From e53b923187b1b07a709ca6e8d76a213b2d91f6e8 Mon Sep 17 00:00:00 2001 From: Lorenzo Pieralisi Date: Fri, 13 Jul 2012 15:55:52 +0100 Subject: drivers: bus: add ARM CCI support On ARM multi-cluster systems coherency between cores running on different clusters is managed by the cache-coherent interconnect (CCI). It allows broadcasting of TLB invalidates and memory barriers and it guarantees cache coherency at system level through snooping of slave interfaces connected to it. This patch enables the basic infrastructure required in Linux to handle and programme the CCI component. Non-local variables used by the CCI management functions called by power down function calls after disabling the cache must be flushed out to main memory in advance, otherwise incoherency of those values may occur if they are sitting in the cache of some other CPU when power down functions execute. Driver code ensures that relevant data structures are flushed from inner and outer caches after the driver probe is completed. CCI slave port resources are linked to set of CPUs through bus masters phandle properties that link the interface resources to masters node in the device tree. Documentation describing the CCI DT bindings is provided with the patch. Signed-off-by: Lorenzo Pieralisi Signed-off-by: Nicolas Pitre --- Documentation/devicetree/bindings/arm/cci.txt | 172 +++++++++++ drivers/bus/Kconfig | 7 + drivers/bus/Makefile | 2 + drivers/bus/arm-cci.c | 426 ++++++++++++++++++++++++++ include/linux/arm-cci.h | 61 ++++ 5 files changed, 668 insertions(+) create mode 100644 Documentation/devicetree/bindings/arm/cci.txt create mode 100644 drivers/bus/arm-cci.c create mode 100644 include/linux/arm-cci.h diff --git a/Documentation/devicetree/bindings/arm/cci.txt b/Documentation/devicetree/bindings/arm/cci.txt new file mode 100644 index 000000000000..92d36e2aa877 --- /dev/null +++ b/Documentation/devicetree/bindings/arm/cci.txt @@ -0,0 +1,172 @@ +======================================================= +ARM CCI cache coherent interconnect binding description +======================================================= + +ARM multi-cluster systems maintain intra-cluster coherency through a +cache coherent interconnect (CCI) that is capable of monitoring bus +transactions and manage coherency, TLB invalidations and memory barriers. + +It allows snooping and distributed virtual memory message broadcast across +clusters, through memory mapped interface, with a global control register +space and multiple sets of interface control registers, one per slave +interface. + +Bindings for the CCI node follow the ePAPR standard, available from: + +www.power.org/documentation/epapr-version-1-1/ + +with the addition of the bindings described in this document which are +specific to ARM. + +* CCI interconnect node + + Description: Describes a CCI cache coherent Interconnect component + + Node name must be "cci". + Node's parent must be the root node /, and the address space visible + through the CCI interconnect is the same as the one seen from the + root node (ie from CPUs perspective as per DT standard). + Every CCI node has to define the following properties: + + - compatible + Usage: required + Value type: + Definition: must be set to + "arm,cci-400" + + - reg + Usage: required + Value type: + Definition: A standard property. Specifies base physical + address of CCI control registers common to all + interfaces. + + - ranges: + Usage: required + Value type: + Definition: A standard property. Follow rules in the ePAPR for + hierarchical bus addressing. CCI interfaces + addresses refer to the parent node addressing + scheme to declare their register bases. + + CCI interconnect node can define the following child nodes: + + - CCI control interface nodes + + Node name must be "slave-if". + Parent node must be CCI interconnect node. + + A CCI control interface node must contain the following + properties: + + - compatible + Usage: required + Value type: + Definition: must be set to + "arm,cci-400-ctrl-if" + + - interface-type: + Usage: required + Value type: + Definition: must be set to one of {"ace", "ace-lite"} + depending on the interface type the node + represents. + + - reg: + Usage: required + Value type: + Definition: the base address and size of the + corresponding interface programming + registers. + +* CCI interconnect bus masters + + Description: masters in the device tree connected to a CCI port + (inclusive of CPUs and their cpu nodes). + + A CCI interconnect bus master node must contain the following + properties: + + - cci-control-port: + Usage: required + Value type: + Definition: a phandle containing the CCI control interface node + the master is connected to. + +Example: + + cpus { + #size-cells = <0>; + #address-cells = <1>; + + CPU0: cpu@0 { + device_type = "cpu"; + compatible = "arm,cortex-a15"; + cci-control-port = <&cci_control1>; + reg = <0x0>; + }; + + CPU1: cpu@1 { + device_type = "cpu"; + compatible = "arm,cortex-a15"; + cci-control-port = <&cci_control1>; + reg = <0x1>; + }; + + CPU2: cpu@100 { + device_type = "cpu"; + compatible = "arm,cortex-a7"; + cci-control-port = <&cci_control2>; + reg = <0x100>; + }; + + CPU3: cpu@101 { + device_type = "cpu"; + compatible = "arm,cortex-a7"; + cci-control-port = <&cci_control2>; + reg = <0x101>; + }; + + }; + + dma0: dma@3000000 { + compatible = "arm,pl330", "arm,primecell"; + cci-control-port = <&cci_control0>; + reg = <0x0 0x3000000 0x0 0x1000>; + interrupts = <10>; + #dma-cells = <1>; + #dma-channels = <8>; + #dma-requests = <32>; + }; + + cci@2c090000 { + compatible = "arm,cci-400"; + #address-cells = <1>; + #size-cells = <1>; + reg = <0x0 0x2c090000 0 0x1000>; + ranges = <0x0 0x0 0x2c090000 0x6000>; + + cci_control0: slave-if@1000 { + compatible = "arm,cci-400-ctrl-if"; + interface-type = "ace-lite"; + reg = <0x1000 0x1000>; + }; + + cci_control1: slave-if@4000 { + compatible = "arm,cci-400-ctrl-if"; + interface-type = "ace"; + reg = <0x4000 0x1000>; + }; + + cci_control2: slave-if@5000 { + compatible = "arm,cci-400-ctrl-if"; + interface-type = "ace"; + reg = <0x5000 0x1000>; + }; + }; + +This CCI node corresponds to a CCI component whose control registers sits +at address 0x000000002c090000. +CCI slave interface @0x000000002c091000 is connected to dma controller dma0. +CCI slave interface @0x000000002c094000 is connected to CPUs {CPU0, CPU1}; +CCI slave interface @0x000000002c095000 is connected to CPUs {CPU2, CPU3}; diff --git a/drivers/bus/Kconfig b/drivers/bus/Kconfig index b05ecab915c4..5286e2d333b0 100644 --- a/drivers/bus/Kconfig +++ b/drivers/bus/Kconfig @@ -26,4 +26,11 @@ config OMAP_INTERCONNECT help Driver to enable OMAP interconnect error handling driver. + +config ARM_CCI + bool "ARM CCI driver support" + depends on ARM + help + Driver supporting the CCI cache coherent interconnect for ARM + platforms. endmenu diff --git a/drivers/bus/Makefile b/drivers/bus/Makefile index 3c7b53c12091..670cea443802 100644 --- a/drivers/bus/Makefile +++ b/drivers/bus/Makefile @@ -7,3 +7,5 @@ obj-$(CONFIG_OMAP_OCP2SCP) += omap-ocp2scp.o # Interconnect bus driver for OMAP SoCs. obj-$(CONFIG_OMAP_INTERCONNECT) += omap_l3_smx.o omap_l3_noc.o +# CCI cache coherent interconnect for ARM platforms +obj-$(CONFIG_ARM_CCI) += arm-cci.o diff --git a/drivers/bus/arm-cci.c b/drivers/bus/arm-cci.c new file mode 100644 index 000000000000..ea81fa4a28db --- /dev/null +++ b/drivers/bus/arm-cci.c @@ -0,0 +1,426 @@ +/* + * CCI cache coherent interconnect driver + * + * Copyright (C) 2013 ARM Ltd. + * Author: Lorenzo Pieralisi + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed "as is" WITHOUT ANY WARRANTY of any + * kind, whether express or implied; without even the implied warranty + * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include +#include +#include +#include +#include + +#include +#include + +#define