2018-08-03 03:04:33

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 00/18] MIPS: intel: add initial support for Intel MIPS SoCs

This patch series is for adding the support for Intel MIPS
interAptiv SoC GRX500 family. It includes CCF support, serial
driver optimization and DTS modification.

This patch series is applied on top of linux v4.18-rc7.
Basic verification is performed on GRX500 board.

We propose merging this patch series into MIPS Linux tree.

Changes in v2:
- Remove unused _END macros
- Remove the redundant check and not accurate comments
- Replace the get_counter_resolution function with fixed value 2
- Use obj-y and split into per line per .o
- Add EVA mapping description in code comments
- Remove unused include header file
- Do a clean-up for grx500_defconfig
- Rewrite clock driver, add platform clock description details in
clock driver.
- Rewrite clock driver's dt-binding document according to Rob Herring's
comments.
- Simplify device tree docoment, remove some clock description.
- New patch split from previous patch
- The memory address is changed to @20000000
- Update to obj-$(CONFIG_BUILTIN_DTB) as per commit fca3aa166422
- New patch split from previous patch
- Add the board and chip compatible in dt document
- New patch to reorder the head files according to the coding style.

Hua Ma (3):
MIPS: intel: Add initial support for Intel MIPS SoCs
MIPS: dts: Add initial support for Intel MIPS SoCs
dt-binding: MIPS: Add documentation of Intel MIPS SoCs

Songjun Wu (13):
MIPS: dts: Change upper case to lower case
MIPS: dts: Add aliases node for lantiq danube serial
serial: intel: Get serial id from dts
serial: intel: Change ltq_w32_mask to asc_update_bits
MIPS: lantiq: Unselect SWAP_IO_SPACE when LANTIQ is selected
serial: intel: Use readl/writel instead of ltq_r32/ltq_w32
serial: intel: Rename fpiclk to freqclk
serial: intel: Replace clk_enable/clk_disable with clk generic API
serial: intel: Add CCF support
serial: intel: Support more platform
serial: intel: Reorder the head files
serial: intel: Change init_lqasc to static declaration
dt-bindings: serial: lantiq: Add optional properties for CCF

Yixin Zhu (2):
clk: intel: Add clock driver for Intel MIPS SoCs
dt-bindings: clk: Add documentation of grx500 clock controller

.../devicetree/bindings/clock/intel,grx500-clk.txt | 39 ++
Documentation/devicetree/bindings/mips/intel.txt | 17 +
.../devicetree/bindings/serial/lantiq_asc.txt | 15 +
arch/mips/Kbuild.platforms | 1 +
arch/mips/Kconfig | 30 +-
arch/mips/boot/dts/Makefile | 1 +
arch/mips/boot/dts/intel-mips/Makefile | 4 +
arch/mips/boot/dts/intel-mips/easy350_anywan.dts | 26 ++
arch/mips/boot/dts/intel-mips/xrx500.dtsi | 66 +++
arch/mips/boot/dts/lantiq/danube.dtsi | 42 +-
arch/mips/boot/dts/lantiq/easy50712.dts | 18 +-
arch/mips/configs/grx500_defconfig | 138 ++++++
.../asm/mach-intel-mips/cpu-feature-overrides.h | 61 +++
arch/mips/include/asm/mach-intel-mips/ioremap.h | 39 ++
arch/mips/include/asm/mach-intel-mips/irq.h | 17 +
.../asm/mach-intel-mips/kernel-entry-init.h | 104 +++++
arch/mips/include/asm/mach-intel-mips/spaces.h | 27 ++
arch/mips/include/asm/mach-intel-mips/war.h | 18 +
arch/mips/intel-mips/Kconfig | 22 +
arch/mips/intel-mips/Makefile | 5 +
arch/mips/intel-mips/Platform | 12 +
arch/mips/intel-mips/irq.c | 35 ++
arch/mips/intel-mips/prom.c | 172 ++++++++
arch/mips/intel-mips/time.c | 42 ++
drivers/clk/Kconfig | 1 +
drivers/clk/Makefile | 3 +
drivers/clk/intel/Kconfig | 20 +
drivers/clk/intel/Makefile | 7 +
drivers/clk/intel/clk-cgu-pll.c | 166 ++++++++
drivers/clk/intel/clk-cgu-pll.h | 34 ++
drivers/clk/intel/clk-cgu.c | 470 +++++++++++++++++++++
drivers/clk/intel/clk-cgu.h | 259 ++++++++++++
drivers/clk/intel/clk-grx500.c | 168 ++++++++
drivers/tty/serial/Kconfig | 2 +-
drivers/tty/serial/lantiq.c | 143 ++++---
include/dt-bindings/clock/intel,grx500-clk.h | 69 +++
36 files changed, 2206 insertions(+), 87 deletions(-)
create mode 100644 Documentation/devicetree/bindings/clock/intel,grx500-clk.txt
create mode 100644 Documentation/devicetree/bindings/mips/intel.txt
create mode 100644 arch/mips/boot/dts/intel-mips/Makefile
create mode 100644 arch/mips/boot/dts/intel-mips/easy350_anywan.dts
create mode 100644 arch/mips/boot/dts/intel-mips/xrx500.dtsi
create mode 100644 arch/mips/configs/grx500_defconfig
create mode 100644 arch/mips/include/asm/mach-intel-mips/cpu-feature-overrides.h
create mode 100644 arch/mips/include/asm/mach-intel-mips/ioremap.h
create mode 100644 arch/mips/include/asm/mach-intel-mips/irq.h
create mode 100644 arch/mips/include/asm/mach-intel-mips/kernel-entry-init.h
create mode 100644 arch/mips/include/asm/mach-intel-mips/spaces.h
create mode 100644 arch/mips/include/asm/mach-intel-mips/war.h
create mode 100644 arch/mips/intel-mips/Kconfig
create mode 100644 arch/mips/intel-mips/Makefile
create mode 100644 arch/mips/intel-mips/Platform
create mode 100644 arch/mips/intel-mips/irq.c
create mode 100644 arch/mips/intel-mips/prom.c
create mode 100644 arch/mips/intel-mips/time.c
create mode 100644 drivers/clk/intel/Kconfig
create mode 100644 drivers/clk/intel/Makefile
create mode 100644 drivers/clk/intel/clk-cgu-pll.c
create mode 100644 drivers/clk/intel/clk-cgu-pll.h
create mode 100644 drivers/clk/intel/clk-cgu.c
create mode 100644 drivers/clk/intel/clk-cgu.h
create mode 100644 drivers/clk/intel/clk-grx500.c
create mode 100644 include/dt-bindings/clock/intel,grx500-clk.h

--
2.11.0



2018-08-03 03:04:46

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 01/18] MIPS: intel: Add initial support for Intel MIPS SoCs

From: Hua Ma <[email protected]>

Add initial support for Intel MIPS interAptiv SoCs made by Intel.
This series will add support for the grx500 family.

The series allows booting a minimal system using a initramfs.

Signed-off-by: Hua Ma <[email protected]>
Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2:
- Remove unused _END macros
- Remove the redundant check and not accurate comments
- Replace the get_counter_resolution function with fixed value 2
- Use obj-y and split into per line per .o
- Add EVA mapping description in code comments
- Remove unused include header file
- Do a clean-up for grx500_defconfig

arch/mips/Kbuild.platforms | 1 +
arch/mips/Kconfig | 29 ++++
arch/mips/configs/grx500_defconfig | 138 +++++++++++++++++
.../asm/mach-intel-mips/cpu-feature-overrides.h | 61 ++++++++
arch/mips/include/asm/mach-intel-mips/ioremap.h | 39 +++++
arch/mips/include/asm/mach-intel-mips/irq.h | 17 ++
.../asm/mach-intel-mips/kernel-entry-init.h | 104 +++++++++++++
arch/mips/include/asm/mach-intel-mips/spaces.h | 27 ++++
arch/mips/include/asm/mach-intel-mips/war.h | 18 +++
arch/mips/intel-mips/Kconfig | 22 +++
arch/mips/intel-mips/Makefile | 5 +
arch/mips/intel-mips/Platform | 12 ++
arch/mips/intel-mips/irq.c | 35 +++++
arch/mips/intel-mips/prom.c | 172 +++++++++++++++++++++
arch/mips/intel-mips/time.c | 42 +++++
15 files changed, 722 insertions(+)
create mode 100644 arch/mips/configs/grx500_defconfig
create mode 100644 arch/mips/include/asm/mach-intel-mips/cpu-feature-overrides.h
create mode 100644 arch/mips/include/asm/mach-intel-mips/ioremap.h
create mode 100644 arch/mips/include/asm/mach-intel-mips/irq.h
create mode 100644 arch/mips/include/asm/mach-intel-mips/kernel-entry-init.h
create mode 100644 arch/mips/include/asm/mach-intel-mips/spaces.h
create mode 100644 arch/mips/include/asm/mach-intel-mips/war.h
create mode 100644 arch/mips/intel-mips/Kconfig
create mode 100644 arch/mips/intel-mips/Makefile
create mode 100644 arch/mips/intel-mips/Platform
create mode 100644 arch/mips/intel-mips/irq.c
create mode 100644 arch/mips/intel-mips/prom.c
create mode 100644 arch/mips/intel-mips/time.c

diff --git a/arch/mips/Kbuild.platforms b/arch/mips/Kbuild.platforms
index ac7ad54f984f..bcd647060f3e 100644
--- a/arch/mips/Kbuild.platforms
+++ b/arch/mips/Kbuild.platforms
@@ -12,6 +12,7 @@ platforms += cobalt
platforms += dec
platforms += emma
platforms += generic
+platforms += intel-mips
platforms += jazz
platforms += jz4740
platforms += lantiq
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 08c10c518f83..2d34f17f3e24 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -409,6 +409,34 @@ config LANTIQ
select ARCH_HAS_RESET_CONTROLLER
select RESET_CONTROLLER

+config INTEL_MIPS
+ bool "Intel MIPS interAptiv SoC based platforms"
+ select BOOT_RAW
+ select CEVT_R4K
+ select COMMON_CLK
+ select CPU_MIPS32_3_5_EVA
+ select CPU_MIPS32_3_5_FEATURES
+ select CPU_MIPSR2_IRQ_EI
+ select CPU_MIPSR2_IRQ_VI
+ select CSRC_R4K
+ select DMA_NONCOHERENT
+ select GENERIC_ISA_DMA
+ select IRQ_MIPS_CPU
+ select MFD_CORE
+ select MFD_SYSCON
+ select MIPS_CPU_SCACHE
+ select MIPS_GIC
+ select SYS_HAS_CPU_MIPS32_R1
+ select SYS_HAS_CPU_MIPS32_R2
+ select SYS_HAS_CPU_MIPS32_R3_5
+ select SYS_SUPPORTS_BIG_ENDIAN
+ select SYS_SUPPORTS_32BIT_KERNEL
+ select SYS_SUPPORTS_MIPS_CPS
+ select SYS_SUPPORTS_MULTITHREADING
+ select SYS_SUPPORTS_ZBOOT
+ select TIMER_OF
+ select USE_OF
+
config LASAT
bool "LASAT Networks platforms"
select CEVT_R4K
@@ -1016,6 +1044,7 @@ source "arch/mips/bcm47xx/Kconfig"
source "arch/mips/bcm63xx/Kconfig"
source "arch/mips/bmips/Kconfig"
source "arch/mips/generic/Kconfig"
+source "arch/mips/intel-mips/Kconfig"
source "arch/mips/jazz/Kconfig"
source "arch/mips/jz4740/Kconfig"
source "arch/mips/lantiq/Kconfig"
diff --git a/arch/mips/configs/grx500_defconfig b/arch/mips/configs/grx500_defconfig
new file mode 100644
index 000000000000..9dd7ba8e1f74
--- /dev/null
+++ b/arch/mips/configs/grx500_defconfig
@@ -0,0 +1,138 @@
+CONFIG_INTEL_MIPS=y
+CONFIG_DTB_INTEL_MIPS_GRX500=y
+CONFIG_CPU_MIPS32_R2=y
+CONFIG_SCHED_SMT=y
+# CONFIG_MIPS_MT_FPAFF is not set
+CONFIG_MIPS_CPS=y
+CONFIG_DEFAULT_MMAP_MIN_ADDR=65536
+CONFIG_NR_CPUS=2
+CONFIG_HZ_100=y
+# CONFIG_SECCOMP is not set
+# CONFIG_LOCALVERSION_AUTO is not set
+CONFIG_DEFAULT_HOSTNAME="GRX500"
+CONFIG_SYSVIPC=y
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_LOG_BUF_SHIFT=18
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_INITRAMFS_SOURCE=""
+# CONFIG_RD_GZIP is not set
+# CONFIG_RD_BZIP2 is not set
+# CONFIG_RD_LZMA is not set
+# CONFIG_RD_LZO is not set
+# CONFIG_RD_LZ4 is not set
+CONFIG_INITRAMFS_COMPRESSION_XZ=y
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_SYSCTL_SYSCALL=y
+# CONFIG_FHANDLE is not set
+# CONFIG_AIO is not set
+CONFIG_EMBEDDED=y
+# CONFIG_SLUB_DEBUG is not set
+# CONFIG_COMPAT_BRK is not set
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+CONFIG_MODVERSIONS=y
+# CONFIG_LBDAF is not set
+# CONFIG_BLK_DEV_BSG is not set
+# CONFIG_BLK_DEBUG_FS is not set
+# CONFIG_SUSPEND is not set
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_INET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
+CONFIG_IPV6_MULTIPLE_TABLES=y
+CONFIG_IPV6_SUBTREES=y
+CONFIG_IPV6_MROUTE=y
+CONFIG_NETFILTER=y
+CONFIG_NF_CONNTRACK=m
+CONFIG_NF_CONNTRACK_MARK=y
+CONFIG_NETFILTER_XT_MARK=m
+CONFIG_NETFILTER_XT_TARGET_LOG=m
+CONFIG_NETFILTER_XT_TARGET_TCPMSS=m
+CONFIG_NETFILTER_XT_MATCH_COMMENT=m
+CONFIG_NETFILTER_XT_MATCH_CONNTRACK=m
+CONFIG_NETFILTER_XT_MATCH_LIMIT=m
+CONFIG_NETFILTER_XT_MATCH_MAC=m
+CONFIG_NETFILTER_XT_MATCH_MULTIPORT=m
+CONFIG_NETFILTER_XT_MATCH_STATE=m
+CONFIG_NETFILTER_XT_MATCH_TIME=m
+CONFIG_NF_CONNTRACK_IPV4=m
+CONFIG_IP_NF_IPTABLES=m
+CONFIG_IP_NF_FILTER=m
+CONFIG_IP_NF_TARGET_REJECT=m
+CONFIG_IP_NF_NAT=m
+CONFIG_IP_NF_TARGET_MASQUERADE=m
+CONFIG_IP_NF_TARGET_REDIRECT=m
+CONFIG_IP_NF_MANGLE=m
+CONFIG_NF_CONNTRACK_IPV6=m
+CONFIG_IP6_NF_IPTABLES=m
+CONFIG_IP6_NF_FILTER=m
+CONFIG_IP6_NF_TARGET_REJECT=m
+CONFIG_IP6_NF_MANGLE=m
+CONFIG_ATM=m
+CONFIG_ATM_BR2684=m
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=4
+CONFIG_BLK_DEV_RAM_SIZE=8192
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_SERIO is not set
+# CONFIG_CONSOLE_TRANSLATIONS is not set
+# CONFIG_VT_CONSOLE is not set
+# CONFIG_LEGACY_PTYS is not set
+# CONFIG_DEVMEM is not set
+CONFIG_SERIAL_LANTIQ=y
+# CONFIG_HW_RANDOM is not set
+# CONFIG_HWMON is not set
+# CONFIG_VGA_CONSOLE is not set
+# CONFIG_HID is not set
+# CONFIG_USB_SUPPORT is not set
+# CONFIG_VIRTIO_MENU is not set
+# CONFIG_MIPS_PLATFORM_DEVICES is not set
+CONFIG_INTEL_CGU_CLK=y
+# CONFIG_IOMMU_SUPPORT is not set
+# CONFIG_MANDATORY_FILE_LOCKING is not set
+CONFIG_QUOTA=y
+# CONFIG_PRINT_QUOTA_WARNING is not set
+CONFIG_AUTOFS4_FS=y
+CONFIG_PROC_KCORE=y
+# CONFIG_PROC_PAGE_MONITOR is not set
+CONFIG_PROC_CHILDREN=y
+CONFIG_TMPFS=y
+CONFIG_TMPFS_XATTR=y
+CONFIG_CONFIGFS_FS=y
+CONFIG_SQUASHFS=y
+CONFIG_SQUASHFS_XATTR=y
+CONFIG_SQUASHFS_LZ4=y
+CONFIG_SQUASHFS_LZO=y
+CONFIG_SQUASHFS_XZ=y
+CONFIG_NLS=y
+CONFIG_PRINTK_TIME=y
+CONFIG_BOOT_PRINTK_DELAY=y
+CONFIG_DEBUG_INFO=y
+# CONFIG_ENABLE_WARN_DEPRECATED is not set
+CONFIG_FRAME_WARN=2048
+CONFIG_STRIP_ASM_SYMS=y
+CONFIG_UNUSED_SYMBOLS=y
+CONFIG_DEBUG_FS=y
+CONFIG_HEADERS_CHECK=y
+CONFIG_MAGIC_SYSRQ=y
+# CONFIG_SCHED_DEBUG is not set
+CONFIG_DEBUG_RT_MUTEXES=y
+CONFIG_DEBUG_ATOMIC_SLEEP=y
+# CONFIG_RCU_TRACE is not set
+# CONFIG_FTRACE is not set
+# CONFIG_RUNTIME_TESTING_MENU is not set
+CONFIG_CRYPTO_CCM=y
+CONFIG_CRYPTO_GCM=y
+# CONFIG_CRYPTO_ECHAINIV is not set
+CONFIG_CRYPTO_ARC4=y
+CONFIG_CRYPTO_LZO=y
+# CONFIG_CRYPTO_HW is not set
+CONFIG_CRC_CCITT=y
+CONFIG_CRC16=y
+CONFIG_CRC_T10DIF=y
+CONFIG_LIBCRC32C=y
+CONFIG_IRQ_POLL=y
diff --git a/arch/mips/include/asm/mach-intel-mips/cpu-feature-overrides.h b/arch/mips/include/asm/mach-intel-mips/cpu-feature-overrides.h
new file mode 100644
index 000000000000..ac5f3b943d2a
--- /dev/null
+++ b/arch/mips/include/asm/mach-intel-mips/cpu-feature-overrides.h
@@ -0,0 +1,61 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * This file was derived from: include/asm-mips/cpu-features.h
+ * Copyright (C) 2003, 2004 Ralf Baechle
+ * Copyright (C) 2004 Maciej W. Rozycki
+ * Copyright (C) 2018 Intel Corporation.
+ */
+
+#ifndef __ASM_MACH_INTEL_MIPS_CPU_FEATURE_OVERRIDES_H
+#define __ASM_MACH_INTEL_MIPS_CPU_FEATURE_OVERRIDES_H
+
+#define cpu_has_tlb 1
+#define cpu_has_4kex 1
+#define cpu_has_3k_cache 0
+#define cpu_has_4k_cache 1
+#define cpu_has_tx39_cache 0
+#define cpu_has_sb1_cache 0
+#define cpu_has_fpu 0
+#define cpu_has_32fpr 0
+#define cpu_has_counter 1
+#define cpu_has_watch 1
+#define cpu_has_divec 1
+
+#define cpu_has_prefetch 1
+#define cpu_has_ejtag 1
+#define cpu_has_llsc 1
+
+#define cpu_has_mips16 0
+#define cpu_has_mdmx 0
+#define cpu_has_mips3d 0
+#define cpu_has_smartmips 0
+#define cpu_has_mmips 0
+#define cpu_has_vz 0
+
+#define cpu_has_mips32r1 1
+#define cpu_has_mips32r2 1
+#define cpu_has_mips64r1 0
+#define cpu_has_mips64r2 0
+
+#define cpu_has_dsp 1
+#define cpu_has_dsp2 0
+#define cpu_has_mipsmt 1
+
+#define cpu_has_vint 1
+#define cpu_has_veic 0
+
+#define cpu_has_64bits 0
+#define cpu_has_64bit_zero_reg 0
+#define cpu_has_64bit_gp_regs 0
+#define cpu_has_64bit_addresses 0
+
+#define cpu_has_cm2 1
+#define cpu_has_cm2_l2sync 1
+#define cpu_has_eva 1
+#define cpu_has_tlbinv 1
+
+#define cpu_dcache_line_size() 32
+#define cpu_icache_line_size() 32
+#define cpu_scache_line_size() 32
+
+#endif /* __ASM_MACH_INTEL_MIPS_CPU_FEATURE_OVERRIDES_H */
diff --git a/arch/mips/include/asm/mach-intel-mips/ioremap.h b/arch/mips/include/asm/mach-intel-mips/ioremap.h
new file mode 100644
index 000000000000..99b20ed0b457
--- /dev/null
+++ b/arch/mips/include/asm/mach-intel-mips/ioremap.h
@@ -0,0 +1,39 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2014 Lei Chuanhua <[email protected]>
+ * Copyright (C) 2018 Intel Corporation.
+ */
+#ifndef __ASM_MACH_INTEL_MIPS_IOREMAP_H
+#define __ASM_MACH_INTEL_MIPS_IOREMAP_H
+
+#include <linux/types.h>
+
+static inline phys_addr_t fixup_bigphys_addr(phys_addr_t phys_addr,
+ phys_addr_t size)
+{
+ return phys_addr;
+}
+
+/*
+ * TOP IO Space definition for SSX7 components /PCIe/ToE/Memcpy
+ * physical 0xa0000000 --> virtual 0xe0000000
+ */
+#define GRX500_TOP_IOREMAP_BASE 0xA0000000
+#define GRX500_TOP_IOREMAP_SIZE 0x20000000
+#define GRX500_TOP_IOREMAP_PHYS_VIRT_OFFSET 0x40000000
+
+static inline void __iomem *plat_ioremap(phys_addr_t offset, unsigned long size,
+ unsigned long flags)
+{
+ if (offset >= GRX500_TOP_IOREMAP_BASE &&
+ offset < (GRX500_TOP_IOREMAP_BASE + GRX500_TOP_IOREMAP_SIZE))
+ return (void __iomem *)(unsigned long)
+ (offset + GRX500_TOP_IOREMAP_PHYS_VIRT_OFFSET);
+ return NULL;
+}
+
+static inline int plat_iounmap(const volatile void __iomem *addr)
+{
+ return (unsigned long)addr >= (unsigned long)GRX500_TOP_IOREMAP_BASE;
+}
+#endif /* __ASM_MACH_INTEL_MIPS_IOREMAP_H */
diff --git a/arch/mips/include/asm/mach-intel-mips/irq.h b/arch/mips/include/asm/mach-intel-mips/irq.h
new file mode 100644
index 000000000000..12a949084856
--- /dev/null
+++ b/arch/mips/include/asm/mach-intel-mips/irq.h
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2014 Lei Chuanhua <[email protected]>
+ * Copyright (C) 2018 Intel Corporation.
+ */
+
+#ifndef __INTEL_MIPS_IRQ_H
+#define __INTEL_MIPS_IRQ_H
+
+#define MIPS_CPU_IRQ_BASE 0
+#define MIPS_GIC_IRQ_BASE (MIPS_CPU_IRQ_BASE + 8)
+
+#define NR_IRQS 256
+
+#include_next <irq.h>
+
+#endif /* __INTEL_MIPS_IRQ_H */
diff --git a/arch/mips/include/asm/mach-intel-mips/kernel-entry-init.h b/arch/mips/include/asm/mach-intel-mips/kernel-entry-init.h
new file mode 100644
index 000000000000..a30542eca9ec
--- /dev/null
+++ b/arch/mips/include/asm/mach-intel-mips/kernel-entry-init.h
@@ -0,0 +1,104 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Chris Dearman ([email protected])
+ * Leonid Yegoshin ([email protected])
+ * Copyright (C) 2012 Mips Technologies, Inc.
+ * Copyright (C) 2018 Intel Corporation.
+ */
+#ifndef __ASM_MACH_INTEL_MIPS_KERNEL_ENTRY_INIT_H
+#define __ASM_MACH_INTEL_MIPS_KERNEL_ENTRY_INIT_H
+/*
+* Prepare segments for EVA boot:
+*
+* This is in case the processor boots in legacy configuration
+* (SI_EVAReset is de-asserted and CONFIG5.K == 0) with 1GB DDR
+*
+* On entry, t1 is loaded with CP0_CONFIG
+*
+* ========================= Mappings =============================
+* Virtual memory Physical memory Mapping
+* 0x00000000 - 0x7fffffff 0x20000000 - 0x9ffffffff MUSUK (kuseg)
+* 0x80000000 - 0x9fffffff 0x80000000 - 0x9ffffffff UK (kseg0)
+* 0xa0000000 - 0xbfffffff 0x20000000 - 0x3ffffffff UK (kseg1)
+* 0xc0000000 - 0xdfffffff - MSK (kseg2)
+* 0xe0000000 - 0xffffffff 0xa0000000 - 0xbfffffff UK 2nd IO
+*
+* user space virtual: 0x00000000 ~ 0x7fffffff
+* kernel space virtual: 0x60000000 ~ 0x9fffffff
+* physical: 0x20000000 ~ 0x5fffffff (flat 1GB)
+* user/kernel space overlapped from 0x60000000 ~ 0x7fffffff (virtual)
+* where physical 0x20000000 ~ 0x2fffffff (cached and uncached)
+* virtual 0xa0000000 ~ 0xafffffff (1st IO space)
+* virtual 0xf0000000 ~ 0xffffffff (2nd IO space)
+*
+* The last 64KB of physical memory are reserved for correct HIGHMEM
+* macros arithmetics.
+* Detailed KSEG and PHYS_OFFSET and PAGE_OFFSEt adaption, refer to
+* asm/mach-intel-mips/spaces.h
+*/
+ .macro platform_eva_init
+
+ .set push
+ .set reorder
+ /*
+ * Get Config.K0 value and use it to program
+ * the segmentation registers
+ */
+ mfc0 t1, CP0_CONFIG
+ andi t1, 0x7 /* CCA */
+ move t2, t1
+ ins t2, t1, 16, 3
+ /* SegCtl0 */
+ li t0, ((MIPS_SEGCFG_UK << MIPS_SEGCFG_AM_SHIFT) | \
+ (5 << MIPS_SEGCFG_PA_SHIFT) | (2 << MIPS_SEGCFG_C_SHIFT) | \
+ (1 << MIPS_SEGCFG_EU_SHIFT)) | \
+ (((MIPS_SEGCFG_MSK << MIPS_SEGCFG_AM_SHIFT) | \
+ (0 << MIPS_SEGCFG_PA_SHIFT) | \
+ (1 << MIPS_SEGCFG_EU_SHIFT)) << 16)
+ ins t0, t1, 16, 3
+ mtc0 t0, $5, 2
+
+ /* SegCtl1 */
+ li t0, ((MIPS_SEGCFG_UK << MIPS_SEGCFG_AM_SHIFT) | \
+ (1 << MIPS_SEGCFG_PA_SHIFT) | (2 << MIPS_SEGCFG_C_SHIFT) | \
+ (1 << MIPS_SEGCFG_EU_SHIFT)) | \
+ (((MIPS_SEGCFG_UK << MIPS_SEGCFG_AM_SHIFT) | \
+ (2 << MIPS_SEGCFG_PA_SHIFT) | \
+ (1 << MIPS_SEGCFG_EU_SHIFT)) << 16)
+ ins t0, t1, 16, 3
+ mtc0 t0, $5, 3
+
+ /* SegCtl2 */
+ li t0, ((MIPS_SEGCFG_MUSUK << MIPS_SEGCFG_AM_SHIFT) | \
+ (0 << MIPS_SEGCFG_PA_SHIFT) | \
+ (1 << MIPS_SEGCFG_EU_SHIFT)) | \
+ (((MIPS_SEGCFG_MUSK << MIPS_SEGCFG_AM_SHIFT) | \
+ (0 << MIPS_SEGCFG_PA_SHIFT)/*| (2 << MIPS_SEGCFG_C_SHIFT)*/ | \
+ (1 << MIPS_SEGCFG_EU_SHIFT)) << 16)
+ ins t0, t1, 0, 3
+ mtc0 t0, $5, 4
+
+ jal mips_ihb
+ mfc0 t0, $16, 5
+ li t2, 0x40000000 /* K bit */
+ or t0, t0, t2
+ mtc0 t0, $16, 5
+ sync
+ jal mips_ihb
+
+ .set pop
+ .endm
+
+ .macro kernel_entry_setup
+ sync
+ ehb
+ platform_eva_init
+ .endm
+
+ .macro smp_slave_setup
+ sync
+ ehb
+ platform_eva_init
+ .endm
+
+#endif /* __ASM_MACH_INTEL_MIPS_KERNEL_ENTRY_INIT_H */
diff --git a/arch/mips/include/asm/mach-intel-mips/spaces.h b/arch/mips/include/asm/mach-intel-mips/spaces.h
new file mode 100644
index 000000000000..80e7b09f712c
--- /dev/null
+++ b/arch/mips/include/asm/mach-intel-mips/spaces.h
@@ -0,0 +1,27 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Author: Leonid Yegoshin ([email protected])
+ * Copyright (C) 2012 MIPS Technologies, Inc.
+ * Copyright (C) 2014 Lei Chuanhua <[email protected]>
+ * Copyright (C) 2018 Intel Corporation.
+ */
+
+#ifndef _ASM_INTEL_MIPS_SPACES_H
+#define _ASM_INTEL_MIPS_SPACES_H
+
+#define PAGE_OFFSET _AC(0x60000000, UL)
+#define PHYS_OFFSET _AC(0x20000000, UL)
+
+/* No Highmem Support */
+#define HIGHMEM_START _AC(0xffff0000, UL)
+
+#define FIXADDR_TOP ((unsigned long)(long)(int)0xcffe0000)
+
+#define IO_SIZE _AC(0x10000000, UL)
+#define IO_SHIFT _AC(0x10000000, UL)
+
+/* IO space one */
+#define __pa_symbol(x) __pa(x)
+
+#include <asm/mach-generic/spaces.h>
+#endif /* __ASM_INTEL_MIPS_SPACES_H */
diff --git a/arch/mips/include/asm/mach-intel-mips/war.h b/arch/mips/include/asm/mach-intel-mips/war.h
new file mode 100644
index 000000000000..1c95553151e1
--- /dev/null
+++ b/arch/mips/include/asm/mach-intel-mips/war.h
@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __ASM_MIPS_MACH_INTEL_MIPS_WAR_H
+#define __ASM_MIPS_MACH_INTEL_MIPS_WAR_H
+
+#define R4600_V1_INDEX_ICACHEOP_WAR 0
+#define R4600_V1_HIT_CACHEOP_WAR 0
+#define R4600_V2_HIT_CACHEOP_WAR 0
+#define R5432_CP0_INTERRUPT_WAR 0
+#define BCM1250_M3_WAR 0
+#define SIBYTE_1956_WAR 0
+#define MIPS4K_ICACHE_REFILL_WAR 0
+#define MIPS_CACHE_SYNC_WAR 0
+#define TX49XX_ICACHE_INDEX_INV_WAR 0
+#define ICACHE_REFILLS_WORKAROUND_WAR 0
+#define R10000_LLSC_WAR 0
+#define MIPS34K_MISSED_ITLB_WAR 0
+
+#endif /* __ASM_MIPS_MACH_INTEL_MIPS_WAR_H */
diff --git a/arch/mips/intel-mips/Kconfig b/arch/mips/intel-mips/Kconfig
new file mode 100644
index 000000000000..35d2ae2b5408
--- /dev/null
+++ b/arch/mips/intel-mips/Kconfig
@@ -0,0 +1,22 @@
+if INTEL_MIPS
+
+choice
+ prompt "Built-in device tree"
+ help
+ Legacy bootloaders do not pass a DTB pointer to the kernel, so
+ if a "wrapper" is not being used, the kernel will need to include
+ a device tree that matches the target board.
+
+ The builtin DTB will only be used if the firmware does not supply
+ a valid DTB.
+
+config DTB_INTEL_MIPS_NONE
+ bool "None"
+
+config DTB_INTEL_MIPS_GRX500
+ bool "Intel MIPS GRX500 Board"
+ select BUILTIN_DTB
+
+endchoice
+
+endif
diff --git a/arch/mips/intel-mips/Makefile b/arch/mips/intel-mips/Makefile
new file mode 100644
index 000000000000..9067d0dd20a0
--- /dev/null
+++ b/arch/mips/intel-mips/Makefile
@@ -0,0 +1,5 @@
+# SPDX-License-Identifier: GPL-2.0
+
+obj-y += prom.o
+obj-y += irq.o
+obj-y += time.o
diff --git a/arch/mips/intel-mips/Platform b/arch/mips/intel-mips/Platform
new file mode 100644
index 000000000000..3976788698e3
--- /dev/null
+++ b/arch/mips/intel-mips/Platform
@@ -0,0 +1,12 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# MIPS SoC platform
+#
+
+platform-$(CONFIG_INTEL_MIPS) += intel-mips/
+cflags-$(CONFIG_INTEL_MIPS) += -I$(srctree)/arch/mips/include/asm/mach-intel-mips
+ifdef CONFIG_EVA
+ load-$(CONFIG_INTEL_MIPS) = 0xffffffff60020000
+else
+ load-$(CONFIG_INTEL_MIPS) = 0xffffffff80020000
+endif
diff --git a/arch/mips/intel-mips/irq.c b/arch/mips/intel-mips/irq.c
new file mode 100644
index 000000000000..b126c98fb391
--- /dev/null
+++ b/arch/mips/intel-mips/irq.c
@@ -0,0 +1,35 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2016 Intel Corporation.
+ */
+#include <linux/init.h>
+#include <linux/irqchip.h>
+#include <linux/of_irq.h>
+#include <asm/irq.h>
+#include <asm/irq_cpu.h>
+
+void __init arch_init_irq(void)
+{
+ struct device_node *intc_node;
+
+ pr_info("EIC is %s\n", cpu_has_veic ? "on" : "off");
+ pr_info("VINT is %s\n", cpu_has_vint ? "on" : "off");
+
+ intc_node = of_find_compatible_node(NULL, NULL,
+ "mti,cpu-interrupt-controller");
+ if (!cpu_has_veic && !intc_node)
+ mips_cpu_irq_init();
+
+ irqchip_init();
+}
+
+int get_c0_perfcount_int(void)
+{
+ return gic_get_c0_perfcount_int();
+}
+EXPORT_SYMBOL_GPL(get_c0_perfcount_int);
+
+unsigned int get_c0_compare_int(void)
+{
+ return gic_get_c0_compare_int();
+}
diff --git a/arch/mips/intel-mips/prom.c b/arch/mips/intel-mips/prom.c
new file mode 100644
index 000000000000..a1b1393c13bc
--- /dev/null
+++ b/arch/mips/intel-mips/prom.c
@@ -0,0 +1,172 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2014 Lei Chuanhua <[email protected]>
+ * Copyright (C) 2016 Intel Corporation.
+ */
+#include <linux/export.h>
+#include <linux/init.h>
+#include <linux/of_platform.h>
+#include <linux/of_fdt.h>
+#include <linux/regmap.h>
+#include <linux/mfd/syscon.h>
+#include <asm/dma-coherence.h>
+#include <asm/mips-cps.h>
+#include <asm/prom.h>
+#include <asm/smp-ops.h>
+
+#define CPC_BASE_ADDR 0x12310000
+#define IOPORT_RESOURCE_START 0x10000000
+#define IOMEM_RESOURCE_START 0x10000000
+
+const char *get_system_type(void)
+{
+ return "Intel MIPS interAptiv SoC";
+}
+
+void prom_free_prom_memory(void)
+{
+}
+
+static void __init prom_init_cmdline(void)
+{
+ int i;
+ int argc;
+ char **argv;
+
+ /*
+ * If u-boot pass parameters, it is ok, however, if without u-boot
+ * JTAG or other tool has to reset all register value before it goes
+ * emulation most likely belongs to this category
+ */
+ if (fw_arg0 == 0 || fw_arg1 == 0)
+ return;
+
+ /*
+ * a0: fw_arg0 - the number of string in init cmdline
+ * a1: fw_arg1 - the address of string in init cmdline
+ *
+ * In accordance with the MIPS UHI specification,
+ * the bootloader can pass the following arguments to the kernel:
+ * - $a0: -2.
+ * - $a1: KSEG0 address of the flattened device-tree blob.
+ */
+ if (fw_arg0 == -2)
+ return;
+
+ argc = fw_arg0;
+ argv = (char **)KSEG1ADDR(fw_arg1);
+
+ arcs_cmdline[0] = '\0';
+
+ for (i = 0; i < argc; i++) {
+ char *p = (char *)KSEG1ADDR(argv[i]);
+
+ if (argv[i] && *p) {
+ strlcat(arcs_cmdline, p, sizeof(arcs_cmdline));
+ strlcat(arcs_cmdline, " ", sizeof(arcs_cmdline));
+ }
+ }
+}
+
+static int __init plat_enable_iocoherency(void)
+{
+ if (!mips_cps_numiocu(0))
+ return 0;
+
+ /* Nothing special needs to be done to enable coherency */
+ pr_info("Coherence Manager IOCU detected\n");
+ /* Second IOCU for MPE or other master access register */
+ write_gcr_reg0_base(0xa0000000);
+ write_gcr_reg0_mask(0xf8000000 | CM_GCR_REGn_MASK_CMTGT_IOCU1);
+ return 1;
+}
+
+static void __init plat_setup_iocoherency(void)
+{
+ if (plat_enable_iocoherency() &&
+ coherentio == IO_COHERENCE_DISABLED) {
+ pr_info("Hardware DMA cache coherency disabled\n");
+ return;
+ }
+ panic("This kind of IO coherency is not supported!");
+}
+
+static void free_init_pages_eva_intel(void *begin, void *end)
+{
+ free_init_pages("unused kernel", __pa_symbol((unsigned long *)begin),
+ __pa_symbol((unsigned long *)end));
+}
+
+static void plat_early_init_devtree(void)
+{
+ void *dtb = NULL;
+
+ /*
+ * Load the builtin devicetree. This causes the chosen node to be
+ * parsed resulting in our memory appearing
+ */
+ if (fw_passed_dtb) /* used by CONFIG_MIPS_APPENDED_RAW_DTB as well */
+ dtb = (void *)fw_passed_dtb;
+ else if (__dtb_start != __dtb_end)
+ dtb = (void *)__dtb_start;
+ else
+ panic("no dtb found");
+
+ if (dtb)
+ __dt_setup_arch(dtb);
+}
+
+void __init plat_mem_setup(void)
+{
+ ioport_resource.start = IOPORT_RESOURCE_START;
+ ioport_resource.end = ~0UL; /* No limit */
+ iomem_resource.start = IOMEM_RESOURCE_START;
+ iomem_resource.end = ~0UL; /* No limit */
+
+ set_io_port_base((unsigned long)KSEG1);
+
+ strlcpy(arcs_cmdline, boot_command_line, COMMAND_LINE_SIZE);
+
+ plat_early_init_devtree();
+ plat_setup_iocoherency();
+
+ if (IS_ENABLED(CONFIG_EVA))
+ free_init_pages_eva = free_init_pages_eva_intel;
+ else
+ free_init_pages_eva = 0;
+}
+
+void __init device_tree_init(void)
+{
+ unflatten_and_copy_device_tree();
+}
+
+phys_addr_t mips_cpc_default_phys_base(void)
+{
+ return CPC_BASE_ADDR;
+}
+
+void __init prom_init(void)
+{
+ prom_init_cmdline();
+
+ mips_cpc_probe();
+
+ if (!register_cps_smp_ops())
+ return;
+
+ if (!register_cmp_smp_ops())
+ return;
+
+ if (!register_vsmp_smp_ops())
+ return;
+}
+
+static int __init plat_publish_devices(void)
+{
+ if (!of_have_populated_dt())
+ return 0;
+ return of_platform_populate(NULL, of_default_bus_match_table, NULL,
+ NULL);
+}
+arch_initcall(plat_publish_devices);
diff --git a/arch/mips/intel-mips/time.c b/arch/mips/intel-mips/time.c
new file mode 100644
index 000000000000..deb462c21df6
--- /dev/null
+++ b/arch/mips/intel-mips/time.c
@@ -0,0 +1,42 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2016 Intel Corporation.
+ */
+
+#include <linux/clk.h>
+#include <linux/clk-provider.h>
+#include <linux/clocksource.h>
+#include <linux/of.h>
+
+#include <asm/time.h>
+
+void __init plat_time_init(void)
+{
+ unsigned long cpuclk;
+ struct device_node *np;
+ struct clk *clk;
+
+ of_clk_init(NULL);
+
+ np = of_get_cpu_node(0, NULL);
+ if (!np) {
+ pr_err("Failed to get CPU node\n");
+ return;
+ }
+
+ clk = of_clk_get(np, 0);
+ if (IS_ERR(clk)) {
+ pr_err("Failed to get CPU clock: %ld\n", PTR_ERR(clk));
+ return;
+ }
+
+ cpuclk = clk_get_rate(clk);
+ /* the chip resolution is the half of the clock*/
+ mips_hpt_frequency = cpuclk / 2;
+ clk_put(clk);
+
+ write_c0_compare(read_c0_count());
+ pr_info("CPU Clock: %ldHz mips_hpt_frequency %dHz\n",
+ cpuclk, mips_hpt_frequency);
+ timer_probe();
+}
--
2.11.0


2018-08-03 03:05:10

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 03/18] dt-bindings: clk: Add documentation of grx500 clock controller

From: Yixin Zhu <[email protected]>

This patch adds binding documentation for grx500 clock controller.

Signed-off-by: YiXin Zhu <[email protected]>
Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2:
- Rewrite clock driver's dt-binding document according to Rob Herring's
comments.
- Simplify device tree docoment, remove some clock description.

.../devicetree/bindings/clock/intel,grx500-clk.txt | 39 ++++++++++++++++++++++
1 file changed, 39 insertions(+)
create mode 100644 Documentation/devicetree/bindings/clock/intel,grx500-clk.txt

diff --git a/Documentation/devicetree/bindings/clock/intel,grx500-clk.txt b/Documentation/devicetree/bindings/clock/intel,grx500-clk.txt
new file mode 100644
index 000000000000..e54e1dad9196
--- /dev/null
+++ b/Documentation/devicetree/bindings/clock/intel,grx500-clk.txt
@@ -0,0 +1,39 @@
+Device Tree Clock bindings for grx500 PLL controller.
+
+This binding uses the common clock binding:
+ Documentation/devicetree/bindings/clock/clock-bindings.txt
+
+The grx500 clock controller supplies clock to various controllers within the
+SoC.
+
+Required properties for clock node
+- compatible: Should be "intel,grx500-cgu".
+- reg: physical base address of the controller and length of memory range.
+- #clock-cells: should be 1.
+
+Optional Propteries:
+- intel,osc-frequency: frequency of the osc clock.
+if missing, driver will use clock rate defined in the driver.
+
+Example: Clock controller node:
+
+ cgu: cgu@16200000 {
+ compatible = "intel,grx500-cgu", "syscon";
+ reg = <0x16200000 0x200>;
+ #clock-cells = <1>;
+ };
+
+
+Example: UART controller node that consumes the clock generated by clock
+ controller.
+
+ asc0: serial@16600000 {
+ compatible = "lantiq,asc";
+ reg = <0x16600000 0x100000>;
+ interrupt-parent = <&gic>;
+ interrupts = <GIC_SHARED 103 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SHARED 105 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SHARED 106 IRQ_TYPE_LEVEL_HIGH>;
+ clocks = <&cgu CLK_SSX4>, <&cgu GCLK_UART>;
+ clock-names = "freq", "asc";
+ };
--
2.11.0


2018-08-03 03:05:24

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 02/18] clk: intel: Add clock driver for Intel MIPS SoCs

From: Yixin Zhu <[email protected]>

This driver provides PLL clock registration as well as various clock
branches, e.g. MUX clock, gate clock, divider clock and so on.

