
--cc=stable@vger.kernel.org \
--cc=tali.perry1@gmail.com \
--cc=thomas.petazzoni@bootlin.com \
--cc=tmaimon77@gmail.com \
--cc=venture@google.com \
--cc=ychuang3@nuvoton.com \
--cc=yuenn@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®