CCI_PORT_CTRL 0x0 +#define CCI_CTRL_STATUS 0xc + +#define CCI_ENABLE_SNOOP_REQ 0x1 +#define CCI_ENABLE_DVM_REQ 0x2 +#define CCI_ENABLE_REQ (CCI_ENABLE_SNOOP_REQ | CCI_ENABLE_DVM_REQ) + +struct cci_nb_ports { + unsigned int nb_ace; + unsigned int nb_ace_lite; +}; + +enum cci_ace_port_type { + ACE_INVALID_PORT = 0x0, + ACE_PORT, + ACE_LITE_PORT, +}; + +struct cci_ace_port { + void __iomem *base; + enum cci_ace_port_type type; + struct device_node *dn; +}; + +static struct cci_ace_port *ports; +static unsigned int nb_cci_ports; + +static void __iomem *cci_ctrl_base; + +struct cpu_port { + u64 mpidr; + u32 port; +}; +/* + * Use the port MSB as valid flag, shift can be made dynamic + * by computing number of bits required for port indexes. + * Code disabling CCI cpu ports runs with D-cache invalidated + * and SCTLR bit clear so data accesses must be kept to a minimum + * to improve performance; for now shift is left static to + * avoid one more data access while disabling the CCI port. + */ +#define PORT_VALID_SHIFT 31 +#define PORT_VALID (0x1 << PORT_VALID_SHIFT) + +static inline void init_cpu_port(struct cpu_port *port, u32 index, u64 mpidr) +{ + port->port = PORT_VALID | index; + port->mpidr = mpidr; +} + +static inline bool cpu_port_is_valid(struct cpu_port *port) +{ + return !!(port->port & PORT_VALID); +} + +static inline bool cpu_port_match(struct cpu_port *port, u64 mpidr) +{ + return port->mpidr == (mpidr & MPIDR_HWID_BITMASK); +} + +static struct cpu_port cpu_port[NR_CPUS]; + +/** + * __cci_ace_get_port - Function to retrieve the port index connected to + * a cpu or device. + * + * @dn: device node of the device to look-up + * @type: port type + * + * Return value: + * - CCI port index if success + * - -ENODEV if failure + */ +static int __cci_ace_get_port(struct device_node *dn, int type) +{ + int i; + bool ace_match; + struct device_node *cci_portn; + + cci_portn = of_parse_phandle(dn, "cci-control-port", 0); + for (i = 0; i < nb_cci_ports; i++) { + ace_match = ports[i].type == type; + if (ace_match && cci_portn == ports[i].dn) + return i; + } + return -ENODEV; +} + +int cci_ace_get_port(struct device_node *dn) +{ + return __cci_ace_get_port(dn, ACE_LITE_PORT); +} +EXPORT_SYMBOL_GPL(cci_ace_get_port); + +static void __init cci_ace_init_ports(void) +{ + int port, ac, cpu; + u64 hwid; + const u32 *cell; + struct device_node *cpun, *cpus; + + cpus = of_find_node_by_path("/cpus"); + if (WARN(!cpus, "Missing cpus node, bailing out\n")) + return; + + if (WARN_ON(of_property_read_u32(cpus, "#address-cells", &ac))) + ac = of_n_addr_cells(cpus); + + /* + * Port index look-up speeds up the function disabling ports by CPU, + * since the logical to port index mapping is done once and does + * not change after system boot. + * The stashed index array is initialized for all possible CPUs + * at probe time. + */ + for_each_child_of_node(cpus, cpun) { + if (of_node_cmp(cpun->type, "cpu")) + continue; + cell = of_get_property(cpun, "reg", NULL); + if (WARN(!cell, "%s: missing reg property\n", cpun->full_name)) + continue; + + hwid = of_read_number(cell, ac); + cpu = get_logical_index(hwid & MPIDR_HWID_BITMASK); + + if (cpu < 0 || !cpu_possible(cpu)) + continue; + port = __cci_ace_get_port(cpun, ACE_PORT); + if (port < 0) + continue; + + init_cpu_port(&cpu_port[cpu], port, cpu_logical_map(cpu)); + } + + for_each_possible_cpu(cpu) { + WARN(!cpu_port_is_valid(&cpu_port[cpu]), + "CPU %u does not have an associated CCI port\n", + cpu); + } +} +/* + * Functions to enable/disable a CCI interconnect slave port + * + * They are called by low-level power management code to disable slave + * interfaces snoops and DVM broadcast. + * Since they may execute with cache data allocation disabled and + * after the caches have been cleaned and invalidated the functions provide + * no explicit locking since they may run with D-cache disabled, so normal + * cacheable kernel locks