PLLs that provide clock to DDR, CPU and peripherals are shown below:

+---------+
|--->| LCPLL3 0|--PCIe clk-->
XO | +---------+
+-----------|
| +---------+
| | 3|--PAE clk-->
|--->| PLL0B 2|--GSWIP clk-->
| | 1|--DDR clk-->DDR PHY clk-->
| | 0|--CPU1 clk--+ +-----+
| +---------+ |--->0 |
| | MUX |--CPU clk-->
| +---------+ |--->1 |
| | 0|--CPU0 clk--+ +-----+
|--->| PLLOA 1|--SSX4 clk-->
| 2|--NGI clk-->
| 3|--CBM clk-->
+---------+

Signed-off-by: Yixin Zhu <[email protected]>
Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2:
- Rewrite clock driver, add platform clock description details in
clock driver.

drivers/clk/Kconfig | 1 +
drivers/clk/Makefile | 3 +
drivers/clk/intel/Kconfig | 20 ++
drivers/clk/intel/Makefile | 7 +
drivers/clk/intel/clk-cgu-pll.c | 166 ++++++++++
drivers/clk/intel/clk-cgu-pll.h | 34 ++
drivers/clk/intel/clk-cgu.c | 470 +++++++++++++++++++++++++++
drivers/clk/intel/clk-cgu.h | 259 +++++++++++++++
drivers/clk/intel/clk-grx500.c | 168 ++++++++++
include/dt-bindings/clock/intel,grx500-clk.h | 69 ++++
10 files changed, 1197 insertions(+)
create mode 100644 drivers/clk/intel/Kconfig
create mode 100644 drivers/clk/intel/Makefile
create mode 100644 drivers/clk/intel/clk-cgu-pll.c
create mode 100644 drivers/clk/intel/clk-cgu-pll.h
create mode 100644 drivers/clk/intel/clk-cgu.c
create mode 100644 drivers/clk/intel/clk-cgu.h
create mode 100644 drivers/clk/intel/clk-grx500.c
create mode 100644 include/dt-bindings/clock/intel,grx500-clk.h

diff --git a/drivers/clk/Kconfig b/drivers/clk/Kconfig
index 721572a8c429..5e0c1597b0d3 100644
--- a/drivers/clk/Kconfig
+++ b/drivers/clk/Kconfig
@@ -281,6 +281,7 @@ source "drivers/clk/actions/Kconfig"
source "drivers/clk/bcm/Kconfig"
source "drivers/clk/hisilicon/Kconfig"
source "drivers/clk/imgtec/Kconfig"
+source "drivers/clk/intel/Kconfig"
source "drivers/clk/keystone/Kconfig"
source "drivers/clk/mediatek/Kconfig"
source "drivers/clk/meson/Kconfig"
diff --git a/drivers/clk/Makefile b/drivers/clk/Makefile
index 0bb25dd009d1..d929ca4607cf 100644
--- a/drivers/clk/Makefile
+++ b/drivers/clk/Makefile
@@ -72,6 +72,9 @@ obj-$(CONFIG_ARCH_HISI) += hisilicon/
obj-y += imgtec/
obj-$(CONFIG_ARCH_MXC) += imx/
obj-$(CONFIG_MACH_INGENIC) += ingenic/
+ifeq ($(CONFIG_COMMON_CLK), y)
+obj-y +=intel/
+endif
obj-$(CONFIG_ARCH_KEYSTONE) += keystone/
obj-$(CONFIG_MACH_LOONGSON32) += loongson1/
obj-y += mediatek/
diff --git a/drivers/clk/intel/Kconfig b/drivers/clk/intel/Kconfig
new file mode 100644
index 000000000000..c7d3fb1721fa
--- /dev/null
+++ b/drivers/clk/intel/Kconfig
@@ -0,0 +1,20 @@
+# SPDX-License-Identifier: GPL-2.0
+config INTEL_CGU_CLK
+ depends on COMMON_CLK
+ depends on INTEL_MIPS || COMPILE_TEST
+ select MFD_SYSCON
+ bool "Intel clock controller support"
+ help
+ This driver support Intel CGU (Clock Generation Unit).
+
+choice
+ prompt "SoC platform selection"
+ depends on INTEL_CGU_CLK
+ default INTEL_GRX500_CGU_CLK
+
+config INTEL_GRX500_CGU_CLK
+ bool "GRX500 CLK"
+ help
+ Clock driver of GRX500 platform.
+
+endchoice
diff --git a/drivers/clk/intel/Makefile b/drivers/clk/intel/Makefile
new file mode 100644
index 000000000000..16a0138e52c2
--- /dev/null
+++ b/drivers/clk/intel/Makefile
@@ -0,0 +1,7 @@
+# SPDX-License-Identifier: GPL-2.0
+# Makefile for intel specific clk
+
+obj-$(CONFIG_INTEL_CGU_CLK) += clk-cgu.o clk-cgu-pll.o
+ifneq ($(CONFIG_INTEL_GRX500_CGU_CLK),)
+ obj-y += clk-grx500.o
+endif
diff --git a/drivers/clk/intel/clk-cgu-pll.c b/drivers/clk/intel/clk-cgu-pll.c
new file mode 100644
index 000000000000..20759bc27e95
--- /dev/null
+++ b/drivers/clk/intel/clk-cgu-pll.c
@@ -0,0 +1,166 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2018 Intel Corporation.
+ * Zhu YiXin <[email protected]>
+ */
+
+#include <linux/clk.h>
+#include <linux/clk-provider.h>
+#include <linux/clkdev.h>
+#include <linux/mfd/syscon.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/regmap.h>
+#include <linux/slab.h>
+
+#include "clk-cgu-pll.h"
+#include "clk-cgu.h"
+
+#define to_intel_clk_pll(_hw) container_of(_hw, struct intel_clk_pll, hw)
+
+/*
+ * Calculate formula:
+ * rate = (prate * mult + (prate * frac) / frac_div) / div
+ */
+static unsigned long
+intel_pll_calc_rate(unsigned long prate, unsigned int mult,
+ unsigned int div, unsigned int frac,
+ unsigned int frac_div)
+{
+ u64 crate, frate, rate64;
+
+ rate64 = prate;
+ crate = rate64 * mult;
+
+ if (frac) {
+ frate = rate64 * frac;
+ do_div(frate, frac_div);
+ crate += frate;
+ }
+ do_div(crate, div);
+
+ return (unsigned long)crate;
+}
+
+static void
+grx500_pll_get_params(struct intel_clk_pll *pll, unsigned int *mult,
+ unsigned int *frac)
+{
+ *mult = intel_get_clk_val(pll->map, pll->reg, 2, 7);
+ *frac = intel_get_clk_val(pll->map, pll->reg, 9, 21);
+}
+
+static int intel_wait_pll_lock(struct intel_clk_pll *pll, int bit_idx)
+{
+ unsigned int val;
+
+ return regmap_read_poll_timeout(pll->map, pll->reg, val,
+ val & BIT(bit_idx), 10, 1000);
+}
+
+static unsigned long
+intel_grx500_pll_recalc_rate(struct clk_hw *hw, unsigned long prate)
+{
+ struct intel_clk_pll *pll = to_intel_clk_pll(hw);
+ unsigned int mult, frac;
+
+ grx500_pll_get_params(pll, &mult, &frac);
+
+ return intel_pll_calc_rate(prate, mult, 1, frac, BIT(20));
+}
+
+static int intel_grx500_pll_is_enabled(struct clk_hw *hw)
+{
+ struct intel_clk_pll *pll = to_intel_clk_pll(hw);
+
+ if (intel_wait_pll_lock(pll, 1)) {
+ pr_err("%s: pll: %s is not locked!\n",
+ __func__, clk_hw_get_name(hw));
+ return 0;
+ }
+
+ return intel_get_clk_val(pll->map, pll->reg, 1, 1);
+}
+
+const static struct clk_ops intel_grx500_pll_ops = {
+ .recalc_rate = intel_grx500_pll_recalc_rate,
+ .is_enabled = intel_grx500_pll_is_enabled,
+};
+
+static struct clk
+*intel_clk_register_pll(struct intel_clk_provider *ctx,
+ enum intel_pll_type type, const char *cname,
+ const char *const *pname, u8 num_parents,
+ unsigned long flags, unsigned int reg,
+ const struct intel_pll_rate_table *table,
+ unsigned int mult, unsigned int div, unsigned int frac)
+{
+ struct clk_init_data init;
+ struct intel_clk_pll *pll;
+ struct clk_hw *hw;
+ int ret, i;
+
+ if (type != pll_grx500) {
+ pr_err("%s: pll type %d not supported!\n",
+ __func__, type);
+ return ERR_PTR(-EINVAL);
+ }
+ init.name = cname;
+ init.ops = &intel_grx500_pll_ops;
+ init.flags = CLK_IS_BASIC;
+ init.parent_names = pname;
+ init.num_parents = num_parents;
+
+ pll = kzalloc(sizeof(*pll), GFP_KERNEL);
+ if (!pll)
+ return ERR_PTR(-ENOMEM);
+ pll->map = ctx->map;
+ pll->reg = reg;
+ pll->flags = flags;
+ pll->mult = mult;
+ pll->div = div;
+ pll->frac = frac;
+ pll->hw.init = &init;
+ if (table) {
+ for (i = 0; table[i].rate != 0; i++)
+ ;
+ pll->table_sz = i;
+ pll->rate_table = kmemdup(table, i * sizeof(table[0]),
+ GFP_KERNEL);
+ if (!pll->rate_table) {
+ ret = -ENOMEM;
+ goto err_free_pll;
+ }
+ }
+ hw = &pll->hw;
+ ret = clk_hw_register(NULL, hw);
+ if (ret)
+ goto err_free_pll;
+
+ return hw->clk;
+
+err_free_pll:
+ kfree(pll);
+ return ERR_PTR(ret);
+}
+
+void intel_clk_register_plls(struct intel_clk_provider *ctx,
+ struct intel_pll_clk *list, unsigned int nr_clk)
+{
+ struct clk *clk;
+ int i;
+
+ for (i = 0; i < nr_clk; i++, list++) {
+ clk = intel_clk_register_pll(ctx, list->type, list->name,
+ list->parent_names, list->num_parents,
+ list->flags, list->reg, list->rate_table,
+ list->mult, list->div, list->frac);
+ if (IS_ERR(clk)) {
+ pr_err("%s: failed to register pll: %s\n",
+ __func__, list->name);
+ continue;
+ }
+
+ intel_clk_add_lookup(ctx, clk, list->id);
+ }
+}
diff --git a/drivers/clk/intel/clk-cgu-pll.h b/drivers/clk/intel/clk-cgu-pll.h
new file mode 100644
index 000000000000..3e7cff1d5e16
--- /dev/null
+++ b/drivers/clk/intel/clk-cgu-pll.h
@@ -0,0 +1,34 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright(c) 2018 Intel Corporation.
+ * Zhu YiXin <[email protected]>
+ */
+
+#ifndef __INTEL_CLK_PLL_H
+#define __INTEL_CLK_PLL_H
+
+enum intel_pll_type {
+ pll_grx500,
+};
+
+struct intel_pll_rate_table {
+ unsigned long prate;
+ unsigned long rate;
+ unsigned int mult;
+ unsigned int div;
+ unsigned int frac;
+};
+
+struct intel_clk_pll {
+ struct clk_hw hw;
+ struct regmap *map;
+ unsigned int reg;
+ unsigned long flags;
+ unsigned int mult;
+ unsigned int div;
+ unsigned int frac;
+ unsigned int table_sz;
+ const struct intel_pll_rate_table *rate_table;
+};
+
+#endif /* __INTEL_CLK_PLL_H */
diff --git a/drivers/clk/intel/clk-cgu.c b/drivers/clk/intel/clk-cgu.c
new file mode 100644
index 000000000000..10cacbe0fbcd
--- /dev/null
+++ b/drivers/clk/intel/clk-cgu.c
@@ -0,0 +1,470 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2018 Intel Corporation.
+ * Zhu YiXin <[email protected]>
+ */
+
+#include <linux/clk.h>
+#include <linux/clk-provider.h>
+#include <linux/clkdev.h>
+#include <linux/mfd/syscon.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/regmap.h>
+#include <linux/slab.h>
+
+#include "clk-cgu-pll.h"
+#include "clk-cgu.h"
+
+#define GATE_HW_REG_STAT(reg) (reg)
+#define GATE_HW_REG_EN(reg) ((reg) + 0x4)
+#define GATE_HW_REG_DIS(reg) ((reg) + 0x8)
+
+#define to_intel_clk_mux(_hw) container_of(_hw, struct intel_clk_mux, hw)
+#define to_intel_clk_divider(_hw) \
+ container_of(_hw, struct intel_clk_divider, hw)
+#define to_intel_clk_gate(_hw) container_of(_hw, struct intel_clk_gate, hw)
+
+void intel_set_clk_val(struct regmap *map, u32 reg, u8 shift,
+ u8 width, u32 set_val)
+{
+ u32 mask = GENMASK(width + shift, shift);
+
+ regmap_update_bits(map, reg, mask, set_val << shift);
+}
+
+u32 intel_get_clk_val(struct regmap *map, u32 reg, u8 shift,
+ u8 width)
+{
+ u32 val;
+
+ if (regmap_read(map, reg, &val)) {
+ WARN_ONCE(1, "Failed to read clk reg: 0x%x\n", reg);
+ return 0;
+ }
+ val >>= shift;
+ val &= BIT(width) - 1;
+
+ return val;
+}
+
+void intel_clk_add_lookup(struct intel_clk_provider *ctx,
+ struct clk *clk, unsigned int id)
+{
+ pr_debug("Add clk: %s, id: %u\n", __clk_get_name(clk), id);
+ if (ctx->clk_data.clks && id)
+ ctx->clk_data.clks[id] = clk;
+}
+
+static struct clk
+*intel_clk_register_fixed(struct intel_clk_provider *ctx,
+ struct intel_clk_branch *list)
+{
+ if (list->div_flags & CLOCK_FLAG_VAL_INIT)
+ intel_set_clk_val(ctx->map, list->div_off, list->div_shift,
+ list->div_width, list->div_val);
+
+ return clk_register_fixed_rate(NULL, list->name, list->parent_names[0],
+ list->flags, list->mux_flags);
+}
+
+static u8 intel_clk_mux_get_parent(struct clk_hw *hw)
+{
+ struct intel_clk_mux *mux = to_intel_clk_mux(hw);
+ u32 val;
+
+ val = intel_get_clk_val(mux->map, mux->reg, mux->shift, mux->width);
+ return clk_mux_val_to_index(hw, NULL, mux->flags, val);
+}
+
+static int intel_clk_mux_set_parent(struct clk_hw *hw, u8 index)
+{
+ struct intel_clk_mux *mux = to_intel_clk_mux(hw);
+ u32 val;
+
+ val = clk_mux_index_to_val(NULL, mux->flags, index);
+ intel_set_clk_val(mux->map, mux->reg, mux->shift, mux->width, val);
+
+ return 0;
+}
+
+static int intel_clk_mux_determine_rate(struct clk_hw *hw,
+ struct clk_rate_request *req)
+{
+ struct intel_clk_mux *mux = to_intel_clk_mux(hw);
+
+ return clk_mux_determine_rate_flags(hw, req, mux->flags);
+}
+
+const static struct clk_ops intel_clk_mux_ops = {
+ .get_parent = intel_clk_mux_get_parent,
+ .set_parent = intel_clk_mux_set_parent,
+ .determine_rate = intel_clk_mux_determine_rate,
+};
+
+static struct clk
+*intel_clk_register_mux(struct intel_clk_provider *ctx,
+ struct intel_clk_branch *list)
+{
+ struct clk_init_data init;
+ struct clk_hw *hw;
+ struct intel_clk_mux *mux;
+ u32 reg = list->mux_off;
+ u8 shift = list->mux_shift;
+ u8 width = list->mux_width;
+ unsigned long cflags = list->mux_flags;
+ int ret;
+
+ mux = kzalloc(sizeof(*mux), GFP_KERNEL);
+ if (!mux)
+ return ERR_PTR(-ENOMEM);
+
+ init.name = list->name;
+ init.ops = &intel_clk_mux_ops;
+ init.flags = list->flags | CLK_IS_BASIC;
+ init.parent_names = list->parent_names;
+ init.num_parents = list->num_parents;
+
+ mux->map = ctx->map;
+ mux->reg = reg;
+ mux->shift = shift;
+ mux->width = width;
+ mux->flags = cflags;
+ mux->hw.init = &init;
+
+ hw = &mux->hw;
+ ret = clk_hw_register(NULL, hw);
+ if (ret) {
+ kfree(mux);
+ return ERR_PTR(ret);
+ }
+
+ if (cflags & CLOCK_FLAG_VAL_INIT)
+ intel_set_clk_val(ctx->map, reg, shift, width, list->mux_val);
+
+ return hw->clk;
+}
+
+static unsigned long
+intel_clk_divider_recalc_rate(struct clk_hw *hw,
+ unsigned long parent_rate)
+{
+ struct intel_clk_divider *divider = to_intel_clk_divider(hw);
+ unsigned int val;
+
+ val = intel_get_clk_val(divider->map, divider->reg,
+ divider->shift, divider->width);
+ return divider_recalc_rate(hw, parent_rate, val, divider->table,
+ divider->flags, divider->width);
+}
+
+static long
+intel_clk_divider_round_rate(struct clk_hw *hw, unsigned long rate,
+ unsigned long *prate)
+{
+ struct intel_clk_divider *divider = to_intel_clk_divider(hw);
+
+ return divider_round_rate(hw, rate, prate, divider->table,
+ divider->width, divider->flags);
+}
+
+static int
+intel_clk_divider_set_rate(struct clk_hw *hw, unsigned long rate,
+ unsigned long prate)
+{
+ struct intel_clk_divider *divider = to_intel_clk_divider(hw);
+ int value;
+
+ value = divider_get_val(rate, prate, divider->table,
+ divider->width, divider->flags);
+ if (value < 0)
+ return value;
+
+ intel_set_clk_val(divider->map, divider->reg,
+ divider->shift, divider->width, value);
+
+ return 0;
+}
+
+const static struct clk_ops intel_clk_divider_ops = {
+ .recalc_rate = intel_clk_divider_recalc_rate,
+ .round_rate = intel_clk_divider_round_rate,
+ .set_rate = intel_clk_divider_set_rate,
+};
+
+static struct clk
+*intel_clk_register_divider(struct intel_clk_provider *ctx,
+ struct intel_clk_branch *list)
+{
+ struct clk_init_data init;
+ struct clk_hw *hw;
+ struct intel_clk_divider *div;
+ u32 reg = list->div_off;
+ u8 shift = list->div_shift;
+ u8 width = list->div_width;
+ unsigned long cflags = list->div_flags;
+ int ret;
+
+ div = kzalloc(sizeof(*div), GFP_KERNEL);
+ if (!div)
+ return ERR_PTR(-ENOMEM);
+
+ init.name = list->name;
+ init.ops = &intel_clk_divider_ops;
+ init.flags = list->flags | CLK_IS_BASIC;
+ init.parent_names = &list->parent_names[0];
+ init.num_parents = 1;
+
+ div->map = ctx->map;
+ div->reg = reg;
+ div->shift = shift;
+ div->width = width;
+ div->flags = cflags;
+ div->table = list->div_table;
+ div->hw.init = &init;
+
+ hw = &div->hw;
+ ret = clk_hw_register(NULL, hw);
+ if (ret) {
+ pr_err("%s: register clk: %s failed!\n",
+ __func__, list->name);
+ kfree(div);
+ return ERR_PTR(ret);
+ }
+
+ if (cflags & CLOCK_FLAG_VAL_INIT)
+ intel_set_clk_val(ctx->map, reg, shift, width, list->div_val);
+
+ return hw->clk;
+}
+
+static struct clk
+*intel_clk_register_fixed_factor(struct intel_clk_provider *ctx,
+ struct intel_clk_branch *list)
+{
+ struct clk_hw *hw;
+
+ hw = clk_hw_register_fixed_factor(NULL, list->name,
+ list->parent_names[0], list->flags,
+ list->mult, list->div);
+ if (IS_ERR(hw))
+ return ERR_CAST(hw);
+
+ if (list->div_flags & CLOCK_FLAG_VAL_INIT)
+ intel_set_clk_val(ctx->map, list->div_off, list->div_shift,
+ list->div_width, list->div_val);
+
+ return hw->clk;
+}
+
+static int
+intel_clk_gate_enable(struct clk_hw *hw)
+{
+ struct intel_clk_gate *gate = to_intel_clk_gate(hw);
+ unsigned int reg;
+
+ if (gate->flags & GATE_CLK_VT) {
+ gate->reg = 1;
+ return 0;
+ }
+
+ if (gate->flags & GATE_CLK_HW) {
+ reg = GATE_HW_REG_EN(gate->reg);
+ } else if (gate->flags & GATE_CLK_SW) {
+ reg = gate->reg;
+ } else {
+ pr_err("%s: gate clk: %s: flag 0x%lx not supported!\n",
+ __func__, clk_hw_get_name(hw), gate->flags);
+ return 0;
+ }
+
+ intel_set_clk_val(gate->map, reg, gate->shift, 1, 1);
+
+ return 0;
+}
+
+static void
+intel_clk_gate_disable(struct clk_hw *hw)
+{
+ struct intel_clk_gate *gate = to_intel_clk_gate(hw);
+ unsigned int reg;
+ unsigned int set;
+
+ if (gate->flags & GATE_CLK_VT) {
+ gate->reg = 0;
+ return;
+ }
+
+ if (gate->flags & GATE_CLK_HW) {
+ reg = GATE_HW_REG_DIS(gate->reg);
+ set = 1;
+ } else if (gate->flags & GATE_CLK_SW) {
+ reg = gate->reg;
+ set = 0;
+ } else {
+ pr_err("%s: gate clk: %s: flag 0x%lx not supported!\n",
+ __func__, clk_hw_get_name(hw), gate->flags);
+ return;
+ }
+
+ intel_set_clk_val(gate->map, reg, gate->shift, 1, set);
+}
+
+static int
+intel_clk_gate_is_enabled(struct clk_hw *hw)
+{
+ struct intel_clk_gate *gate = to_intel_clk_gate(hw);
+ unsigned int reg;
+
+ if (gate->flags & GATE_CLK_VT)
+ return gate->reg;
+
+ if (gate->flags & GATE_CLK_HW) {
+ reg = GATE_HW_REG_STAT(gate->reg);
+ } else if (gate->flags & GATE_CLK_SW) {
+ reg = gate->reg;
+ } else {
+ pr_err("%s: gate clk: %s: flag 0x%lx not supported!\n",
+ __func__, clk_hw_get_name(hw), gate->flags);
+ return 0;
+ }
+
+ return intel_get_clk_val(gate->map, reg, gate->shift, 1);
+}
+
+const static struct clk_ops intel_clk_gate_ops = {
+ .enable = intel_clk_gate_enable,
+ .disable = intel_clk_gate_disable,
+ .is_enabled = intel_clk_gate_is_enabled,
+};
+
+static struct clk
+*intel_clk_register_gate(struct intel_clk_provider *ctx,
+ struct intel_clk_branch *list)
+{
+ struct clk_init_data init;
+ struct clk_hw *hw;
+ struct intel_clk_gate *gate;
+ u32 reg = list->gate_off;
+ u8 shift = list->gate_shift;
+ unsigned long cflags = list->gate_flags;
+ const char *pname = list->parent_names[0];
+ int ret;
+
+ gate = kzalloc(sizeof(*gate), GFP_KERNEL);
+ if (!gate)
+ return ERR_PTR(-ENOMEM);
+
+ init.name = list->name;
+ init.ops = &intel_clk_gate_ops;
+ init.flags = list->flags | CLK_IS_BASIC;
+ init.parent_names = pname ? &pname : NULL;
+ init.num_parents = pname ? 1 : 0;
+
+ gate->map = ctx->map;
+ gate->reg = reg;
+ gate->shift = shift;
+ gate->flags = cflags;
+ gate->hw.init = &init;
+
+ hw = &gate->hw;
+ ret = clk_hw_register(NULL, hw);
+ if (ret) {
+ kfree(gate);
+ return ERR_PTR(ret);
+ }
+
+ if (cflags & CLOCK_FLAG_VAL_INIT)
+ intel_set_clk_val(ctx->map, reg, shift, 1, list->gate_val);
+
+ return hw->clk;
+}
+
+void intel_clk_register_branches(struct intel_clk_provider *ctx,
+ struct intel_clk_branch *list,
+ unsigned int nr_clk)
+{
+ struct clk *clk;
+ unsigned int idx;
+
+ for (idx = 0; idx < nr_clk; idx++, list++) {
+ switch (list->type) {
+ case intel_clk_fixed:
+ clk = intel_clk_register_fixed(ctx, list);
+ break;
+ case intel_clk_mux:
+ clk = intel_clk_register_mux(ctx, list);
+ break;
+ case intel_clk_divider:
+ clk = intel_clk_register_divider(ctx, list);
+ break;
+ case intel_clk_fixed_factor:
+ clk = intel_clk_register_fixed_factor(ctx, list);
+ break;
+ case intel_clk_gate:
+ clk = intel_clk_register_gate(ctx, list);
+ break;
+ default:
+ pr_err("%s: type: %u not supported!\n",
+ __func__, list->type);
+ return;
+ }
+
+ if (IS_ERR(clk)) {
+ pr_err("%s: register clk: %s, type: %u failed!\n",
+ __func__, list->name, list->type);
+ return;
+ }
+
+ intel_clk_add_lookup(ctx, clk, list->id);
+ }
+}
+
+struct intel_clk_provider * __init
+intel_clk_init(struct device_node *np, struct regmap *map, unsigned int nr_clks)
+{
+ struct intel_clk_provider *ctx;
+ struct clk **clks;
+
+ ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+ if (!ctx)
+ return ERR_PTR(-ENOMEM);
+
+ clks = kcalloc(nr_clks, sizeof(*clks), GFP_KERNEL);
+ if (!clks) {
+ kfree(ctx);
+ return ERR_PTR(-ENOMEM);
+ }
+
+ memset_p((void **)clks, ERR_PTR(-ENOENT), nr_clks);
+ ctx->map = map;
+ ctx->clk_data.clks = clks;
+ ctx->clk_data.clk_num = nr_clks;
+ ctx->np = np;
+
+ return ctx;
+}
+
+void __init intel_clk_register_osc(struct intel_clk_provider *ctx,
+ struct intel_osc_clk *osc,
+ unsigned int nr_clks)
+{
+ u32 freq;
+ struct clk *clk;
+ int idx;
+
+ for (idx = 0; idx < nr_clks; idx++, osc++) {
+ if (!osc->dt_freq ||
+ of_property_read_u32(ctx->np, osc->dt_freq, &freq))
+ freq = osc->def_rate;
+
+ clk = clk_register_fixed_rate(NULL, osc->name, NULL, 0, freq);
+ if (IS_ERR(clk)) {
+ pr_err("%s: Failed to register clock: %s\n",
+ __func__, osc->name);
+ return;
+ }
+
+ intel_clk_add_lookup(ctx, clk, osc->id);
+ }
+}
diff --git a/drivers/clk/intel/clk-cgu.h b/drivers/clk/intel/clk-cgu.h
new file mode 100644
index 000000000000..6dc4e45fc499
--- /dev/null
+++ b/drivers/clk/intel/clk-cgu.h
@@ -0,0 +1,259 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright(c) 2018 Intel Corporation.
+ * Zhu YiXin <[email protected]>
+ */
+
+#ifndef __INTEL_CLK_H
+#define __INTEL_CLK_H
+
+#define PNAME(x) static const char *const x[] __initconst
+
+struct intel_clk_mux {
+ struct clk_hw hw;
+ struct regmap *map;
+ unsigned int reg;
+ u8 shift;
+ u8 width;
+ unsigned long flags;
+};
+
+struct intel_clk_divider {
+ struct clk_hw hw;
+ struct regmap *map;
+ unsigned int reg;
+ u8 shift;
+ u8 width;
+ unsigned long flags;
+ const struct clk_div_table *table;
+};
+
+struct intel_clk_gate {
+ struct clk_hw hw;
+ struct regmap *map;
+ unsigned int reg;
+ u8 shift;
+ unsigned long flags;
+};
+
+enum intel_clk_type {
+ intel_clk_fixed,
+ intel_clk_mux,
+ intel_clk_divider,
+ intel_clk_fixed_factor,
+ intel_clk_gate,
+};
+
+/**
+ * struct intel_clk_provider
+ * @map: regmap type base address for register.
+ * @np: device node
+ * @clk_data: array of hw clocks and clk number.
+ */
+struct intel_clk_provider {
+ struct regmap *map;
+ struct device_node *np;
+ struct clk_onecell_data clk_data;
+};
+
+/**
+ * struct intel_pll_clk
+ * @id: plaform specific id of the clock.
+ * @name: name of this pll clock.
+ * @parent_names: name of the parent clock.
+ * @num_parents: number of parents.
+ * @flags: optional flags for basic clock.
+ * @type: platform type of pll.
+ * @reg: offset of the register.
+ * @mult: init value of mulitplier.
+ * @div: init value of divider.
+ * @frac: init value of fraction.
+ * @rate_table: table of pll clock rate.
+ */
+struct intel_pll_clk {
+ unsigned int id;
+ const char *name;
+ const char *const *parent_names;
+ u8 num_parents;
+ unsigned long flags;
+ enum intel_pll_type type;
+ int reg;
+ unsigned int mult;
+ unsigned int div;
+ unsigned int frac;
+ const struct intel_pll_rate_table *rate_table;
+};
+
+#define INTEL_PLL(_id, _type, _name, _pnames, _flags, \
+ _reg, _rtable, _mult, _div, _frac) \
+ { \
+ .id = _id, \
+ .type = _type, \
+ .name = _name, \
+ .parent_names = _pnames, \
+ .num_parents = ARRAY_SIZE(_pnames), \
+ .flags = _flags, \
+ .reg = _reg, \
+ .rate_table = _rtable, \
+ .mult = _mult, \
+ .div = _div, \
+ .frac = _frac \
+ }
+
+/**
+ * struct intel_osc_clk
+ * @id: platform specific id of the clock.
+ * @name: name of the osc clock.
+ * @dt_freq: frequency node name in device tree.
+ * @def_rate: default rate of the osc clock.
+ * @flags: optional flags for basic clock.
+ */
+struct intel_osc_clk {
+ unsigned int id;
+ const char *name;
+ const char *dt_freq;
+ const u32 def_rate;
+};
+
+#define INTEL_OSC(_id, _name, _freq, _rate) \
+ { \
+ .id = _id, \
+ .name = _name, \
+ .dt_freq = _freq, \
+ .def_rate = _rate, \
+ }
+
+struct intel_clk_branch {
+ unsigned int id;
+ enum intel_clk_type type;
+ const char *name;
+ const char *const *parent_names;
+ u8 num_parents;
+ unsigned long flags;
+ unsigned int mux_off;
+ u8 mux_shift;
+ u8 mux_width;
+ unsigned long mux_flags;
+ unsigned int mux_val;
+ unsigned int div_off;
+ u8 div_shift;
+ u8 div_width;
+ unsigned long div_flags;
+ unsigned int div_val;
+ const struct clk_div_table *div_table;
+ unsigned int gate_off;
+ u8 gate_shift;
+ unsigned long gate_flags;
+ unsigned int gate_val;
+ unsigned int mult;
+ unsigned int div;
+};
+
+/* clock flags definition */
+#define CLOCK_FLAG_VAL_INIT BIT(16)
+#define GATE_CLK_HW BIT(17)
+#define GATE_CLK_SW BIT(18)
+#define GATE_CLK_VT BIT(19)
+
+#define INTEL_MUX(_id, _name, _pname, _f, _reg, \
+ _shift, _width, _cf, _v) \
+ { \
+ .id = _id, \
+ .type = intel_clk_mux, \
+ .name = _name, \
+ .parent_names = _pname, \
+ .num_parents = ARRAY_SIZE(_pname), \
+ .flags = _f, \
+ .mux_off = _reg, \
+ .mux_shift = _shift, \
+ .mux_width = _width, \
+ .mux_flags = _cf, \
+ .mux_val = _v, \
+ }
+
+#define INTEL_DIV(_id, _name, _pname, _f, _reg, \
+ _shift, _width, _cf, _v, _dtable) \
+ { \
+ .id = _id, \
+ .type = intel_clk_divider, \
+ .name = _name, \
+ .parent_names = (const char *[]) { _pname }, \
+ .num_parents = 1, \
+ .flags = _f, \
+ .div_off = _reg, \
+ .div_shift = _shift, \
+ .div_width = _width, \
+ .div_flags = _cf, \
+ .div_val = _v, \
+ .div_table = _dtable, \
+ }
+
+#define INTEL_GATE(_id, _name, _pname, _f, _reg, \
+ _shift, _cf, _v) \
+ { \
+ .id = _id, \
+ .type = intel_clk_gate, \
+ .name = _name, \
+ .parent_names = (const char *[]) { _pname }, \
+ .num_parents = !_pname ? 0 : 1, \
+ .flags = _f, \
+ .gate_off = _reg, \
+ .gate_shift = _shift, \
+ .gate_flags = _cf, \
+ .gate_val = _v, \
+ }
+
+#define INTEL_FIXED(_id, _name, _pname, _f, _reg, \
+ _shift, _width, _cf, _freq, _v) \
+ { \
+ .id = _id, \
+ .type = intel_clk_fixed, \
+ .name = _name, \
+ .parent_names = (const char *[]) { _pname }, \
+ .num_parents = !_pname ? 0 : 1, \
+ .flags = _f, \
+ .div_off = _reg, \
+ .div_shift = _shift, \
+ .div_width = _width, \
+ .div_flags = _cf, \
+ .div_val = _v, \
+ .mux_flags = _freq, \
+ }
+
+#define INTEL_FIXED_FACTOR(_id, _name, _pname, _f, _reg, \
+ _shift, _width, _cf, _v, _m, _d) \
+ { \
+ .id = _id, \
+ .type = intel_clk_fixed_factor, \
+ .name = _name, \
+ .parent_names = (const char *[]) { _pname }, \
+ .num_parents = 1, \
+ .flags = _f, \
+ .div_off = _reg, \
+ .div_shift = _shift, \
+ .div_width = _width, \
+ .div_flags = _cf, \
+ .div_val = _v, \
+ .mult = _m, \
+ .div = _d, \
+ }
+
+void intel_set_clk_val(struct regmap *map, u32 reg, u8 shift,
+ u8 width, u32 set_val);
+u32 intel_get_clk_val(struct regmap *map, u32 reg, u8 shift, u8 width);
+void intel_clk_add_lookup(struct intel_clk_provider *ctx,
+ struct clk *clk, unsigned int id);
+void __init intel_clk_of_add_provider(struct device_node *np,
+ struct intel_clk_provider *ctx);
+struct intel_clk_provider * __init
+intel_clk_init(struct device_node *np, struct regmap *map,
+ unsigned int nr_clks);
+void __init intel_clk_register_osc(struct intel_clk_provider *ctx,
+ struct intel_osc_clk *osc,
+ unsigned int nr_clks);
+void intel_clk_register_branches(struct intel_clk_provider *ctx,
+ struct intel_clk_branch *list,
+ unsigned int nr_clk);
+void intel_clk_register_plls(struct intel_clk_provider *ctx,
+ struct intel_pll_clk *list, unsigned int nr_clk);
+#endif /* __INTEL_CLK_H */
diff --git a/drivers/clk/intel/clk-grx500.c b/drivers/clk/intel/clk-grx500.c
new file mode 100644
index 000000000000..5c2546f82579
--- /dev/null
+++ b/drivers/clk/intel/clk-grx500.c
@@ -0,0 +1,168 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2018 Intel Corporation.
+ * Zhu YiXin <[email protected]>
+ */
+
+#include <linux/clk-provider.h>
+#include <linux/mfd/syscon.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/regmap.h>
+#include <linux/spinlock.h>
+#include <dt-bindings/clock/intel,grx500-clk.h>
+
+#include "clk-cgu-pll.h"
+#include "clk-cgu.h"
+
+#define PLL_DIV_WIDTH 4
+
+/* Gate1 clock shift */
+#define G_VCODEC_SHIFT 2
+#define G_DMA0_SHIFT 5
+#define G_USB0_SHIFT 6
+#define G_SPI1_SHIFT 7
+#define G_SPI0_SHIFT 8
+#define G_CBM_SHIFT 9
+#define G_EBU_SHIFT 10
+#define G_SSO_SHIFT 11
+#define G_GPTC0_SHIFT 12
+#define G_GPTC1_SHIFT 13
+#define G_GPTC2_SHIFT 14
+#define G_UART_SHIFT 17
+#define G_CPYTO_SHIFT 20
+#define G_SECPT_SHIFT 21
+#define G_TOE_SHIFT 22
+#define G_MPE_SHIFT 23
+#define G_TDM_SHIFT 25
+#define G_PAE_SHIFT 26
+#define G_USB1_SHIFT 27
+#define G_SWITCH_SHIFT 28
+
+/* Gate2 clock shift */
+#define G_PCIE0_SHIFT 1
+#define G_PCIE1_SHIFT 17
+#define G_PCIE2_SHIFT 25
+
+/* Register definition */
+#define GRX500_PLL0A_CFG0 0x0004
+#define GRX500_PLL0A_CFG1 0x0008
+#define GRX500_PLL0B_CFG0 0x0034
+#define GRX500_PLL0B_CFG1 0x0038
+#define GRX500_LCPLL_CFG0 0x0094
+#define GRX500_LCPLL_CFG1 0x0098
+#define GRX500_IF_CLK 0x00c4
+#define GRX500_CLK_GSR1 0x0120
+#define GRX500_CLK_GSR2 0x0130
+
+static const struct clk_div_table pll_div[] = {
+ {1, 2},
+ {2, 3},
+ {3, 4},
+ {4, 5},
+ {5, 6},
+ {6, 8},
+ {7, 10},
+ {8, 12},
+ {9, 16},
+ {10, 20},
+ {11, 24},
+ {12, 32},
+ {13, 40},
+ {14, 48},
+ {15, 64}
+};
+
+enum grx500_plls {
+ pll0a, pll0b, pll3,
+};
+
+PNAME(pll_p) = { "osc" };
+PNAME(cpu_p) = { "cpu0", "cpu1" };
+
+static struct intel_osc_clk grx500_osc_clks[] __initdata = {
+ INTEL_OSC(CLK_OSC, "osc", "intel,osc-frequency", 40000000),
+};
+
+static struct intel_pll_clk grx500_pll_clks[] __initdata = {
+ [pll0a] = INTEL_PLL(CLK_PLL0A, pll_grx500, "pll0a",
+ pll_p, 0, GRX500_PLL0A_CFG0, NULL, 0, 0, 0),
+ [pll0b] = INTEL_PLL(CLK_PLL0B, pll_grx500, "pll0b",
+ pll_p, 0, GRX500_PLL0B_CFG0, NULL, 0, 0, 0),
+ [pll3] = INTEL_PLL(CLK_PLL3, pll_grx500, "pll3",
+ pll_p, 0, GRX500_LCPLL_CFG0, NULL, 0, 0, 0),
+};
+
+static struct intel_clk_branch grx500_branch_clks[] __initdata = {
+ INTEL_DIV(CLK_CBM, "cbm", "pll0a", 0, GRX500_PLL0A_CFG1,
+ 0, PLL_DIV_WIDTH, 0, 0, pll_div),
+ INTEL_DIV(CLK_NGI, "ngi", "pll0a", 0, GRX500_PLL0A_CFG1,
+ 4, PLL_DIV_WIDTH, 0, 0, pll_div),
+ INTEL_DIV(CLK_SSX4, "ssx4", "pll0a", 0, GRX500_PLL0A_CFG1,
+ 8, PLL_DIV_WIDTH, 0, 0, pll_div),
+ INTEL_DIV(CLK_CPU0, "cpu0", "pll0a", 0, GRX500_PLL0A_CFG1,
+ 12, PLL_DIV_WIDTH, 0, 0, pll_div),
+ INTEL_DIV(CLK_PAE, "pae", "pll0b", 0, GRX500_PLL0B_CFG1,
+ 0, PLL_DIV_WIDTH, 0, 0, pll_div),
+ INTEL_DIV(CLK_GSWIP, "gswip", "pll0b", 0, GRX500_PLL0B_CFG1,
+ 4, PLL_DIV_WIDTH, 0, 0, pll_div),
+ INTEL_DIV(CLK_DDR, "ddr", "pll0b", 0, GRX500_PLL0B_CFG1,
+ 8, PLL_DIV_WIDTH, 0, 0, pll_div),
+ INTEL_DIV(CLK_CPU1, "cpu1", "pll0b", 0, GRX500_PLL0B_CFG1,
+ 12, PLL_DIV_WIDTH, 0, 0, pll_div),
+ INTEL_MUX(CLK_CPU, "cpu", cpu_p, CLK_SET_RATE_PARENT,
+ GRX500_PLL0A_CFG1, 29, 1, 0, 0),
+ INTEL_GATE(GCLK_DMA0, "g_dma0", NULL, 0, GRX500_CLK_GSR1,
+ G_DMA0_SHIFT, GATE_CLK_HW, 0),
+ INTEL_GATE(GCLK_USB0, "g_usb0", NULL, 0, GRX500_CLK_GSR1,
+ G_USB0_SHIFT, GATE_CLK_HW, 0),
+ INTEL_GATE(GCLK_GPTC0, "g_gptc0", NULL, 0, GRX500_CLK_GSR1,
+ G_GPTC0_SHIFT, GATE_CLK_HW, 0),
+ INTEL_GATE(GCLK_GPTC1, "g_gptc1", NULL, 0, GRX500_CLK_GSR1,
+ G_GPTC1_SHIFT, GATE_CLK_HW, 0),
+ INTEL_GATE(GCLK_GPTC2, "g_gptc2", NULL, 0, GRX500_CLK_GSR1,
+ G_GPTC2_SHIFT, GATE_CLK_HW, 0),
+ INTEL_GATE(GCLK_UART, "g_uart", NULL, 0, GRX500_CLK_GSR1,
+ G_UART_SHIFT, GATE_CLK_HW, 0),
+ INTEL_GATE(GCLK_PCIE0, "g_pcie0", NULL, 0, GRX500_CLK_GSR2,
+ G_PCIE0_SHIFT, GATE_CLK_HW, 0),
+ INTEL_GATE(GCLK_PCIE1, "g_pcie1", NULL, 0, GRX500_CLK_GSR2,
+ G_PCIE1_SHIFT, GATE_CLK_HW, 0),
+ INTEL_GATE(GCLK_PCIE2, "g_pcie2", NULL, 0, GRX500_CLK_GSR2,
+ G_PCIE2_SHIFT, GATE_CLK_HW, 0),
+ INTEL_GATE(GCLK_I2C, "g_i2c", NULL, 0, 0, 0, GATE_CLK_VT, 0),
+ INTEL_FIXED(CLK_VOICE, "voice", NULL, 0, GRX500_IF_CLK, 14, 2,
+ CLOCK_FLAG_VAL_INIT, 8192000, 2),
+ INTEL_FIXED_FACTOR(CLK_DDRPHY, "ddrphy", "ddr", 0, 0, 0,
+ 0, 0, 0, 2, 1),
+ INTEL_FIXED_FACTOR(CLK_PCIE, "pcie", "pll3", 0, 0, 0,
+ 0, 0, 0, 1, 40),
+};
+
+static void __init grx500_clk_init(struct device_node *np)
+{
+ struct intel_clk_provider *ctx;
+ struct regmap *map;
+
+ map = syscon_node_to_regmap(np);
+ if (IS_ERR(map))
+ return;
+
+ ctx = intel_clk_init(np, map, CLK_NR_CLKS);
+ if (IS_ERR(ctx)) {
+ regmap_exit(map);
+ return;
+ }
+
+ intel_clk_register_osc(ctx, grx500_osc_clks,
+ ARRAY_SIZE(grx500_osc_clks));
+ intel_clk_register_plls(ctx, grx500_pll_clks,
+ ARRAY_SIZE(grx500_pll_clks));
+ intel_clk_register_branches(ctx, grx500_branch_clks,
+ ARRAY_SIZE(grx500_branch_clks));
+ of_clk_add_provider(np, of_clk_src_onecell_get, &ctx->clk_data);
+
+ pr_debug("%s clk init done!\n", __func__);
+}
+
+CLK_OF_DECLARE(intel_grx500_cgu, "intel,grx500-cgu", grx500_clk_init);
diff --git a/include/dt-bindings/clock/intel,grx500-clk.h b/include/dt-bindings/clock/intel,grx500-clk.h
new file mode 100644
index 000000000000..eb7daecea7dc
--- /dev/null
+++ b/include/dt-bindings/clock/intel,grx500-clk.h
@@ -0,0 +1,69 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2018 Intel Corporation.
+ * Zhu YiXin <[email protected]>
+ *
+ */
+
+#ifndef __INTEL_GRX500_CLK_H
+#define __INTEL_GRX500_CLK_H
+
+/* OSC clock */
+#define CLK_OSC 1
+
+/* PLL clocks */
+#define CLK_PLL0A 2
+#define CLK_PLL0B 3
+#define CLK_PLL3 4
+
+/* clocks under pll0a */
+#define CLK_CBM 11
+#define CLK_NGI 12
+#define CLK_SSX4 13
+#define CLK_CPU0 14
+
+/* clocks under pll0b */
+#define CLK_PAE 21
+#define CLK_GSWIP 22
+#define CLK_DDR 23
+#define CLK_CPU1 24
+
+/* clocks under pll3 */
+#define CLK_PCIE 33
+
+/* clocks under gate1 */
+#define GCLK_V_CODEC 106
+#define GCLK_DMA0 107
+#define GCLK_USB0 108
+#define GCLK_SPI1 109
+#define GCLK_SPI0 110
+#define GCLK_CBM 111
+#define GCLK_EBU 112
+#define GCLK_SSO 113
+#define GCLK_GPTC0 114
+#define GCLK_GPTC1 115
+#define GCLK_GPTC2 116
+#define GCLK_UART 117
+#define GCLK_EIP97 118
+#define GCLK_EIP123 119
+#define GCLK_TOE 120
+#define GCLK_MPE 121
+#define GCLK_TDM 122
+#define GCLK_PAE 123
+#define GCLK_USB1 124
+#define GCLK_GSWIP 125
+
+/* clocks under gate2 */
+#define GCLK_PCIE0 126
+#define GCLK_PCIE1 127
+#define GCLK_PCIE2 128
+
+/* other clocks */
+#define CLK_CPU 150
+#define CLK_DDRPHY 151
+#define GCLK_I2C 152
+#define CLK_VOICE 153
+
+#define CLK_NR_CLKS 200
+
+#endif /* __INTEL_GRX500_CLK_H */
--
2.11.0


2018-08-03 03:05:31

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 06/18] MIPS: dts: Change upper case to lower case

All the upper case in unit-address and hex constants are
changed to lower case according to the Linux conventions.

Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2: None

arch/mips/boot/dts/lantiq/danube.dtsi | 42 ++++++++++++++++-----------------
arch/mips/boot/dts/lantiq/easy50712.dts | 14 +++++------
2 files changed, 28 insertions(+), 28 deletions(-)

diff --git a/arch/mips/boot/dts/lantiq/danube.dtsi b/arch/mips/boot/dts/lantiq/danube.dtsi
index 2dd950181f8a..510be63c8bdf 100644
--- a/arch/mips/boot/dts/lantiq/danube.dtsi
+++ b/arch/mips/boot/dts/lantiq/danube.dtsi
@@ -10,12 +10,12 @@
};
};

- biu@1F800000 {
+ biu@1f800000 {
#address-cells = <1>;
#size-cells = <1>;
compatible = "lantiq,biu", "simple-bus";
- reg = <0x1F800000 0x800000>;
- ranges = <0x0 0x1F800000 0x7FFFFF>;
+ reg = <0x1f800000 0x800000>;
+ ranges = <0x0 0x1f800000 0x7fffff>;

icu0: icu@80200 {
#interrupt-cells = <1>;
@@ -24,18 +24,18 @@
reg = <0x80200 0x120>;
};

- watchdog@803F0 {
+ watchdog@803f0 {
compatible = "lantiq,wdt";
- reg = <0x803F0 0x10>;
+ reg = <0x803f0 0x10>;
};
};

- sram@1F000000 {
+ sram@1f000000 {
#address-cells = <1>;
#size-cells = <1>;
compatible = "lantiq,sram";
- reg = <0x1F000000 0x800000>;
- ranges = <0x0 0x1F000000 0x7FFFFF>;
+ reg = <0x1f000000 0x800000>;
+ ranges = <0x0 0x1f000000 0x7fffff>;

eiu0: eiu@101000 {
#interrupt-cells = <1>;
@@ -66,41 +66,41 @@
#address-cells = <1>;
#size-cells = <1>;
compatible = "lantiq,fpi", "simple-bus";
- ranges = <0x0 0x10000000 0xEEFFFFF>;
- reg = <0x10000000 0xEF00000>;
+ ranges = <0x0 0x10000000 0xeefffff>;
+ reg = <0x10000000 0xef00000>;

- gptu@E100A00 {
+ gptu@e100a00 {
compatible = "lantiq,gptu-xway";
- reg = <0xE100A00 0x100>;
+ reg = <0xe100a00 0x100>;
};

- serial@E100C00 {
+ serial@e100c00 {
compatible = "lantiq,asc";
- reg = <0xE100C00 0x400>;
+ reg = <0xe100c00 0x400>;
interrupt-parent = <&icu0>;
interrupts = <112 113 114>;
};

- dma0: dma@E104100 {
+ dma0: dma@e104100 {
compatible = "lantiq,dma-xway";
- reg = <0xE104100 0x800>;
+ reg = <0xe104100 0x800>;
};

- ebu0: ebu@E105300 {
+ ebu0: ebu@e105300 {
compatible = "lantiq,ebu-xway";
- reg = <0xE105300 0x100>;
+ reg = <0xe105300 0x100>;
};

- pci0: pci@E105400 {
+ pci0: pci@e105400 {
#address-cells = <3>;
#size-cells = <2>;
#interrupt-cells = <1>;
compatible = "lantiq,pci-xway";
bus-range = <0x0 0x0>;
ranges = <0x2000000 0 0x8000000 0x8000000 0 0x2000000 /* pci memory */
- 0x1000000 0 0x00000000 0xAE00000 0 0x200000>; /* io space */
+ 0x1000000 0 0x00000000 0xae00000 0 0x200000>; /* io space */
reg = <0x7000000 0x8000 /* config space */
- 0xE105400 0x400>; /* pci bridge */
+ 0xe105400 0x400>; /* pci bridge */
};
};
};
diff --git a/arch/mips/boot/dts/lantiq/easy50712.dts b/arch/mips/boot/dts/lantiq/easy50712.dts
index c37a33962f28..1ce20b7d05cb 100644
--- a/arch/mips/boot/dts/lantiq/easy50712.dts
+++ b/arch/mips/boot/dts/lantiq/easy50712.dts
@@ -52,14 +52,14 @@
};
};

- gpio: pinmux@E100B10 {
+ gpio: pinmux@e100b10 {
compatible = "lantiq,danube-pinctrl";
pinctrl-names = "default";
pinctrl-0 = <&state_default>;

#gpio-cells = <2>;
gpio-controller;
- reg = <0xE100B10 0xA0>;
+ reg = <0xe100b10 0xa0>;

state_default: pinmux {
stp {
@@ -82,26 +82,26 @@
};
};

- etop@E180000 {
+ etop@e180000 {
compatible = "lantiq,etop-xway";
- reg = <0xE180000 0x40000>;
+ reg = <0xe180000 0x40000>;
interrupt-parent = <&icu0>;
interrupts = <73 78>;
phy-mode = "rmii";
mac-address = [ 00 11 22 33 44 55 ];
};

- stp0: stp@E100BB0 {
+ stp0: stp@e100bb0 {
#gpio-cells = <2>;
compatible = "lantiq,gpio-stp-xway";
gpio-controller;
- reg = <0xE100BB0 0x40>;
+ reg = <0xe100bb0 0x40>;

lantiq,shadow = <0xfff>;
lantiq,groups = <0x3>;
};

- pci@E105400 {
+ pci@e105400 {
lantiq,bus-clock = <33333333>;
interrupt-map-mask = <0xf800 0x0 0x0 0x7>;
interrupt-map = <
--
2.11.0


2018-08-03 03:05:41

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 04/18] MIPS: dts: Add initial support for Intel MIPS SoCs

From: Hua Ma <[email protected]>

Add dts files to support Intel MIPS SoCs:
- xrx500.dtsi is the chip dts
- easy350_anywan.dts is the board dts

Signed-off-by: Hua Ma <[email protected]>
Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2:
- New patch split from previous patch
- The memory address is changed to @20000000
- Update to obj-$(CONFIG_BUILTIN_DTB) as per commit fca3aa166422

arch/mips/boot/dts/Makefile | 1 +
arch/mips/boot/dts/intel-mips/Makefile | 4 ++
arch/mips/boot/dts/intel-mips/easy350_anywan.dts | 26 ++++++++++
arch/mips/boot/dts/intel-mips/xrx500.dtsi | 66 ++++++++++++++++++++++++
4 files changed, 97 insertions(+)
create mode 100644 arch/mips/boot/dts/intel-mips/Makefile
create mode 100644 arch/mips/boot/dts/intel-mips/easy350_anywan.dts
create mode 100644 arch/mips/boot/dts/intel-mips/xrx500.dtsi

diff --git a/arch/mips/boot/dts/Makefile b/arch/mips/boot/dts/Makefile
index 1e79cab8e269..05f52f279047 100644
--- a/arch/mips/boot/dts/Makefile
+++ b/arch/mips/boot/dts/Makefile
@@ -3,6 +3,7 @@ subdir-y += brcm
subdir-y += cavium-octeon
subdir-y += img
subdir-y += ingenic
+subdir-y += intel-mips
subdir-y += lantiq
subdir-y += mscc
subdir-y += mti
diff --git a/arch/mips/boot/dts/intel-mips/Makefile b/arch/mips/boot/dts/intel-mips/Makefile
new file mode 100644
index 000000000000..adfaabbbb07c
--- /dev/null
+++ b/arch/mips/boot/dts/intel-mips/Makefile
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
+dtb-$(CONFIG_DTB_INTEL_MIPS_GRX500) += easy350_anywan.dtb
+
+obj-$(CONFIG_BUILTIN_DTB) += $(addsuffix .o, $(dtb-y))
diff --git a/arch/mips/boot/dts/intel-mips/easy350_anywan.dts b/arch/mips/boot/dts/intel-mips/easy350_anywan.dts
new file mode 100644
index 000000000000..e5e95f90c5e7
--- /dev/null
+++ b/arch/mips/boot/dts/intel-mips/easy350_anywan.dts
@@ -0,0 +1,26 @@
+// SPDX-License-Identifier: GPL-2.0
+/dts-v1/;
+
+#include <dt-bindings/interrupt-controller/mips-gic.h>
+#include <dt-bindings/clock/intel,grx500-clk.h>
+
+#include "xrx500.dtsi"
+
+/ {
+ model = "EASY350 ANYWAN (GRX350) Main model";
+ compatible = "intel,easy350-anywan";
+
+ aliases {
+ serial0 = &asc0;
+ };
+
+ chosen {
+ bootargs = "earlycon=lantiq,0x16600000 clk_ignore_unused";
+ stdout-path = "serial0";
+ };
+
+ memory@20000000 {
+ device_type = "memory";
+ reg = <0x20000000 0x0e000000>;
+ };
+};
diff --git a/arch/mips/boot/dts/intel-mips/xrx500.dtsi b/arch/mips/boot/dts/intel-mips/xrx500.dtsi
new file mode 100644
index 000000000000..54c5f8f8b604
--- /dev/null
+++ b/arch/mips/boot/dts/intel-mips/xrx500.dtsi
@@ -0,0 +1,66 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/ {
+ #address-cells = <1>;
+ #size-cells = <1>;
+ compatible = "intel,xrx500";
+
+ cpus {
+ #address-cells = <1>;
+ #size-cells = <0>;
+
+ cpu0: cpu@0 {
+ device_type = "cpu";
+ compatible = "mti,interaptiv";
+ clocks = <&cgu CLK_CPU>;
+ reg = <0>;
+ };
+
+ cpu1: cpu@1 {
+ device_type = "cpu";
+ compatible = "mti,interaptiv";
+ reg = <1>;
+ };
+ };
+
+ cpu_intc: interrupt-controller {
+ compatible = "mti,cpu-interrupt-controller";
+
+ interrupt-controller;
+ #interrupt-cells = <1>;
+ };
+
+ gic: gic@12320000 {
+ compatible = "mti,gic";
+ reg = <0x12320000 0x20000>;
+
+ interrupt-controller;
+ #interrupt-cells = <3>;
+ /*
+ * Declare the interrupt-parent even though the mti,gic
+ * binding doesn't require it, such that the kernel can
+ * figure out that cpu_intc is the root interrupt
+ * controller & should be probed first.
+ */
+ interrupt-parent = <&cpu_intc>;
+ mti,reserved-ipi-vectors = <56 8>;
+ };
+
+ cgu: cgu@16200000 {
+ compatible = "intel,grx500-cgu", "syscon";
+ reg = <0x16200000 0x200>;
+ #clock-cells = <1>;
+ };
+
+ asc0: serial@16600000 {
+ compatible = "lantiq,asc";
+ reg = <0x16600000 0x100000>;
+
+ interrupt-parent = <&gic>;
+ interrupts = <GIC_SHARED 103 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SHARED 105 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SHARED 106 IRQ_TYPE_LEVEL_HIGH>;
+ clocks = <&cgu CLK_SSX4>, <&cgu GCLK_UART>;
+ clock-names = "freq", "asc";
+ };
+};
--
2.11.0


2018-08-03 03:05:42

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 09/18] serial: intel: Change ltq_w32_mask to asc_update_bits

ltq prefix is platform specific function, asc prefix
is more generic.

Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2: None

drivers/tty/serial/lantiq.c | 33 ++++++++++++++++++++-------------
1 file changed, 20 insertions(+), 13 deletions(-)

diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
index 836ca51460f2..e36e6a267e7a 100644
--- a/drivers/tty/serial/lantiq.c
+++ b/drivers/tty/serial/lantiq.c
@@ -114,6 +114,13 @@ struct ltq_uart_port {
unsigned int err_irq;
};

+static inline void asc_update_bits(u32 clear, u32 set, void __iomem *reg)
+{
+ u32 tmp = readl(reg);
+
+ writel((tmp & ~clear) | set, reg);
+}
+
static inline struct
ltq_uart_port *to_ltq_uart_port(struct uart_port *port)
{
@@ -164,16 +171,16 @@ lqasc_rx_chars(struct uart_port *port)
if (rsr & ASCSTATE_ANY) {
if (rsr & ASCSTATE_PE) {
port->icount.parity++;
- ltq_w32_mask(0, ASCWHBSTATE_CLRPE,
+ asc_update_bits(0, ASCWHBSTATE_CLRPE,
port->membase + LTQ_ASC_WHBSTATE);
} else if (rsr & ASCSTATE_FE) {
port->icount.frame++;
- ltq_w32_mask(0, ASCWHBSTATE_CLRFE,
+ asc_update_bits(0, ASCWHBSTATE_CLRFE,
port->membase + LTQ_ASC_WHBSTATE);
}
if (rsr & ASCSTATE_ROE) {
port->icount.overrun++;
- ltq_w32_mask(0, ASCWHBSTATE_CLRROE,
+ asc_update_bits(0, ASCWHBSTATE_CLRROE,
port->membase + LTQ_ASC_WHBSTATE);
}

@@ -253,7 +260,7 @@ lqasc_err_int(int irq, void *_port)
struct uart_port *port = (struct uart_port *)_port;
spin_lock_irqsave(&ltq_asc_lock, flags);
/* clear any pending interrupts */
- ltq_w32_mask(0, ASCWHBSTATE_CLRPE | ASCWHBSTATE_CLRFE |
+ asc_update_bits(0, ASCWHBSTATE_CLRPE | ASCWHBSTATE_CLRFE |
ASCWHBSTATE_CLRROE, port->membase + LTQ_ASC_WHBSTATE);
spin_unlock_irqrestore(&ltq_asc_lock, flags);
return IRQ_HANDLED;
@@ -305,7 +312,7 @@ lqasc_startup(struct uart_port *port)
clk_enable(ltq_port->clk);
port->uartclk = clk_get_rate(ltq_port->fpiclk);

- ltq_w32_mask(ASCCLC_DISS | ASCCLC_RMCMASK, (1 << ASCCLC_RMCOFFSET),
+ asc_update_bits(ASCCLC_DISS | ASCCLC_RMCMASK, (1 << ASCCLC_RMCOFFSET),
port->membase + LTQ_ASC_CLC);

ltq_w32(0, port->membase + LTQ_ASC_PISEL);
@@ -321,7 +328,7 @@ lqasc_startup(struct uart_port *port)
* setting enable bits
*/
wmb();
- ltq_w32_mask(0, ASCCON_M_8ASYNC | ASCCON_FEN | ASCCON_TOEN |
+ asc_update_bits(0, ASCCON_M_8ASYNC | ASCCON_FEN | ASCCON_TOEN |
ASCCON_ROEN, port->membase + LTQ_ASC_CON);

retval = request_irq(ltq_port->tx_irq, lqasc_tx_int,
@@ -365,9 +372,9 @@ lqasc_shutdown(struct uart_port *port)
free_irq(ltq_port->err_irq, port);

ltq_w32(0, port->membase + LTQ_ASC_CON);
- ltq_w32_mask(ASCRXFCON_RXFEN, ASCRXFCON_RXFFLU,
+ asc_update_bits(ASCRXFCON_RXFEN, ASCRXFCON_RXFFLU,
port->membase + LTQ_ASC_RXFCON);
- ltq_w32_mask(ASCTXFCON_TXFEN, ASCTXFCON_TXFFLU,
+ asc_update_bits(ASCTXFCON_TXFEN, ASCTXFCON_TXFFLU,
port->membase + LTQ_ASC_TXFCON);
if (!IS_ERR(ltq_port->clk))
clk_disable(ltq_port->clk);
@@ -439,7 +446,7 @@ lqasc_set_termios(struct uart_port *port,
spin_lock_irqsave(&ltq_asc_lock, flags);

/* set up CON */
- ltq_w32_mask(0, con, port->membase + LTQ_ASC_CON);
+ asc_update_bits(0, con, port->membase + LTQ_ASC_CON);

/* Set baud rate - take a divider of 2 into account */
baud = uart_get_baud_rate(port, new, old, 0, port->uartclk / 16);
@@ -447,19 +454,19 @@ lqasc_set_termios(struct uart_port *port,
divisor = divisor / 2 - 1;

/* disable the baudrate generator */
- ltq_w32_mask(ASCCON_R, 0, port->membase + LTQ_ASC_CON);
+ asc_update_bits(ASCCON_R, 0, port->membase + LTQ_ASC_CON);

/* make sure the fractional divider is off */
- ltq_w32_mask(ASCCON_FDE, 0, port->membase + LTQ_ASC_CON);
+ asc_update_bits(ASCCON_FDE, 0, port->membase + LTQ_ASC_CON);

/* set up to use divisor of 2 */
- ltq_w32_mask(ASCCON_BRS, 0, port->membase + LTQ_ASC_CON);
+ asc_update_bits(ASCCON_BRS, 0, port->membase + LTQ_ASC_CON);

/* now we can write the new baudrate into the register */
ltq_w32(divisor, port->membase + LTQ_ASC_BG);

/* turn the baudrate generator back on */
- ltq_w32_mask(0, ASCCON_R, port->membase + LTQ_ASC_CON);
+ asc_update_bits(0, ASCCON_R, port->membase + LTQ_ASC_CON);

/* enable rx */
ltq_w32(ASCWHBSTATE_SETREN, port->membase + LTQ_ASC_WHBSTATE);
--
2.11.0


2018-08-03 03:05:43

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 10/18] MIPS: lantiq: Unselect SWAP_IO_SPACE when LANTIQ is selected

SWAP_IO_SPACE macro prevents serial driver /drivers/tty/serial/lantiq.c
to use readl/writel to replace ltq_r32/w32 which are SoC or platform
specific APIs.

readl/writel are used for this serial driver to support multiple
platforms and multiple architectures. The legacy lantiq platform(Danube)
enables SWAP_IO_SPACE for supporting PCI due to some hardware bugs.

It's a little-endian bus plus PCI TX/RX swap enable impacted both data
and control path for MIPS based platforms. But it is better to let PCI
device driver to do endian swap since SWAP_IO_SPACE is a global wide macro
which potentially impacts other peripheral like USB.
ltq_r32/ltq_w32 is not impacted in other device drivers based on MIPS when
SWAP_IO_SPACE is not selected as they use non-byte swapping OS API
(__raw_read/__raw_writel).

Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2: None

arch/mips/Kconfig | 1 -
1 file changed, 1 deletion(-)

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 2d34f17f3e24..2e7e0b538e52 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -400,7 +400,6 @@ config LANTIQ
select SYS_SUPPORTS_VPE_LOADER
select SYS_HAS_EARLY_PRINTK
select GPIOLIB
- select SWAP_IO_SPACE
select BOOT_RAW
select CLKDEV_LOOKUP
select USE_OF
--
2.11.0


2018-08-03 03:05:45

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 11/18] serial: intel: Use readl/writel instead of ltq_r32/ltq_w32

Previous implementation uses platform-dependent functions
ltq_w32()/ltq_r32() to access registers. Those functions are not
available for other SoC which uses the same IP.
Change to OS provided readl()/writel() and readb()/writeb(), so
that different SoCs can use the same driver.

Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2: None

drivers/tty/serial/lantiq.c | 38 +++++++++++++++++++-------------------
1 file changed, 19 insertions(+), 19 deletions(-)

diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
index e36e6a267e7a..2e1b35b1cf4d 100644
--- a/drivers/tty/serial/lantiq.c
+++ b/drivers/tty/serial/lantiq.c
@@ -146,7 +146,7 @@ lqasc_start_tx(struct uart_port *port)
static void
lqasc_stop_rx(struct uart_port *port)
{
- ltq_w32(ASCWHBSTATE_CLRREN, port->membase + LTQ_ASC_WHBSTATE);
+ writel(ASCWHBSTATE_CLRREN, port->membase + LTQ_ASC_WHBSTATE);
}

static int
@@ -155,11 +155,11 @@ lqasc_rx_chars(struct uart_port *port)
struct tty_port *tport = &port->state->port;
unsigned int ch = 0, rsr = 0, fifocnt;

- fifocnt = ltq_r32(port->membase + LTQ_ASC_FSTAT) & ASCFSTAT_RXFFLMASK;
+ fifocnt = readl(port->membase + LTQ_ASC_FSTAT) & ASCFSTAT_RXFFLMASK;
while (fifocnt--) {
u8 flag = TTY_NORMAL;
- ch = ltq_r8(port->membase + LTQ_ASC_RBUF);
- rsr = (ltq_r32(port->membase + LTQ_ASC_STATE)
+ ch = readb(port->membase + LTQ_ASC_RBUF);
+ rsr = (readl(port->membase + LTQ_ASC_STATE)
& ASCSTATE_ANY) | UART_DUMMY_UER_RX;
tty_flip_buffer_push(tport);
port->icount.rx++;
@@ -219,10 +219,10 @@ lqasc_tx_chars(struct uart_port *port)
return;
}

- while (((ltq_r32(port->membase + LTQ_ASC_FSTAT) &
+ while (((readl(port->membase + LTQ_ASC_FSTAT) &
ASCFSTAT_TXFREEMASK) >> ASCFSTAT_TXFREEOFF) != 0) {
if (port->x_char) {
- ltq_w8(port->x_char, port->membase + LTQ_ASC_TBUF);
+ writeb(port->x_char, port->membase + LTQ_ASC_TBUF);
port->icount.tx++;
port->x_char = 0;
continue;
@@ -231,7 +231,7 @@ lqasc_tx_chars(struct uart_port *port)
if (uart_circ_empty(xmit))
break;

- ltq_w8(port->state->xmit.buf[port->state->xmit.tail],
+ writeb(port->state->xmit.buf[port->state->xmit.tail],
port->membase + LTQ_ASC_TBUF);
xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
port->icount.tx++;
@@ -247,7 +247,7 @@ lqasc_tx_int(int irq, void *_port)
unsigned long flags;
struct uart_port *port = (struct uart_port *)_port;
spin_lock_irqsave(&ltq_asc_lock, flags);
- ltq_w32(ASC_IRNCR_TIR, port->membase + LTQ_ASC_IRNCR);
+ writel(ASC_IRNCR_TIR, port->membase + LTQ_ASC_IRNCR);
spin_unlock_irqrestore(&ltq_asc_lock, flags);
lqasc_start_tx(port);
return IRQ_HANDLED;
@@ -272,7 +272,7 @@ lqasc_rx_int(int irq, void *_port)
unsigned long flags;
struct uart_port *port = (struct uart_port *)_port;
spin_lock_irqsave(&ltq_asc_lock, flags);
- ltq_w32(ASC_IRNCR_RIR, port->membase + LTQ_ASC_IRNCR);
+ writel(ASC_IRNCR_RIR, port->membase + LTQ_ASC_IRNCR);
lqasc_rx_chars(port);
spin_unlock_irqrestore(&ltq_asc_lock, flags);
return IRQ_HANDLED;
@@ -282,7 +282,7 @@ static unsigned int
lqasc_tx_empty(struct uart_port *port)
{
int status;
- status = ltq_r32(port->membase + LTQ_ASC_FSTAT) & ASCFSTAT_TXFFLMASK;
+ status = readl(port->membase + LTQ_ASC_FSTAT) & ASCFSTAT_TXFFLMASK;
return status ? 0 : TIOCSER_TEMT;
}

@@ -315,12 +315,12 @@ lqasc_startup(struct uart_port *port)
asc_update_bits(ASCCLC_DISS | ASCCLC_RMCMASK, (1 << ASCCLC_RMCOFFSET),
port->membase + LTQ_ASC_CLC);

- ltq_w32(0, port->membase + LTQ_ASC_PISEL);
- ltq_w32(
+ writel(0, port->membase + LTQ_ASC_PISEL);
+ writel(
((TXFIFO_FL << ASCTXFCON_TXFITLOFF) & ASCTXFCON_TXFITLMASK) |
ASCTXFCON_TXFEN | ASCTXFCON_TXFFLU,
port->membase + LTQ_ASC_TXFCON);
- ltq_w32(
+ writel(
((RXFIFO_FL << ASCRXFCON_RXFITLOFF) & ASCRXFCON_RXFITLMASK)
| ASCRXFCON_RXFEN | ASCRXFCON_RXFFLU,
port->membase + LTQ_ASC_RXFCON);
@@ -352,7 +352,7 @@ lqasc_startup(struct uart_port *port)
goto err2;
}

- ltq_w32(ASC_IRNREN_RX | ASC_IRNREN_ERR | ASC_IRNREN_TX,
+ writel(ASC_IRNREN_RX | ASC_IRNREN_ERR | ASC_IRNREN_TX,
port->membase + LTQ_ASC_IRNREN);
return 0;

@@ -371,7 +371,7 @@ lqasc_shutdown(struct uart_port *port)
free_irq(ltq_port->rx_irq, port);
free_irq(ltq_port->err_irq, port);

- ltq_w32(0, port->membase + LTQ_ASC_CON);
+ writel(0, port->membase + LTQ_ASC_CON);
asc_update_bits(ASCRXFCON_RXFEN, ASCRXFCON_RXFFLU,
port->membase + LTQ_ASC_RXFCON);
asc_update_bits(ASCTXFCON_TXFEN, ASCTXFCON_TXFFLU,
@@ -463,13 +463,13 @@ lqasc_set_termios(struct uart_port *port,
asc_update_bits(ASCCON_BRS, 0, port->membase + LTQ_ASC_CON);

/* now we can write the new baudrate into the register */
- ltq_w32(divisor, port->membase + LTQ_ASC_BG);
+ writel(divisor, port->membase + LTQ_ASC_BG);

/* turn the baudrate generator back on */
asc_update_bits(0, ASCCON_R, port->membase + LTQ_ASC_CON);

/* enable rx */
- ltq_w32(ASCWHBSTATE_SETREN, port->membase + LTQ_ASC_WHBSTATE);
+ writel(ASCWHBSTATE_SETREN, port->membase + LTQ_ASC_WHBSTATE);

spin_unlock_irqrestore(&ltq_asc_lock, flags);

@@ -580,10 +580,10 @@ lqasc_console_putchar(struct uart_port *port, int ch)
return;

do {
- fifofree = (ltq_r32(port->membase + LTQ_ASC_FSTAT)
+ fifofree = (readl(port->membase + LTQ_ASC_FSTAT)
& ASCFSTAT_TXFREEMASK) >> ASCFSTAT_TXFREEOFF;
} while (fifofree == 0);
- ltq_w8(ch, port->membase + LTQ_ASC_TBUF);
+ writeb(ch, port->membase + LTQ_ASC_TBUF);
}

static void lqasc_serial_port_write(struct uart_port *port, const char *s,
--
2.11.0


2018-08-03 03:05:45

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 05/18] dt-binding: MIPS: Add documentation of Intel MIPS SoCs

From: Hua Ma <[email protected]>

This patch adds binding documentation for the
compatible values of the Intel MIPS SoCs.

Signed-off-by: Hua Ma <[email protected]>
Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2:
- New patch split from previous patch
- Add the board and chip compatible in dt document

Documentation/devicetree/bindings/mips/intel.txt | 17 +++++++++++++++++
1 file changed, 17 insertions(+)
create mode 100644 Documentation/devicetree/bindings/mips/intel.txt

diff --git a/Documentation/devicetree/bindings/mips/intel.txt b/Documentation/devicetree/bindings/mips/intel.txt
new file mode 100644
index 000000000000..ac594ef303b7
--- /dev/null
+++ b/Documentation/devicetree/bindings/mips/intel.txt
@@ -0,0 +1,17 @@
+Intel MIPS SoC device tree bindings
+
+1, SoCs
+
+Each device tree must specify a compatible value for the Intel SoC
+it uses in the compatible property of the root node. The compatible
+value must be one of the following values:
+
+ intel,xrx500
+
+2, Boards
+
+Each device tree must specify a compatible value for the Intel Board
+it uses in the compatible property of the root node. The compatible
+value must be one of the following values:
+
+ intel,easy350-anywan
--
2.11.0


2018-08-03 03:05:51

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 13/18] serial: intel: Replace clk_enable/clk_disable with clk generic API

The clk driver has introduced new clock APIs that replace
the existing clk_enable and clk_disable.
-clk_enable() APIs is replaced with clk_prepare_enable().
-clk_disable() API is replaced with clk_disable_unprepare().

Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2: None

drivers/tty/serial/lantiq.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
index 28086d52e980..36479d66fb7c 100644
--- a/drivers/tty/serial/lantiq.c
+++ b/drivers/tty/serial/lantiq.c
@@ -309,7 +309,7 @@ lqasc_startup(struct uart_port *port)
int retval;

if (!IS_ERR(ltq_port->clk))
- clk_enable(ltq_port->clk);
+ clk_prepare_enable(ltq_port->clk);
port->uartclk = clk_get_rate(ltq_port->freqclk);

asc_update_bits(ASCCLC_DISS | ASCCLC_RMCMASK, (1 << ASCCLC_RMCOFFSET),
@@ -377,7 +377,7 @@ lqasc_shutdown(struct uart_port *port)
asc_update_bits(ASCTXFCON_TXFEN, ASCTXFCON_TXFFLU,
port->membase + LTQ_ASC_TXFCON);
if (!IS_ERR(ltq_port->clk))
- clk_disable(ltq_port->clk);
+ clk_disable_unprepare(ltq_port->clk);
}

static void
@@ -631,7 +631,7 @@ lqasc_console_setup(struct console *co, char *options)
port = &ltq_port->port;

if (!IS_ERR(ltq_port->clk))
- clk_enable(ltq_port->clk);
+ clk_prepare_enable(ltq_port->clk);

port->uartclk = clk_get_rate(ltq_port->freqclk);

--
2.11.0


2018-08-03 03:05:56

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 14/18] serial: intel: Add CCF support

Previous implementation uses platform-dependent API to get the clock.
Those functions are not available for other SoC which uses the same IP.
The CCF (Common Clock Framework) have an abstraction based APIs for
clock. In future, the platform specific code will be removed when the
legacy soc use CCF as well.
Change to use CCF APIs to get clock and rate. So that different SoCs
can use the same driver.

Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2: None

drivers/tty/serial/lantiq.c | 11 +++++++++++
1 file changed, 11 insertions(+)

diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
index 36479d66fb7c..35518ab3a80d 100644
--- a/drivers/tty/serial/lantiq.c
+++ b/drivers/tty/serial/lantiq.c
@@ -26,7 +26,9 @@
#include <linux/clk.h>
#include <linux/gpio.h>

+#ifdef CONFIG_LANTIQ
#include <lantiq_soc.h>
+#endif

#define PORT_LTQ_ASC 111
#define MAXPORTS 2
@@ -744,14 +746,23 @@ lqasc_probe(struct platform_device *pdev)
port->irq = irqres[0].start;
port->mapbase = mmres->start;

+#if (IS_ENABLED(CONFIG_LANTIQ) && !IS_ENABLED(CONFIG_COMMON_CLK))
ltq_port->freqclk = clk_get_fpi();
+#else
+ ltq_port->freqclk = devm_clk_get(&pdev->dev, "freq");
+#endif
+
if (IS_ERR(ltq_port->freqclk)) {
pr_err("failed to get fpi clk\n");
return -ENOENT;
}

/* not all asc ports have clock gates, lets ignore the return code */
+#if (IS_ENABLED(CONFIG_LANTIQ) && !IS_ENABLED(CONFIG_COMMON_CLK))
ltq_port->clk = clk_get(&pdev->dev, NULL);
+#else
+ ltq_port->clk = devm_clk_get(&pdev->dev, "asc");
+#endif

ltq_port->tx_irq = irqres[0].start;
ltq_port->rx_irq = irqres[1].start;
--
2.11.0


2018-08-03 03:06:07

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 17/18] serial: intel: Change init_lqasc to static declaration

init_lqasc() is only used internally, change to static declaration.

Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2: None

drivers/tty/serial/lantiq.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
index 804aad60ed80..2bb8e37e6886 100644
--- a/drivers/tty/serial/lantiq.c
+++ b/drivers/tty/serial/lantiq.c
@@ -788,7 +788,7 @@ static struct platform_driver lqasc_driver = {
},
};

-int __init
+static int __init
init_lqasc(void)
{
int ret;
--
2.11.0


2018-08-03 03:06:09

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 07/18] MIPS: dts: Add aliases node for lantiq danube serial

Previous implementation uses a hard-coded register value to check
if the current serial entity is the console entity.
Now the lantiq serial driver uses the aliases for the index of the
serial port.
The lantiq danube serial dts are updated with aliases to support this.

Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2: None

arch/mips/boot/dts/lantiq/danube.dtsi | 2 +-
arch/mips/boot/dts/lantiq/easy50712.dts | 4 ++++
2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/arch/mips/boot/dts/lantiq/danube.dtsi b/arch/mips/boot/dts/lantiq/danube.dtsi
index 510be63c8bdf..73746d7577d7 100644
--- a/arch/mips/boot/dts/lantiq/danube.dtsi
+++ b/arch/mips/boot/dts/lantiq/danube.dtsi
@@ -74,7 +74,7 @@
reg = <0xe100a00 0x100>;
};

- serial@e100c00 {
+ asc1: serial@e100c00 {
compatible = "lantiq,asc";
reg = <0xe100c00 0x400>;
interrupt-parent = <&icu0>;
diff --git a/arch/mips/boot/dts/lantiq/easy50712.dts b/arch/mips/boot/dts/lantiq/easy50712.dts
index 1ce20b7d05cb..452860ca1868 100644
--- a/arch/mips/boot/dts/lantiq/easy50712.dts
+++ b/arch/mips/boot/dts/lantiq/easy50712.dts
@@ -4,6 +4,10 @@
/include/ "danube.dtsi"

/ {
+ aliases {
+ serial0 = &asc1;
+ };
+
chosen {
bootargs = "console=ttyLTQ0,115200 init=/etc/preinit";
};
--
2.11.0


2018-08-03 03:06:11

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 08/18] serial: intel: Get serial id from dts

Get serial id from dts.

"#ifdef CONFIG_LANTIQ" preprocessor is used because LTQ_EARLY_ASC
macro is defined in lantiq_soc.h.
lantiq_soc.h is in arch path for legacy product support.

arch/mips/include/asm/mach-lantiq/xway/lantiq_soc.h

If "#ifdef preprocessor" is changed to
"if (IS_ENABLED(CONFIG_LANTIQ))", when CONFIG_LANTIQ is not enabled,
code using LTQ_EARLY_ASC is compiled.
Compilation will fail for no LTQ_EARLY_ASC defined.

Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2: None

drivers/tty/serial/lantiq.c | 19 +++++++++++++++----
1 file changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
index 044128277248..836ca51460f2 100644
--- a/drivers/tty/serial/lantiq.c
+++ b/drivers/tty/serial/lantiq.c
@@ -6,6 +6,7 @@
* Copyright (C) 2007 Felix Fietkau <[email protected]>
* Copyright (C) 2007 John Crispin <[email protected]>
* Copyright (C) 2010 Thomas Langer, <[email protected]>
+ * Copyright (C) 2018 Intel Corporation.
*/

#include <linux/slab.h>
@@ -688,7 +689,7 @@ lqasc_probe(struct platform_device *pdev)
struct ltq_uart_port *ltq_port;
struct uart_port *port;
struct resource *mmres, irqres[3];
- int line = 0;
+ int line;
int ret;

mmres = platform_get_resource(pdev, IORESOURCE_MEM, 0);
@@ -699,9 +700,19 @@ lqasc_probe(struct platform_device *pdev)
return -ENODEV;
}

- /* check if this is the console port */
- if (mmres->start != CPHYSADDR(LTQ_EARLY_ASC))
- line = 1;
+ /* get serial id */
+ line = of_alias_get_id(node, "serial");
+ if (line < 0) {
+#ifdef CONFIG_LANTIQ
+ if (mmres->start == CPHYSADDR(LTQ_EARLY_ASC))
+ line = 0;
+ else
+ line = 1;
+#else
+ dev_err(&pdev->dev, "failed to get alias id, errno %d\n", line);
+ return line;
+#endif
+ }

if (lqasc_port[line]) {
dev_err(&pdev->dev, "port %d already allocated\n", line);
--
2.11.0


2018-08-03 03:06:30

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 15/18] serial: intel: Support more platform

Support more platform.

Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2: None

drivers/tty/serial/Kconfig | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/tty/serial/Kconfig b/drivers/tty/serial/Kconfig
index df8bd0c7b97d..564c71fc24bb 100644
--- a/drivers/tty/serial/Kconfig
+++ b/drivers/tty/serial/Kconfig
@@ -1062,7 +1062,7 @@ config SERIAL_OMAP_CONSOLE

config SERIAL_LANTIQ
bool "Lantiq serial driver"
- depends on LANTIQ
+ depends on LANTIQ || INTEL_MIPS || COMPILE_TEST
select SERIAL_CORE
select SERIAL_CORE_CONSOLE
select SERIAL_EARLYCON
--
2.11.0


2018-08-03 03:06:31

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 16/18] serial: intel: Reorder the head files

Reorder the head files according to the coding style.

Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2:
- New patch to reorder the head files according to the coding style.

drivers/tty/serial/lantiq.c | 24 ++++++++++++------------
1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
index 35518ab3a80d..804aad60ed80 100644
--- a/drivers/tty/serial/lantiq.c
+++ b/drivers/tty/serial/lantiq.c
@@ -9,22 +9,22 @@
* Copyright (C) 2018 Intel Corporation.
*/

-#include <linux/slab.h>
-#include <linux/ioport.h>
-#include <linux/init.h>
+#include <linux/clk.h>
#include <linux/console.h>
-#include <linux/sysrq.h>
#include <linux/device.h>
-#include <linux/tty.h>
-#include <linux/tty_flip.h>
-#include <linux/serial_core.h>
-#include <linux/serial.h>
-#include <linux/of_platform.h>
+#include <linux/gpio.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/ioport.h>
#include <linux/of_address.h>
#include <linux/of_irq.h>
-#include <linux/io.h>
-#include <linux/clk.h>
-#include <linux/gpio.h>
+#include <linux/of_platform.h>
+#include <linux/serial.h>
+#include <linux/serial_core.h>
+#include <linux/slab.h>
+#include <linux/sysrq.h>
+#include <linux/tty.h>
+#include <linux/tty_flip.h>

#ifdef CONFIG_LANTIQ
#include <lantiq_soc.h>
--
2.11.0


2018-08-03 03:06:36

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 18/18] dt-bindings: serial: lantiq: Add optional properties for CCF

Clocks and clock-names are updated in device tree binding.

Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2: None

Documentation/devicetree/bindings/serial/lantiq_asc.txt | 15 +++++++++++++++
1 file changed, 15 insertions(+)

diff --git a/Documentation/devicetree/bindings/serial/lantiq_asc.txt b/Documentation/devicetree/bindings/serial/lantiq_asc.txt
index 3acbd309ab9d..40e81a5818f6 100644
--- a/Documentation/devicetree/bindings/serial/lantiq_asc.txt
+++ b/Documentation/devicetree/bindings/serial/lantiq_asc.txt
@@ -6,8 +6,23 @@ Required properties:
- interrupts: the 3 (tx rx err) interrupt numbers. The interrupt specifier
depends on the interrupt-parent interrupt controller.

+Optional properties:
+- clocks: Should contain frequency clock and gate clock
+- clock-names: Should be "freq" and "asc"
+
Example:

+asc0: serial@16600000 {
+ compatible = "lantiq,asc";
+ reg = <0x16600000 0x100000>;
+ interrupt-parent = <&gic>;
+ interrupts = <GIC_SHARED 103 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SHARED 105 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SHARED 106 IRQ_TYPE_LEVEL_HIGH>;
+ clocks = <&cgu CLK_SSX4>, <&cgu GCLK_UART>;
+ clock-names = "freq", "asc";
+};
+
asc1: serial@e100c00 {
compatible = "lantiq,asc";
reg = <0xE100C00 0x400>;
--
2.11.0


2018-08-03 03:07:22

by Songjun Wu

[permalink] [raw]
Subject: [PATCH v2 12/18] serial: intel: Rename fpiclk to freqclk

Rename fpiclk to freqclk.

Signed-off-by: Songjun Wu <[email protected]>
---

Changes in v2: None

drivers/tty/serial/lantiq.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
index 2e1b35b1cf4d..28086d52e980 100644
--- a/drivers/tty/serial/lantiq.c
+++ b/drivers/tty/serial/lantiq.c
@@ -106,7 +106,7 @@ static DEFINE_SPINLOCK(ltq_asc_lock);
struct ltq_uart_port {
struct uart_port port;
/* clock used to derive divider */
- struct clk *fpiclk;
+ struct clk *freqclk;
/* clock gating of the ASC core */
struct clk *clk;
unsigned int tx_irq;
@@ -310,7 +310,7 @@ lqasc_startup(struct uart_port *port)

if (!IS_ERR(ltq_port->clk))
clk_enable(ltq_port->clk);
- port->uartclk = clk_get_rate(ltq_port->fpiclk);
+ port->uartclk = clk_get_rate(ltq_port->freqclk);

asc_update_bits(ASCCLC_DISS | ASCCLC_RMCMASK, (1 << ASCCLC_RMCOFFSET),
port->membase + LTQ_ASC_CLC);
@@ -633,7 +633,7 @@ lqasc_console_setup(struct console *co, char *options)
if (!IS_ERR(ltq_port->clk))
clk_enable(ltq_port->clk);

- port->uartclk = clk_get_rate(ltq_port->fpiclk);
+ port->uartclk = clk_get_rate(ltq_port->freqclk);

if (options)
uart_parse_options(options, &baud, &parity, &bits, &flow);
@@ -744,8 +744,8 @@ lqasc_probe(struct platform_device *pdev)
port->irq = irqres[0].start;
port->mapbase = mmres->start;

- ltq_port->fpiclk = clk_get_fpi();
- if (IS_ERR(ltq_port->fpiclk)) {
+ ltq_port->freqclk = clk_get_fpi();
+ if (IS_ERR(ltq_port->freqclk)) {
pr_err("failed to get fpi clk\n");
return -ENOENT;
}
--
2.11.0


2018-08-03 05:44:08

by Greg KH

[permalink] [raw]
Subject: Re: [PATCH v2 08/18] serial: intel: Get serial id from dts

On Fri, Aug 03, 2018 at 11:02:27AM +0800, Songjun Wu wrote:
> Get serial id from dts.
>
> "#ifdef CONFIG_LANTIQ" preprocessor is used because LTQ_EARLY_ASC
> macro is defined in lantiq_soc.h.
> lantiq_soc.h is in arch path for legacy product support.
>
> arch/mips/include/asm/mach-lantiq/xway/lantiq_soc.h
>
> If "#ifdef preprocessor" is changed to
> "if (IS_ENABLED(CONFIG_LANTIQ))", when CONFIG_LANTIQ is not enabled,
> code using LTQ_EARLY_ASC is compiled.
> Compilation will fail for no LTQ_EARLY_ASC defined.
>
> Signed-off-by: Songjun Wu <[email protected]>
> ---
>
> Changes in v2: None
>
> drivers/tty/serial/lantiq.c | 19 +++++++++++++++----
> 1 file changed, 15 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
> index 044128277248..836ca51460f2 100644
> --- a/drivers/tty/serial/lantiq.c
> +++ b/drivers/tty/serial/lantiq.c
> @@ -6,6 +6,7 @@
> * Copyright (C) 2007 Felix Fietkau <[email protected]>
> * Copyright (C) 2007 John Crispin <[email protected]>
> * Copyright (C) 2010 Thomas Langer, <[email protected]>
> + * Copyright (C) 2018 Intel Corporation.

Your changes here do not warrent the addition of a copyright line, don't
you agree? If not, please get a signed-off-by from your corporate
lawyer who does this this is warrented when you resend this patch.

thanks,

greg k-h

2018-08-03 05:57:54

by Greg KH

[permalink] [raw]
Subject: Re: [PATCH v2 14/18] serial: intel: Add CCF support

On Fri, Aug 03, 2018 at 11:02:33AM +0800, Songjun Wu wrote:
> Previous implementation uses platform-dependent API to get the clock.
> Those functions are not available for other SoC which uses the same IP.
> The CCF (Common Clock Framework) have an abstraction based APIs for
> clock. In future, the platform specific code will be removed when the
> legacy soc use CCF as well.
> Change to use CCF APIs to get clock and rate. So that different SoCs
> can use the same driver.
>
> Signed-off-by: Songjun Wu <[email protected]>
> ---
>
> Changes in v2: None
>
> drivers/tty/serial/lantiq.c | 11 +++++++++++
> 1 file changed, 11 insertions(+)
>
> diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
> index 36479d66fb7c..35518ab3a80d 100644
> --- a/drivers/tty/serial/lantiq.c
> +++ b/drivers/tty/serial/lantiq.c
> @@ -26,7 +26,9 @@
> #include <linux/clk.h>
> #include <linux/gpio.h>
>
> +#ifdef CONFIG_LANTIQ
> #include <lantiq_soc.h>
> +#endif

That is never how you do this in Linux, you know better.

Please go and get this patchset reviewed and signed-off-by from other
internal Intel kernel developers before resending it next time. It is
their job to find and fix your basic errors like this, not ours.

greg k-h

2018-08-03 05:58:18

by Greg KH

[permalink] [raw]
Subject: Re: [PATCH v2 15/18] serial: intel: Support more platform

On Fri, Aug 03, 2018 at 11:02:34AM +0800, Songjun Wu wrote:
> Support more platform.
>
> Signed-off-by: Songjun Wu <[email protected]>
> ---

Your changelog text makes no sense, sorry.

greg k-h

2018-08-03 07:24:47

by Songjun Wu

[permalink] [raw]
Subject: Re: [PATCH v2 15/18] serial: intel: Support more platform



On 8/3/2018 1:57 PM, Greg Kroah-Hartman wrote:
> On Fri, Aug 03, 2018 at 11:02:34AM +0800, Songjun Wu wrote:
>> Support more platform.
>>
>> Signed-off-by: Songjun Wu <[email protected]>
>> ---
> Your changelog text makes no sense, sorry.
Thanks for your comment.
I will describe it more clearly.

2018-08-03 07:34:57

by Songjun Wu

[permalink] [raw]
Subject: Re: [PATCH v2 14/18] serial: intel: Add CCF support



On 8/3/2018 1:56 PM, Greg Kroah-Hartman wrote:
> On Fri, Aug 03, 2018 at 11:02:33AM +0800, Songjun Wu wrote:
>> Previous implementation uses platform-dependent API to get the clock.
>> Those functions are not available for other SoC which uses the same IP.
>> The CCF (Common Clock Framework) have an abstraction based APIs for
>> clock. In future, the platform specific code will be removed when the
>> legacy soc use CCF as well.
>> Change to use CCF APIs to get clock and rate. So that different SoCs
>> can use the same driver.
>>
>> Signed-off-by: Songjun Wu <[email protected]>
>> ---
>>
>> Changes in v2: None
>>
>> drivers/tty/serial/lantiq.c | 11 +++++++++++
>> 1 file changed, 11 insertions(+)
>>
>> diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
>> index 36479d66fb7c..35518ab3a80d 100644
>> --- a/drivers/tty/serial/lantiq.c
>> +++ b/drivers/tty/serial/lantiq.c
>> @@ -26,7 +26,9 @@
>> #include <linux/clk.h>
>> #include <linux/gpio.h>
>>
>> +#ifdef CONFIG_LANTIQ
>> #include <lantiq_soc.h>
>> +#endif
> That is never how you do this in Linux, you know better.
>
> Please go and get this patchset reviewed and signed-off-by from other
> internal Intel kernel developers before resending it next time. It is
> their job to find and fix your basic errors like this, not ours.
Thank you for your comment.
Actually, we have discussed this issue internally.
We put the reason why we use "#ifdef CONFIG_LANTIQ" preprocessor in commit
message in "[PATCH v2 08/18] serial: intel: Get serial id from dts".
Please refer the commit message below.

"#ifdef CONFIG_LANTIQ" preprocessor is used because LTQ_EARLY_ASC
macro is defined in lantiq_soc.h.
lantiq_soc.h is in arch path for legacy product support.

arch/mips/include/asm/mach-lantiq/xway/lantiq_soc.h

If "#ifdef preprocessor" is changed to
"if (IS_ENABLED(CONFIG_LANTIQ))", when CONFIG_LANTIQ is not enabled,
code using LTQ_EARLY_ASC is compiled.
Compilation will fail for no LTQ_EARLY_ASC defined.


2018-08-03 10:31:31

by Greg KH

[permalink] [raw]
Subject: Re: [PATCH v2 14/18] serial: intel: Add CCF support

On Fri, Aug 03, 2018 at 03:33:38PM +0800, Wu, Songjun wrote:
>
>
> On 8/3/2018 1:56 PM, Greg Kroah-Hartman wrote:
> > On Fri, Aug 03, 2018 at 11:02:33AM +0800, Songjun Wu wrote:
> > > Previous implementation uses platform-dependent API to get the clock.
> > > Those functions are not available for other SoC which uses the same IP.
> > > The CCF (Common Clock Framework) have an abstraction based APIs for
> > > clock. In future, the platform specific code will be removed when the
> > > legacy soc use CCF as well.
> > > Change to use CCF APIs to get clock and rate. So that different SoCs
> > > can use the same driver.
> > >
> > > Signed-off-by: Songjun Wu <[email protected]>
> > > ---
> > >
> > > Changes in v2: None
> > >
> > > drivers/tty/serial/lantiq.c | 11 +++++++++++
> > > 1 file changed, 11 insertions(+)
> > >
> > > diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
> > > index 36479d66fb7c..35518ab3a80d 100644
> > > --- a/drivers/tty/serial/lantiq.c
> > > +++ b/drivers/tty/serial/lantiq.c
> > > @@ -26,7 +26,9 @@
> > > #include <linux/clk.h>
> > > #include <linux/gpio.h>
> > > +#ifdef CONFIG_LANTIQ
> > > #include <lantiq_soc.h>
> > > +#endif
> > That is never how you do this in Linux, you know better.
> >
> > Please go and get this patchset reviewed and signed-off-by from other
> > internal Intel kernel developers before resending it next time. It is
> > their job to find and fix your basic errors like this, not ours.
> Thank you for your comment.
> Actually, we have discussed this issue internally.
> We put the reason why we use "#ifdef CONFIG_LANTIQ" preprocessor in commit
> message in "[PATCH v2 08/18] serial: intel: Get serial id from dts".
> Please refer the commit message below.
>
> "#ifdef CONFIG_LANTIQ" preprocessor is used because LTQ_EARLY_ASC
> macro is defined in lantiq_soc.h.
> lantiq_soc.h is in arch path for legacy product support.
>
> arch/mips/include/asm/mach-lantiq/xway/lantiq_soc.h
>
> If "#ifdef preprocessor" is changed to
> "if (IS_ENABLED(CONFIG_LANTIQ))", when CONFIG_LANTIQ is not enabled,
> code using LTQ_EARLY_ASC is compiled.
> Compilation will fail for no LTQ_EARLY_ASC defined.

Sorry, but no. Why is this one tiny driver/chip somehow more "special"
than all of the tens of thousands of other devices we support to warrent
it getting some sort of special exception to do things differently?
What happens to the next device that wants to do it this way?

Our coding style and rules are there for a reason, do not violate them
thinking your device is the only one that matters.

Do it properly, again, you all know better than this.

greg k-h

2018-08-03 17:50:51

by Paul Burton

[permalink] [raw]
Subject: Re: [PATCH v2 01/18] MIPS: intel: Add initial support for Intel MIPS SoCs

Hi Songjun / Hua,

On Fri, Aug 03, 2018 at 11:02:20AM +0800, Songjun Wu wrote:
> From: Hua Ma <[email protected]>
>
> Add initial support for Intel MIPS interAptiv SoCs made by Intel.
> This series will add support for the grx500 family.
>
> The series allows booting a minimal system using a initramfs.

Thanks for submitting this - I have some comments below.

> diff --git a/arch/mips/Kbuild.platforms b/arch/mips/Kbuild.platforms
> index ac7ad54f984f..bcd647060f3e 100644
> --- a/arch/mips/Kbuild.platforms
> +++ b/arch/mips/Kbuild.platforms
> @@ -12,6 +12,7 @@ platforms += cobalt
> platforms += dec
> platforms += emma
> platforms += generic
> +platforms += intel-mips
> platforms += jazz
> platforms += jz4740
> platforms += lantiq

Oh EVA, why must you ruin nice things... Ideally I'd be suggesting that
we use the generic platform but it doesn't yet have a nice way to deal
with non-standard EVA setups.

It would be good if we could make sure that's the only reason for your
custom platform though, so that once generic does support EVA we can
migrate your system over. Most notably, it would be good to make use of
the UHI-specified boot protocol if possible (ie. $r4==-2, $r5==&dtb).

It looks like your prom_init_cmdline() supports multiple boot protocols
- could you clarify which is actually used?

> diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
> index 08c10c518f83..2d34f17f3e24 100644
> --- a/arch/mips/Kconfig
> +++ b/arch/mips/Kconfig
> @@ -409,6 +409,34 @@ config LANTIQ
> select ARCH_HAS_RESET_CONTROLLER
> select RESET_CONTROLLER
>
> +config INTEL_MIPS
> + bool "Intel MIPS interAptiv SoC based platforms"
> + select BOOT_RAW
> + select CEVT_R4K
> + select COMMON_CLK
> + select CPU_MIPS32_3_5_EVA
> + select CPU_MIPS32_3_5_FEATURES
> + select CPU_MIPSR2_IRQ_EI
> + select CPU_MIPSR2_IRQ_VI
> + select CSRC_R4K
> + select DMA_NONCOHERENT
> + select GENERIC_ISA_DMA
> + select IRQ_MIPS_CPU
> + select MFD_CORE
> + select MFD_SYSCON
> + select MIPS_CPU_SCACHE
> + select MIPS_GIC
> + select SYS_HAS_CPU_MIPS32_R1

For a system based on interAptiv you should never need to build a
MIPS32r1 kernel, so you should remove the above select.

> diff --git a/arch/mips/include/asm/mach-intel-mips/cpu-feature-overrides.h b/arch/mips/include/asm/mach-intel-mips/cpu-feature-overrides.h
> new file mode 100644
> index 000000000000..ac5f3b943d2a
> --- /dev/null
> +++ b/arch/mips/include/asm/mach-intel-mips/cpu-feature-overrides.h
> @@ -0,0 +1,61 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +/*
> + * This file was derived from: include/asm-mips/cpu-features.h
> + * Copyright (C) 2003, 2004 Ralf Baechle
> + * Copyright (C) 2004 Maciej W. Rozycki
> + * Copyright (C) 2018 Intel Corporation.
> + */
> +
> +#ifndef __ASM_MACH_INTEL_MIPS_CPU_FEATURE_OVERRIDES_H
> +#define __ASM_MACH_INTEL_MIPS_CPU_FEATURE_OVERRIDES_H
> +
> +#define cpu_has_tlb 1
> +#define cpu_has_4kex 1
> +#define cpu_has_3k_cache 0
> +#define cpu_has_4k_cache 1
> +#define cpu_has_tx39_cache 0
> +#define cpu_has_sb1_cache 0
> +#define cpu_has_fpu 0
> +#define cpu_has_32fpr 0
> +#define cpu_has_counter 1
> +#define cpu_has_watch 1
> +#define cpu_has_divec 1
> +
> +#define cpu_has_prefetch 1
> +#define cpu_has_ejtag 1
> +#define cpu_has_llsc 1
> +
> +#define cpu_has_mips16 0
> +#define cpu_has_mdmx 0
> +#define cpu_has_mips3d 0
> +#define cpu_has_smartmips 0
> +#define cpu_has_mmips 0
> +#define cpu_has_vz 0
> +
> +#define cpu_has_mips32r1 1
> +#define cpu_has_mips32r2 1
> +#define cpu_has_mips64r1 0
> +#define cpu_has_mips64r2 0
> +
> +#define cpu_has_dsp 1
> +#define cpu_has_dsp2 0
> +#define cpu_has_mipsmt 1
> +
> +#define cpu_has_vint 1
> +#define cpu_has_veic 0
> +
> +#define cpu_has_64bits 0
> +#define cpu_has_64bit_zero_reg 0
> +#define cpu_has_64bit_gp_regs 0
> +#define cpu_has_64bit_addresses 0
> +
> +#define cpu_has_cm2 1
> +#define cpu_has_cm2_l2sync 1
> +#define cpu_has_eva 1
> +#define cpu_has_tlbinv 1
> +
> +#define cpu_dcache_line_size() 32
> +#define cpu_icache_line_size() 32
> +#define cpu_scache_line_size() 32

If you rebase atop linux-next or mips-next then you should find that
many of these defines are now redundant, especially after removing the
SYS_HAS_CPU_MIPS32_R1 select which means your kernel build will always
target MIPS32r2.

Due to architectural restrictions on where various options can be
present, you should be able to remove:

- cpu_has_4kex
- cpu_has_3k_cache
- cpu_has_4k_cache
- cpu_has_32fpr
- cpu_has_divec
- cpu_has_prefetch
- cpu_has_llsc

cpu_has_mmips defaults to a compile-time zero unless you select
CONFIG_SYS_SUPPORTS_MICROMIPS, so please remove that one.

cpu_has_64bit_gp_regs & cpu_has_64bit_addresses both default to zero
already for 32bit kernel builds, so please remove those.

cpu_has_cm2 & cpu_has_cm2_l2sync don't exist anywhere in-tree, so please
remove those.

Additionally cpu_has_sb1_cache is not used anywhere, or defined by
asm/cpu-features.h & seems to just be left over in some platform
override files including presumably one you based yours on. Please
remove it too.

Also you select CPU_MIPSR2_IRQ_EI but define cpu_has_veic as 0, could
you check that mismatch?

> diff --git a/arch/mips/include/asm/mach-intel-mips/irq.h b/arch/mips/include/asm/mach-intel-mips/irq.h
> new file mode 100644
> index 000000000000..12a949084856
> --- /dev/null
> +++ b/arch/mips/include/asm/mach-intel-mips/irq.h
> @@ -0,0 +1,17 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +/*
> + * Copyright (C) 2014 Lei Chuanhua <[email protected]>
> + * Copyright (C) 2018 Intel Corporation.
> + */
> +
> +#ifndef __INTEL_MIPS_IRQ_H
> +#define __INTEL_MIPS_IRQ_H
> +
> +#define MIPS_CPU_IRQ_BASE 0
> +#define MIPS_GIC_IRQ_BASE (MIPS_CPU_IRQ_BASE + 8)

These 2 defines are the defaults anyway, so please remove them.

> +#define NR_IRQS 256

And if you don't actually need this then you could remove irq.h entirely
- do you actually use more than 128 IRQs?

> diff --git a/arch/mips/include/asm/mach-intel-mips/spaces.h b/arch/mips/include/asm/mach-intel-mips/spaces.h
> new file mode 100644
> index 000000000000..80e7b09f712c
> --- /dev/null
> +++ b/arch/mips/include/asm/mach-intel-mips/spaces.h
>% >% >%
> +#define IO_SIZE _AC(0x10000000, UL)
> +#define IO_SHIFT _AC(0x10000000, UL)

These IO_ defines don't appear to be used anywhere?

> +/* IO space one */
> +#define __pa_symbol(x) __pa(x)

Can you explain why you need this, rather than the default definition of
__pa_symbol()? The comment doesn't seem to describe much of anything.

> diff --git a/arch/mips/include/asm/mach-intel-mips/war.h b/arch/mips/include/asm/mach-intel-mips/war.h
> new file mode 100644
> index 000000000000..1c95553151e1
> --- /dev/null
> +++ b/arch/mips/include/asm/mach-intel-mips/war.h
> @@ -0,0 +1,18 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +#ifndef __ASM_MIPS_MACH_INTEL_MIPS_WAR_H
> +#define __ASM_MIPS_MACH_INTEL_MIPS_WAR_H
> +
> +#define R4600_V1_INDEX_ICACHEOP_WAR 0
> +#define R4600_V1_HIT_CACHEOP_WAR 0
> +#define R4600_V2_HIT_CACHEOP_WAR 0
> +#define R5432_CP0_INTERRUPT_WAR 0
> +#define BCM1250_M3_WAR 0
> +#define SIBYTE_1956_WAR 0
> +#define MIPS4K_ICACHE_REFILL_WAR 0
> +#define MIPS_CACHE_SYNC_WAR 0
> +#define TX49XX_ICACHE_INDEX_INV_WAR 0
> +#define ICACHE_REFILLS_WORKAROUND_WAR 0
> +#define R10000_LLSC_WAR 0
> +#define MIPS34K_MISSED_ITLB_WAR 0
> +
> +#endif /* __ASM_MIPS_MACH_INTEL_MIPS_WAR_H */

Since you need none of these workarounds, you shouldn't need war.h at
all.

> diff --git a/arch/mips/intel-mips/Kconfig b/arch/mips/intel-mips/Kconfig
> new file mode 100644
> index 000000000000..35d2ae2b5408
> --- /dev/null
> +++ b/arch/mips/intel-mips/Kconfig
> @@ -0,0 +1,22 @@
> +if INTEL_MIPS
> +
> +choice
> + prompt "Built-in device tree"
> + help
> + Legacy bootloaders do not pass a DTB pointer to the kernel, so
> + if a "wrapper" is not being used, the kernel will need to include
> + a device tree that matches the target board.
> +
> + The builtin DTB will only be used if the firmware does not supply
> + a valid DTB.
> +
> +config DTB_INTEL_MIPS_NONE
> + bool "None"
> +
> +config DTB_INTEL_MIPS_GRX500
> + bool "Intel MIPS GRX500 Board"
> + select BUILTIN_DTB
> +
> +endchoice
> +
> +endif

So do you actually have both styles of bootloader?

> diff --git a/arch/mips/intel-mips/prom.c b/arch/mips/intel-mips/prom.c
> new file mode 100644
> index 000000000000..a1b1393c13bc
> --- /dev/null
> +++ b/arch/mips/intel-mips/prom.c
>% >% >%
> +static void __init prom_init_cmdline(void)
> +{
> + int i;
> + int argc;
> + char **argv;
> +
> + /*
> + * If u-boot pass parameters, it is ok, however, if without u-boot
> + * JTAG or other tool has to reset all register value before it goes
> + * emulation most likely belongs to this category
> + */
> + if (fw_arg0 == 0 || fw_arg1 == 0)
> + return;

I don't understand what you're trying to say here, or why this check
exists. If you boot with fw_arg0 == fw_arg1 == 0 then you'd just hit the
loop below right, and execute zero iterations of it? That seems like it
would be fine without this special case.

> + /*
> + * a0: fw_arg0 - the number of string in init cmdline
> + * a1: fw_arg1 - the address of string in init cmdline
> + *
> + * In accordance with the MIPS UHI specification,
> + * the bootloader can pass the following arguments to the kernel:
> + * - $a0: -2.
> + * - $a1: KSEG0 address of the flattened device-tree blob.
> + */
> + if (fw_arg0 == -2)
> + return;
> +
> + argc = fw_arg0;
> + argv = (char **)KSEG1ADDR(fw_arg1);
> +
> + arcs_cmdline[0] = '\0';
> +
> + for (i = 0; i < argc; i++) {
> + char *p = (char *)KSEG1ADDR(argv[i]);
> +
> + if (argv[i] && *p) {
> + strlcat(arcs_cmdline, p, sizeof(arcs_cmdline));
> + strlcat(arcs_cmdline, " ", sizeof(arcs_cmdline));
> + }
> + }

Why do you need to use kseg1? Why can't the arguments be accessed
cached?

Are the arguments passed as virtual or physical addresses? If virtual &
we can access them cached then you could replace all of this with a call
to fw_init_cmdline().

> +static int __init plat_enable_iocoherency(void)
> +{
> + if (!mips_cps_numiocu(0))
> + return 0;
> +
> + /* Nothing special needs to be done to enable coherency */
> + pr_info("Coherence Manager IOCU detected\n");
> + /* Second IOCU for MPE or other master access register */
> + write_gcr_reg0_base(0xa0000000);
> + write_gcr_reg0_mask(0xf8000000 | CM_GCR_REGn_MASK_CMTGT_IOCU1);
> + return 1;
> +}
> +
> +static void __init plat_setup_iocoherency(void)
> +{
> + if (plat_enable_iocoherency() &&
> + coherentio == IO_COHERENCE_DISABLED) {
> + pr_info("Hardware DMA cache coherency disabled\n");
> + return;
> + }
> + panic("This kind of IO coherency is not supported!");
> +}

Since you select CONFIG_DMA_NONCOHERENT in Kconfig, coherentio will
always equal IO_COHERENCE_DISABLED. That suggests to me that the above 2
functions are probably useless, or at least needlessly convoluted.

> +static int __init plat_publish_devices(void)
> +{
> + if (!of_have_populated_dt())
> + return 0;
> + return of_platform_populate(NULL, of_default_bus_match_table, NULL,
> + NULL);
> +}
> +arch_initcall(plat_publish_devices);

The core DT code calls of_platform_populate() already (see
of_platform_default_populate_init()), so you can remove this function.

Thanks,
Paul

2018-08-04 10:55:44

by Hauke Mehrtens

[permalink] [raw]
Subject: Re: [PATCH v2 14/18] serial: intel: Add CCF support

On 08/03/2018 12:30 PM, Greg Kroah-Hartman wrote:
> On Fri, Aug 03, 2018 at 03:33:38PM +0800, Wu, Songjun wrote:
>>
>>
>> On 8/3/2018 1:56 PM, Greg Kroah-Hartman wrote:
>>> On Fri, Aug 03, 2018 at 11:02:33AM +0800, Songjun Wu wrote:
>>>> Previous implementation uses platform-dependent API to get the clock.
>>>> Those functions are not available for other SoC which uses the same IP.
>>>> The CCF (Common Clock Framework) have an abstraction based APIs for
>>>> clock. In future, the platform specific code will be removed when the
>>>> legacy soc use CCF as well.
>>>> Change to use CCF APIs to get clock and rate. So that different SoCs
>>>> can use the same driver.
>>>>
>>>> Signed-off-by: Songjun Wu <[email protected]>
>>>> ---
>>>>
>>>> Changes in v2: None
>>>>
>>>> drivers/tty/serial/lantiq.c | 11 +++++++++++
>>>> 1 file changed, 11 insertions(+)
>>>>
>>>> diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
>>>> index 36479d66fb7c..35518ab3a80d 100644
>>>> --- a/drivers/tty/serial/lantiq.c
>>>> +++ b/drivers/tty/serial/lantiq.c
>>>> @@ -26,7 +26,9 @@
>>>> #include <linux/clk.h>
>>>> #include <linux/gpio.h>
>>>> +#ifdef CONFIG_LANTIQ
>>>> #include <lantiq_soc.h>
>>>> +#endif
>>> That is never how you do this in Linux, you know better.
>>>
>>> Please go and get this patchset reviewed and signed-off-by from other
>>> internal Intel kernel developers before resending it next time. It is
>>> their job to find and fix your basic errors like this, not ours.
>> Thank you for your comment.
>> Actually, we have discussed this issue internally.
>> We put the reason why we use "#ifdef CONFIG_LANTIQ" preprocessor in commit
>> message in "[PATCH v2 08/18] serial: intel: Get serial id from dts".
>> Please refer the commit message below.
>>
>> "#ifdef CONFIG_LANTIQ" preprocessor is used because LTQ_EARLY_ASC
>> macro is defined in lantiq_soc.h.
>> lantiq_soc.h is in arch path for legacy product support.
>>
>> arch/mips/include/asm/mach-lantiq/xway/lantiq_soc.h
>>
>> If "#ifdef preprocessor" is changed to
>> "if (IS_ENABLED(CONFIG_LANTIQ))", when CONFIG_LANTIQ is not enabled,
>> code using LTQ_EARLY_ASC is compiled.
>> Compilation will fail for no LTQ_EARLY_ASC defined.
>
> Sorry, but no. Why is this one tiny driver/chip somehow more "special"
> than all of the tens of thousands of other devices we support to warrent
> it getting some sort of special exception to do things differently?
> What happens to the next device that wants to do it this way?
>
> Our coding style and rules are there for a reason, do not violate them
> thinking your device is the only one that matters.
>
> Do it properly, again, you all know better than this.
>
> greg k-h
>
Hi Greg,

The problem is that the Lantiq SoC code in arch/mips/lantiq does not use
the common clock framework, but it uses the clk framework directly. It
defines CONFIG_HAVE_CLK and CONFIG_CLKDEV_LOOKUP, but not
CONFIG_COMMON_CLK. The xRX500 SoC which is being added here is about 2
generations more recent than the VR9/xRX200 SoC which is the latest
which is supported by the code in arch/mips/lantiq. With this new SoC we
switched to the common clock framework. This driver is used by the older
SoC and also by the new ones because this IP core is pretty similar in
all the SoCs.
This patch makes it possible to use it with the legacy lantiq code and
also with the common clock framework. I see multiple options to fix this
problem.

1. The current approach to have it as a compile variant for a) legacy
lantiq arch code without common clock framework and b) support for SoCs
using the common clock framework.
2. Convert the lantiq arch code to the common clock framework. This
would be a good approach, but it need some efforts.
3. Remove the arch/mips/lantiq code. There are still users of this code.
4. Use the old APIs also for the new xRX500 SoC, I do not like this
approach.
5. Move lantiq_soc.h to somewhere in include/linux/ so it is globally
available and provide some better wrapper code.

Hauke


Attachments:
signature.asc (499.00 B)
OpenPGP digital signature

2018-08-04 11:12:28

by Hauke Mehrtens

[permalink] [raw]
Subject: Re: [PATCH v2 04/18] MIPS: dts: Add initial support for Intel MIPS SoCs

On 08/03/2018 05:02 AM, Songjun Wu wrote:
> From: Hua Ma <[email protected]>
>
> Add dts files to support Intel MIPS SoCs:
> - xrx500.dtsi is the chip dts
> - easy350_anywan.dts is the board dts
>
> Signed-off-by: Hua Ma <[email protected]>
> Signed-off-by: Songjun Wu <[email protected]>
> ---
>
> Changes in v2:
> - New patch split from previous patch
> - The memory address is changed to @20000000
> - Update to obj-$(CONFIG_BUILTIN_DTB) as per commit fca3aa166422
>
> arch/mips/boot/dts/Makefile | 1 +
> arch/mips/boot/dts/intel-mips/Makefile | 4 ++
> arch/mips/boot/dts/intel-mips/easy350_anywan.dts | 26 ++++++++++
> arch/mips/boot/dts/intel-mips/xrx500.dtsi | 66 ++++++++++++++++++++++++
> 4 files changed, 97 insertions(+)
> create mode 100644 arch/mips/boot/dts/intel-mips/Makefile
> create mode 100644 arch/mips/boot/dts/intel-mips/easy350_anywan.dts
> create mode 100644 arch/mips/boot/dts/intel-mips/xrx500.dtsi
>
> diff --git a/arch/mips/boot/dts/Makefile b/arch/mips/boot/dts/Makefile
> index 1e79cab8e269..05f52f279047 100644
> --- a/arch/mips/boot/dts/Makefile
> +++ b/arch/mips/boot/dts/Makefile
> @@ -3,6 +3,7 @@ subdir-y += brcm
> subdir-y += cavium-octeon
> subdir-y += img
> subdir-y += ingenic
> +subdir-y += intel-mips
> subdir-y += lantiq
> subdir-y += mscc
> subdir-y += mti
> diff --git a/arch/mips/boot/dts/intel-mips/Makefile b/arch/mips/boot/dts/intel-mips/Makefile
> new file mode 100644
> index 000000000000..adfaabbbb07c
> --- /dev/null
> +++ b/arch/mips/boot/dts/intel-mips/Makefile
> @@ -0,0 +1,4 @@
> +# SPDX-License-Identifier: GPL-2.0
> +dtb-$(CONFIG_DTB_INTEL_MIPS_GRX500) += easy350_anywan.dtb
> +
> +obj-$(CONFIG_BUILTIN_DTB) += $(addsuffix .o, $(dtb-y))
> diff --git a/arch/mips/boot/dts/intel-mips/easy350_anywan.dts b/arch/mips/boot/dts/intel-mips/easy350_anywan.dts
> new file mode 100644
> index 000000000000..e5e95f90c5e7
> --- /dev/null
> +++ b/arch/mips/boot/dts/intel-mips/easy350_anywan.dts
> @@ -0,0 +1,26 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/dts-v1/;
> +
> +#include <dt-bindings/interrupt-controller/mips-gic.h>
> +#include <dt-bindings/clock/intel,grx500-clk.h>
> +
> +#include "xrx500.dtsi"
> +
> +/ {
> + model = "EASY350 ANYWAN (GRX350) Main model";

Main model can be removed, it does not identify the board.

> + compatible = "intel,easy350-anywan";

I think this should be
compatible = "intel,easy350-anywan", "intel,xrx500";

Are there different revisions of the EASY350 Anywan board or only of the
EASY550 board?There are at least some differences in the power supply on
the EASY550 V1 and EASY550 V2 board. I would suggest to be here very
specific to make it easier when adding more boards.

> +
> + aliases {
> + serial0 = &asc0;
> + };
> +
> + chosen {
> + bootargs = "earlycon=lantiq,0x16600000 clk_ignore_unused";

What happens when you remove clk_ignore_unused?
If it crashes we should probably define some of the clock to be always
active.

> + stdout-path = "serial0";
> + };
> +
> + memory@20000000 {
> + device_type = "memory";
> + reg = <0x20000000 0x0e000000>;
> + };
> +};
> diff --git a/arch/mips/boot/dts/intel-mips/xrx500.dtsi b/arch/mips/boot/dts/intel-mips/xrx500.dtsi
> new file mode 100644
> index 000000000000..54c5f8f8b604
> --- /dev/null
> +++ b/arch/mips/boot/dts/intel-mips/xrx500.dtsi
> @@ -0,0 +1,66 @@
> +// SPDX-License-Identifier: GPL-2.0
> +
> +/ {
> + #address-cells = <1>;
> + #size-cells = <1>;
> + compatible = "intel,xrx500";
> +
> + cpus {
> + #address-cells = <1>;
> + #size-cells = <0>;
> +
> + cpu0: cpu@0 {
> + device_type = "cpu";
> + compatible = "mti,interaptiv";
> + clocks = <&cgu CLK_CPU>;
> + reg = <0>;
> + };
> +
> + cpu1: cpu@1 {
> + device_type = "cpu";
> + compatible = "mti,interaptiv";
> + reg = <1>;
> + };
> + };
> +
> + cpu_intc: interrupt-controller {
> + compatible = "mti,cpu-interrupt-controller";
> +
> + interrupt-controller;
> + #interrupt-cells = <1>;
> + };
> +
> + gic: gic@12320000 {
> + compatible = "mti,gic";
> + reg = <0x12320000 0x20000>;
> +
> + interrupt-controller;
> + #interrupt-cells = <3>;
> + /*
> + * Declare the interrupt-parent even though the mti,gic
> + * binding doesn't require it, such that the kernel can
> + * figure out that cpu_intc is the root interrupt
> + * controller & should be probed first.
> + */
> + interrupt-parent = <&cpu_intc>;
> + mti,reserved-ipi-vectors = <56 8>;
> + };
> +
> + cgu: cgu@16200000 {
> + compatible = "intel,grx500-cgu", "syscon";
> + reg = <0x16200000 0x200>;
> + #clock-cells = <1>;
> + };
> +
> + asc0: serial@16600000 {
> + compatible = "lantiq,asc";
> + reg = <0x16600000 0x100000>;
> +
> + interrupt-parent = <&gic>;
> + interrupts = <GIC_SHARED 103 IRQ_TYPE_LEVEL_HIGH>,
> + <GIC_SHARED 105 IRQ_TYPE_LEVEL_HIGH>,
> + <GIC_SHARED 106 IRQ_TYPE_LEVEL_HIGH>;
> + clocks = <&cgu CLK_SSX4>, <&cgu GCLK_UART>;
> + clock-names = "freq", "asc";
> + };
> +};
>



Attachments:
signature.asc (499.00 B)
OpenPGP digital signature

2018-08-04 12:44:17

by Greg KH

[permalink] [raw]
Subject: Re: [PATCH v2 14/18] serial: intel: Add CCF support

On Sat, Aug 04, 2018 at 12:54:22PM +0200, Hauke Mehrtens wrote:
> On 08/03/2018 12:30 PM, Greg Kroah-Hartman wrote:
> > On Fri, Aug 03, 2018 at 03:33:38PM +0800, Wu, Songjun wrote:
> >>
> >>
> >> On 8/3/2018 1:56 PM, Greg Kroah-Hartman wrote:
> >>> On Fri, Aug 03, 2018 at 11:02:33AM +0800, Songjun Wu wrote:
> >>>> Previous implementation uses platform-dependent API to get the clock.
> >>>> Those functions are not available for other SoC which uses the same IP.
> >>>> The CCF (Common Clock Framework) have an abstraction based APIs for
> >>>> clock. In future, the platform specific code will be removed when the
> >>>> legacy soc use CCF as well.
> >>>> Change to use CCF APIs to get clock and rate. So that different SoCs
> >>>> can use the same driver.
> >>>>
> >>>> Signed-off-by: Songjun Wu <[email protected]>
> >>>> ---
> >>>>
> >>>> Changes in v2: None
> >>>>
> >>>> drivers/tty/serial/lantiq.c | 11 +++++++++++
> >>>> 1 file changed, 11 insertions(+)
> >>>>
> >>>> diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
> >>>> index 36479d66fb7c..35518ab3a80d 100644
> >>>> --- a/drivers/tty/serial/lantiq.c
> >>>> +++ b/drivers/tty/serial/lantiq.c
> >>>> @@ -26,7 +26,9 @@
> >>>> #include <linux/clk.h>
> >>>> #include <linux/gpio.h>
> >>>> +#ifdef CONFIG_LANTIQ
> >>>> #include <lantiq_soc.h>
> >>>> +#endif
> >>> That is never how you do this in Linux, you know better.
> >>>
> >>> Please go and get this patchset reviewed and signed-off-by from other
> >>> internal Intel kernel developers before resending it next time. It is
> >>> their job to find and fix your basic errors like this, not ours.
> >> Thank you for your comment.
> >> Actually, we have discussed this issue internally.
> >> We put the reason why we use "#ifdef CONFIG_LANTIQ" preprocessor in commit
> >> message in "[PATCH v2 08/18] serial: intel: Get serial id from dts".
> >> Please refer the commit message below.
> >>
> >> "#ifdef CONFIG_LANTIQ" preprocessor is used because LTQ_EARLY_ASC
> >> macro is defined in lantiq_soc.h.
> >> lantiq_soc.h is in arch path for legacy product support.
> >>
> >> arch/mips/include/asm/mach-lantiq/xway/lantiq_soc.h
> >>
> >> If "#ifdef preprocessor" is changed to
> >> "if (IS_ENABLED(CONFIG_LANTIQ))", when CONFIG_LANTIQ is not enabled,
> >> code using LTQ_EARLY_ASC is compiled.
> >> Compilation will fail for no LTQ_EARLY_ASC defined.
> >
> > Sorry, but no. Why is this one tiny driver/chip somehow more "special"
> > than all of the tens of thousands of other devices we support to warrent
> > it getting some sort of special exception to do things differently?
> > What happens to the next device that wants to do it this way?
> >
> > Our coding style and rules are there for a reason, do not violate them
> > thinking your device is the only one that matters.
> >
> > Do it properly, again, you all know better than this.
> >
> > greg k-h
> >
> Hi Greg,
>
> The problem is that the Lantiq SoC code in arch/mips/lantiq does not use
> the common clock framework, but it uses the clk framework directly. It
> defines CONFIG_HAVE_CLK and CONFIG_CLKDEV_LOOKUP, but not
> CONFIG_COMMON_CLK. The xRX500 SoC which is being added here is about 2
> generations more recent than the VR9/xRX200 SoC which is the latest
> which is supported by the code in arch/mips/lantiq. With this new SoC we
> switched to the common clock framework. This driver is used by the older
> SoC and also by the new ones because this IP core is pretty similar in
> all the SoCs.
> This patch makes it possible to use it with the legacy lantiq code and
> also with the common clock framework. I see multiple options to fix this
> problem.
>
> 1. The current approach to have it as a compile variant for a) legacy
> lantiq arch code without common clock framework and b) support for SoCs
> using the common clock framework.
> 2. Convert the lantiq arch code to the common clock framework. This
> would be a good approach, but it need some efforts.
> 3. Remove the arch/mips/lantiq code. There are still users of this code.
> 4. Use the old APIs also for the new xRX500 SoC, I do not like this
> approach.
> 5. Move lantiq_soc.h to somewhere in include/linux/ so it is globally
> available and provide some better wrapper code.

I don't really care what you do at this point in time, but you all
should know better than the crazy #ifdef is not allowed to try to
prevent/allow the inclusion of a .h file. Checkpatch might have even
warned you about it, right?

So do it correctly, odds are #5 is correct, as that makes it work like
any other device in the kernel. You are not unique here.

greg k-h

2018-08-04 21:04:49

by Arnd Bergmann

[permalink] [raw]
Subject: Re: [PATCH v2 14/18] serial: intel: Add CCF support

On Sat, Aug 4, 2018 at 2:43 PM, Greg Kroah-Hartman
<[email protected]> wrote:
> On Sat, Aug 04, 2018 at 12:54:22PM +0200, Hauke Mehrtens wrote:
>> On 08/03/2018 12:30 PM, Greg Kroah-Hartman wrote:
>> > On Fri, Aug 03, 2018 at 03:33:38PM +0800, Wu, Songjun wrote:

>> This patch makes it possible to use it with the legacy lantiq code and
>> also with the common clock framework. I see multiple options to fix this
>> problem.
>>
>> 1. The current approach to have it as a compile variant for a) legacy
>> lantiq arch code without common clock framework and b) support for SoCs
>> using the common clock framework.
>> 2. Convert the lantiq arch code to the common clock framework. This
>> would be a good approach, but it need some efforts.
>> 3. Remove the arch/mips/lantiq code. There are still users of this code.
>> 4. Use the old APIs also for the new xRX500 SoC, I do not like this
>> approach.
>> 5. Move lantiq_soc.h to somewhere in include/linux/ so it is globally
>> available and provide some better wrapper code.
>
> I don't really care what you do at this point in time, but you all
> should know better than the crazy #ifdef is not allowed to try to
> prevent/allow the inclusion of a .h file. Checkpatch might have even
> warned you about it, right?
>
> So do it correctly, odds are #5 is correct, as that makes it work like
> any other device in the kernel. You are not unique here.

The best approach here would clearly be 2. We don't want platform
specific header files for doing things that should be completely generic.

Converting lantiq to the common-clk framework obviously requires
some work, but then again the whole arch/mips/lantiq/clk.c file
is fairly short and maybe not that hard to convert.

From looking at arch/mips/lantiq/xway/sysctrl.c, it appears that you
already use the clkdev lookup mechanism for some devices without
using COMMON_CLK, so I would assume that you can also use those
for the remaining clks, which would be much simpler. It registers
one anonymous clk there as

clkdev_add_pmu("1e100c00.serial", NULL, 0, 0, PMU_ASC1);

so why not add replace that with two named clocks and just use
the same names in the DT for the newer chip?

Arnd

2018-08-05 08:38:36

by Christoph Hellwig

[permalink] [raw]
Subject: Re: [PATCH v2 15/18] serial: intel: Support more platform

The subject line also seems odd, your are changing deps on the lantiq
driver, not some (nonexistent) intel serial driver.

2018-08-06 07:16:26

by Songjun Wu

[permalink] [raw]
Subject: Re: [PATCH v2 14/18] serial: intel: Add CCF support



On 8/5/2018 5:03 AM, Arnd Bergmann wrote:
> On Sat, Aug 4, 2018 at 2:43 PM, Greg Kroah-Hartman
> <[email protected]> wrote:
>> On Sat, Aug 04, 2018 at 12:54:22PM +0200, Hauke Mehrtens wrote:
>>> On 08/03/2018 12:30 PM, Greg Kroah-Hartman wrote:
>>>> On Fri, Aug 03, 2018 at 03:33:38PM +0800, Wu, Songjun wrote:
>>> This patch makes it possible to use it with the legacy lantiq code and
>>> also with the common clock framework. I see multiple options to fix this
>>> problem.
>>>
>>> 1. The current approach to have it as a compile variant for a) legacy
>>> lantiq arch code without common clock framework and b) support for SoCs
>>> using the common clock framework.
>>> 2. Convert the lantiq arch code to the common clock framework. This
>>> would be a good approach, but it need some efforts.
>>> 3. Remove the arch/mips/lantiq code. There are still users of this code.
>>> 4. Use the old APIs also for the new xRX500 SoC, I do not like this
>>> approach.
>>> 5. Move lantiq_soc.h to somewhere in include/linux/ so it is globally
>>> available and provide some better wrapper code.
>> I don't really care what you do at this point in time, but you all
>> should know better than the crazy #ifdef is not allowed to try to
>> prevent/allow the inclusion of a .h file. Checkpatch might have even
>> warned you about it, right?
>>
>> So do it correctly, odds are #5 is correct, as that makes it work like
>> any other device in the kernel. You are not unique here.
> The best approach here would clearly be 2. We don't want platform
> specific header files for doing things that should be completely generic.
>
> Converting lantiq to the common-clk framework obviously requires
> some work, but then again the whole arch/mips/lantiq/clk.c file
> is fairly short and maybe not that hard to convert.
>
> >From looking at arch/mips/lantiq/xway/sysctrl.c, it appears that you
> already use the clkdev lookup mechanism for some devices without
> using COMMON_CLK, so I would assume that you can also use those
> for the remaining clks, which would be much simpler. It registers
> one anonymous clk there as
>
> clkdev_add_pmu("1e100c00.serial", NULL, 0, 0, PMU_ASC1);
>
> so why not add replace that with two named clocks and just use
> the same names in the DT for the newer chip?
>
> Arnd
We discussed internally and have another solution for this issue.
Add one lantiq.h in the serial folder, and use "#ifdef preprocessor" in
lantiq.h,
also providing no-op stub functions in the #else case, then call those
functions
unconditionally from lantiq.c to avoid #ifdef in C file.

To support CCF in legacy product is another topic, is not included in
this patch.