based on ldrex/strex may not work. + * Locking has to be provided by BSP implementations to ensure proper + * operations. + */ + +/** + * cci_port_control() - function to control a CCI port + * + * @port: index of the port to setup + * @enable: if true enables the port, if false disables it + */ +static void notrace cci_port_control(unsigned int port, bool enable) +{ + void __iomem *base = ports[port].base; + + writel_relaxed(enable ? CCI_ENABLE_REQ : 0, base + CCI_PORT_CTRL); + /* + * This function is called from power down procedures + * and must not execute any instruction that might + * cause the processor to be put in a quiescent state + * (eg wfi). Hence, cpu_relax() can not be added to this + * read loop to optimize power, since it might hide possibly + * disruptive operations. + */ + while (readl_relaxed(cci_ctrl_base + CCI_CTRL_STATUS) & 0x1) + ; +} + +/** + * cci_disable_port_by_cpu() - function to disable a CCI port by CPU + * reference + * + * @mpidr: mpidr of the CPU whose CCI port should be disabled + * + * Disabling a CCI port for a CPU implies disabling the CCI port + * controlling that CPU cluster. Code disabling CPU CCI ports + * must make sure that the CPU running the code is the last active CPU + * in the cluster ie all other CPUs are quiescent in a low power state. + * + * Return: + * 0 on success + * -ENODEV on port look-up failure + */ +int notrace cci_disable_port_by_cpu(u64 mpidr) +{ + int cpu; + bool is_valid; + for (cpu = 0; cpu < nr_cpu_ids; cpu++) { + is_valid = cpu_port_is_valid(&cpu_port[cpu]); + if (is_valid && cpu_port_match(&cpu_port[cpu], mpidr)) { + cci_port_control(cpu_port[cpu].port, false); + return 0; + } + } + return -ENODEV; +} +EXPORT_SYMBOL_GPL(cci_disable_port_by_cpu); + +/** + * __cci_control_port_by_device() - function to control a CCI port by device + * reference + * + * @dn: device node pointer of the device whose CCI port should be + * controlled + * @enable: if true enables the port, if false disables it + * + * Return: + * 0 on success + * -ENODEV on port look-up failure + */ +int notrace __cci_control_port_by_device(struct device_node *dn, bool enable) +{ + int port; + + if (!dn) + return -ENODEV; + + port = __cci_ace_get_port(dn, ACE_LITE_PORT); + if (WARN_ONCE(port < 0, "node %s ACE lite port look-up failure\n", + dn->full_name)) + return -ENODEV; + cci_port_control(port, enable); + return 0; +} +EXPORT_SYMBOL_GPL(__cci_control_port_by_device); + +/** + * __cci_control_port_by_index() - function to control a CCI port by port index + * + * @port: port index previously retrieved with cci_ace_get_port() + * @enable: if true enables the port, if false disables it + * + * Return: + * 0 on success + * -ENODEV on port index out of range + * -EPERM if operation carried out on an ACE PORT + */ +int notrace __cci_control_port_by_index(u32 port, bool enable) +{ + if (port >= nb_cci_ports || ports[port].type == ACE_INVALID_PORT) + return -ENODEV; + /* + * CCI control for ports connected to CPUS is extremely fragile + * and must be made to go through a specific and controlled + * interface (ie cci_disable_port_by_cpu(); control by general purpose + * indexing is therefore disabled for ACE ports. + */ + if (ports[port].type == ACE_PORT) + return -EPERM; + + cci_port_control(port, enable); + return 0; +} +EXPORT_SYMBOL_GPL(__cci_control_port_by_index); + +static const struct cci_nb_ports cci400_ports = { + .nb_ace = 2, + .nb_ace_lite = 3 +}; + +static const struct of_device_id arm_cci_matches[] = { + {.compatible = "arm,cci-400", .data = &cci400_ports }, + {}, +}; + +static const struct of_device_id arm_cci_ctrl_if_matches[] = { + {.compatible = "arm,cci-400-ctrl-if", }, + {}, +}; + +static int __init cci_probe(void) +{ + struct cci_nb_ports const *cci_config; + int ret, i, nb_ace = 0, nb_ace_lite = 0; + struct device_node *np, *cp; + const char *match_str; + bool is_ace; + + np = of_find_matching_node(NULL, arm_cci_matches); + if (!np) + return -ENODEV; + + cci_config = of_match_node(arm_cci_matches, np)->data; + if (!cci_config) + return -ENODEV; + + nb_cci_ports = cci_config->nb_ace + cci_config->nb_ace_lite; + + ports = kcalloc(sizeof(*ports), nb_cci_ports, GFP_KERNEL); + if (!ports) + return -ENOMEM; + + cci_ctrl_base = of_iomap(np, 0); + + if (!cci_ctrl_base) { + WARN(1, "unable to ioremap CCI ctrl\n"); + ret = -ENXIO; + goto memalloc_err; + } + + for_each_child_of_node(np, cp) { + if (!of_match_node(arm_cci_ctrl_if_matches, cp)) + continue; + + i = nb_ace + nb_ace_lite; + + if (i >= nb_cci_ports) + break; + + if (of_property_read_string(cp, "interface-type", + &match_str)) { + WARN(1, "node %s missing interface-type property\n", + cp->full_name); + continue; + } + is_ace = strcmp(match_str, "ace") == 0; + if (!is_ace && strcmp(match_str, "ace-lite")) { + WARN(1, "node %s containing invalid interface-type property, skipping it\n", + cp->full_name); + continue; + } + + ports[i].base = of_iomap(cp, 0); + + if (!ports[i].base) { + WARN(1, "unable to ioremap CCI port %d\n", i); + continue; + } + + if (is_ace) { + if (WARN_ON(nb_ace >= cci_config->nb_ace)) + continue; + ports[i].type = ACE_PORT; + ++nb_ace; + } else { + if (WARN_ON(nb_ace_lite >= cci_config->nb_ace_lite)) + continue; + ports[i].type = ACE_LITE_PORT; + ++nb_ace_lite; + } + ports[i].dn = cp; + } + + /* initialize a stashed array of ACE ports to speed-up look-up */ + cci_ace_init_ports(); + + /* + * Multi-cluster systems may need this data when non-coherent, during + * cluster power-up/power-down. Make sure it reaches main memory. + */ + sync_cache_w(&cci_ctrl_base); + sync_cache_w(&ports); + sync_cache_w(&cpu_port); + __sync_cache_range_w(ports, sizeof(*ports) * nb_cci_ports); + pr_info("ARM CCI driver probed\n"); + return 0; + +memalloc_err: + + kfree(ports); + return ret; +} + +static int cci_init_status = -EAGAIN; +static DEFINE_MUTEX(cci_probing); + +static int __init cci_init(void) +{ + if (cci_init_status != -EAGAIN) + return cci_init_status; + + mutex_lock(&cci_probing); + if (cci_init_status == -EAGAIN) + cci_init_status = cci_probe(); + mutex_unlock(&cci_probing); + return cci_init_status; +} + +/* + * To sort out early init calls ordering a helper function is provided to + * check if the CCI driver has beed initialized. Function check if the driver + * has been initialized, if not it calls the init function that probes + * the driver and updates the return value. + */ +bool __init cci_probed(void) +{ + return cci_init() == 0; +} +EXPORT_SYMBOL_GPL(cci_probed); + +early_initcall(cci_init); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("ARM CCI support"); diff --git a/include/linux/arm-cci.h b/include/linux/arm-cci.h new file mode 100644 index 000000000000..79d6edf446d5 --- /dev/null +++ b/include/linux/arm-cci.h @@ -0,0 +1,61 @@ +/* + * CCI cache coherent interconnect support + * + * Copyright (C) 2013 ARM Ltd. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#ifndef __LINUX_ARM_CCI_H +#define __LINUX_ARM_CCI_H + +#include +#include + +struct device_node; + +#ifdef CONFIG_ARM_CCI +extern bool cci_probed(void); +extern int cci_ace_get_port(struct device_node *dn); +extern int cci_disable_port_by_cpu(u64 mpidr); +extern int __cci_control_port_by_device(struct device_node *dn, bool enable); +extern int __cci_control_port_by_index(u32 port, bool enable); +#else +static inline bool cci_probed(void) { return false; } +static inline int cci_ace_get_port(struct device_node *dn) +{ + return -ENODEV; +} +static inline int cci_disable_port_by_cpu(u64 mpidr) { return -ENODEV; } +static inline int __cci_control_port_by_device(struct device_node *dn, + bool enable) +{ + return -ENODEV; +} +static inline int __cci_control_port_by_index(u32 port, bool enable) +{ + return -ENODEV; +} +#endif +#define cci_disable_port_by_device(dev) \ + __cci_control_port_by_device(dev, false) +#define