The implementation is as following:
#ifdef CONFIG_LANTIQ
#include <lantiq_soc.h>
#else
#define LTQ_EARLY_ASC 0
#define CPHYSADDR(_val) 0

static inline struct clk *clk_get_fpi(void)
{
    return NULL;
}
#endif

2018-08-06 07:21:35

by Geert Uytterhoeven

[permalink] [raw]
Subject: Re: [PATCH v2 14/18] serial: intel: Add CCF support

Hi Songjun,

On Mon, Aug 6, 2018 at 9:15 AM Wu, Songjun <[email protected]> wrote:
> On 8/5/2018 5:03 AM, Arnd Bergmann wrote:
> > On Sat, Aug 4, 2018 at 2:43 PM, Greg Kroah-Hartman
> > <[email protected]> wrote:
> >> On Sat, Aug 04, 2018 at 12:54:22PM +0200, Hauke Mehrtens wrote:
> >>> On 08/03/2018 12:30 PM, Greg Kroah-Hartman wrote:
> >>>> On Fri, Aug 03, 2018 at 03:33:38PM +0800, Wu, Songjun wrote:
> >>> This patch makes it possible to use it with the legacy lantiq code and
> >>> also with the common clock framework. I see multiple options to fix this
> >>> problem.
> >>>
> >>> 1. The current approach to have it as a compile variant for a) legacy
> >>> lantiq arch code without common clock framework and b) support for SoCs
> >>> using the common clock framework.
> >>> 2. Convert the lantiq arch code to the common clock framework. This
> >>> would be a good approach, but it need some efforts.
> >>> 3. Remove the arch/mips/lantiq code. There are still users of this code.
> >>> 4. Use the old APIs also for the new xRX500 SoC, I do not like this
> >>> approach.
> >>> 5. Move lantiq_soc.h to somewhere in include/linux/ so it is globally
> >>> available and provide some better wrapper code.
> >> I don't really care what you do at this point in time, but you all
> >> should know better than the crazy #ifdef is not allowed to try to
> >> prevent/allow the inclusion of a .h file. Checkpatch might have even
> >> warned you about it, right?
> >>
> >> So do it correctly, odds are #5 is correct, as that makes it work like
> >> any other device in the kernel. You are not unique here.
> > The best approach here would clearly be 2. We don't want platform
> > specific header files for doing things that should be completely generic.
> >
> > Converting lantiq to the common-clk framework obviously requires
> > some work, but then again the whole arch/mips/lantiq/clk.c file
> > is fairly short and maybe not that hard to convert.
> >
> > >From looking at arch/mips/lantiq/xway/sysctrl.c, it appears that you
> > already use the clkdev lookup mechanism for some devices without
> > using COMMON_CLK, so I would assume that you can also use those
> > for the remaining clks, which would be much simpler. It registers
> > one anonymous clk there as
> >
> > clkdev_add_pmu("1e100c00.serial", NULL, 0, 0, PMU_ASC1);
> >
> > so why not add replace that with two named clocks and just use
> > the same names in the DT for the newer chip?
> >
> > Arnd
> We discussed internally and have another solution for this issue.
> Add one lantiq.h in the serial folder, and use "#ifdef preprocessor" in
> lantiq.h,
> also providing no-op stub functions in the #else case, then call those
> functions
> unconditionally from lantiq.c to avoid #ifdef in C file.
>
> To support CCF in legacy product is another topic, is not included in
> this patch.
>
> The implementation is as following:
> #ifdef CONFIG_LANTIQ
> #include <lantiq_soc.h>
> #else
> #define LTQ_EARLY_ASC 0
> #define CPHYSADDR(_val) 0
>
> static inline struct clk *clk_get_fpi(void)
> {
> return NULL;
> }
> #endif

Why not use clkdev_add(), as Arnd suggested?
That would be a 3-line patch without introducing a new header file and an ugly
#ifdef, which complicates compile coverage testing?

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

2018-08-06 07:22:28

by Songjun Wu

[permalink] [raw]
Subject: Re: [PATCH v2 15/18] serial: intel: Support more platform



On 8/5/2018 4:37 PM, Christoph Hellwig wrote:
> The subject line also seems odd, your are changing deps on the lantiq
> driver, not some (nonexistent) intel serial driver.
>
Your suggestion is reasonable, it will be changed to "serial: lantiq".

2018-08-06 09:00:34

by Songjun Wu

[permalink] [raw]
Subject: Re: [PATCH v2 14/18] serial: intel: Add CCF support



On 8/6/2018 3:20 PM, Geert Uytterhoeven wrote:
> Hi Songjun,
>
> On Mon, Aug 6, 2018 at 9:15 AM Wu, Songjun <[email protected]> wrote:
>> On 8/5/2018 5:03 AM, Arnd Bergmann wrote:
>>> On Sat, Aug 4, 2018 at 2:43 PM, Greg Kroah-Hartman
>>> <[email protected]> wrote:
>>>> On Sat, Aug 04, 2018 at 12:54:22PM +0200, Hauke Mehrtens wrote:
>>>>> On 08/03/2018 12:30 PM, Greg Kroah-Hartman wrote:
>>>>>> On Fri, Aug 03, 2018 at 03:33:38PM +0800, Wu, Songjun wrote:
>>>>> This patch makes it possible to use it with the legacy lantiq code and
>>>>> also with the common clock framework. I see multiple options to fix this
>>>>> problem.
>>>>>
>>>>> 1. The current approach to have it as a compile variant for a) legacy
>>>>> lantiq arch code without common clock framework and b) support for SoCs
>>>>> using the common clock framework.
>>>>> 2. Convert the lantiq arch code to the common clock framework. This
>>>>> would be a good approach, but it need some efforts.
>>>>> 3. Remove the arch/mips/lantiq code. There are still users of this code.
>>>>> 4. Use the old APIs also for the new xRX500 SoC, I do not like this
>>>>> approach.
>>>>> 5. Move lantiq_soc.h to somewhere in include/linux/ so it is globally
>>>>> available and provide some better wrapper code.
>>>> I don't really care what you do at this point in time, but you all
>>>> should know better than the crazy #ifdef is not allowed to try to
>>>> prevent/allow the inclusion of a .h file. Checkpatch might have even
>>>> warned you about it, right?
>>>>
>>>> So do it correctly, odds are #5 is correct, as that makes it work like
>>>> any other device in the kernel. You are not unique here.
>>> The best approach here would clearly be 2. We don't want platform
>>> specific header files for doing things that should be completely generic.
>>>
>>> Converting lantiq to the common-clk framework obviously requires
>>> some work, but then again the whole arch/mips/lantiq/clk.c file
>>> is fairly short and maybe not that hard to convert.
>>>
>>> >From looking at arch/mips/lantiq/xway/sysctrl.c, it appears that you
>>> already use the clkdev lookup mechanism for some devices without
>>> using COMMON_CLK, so I would assume that you can also use those
>>> for the remaining clks, which would be much simpler. It registers
>>> one anonymous clk there as
>>>
>>> clkdev_add_pmu("1e100c00.serial", NULL, 0, 0, PMU_ASC1);
>>>
>>> so why not add replace that with two named clocks and just use
>>> the same names in the DT for the newer chip?
>>>
>>> Arnd
>> We discussed internally and have another solution for this issue.
>> Add one lantiq.h in the serial folder, and use "#ifdef preprocessor" in
>> lantiq.h,
>> also providing no-op stub functions in the #else case, then call those
>> functions
>> unconditionally from lantiq.c to avoid #ifdef in C file.
>>
>> To support CCF in legacy product is another topic, is not included in
>> this patch.
>>
>> The implementation is as following:
>> #ifdef CONFIG_LANTIQ
>> #include <lantiq_soc.h>
>> #else
>> #define LTQ_EARLY_ASC 0
>> #define CPHYSADDR(_val) 0
>>
>> static inline struct clk *clk_get_fpi(void)
>> {
>> return NULL;
>> }
>> #endif
> Why not use clkdev_add(), as Arnd suggested?
> That would be a 3-line patch without introducing a new header file and an ugly
> #ifdef, which complicates compile coverage testing?
>
> Gr{oetje,eeting}s,
>
> Geert
The reason we add a new head file is also for two macros(LTQ_EARLY_ASC
and CPHYSADDR)
used by legacy product. We need to provide the no-op stub for these two
macro for new product.


2018-08-06 09:13:56

by Hua Ma

[permalink] [raw]
Subject: Re: [PATCH v2 01/18] MIPS: intel: Add initial support for Intel MIPS SoCs



On 8/4/2018 1:49 AM, Paul Burton wrote:
> Hi Songjun / Hua,
>
> On Fri, Aug 03, 2018 at 11:02:20AM +0800, Songjun Wu wrote:
>> From: Hua Ma <[email protected]>
>>
>> Add initial support for Intel MIPS interAptiv SoCs made by Intel.
>> This series will add support for the grx500 family.
>>
>> The series allows booting a minimal system using a initramfs.
> Thanks for submitting this - I have some comments below.
Thanks for the review.

>> diff --git a/arch/mips/Kbuild.platforms b/arch/mips/Kbuild.platforms
>> index ac7ad54f984f..bcd647060f3e 100644
>> --- a/arch/mips/Kbuild.platforms
>> +++ b/arch/mips/Kbuild.platforms
>> @@ -12,6 +12,7 @@ platforms += cobalt
>> platforms += dec
>> platforms += emma
>> platforms += generic
>> +platforms += intel-mips
>> platforms += jazz
>> platforms += jz4740
>> platforms += lantiq
> Oh EVA, why must you ruin nice things... Ideally I'd be suggesting that
> we use the generic platform but it doesn't yet have a nice way to deal
> with non-standard EVA setups.
yes, we only support EVA.

> It would be good if we could make sure that's the only reason for your
> custom platform though, so that once generic does support EVA we can
> migrate your system over. Most notably, it would be good to make use of
> the UHI-specified boot protocol if possible (ie. $r4==-2, $r5==&dtb).
>
> It looks like your prom_init_cmdline() supports multiple boot protocols
> - could you clarify which is actually used?
this patch only support build-in dts, we will do a clean up.

>> diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
>> index 08c10c518f83..2d34f17f3e24 100644
>> --- a/arch/mips/Kconfig
>> +++ b/arch/mips/Kconfig
>> @@ -409,6 +409,34 @@ config LANTIQ
>> select ARCH_HAS_RESET_CONTROLLER
>> select RESET_CONTROLLER
>>
>> +config INTEL_MIPS
>> + bool "Intel MIPS interAptiv SoC based platforms"
>> + select BOOT_RAW
>> + select CEVT_R4K
>> + select COMMON_CLK
>> + select CPU_MIPS32_3_5_EVA
>> + select CPU_MIPS32_3_5_FEATURES
>> + select CPU_MIPSR2_IRQ_EI
>> + select CPU_MIPSR2_IRQ_VI
>> + select CSRC_R4K
>> + select DMA_NONCOHERENT
>> + select GENERIC_ISA_DMA
>> + select IRQ_MIPS_CPU
>> + select MFD_CORE
>> + select MFD_SYSCON
>> + select MIPS_CPU_SCACHE
>> + select MIPS_GIC
>> + select SYS_HAS_CPU_MIPS32_R1
> For a system based on interAptiv you should never need to build a
> MIPS32r1 kernel, so you should remove the above select.
will remove.

>> diff --git a/arch/mips/include/asm/mach-intel-mips/cpu-feature-overrides.h b/arch/mips/include/asm/mach-intel-mips/cpu-feature-overrides.h
>> new file mode 100644
>> index 000000000000..ac5f3b943d2a
>> --- /dev/null
>> +++ b/arch/mips/include/asm/mach-intel-mips/cpu-feature-overrides.h
>> @@ -0,0 +1,61 @@
>> +/* SPDX-License-Identifier: GPL-2.0 */
>> +/*
>> + * This file was derived from: include/asm-mips/cpu-features.h
>> + * Copyright (C) 2003, 2004 Ralf Baechle
>> + * Copyright (C) 2004 Maciej W. Rozycki
>> + * Copyright (C) 2018 Intel Corporation.
>> + */
>> +
>> +#ifndef __ASM_MACH_INTEL_MIPS_CPU_FEATURE_OVERRIDES_H
>> +#define __ASM_MACH_INTEL_MIPS_CPU_FEATURE_OVERRIDES_H
>> +
>> +#define cpu_has_tlb 1
>> +#define cpu_has_4kex 1
>> +#define cpu_has_3k_cache 0
>> +#define cpu_has_4k_cache 1
>> +#define cpu_has_tx39_cache 0
>> +#define cpu_has_sb1_cache 0
>> +#define cpu_has_fpu 0
>> +#define cpu_has_32fpr 0
>> +#define cpu_has_counter 1
>> +#define cpu_has_watch 1
>> +#define cpu_has_divec 1
>> +
>> +#define cpu_has_prefetch 1
>> +#define cpu_has_ejtag 1
>> +#define cpu_has_llsc 1
>> +
>> +#define cpu_has_mips16 0
>> +#define cpu_has_mdmx 0
>> +#define cpu_has_mips3d 0
>> +#define cpu_has_smartmips 0
>> +#define cpu_has_mmips 0
>> +#define cpu_has_vz 0
>> +
>> +#define cpu_has_mips32r1 1
>> +#define cpu_has_mips32r2 1
>> +#define cpu_has_mips64r1 0
>> +#define cpu_has_mips64r2 0
>> +
>> +#define cpu_has_dsp 1
>> +#define cpu_has_dsp2 0
>> +#define cpu_has_mipsmt 1
>> +
>> +#define cpu_has_vint 1
>> +#define cpu_has_veic 0
>> +
>> +#define cpu_has_64bits 0
>> +#define cpu_has_64bit_zero_reg 0
>> +#define cpu_has_64bit_gp_regs 0
>> +#define cpu_has_64bit_addresses 0
>> +
>> +#define cpu_has_cm2 1
>> +#define cpu_has_cm2_l2sync 1
>> +#define cpu_has_eva 1
>> +#define cpu_has_tlbinv 1
>> +
>> +#define cpu_dcache_line_size() 32
>> +#define cpu_icache_line_size() 32
>> +#define cpu_scache_line_size() 32
> If you rebase atop linux-next or mips-next then you should find that
> many of these defines are now redundant, especially after removing the
> SYS_HAS_CPU_MIPS32_R1 select which means your kernel build will always
> target MIPS32r2.
>
> Due to architectural restrictions on where various options can be
> present, you should be able to remove:
>
> - cpu_has_4kex
> - cpu_has_3k_cache
> - cpu_has_4k_cache
> - cpu_has_32fpr
> - cpu_has_divec
> - cpu_has_prefetch
> - cpu_has_llsc
>
> cpu_has_mmips defaults to a compile-time zero unless you select
> CONFIG_SYS_SUPPORTS_MICROMIPS, so please remove that one.
>
> cpu_has_64bit_gp_regs & cpu_has_64bit_addresses both default to zero
> already for 32bit kernel builds, so please remove those.
>
> cpu_has_cm2 & cpu_has_cm2_l2sync don't exist anywhere in-tree, so please
> remove those.
>
> Additionally cpu_has_sb1_cache is not used anywhere, or defined by
> asm/cpu-features.h & seems to just be left over in some platform
> override files including presumably one you based yours on. Please
> remove it too.
Thanks, will remove.

> Also you select CPU_MIPSR2_IRQ_EI but define cpu_has_veic as 0, could
> you check that mismatch?
The hardware does support, but the software does not support.

>> diff --git a/arch/mips/include/asm/mach-intel-mips/irq.h b/arch/mips/include/asm/mach-intel-mips/irq.h
>> new file mode 100644
>> index 000000000000..12a949084856
>> --- /dev/null
>> +++ b/arch/mips/include/asm/mach-intel-mips/irq.h
>> @@ -0,0 +1,17 @@
>> +/* SPDX-License-Identifier: GPL-2.0 */
>> +/*
>> + * Copyright (C) 2014 Lei Chuanhua <[email protected]>
>> + * Copyright (C) 2018 Intel Corporation.
>> + */
>> +
>> +#ifndef __INTEL_MIPS_IRQ_H
>> +#define __INTEL_MIPS_IRQ_H
>> +
>> +#define MIPS_CPU_IRQ_BASE 0
>> +#define MIPS_GIC_IRQ_BASE (MIPS_CPU_IRQ_BASE + 8)
> These 2 defines are the defaults anyway, so please remove them.
Thanks, will remove.
>> +#define NR_IRQS 256
> And if you don't actually need this then you could remove irq.h entirely
> - do you actually use more than 128 IRQs?
Yes, the hardware support 256 IRQs.

>> diff --git a/arch/mips/include/asm/mach-intel-mips/spaces.h b/arch/mips/include/asm/mach-intel-mips/spaces.h
>> new file mode 100644
>> index 000000000000..80e7b09f712c
>> --- /dev/null
>> +++ b/arch/mips/include/asm/mach-intel-mips/spaces.h
>> % >% >%
>> +#define IO_SIZE _AC(0x10000000, UL)
>> +#define IO_SHIFT _AC(0x10000000, UL)
> These IO_ defines don't appear to be used anywhere?
Thanks, will remove.

>> +/* IO space one */
>> +#define __pa_symbol(x) __pa(x)
> Can you explain why you need this, rather than the default definition of
> __pa_symbol()? The comment doesn't seem to describe much of anything.
Thanks, will remove.

>> diff --git a/arch/mips/include/asm/mach-intel-mips/war.h b/arch/mips/include/asm/mach-intel-mips/war.h
>> new file mode 100644
>> index 000000000000..1c95553151e1
>> --- /dev/null
>> +++ b/arch/mips/include/asm/mach-intel-mips/war.h
>> @@ -0,0 +1,18 @@
>> +/* SPDX-License-Identifier: GPL-2.0 */
>> +#ifndef __ASM_MIPS_MACH_INTEL_MIPS_WAR_H
>> +#define __ASM_MIPS_MACH_INTEL_MIPS_WAR_H
>> +
>> +#define R4600_V1_INDEX_ICACHEOP_WAR 0
>> +#define R4600_V1_HIT_CACHEOP_WAR 0
>> +#define R4600_V2_HIT_CACHEOP_WAR 0
>> +#define R5432_CP0_INTERRUPT_WAR 0
>> +#define BCM1250_M3_WAR 0
>> +#define SIBYTE_1956_WAR 0
>> +#define MIPS4K_ICACHE_REFILL_WAR 0
>> +#define MIPS_CACHE_SYNC_WAR 0
>> +#define TX49XX_ICACHE_INDEX_INV_WAR 0
>> +#define ICACHE_REFILLS_WORKAROUND_WAR 0
>> +#define R10000_LLSC_WAR 0
>> +#define MIPS34K_MISSED_ITLB_WAR 0
>> +
>> +#endif /* __ASM_MIPS_MACH_INTEL_MIPS_WAR_H */
> Since you need none of these workarounds, you shouldn't need war.h at
> all.
Thanks, will remove this file.

>> diff --git a/arch/mips/intel-mips/Kconfig b/arch/mips/intel-mips/Kconfig
>> new file mode 100644
>> index 000000000000..35d2ae2b5408
>> --- /dev/null
>> +++ b/arch/mips/intel-mips/Kconfig
>> @@ -0,0 +1,22 @@
>> +if INTEL_MIPS
>> +
>> +choice
>> + prompt "Built-in device tree"
>> + help
>> + Legacy bootloaders do not pass a DTB pointer to the kernel, so
>> + if a "wrapper" is not being used, the kernel will need to include
>> + a device tree that matches the target board.
>> +
>> + The builtin DTB will only be used if the firmware does not supply
>> + a valid DTB.
>> +
>> +config DTB_INTEL_MIPS_NONE
>> + bool "None"
>> +
>> +config DTB_INTEL_MIPS_GRX500
>> + bool "Intel MIPS GRX500 Board"
>> + select BUILTIN_DTB
>> +
>> +endchoice
>> +
>> +endif
> So do you actually have both styles of bootloader?
this patch only support the build-in, will do a clean up.

>> diff --git a/arch/mips/intel-mips/prom.c b/arch/mips/intel-mips/prom.c
>> new file mode 100644
>> index 000000000000..a1b1393c13bc
>> --- /dev/null
>> +++ b/arch/mips/intel-mips/prom.c
>> % >% >%
>> +static void __init prom_init_cmdline(void)
>> +{
>> + int i;
>> + int argc;
>> + char **argv;
>> +
>> + /*
>> + * If u-boot pass parameters, it is ok, however, if without u-boot
>> + * JTAG or other tool has to reset all register value before it goes
>> + * emulation most likely belongs to this category
>> + */
>> + if (fw_arg0 == 0 || fw_arg1 == 0)
>> + return;
> I don't understand what you're trying to say here, or why this check
> exists. If you boot with fw_arg0 == fw_arg1 == 0 then you'd just hit the
> loop below right, and execute zero iterations of it? That seems like it
> would be fine without this special case.
this patch do not support this , will remove.

>> + /*
>> + * a0: fw_arg0 - the number of string in init cmdline
>> + * a1: fw_arg1 - the address of string in init cmdline
>> + *
>> + * In accordance with the MIPS UHI specification,
>> + * the bootloader can pass the following arguments to the kernel:
>> + * - $a0: -2.
>> + * - $a1: KSEG0 address of the flattened device-tree blob.
>> + */
>> + if (fw_arg0 == -2)
>> + return;
>> +
>> + argc = fw_arg0;
>> + argv = (char **)KSEG1ADDR(fw_arg1);
>> +
>> + arcs_cmdline[0] = '\0';
>> +
>> + for (i = 0; i < argc; i++) {
>> + char *p = (char *)KSEG1ADDR(argv[i]);
>> +
>> + if (argv[i] && *p) {
>> + strlcat(arcs_cmdline, p, sizeof(arcs_cmdline));
>> + strlcat(arcs_cmdline, " ", sizeof(arcs_cmdline));
>> + }
>> + }
> Why do you need to use kseg1? Why can't the arguments be accessed
> cached?
>
> Are the arguments passed as virtual or physical addresses? If virtual &
> we can access them cached then you could replace all of this with a call
> to fw_init_cmdline().
this patch only support build-in dts, will remove .

>> +static int __init plat_enable_iocoherency(void)
>> +{
>> + if (!mips_cps_numiocu(0))
>> + return 0;
>> +
>> + /* Nothing special needs to be done to enable coherency */
>> + pr_info("Coherence Manager IOCU detected\n");
>> + /* Second IOCU for MPE or other master access register */
>> + write_gcr_reg0_base(0xa0000000);
>> + write_gcr_reg0_mask(0xf8000000 | CM_GCR_REGn_MASK_CMTGT_IOCU1);
>> + return 1;
>> +}
>> +
>> +static void __init plat_setup_iocoherency(void)
>> +{
>> + if (plat_enable_iocoherency() &&
>> + coherentio == IO_COHERENCE_DISABLED) {
>> + pr_info("Hardware DMA cache coherency disabled\n");
>> + return;
>> + }
>> + panic("This kind of IO coherency is not supported!");
>> +}
> Since you select CONFIG_DMA_NONCOHERENT in Kconfig, coherentio will
> always equal IO_COHERENCE_DISABLED. That suggests to me that the above 2
> functions are probably useless, or at least needlessly convoluted.
Thanks, will remove.

>> +static int __init plat_publish_devices(void)
>> +{
>> + if (!of_have_populated_dt())
>> + return 0;
>> + return of_platform_populate(NULL, of_default_bus_match_table, NULL,
>> + NULL);
>> +}
>> +arch_initcall(plat_publish_devices);
> The core DT code calls of_platform_populate() already (see
> of_platform_default_populate_init()), so you can remove this function.
>
> Thanks,
> Paul
Thanks, will remove.


2018-08-06 09:23:46

by Hua Ma

[permalink] [raw]
Subject: Re: [PATCH v2 04/18] MIPS: dts: Add initial support for Intel MIPS SoCs



On 8/4/2018 7:11 PM, Hauke Mehrtens wrote:
> On 08/03/2018 05:02 AM, Songjun Wu wrote:
>> From: Hua Ma <[email protected]>
>>
>> Add dts files to support Intel MIPS SoCs:
>> - xrx500.dtsi is the chip dts
>> - easy350_anywan.dts is the board dts
>>
>> Signed-off-by: Hua Ma <[email protected]>
>> Signed-off-by: Songjun Wu <[email protected]>
>> ---
>>
>> Changes in v2:
>> - New patch split from previous patch
>> - The memory address is changed to @20000000
>> - Update to obj-$(CONFIG_BUILTIN_DTB) as per commit fca3aa166422
>>
>> arch/mips/boot/dts/Makefile | 1 +
>> arch/mips/boot/dts/intel-mips/Makefile | 4 ++
>> arch/mips/boot/dts/intel-mips/easy350_anywan.dts | 26 ++++++++++
>> arch/mips/boot/dts/intel-mips/xrx500.dtsi | 66 ++++++++++++++++++++++++
>> 4 files changed, 97 insertions(+)
>> create mode 100644 arch/mips/boot/dts/intel-mips/Makefile
>> create mode 100644 arch/mips/boot/dts/intel-mips/easy350_anywan.dts
>> create mode 100644 arch/mips/boot/dts/intel-mips/xrx500.dtsi
>>
>> diff --git a/arch/mips/boot/dts/intel-mips/easy350_anywan.dts b/arch/mips/boot/dts/intel-mips/easy350_anywan.dts
>> new file mode 100644
>> index 000000000000..e5e95f90c5e7
>> --- /dev/null
>> +++ b/arch/mips/boot/dts/intel-mips/easy350_anywan.dts
>> @@ -0,0 +1,26 @@
>> +// SPDX-License-Identifier: GPL-2.0
>> +/dts-v1/;
>> +
>> +#include <dt-bindings/interrupt-controller/mips-gic.h>
>> +#include <dt-bindings/clock/intel,grx500-clk.h>
>> +
>> +#include "xrx500.dtsi"
>> +
>> +/ {
>> + model = "EASY350 ANYWAN (GRX350) Main model";
> Main model can be removed, it does not identify the board.
Thanks, will remove.

>> + compatible = "intel,easy350-anywan";
> I think this should be
> compatible = "intel,easy350-anywan", "intel,xrx500";
>
> Are there different revisions of the EASY350 Anywan board or only of the
> EASY550 board?There are at least some differences in the power supply on
> the EASY550 V1 and EASY550 V2 board. I would suggest to be here very
> specific to make it easier when adding more boards.
OK, thanks.

>> +
>> + aliases {
>> + serial0 = &asc0;
>> + };
>> +
>> + chosen {
>> + bootargs = "earlycon=lantiq,0x16600000 clk_ignore_unused";
> What happens when you remove clk_ignore_unused?
> If it crashes we should probably define some of the clock to be always
> active.
OK, will check and improve if possible.


2018-08-06 09:31:35

by Geert Uytterhoeven

[permalink] [raw]
Subject: Re: [PATCH v2 14/18] serial: intel: Add CCF support

Hi Songjun,

On Mon, Aug 6, 2018 at 10:58 AM Wu, Songjun <[email protected]> wrote:
> On 8/6/2018 3:20 PM, Geert Uytterhoeven wrote:
> > On Mon, Aug 6, 2018 at 9:15 AM Wu, Songjun <[email protected]> wrote:
> >> On 8/5/2018 5:03 AM, Arnd Bergmann wrote:
> >>> On Sat, Aug 4, 2018 at 2:43 PM, Greg Kroah-Hartman
> >>> <[email protected]> wrote:
> >>>> On Sat, Aug 04, 2018 at 12:54:22PM +0200, Hauke Mehrtens wrote:
> >>>>> On 08/03/2018 12:30 PM, Greg Kroah-Hartman wrote:
> >>>>>> On Fri, Aug 03, 2018 at 03:33:38PM +0800, Wu, Songjun wrote:
> >>>>> This patch makes it possible to use it with the legacy lantiq code and
> >>>>> also with the common clock framework. I see multiple options to fix this
> >>>>> problem.
> >>>>>
> >>>>> 1. The current approach to have it as a compile variant for a) legacy
> >>>>> lantiq arch code without common clock framework and b) support for SoCs
> >>>>> using the common clock framework.
> >>>>> 2. Convert the lantiq arch code to the common clock framework. This
> >>>>> would be a good approach, but it need some efforts.
> >>>>> 3. Remove the arch/mips/lantiq code. There are still users of this code.
> >>>>> 4. Use the old APIs also for the new xRX500 SoC, I do not like this
> >>>>> approach.
> >>>>> 5. Move lantiq_soc.h to somewhere in include/linux/ so it is globally
> >>>>> available and provide some better wrapper code.
> >>>> I don't really care what you do at this point in time, but you all
> >>>> should know better than the crazy #ifdef is not allowed to try to
> >>>> prevent/allow the inclusion of a .h file. Checkpatch might have even
> >>>> warned you about it, right?
> >>>>
> >>>> So do it correctly, odds are #5 is correct, as that makes it work like
> >>>> any other device in the kernel. You are not unique here.
> >>> The best approach here would clearly be 2. We don't want platform
> >>> specific header files for doing things that should be completely generic.
> >>>
> >>> Converting lantiq to the common-clk framework obviously requires
> >>> some work, but then again the whole arch/mips/lantiq/clk.c file
> >>> is fairly short and maybe not that hard to convert.
> >>>
> >>> >From looking at arch/mips/lantiq/xway/sysctrl.c, it appears that you
> >>> already use the clkdev lookup mechanism for some devices without
> >>> using COMMON_CLK, so I would assume that you can also use those
> >>> for the remaining clks, which would be much simpler. It registers
> >>> one anonymous clk there as
> >>>
> >>> clkdev_add_pmu("1e100c00.serial", NULL, 0, 0, PMU_ASC1);
> >>>
> >>> so why not add replace that with two named clocks and just use
> >>> the same names in the DT for the newer chip?
> >>>
> >>> Arnd
> >> We discussed internally and have another solution for this issue.
> >> Add one lantiq.h in the serial folder, and use "#ifdef preprocessor" in
> >> lantiq.h,
> >> also providing no-op stub functions in the #else case, then call those
> >> functions
> >> unconditionally from lantiq.c to avoid #ifdef in C file.
> >>
> >> To support CCF in legacy product is another topic, is not included in
> >> this patch.
> >>
> >> The implementation is as following:
> >> #ifdef CONFIG_LANTIQ
> >> #include <lantiq_soc.h>
> >> #else
> >> #define LTQ_EARLY_ASC 0
> >> #define CPHYSADDR(_val) 0
> >>
> >> static inline struct clk *clk_get_fpi(void)
> >> {
> >> return NULL;
> >> }
> >> #endif
> > Why not use clkdev_add(), as Arnd suggested?
> > That would be a 3-line patch without introducing a new header file and an ugly
> > #ifdef, which complicates compile coverage testing?
> >
> The reason we add a new head file is also for two macros(LTQ_EARLY_ASC
> and CPHYSADDR)
> used by legacy product. We need to provide the no-op stub for these two
> macro for new product.

No you don't. The line number should not be obtained by comparing the
resource address with a hardcoded base address.

Perhaps the override of port->line should just be removed, as IIRC, the serial
core has already filled in that field with the (next available) line number?

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

2018-08-06 09:33:48

by Songjun Wu

[permalink] [raw]
Subject: Re: [PATCH v2 08/18] serial: intel: Get serial id from dts



On 8/3/2018 1:43 PM, Greg Kroah-Hartman wrote:
> On Fri, Aug 03, 2018 at 11:02:27AM +0800, Songjun Wu wrote:
>> Get serial id from dts.
>>
>> "#ifdef CONFIG_LANTIQ" preprocessor is used because LTQ_EARLY_ASC
>> macro is defined in lantiq_soc.h.
>> lantiq_soc.h is in arch path for legacy product support.
>>
>> arch/mips/include/asm/mach-lantiq/xway/lantiq_soc.h
>>
>> If "#ifdef preprocessor" is changed to
>> "if (IS_ENABLED(CONFIG_LANTIQ))", when CONFIG_LANTIQ is not enabled,
>> code using LTQ_EARLY_ASC is compiled.
>> Compilation will fail for no LTQ_EARLY_ASC defined.
>>
>> Signed-off-by: Songjun Wu <[email protected]>
>> ---
>>
>> Changes in v2: None
>>
>> drivers/tty/serial/lantiq.c | 19 +++++++++++++++----
>> 1 file changed, 15 insertions(+), 4 deletions(-)
>>
>> diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
>> index 044128277248..836ca51460f2 100644
>> --- a/drivers/tty/serial/lantiq.c
>> +++ b/drivers/tty/serial/lantiq.c
>> @@ -6,6 +6,7 @@
>> * Copyright (C) 2007 Felix Fietkau <[email protected]>
>> * Copyright (C) 2007 John Crispin <[email protected]>
>> * Copyright (C) 2010 Thomas Langer, <[email protected]>
>> + * Copyright (C) 2018 Intel Corporation.
> Your changes here do not warrent the addition of a copyright line, don't
> you agree? If not, please get a signed-off-by from your corporate
> lawyer who does this this is warrented when you resend this patch.
>
> thanks,
>
> greg k-h
>
Thanks.
The copyright line will be removed when we resend this patch.

2018-08-06 15:21:14

by Rob Herring

[permalink] [raw]
Subject: Re: [PATCH v2 06/18] MIPS: dts: Change upper case to lower case

On Thu, Aug 2, 2018 at 9:03 PM Songjun Wu <[email protected]> wrote:
>
> All the upper case in unit-address and hex constants are
> changed to lower case according to the Linux conventions.

It is DT conventions, not Linux.

Otherwise,

Reviewed-by: Rob Herring <[email protected]>

>
> Signed-off-by: Songjun Wu <[email protected]>
> ---
>
> Changes in v2: None
>
> arch/mips/boot/dts/lantiq/danube.dtsi | 42 ++++++++++++++++-----------------
> arch/mips/boot/dts/lantiq/easy50712.dts | 14 +++++------
> 2 files changed, 28 insertions(+), 28 deletions(-)

2018-08-06 15:22:04

by Rob Herring

[permalink] [raw]
Subject: Re: [PATCH v2 05/18] dt-binding: MIPS: Add documentation of Intel MIPS SoCs

On Thu, Aug 2, 2018 at 9:03 PM Songjun Wu <[email protected]> wrote:
>
> From: Hua Ma <[email protected]>
>
> This patch adds binding documentation for the
> compatible values of the Intel MIPS SoCs.
>
> Signed-off-by: Hua Ma <[email protected]>
> Signed-off-by: Songjun Wu <[email protected]>
> ---
>
> Changes in v2:
> - New patch split from previous patch
> - Add the board and chip compatible in dt document
>
> Documentation/devicetree/bindings/mips/intel.txt | 17 +++++++++++++++++
> 1 file changed, 17 insertions(+)
> create mode 100644 Documentation/devicetree/bindings/mips/intel.txt

Reviewed-by: Rob Herring <[email protected]>

2018-08-06 15:24:33

by Rob Herring

[permalink] [raw]
Subject: Re: [PATCH v2 03/18] dt-bindings: clk: Add documentation of grx500 clock controller

On Thu, Aug 2, 2018 at 9:03 PM Songjun Wu <[email protected]> wrote:
>
> From: Yixin Zhu <[email protected]>
>
> This patch adds binding documentation for grx500 clock controller.
>
> Signed-off-by: YiXin Zhu <[email protected]>
> Signed-off-by: Songjun Wu <[email protected]>
> ---
>
> Changes in v2:
> - Rewrite clock driver's dt-binding document according to Rob Herring's
> comments.
> - Simplify device tree docoment, remove some clock description.
>
> .../devicetree/bindings/clock/intel,grx500-clk.txt | 39 ++++++++++++++++++++++

Please match the compatible string: intel,grx500-cgu.txt

> 1 file changed, 39 insertions(+)
> create mode 100644 Documentation/devicetree/bindings/clock/intel,grx500-clk.txt
>
> diff --git a/Documentation/devicetree/bindings/clock/intel,grx500-clk.txt b/Documentation/devicetree/bindings/clock/intel,grx500-clk.txt
> new file mode 100644
> index 000000000000..e54e1dad9196
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/clock/intel,grx500-clk.txt
> @@ -0,0 +1,39 @@
> +Device Tree Clock bindings for grx500 PLL controller.
> +
> +This binding uses the common clock binding:
> + Documentation/devicetree/bindings/clock/clock-bindings.txt
> +
> +The grx500 clock controller supplies clock to various controllers within the
> +SoC.
> +
> +Required properties for clock node
> +- compatible: Should be "intel,grx500-cgu".
> +- reg: physical base address of the controller and length of memory range.
> +- #clock-cells: should be 1.
> +
> +Optional Propteries:
> +- intel,osc-frequency: frequency of the osc clock.
> +if missing, driver will use clock rate defined in the driver.

This should use a fixed-clock node instead.

> +
> +Example: Clock controller node:
> +
> + cgu: cgu@16200000 {
> + compatible = "intel,grx500-cgu", "syscon";
> + reg = <0x16200000 0x200>;
> + #clock-cells = <1>;
> + };
> +
> +
> +Example: UART controller node that consumes the clock generated by clock
> + controller.
> +
> + asc0: serial@16600000 {
> + compatible = "lantiq,asc";
> + reg = <0x16600000 0x100000>;
> + interrupt-parent = <&gic>;
> + interrupts = <GIC_SHARED 103 IRQ_TYPE_LEVEL_HIGH>,
> + <GIC_SHARED 105 IRQ_TYPE_LEVEL_HIGH>,
> + <GIC_SHARED 106 IRQ_TYPE_LEVEL_HIGH>;
> + clocks = <&cgu CLK_SSX4>, <&cgu GCLK_UART>;
> + clock-names = "freq", "asc";
> + };
> --
> 2.11.0
>

2018-08-06 16:43:20

by Rob Herring

[permalink] [raw]
Subject: Re: [PATCH v2 02/18] clk: intel: Add clock driver for Intel MIPS SoCs

On Thu, Aug 2, 2018 at 9:03 PM Songjun Wu <[email protected]> wrote:
>
> From: Yixin Zhu <[email protected]>
>
> This driver provides PLL clock registration as well as various clock
> branches, e.g. MUX clock, gate clock, divider clock and so on.
>
> PLLs that provide clock to DDR, CPU and peripherals are shown below:
>
> +---------+
> |--->| LCPLL3 0|--PCIe clk-->
> XO | +---------+
> +-----------|
> | +---------+
> | | 3|--PAE clk-->
> |--->| PLL0B 2|--GSWIP clk-->
> | | 1|--DDR clk-->DDR PHY clk-->
> | | 0|--CPU1 clk--+ +-----+
> | +---------+ |--->0 |
> | | MUX |--CPU clk-->
> | +---------+ |--->1 |
> | | 0|--CPU0 clk--+ +-----+
> |--->| PLLOA 1|--SSX4 clk-->
> | 2|--NGI clk-->
> | 3|--CBM clk-->
> +---------+
>
> Signed-off-by: Yixin Zhu <[email protected]>
> Signed-off-by: Songjun Wu <[email protected]>
> ---
>
> Changes in v2:
> - Rewrite clock driver, add platform clock description details in
> clock driver.
>
> drivers/clk/Kconfig | 1 +
> drivers/clk/Makefile | 3 +
> drivers/clk/intel/Kconfig | 20 ++
> drivers/clk/intel/Makefile | 7 +
> drivers/clk/intel/clk-cgu-pll.c | 166 ++++++++++
> drivers/clk/intel/clk-cgu-pll.h | 34 ++
> drivers/clk/intel/clk-cgu.c | 470 +++++++++++++++++++++++++++
> drivers/clk/intel/clk-cgu.h | 259 +++++++++++++++
> drivers/clk/intel/clk-grx500.c | 168 ++++++++++

> include/dt-bindings/clock/intel,grx500-clk.h | 69 ++++

This belongs with the clk binding patch.

Rob

2018-08-07 07:19:57

by Songjun Wu

[permalink] [raw]
Subject: Re: [PATCH v2 14/18] serial: intel: Add CCF support



On 8/6/2018 5:29 PM, Geert Uytterhoeven wrote:
> Hi Songjun,
>
> On Mon, Aug 6, 2018 at 10:58 AM Wu, Songjun <[email protected]> wrote:
>> On 8/6/2018 3:20 PM, Geert Uytterhoeven wrote:
>>> On Mon, Aug 6, 2018 at 9:15 AM Wu, Songjun <[email protected]> wrote:
>>>> On 8/5/2018 5:03 AM, Arnd Bergmann wrote:
>>>>> On Sat, Aug 4, 2018 at 2:43 PM, Greg Kroah-Hartman
>>>>> <[email protected]> wrote:
>>>>>> On Sat, Aug 04, 2018 at 12:54:22PM +0200, Hauke Mehrtens wrote:
>>>>>>> On 08/03/2018 12:30 PM, Greg Kroah-Hartman wrote:
>>>>>>>> On Fri, Aug 03, 2018 at 03:33:38PM +0800, Wu, Songjun wrote:
>>>>>>> This patch makes it possible to use it with the legacy lantiq code and
>>>>>>> also with the common clock framework. I see multiple options to fix this
>>>>>>> problem.
>>>>>>>
>>>>>>> 1. The current approach to have it as a compile variant for a) legacy
>>>>>>> lantiq arch code without common clock framework and b) support for SoCs
>>>>>>> using the common clock framework.
>>>>>>> 2. Convert the lantiq arch code to the common clock framework. This
>>>>>>> would be a good approach, but it need some efforts.
>>>>>>> 3. Remove the arch/mips/lantiq code. There are still users of this code.
>>>>>>> 4. Use the old APIs also for the new xRX500 SoC, I do not like this
>>>>>>> approach.
>>>>>>> 5. Move lantiq_soc.h to somewhere in include/linux/ so it is globally
>>>>>>> available and provide some better wrapper code.
>>>>>> I don't really care what you do at this point in time, but you all
>>>>>> should know better than the crazy #ifdef is not allowed to try to
>>>>>> prevent/allow the inclusion of a .h file. Checkpatch might have even
>>>>>> warned you about it, right?
>>>>>>
>>>>>> So do it correctly, odds are #5 is correct, as that makes it work like
>>>>>> any other device in the kernel. You are not unique here.
>>>>> The best approach here would clearly be 2. We don't want platform
>>>>> specific header files for doing things that should be completely generic.
>>>>>
>>>>> Converting lantiq to the common-clk framework obviously requires
>>>>> some work, but then again the whole arch/mips/lantiq/clk.c file
>>>>> is fairly short and maybe not that hard to convert.
>>>>>
>>>>> >From looking at arch/mips/lantiq/xway/sysctrl.c, it appears that you
>>>>> already use the clkdev lookup mechanism for some devices without
>>>>> using COMMON_CLK, so I would assume that you can also use those
>>>>> for the remaining clks, which would be much simpler. It registers
>>>>> one anonymous clk there as
>>>>>
>>>>> clkdev_add_pmu("1e100c00.serial", NULL, 0, 0, PMU_ASC1);
>>>>>
>>>>> so why not add replace that with two named clocks and just use
>>>>> the same names in the DT for the newer chip?
>>>>>
>>>>> Arnd
>>>> We discussed internally and have another solution for this issue.
>>>> Add one lantiq.h in the serial folder, and use "#ifdef preprocessor" in
>>>> lantiq.h,
>>>> also providing no-op stub functions in the #else case, then call those
>>>> functions
>>>> unconditionally from lantiq.c to avoid #ifdef in C file.
>>>>
>>>> To support CCF in legacy product is another topic, is not included in
>>>> this patch.
>>>>
>>>> The implementation is as following:
>>>> #ifdef CONFIG_LANTIQ
>>>> #include <lantiq_soc.h>
>>>> #else
>>>> #define LTQ_EARLY_ASC 0
>>>> #define CPHYSADDR(_val) 0
>>>>
>>>> static inline struct clk *clk_get_fpi(void)
>>>> {
>>>> return NULL;
>>>> }
>>>> #endif
>>> Why not use clkdev_add(), as Arnd suggested?
>>> That would be a 3-line patch without introducing a new header file and an ugly
>>> #ifdef, which complicates compile coverage testing?
>>>
>> The reason we add a new head file is also for two macros(LTQ_EARLY_ASC
>> and CPHYSADDR)
>> used by legacy product. We need to provide the no-op stub for these two
>> macro for new product.
> No you don't. The line number should not be obtained by comparing the
> resource address with a hardcoded base address.
This is the previous code. Now the line number is obtained from dts.
We keep this code for the compatibility.

Referring to the conditional-compilation part in coding-style,
We add a header file to avoid using “#ifdef” in C file.
> Perhaps the override of port->line should just be removed, as IIRC, the serial
> core has already filled in that field with the (next available) line number?
>
> Gr{oetje,eeting}s,
>
> Geert

2018-08-07 07:35:32

by Geert Uytterhoeven

[permalink] [raw]
Subject: Re: [PATCH v2 14/18] serial: intel: Add CCF support

Hi Songjun,

On Tue, Aug 7, 2018 at 9:18 AM Wu, Songjun <[email protected]> wrote:
> On 8/6/2018 5:29 PM, Geert Uytterhoeven wrote:
> > On Mon, Aug 6, 2018 at 10:58 AM Wu, Songjun <[email protected]> wrote:
> >> The reason we add a new head file is also for two macros(LTQ_EARLY_ASC
> >> and CPHYSADDR)
> >> used by legacy product. We need to provide the no-op stub for these two
> >> macro for new product.
> > No you don't. The line number should not be obtained by comparing the
> > resource address with a hardcoded base address.
> This is the previous code. Now the line number is obtained from dts.

Note that obtaining line numbers from DTS has its own share of problems, when
considering DT overlays. I've replied to the patch adding the call to
of_alias_get_id().

> We keep this code for the compatibility.
>
> Referring to the conditional-compilation part in coding-style,
> We add a header file to avoid using “#ifdef” in C file.
> > Perhaps the override of port->line should just be removed, as IIRC, the serial
> > core has already filled in that field with the (next available) line number?

Gr{oetje,eeting}s,

Geert--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

2018-08-07 08:08:44

by Geert Uytterhoeven

[permalink] [raw]
Subject: Re: [PATCH v2 08/18] serial: intel: Get serial id from dts

Hi Songjun,

On Fri, Aug 3, 2018 at 5:04 AM Songjun Wu <[email protected]> wrote:
> Get serial id from dts.
>
> "#ifdef CONFIG_LANTIQ" preprocessor is used because LTQ_EARLY_ASC
> macro is defined in lantiq_soc.h.
> lantiq_soc.h is in arch path for legacy product support.
>
> arch/mips/include/asm/mach-lantiq/xway/lantiq_soc.h
>
> If "#ifdef preprocessor" is changed to
> "if (IS_ENABLED(CONFIG_LANTIQ))", when CONFIG_LANTIQ is not enabled,
> code using LTQ_EARLY_ASC is compiled.
> Compilation will fail for no LTQ_EARLY_ASC defined.
>
> Signed-off-by: Songjun Wu <[email protected]>

Thanks for your patch!

> @@ -699,9 +700,19 @@ lqasc_probe(struct platform_device *pdev)
> return -ENODEV;
> }
>
> - /* check if this is the console port */
> - if (mmres->start != CPHYSADDR(LTQ_EARLY_ASC))
> - line = 1;
> + /* get serial id */
> + line = of_alias_get_id(node, "serial");
> + if (line < 0) {
> +#ifdef CONFIG_LANTIQ
> + if (mmres->start == CPHYSADDR(LTQ_EARLY_ASC))
> + line = 0;
> + else
> + line = 1;
> +#else
> + dev_err(&pdev->dev, "failed to get alias id, errno %d\n", line);
> + return line;

Please note that not providing a fallback here makes life harder when using
DT overlays.
See the description of commit 7678f4c20fa7670f ("serial: sh-sci: Add support
for dynamic instances") for background info.

> +#endif
> + }
>
> if (lqasc_port[line]) {
> dev_err(&pdev->dev, "port %d already allocated\n", line);

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

2018-08-08 02:53:17

by Zhu, Yi Xin

[permalink] [raw]
Subject: Re: [PATCH v2 02/18] clk: intel: Add clock driver for Intel MIPS SoCs



On 8/6/2018 11:19 PM, Rob Herring wrote:
> On Thu, Aug 2, 2018 at 9:03 PM Songjun Wu <[email protected]> wrote:
>> From: Yixin Zhu <[email protected]>
>>
>> This driver provides PLL clock registration as well as various clock
>> branches, e.g. MUX clock, gate clock, divider clock and so on.
>>
>> PLLs that provide clock to DDR, CPU and peripherals are shown below:
>>
>> +---------+
>> |--->| LCPLL3 0|--PCIe clk-->
>> XO | +---------+
>> +-----------|
>> | +---------+
>> | | 3|--PAE clk-->
>> |--->| PLL0B 2|--GSWIP clk-->
>> | | 1|--DDR clk-->DDR PHY clk-->
>> | | 0|--CPU1 clk--+ +-----+
>> | +---------+ |--->0 |
>> | | MUX |--CPU clk-->
>> | +---------+ |--->1 |
>> | | 0|--CPU0 clk--+ +-----+
>> |--->| PLLOA 1|--SSX4 clk-->
>> | 2|--NGI clk-->
>> | 3|--CBM clk-->
>> +---------+
>>
>> Signed-off-by: Yixin Zhu <[email protected]>
>> Signed-off-by: Songjun Wu <[email protected]>
>> ---
>>
>> Changes in v2:
>> - Rewrite clock driver, add platform clock description details in
>> clock driver.
>>
>> drivers/clk/Kconfig | 1 +
>> drivers/clk/Makefile | 3 +
>> drivers/clk/intel/Kconfig | 20 ++
>> drivers/clk/intel/Makefile | 7 +
>> drivers/clk/intel/clk-cgu-pll.c | 166 ++++++++++
>> drivers/clk/intel/clk-cgu-pll.h | 34 ++
>> drivers/clk/intel/clk-cgu.c | 470 +++++++++++++++++++++++++++
>> drivers/clk/intel/clk-cgu.h | 259 +++++++++++++++
>> drivers/clk/intel/clk-grx500.c | 168 ++++++++++
>> include/dt-bindings/clock/intel,grx500-clk.h | 69 ++++
> This belongs with the clk binding patch.
>
> Rob
Thanks for review.
Will move it to clock binding patch.

>


2018-08-08 03:10:00

by Zhu, Yi Xin

[permalink] [raw]
Subject: Re: [PATCH v2 03/18] dt-bindings: clk: Add documentation of grx500 clock controller



On 8/6/2018 11:18 PM, Rob Herring wrote:
> On Thu, Aug 2, 2018 at 9:03 PM Songjun Wu <[email protected]> wrote:
>> From: Yixin Zhu <[email protected]>
>>
>> This patch adds binding documentation for grx500 clock controller.
>>
>> Signed-off-by: YiXin Zhu <[email protected]>
>> Signed-off-by: Songjun Wu <[email protected]>
>> ---
>>
>> Changes in v2:
>> - Rewrite clock driver's dt-binding document according to Rob Herring's
>> comments.
>> - Simplify device tree docoment, remove some clock description.
>>
>> .../devicetree/bindings/clock/intel,grx500-clk.txt | 39 ++++++++++++++++++++++
> Please match the compatible string: intel,grx500-cgu.txt
Will update to use same name.

>
>> 1 file changed, 39 insertions(+)
>> create mode 100644 Documentation/devicetree/bindings/clock/intel,grx500-clk.txt
>>
>> diff --git a/Documentation/devicetree/bindings/clock/intel,grx500-clk.txt b/Documentation/devicetree/bindings/clock/intel,grx500-clk.txt
>> new file mode 100644
>> index 000000000000..e54e1dad9196
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/clock/intel,grx500-clk.txt
>> @@ -0,0 +1,39 @@
>> +Device Tree Clock bindings for grx500 PLL controller.
>> +
>> +This binding uses the common clock binding:
>> + Documentation/devicetree/bindings/clock/clock-bindings.txt
>> +
>> +The grx500 clock controller supplies clock to various controllers within the
>> +SoC.
>> +
>> +Required properties for clock node
>> +- compatible: Should be "intel,grx500-cgu".
>> +- reg: physical base address of the controller and length of memory range.
>> +- #clock-cells: should be 1.
>> +
>> +Optional Propteries:
>> +- intel,osc-frequency: frequency of the osc clock.
>> +if missing, driver will use clock rate defined in the driver.
> This should use a fixed-clock node instead.
Yes, This is a fixed clock node registered in driver code.
The frequency of the fixed clock is designed to be overwritten by device
tree in case some one verify
clock driver in the emulation platform or in some cases frequency other
than driver defined one is preferred.
These kinds of cases are very rare. But I feel it would be better to
have a way to use customized frequency.
The frequency defined in device tree will overwritten driver defined
frequency before registering fixed-clock node.

>> +
>> +Example: Clock controller node:
>> +
>> + cgu: cgu@16200000 {
>> + compatible = "intel,grx500-cgu", "syscon";
>> + reg = <0x16200000 0x200>;
>> + #clock-cells = <1>;
>> + };
>> +
>> +
>> +Example: UART controller node that consumes the clock generated by clock
>> + controller.
>> +
>> + asc0: serial@16600000 {
>> + compatible = "lantiq,asc";
>> + reg = <0x16600000 0x100000>;
>> + interrupt-parent = <&gic>;
>> + interrupts = <GIC_SHARED 103 IRQ_TYPE_LEVEL_HIGH>,
>> + <GIC_SHARED 105 IRQ_TYPE_LEVEL_HIGH>,
>> + <GIC_SHARED 106 IRQ_TYPE_LEVEL_HIGH>;
>> + clocks = <&cgu CLK_SSX4>, <&cgu GCLK_UART>;
>> + clock-names = "freq", "asc";
>> + };
>> --
>> 2.11.0
>>


2018-08-08 04:06:55

by Songjun Wu

[permalink] [raw]
Subject: Re: [PATCH v2 08/18] serial: intel: Get serial id from dts



On 8/7/2018 3:33 PM, Geert Uytterhoeven wrote:
> Hi Songjun,
>
> On Fri, Aug 3, 2018 at 5:04 AM Songjun Wu <[email protected]> wrote:
>> Get serial id from dts.
>>
>> "#ifdef CONFIG_LANTIQ" preprocessor is used because LTQ_EARLY_ASC
>> macro is defined in lantiq_soc.h.
>> lantiq_soc.h is in arch path for legacy product support.
>>
>> arch/mips/include/asm/mach-lantiq/xway/lantiq_soc.h
>>
>> If "#ifdef preprocessor" is changed to
>> "if (IS_ENABLED(CONFIG_LANTIQ))", when CONFIG_LANTIQ is not enabled,
>> code using LTQ_EARLY_ASC is compiled.
>> Compilation will fail for no LTQ_EARLY_ASC defined.
>>
>> Signed-off-by: Songjun Wu <[email protected]>
> Thanks for your patch!
>
>> @@ -699,9 +700,19 @@ lqasc_probe(struct platform_device *pdev)
>> return -ENODEV;
>> }
>>
>> - /* check if this is the console port */
>> - if (mmres->start != CPHYSADDR(LTQ_EARLY_ASC))
>> - line = 1;
>> + /* get serial id */
>> + line = of_alias_get_id(node, "serial");
>> + if (line < 0) {
>> +#ifdef CONFIG_LANTIQ
>> + if (mmres->start == CPHYSADDR(LTQ_EARLY_ASC))
>> + line = 0;
>> + else
>> + line = 1;
>> +#else
>> + dev_err(&pdev->dev, "failed to get alias id, errno %d\n", line);
>> + return line;
> Please note that not providing a fallback here makes life harder when using
> DT overlays.
> See the description of commit 7678f4c20fa7670f ("serial: sh-sci: Add support
> for dynamic instances") for background info.
Thanks for your comment.
The logic in commit 7678f4c20fa7670f is not suitable here.
We need to know which serial instance is used for console.
We cannot use dynamic serial instance here.



2018-08-08 05:51:31

by Stephen Boyd

[permalink] [raw]
Subject: Re: [PATCH v2 02/18] clk: intel: Add clock driver for Intel MIPS SoCs

Quoting Songjun Wu (2018-08-02 20:02:21)
> From: Yixin Zhu <[email protected]>
>
> This driver provides PLL clock registration as well as various clock
> branches, e.g. MUX clock, gate clock, divider clock and so on.
>
> PLLs that provide clock to DDR, CPU and peripherals are shown below:
>
> +---------+
> |--->| LCPLL3 0|--PCIe clk-->
> XO | +---------+
> +-----------|
> | +---------+
> | | 3|--PAE clk-->
> |--->| PLL0B 2|--GSWIP clk-->
> | | 1|--DDR clk-->DDR PHY clk-->
> | | 0|--CPU1 clk--+ +-----+
> | +---------+ |--->0 |
> | | MUX |--CPU clk-->
> | +---------+ |--->1 |
> | | 0|--CPU0 clk--+ +-----+
> |--->| PLLOA 1|--SSX4 clk-->
> | 2|--NGI clk-->
> | 3|--CBM clk-->
> +---------+

Thanks for the picture!

>
> Signed-off-by: Yixin Zhu <[email protected]>
> Signed-off-by: Songjun Wu <[email protected]>
> diff --git a/drivers/clk/Makefile b/drivers/clk/Makefile
> index 0bb25dd009d1..d929ca4607cf 100644
> --- a/drivers/clk/Makefile
> +++ b/drivers/clk/Makefile
> @@ -72,6 +72,9 @@ obj-$(CONFIG_ARCH_HISI) += hisilicon/
> obj-y += imgtec/
> obj-$(CONFIG_ARCH_MXC) += imx/
> obj-$(CONFIG_MACH_INGENIC) += ingenic/
> +ifeq ($(CONFIG_COMMON_CLK), y)
> +obj-y +=intel/
> +endif

Why not obj-$(CONFIG_INTEL_CCF) or something like that?

> obj-$(CONFIG_ARCH_KEYSTONE) += keystone/
> obj-$(CONFIG_MACH_LOONGSON32) += loongson1/
> obj-y += mediatek/
> diff --git a/drivers/clk/intel/Kconfig b/drivers/clk/intel/Kconfig
> new file mode 100644
> index 000000000000..c7d3fb1721fa
> --- /dev/null
> +++ b/drivers/clk/intel/Kconfig
> @@ -0,0 +1,20 @@
> +# SPDX-License-Identifier: GPL-2.0
> +config INTEL_CGU_CLK
> + depends on COMMON_CLK
> + depends on INTEL_MIPS || COMPILE_TEST
> + select MFD_SYSCON
> + bool "Intel clock controller support"
> + help
> + This driver support Intel CGU (Clock Generation Unit).

Is it really called a clock generation unit? Or that's just copied from
sunxi driver?

> +
> +choice
> + prompt "SoC platform selection"
> + depends on INTEL_CGU_CLK
> + default INTEL_GRX500_CGU_CLK
> +
> +config INTEL_GRX500_CGU_CLK
> + bool "GRX500 CLK"
> + help
> + Clock driver of GRX500 platform.
> +
> +endchoice
> diff --git a/drivers/clk/intel/Makefile b/drivers/clk/intel/Makefile
> new file mode 100644
> index 000000000000..16a0138e52c2
> --- /dev/null
> +++ b/drivers/clk/intel/Makefile
> @@ -0,0 +1,7 @@
> +# SPDX-License-Identifier: GPL-2.0
> +# Makefile for intel specific clk
> +
> +obj-$(CONFIG_INTEL_CGU_CLK) += clk-cgu.o clk-cgu-pll.o
> +ifneq ($(CONFIG_INTEL_GRX500_CGU_CLK),)
> + obj-y += clk-grx500.o
> +endif
> diff --git a/drivers/clk/intel/clk-cgu-pll.c b/drivers/clk/intel/clk-cgu-pll.c
> new file mode 100644
> index 000000000000..20759bc27e95
> --- /dev/null
> +++ b/drivers/clk/intel/clk-cgu-pll.c
> @@ -0,0 +1,166 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Copyright (C) 2018 Intel Corporation.
> + * Zhu YiXin <[email protected]>
> + */
> +
> +#include <linux/clk.h>

Is this include used?

> +#include <linux/clk-provider.h>
> +#include <linux/clkdev.h>
> +#include <linux/mfd/syscon.h>
> +#include <linux/of.h>
> +#include <linux/of_address.h>
> +#include <linux/regmap.h>
> +#include <linux/slab.h>
> +
> +#include "clk-cgu-pll.h"
> +#include "clk-cgu.h"
> +
> +#define to_intel_clk_pll(_hw) container_of(_hw, struct intel_clk_pll, hw)
> +
> +/*
> + * Calculate formula:
> + * rate = (prate * mult + (prate * frac) / frac_div) / div
> + */
> +static unsigned long
> +intel_pll_calc_rate(unsigned long prate, unsigned int mult,
> + unsigned int div, unsigned int frac,
> + unsigned int frac_div)
> +{
> + u64 crate, frate, rate64;
> +
> + rate64 = prate;
> + crate = rate64 * mult;
> +
> + if (frac) {
> + frate = rate64 * frac;
> + do_div(frate, frac_div);
> + crate += frate;
> + }
> + do_div(crate, div);
> +
> + return (unsigned long)crate;
> +}
> +
> +static void
> +grx500_pll_get_params(struct intel_clk_pll *pll, unsigned int *mult,
> + unsigned int *frac)
> +{
> + *mult = intel_get_clk_val(pll->map, pll->reg, 2, 7);
> + *frac = intel_get_clk_val(pll->map, pll->reg, 9, 21);
> +}
> +
> +static int intel_wait_pll_lock(struct intel_clk_pll *pll, int bit_idx)
> +{
> + unsigned int val;
> +
> + return regmap_read_poll_timeout(pll->map, pll->reg, val,
> + val & BIT(bit_idx), 10, 1000);
> +}
> +
> +static unsigned long
> +intel_grx500_pll_recalc_rate(struct clk_hw *hw, unsigned long prate)
> +{
> + struct intel_clk_pll *pll = to_intel_clk_pll(hw);
> + unsigned int mult, frac;
> +
> + grx500_pll_get_params(pll, &mult, &frac);
> +
> + return intel_pll_calc_rate(prate, mult, 1, frac, BIT(20));
> +}
> +
> +static int intel_grx500_pll_is_enabled(struct clk_hw *hw)
> +{
> + struct intel_clk_pll *pll = to_intel_clk_pll(hw);
> +
> + if (intel_wait_pll_lock(pll, 1)) {
> + pr_err("%s: pll: %s is not locked!\n",
> + __func__, clk_hw_get_name(hw));
> + return 0;
> + }
> +
> + return intel_get_clk_val(pll->map, pll->reg, 1, 1);
> +}
> +
> +const static struct clk_ops intel_grx500_pll_ops = {

Should be static const struct ...

> + .recalc_rate = intel_grx500_pll_recalc_rate,
> + .is_enabled = intel_grx500_pll_is_enabled,
> +};
> +
> +static struct clk
> +*intel_clk_register_pll(struct intel_clk_provider *ctx,
> + enum intel_pll_type type, const char *cname,
> + const char *const *pname, u8 num_parents,
> + unsigned long flags, unsigned int reg,
> + const struct intel_pll_rate_table *table,
> + unsigned int mult, unsigned int div, unsigned int frac)
> +{
> + struct clk_init_data init;
> + struct intel_clk_pll *pll;
> + struct clk_hw *hw;
> + int ret, i;
> +
> + if (type != pll_grx500) {
> + pr_err("%s: pll type %d not supported!\n",
> + __func__, type);
> + return ERR_PTR(-EINVAL);
> + }
> + init.name = cname;
> + init.ops = &intel_grx500_pll_ops;
> + init.flags = CLK_IS_BASIC;

Don't use this flag unless you have some reason to need it.

> + init.parent_names = pname;
> + init.num_parents = num_parents;
> +
> + pll = kzalloc(sizeof(*pll), GFP_KERNEL);
> + if (!pll)
> + return ERR_PTR(-ENOMEM);
> + pll->map = ctx->map;
> + pll->reg = reg;
> + pll->flags = flags;
> + pll->mult = mult;
> + pll->div = div;
> + pll->frac = frac;
> + pll->hw.init = &init;
> + if (table) {
> + for (i = 0; table[i].rate != 0; i++)
> + ;
> + pll->table_sz = i;
> + pll->rate_table = kmemdup(table, i * sizeof(table[0]),
> + GFP_KERNEL);
> + if (!pll->rate_table) {
> + ret = -ENOMEM;
> + goto err_free_pll;
> + }
> + }
> + hw = &pll->hw;
> + ret = clk_hw_register(NULL, hw);
> + if (ret)
> + goto err_free_pll;
> +
> + return hw->clk;
> +
> +err_free_pll:
> + kfree(pll);
> + return ERR_PTR(ret);
> +}
> +
> +void intel_clk_register_plls(struct intel_clk_provider *ctx,
> + struct intel_pll_clk *list, unsigned int nr_clk)
> +{
> + struct clk *clk;
> + int i;
> +
> + for (i = 0; i < nr_clk; i++, list++) {
> + clk = intel_clk_register_pll(ctx, list->type, list->name,
> + list->parent_names, list->num_parents,
> + list->flags, list->reg, list->rate_table,
> + list->mult, list->div, list->frac);
> + if (IS_ERR(clk)) {
> + pr_err("%s: failed to register pll: %s\n",
> + __func__, list->name);
> + continue;
> + }
> +
> + intel_clk_add_lookup(ctx, clk, list->id);
> + }
> +}
> diff --git a/drivers/clk/intel/clk-cgu-pll.h b/drivers/clk/intel/clk-cgu-pll.h
> new file mode 100644
> index 000000000000..3e7cff1d5e16
> --- /dev/null
> +++ b/drivers/clk/intel/clk-cgu-pll.h
> @@ -0,0 +1,34 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +/*
> + * Copyright(c) 2018 Intel Corporation.
> + * Zhu YiXin <[email protected]>
> + */
> +
> +#ifndef __INTEL_CLK_PLL_H
> +#define __INTEL_CLK_PLL_H
> +
> +enum intel_pll_type {
> + pll_grx500,
> +};
> +
> +struct intel_pll_rate_table {
> + unsigned long prate;
> + unsigned long rate;
> + unsigned int mult;
> + unsigned int div;
> + unsigned int frac;
> +};
> +
> +struct intel_clk_pll {
> + struct clk_hw hw;
> + struct regmap *map;
> + unsigned int reg;
> + unsigned long flags;
> + unsigned int mult;
> + unsigned int div;
> + unsigned int frac;
> + unsigned int table_sz;
> + const struct intel_pll_rate_table *rate_table;
> +};
> +
> +#endif /* __INTEL_CLK_PLL_H */
> diff --git a/drivers/clk/intel/clk-cgu.c b/drivers/clk/intel/clk-cgu.c
> new file mode 100644
> index 000000000000..10cacbe0fbcd
> --- /dev/null
> +++ b/drivers/clk/intel/clk-cgu.c
> @@ -0,0 +1,470 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Copyright (C) 2018 Intel Corporation.
> + * Zhu YiXin <[email protected]>
> + */
> +
> +#include <linux/clk.h>
> +#include <linux/clk-provider.h>
> +#include <linux/clkdev.h>
> +#include <linux/mfd/syscon.h>
> +#include <linux/of.h>
> +#include <linux/of_address.h>
> +#include <linux/regmap.h>
> +#include <linux/slab.h>
> +
> +#include "clk-cgu-pll.h"
> +#include "clk-cgu.h"
> +
> +#define GATE_HW_REG_STAT(reg) (reg)
> +#define GATE_HW_REG_EN(reg) ((reg) + 0x4)
> +#define GATE_HW_REG_DIS(reg) ((reg) + 0x8)
> +
> +#define to_intel_clk_mux(_hw) container_of(_hw, struct intel_clk_mux, hw)
> +#define to_intel_clk_divider(_hw) \
> + container_of(_hw, struct intel_clk_divider, hw)
> +#define to_intel_clk_gate(_hw) container_of(_hw, struct intel_clk_gate, hw)
> +
> +void intel_set_clk_val(struct regmap *map, u32 reg, u8 shift,
> + u8 width, u32 set_val)
> +{
> + u32 mask = GENMASK(width + shift, shift);
> +
> + regmap_update_bits(map, reg, mask, set_val << shift);
> +}
> +
> +u32 intel_get_clk_val(struct regmap *map, u32 reg, u8 shift,
> + u8 width)
> +{
> + u32 val;
> +
> + if (regmap_read(map, reg, &val)) {
> + WARN_ONCE(1, "Failed to read clk reg: 0x%x\n", reg);
> + return 0;
> + }
> + val >>= shift;
> + val &= BIT(width) - 1;
> +
> + return val;
> +}
> +
> +void intel_clk_add_lookup(struct intel_clk_provider *ctx,
> + struct clk *clk, unsigned int id)
> +{
> + pr_debug("Add clk: %s, id: %u\n", __clk_get_name(clk), id);
> + if (ctx->clk_data.clks && id)
> + ctx->clk_data.clks[id] = clk;
> +}
> +
> +static struct clk
> +*intel_clk_register_fixed(struct intel_clk_provider *ctx,
> + struct intel_clk_branch *list)
> +{
> + if (list->div_flags & CLOCK_FLAG_VAL_INIT)
> + intel_set_clk_val(ctx->map, list->div_off, list->div_shift,
> + list->div_width, list->div_val);
> +
> + return clk_register_fixed_rate(NULL, list->name, list->parent_names[0],
> + list->flags, list->mux_flags);
> +}
> +
> +static u8 intel_clk_mux_get_parent(struct clk_hw *hw)
> +{
> + struct intel_clk_mux *mux = to_intel_clk_mux(hw);
> + u32 val;
> +
> + val = intel_get_clk_val(mux->map, mux->reg, mux->shift, mux->width);
> + return clk_mux_val_to_index(hw, NULL, mux->flags, val);
> +}
> +
> +static int intel_clk_mux_set_parent(struct clk_hw *hw, u8 index)
> +{
> + struct intel_clk_mux *mux = to_intel_clk_mux(hw);
> + u32 val;
> +
> + val = clk_mux_index_to_val(NULL, mux->flags, index);
> + intel_set_clk_val(mux->map, mux->reg, mux->shift, mux->width, val);
> +
> + return 0;
> +}
> +
> +static int intel_clk_mux_determine_rate(struct clk_hw *hw,
> + struct clk_rate_request *req)
> +{
> + struct intel_clk_mux *mux = to_intel_clk_mux(hw);
> +
> + return clk_mux_determine_rate_flags(hw, req, mux->flags);
> +}
> +
> +const static struct clk_ops intel_clk_mux_ops = {
> + .get_parent = intel_clk_mux_get_parent,
> + .set_parent = intel_clk_mux_set_parent,
> + .determine_rate = intel_clk_mux_determine_rate,
> +};
> +
> +static struct clk
> +*intel_clk_register_mux(struct intel_clk_provider *ctx,
> + struct intel_clk_branch *list)
> +{
> + struct clk_init_data init;
> + struct clk_hw *hw;
> + struct intel_clk_mux *mux;
> + u32 reg = list->mux_off;
> + u8 shift = list->mux_shift;
> + u8 width = list->mux_width;
> + unsigned long cflags = list->mux_flags;
> + int ret;
> +
> + mux = kzalloc(sizeof(*mux), GFP_KERNEL);
> + if (!mux)
> + return ERR_PTR(-ENOMEM);
> +
> + init.name = list->name;
> + init.ops = &intel_clk_mux_ops;
> + init.flags = list->flags | CLK_IS_BASIC;
> + init.parent_names = list->parent_names;
> + init.num_parents = list->num_parents;
> +
> + mux->map = ctx->map;
> + mux->reg = reg;
> + mux->shift = shift;
> + mux->width = width;
> + mux->flags = cflags;
> + mux->hw.init = &init;
> +
> + hw = &mux->hw;
> + ret = clk_hw_register(NULL, hw);
> + if (ret) {
> + kfree(mux);
> + return ERR_PTR(ret);
> + }
> +
> + if (cflags & CLOCK_FLAG_VAL_INIT)
> + intel_set_clk_val(ctx->map, reg, shift, width, list->mux_val);
> +
> + return hw->clk;
> +}
> +
> +static unsigned long
> +intel_clk_divider_recalc_rate(struct clk_hw *hw,
> + unsigned long parent_rate)
> +{
> + struct intel_clk_divider *divider = to_intel_clk_divider(hw);
> + unsigned int val;
> +
> + val = intel_get_clk_val(divider->map, divider->reg,
> + divider->shift, divider->width);
> + return divider_recalc_rate(hw, parent_rate, val, divider->table,
> + divider->flags, divider->width);
> +}
> +
> +static long
> +intel_clk_divider_round_rate(struct clk_hw *hw, unsigned long rate,
> + unsigned long *prate)
> +{
> + struct intel_clk_divider *divider = to_intel_clk_divider(hw);
> +
> + return divider_round_rate(hw, rate, prate, divider->table,
> + divider->width, divider->flags);
> +}
> +
> +static int
> +intel_clk_divider_set_rate(struct clk_hw *hw, unsigned long rate,
> + unsigned long prate)
> +{
> + struct intel_clk_divider *divider = to_intel_clk_divider(hw);
> + int value;
> +
> + value = divider_get_val(rate, prate, divider->table,
> + divider->width, divider->flags);
> + if (value < 0)
> + return value;
> +
> + intel_set_clk_val(divider->map, divider->reg,
> + divider->shift, divider->width, value);
> +
> + return 0;
> +}
> +
> +const static struct clk_ops intel_clk_divider_ops = {
> + .recalc_rate = intel_clk_divider_recalc_rate,
> + .round_rate = intel_clk_divider_round_rate,
> + .set_rate = intel_clk_divider_set_rate,
> +};
> +
> +static struct clk
> +*intel_clk_register_divider(struct intel_clk_provider *ctx,
> + struct intel_clk_branch *list)
> +{
> + struct clk_init_data init;
> + struct clk_hw *hw;
> + struct intel_clk_divider *div;
> + u32 reg = list->div_off;
> + u8 shift = list->div_shift;
> + u8 width = list->div_width;
> + unsigned long cflags = list->div_flags;
> + int ret;
> +
> + div = kzalloc(sizeof(*div), GFP_KERNEL);
> + if (!div)
> + return ERR_PTR(-ENOMEM);
> +
> + init.name = list->name;
> + init.ops = &intel_clk_divider_ops;
> + init.flags = list->flags | CLK_IS_BASIC;
> + init.parent_names = &list->parent_names[0];
> + init.num_parents = 1;
> +
> + div->map = ctx->map;
> + div->reg = reg;
> + div->shift = shift;
> + div->width = width;
> + div->flags = cflags;
> + div->table = list->div_table;
> + div->hw.init = &init;
> +
> + hw = &div->hw;
> + ret = clk_hw_register(NULL, hw);
> + if (ret) {
> + pr_err("%s: register clk: %s failed!\n",
> + __func__, list->name);
> + kfree(div);
> + return ERR_PTR(ret);
> + }
> +
> + if (cflags & CLOCK_FLAG_VAL_INIT)
> + intel_set_clk_val(ctx->map, reg, shift, width, list->div_val);
> +
> + return hw->clk;
> +}
> +
> +static struct clk
> +*intel_clk_register_fixed_factor(struct intel_clk_provider *ctx,
> + struct intel_clk_branch *list)
> +{
> + struct clk_hw *hw;
> +
> + hw = clk_hw_register_fixed_factor(NULL, list->name,
> + list->parent_names[0], list->flags,
> + list->mult, list->div);
> + if (IS_ERR(hw))
> + return ERR_CAST(hw);
> +
> + if (list->div_flags & CLOCK_FLAG_VAL_INIT)
> + intel_set_clk_val(ctx->map, list->div_off, list->div_shift,
> + list->div_width, list->div_val);
> +
> + return hw->clk;
> +}
> +
> +static int
> +intel_clk_gate_enable(struct clk_hw *hw)
> +{
> + struct intel_clk_gate *gate = to_intel_clk_gate(hw);
> + unsigned int reg;
> +
> + if (gate->flags & GATE_CLK_VT) {
> + gate->reg = 1;
> + return 0;
> + }
> +
> + if (gate->flags & GATE_CLK_HW) {
> + reg = GATE_HW_REG_EN(gate->reg);
> + } else if (gate->flags & GATE_CLK_SW) {
> + reg = gate->reg;
> + } else {
> + pr_err("%s: gate clk: %s: flag 0x%lx not supported!\n",
> + __func__, clk_hw_get_name(hw), gate->flags);
> + return 0;
> + }
> +
> + intel_set_clk_val(gate->map, reg, gate->shift, 1, 1);
> +
> + return 0;
> +}
> +
> +static void
> +intel_clk_gate_disable(struct clk_hw *hw)
> +{
> + struct intel_clk_gate *gate = to_intel_clk_gate(hw);
> + unsigned int reg;
> + unsigned int set;
> +
> + if (gate->flags & GATE_CLK_VT) {
> + gate->reg = 0;
> + return;
> + }
> +
> + if (gate->flags & GATE_CLK_HW) {
> + reg = GATE_HW_REG_DIS(gate->reg);
> + set = 1;
> + } else if (gate->flags & GATE_CLK_SW) {
> + reg = gate->reg;
> + set = 0;
> + } else {
> + pr_err("%s: gate clk: %s: flag 0x%lx not supported!\n",
> + __func__, clk_hw_get_name(hw), gate->flags);
> + return;
> + }
> +
> + intel_set_clk_val(gate->map, reg, gate->shift, 1, set);
> +}
> +
> +static int
> +intel_clk_gate_is_enabled(struct clk_hw *hw)
> +{
> + struct intel_clk_gate *gate = to_intel_clk_gate(hw);
> + unsigned int reg;
> +
> + if (gate->flags & GATE_CLK_VT)
> + return gate->reg;
> +
> + if (gate->flags & GATE_CLK_HW) {
> + reg = GATE_HW_REG_STAT(gate->reg);
> + } else if (gate->flags & GATE_CLK_SW) {
> + reg = gate->reg;
> + } else {
> + pr_err("%s: gate clk: %s: flag 0x%lx not supported!\n",
> + __func__, clk_hw_get_name(hw), gate->flags);
> + return 0;
> + }
> +
> + return intel_get_clk_val(gate->map, reg, gate->shift, 1);
> +}
> +
> +const static struct clk_ops intel_clk_gate_ops = {
> + .enable = intel_clk_gate_enable,
> + .disable = intel_clk_gate_disable,
> + .is_enabled = intel_clk_gate_is_enabled,
> +};
> +
> +static struct clk
> +*intel_clk_register_gate(struct intel_clk_provider *ctx,
> + struct intel_clk_branch *list)
> +{
> + struct clk_init_data init;

Please init the init struct with { } so that future possible additions
to the structure don't require us to hunt this silent corruption down
later.

> + struct clk_hw *hw;
> + struct intel_clk_gate *gate;
> + u32 reg = list->gate_off;
> + u8 shift = list->gate_shift;
> + unsigned long cflags = list->gate_flags;
> + const char *pname = list->parent_names[0];
> + int ret;
> +
> + gate = kzalloc(sizeof(*gate), GFP_KERNEL);
> + if (!gate)
> + return ERR_PTR(-ENOMEM);
> +
> + init.name = list->name;
> + init.ops = &intel_clk_gate_ops;
> + init.flags = list->flags | CLK_IS_BASIC;
> + init.parent_names = pname ? &pname : NULL;
> + init.num_parents = pname ? 1 : 0;
> +
> + gate->map = ctx->map;
> + gate->reg = reg;
> + gate->shift = shift;
> + gate->flags = cflags;
> + gate->hw.init = &init;
> +
> + hw = &gate->hw;
> + ret = clk_hw_register(NULL, hw);
> + if (ret) {
> + kfree(gate);
> + return ERR_PTR(ret);
> + }
> +
> + if (cflags & CLOCK_FLAG_VAL_INIT)
> + intel_set_clk_val(ctx->map, reg, shift, 1, list->gate_val);
> +
> + return hw->clk;
> +}
> +
> +void intel_clk_register_branches(struct intel_clk_provider *ctx,
> + struct intel_clk_branch *list,
> + unsigned int nr_clk)
> +{
> + struct clk *clk;
> + unsigned int idx;
> +
> + for (idx = 0; idx < nr_clk; idx++, list++) {
> + switch (list->type) {
> + case intel_clk_fixed:

Please use uppercase for enums.

> + clk = intel_clk_register_fixed(ctx, list);
> + break;
> + case intel_clk_mux:
> + clk = intel_clk_register_mux(ctx, list);
> + break;
> + case intel_clk_divider:
> + clk = intel_clk_register_divider(ctx, list);
> + break;
> + case intel_clk_fixed_factor:
> + clk = intel_clk_register_fixed_factor(ctx, list);
> + break;
> + case intel_clk_gate:
> + clk = intel_clk_register_gate(ctx, list);
> + break;
> + default:
> + pr_err("%s: type: %u not supported!\n",
> + __func__, list->type);
> + return;
> + }
> +
> + if (IS_ERR(clk)) {
> + pr_err("%s: register clk: %s, type: %u failed!\n",
> + __func__, list->name, list->type);
> + return;
> + }
> +
> + intel_clk_add_lookup(ctx, clk, list->id);
> + }
> +}
> +
> +struct intel_clk_provider * __init
> +intel_clk_init(struct device_node *np, struct regmap *map, unsigned int nr_clks)
> +{
> + struct intel_clk_provider *ctx;
> + struct clk **clks;
> +
> + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
> + if (!ctx)
> + return ERR_PTR(-ENOMEM);
> +
> + clks = kcalloc(nr_clks, sizeof(*clks), GFP_KERNEL);
> + if (!clks) {
> + kfree(ctx);
> + return ERR_PTR(-ENOMEM);
> + }
> +
> + memset_p((void **)clks, ERR_PTR(-ENOENT), nr_clks);
> + ctx->map = map;
> + ctx->clk_data.clks = clks;
> + ctx->clk_data.clk_num = nr_clks;
> + ctx->np = np;
> +
> + return ctx;
> +}
> +
> +void __init intel_clk_register_osc(struct intel_clk_provider *ctx,
> + struct intel_osc_clk *osc,
> + unsigned int nr_clks)
> +{
> + u32 freq;
> + struct clk *clk;
> + int idx;
> +
> + for (idx = 0; idx < nr_clks; idx++, osc++) {
> + if (!osc->dt_freq ||
> + of_property_read_u32(ctx->np, osc->dt_freq, &freq))
> + freq = osc->def_rate;
> +
> + clk = clk_register_fixed_rate(NULL, osc->name, NULL, 0, freq);

Should come from DT itself.

> + if (IS_ERR(clk)) {
> + pr_err("%s: Failed to register clock: %s\n",
> + __func__, osc->name);
> + return;
> + }
> +
> + intel_clk_add_lookup(ctx, clk, osc->id);
> + }
> +}
> diff --git a/drivers/clk/intel/clk-cgu.h b/drivers/clk/intel/clk-cgu.h
> new file mode 100644
> index 000000000000..6dc4e45fc499
> --- /dev/null
> +++ b/drivers/clk/intel/clk-cgu.h
> @@ -0,0 +1,259 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +/*
> + * Copyright(c) 2018 Intel Corporation.
> + * Zhu YiXin <[email protected]>
> + */
> +
> +#ifndef __INTEL_CLK_H
> +#define __INTEL_CLK_H
> +
> +#define PNAME(x) static const char *const x[] __initconst
> +
> +struct intel_clk_mux {
> + struct clk_hw hw;
> + struct regmap *map;
> + unsigned int reg;
> + u8 shift;
> + u8 width;
> + unsigned long flags;
> +};
> +
> +struct intel_clk_divider {
> + struct clk_hw hw;
> + struct regmap *map;
> + unsigned int reg;
> + u8 shift;
> + u8 width;
> + unsigned long flags;
> + const struct clk_div_table *table;
> +};
> +
> +struct intel_clk_gate {
> + struct clk_hw hw;
> + struct regmap *map;
> + unsigned int reg;
> + u8 shift;
> + unsigned long flags;
> +};
> +
> +enum intel_clk_type {
> + intel_clk_fixed,
> + intel_clk_mux,
> + intel_clk_divider,
> + intel_clk_fixed_factor,
> + intel_clk_gate,
> +};
> +
> +/**
> + * struct intel_clk_provider
> + * @map: regmap type base address for register.
> + * @np: device node
> + * @clk_data: array of hw clocks and clk number.
> + */
> +struct intel_clk_provider {
> + struct regmap *map;
> + struct device_node *np;
> + struct clk_onecell_data clk_data;

Please register clk_hw pointers instead of clk pointers with the of
provider APIs.

> +};
> +
> +/**
> + * struct intel_pll_clk
> + * @id: plaform specific id of the clock.
> + * @name: name of this pll clock.
> + * @parent_names: name of the parent clock.
> + * @num_parents: number of parents.
> + * @flags: optional flags for basic clock.
> + * @type: platform type of pll.
> + * @reg: offset of the register.
> + * @mult: init value of mulitplier.
> + * @div: init value of divider.
> + * @frac: init value of fraction.
> + * @rate_table: table of pll clock rate.

Please drop the full-stop on kernel doc one-liners like this.

> + */
> +struct intel_pll_clk {
> + unsigned int id;
> + const char *name;
> + const char *const *parent_names;
> + u8 num_parents;

Can the PLL have multiple parents?

> + unsigned long flags;
> + enum intel_pll_type type;
> + int reg;
> + unsigned int mult;
> + unsigned int div;
> + unsigned int frac;
> + const struct intel_pll_rate_table *rate_table;
> +};
> +
> +#define INTEL_PLL(_id, _type, _name, _pnames, _flags, \
> + _reg, _rtable, _mult, _div, _frac) \
> + { \
> + .id = _id, \
> + .type = _type, \
> + .name = _name, \
> + .parent_names = _pnames, \
> + .num_parents = ARRAY_SIZE(_pnames), \
> + .flags = _flags, \
> + .reg = _reg, \
> + .rate_table = _rtable, \
> + .mult = _mult, \
> + .div = _div, \
> + .frac = _frac \
> + }
> +
> +/**
> + * struct intel_osc_clk
> + * @id: platform specific id of the clock.
> + * @name: name of the osc clock.
> + * @dt_freq: frequency node name in device tree.
> + * @def_rate: default rate of the osc clock.
> + * @flags: optional flags for basic clock.

There aren't flags though. I'm very confused by this kernel-doc too.
Looks like something that should be done with a fixed rate clk in DT.

> + */
> +struct intel_osc_clk {
> + unsigned int id;
> + const char *name;
> + const char *dt_freq;
> + const u32 def_rate;
> +};
> +
> +#define INTEL_OSC(_id, _name, _freq, _rate) \
> + { \
> + .id = _id, \
> + .name = _name, \
> + .dt_freq = _freq, \
> + .def_rate = _rate, \
> + }
> +
> +struct intel_clk_branch {

Seems to be more like intel_clk instead of intel_clk_branch because it
does lots of stuff.

> + unsigned int id;
> + enum intel_clk_type type;
> + const char *name;
> + const char *const *parent_names;
> + u8 num_parents;
> + unsigned long flags;
> + unsigned int mux_off;
> + u8 mux_shift;
> + u8 mux_width;
> + unsigned long mux_flags;
> + unsigned int mux_val;
> + unsigned int div_off;
> + u8 div_shift;
> + u8 div_width;
> + unsigned long div_flags;
> + unsigned int div_val;
> + const struct clk_div_table *div_table;
> + unsigned int gate_off;
> + u8 gate_shift;
> + unsigned long gate_flags;
> + unsigned int gate_val;
> + unsigned int mult;
> + unsigned int div;
> +};
> +
> +/* clock flags definition */
> +#define CLOCK_FLAG_VAL_INIT BIT(16)
> +#define GATE_CLK_HW BIT(17)
> +#define GATE_CLK_SW BIT(18)
> +#define GATE_CLK_VT BIT(19)

What does VT mean? Virtual?

> +
> +#define INTEL_MUX(_id, _name, _pname, _f, _reg, \
> + _shift, _width, _cf, _v) \
> + { \
> + .id = _id, \
> + .type = intel_clk_mux, \
> + .name = _name, \
> + .parent_names = _pname, \
> + .num_parents = ARRAY_SIZE(_pname), \
> + .flags = _f, \
> + .mux_off = _reg, \
> + .mux_shift = _shift, \
> + .mux_width = _width, \
> + .mux_flags = _cf, \
> + .mux_val = _v, \
> + }
> +
> +#define INTEL_DIV(_id, _name, _pname, _f, _reg, \
> + _shift, _width, _cf, _v, _dtable) \
> + { \
> + .id = _id, \
> + .type = intel_clk_divider, \
> + .name = _name, \
> + .parent_names = (const char *[]) { _pname }, \
> + .num_parents = 1, \
> + .flags = _f, \
> + .div_off = _reg, \
> + .div_shift = _shift, \
> + .div_width = _width, \
> + .div_flags = _cf, \
> + .div_val = _v, \
> + .div_table = _dtable, \
> + }
> +
> +#define INTEL_GATE(_id, _name, _pname, _f, _reg, \
> + _shift, _cf, _v) \
> + { \
> + .id = _id, \
> + .type = intel_clk_gate, \
> + .name = _name, \
> + .parent_names = (const char *[]) { _pname }, \
> + .num_parents = !_pname ? 0 : 1, \
> + .flags = _f, \
> + .gate_off = _reg, \
> + .gate_shift = _shift, \
> + .gate_flags = _cf, \
> + .gate_val = _v, \
> + }
> +
> +#define INTEL_FIXED(_id, _name, _pname, _f, _reg, \
> + _shift, _width, _cf, _freq, _v) \
> + { \
> + .id = _id, \
> + .type = intel_clk_fixed, \
> + .name = _name, \
> + .parent_names = (const char *[]) { _pname }, \
> + .num_parents = !_pname ? 0 : 1, \
> + .flags = _f, \
> + .div_off = _reg, \
> + .div_shift = _shift, \
> + .div_width = _width, \
> + .div_flags = _cf, \
> + .div_val = _v, \
> + .mux_flags = _freq, \
> + }
> +
> +#define INTEL_FIXED_FACTOR(_id, _name, _pname, _f, _reg, \
> + _shift, _width, _cf, _v, _m, _d) \
> + { \
> + .id = _id, \
> + .type = intel_clk_fixed_factor, \
> + .name = _name, \
> + .parent_names = (const char *[]) { _pname }, \
> + .num_parents = 1, \
> + .flags = _f, \
> + .div_off = _reg, \
> + .div_shift = _shift, \
> + .div_width = _width, \
> + .div_flags = _cf, \
> + .div_val = _v, \
> + .mult = _m, \
> + .div = _d, \
> + }
> +
> +void intel_set_clk_val(struct regmap *map, u32 reg, u8 shift,
> + u8 width, u32 set_val);
> +u32 intel_get_clk_val(struct regmap *map, u32 reg, u8 shift, u8 width);
> +void intel_clk_add_lookup(struct intel_clk_provider *ctx,
> + struct clk *clk, unsigned int id);
> +void __init intel_clk_of_add_provider(struct device_node *np,
> + struct intel_clk_provider *ctx);
> +struct intel_clk_provider * __init
> +intel_clk_init(struct device_node *np, struct regmap *map,
> + unsigned int nr_clks);
> +void __init intel_clk_register_osc(struct intel_clk_provider *ctx,
> + struct intel_osc_clk *osc,
> + unsigned int nr_clks);

Remove __init from headers files. It does nothing.

> +void intel_clk_register_branches(struct intel_clk_provider *ctx,
> + struct intel_clk_branch *list,
> + unsigned int nr_clk);
> +void intel_clk_register_plls(struct intel_clk_provider *ctx,
> + struct intel_pll_clk *list, unsigned int nr_clk);
> +#endif /* __INTEL_CLK_H */
> diff --git a/drivers/clk/intel/clk-grx500.c b/drivers/clk/intel/clk-grx500.c
> new file mode 100644
> index 000000000000..5c2546f82579
> --- /dev/null
> +++ b/drivers/clk/intel/clk-grx500.c
> @@ -0,0 +1,168 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Copyright (C) 2018 Intel Corporation.
> + * Zhu YiXin <[email protected]>
> + */
> +
> +#include <linux/clk-provider.h>
> +#include <linux/mfd/syscon.h>
> +#include <linux/of.h>
> +#include <linux/of_address.h>
> +#include <linux/regmap.h>
> +#include <linux/spinlock.h>

Used?

> +#include <dt-bindings/clock/intel,grx500-clk.h>
> +
> +#include "clk-cgu-pll.h"
> +#include "clk-cgu.h"
> +
> +#define PLL_DIV_WIDTH 4
> +
> +/* Gate1 clock shift */
> +#define G_VCODEC_SHIFT 2
> +#define G_DMA0_SHIFT 5
> +#define G_USB0_SHIFT 6
> +#define G_SPI1_SHIFT 7
> +#define G_SPI0_SHIFT 8
> +#define G_CBM_SHIFT 9
> +#define G_EBU_SHIFT 10
> +#define G_SSO_SHIFT 11
> +#define G_GPTC0_SHIFT 12
> +#define G_GPTC1_SHIFT 13
> +#define G_GPTC2_SHIFT 14
> +#define G_UART_SHIFT 17
> +#define G_CPYTO_SHIFT 20
> +#define G_SECPT_SHIFT 21
> +#define G_TOE_SHIFT 22
> +#define G_MPE_SHIFT 23
> +#define G_TDM_SHIFT 25
> +#define G_PAE_SHIFT 26
> +#define G_USB1_SHIFT 27
> +#define G_SWITCH_SHIFT 28
> +
> +/* Gate2 clock shift */
> +#define G_PCIE0_SHIFT 1
> +#define G_PCIE1_SHIFT 17
> +#define G_PCIE2_SHIFT 25
> +
> +/* Register definition */
> +#define GRX500_PLL0A_CFG0 0x0004
> +#define GRX500_PLL0A_CFG1 0x0008
> +#define GRX500_PLL0B_CFG0 0x0034
> +#define GRX500_PLL0B_CFG1 0x0038
> +#define GRX500_LCPLL_CFG0 0x0094
> +#define GRX500_LCPLL_CFG1 0x0098
> +#define GRX500_IF_CLK 0x00c4
> +#define GRX500_CLK_GSR1 0x0120
> +#define GRX500_CLK_GSR2 0x0130
> +
> +static const struct clk_div_table pll_div[] = {
> + {1, 2},

Please write it like

{ 1, 2 },

instead.

> + {2, 3},
> + {3, 4},
> + {4, 5},
> + {5, 6},
> + {6, 8},
> + {7, 10},
> + {8, 12},
> + {9, 16},
> + {10, 20},
> + {11, 24},
> + {12, 32},
> + {13, 40},
> + {14, 48},
> + {15, 64}
> +};
> +
> +enum grx500_plls {
> + pll0a, pll0b, pll3,
> +};

What's the point of the enum?

> +
> +PNAME(pll_p) = { "osc" };
> +PNAME(cpu_p) = { "cpu0", "cpu1" };
> +
> +static struct intel_osc_clk grx500_osc_clks[] __initdata = {
> + INTEL_OSC(CLK_OSC, "osc", "intel,osc-frequency", 40000000),
> +};
> +
> +static struct intel_pll_clk grx500_pll_clks[] __initdata = {
> + [pll0a] = INTEL_PLL(CLK_PLL0A, pll_grx500, "pll0a",
> + pll_p, 0, GRX500_PLL0A_CFG0, NULL, 0, 0, 0),
> + [pll0b] = INTEL_PLL(CLK_PLL0B, pll_grx500, "pll0b",
> + pll_p, 0, GRX500_PLL0B_CFG0, NULL, 0, 0, 0),
> + [pll3] = INTEL_PLL(CLK_PLL3, pll_grx500, "pll3",
> + pll_p, 0, GRX500_LCPLL_CFG0, NULL, 0, 0, 0),
> +};
> +
> +static struct intel_clk_branch grx500_branch_clks[] __initdata = {
> + INTEL_DIV(CLK_CBM, "cbm", "pll0a", 0, GRX500_PLL0A_CFG1,
> + 0, PLL_DIV_WIDTH, 0, 0, pll_div),
> + INTEL_DIV(CLK_NGI, "ngi", "pll0a", 0, GRX500_PLL0A_CFG1,
> + 4, PLL_DIV_WIDTH, 0, 0, pll_div),
> + INTEL_DIV(CLK_SSX4, "ssx4", "pll0a", 0, GRX500_PLL0A_CFG1,
> + 8, PLL_DIV_WIDTH, 0, 0, pll_div),
> + INTEL_DIV(CLK_CPU0, "cpu0", "pll0a", 0, GRX500_PLL0A_CFG1,
> + 12, PLL_DIV_WIDTH, 0, 0, pll_div),
> + INTEL_DIV(CLK_PAE, "pae", "pll0b", 0, GRX500_PLL0B_CFG1,
> + 0, PLL_DIV_WIDTH, 0, 0, pll_div),
> + INTEL_DIV(CLK_GSWIP, "gswip", "pll0b", 0, GRX500_PLL0B_CFG1,
> + 4, PLL_DIV_WIDTH, 0, 0, pll_div),
> + INTEL_DIV(CLK_DDR, "ddr", "pll0b", 0, GRX500_PLL0B_CFG1,
> + 8, PLL_DIV_WIDTH, 0, 0, pll_div),
> + INTEL_DIV(CLK_CPU1, "cpu1", "pll0b", 0, GRX500_PLL0B_CFG1,
> + 12, PLL_DIV_WIDTH, 0, 0, pll_div),
> + INTEL_MUX(CLK_CPU, "cpu", cpu_p, CLK_SET_RATE_PARENT,
> + GRX500_PLL0A_CFG1, 29, 1, 0, 0),
> + INTEL_GATE(GCLK_DMA0, "g_dma0", NULL, 0, GRX500_CLK_GSR1,
> + G_DMA0_SHIFT, GATE_CLK_HW, 0),
> + INTEL_GATE(GCLK_USB0, "g_usb0", NULL, 0, GRX500_CLK_GSR1,
> + G_USB0_SHIFT, GATE_CLK_HW, 0),
> + INTEL_GATE(GCLK_GPTC0, "g_gptc0", NULL, 0, GRX500_CLK_GSR1,
> + G_GPTC0_SHIFT, GATE_CLK_HW, 0),
> + INTEL_GATE(GCLK_GPTC1, "g_gptc1", NULL, 0, GRX500_CLK_GSR1,
> + G_GPTC1_SHIFT, GATE_CLK_HW, 0),
> + INTEL_GATE(GCLK_GPTC2, "g_gptc2", NULL, 0, GRX500_CLK_GSR1,
> + G_GPTC2_SHIFT, GATE_CLK_HW, 0),
> + INTEL_GATE(GCLK_UART, "g_uart", NULL, 0, GRX500_CLK_GSR1,
> + G_UART_SHIFT, GATE_CLK_HW, 0),
> + INTEL_GATE(GCLK_PCIE0, "g_pcie0", NULL, 0, GRX500_CLK_GSR2,
> + G_PCIE0_SHIFT, GATE_CLK_HW, 0),
> + INTEL_GATE(GCLK_PCIE1, "g_pcie1", NULL, 0, GRX500_CLK_GSR2,
> + G_PCIE1_SHIFT, GATE_CLK_HW, 0),
> + INTEL_GATE(GCLK_PCIE2, "g_pcie2", NULL, 0, GRX500_CLK_GSR2,
> + G_PCIE2_SHIFT, GATE_CLK_HW, 0),
> + INTEL_GATE(GCLK_I2C, "g_i2c", NULL, 0, 0, 0, GATE_CLK_VT, 0),
> + INTEL_FIXED(CLK_VOICE, "voice", NULL, 0, GRX500_IF_CLK, 14, 2,
> + CLOCK_FLAG_VAL_INIT, 8192000, 2),
> + INTEL_FIXED_FACTOR(CLK_DDRPHY, "ddrphy", "ddr", 0, 0, 0,
> + 0, 0, 0, 2, 1),
> + INTEL_FIXED_FACTOR(CLK_PCIE, "pcie", "pll3", 0, 0, 0,
> + 0, 0, 0, 1, 40),
> +};
> +
> +static void __init grx500_clk_init(struct device_node *np)
> +{
> + struct intel_clk_provider *ctx;
> + struct regmap *map;
> +
> + map = syscon_node_to_regmap(np);
> + if (IS_ERR(map))
> + return;
> +
> + ctx = intel_clk_init(np, map, CLK_NR_CLKS);
> + if (IS_ERR(ctx)) {
> + regmap_exit(map);
> + return;
> + }
> +
> + intel_clk_register_osc(ctx, grx500_osc_clks,
> + ARRAY_SIZE(grx500_osc_clks));
> + intel_clk_register_plls(ctx, grx500_pll_clks,
> + ARRAY_SIZE(grx500_pll_clks));
> + intel_clk_register_branches(ctx, grx500_branch_clks,
> + ARRAY_SIZE(grx500_branch_clks));
> + of_clk_add_provider(np, of_clk_src_onecell_get, &ctx->clk_data);
> +
> + pr_debug("%s clk init done!\n", __func__);

Yay!!!

> +}
> +
> +CLK_OF_DECLARE(intel_grx500_cgu, "intel,grx500-cgu", grx500_clk_init);

Any reason a platform driver can't be used instead of CLK_OF_DECLARE()?


2018-08-08 08:36:39

by Geert Uytterhoeven

[permalink] [raw]
Subject: Re: [PATCH v2 08/18] serial: intel: Get serial id from dts

Hi Songjun,

On Wed, Aug 8, 2018 at 6:05 AM Wu, Songjun <[email protected]> wrote:
> On 8/7/2018 3:33 PM, Geert Uytterhoeven wrote:
> > On Fri, Aug 3, 2018 at 5:04 AM Songjun Wu <[email protected]> wrote:
> >> Get serial id from dts.
> >>
> >> "#ifdef CONFIG_LANTIQ" preprocessor is used because LTQ_EARLY_ASC
> >> macro is defined in lantiq_soc.h.
> >> lantiq_soc.h is in arch path for legacy product support.
> >>
> >> arch/mips/include/asm/mach-lantiq/xway/lantiq_soc.h
> >>
> >> If "#ifdef preprocessor" is changed to
> >> "if (IS_ENABLED(CONFIG_LANTIQ))", when CONFIG_LANTIQ is not enabled,
> >> code using LTQ_EARLY_ASC is compiled.
> >> Compilation will fail for no LTQ_EARLY_ASC defined.
> >>
> >> Signed-off-by: Songjun Wu <[email protected]>
> > Thanks for your patch!
> >
> >> @@ -699,9 +700,19 @@ lqasc_probe(struct platform_device *pdev)
> >> return -ENODEV;
> >> }
> >>
> >> - /* check if this is the console port */
> >> - if (mmres->start != CPHYSADDR(LTQ_EARLY_ASC))
> >> - line = 1;
> >> + /* get serial id */
> >> + line = of_alias_get_id(node, "serial");
> >> + if (line < 0) {
> >> +#ifdef CONFIG_LANTIQ
> >> + if (mmres->start == CPHYSADDR(LTQ_EARLY_ASC))
> >> + line = 0;
> >> + else
> >> + line = 1;
> >> +#else
> >> + dev_err(&pdev->dev, "failed to get alias id, errno %d\n", line);
> >> + return line;
> > Please note that not providing a fallback here makes life harder when using
> > DT overlays.
> > See the description of commit 7678f4c20fa7670f ("serial: sh-sci: Add support
> > for dynamic instances") for background info.
> Thanks for your comment.
> The logic in commit 7678f4c20fa7670f is not suitable here.
> We need to know which serial instance is used for console.
> We cannot use dynamic serial instance here.

Why does the driver need to use which serial instance is used for the console?
Hardcoding that is not an option, as the board DTS may specify the console using
chosen/stdout-path.

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

2018-08-08 08:53:32

by Zhu, Yi Xin

[permalink] [raw]
Subject: Re: [PATCH v2 02/18] clk: intel: Add clock driver for Intel MIPS SoCs



On 8/8/2018 1:50 PM, Stephen Boyd wrote:
> Quoting Songjun Wu (2018-08-02 20:02:21)
>> From: Yixin Zhu <[email protected]>
>>
>> This driver provides PLL clock registration as well as various clock
>> branches, e.g. MUX clock, gate clock, divider clock and so on.
>>
>> PLLs that provide clock to DDR, CPU and peripherals are shown below:
>>
>> +---------+
>> |--->| LCPLL3 0|--PCIe clk-->
>> XO | +---------+
>> +-----------|
>> | +---------+
>> | | 3|--PAE clk-->
>> |--->| PLL0B 2|--GSWIP clk-->
>> | | 1|--DDR clk-->DDR PHY clk-->
>> | | 0|--CPU1 clk--+ +-----+
>> | +---------+ |--->0 |
>> | | MUX |--CPU clk-->
>> | +---------+ |--->1 |
>> | | 0|--CPU0 clk--+ +-----+
>> |--->| PLLOA 1|--SSX4 clk-->
>> | 2|--NGI clk-->
>> | 3|--CBM clk-->
>> +---------+
> Thanks for the picture!
Thanks for the review.

>
>> Signed-off-by: Yixin Zhu <[email protected]>
>> Signed-off-by: Songjun Wu <[email protected]>
>> diff --git a/drivers/clk/Makefile b/drivers/clk/Makefile
>> index 0bb25dd009d1..d929ca4607cf 100644
>> --- a/drivers/clk/Makefile
>> +++ b/drivers/clk/Makefile
>> @@ -72,6 +72,9 @@ obj-$(CONFIG_ARCH_HISI) += hisilicon/
>> obj-y += imgtec/
>> obj-$(CONFIG_ARCH_MXC) += imx/
>> obj-$(CONFIG_MACH_INGENIC) += ingenic/
>> +ifeq ($(CONFIG_COMMON_CLK), y)
>> +obj-y +=intel/
>> +endif
> Why not obj-$(CONFIG_INTEL_CCF) or something like that?
Will use obj-$(CONFIG_INTEL_CCF)

>> obj-$(CONFIG_ARCH_KEYSTONE) += keystone/
>> obj-$(CONFIG_MACH_LOONGSON32) += loongson1/
>> obj-y += mediatek/
>> diff --git a/drivers/clk/intel/Kconfig b/drivers/clk/intel/Kconfig
>> new file mode 100644
>> index 000000000000..c7d3fb1721fa
>> --- /dev/null
>> +++ b/drivers/clk/intel/Kconfig
>> @@ -0,0 +1,20 @@
>> +# SPDX-License-Identifier: GPL-2.0
>> +config INTEL_CGU_CLK
>> + depends on COMMON_CLK
>> + depends on INTEL_MIPS || COMPILE_TEST
>> + select MFD_SYSCON
>> + bool "Intel clock controller support"
>> + help
>> + This driver support Intel CGU (Clock Generation Unit).
> Is it really called a clock generation unit? Or that's just copied from
> sunxi driver?
Yes,  It's called clock generation unit(CGU) in our HW chip spec.

>> +
>> +choice
>> + prompt "SoC platform selection"
>> + depends on INTEL_CGU_CLK
>> + default INTEL_GRX500_CGU_CLK
>> +
>> +config INTEL_GRX500_CGU_CLK
>> + bool "GRX500 CLK"
>> + help
>> + Clock driver of GRX500 platform.
>> +
>> +endchoice
>> diff --git a/drivers/clk/intel/Makefile b/drivers/clk/intel/Makefile
>> new file mode 100644
>> index 000000000000..16a0138e52c2
>> --- /dev/null
>> +++ b/drivers/clk/intel/Makefile
>> @@ -0,0 +1,7 @@
>> +# SPDX-License-Identifier: GPL-2.0
>> +# Makefile for intel specific clk
>> +
>> +obj-$(CONFIG_INTEL_CGU_CLK) += clk-cgu.o clk-cgu-pll.o
>> +ifneq ($(CONFIG_INTEL_GRX500_CGU_CLK),)
>> + obj-y += clk-grx500.o
>> +endif
>> diff --git a/drivers/clk/intel/clk-cgu-pll.c b/drivers/clk/intel/clk-cgu-pll.c
>> new file mode 100644
>> index 000000000000..20759bc27e95
>> --- /dev/null
>> +++ b/drivers/clk/intel/clk-cgu-pll.c
>> @@ -0,0 +1,166 @@
>> +// SPDX-License-Identifier: GPL-2.0
>> +/*
>> + * Copyright (C) 2018 Intel Corporation.
>> + * Zhu YiXin <[email protected]>
>> + */
>> +
>> +#include <linux/clk.h>
> Is this include used?
Not used. Will remove it.

>> +#include <linux/clk-provider.h>
>> +#include <linux/clkdev.h>
>> +#include <linux/mfd/syscon.h>
>> +#include <linux/of.h>
>> +#include <linux/of_address.h>
>> +#include <linux/regmap.h>
>> +#include <linux/slab.h>
>> +
>> +#include "clk-cgu-pll.h"
>> +#include "clk-cgu.h"
>> +
>> +#define to_intel_clk_pll(_hw) container_of(_hw, struct intel_clk_pll, hw)
>> +
>> +/*
>> + * Calculate formula:
>> + * rate = (prate * mult + (prate * frac) / frac_div) / div
>> + */
>> +static unsigned long
>> +intel_pll_calc_rate(unsigned long prate, unsigned int mult,
>> + unsigned int div, unsigned int frac,
>> + unsigned int frac_div)
>> +{
>> + u64 crate, frate, rate64;
>> +
>> + rate64 = prate;
>> + crate = rate64 * mult;
>> +
>> + if (frac) {
>> + frate = rate64 * frac;
>> + do_div(frate, frac_div);
>> + crate += frate;
>> + }
>> + do_div(crate, div);
>> +
>> + return (unsigned long)crate;
>> +}
>> +
>> +static void
>> +grx500_pll_get_params(struct intel_clk_pll *pll, unsigned int *mult,
>> + unsigned int *frac)
>> +{
>> + *mult = intel_get_clk_val(pll->map, pll->reg, 2, 7);
>> + *frac = intel_get_clk_val(pll->map, pll->reg, 9, 21);
>> +}
>> +
>> +static int intel_wait_pll_lock(struct intel_clk_pll *pll, int bit_idx)
>> +{
>> + unsigned int val;
>> +
>> + return regmap_read_poll_timeout(pll->map, pll->reg, val,
>> + val & BIT(bit_idx), 10, 1000);
>> +}
>> +
>> +static unsigned long
>> +intel_grx500_pll_recalc_rate(struct clk_hw *hw, unsigned long prate)
>> +{
>> + struct intel_clk_pll *pll = to_intel_clk_pll(hw);
>> + unsigned int mult, frac;
>> +
>> + grx500_pll_get_params(pll, &mult, &frac);
>> +
>> + return intel_pll_calc_rate(prate, mult, 1, frac, BIT(20));
>> +}
>> +
>> +static int intel_grx500_pll_is_enabled(struct clk_hw *hw)
>> +{
>> + struct intel_clk_pll *pll = to_intel_clk_pll(hw);
>> +
>> + if (intel_wait_pll_lock(pll, 1)) {
>> + pr_err("%s: pll: %s is not locked!\n",
>> + __func__, clk_hw_get_name(hw));
>> + return 0;
>> + }
>> +
>> + return intel_get_clk_val(pll->map, pll->reg, 1, 1);
>> +}
>> +
>> +const static struct clk_ops intel_grx500_pll_ops = {
> Should be static const struct ...
Will update it.

>> + .recalc_rate = intel_grx500_pll_recalc_rate,
>> + .is_enabled = intel_grx500_pll_is_enabled,
>> +};
>> +
>> +static struct clk
>> +*intel_clk_register_pll(struct intel_clk_provider *ctx,
>> + enum intel_pll_type type, const char *cname,
>> + const char *const *pname, u8 num_parents,
>> + unsigned long flags, unsigned int reg,
>> + const struct intel_pll_rate_table *table,
>> + unsigned int mult, unsigned int div, unsigned int frac)
>> +{
>> + struct clk_init_data init;
>> + struct intel_clk_pll *pll;
>> + struct clk_hw *hw;
>> + int ret, i;
>> +
>> + if (type != pll_grx500) {
>> + pr_err("%s: pll type %d not supported!\n",
>> + __func__, type);
>> + return ERR_PTR(-EINVAL);
>> + }
>> + init.name = cname;
>> + init.ops = &intel_grx500_pll_ops;
>> + init.flags = CLK_IS_BASIC;
> Don't use this flag unless you have some reason to need it.
Will remove it.

>> + init.parent_names = pname;
>> + init.num_parents = num_parents;
>> +
>> + pll = kzalloc(sizeof(*pll), GFP_KERNEL);
>> + if (!pll)
>> + return ERR_PTR(-ENOMEM);
>> + pll->map = ctx->map;
>> + pll->reg = reg;
>> + pll->flags = flags;
>> + pll->mult = mult;
>> + pll->div = div;
>> + pll->frac = frac;
>> + pll->hw.init = &init;
>> + if (table) {
>> + for (i = 0; table[i].rate != 0; i++)
>> + ;
>> + pll->table_sz = i;
>> + pll->rate_table = kmemdup(table, i * sizeof(table[0]),
>> + GFP_KERNEL);
>> + if (!pll->rate_table) {
>> + ret = -ENOMEM;
>> + goto err_free_pll;
>> + }
>> + }
>> + hw = &pll->hw;
>> + ret = clk_hw_register(NULL, hw);
>> + if (ret)
>> + goto err_free_pll;
>> +
>> + return hw->clk;
>> +
>> +err_free_pll:
>> + kfree(pll);
>> + return ERR_PTR(ret);
>> +}
>> +
>> +void intel_clk_register_plls(struct intel_clk_provider *ctx,
>> + struct intel_pll_clk *list, unsigned int nr_clk)
>> +{
>> + struct clk *clk;
>> + int i;
>> +
>> + for (i = 0; i < nr_clk; i++, list++) {
>> + clk = intel_clk_register_pll(ctx, list->type, list->name,
>> + list->parent_names, list->num_parents,
>> + list->flags, list->reg, list->rate_table,
>> + list->mult, list->div, list->frac);
>> + if (IS_ERR(clk)) {
>> + pr_err("%s: failed to register pll: %s\n",
>> + __func__, list->name);
>> + continue;
>> + }
>> +
>> + intel_clk_add_lookup(ctx, clk, list->id);
>> + }
>> +}
>> diff --git a/drivers/clk/intel/clk-cgu-pll.h b/drivers/clk/intel/clk-cgu-pll.h
>> new file mode 100644
>> index 000000000000..3e7cff1d5e16
>> --- /dev/null
>> +++ b/drivers/clk/intel/clk-cgu-pll.h
>> @@ -0,0 +1,34 @@
>> +/* SPDX-License-Identifier: GPL-2.0 */
>> +/*
>> + * Copyright(c) 2018 Intel Corporation.
>> + * Zhu YiXin <[email protected]>
>> + */
>> +
>> +#ifndef __INTEL_CLK_PLL_H
>> +#define __INTEL_CLK_PLL_H
>> +
>> +enum intel_pll_type {
>> + pll_grx500,
>> +};
>> +
>> +struct intel_pll_rate_table {
>> + unsigned long prate;
>> + unsigned long rate;
>> + unsigned int mult;
>> + unsigned int div;
>> + unsigned int frac;
>> +};
>> +
>> +struct intel_clk_pll {
>> + struct clk_hw hw;
>> + struct regmap *map;
>> + unsigned int reg;
>> + unsigned long flags;
>> + unsigned int mult;
>> + unsigned int div;
>> + unsigned int frac;
>> + unsigned int table_sz;
>> + const struct intel_pll_rate_table *rate_table;
>> +};
>> +
>> +#endif /* __INTEL_CLK_PLL_H */
>> diff --git a/drivers/clk/intel/clk-cgu.c b/drivers/clk/intel/clk-cgu.c
>> new file mode 100644
>> index 000000000000..10cacbe0fbcd
>> --- /dev/null
>> +++ b/drivers/clk/intel/clk-cgu.c
>> @@ -0,0 +1,470 @@
>> +// SPDX-License-Identifier: GPL-2.0
>> +/*
>> + * Copyright (C) 2018 Intel Corporation.
>> + * Zhu YiXin <[email protected]>
>> + */
>> +
>> +#include <linux/clk.h>
>> +#include <linux/clk-provider.h>
>> +#include <linux/clkdev.h>
>> +#include <linux/mfd/syscon.h>
>> +#include <linux/of.h>
>> +#include <linux/of_address.h>
>> +#include <linux/regmap.h>
>> +#include <linux/slab.h>
>> +
>> +#include "clk-cgu-pll.h"
>> +#include "clk-cgu.h"
>> +
>> +#define GATE_HW_REG_STAT(reg) (reg)
>> +#define GATE_HW_REG_EN(reg) ((reg) + 0x4)
>> +#define GATE_HW_REG_DIS(reg) ((reg) + 0x8)
>> +
>> +#define to_intel_clk_mux(_hw) container_of(_hw, struct intel_clk_mux, hw)
>> +#define to_intel_clk_divider(_hw) \
>> + container_of(_hw, struct intel_clk_divider, hw)
>> +#define to_intel_clk_gate(_hw) container_of(_hw, struct intel_clk_gate, hw)
>> +
>> +void intel_set_clk_val(struct regmap *map, u32 reg, u8 shift,
>> + u8 width, u32 set_val)
>> +{
>> + u32 mask = GENMASK(width + shift, shift);
>> +
>> + regmap_update_bits(map, reg, mask, set_val << shift);
>> +}
>> +
>> +u32 intel_get_clk_val(struct regmap *map, u32 reg, u8 shift,
>> + u8 width)
>> +{
>> + u32 val;
>> +
>> + if (regmap_read(map, reg, &val)) {
>> + WARN_ONCE(1, "Failed to read clk reg: 0x%x\n", reg);
>> + return 0;
>> + }
>> + val >>= shift;
>> + val &= BIT(width) - 1;
>> +
>> + return val;
>> +}
>> +
>> +void intel_clk_add_lookup(struct intel_clk_provider *ctx,
>> + struct clk *clk, unsigned int id)
>> +{
>> + pr_debug("Add clk: %s, id: %u\n", __clk_get_name(clk), id);
>> + if (ctx->clk_data.clks && id)
>> + ctx->clk_data.clks[id] = clk;
>> +}
>> +
>> +static struct clk
>> +*intel_clk_register_fixed(struct intel_clk_provider *ctx,
>> + struct intel_clk_branch *list)
>> +{
>> + if (list->div_flags & CLOCK_FLAG_VAL_INIT)
>> + intel_set_clk_val(ctx->map, list->div_off, list->div_shift,
>> + list->div_width, list->div_val);
>> +
>> + return clk_register_fixed_rate(NULL, list->name, list->parent_names[0],
>> + list->flags, list->mux_flags);
>> +}
>> +
>> +static u8 intel_clk_mux_get_parent(struct clk_hw *hw)
>> +{
>> + struct intel_clk_mux *mux = to_intel_clk_mux(hw);
>> + u32 val;
>> +
>> + val = intel_get_clk_val(mux->map, mux->reg, mux->shift, mux->width);
>> + return clk_mux_val_to_index(hw, NULL, mux->flags, val);
>> +}
>> +
>> +static int intel_clk_mux_set_parent(struct clk_hw *hw, u8 index)
>> +{
>> + struct intel_clk_mux *mux = to_intel_clk_mux(hw);
>> + u32 val;
>> +
>> + val = clk_mux_index_to_val(NULL, mux->flags, index);
>> + intel_set_clk_val(mux->map, mux->reg, mux->shift, mux->width, val);
>> +
>> + return 0;
>> +}
>> +
>> +static int intel_clk_mux_determine_rate(struct clk_hw *hw,
>> + struct clk_rate_request *req)
>> +{
>> + struct intel_clk_mux *mux = to_intel_clk_mux(hw);
>> +
>> + return clk_mux_determine_rate_flags(hw, req, mux->flags);
>> +}
>> +
>> +const static struct clk_ops intel_clk_mux_ops = {
>> + .get_parent = intel_clk_mux_get_parent,
>> + .set_parent = intel_clk_mux_set_parent,
>> + .determine_rate = intel_clk_mux_determine_rate,
>> +};
>> +
>> +static struct clk
>> +*intel_clk_register_mux(struct intel_clk_provider *ctx,
>> + struct intel_clk_branch *list)
>> +{
>> + struct clk_init_data init;
>> + struct clk_hw *hw;
>> + struct intel_clk_mux *mux;
>> + u32 reg = list->mux_off;
>> + u8 shift = list->mux_shift;
>> + u8 width = list->mux_width;
>> + unsigned long cflags = list->mux_flags;
>> + int ret;
>> +
>> + mux = kzalloc(sizeof(*mux), GFP_KERNEL);
>> + if (!mux)
>> + return ERR_PTR(-ENOMEM);
>> +
>> + init.name = list->name;
>> + init.ops = &intel_clk_mux_ops;
>> + init.flags = list->flags | CLK_IS_BASIC;
>> + init.parent_names = list->parent_names;
>> + init.num_parents = list->num_parents;
>> +
>> + mux->map = ctx->map;
>> + mux->reg = reg;
>> + mux->shift = shift;
>> + mux->width = width;
>> + mux->flags = cflags;
>> + mux->hw.init = &init;
>> +
>> + hw = &mux->hw;
>> + ret = clk_hw_register(NULL, hw);
>> + if (ret) {
>> + kfree(mux);
>> + return ERR_PTR(ret);
>> + }
>> +
>> + if (cflags & CLOCK_FLAG_VAL_INIT)
>> + intel_set_clk_val(ctx->map, reg, shift, width, list->mux_val);
>> +
>> + return hw->clk;
>> +}
>> +
>> +static unsigned long
>> +intel_clk_divider_recalc_rate(struct clk_hw *hw,
>> + unsigned long parent_rate)
>> +{
>> + struct intel_clk_divider *divider = to_intel_clk_divider(hw);
>> + unsigned int val;
>> +
>> + val = intel_get_clk_val(divider->map, divider->reg,
>> + divider->shift, divider->width);
>> + return divider_recalc_rate(hw, parent_rate, val, divider->table,
>> + divider->flags, divider->width);
>> +}
>> +
>> +static long
>> +intel_clk_divider_round_rate(struct clk_hw *hw, unsigned long rate,
>> + unsigned long *prate)
>> +{
>> + struct intel_clk_divider *divider = to_intel_clk_divider(hw);
>> +
>> + return divider_round_rate(hw, rate, prate, divider->table,
>> + divider->width, divider->flags);
>> +}
>> +
>> +static int
>> +intel_clk_divider_set_rate(struct clk_hw *hw, unsigned long rate,
>> + unsigned long prate)
>> +{
>> + struct intel_clk_divider *divider = to_intel_clk_divider(hw);
>> + int value;
>> +
>> + value = divider_get_val(rate, prate, divider->table,
>> + divider->width, divider->flags);
>> + if (value < 0)
>> + return value;
>> +
>> + intel_set_clk_val(divider->map, divider->reg,
>> + divider->shift, divider->width, value);
>> +
>> + return 0;
>> +}
>> +
>> +const static struct clk_ops intel_clk_divider_ops = {
>> + .recalc_rate = intel_clk_divider_recalc_rate,
>> + .round_rate = intel_clk_divider_round_rate,
>> + .set_rate = intel_clk_divider_set_rate,
>> +};
>> +
>> +static struct clk
>> +*intel_clk_register_divider(struct intel_clk_provider *ctx,
>> + struct intel_clk_branch *list)
>> +{
>> + struct clk_init_data init;
>> + struct clk_hw *hw;
>> + struct intel_clk_divider *div;
>> + u32 reg = list->div_off;
>> + u8 shift = list->div_shift;
>> + u8 width = list->div_width;
>> + unsigned long cflags = list->div_flags;
>> + int ret;
>> +
>> + div = kzalloc(sizeof(*div), GFP_KERNEL);
>> + if (!div)
>> + return ERR_PTR(-ENOMEM);
>> +
>> + init.name = list->name;
>> + init.ops = &intel_clk_divider_ops;
>> + init.flags = list->flags | CLK_IS_BASIC;
>> + init.parent_names = &list->parent_names[0];
>> + init.num_parents = 1;
>> +
>> + div->map = ctx->map;
>> + div->reg = reg;
>> + div->shift = shift;
>> + div->width = width;
>> + div->flags = cflags;
>> + div->table = list->div_table;
>> + div->hw.init = &init;
>> +
>> + hw = &div->hw;
>> + ret = clk_hw_register(NULL, hw);
>> + if (ret) {
>> + pr_err("%s: register clk: %s failed!\n",
>> + __func__, list->name);
>> + kfree(div);
>> + return ERR_PTR(ret);
>> + }
>> +
>> + if (cflags & CLOCK_FLAG_VAL_INIT)
>> + intel_set_clk_val(ctx->map, reg, shift, width, list->div_val);
>> +
>> + return hw->clk;
>> +}
>> +
>> +static struct clk
>> +*intel_clk_register_fixed_factor(struct intel_clk_provider *ctx,
>> + struct intel_clk_branch *list)
>> +{
>> + struct clk_hw *hw;
>> +
>> + hw = clk_hw_register_fixed_factor(NULL, list->name,
>> + list->parent_names[0], list->flags,
>> + list->mult, list->div);
>> + if (IS_ERR(hw))
>> + return ERR_CAST(hw);
>> +
>> + if (list->div_flags & CLOCK_FLAG_VAL_INIT)
>> + intel_set_clk_val(ctx->map, list->div_off, list->div_shift,
>> + list->div_width, list->div_val);
>> +
>> + return hw->clk;
>> +}
>> +
>> +static int
>> +intel_clk_gate_enable(struct clk_hw *hw)
>> +{
>> + struct intel_clk_gate *gate = to_intel_clk_gate(hw);
>> + unsigned int reg;
>> +
>> + if (gate->flags & GATE_CLK_VT) {
>> + gate->reg = 1;
>> + return 0;
>> + }
>> +
>> + if (gate->flags & GATE_CLK_HW) {
>> + reg = GATE_HW_REG_EN(gate->reg);
>> + } else if (gate->flags & GATE_CLK_SW) {
>> + reg = gate->reg;
>> + } else {
>> + pr_err("%s: gate clk: %s: flag 0x%lx not supported!\n",
>> + __func__, clk_hw_get_name(hw), gate->flags);
>> + return 0;
>> + }
>> +
>> + intel_set_clk_val(gate->map, reg, gate->shift, 1, 1);
>> +
>> + return 0;
>> +}
>> +
>> +static void
>> +intel_clk_gate_disable(struct clk_hw *hw)
>> +{
>> + struct intel_clk_gate *gate = to_intel_clk_gate(hw);
>> + unsigned int reg;
>> + unsigned int set;
>> +
>> + if (gate->flags & GATE_CLK_VT) {
>> + gate->reg = 0;
>> + return;
>> + }
>> +
>> + if (gate->flags & GATE_CLK_HW) {
>> + reg = GATE_HW_REG_DIS(gate->reg);
>> + set = 1;
>> + } else if (gate->flags & GATE_CLK_SW) {
>> + reg = gate->reg;
>> + set = 0;
>> + } else {
>> + pr_err("%s: gate clk: %s: flag 0x%lx not supported!\n",
>> + __func__, clk_hw_get_name(hw), gate->flags);
>> + return;
>> + }
>> +
>> + intel_set_clk_val(gate->map, reg, gate->shift, 1, set);
>> +}
>> +
>> +static int
>> +intel_clk_gate_is_enabled(struct clk_hw *hw)
>> +{
>> + struct intel_clk_gate *gate = to_intel_clk_gate(hw);
>> + unsigned int reg;
>> +
>> + if (gate->flags & GATE_CLK_VT)
>> + return gate->reg;
>> +
>> + if (gate->flags & GATE_CLK_HW) {
>> + reg = GATE_HW_REG_STAT(gate->reg);
>> + } else if (gate->flags & GATE_CLK_SW) {
>> + reg = gate->reg;
>> + } else {
>> + pr_err("%s: gate clk: %s: flag 0x%lx not supported!\n",
>> + __func__, clk_hw_get_name(hw), gate->flags);
>> + return 0;
>> + }
>> +
>> + return intel_get_clk_val(gate->map, reg, gate->shift, 1);
>> +}
>> +
>> +const static struct clk_ops intel_clk_gate_ops = {
>> + .enable = intel_clk_gate_enable,
>> + .disable = intel_clk_gate_disable,
>> + .is_enabled = intel_clk_gate_is_enabled,
>> +};
>> +
>> +static struct clk
>> +*intel_clk_register_gate(struct intel_clk_provider *ctx,
>> + struct intel_clk_branch *list)
>> +{
>> + struct clk_init_data init;
> Please init the init struct with { } so that future possible additions
> to the structure don't require us to hunt this silent corruption down
> later.
Will update it.

>> + struct clk_hw *hw;
>> + struct intel_clk_gate *gate;
>> + u32 reg = list->gate_off;
>> + u8 shift = list->gate_shift;
>> + unsigned long cflags = list->gate_flags;
>> + const char *pname = list->parent_names[0];
>> + int ret;
>> +
>> + gate = kzalloc(sizeof(*gate), GFP_KERNEL);
>> + if (!gate)
>> + return ERR_PTR(-ENOMEM);
>> +
>> + init.name = list->name;
>> + init.ops = &intel_clk_gate_ops;
>> + init.flags = list->flags | CLK_IS_BASIC;
>> + init.parent_names = pname ? &pname : NULL;
>> + init.num_parents = pname ? 1 : 0;
>> +
>> + gate->map = ctx->map;
>> + gate->reg = reg;
>> + gate->shift = shift;
>> + gate->flags = cflags;
>> + gate->hw.init = &init;
>> +
>> + hw = &gate->hw;
>> + ret = clk_hw_register(NULL, hw);
>> + if (ret) {
>> + kfree(gate);
>> + return ERR_PTR(ret);
>> + }
>> +
>> + if (cflags & CLOCK_FLAG_VAL_INIT)
>> + intel_set_clk_val(ctx->map, reg, shift, 1, list->gate_val);
>> +
>> + return hw->clk;
>> +}
>> +
>> +void intel_clk_register_branches(struct intel_clk_provider *ctx,
>> + struct intel_clk_branch *list,
>> + unsigned int nr_clk)
>> +{
>> + struct clk *clk;
>> + unsigned int idx;
>> +
>> + for (idx = 0; idx < nr_clk; idx++, list++) {
>> + switch (list->type) {
>> + case intel_clk_fixed:
> Please use uppercase for enums.
Will update.

>
>> + clk = intel_clk_register_fixed(ctx, list);
>> + break;
>> + case intel_clk_mux:
>> + clk = intel_clk_register_mux(ctx, list);
>> + break;
>> + case intel_clk_divider:
>> + clk = intel_clk_register_divider(ctx, list);
>> + break;
>> + case intel_clk_fixed_factor:
>> + clk = intel_clk_register_fixed_factor(ctx, list);
>> + break;
>> + case intel_clk_gate:
>> + clk = intel_clk_register_gate(ctx, list);
>> + break;
>> + default:
>> + pr_err("%s: type: %u not supported!\n",
>> + __func__, list->type);
>> + return;
>> + }
>> +
>> + if (IS_ERR(clk)) {
>> + pr_err("%s: register clk: %s, type: %u failed!\n",
>> + __func__, list->name, list->type);
>> + return;
>> + }
>> +
>> + intel_clk_add_lookup(ctx, clk, list->id);
>> + }
>> +}
>> +
>> +struct intel_clk_provider * __init
>> +intel_clk_init(struct device_node *np, struct regmap *map, unsigned int nr_clks)
>> +{
>> + struct intel_clk_provider *ctx;
>> + struct clk **clks;
>> +
>> + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
>> + if (!ctx)
>> + return ERR_PTR(-ENOMEM);
>> +
>> + clks = kcalloc(nr_clks, sizeof(*clks), GFP_KERNEL);
>> + if (!clks) {
>> + kfree(ctx);
>> + return ERR_PTR(-ENOMEM);
>> + }
>> +
>> + memset_p((void **)clks, ERR_PTR(-ENOENT), nr_clks);
>> + ctx->map = map;
>> + ctx->clk_data.clks = clks;
>> + ctx->clk_data.clk_num = nr_clks;
>> + ctx->np = np;
>> +
>> + return ctx;
>> +}
>> +
>> +void __init intel_clk_register_osc(struct intel_clk_provider *ctx,
>> + struct intel_osc_clk *osc,
>> + unsigned int nr_clks)
>> +{
>> + u32 freq;
>> + struct clk *clk;
>> + int idx;
>> +
>> + for (idx = 0; idx < nr_clks; idx++, osc++) {
>> + if (!osc->dt_freq ||
>> + of_property_read_u32(ctx->np, osc->dt_freq, &freq))
>> + freq = osc->def_rate;
>> +
>> + clk = clk_register_fixed_rate(NULL, osc->name, NULL, 0, freq);
> Should come from DT itself.
Yes. It can be defined as fixed-clock node in device tree.
Do you mean it should be defined in device tree and driver reference it
via device tree?

>> + iS_ERR(clk)) {
>> + pr_err("%s: Failed to register clock: %s\n",
>> + __func__, osc->name);
>> + return;
>> + }
>> +
>> + intel_clk_add_lookup(ctx, clk, osc->id);
>> + }
>> +}
>> diff --git a/drivers/clk/intel/clk-cgu.h b/drivers/clk/intel/clk-cgu.h
>> new file mode 100644
>> index 000000000000..6dc4e45fc499
>> --- /dev/null
>> +++ b/drivers/clk/intel/clk-cgu.h
>> @@ -0,0 +1,259 @@
>> +/* SPDX-License-Identifier: GPL-2.0 */
>> +/*
>> + * Copyright(c) 2018 Intel Corporation.
>> + * Zhu YiXin <[email protected]>
>> + */
>> +
>> +#ifndef __INTEL_CLK_H
>> +#define __INTEL_CLK_H
>> +
>> +#define PNAME(x) static const char *const x[] __initconst
>> +
>> +struct intel_clk_mux {
>> + struct clk_hw hw;
>> + struct regmap *map;
>> + unsigned int reg;
>> + u8 shift;
>> + u8 width;
>> + unsigned long flags;
>> +};
>> +
>> +struct intel_clk_divider {
>> + struct clk_hw hw;
>> + struct regmap *map;
>> + unsigned int reg;
>> + u8 shift;
>> + u8 width;
>> + unsigned long flags;
>> + const struct clk_div_table *table;
>> +};
>> +
>> +struct intel_clk_gate {
>> + struct clk_hw hw;
>> + struct regmap *map;
>> + unsigned int reg;
>> + u8 shift;
>> + unsigned long flags;
>> +};
>> +
>> +enum intel_clk_type {
>> + intel_clk_fixed,
>> + intel_clk_mux,
>> + intel_clk_divider,
>> + intel_clk_fixed_factor,
>> + intel_clk_gate,
>> +};
>> +
>> +/**
>> + * struct intel_clk_provider
>> + * @map: regmap type base address for register.
>> + * @np: device node
>> + * @clk_data: array of hw clocks and clk number.
>> + */
>> +struct intel_clk_provider {
>> + struct regmap *map;
>> + struct device_node *np;
>> + struct clk_onecell_data clk_data;
> Please register clk_hw pointers instead of clk pointers with the of
> provider APIs.
Sorry.  I'm not sure I understand you correctly.
If only registering clk_hw pointer,  not registering of_provider API, then
how to reference it in the user drivers ?
Could you please give me more hints ?

>> +};
>> +
>> +/**
>> + * struct intel_pll_clk
>> + * @id: plaform specific id of the clock.
>> + * @name: name of this pll clock.
>> + * @parent_names: name of the parent clock.
>> + * @num_parents: number of parents.
>> + * @flags: optional flags for basic clock.
>> + * @type: platform type of pll.
>> + * @reg: offset of the register.
>> + * @mult: init value of mulitplier.
>> + * @div: init value of divider.
>> + * @frac: init value of fraction.
>> + * @rate_table: table of pll clock rate.
> Please drop the full-stop on kernel doc one-liners like this.
Will update it.

>
>> + */
>> +struct intel_pll_clk {
>> + unsigned int id;
>> + const char *name;
>> + const char *const *parent_names;
>> + u8 num_parents;
> Can the PLL have multiple parents?
Yes. But not in this platform.
The define here make it easy to expand to support new platform.

>> + unsigned long flags;
>> + enum intel_pll_type type;
>> + int reg;
>> + unsigned int mult;
>> + unsigned int div;
>> + unsigned int frac;
>> + const struct intel_pll_rate_table *rate_table;
>> +};
>> +
>> +#define INTEL_PLL(_id, _type, _name, _pnames, _flags, \
>> + _reg, _rtable, _mult, _div, _frac) \
>> + { \
>> + .id = _id, \
>> + .type = _type, \
>> + .name = _name, \
>> + .parent_names = _pnames, \
>> + .num_parents = ARRAY_SIZE(_pnames), \
>> + .flags = _flags, \
>> + .reg = _reg, \
>> + .rate_table = _rtable, \
>> + .mult = _mult, \
>> + .div = _div, \
>> + .frac = _frac \
>> + }
>> +
>> +/**
>> + * struct intel_osc_clk
>> + * @id: platform specific id of the clock.
>> + * @name: name of the osc clock.
>> + * @dt_freq: frequency node name in device tree.
>> + * @def_rate: default rate of the osc clock.
>> + * @flags: optional flags for basic clock.
> There aren't flags though. I'm very confused by this kernel-doc too.
> Looks like something that should be done with a fixed rate clk in DT.
Will remove the flags comments.

>> + */
>> +struct intel_osc_clk {
>> + unsigned int id;
>> + const char *name;
>> + const char *dt_freq;
>> + const u32 def_rate;
>> +};
>> +
>> +#define INTEL_OSC(_id, _name, _freq, _rate) \
>> + { \
>> + .id = _id, \
>> + .name = _name, \
>> + .dt_freq = _freq, \
>> + .def_rate = _rate, \
>> + }
>> +
>> +struct intel_clk_branch {
> Seems to be more like intel_clk instead of intel_clk_branch because it
> does lots of stuff.
Will update.

>> + unsigned int id;
>> + enum intel_clk_type type;
>> + const char *name;
>> + const char *const *parent_names;
>> + u8 num_parents;
>> + unsigned long flags;
>> + unsigned int mux_off;
>> + u8 mux_shift;
>> + u8 mux_width;
>> + unsigned long mux_flags;
>> + unsigned int mux_val;
>> + unsigned int div_off;
>> + u8 div_shift;
>> + u8 div_width;
>> + unsigned long div_flags;
>> + unsigned int div_val;
>> + const struct clk_div_table *div_table;
>> + unsigned int gate_off;
>> + u8 gate_shift;
>> + unsigned long gate_flags;
>> + unsigned int gate_val;
>> + unsigned int mult;
>> + unsigned int div;
>> +};
>> +
>> +/* clock flags definition */
>> +#define CLOCK_FLAG_VAL_INIT BIT(16)
>> +#define GATE_CLK_HW BIT(17)
>> +#define GATE_CLK_SW BIT(18)
>> +#define GATE_CLK_VT BIT(19)
> What does VT mean? Virtual?
Yes. VT means virtual here.
Will change to GATE_CLK_VIRT.

>> +
>> +#define INTEL_MUX(_id, _name, _pname, _f, _reg, \
>> + _shift, _width, _cf, _v) \
>> + { \
>> + .id = _id, \
>> + .type = intel_clk_mux, \
>> + .name = _name, \
>> + .parent_names = _pname, \
>> + .num_parents = ARRAY_SIZE(_pname), \
>> + .flags = _f, \
>> + .mux_off = _reg, \
>> + .mux_shift = _shift, \
>> + .mux_width = _width, \
>> + .mux_flags = _cf, \
>> + .mux_val = _v, \
>> + }
>> +
>> +#define INTEL_DIV(_id, _name, _pname, _f, _reg, \
>> + _shift, _width, _cf, _v, _dtable) \
>> + { \
>> + .id = _id, \
>> + .type = intel_clk_divider, \
>> + .name = _name, \
>> + .parent_names = (const char *[]) { _pname }, \
>> + .num_parents = 1, \
>> + .flags = _f, \
>> + .div_off = _reg, \
>> + .div_shift = _shift, \
>> + .div_width = _width, \
>> + .div_flags = _cf, \
>> + .div_val = _v, \
>> + .div_table = _dtable, \
>> + }
>> +
>> +#define INTEL_GATE(_id, _name, _pname, _f, _reg, \
>> + _shift, _cf, _v) \
>> + { \
>> + .id = _id, \
>> + .type = intel_clk_gate, \
>> + .name = _name, \
>> + .parent_names = (const char *[]) { _pname }, \
>> + .num_parents = !_pname ? 0 : 1, \
>> + .flags = _f, \
>> + .gate_off = _reg, \
>> + .gate_shift = _shift, \
>> + .gate_flags = _cf, \
>> + .gate_val = _v, \
>> + }
>> +
>> +#define INTEL_FIXED(_id, _name, _pname, _f, _reg, \
>> + _shift, _width, _cf, _freq, _v) \
>> + { \
>> + .id = _id, \
>> + .type = intel_clk_fixed, \
>> + .name = _name, \
>> + .parent_names = (const char *[]) { _pname }, \
>> + .num_parents = !_pname ? 0 : 1, \
>> + .flags = _f, \
>> + .div_off = _reg, \
>> + .div_shift = _shift, \
>> + .div_width = _width, \
>> + .div_flags = _cf, \
>> + .div_val = _v, \
>> + .mux_flags = _freq, \
>> + }
>> +
>> +#define INTEL_FIXED_FACTOR(_id, _name, _pname, _f, _reg, \
>> + _shift, _width, _cf, _v, _m, _d) \
>> + { \
>> + .id = _id, \
>> + .type = intel_clk_fixed_factor, \
>> + .name = _name, \
>> + .parent_names = (const char *[]) { _pname }, \
>> + .num_parents = 1, \
>> + .flags = _f, \
>> + .div_off = _reg, \
>> + .div_shift = _shift, \
>> + .div_width = _width, \
>> + .div_flags = _cf, \
>> + .div_val = _v, \
>> + .mult = _m, \
>> + .div = _d, \
>> + }
>> +
>> +void intel_set_clk_val(struct regmap *map, u32 reg, u8 shift,
>> + u8 width, u32 set_val);
>> +u32 intel_get_clk_val(struct regmap *map, u32 reg, u8 shift, u8 width);
>> +void intel_clk_add_lookup(struct intel_clk_provider *ctx,
>> + struct clk *clk, unsigned int id);
>> +void __init intel_clk_of_add_provider(struct device_node *np,
>> + struct intel_clk_provider *ctx);
>> +struct intel_clk_provider * __init
>> +intel_clk_init(struct device_node *np, struct regmap *map,
>> + unsigned int nr_clks);
>> +void __init intel_clk_register_osc(struct intel_clk_provider *ctx,
>> + struct intel_osc_clk *osc,
>> + unsigned int nr_clks);
> Remove __init from headers files. It does nothing.
Will remove it.

>
>> +void intel_clk_register_branches(struct intel_clk_provider *ctx,
>> + struct intel_clk_branch *list,
>> + unsigned int nr_clk);
>> +void intel_clk_register_plls(struct intel_clk_provider *ctx,
>> + struct intel_pll_clk *list, unsigned int nr_clk);
>> +#endif /* __INTEL_CLK_H */
>> diff --git a/drivers/clk/intel/clk-grx500.c b/drivers/clk/intel/clk-grx500.c
>> new file mode 100644
>> index 000000000000..5c2546f82579
>> --- /dev/null
>> +++ b/drivers/clk/intel/clk-grx500.c
>> @@ -0,0 +1,168 @@
>> +// SPDX-License-Identifier: GPL-2.0
>> +/*
>> + * Copyright (C) 2018 Intel Corporation.
>> + * Zhu YiXin <[email protected]>
>> + */
>> +
>> +#include <linux/clk-provider.h>
>> +#include <linux/mfd/syscon.h>
>> +#include <linux/of.h>
>> +#include <linux/of_address.h>
>> +#include <linux/regmap.h>
>> +#include <linux/spinlock.h>
> Used?
Will remove it

>> +#include <dt-bindings/clock/intel,grx500-clk.h>
>> +
>> +#include "clk-cgu-pll.h"
>> +#include "clk-cgu.h"
>> +
>> +#define PLL_DIV_WIDTH 4
>> +
>> +/* Gate1 clock shift */
>> +#define G_VCODEC_SHIFT 2
>> +#define G_DMA0_SHIFT 5
>> +#define G_USB0_SHIFT 6
>> +#define G_SPI1_SHIFT 7
>> +#define G_SPI0_SHIFT 8
>> +#define G_CBM_SHIFT 9
>> +#define G_EBU_SHIFT 10
>> +#define G_SSO_SHIFT 11
>> +#define G_GPTC0_SHIFT 12
>> +#define G_GPTC1_SHIFT 13
>> +#define G_GPTC2_SHIFT 14
>> +#define G_UART_SHIFT 17
>> +#define G_CPYTO_SHIFT 20
>> +#define G_SECPT_SHIFT 21
>> +#define G_TOE_SHIFT 22
>> +#define G_MPE_SHIFT 23
>> +#define G_TDM_SHIFT 25
>> +#define G_PAE_SHIFT 26
>> +#define G_USB1_SHIFT 27
>> +#define G_SWITCH_SHIFT 28
>> +
>> +/* Gate2 clock shift */
>> +#define G_PCIE0_SHIFT 1
>> +#define G_PCIE1_SHIFT 17
>> +#define G_PCIE2_SHIFT 25
>> +
>> +/* Register definition */
>> +#define GRX500_PLL0A_CFG0 0x0004
>> +#define GRX500_PLL0A_CFG1 0x0008
>> +#define GRX500_PLL0B_CFG0 0x0034
>> +#define GRX500_PLL0B_CFG1 0x0038
>> +#define GRX500_LCPLL_CFG0 0x0094
>> +#define GRX500_LCPLL_CFG1 0x0098
>> +#define GRX500_IF_CLK 0x00c4
>> +#define GRX500_CLK_GSR1 0x0120
>> +#define GRX500_CLK_GSR2 0x0130
>> +
>> +static const struct clk_div_table pll_div[] = {
>> + {1, 2},
> Please write it like
>
> { 1, 2 },
>
> instead.
Will update.

>> + {2, 3},
>> + {3, 4},
>> + {4, 5},
>> + {5, 6},
>> + {6, 8},
>> + {7, 10},
>> + {8, 12},
>> + {9, 16},
>> + {10, 20},
>> + {11, 24},
>> + {12, 32},
>> + {13, 40},
>> + {14, 48},
>> + {15, 64}
>> +};
>> +
>> +enum grx500_plls {
>> + pll0a, pll0b, pll3,
>> +};
> What's the point of the enum?
Will remove it.

>> +
>> +PNAME(pll_p) = { "osc" };
>> +PNAME(cpu_p) = { "cpu0", "cpu1" };
>> +
>> +static struct intel_osc_clk grx500_osc_clks[] __initdata = {
>> + INTEL_OSC(CLK_OSC, "osc", "intel,osc-frequency", 40000000),
>> +};
>> +
>> +static struct intel_pll_clk grx500_pll_clks[] __initdata = {
>> + [pll0a] = INTEL_PLL(CLK_PLL0A, pll_grx500, "pll0a",
>> + pll_p, 0, GRX500_PLL0A_CFG0, NULL, 0, 0, 0),
>> + [pll0b] = INTEL_PLL(CLK_PLL0B, pll_grx500, "pll0b",
>> + pll_p, 0, GRX500_PLL0B_CFG0, NULL, 0, 0, 0),
>> + [pll3] = INTEL_PLL(CLK_PLL3, pll_grx500, "pll3",
>> + pll_p, 0, GRX500_LCPLL_CFG0, NULL, 0, 0, 0),
>> +};
>> +
>> +static struct intel_clk_branch grx500_branch_clks[] __initdata = {
>> + INTEL_DIV(CLK_CBM, "cbm", "pll0a", 0, GRX500_PLL0A_CFG1,
>> + 0, PLL_DIV_WIDTH, 0, 0, pll_div),
>> + INTEL_DIV(CLK_NGI, "ngi", "pll0a", 0, GRX500_PLL0A_CFG1,
>> + 4, PLL_DIV_WIDTH, 0, 0, pll_div),
>> + INTEL_DIV(CLK_SSX4, "ssx4", "pll0a", 0, GRX500_PLL0A_CFG1,
>> + 8, PLL_DIV_WIDTH, 0, 0, pll_div),
>> + INTEL_DIV(CLK_CPU0, "cpu0", "pll0a", 0, GRX500_PLL0A_CFG1,
>> + 12, PLL_DIV_WIDTH, 0, 0, pll_div),
>> + INTEL_DIV(CLK_PAE, "pae", "pll0b", 0, GRX500_PLL0B_CFG1,
>> + 0, PLL_DIV_WIDTH, 0, 0, pll_div),
>> + INTEL_DIV(CLK_GSWIP, "gswip", "pll0b", 0, GRX500_PLL0B_CFG1,
>> + 4, PLL_DIV_WIDTH, 0, 0, pll_div),
>> + INTEL_DIV(CLK_DDR, "ddr", "pll0b", 0, GRX500_PLL0B_CFG1,
>> + 8, PLL_DIV_WIDTH, 0, 0, pll_div),
>> + INTEL_DIV(CLK_CPU1, "cpu1", "pll0b", 0, GRX500_PLL0B_CFG1,
>> + 12, PLL_DIV_WIDTH, 0, 0, pll_div),
>> + INTEL_MUX(CLK_CPU, "cpu", cpu_p, CLK_SET_RATE_PARENT,
>> + GRX500_PLL0A_CFG1, 29, 1, 0, 0),
>> + INTEL_GATE(GCLK_DMA0, "g_dma0", NULL, 0, GRX500_CLK_GSR1,
>> + G_DMA0_SHIFT, GATE_CLK_HW, 0),
>> + INTEL_GATE(GCLK_USB0, "g_usb0", NULL, 0, GRX500_CLK_GSR1,
>> + G_USB0_SHIFT, GATE_CLK_HW, 0),
>> + INTEL_GATE(GCLK_GPTC0, "g_gptc0", NULL, 0, GRX500_CLK_GSR1,
>> + G_GPTC0_SHIFT, GATE_CLK_HW, 0),
>> + INTEL_GATE(GCLK_GPTC1, "g_gptc1", NULL, 0, GRX500_CLK_GSR1,
>> + G_GPTC1_SHIFT, GATE_CLK_HW, 0),
>> + INTEL_GATE(GCLK_GPTC2, "g_gptc2", NULL, 0, GRX500_CLK_GSR1,
>> + G_GPTC2_SHIFT, GATE_CLK_HW, 0),
>> + INTEL_GATE(GCLK_UART, "g_uart", NULL, 0, GRX500_CLK_GSR1,
>> + G_UART_SHIFT, GATE_CLK_HW, 0),
>> + INTEL_GATE(GCLK_PCIE0, "g_pcie0", NULL, 0, GRX500_CLK_GSR2,
>> + G_PCIE0_SHIFT, GATE_CLK_HW, 0),
>> + INTEL_GATE(GCLK_PCIE1, "g_pcie1", NULL, 0, GRX500_CLK_GSR2,
>> + G_PCIE1_SHIFT, GATE_CLK_HW, 0),
>> + INTEL_GATE(GCLK_PCIE2, "g_pcie2", NULL, 0, GRX500_CLK_GSR2,
>> + G_PCIE2_SHIFT, GATE_CLK_HW, 0),
>> + INTEL_GATE(GCLK_I2C, "g_i2c", NULL, 0, 0, 0, GATE_CLK_VT, 0),
>> + INTEL_FIXED(CLK_VOICE, "voice", NULL, 0, GRX500_IF_CLK, 14, 2,
>> + CLOCK_FLAG_VAL_INIT, 8192000, 2),
>> + INTEL_FIXED_FACTOR(CLK_DDRPHY, "ddrphy", "ddr", 0, 0, 0,
>> + 0, 0, 0, 2, 1),
>> + INTEL_FIXED_FACTOR(CLK_PCIE, "pcie", "pll3", 0, 0, 0,
>> + 0, 0, 0, 1, 40),
>> +};
>> +
>> +static void __init grx500_clk_init(struct device_node *np)
>> +{
>> + struct intel_clk_provider *ctx;
>> + struct regmap *map;
>> +
>> + map = syscon_node_to_regmap(np);
>> + if (IS_ERR(map))
>> + return;
>> +
>> + ctx = intel_clk_init(np, map, CLK_NR_CLKS);
>> + if (IS_ERR(ctx)) {
>> + regmap_exit(map);
>> + return;
>> + }
>> +
>> + intel_clk_register_osc(ctx, grx500_osc_clks,
>> + ARRAY_SIZE(grx500_osc_clks));
>> + intel_clk_register_plls(ctx, grx500_pll_clks,
>> + ARRAY_SIZE(grx500_pll_clks));
>> + intel_clk_register_branches(ctx, grx500_branch_clks,
>> + ARRAY_SIZE(grx500_branch_clks));
>> + of_clk_add_provider(np, of_clk_src_onecell_get, &ctx->clk_data);
>> +
>> + pr_debug("%s clk init done!\n", __func__);
> Yay!!!
>
>> +}
>> +
>> +CLK_OF_DECLARE(intel_grx500_cgu, "intel,grx500-cgu", grx500_clk_init);
> Any reason a platform driver can't be used instead of CLK_OF_DECLARE()?
It provides CPU clock which is used in early boot stage.


2018-08-08 14:55:29

by Rob Herring

[permalink] [raw]
Subject: Re: [PATCH v2 03/18] dt-bindings: clk: Add documentation of grx500 clock controller

On Tue, Aug 7, 2018 at 9:08 PM yixin zhu <[email protected]> wrote:
>
>
>
> On 8/6/2018 11:18 PM, Rob Herring wrote:
> > On Thu, Aug 2, 2018 at 9:03 PM Songjun Wu <[email protected]> wrote:
> >> From: Yixin Zhu <[email protected]>
> >>
> >> This patch adds binding documentation for grx500 clock controller.
> >>
> >> Signed-off-by: YiXin Zhu <[email protected]>
> >> Signed-off-by: Songjun Wu <[email protected]>
> >> ---
> >>
> >> Changes in v2:
> >> - Rewrite clock driver's dt-binding document according to Rob Herring's
> >> comments.
> >> - Simplify device tree docoment, remove some clock description.
> >>
> >> .../devicetree/bindings/clock/intel,grx500-clk.txt | 39 ++++++++++++++++++++++
> > Please match the compatible string: intel,grx500-cgu.txt
> Will update to use same name.
>
> >
> >> 1 file changed, 39 insertions(+)
> >> create mode 100644 Documentation/devicetree/bindings/clock/intel,grx500-clk.txt
> >>
> >> diff --git a/Documentation/devicetree/bindings/clock/intel,grx500-clk.txt b/Documentation/devicetree/bindings/clock/intel,grx500-clk.txt
> >> new file mode 100644
> >> index 000000000000..e54e1dad9196
> >> --- /dev/null
> >> +++ b/Documentation/devicetree/bindings/clock/intel,grx500-clk.txt
> >> @@ -0,0 +1,39 @@
> >> +Device Tree Clock bindings for grx500 PLL controller.
> >> +
> >> +This binding uses the common clock binding:
> >> + Documentation/devicetree/bindings/clock/clock-bindings.txt
> >> +
> >> +The grx500 clock controller supplies clock to various controllers within the
> >> +SoC.
> >> +
> >> +Required properties for clock node
> >> +- compatible: Should be "intel,grx500-cgu".
> >> +- reg: physical base address of the controller and length of memory range.
> >> +- #clock-cells: should be 1.
> >> +
> >> +Optional Propteries:
> >> +- intel,osc-frequency: frequency of the osc clock.
> >> +if missing, driver will use clock rate defined in the driver.
> > This should use a fixed-clock node instead.
> Yes, This is a fixed clock node registered in driver code.
> The frequency of the fixed clock is designed to be overwritten by device
> tree in case some one verify
> clock driver in the emulation platform or in some cases frequency other
> than driver defined one is preferred.

Emulation platforms often need several hacks that shouldn't be upstream...

> These kinds of cases are very rare. But I feel it would be better to
> have a way to use customized frequency.
> The frequency defined in device tree will overwritten driver defined
> frequency before registering fixed-clock node.

I don't think you understand what I meant. Add a DT node with
"fixed-clock" compatible and have this node refer to it with "clocks"
property. The frequency is still in DT, but uses the clock binding.

Rob

2018-08-10 08:14:43

by Songjun Wu

[permalink] [raw]
Subject: Re: [PATCH v2 08/18] serial: intel: Get serial id from dts



On 8/8/2018 4:33 PM, Geert Uytterhoeven wrote:
> Hi Songjun,
>
> On Wed, Aug 8, 2018 at 6:05 AM Wu, Songjun <[email protected]> wrote:
>> On 8/7/2018 3:33 PM, Geert Uytterhoeven wrote:
>>> On Fri, Aug 3, 2018 at 5:04 AM Songjun Wu <[email protected]> wrote:
>>>> Get serial id from dts.
>>>>
>>>> "#ifdef CONFIG_LANTIQ" preprocessor is used because LTQ_EARLY_ASC
>>>> macro is defined in lantiq_soc.h.
>>>> lantiq_soc.h is in arch path for legacy product support.
>>>>
>>>> arch/mips/include/asm/mach-lantiq/xway/lantiq_soc.h
>>>>
>>>> If "#ifdef preprocessor" is changed to
>>>> "if (IS_ENABLED(CONFIG_LANTIQ))", when CONFIG_LANTIQ is not enabled,
>>>> code using LTQ_EARLY_ASC is compiled.
>>>> Compilation will fail for no LTQ_EARLY_ASC defined.
>>>>
>>>> Signed-off-by: Songjun Wu <[email protected]>
>>> Thanks for your patch!
>>>
>>>> @@ -699,9 +700,19 @@ lqasc_probe(struct platform_device *pdev)
>>>> return -ENODEV;
>>>> }
>>>>
>>>> - /* check if this is the console port */
>>>> - if (mmres->start != CPHYSADDR(LTQ_EARLY_ASC))
>>>> - line = 1;
>>>> + /* get serial id */
>>>> + line = of_alias_get_id(node, "serial");
>>>> + if (line < 0) {
>>>> +#ifdef CONFIG_LANTIQ
>>>> + if (mmres->start == CPHYSADDR(LTQ_EARLY_ASC))
>>>> + line = 0;
>>>> + else
>>>> + line = 1;
>>>> +#else
>>>> + dev_err(&pdev->dev, "failed to get alias id, errno %d\n", line);
>>>> + return line;
>>> Please note that not providing a fallback here makes life harder when using
>>> DT overlays.
>>> See the description of commit 7678f4c20fa7670f ("serial: sh-sci: Add support
>>> for dynamic instances") for background info.
>> Thanks for your comment.
>> The logic in commit 7678f4c20fa7670f is not suitable here.
>> We need to know which serial instance is used for console.
>> We cannot use dynamic serial instance here.
> Why does the driver need to use which serial instance is used for the console?
> Hardcoding that is not an option, as the board DTS may specify the console using
> chosen/stdout-path.
In legacy platform in open source, it only defined asc1 in dts.
There's no asc0 in legacy dts.While in the new platform, asc0
is defined in dts. There's no asc1 in new platform dts.
To avoid hard code in driver, alias serial0 is used to unified
driver code. Actually only one serial is supported in SoC.

aliases {
        serial0 = &asc0;
};

chosen {
    bootargs = "earlycon  clk_ignore_unused";
    stdout-path = "serial0";
};



2018-08-13 18:14:56

by Rob Herring (Arm)

[permalink] [raw]
Subject: Re: [PATCH v2 18/18] dt-bindings: serial: lantiq: Add optional properties for CCF

On Fri, Aug 03, 2018 at 11:02:37AM +0800, Songjun Wu wrote:
> Clocks and clock-names are updated in device tree binding.
>
> Signed-off-by: Songjun Wu <[email protected]>
> ---
>
> Changes in v2: None
>
> Documentation/devicetree/bindings/serial/lantiq_asc.txt | 15 +++++++++++++++
> 1 file changed, 15 insertions(+)

Reviewed-by: Rob Herring <[email protected]>

2018-08-27 19:11:24

by Stephen Boyd

[permalink] [raw]
Subject: Re: [PATCH v2 02/18] clk: intel: Add clock driver for Intel MIPS SoCs

Quoting yixin zhu (2018-08-08 01:52:20)
> On 8/8/2018 1:50 PM, Stephen Boyd wrote:
> > Quoting Songjun Wu (2018-08-02 20:02:21)
> >> + struct clk *clk;
> >> + int idx;
> >> +
> >> + for (idx = 0; idx < nr_clks; idx++, osc++) {
> >> + if (!osc->dt_freq ||
> >> + of_property_read_u32(ctx->np, osc->dt_freq, &freq))
> >> + freq = osc->def_rate;
> >> +
> >> + clk = clk_register_fixed_rate(NULL, osc->name, NULL, 0, freq);
> > Should come from DT itself.
> Yes. It can be defined as fixed-clock node in device tree.
> Do you mean it should be defined in device tree and driver reference it
> via device tree?

Yes the oscillator should be in DT and then the DT node here can call
clk_get() or just hardcode the parent name to be what it knows it is.
Eventually we'd like to be able to move away from string names for
hierarchy descriptions but that's far off. To get there, we would need
DT nodes for clock controllers to indicate their clk parents with the
clocks and clock-names properties. So for the oscillator, DT would
define it and then the driver would eventually have a way to specify
that some parent is index 5 or clock name "foo" and then the clk core
could figure out the linkage. I haven't written that code yet, but I'll
probably do it soon if nobody beats me to it.

> >> +/**
> >> + * struct intel_clk_provider
> >> + * @map: regmap type base address for register.
> >> + * @np: device node
> >> + * @clk_data: array of hw clocks and clk number.
> >> + */
> >> +struct intel_clk_provider {
> >> + struct regmap *map;
> >> + struct device_node *np;
> >> + struct clk_onecell_data clk_data;
> > Please register clk_hw pointers instead of clk pointers with the of
> > provider APIs.
> Sorry.  I'm not sure I understand you correctly.
> If only registering clk_hw pointer,  not registering of_provider API, then
> how to reference it in the user drivers ?
> Could you please give me more hints ?

Clk provider drivers shouldn't be using clk pointers directly. Usually
when that happens something is wrong. So new clk drivers should register
clk_hw pointers and pretty much only deal with clk_hw pointers instead
of struct clk pointers. You still register an of_provider, but that
provider hands out clk_hw pointers so that clk provider drivers aren't
tempted to use struct clk pointers.

>
>
> >
> >> + */
> >> +struct intel_pll_clk {
> >> + unsigned int id;
> >> + const char *name;
> >> + const char *const *parent_names;
> >> + u8 num_parents;
> > Can the PLL have multiple parents?
> Yes. But not in this platform.
> The define here make it easy to expand to support new platform.
>

Ok, so it has a mux inside.

>
> >> + unsigned int id;
> >> + enum intel_clk_type type;
> >> + const char *name;
> >> + const char *const *parent_names;
> >> + u8 num_parents;
> >> + unsigned long flags;
> >> + unsigned int mux_off;
> >> + u8 mux_shift;
> >> + u8 mux_width;
> >> + unsigned long mux_flags;
> >> + unsigned int mux_val;
> >> + unsigned int div_off;
> >> + u8 div_shift;
> >> + u8 div_width;
> >> + unsigned long div_flags;
> >> + unsigned int div_val;
> >> + const struct clk_div_table *div_table;
> >> + unsigned int gate_off;
> >> + u8 gate_shift;
> >> + unsigned long gate_flags;
> >> + unsigned int gate_val;
> >> + unsigned int mult;
> >> + unsigned int div;
> >> +};
> >> +
> >> +/* clock flags definition */
> >> +#define CLOCK_FLAG_VAL_INIT BIT(16)
> >> +#define GATE_CLK_HW BIT(17)
> >> +#define GATE_CLK_SW BIT(18)
> >> +#define GATE_CLK_VT BIT(19)
> > What does VT mean? Virtual?
> Yes. VT means virtual here.
> Will change to GATE_CLK_VIRT.
>

Is it a hardware concept? Or virtualization with hypervisor?

> >
> >> +}
> >> +
> >> +CLK_OF_DECLARE(intel_grx500_cgu, "intel,grx500-cgu", grx500_clk_init);
> > Any reason a platform driver can't be used instead of CLK_OF_DECLARE()?
> It provides CPU clock which is used in early boot stage.
>

Ok. What is the CPU clock doing in early boot stage? Some sort of timer
frequency? If the driver can be split into two pieces, one to handle the
really early stuff that must be in place to get timers up and running
and the other to register the rest of the clks that aren't critical from
a regular platform driver it would be good. That's preferred model if
something is super critical.


2018-08-29 06:57:48

by Zhu, Yi Xin

[permalink] [raw]
Subject: Re: [PATCH v2 02/18] clk: intel: Add clock driver for Intel MIPS SoCs


On 8/28/2018 3:09 AM, Stephen Boyd wrote:
> Quoting yixin zhu (2018-08-08 01:52:20)
>> On 8/8/2018 1:50 PM, Stephen Boyd wrote:
>>> Quoting Songjun Wu (2018-08-02 20:02:21)
>>>> + struct clk *clk;
>>>> + int idx;
>>>> +
>>>> + for (idx = 0; idx < nr_clks; idx++, osc++) {
>>>> + if (!osc->dt_freq ||
>>>> + of_property_read_u32(ctx->np, osc->dt_freq, &freq))
>>>> + freq = osc->def_rate;
>>>> +
>>>> + clk = clk_register_fixed_rate(NULL, osc->name, NULL, 0, freq);
>>> Should come from DT itself.
>> Yes. It can be defined as fixed-clock node in device tree.
>> Do you mean it should be defined in device tree and driver reference it
>> via device tree?
> Yes the oscillator should be in DT and then the DT node here can call
> clk_get() or just hardcode the parent name to be what it knows it is.
> Eventually we'd like to be able to move away from string names for
> hierarchy descriptions but that's far off. To get there, we would need
> DT nodes for clock controllers to indicate their clk parents with the
> clocks and clock-names properties. So for the oscillator, DT would
> define it and then the driver would eventually have a way to specify
> that some parent is index 5 or clock name "foo" and then the clk core
> could figure out the linkage. I haven't written that code yet, but I'll
> probably do it soon if nobody beats me to it.

Thanks.  Will update.


>
>>>> +/**
>>>> + * struct intel_clk_provider
>>>> + * @map: regmap type base address for register.
>>>> + * @np: device node
>>>> + * @clk_data: array of hw clocks and clk number.
>>>> + */
>>>> +struct intel_clk_provider {
>>>> + struct regmap *map;
>>>> + struct device_node *np;
>>>> + struct clk_onecell_data clk_data;
>>> Please register clk_hw pointers instead of clk pointers with the of
>>> provider APIs.
>> Sorry.  I'm not sure I understand you correctly.
>> If only registering clk_hw pointer,  not registering of_provider API, then
>> how to reference it in the user drivers ?
>> Could you please give me more hints ?
> Clk provider drivers shouldn't be using clk pointers directly. Usually
> when that happens something is wrong. So new clk drivers should register
> clk_hw pointers and pretty much only deal with clk_hw pointers instead
> of struct clk pointers. You still register an of_provider, but that
> provider hands out clk_hw pointers so that clk provider drivers aren't
> tempted to use struct clk pointers.

Understood.  Will update to use clk_hw_onecell_data and change the
registration accordingly.


>>
>>>> + */
>>>> +struct intel_pll_clk {
>>>> + unsigned int id;
>>>> + const char *name;
>>>> + const char *const *parent_names;
>>>> + u8 num_parents;
>>> Can the PLL have multiple parents?
>> Yes. But not in this platform.
>> The define here make it easy to expand to support new platform.
>>
> Ok, so it has a mux inside.
>
>>>> + unsigned int id;
>>>> + enum intel_clk_type type;
>>>> + const char *name;
>>>> + const char *const *parent_names;
>>>> + u8 num_parents;
>>>> + unsigned long flags;
>>>> + unsigned int mux_off;
>>>> + u8 mux_shift;
>>>> + u8 mux_width;
>>>> + unsigned long mux_flags;
>>>> + unsigned int mux_val;
>>>> + unsigned int div_off;
>>>> + u8 div_shift;
>>>> + u8 div_width;
>>>> + unsigned long div_flags;
>>>> + unsigned int div_val;
>>>> + const struct clk_div_table *div_table;
>>>> + unsigned int gate_off;
>>>> + u8 gate_shift;
>>>> + unsigned long gate_flags;
>>>> + unsigned int gate_val;
>>>> + unsigned int mult;
>>>> + unsigned int div;
>>>> +};
>>>> +
>>>> +/* clock flags definition */
>>>> +#define CLOCK_FLAG_VAL_INIT BIT(16)
>>>> +#define GATE_CLK_HW BIT(17)
>>>> +#define GATE_CLK_SW BIT(18)
>>>> +#define GATE_CLK_VT BIT(19)
>>> What does VT mean? Virtual?
>> Yes. VT means virtual here.
>> Will change to GATE_CLK_VIRT.
>>
> Is it a hardware concept? Or virtualization with hypervisor?

Some peripheral drivers want to use same code cross platforms.

But not all platforms provide HW gate clock.  So in this case, clock
driver creates

a virtual gate clock to make it work if no HW gate clock in the SoC.


>
>>>> +}
>>>> +
>>>> +CLK_OF_DECLARE(intel_grx500_cgu, "intel,grx500-cgu", grx500_clk_init);
>>> Any reason a platform driver can't be used instead of CLK_OF_DECLARE()?
>> It provides CPU clock which is used in early boot stage.
>>
> Ok. What is the CPU clock doing in early boot stage? Some sort of timer
> frequency? If the driver can be split into two pieces, one to handle the
> really early stuff that must be in place to get timers up and running
> and the other to register the rest of the clks that aren't critical from
> a regular platform driver it would be good. That's preferred model if
> something is super critical.

Yes, CPU clock is providing CPU frequency in the early boot stage.

Will put the non-critical clocks in the platform driver.


>

2018-08-29 10:35:50

by Zhu, Yi Xin

[permalink] [raw]
Subject: Re: [PATCH v2 02/18] clk: intel: Add clock driver for Intel MIPS SoCs

>>>> +}
>>>> +
>>>> +CLK_OF_DECLARE(intel_grx500_cgu, "intel,grx500-cgu", grx500_clk_init);
>>> Any reason a platform driver can't be used instead of CLK_OF_DECLARE()?
>> It provides CPU clock which is used in early boot stage.
>>
> Ok. What is the CPU clock doing in early boot stage? Some sort of timer
> frequency? If the driver can be split into two pieces, one to handle the
> really early stuff that must be in place to get timers up and running
> and the other to register the rest of the clks that aren't critical from
> a regular platform driver it would be good. That's preferred model if
> something is super critical.
>
Just to make sure my approach is same as you think.

In the driver, there's two clock registrations.

- One through CLK_OF_DECLARE for early stage clocks.

- The other via platform driver for the non-critical clocks.

In the device tree,  two clock device nodes are required.

e.g. device tree:

cgu: cgu@16200000 {
                compatible = "intel,grx500-clk", "syscon";
                reg = <0x16200000 0x200>;
                #clock-cells = <1>;
};

clk: clk {
                compatible = "intel,grx500-cgu";
                #clock-cells = <1>;
                intel,cgu-syscon = <&cgu>;
};

source code:

CLK_OF_DECLARE(intel_grx500_cgu, "intel,grx500-cgu", grx500_clk_init);

static const struct of_device_id of_intel_grx500_cgu_match[] = {
        { .compatible = "intel,grx500-clk" },
        {}
};

static struct platform_driver intel_grx500_clk_driver = {
        .probe  = intel_grx500_clk_probe,
        .driver = {
                .name = "grx500-cgu",
                .of_match_table = of_match_ptr(of_intel_grx500_cgu_match),
        },
};

static int __init intel_grx500_cgu_init(void)
{
        return platform_driver_register(&intel_grx500_clk_driver);
}
arch_initcall(intel_grx500_cgu_init);



2018-08-31 17:17:15

by Stephen Boyd

[permalink] [raw]
Subject: Re: [PATCH v2 02/18] clk: intel: Add clock driver for Intel MIPS SoCs

Quoting Zhu, Yi Xin (2018-08-29 03:34:26)
> >>>> +}
> >>>> +
> >>>> +CLK_OF_DECLARE(intel_grx500_cgu, "intel,grx500-cgu", grx500_clk_init);
> >>> Any reason a platform driver can't be used instead of CLK_OF_DECLARE()?
> >> It provides CPU clock which is used in early boot stage.
> >>
> > Ok. What is the CPU clock doing in early boot stage? Some sort of timer
> > frequency? If the driver can be split into two pieces, one to handle the
> > really early stuff that must be in place to get timers up and running
> > and the other to register the rest of the clks that aren't critical from
> > a regular platform driver it would be good. That's preferred model if
> > something is super critical.
> >
> Just to make sure my approach is same as you think.
>
> In the driver, there's two clock registrations.
>
> - One through CLK_OF_DECLARE for early stage clocks.
>
> - The other via platform driver for the non-critical clocks.

You can use the same DT node for both parts, no need to split the node
into two and use syscon here.


2018-08-31 17:17:23

by Stephen Boyd

[permalink] [raw]
Subject: Re: [PATCH v2 02/18] clk: intel: Add clock driver for Intel MIPS SoCs

Quoting Zhu, Yi Xin (2018-08-28 23:56:22)
>
> On 8/28/2018 3:09 AM, Stephen Boyd wrote:
> > Quoting yixin zhu (2018-08-08 01:52:20)
> >> On 8/8/2018 1:50 PM, Stephen Boyd wrote:
> >>>> +/* clock flags definition */
> >>>> +#define CLOCK_FLAG_VAL_INIT BIT(16)
> >>>> +#define GATE_CLK_HW BIT(17)
> >>>> +#define GATE_CLK_SW BIT(18)
> >>>> +#define GATE_CLK_VT BIT(19)
> >>> What does VT mean? Virtual?
> >> Yes. VT means virtual here.
> >> Will change to GATE_CLK_VIRT.
> >>
> > Is it a hardware concept? Or virtualization with hypervisor?
>
> Some peripheral drivers want to use same code cross platforms.
>
> But not all platforms provide HW gate clock.  So in this case, clock
> driver creates
>
> a virtual gate clock to make it work if no HW gate clock in the SoC.

That's not how things are supposed to work. If a clk isn't there in the
hardware we don't make them up in software so that the consumer software
drivers can keep requesting clks on different platforms. On a different
platform, the driver needs to know that the clks aren't there with a
different compatible string.

>
>
> >
> >>>> +}
> >>>> +
> >>>> +CLK_OF_DECLARE(intel_grx500_cgu, "intel,grx500-cgu", grx500_clk_init);
> >>> Any reason a platform driver can't be used instead of CLK_OF_DECLARE()?
> >> It provides CPU clock which is used in early boot stage.
> >>
> > Ok. What is the CPU clock doing in early boot stage? Some sort of timer
> > frequency? If the driver can be split into two pieces, one to handle the
> > really early stuff that must be in place to get timers up and running
> > and the other to register the rest of the clks that aren't critical from
> > a regular platform driver it would be good. That's preferred model if
> > something is super critical.
>
> Yes, CPU clock is providing CPU frequency in the early boot stage.
>
> Will put the non-critical clocks in the platform driver.
>
>

Sure the CPU clock is handling frequency, but does that matter for early
boot to get going? If timers aren't involved here then it doesn't sound
like this needs CLK_OF_DECLARE.


2018-09-03 10:49:05

by Zhu, Yi Xin

[permalink] [raw]
Subject: Re: [PATCH v2 02/18] clk: intel: Add clock driver for Intel MIPS SoCs


On 9/1/2018 1:10 AM, Stephen Boyd wrote:
> Quoting Zhu, Yi Xin (2018-08-28 23:56:22)
>> On 8/28/2018 3:09 AM, Stephen Boyd wrote:
>>> Quoting yixin zhu (2018-08-08 01:52:20)
>>>> On 8/8/2018 1:50 PM, Stephen Boyd wrote:
>>>>>> +/* clock flags definition */
>>>>>> +#define CLOCK_FLAG_VAL_INIT BIT(16)
>>>>>> +#define GATE_CLK_HW BIT(17)
>>>>>> +#define GATE_CLK_SW BIT(18)
>>>>>> +#define GATE_CLK_VT BIT(19)
>>>>> What does VT mean? Virtual?
>>>> Yes. VT means virtual here.
>>>> Will change to GATE_CLK_VIRT.
>>>>
>>> Is it a hardware concept? Or virtualization with hypervisor?
>> Some peripheral drivers want to use same code cross platforms.
>>
>> But not all platforms provide HW gate clock.  So in this case, clock
>> driver creates
>>
>> a virtual gate clock to make it work if no HW gate clock in the SoC.
> That's not how things are supposed to work. If a clk isn't there in the
> hardware we don't make them up in software so that the consumer software
> drivers can keep requesting clks on different platforms. On a different
> platform, the driver needs to know that the clks aren't there with a
> different compatible string.

OK. Will remove virtual gate clock.


>>
>>>>>> +}
>>>>>> +
>>>>>> +CLK_OF_DECLARE(intel_grx500_cgu, "intel,grx500-cgu", grx500_clk_init);
>>>>> Any reason a platform driver can't be used instead of CLK_OF_DECLARE()?
>>>> It provides CPU clock which is used in early boot stage.
>>>>
>>> Ok. What is the CPU clock doing in early boot stage? Some sort of timer
>>> frequency? If the driver can be split into two pieces, one to handle the
>>> really early stuff that must be in place to get timers up and running
>>> and the other to register the rest of the clks that aren't critical from
>>> a regular platform driver it would be good. That's preferred model if
>>> something is super critical.
>> Yes, CPU clock is providing CPU frequency in the early boot stage.
>>
>> Will put the non-critical clocks in the platform driver.
>>
>>
> Sure the CPU clock is handling frequency, but does that matter for early
> boot to get going? If timers aren't involved here then it doesn't sound
> like this needs CLK_OF_DECLARE.

Yes, timer is involved here.

CPU frequency get by early stage platform code used in clockevent
registration.


>

2018-09-03 10:53:47

by Zhu, Yi Xin

[permalink] [raw]
Subject: Re: [PATCH v2 02/18] clk: intel: Add clock driver for Intel MIPS SoCs


On 9/1/2018 1:13 AM, Stephen Boyd wrote:
> Quoting Zhu, Yi Xin (2018-08-29 03:34:26)
>>>>>> +}
>>>>>> +
>>>>>> +CLK_OF_DECLARE(intel_grx500_cgu, "intel,grx500-cgu", grx500_clk_init);
>>>>> Any reason a platform driver can't be used instead of CLK_OF_DECLARE()?
>>>> It provides CPU clock which is used in early boot stage.
>>>>
>>> Ok. What is the CPU clock doing in early boot stage? Some sort of timer
>>> frequency? If the driver can be split into two pieces, one to handle the
>>> really early stuff that must be in place to get timers up and running
>>> and the other to register the rest of the clks that aren't critical from
>>> a regular platform driver it would be good. That's preferred model if
>>> something is super critical.
>>>
>> Just to make sure my approach is same as you think.
>>
>> In the driver, there's two clock registrations.
>>
>> - One through CLK_OF_DECLARE for early stage clocks.
>>
>> - The other via platform driver for the non-critical clocks.
> You can use the same DT node for both parts, no need to split the node
> into two and use syscon here.
>
Thank you.

Will use CLK_OF_DECLARE_DRIVER to use the same DT node.