cci_enable_port_by_device(dev) \ + __cci_control_port_by_device(dev, true) +#define cci_disable_port_by_index(dev) \ + __cci_control_port_by_index(dev, false) +#define cci_enable_port_by_index(dev) \ + __cci_control_port_by_index(dev, true) + +#endif -- cgit v1.2.3 From ab28e08f102695a7f81cc5618e1f78fbbfba92a0 Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Tue, 21 May 2013 23:34:41 -0400 Subject: drivers/bus: arm-cci: function to enable CCI ports from early boot code This provides cci_enable_port_for_self(). This is the counterpart to cci_disable_port_by_cpu(self). This is meant to be called from the MCPM machine specific power_up_setup callback code when the appropriate affinity level needs to be initialized. The code therefore has to be position independent as the MMU is still off and it cannot rely on any stack space. Signed-off-by: Nicolas Pitre Reviewed-by: Dave Martin --- drivers/bus/arm-cci.c | 119 +++++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 113 insertions(+), 6 deletions(-) diff --git a/drivers/bus/arm-cci.c b/drivers/bus/arm-cci.c index ea81fa4a28db..380319ebf729 100644 --- a/drivers/bus/arm-cci.c +++ b/drivers/bus/arm-cci.c @@ -43,6 +43,7 @@ enum cci_ace_port_type { struct cci_ace_port { void __iomem *base; + unsigned long phys; enum cci_ace_port_type type; struct device_node *dn; }; @@ -51,11 +52,13 @@ static struct cci_ace_port *ports; static unsigned int nb_cci_ports; static void __iomem *cci_ctrl_base; +static unsigned long cci_ctrl_phys; struct cpu_port { u64 mpidr; u32 port; }; + /* * Use the port MSB as valid flag, shift can be made dynamic * by computing number of bits required for port indexes. @@ -229,6 +232,102 @@ int notrace cci_disable_port_by_cpu(u64 mpidr) } EXPORT_SYMBOL_GPL(cci_disable_port_by_cpu); +/** + * cci_enable_port_for_self() - enable a CCI port for calling CPU + * + * Enabling a CCI port for the calling CPU implies enabling the CCI + * port controlling that CPU's cluster. Caller must make sure that the + * CPU running the code is the first active CPU in the cluster and all + * other CPUs are quiescent in a low power state or waiting for this CPU + * to complete the CCI initialization. + * + * Because this is called when the MMU is still off and with no stack, + * the code must be position independent and ideally rely on callee + * clobbered registers only. To achieve this we must code this function + * entirely in assembler. + * + * On success this returns with the proper CCI port enabled. In case of + * any failure this never returns as the inability to enable the CCI is + * fatal and there is no possible recovery at this stage. + */ +asmlinkage void __naked cci_enable_port_for_self(void) +{ + asm volatile ("\n" + +" mrc p15, 0, r0, c0, c0, 5 @ get MPIDR value \n" +" and r0, r0, #"__stringify(MPIDR_HWID_BITMASK)" \n" +" adr r1, 5f \n" +" ldr r2, [r1] \n" +" add r1, r1, r2 @ &cpu_port \n" +" add ip, r1, %[sizeof_cpu_port] \n" + + /* Loop over the cpu_port array looking for a matching MPIDR */ +"1: ldr r2, [r1, %[offsetof_cpu_port_mpidr_lsb]] \n" +" cmp r2, r0 @ compare MPIDR \n" +" bne 2f \n" + + /* Found a match, now test port validity */ +" ldr r3, [r1, %[offsetof_cpu_port_port]] \n" +" tst r3, #"__stringify(PORT_VALID)" \n" +" bne 3f \n" + + /* no match, loop with the next cpu_port entry */ +"2: add r1, r1, %[sizeof_struct_cpu_port] \n" +" cmp r1, ip @ done? \n" +" blo 1b \n" + + /* CCI port not found -- cheaply try to stall this CPU */ +"cci_port_not_found: \n" +" wfi \n" +" wfe \n" +" b cci_port_not_found \n" + + /* Use matched port index to look up the corresponding ports entry */ +"3: bic r3, r3, #"__stringify(PORT_VALID)" \n" +" adr r0, 6f \n" +" ldmia r0, {r1, r2} \n" +" sub r1, r1, r0 @ virt - phys \n" +" ldr r0, [r0, r2] @ *(&ports) \n" +" mov r2, %[sizeof_struct_ace_port] \n" +" mla r0, r2, r3, r0 @ &ports[index] \n" +" sub r0, r0, r1 @ virt_to_phys() \n" + + /* Enable the CCI port */ +" ldr r0, [r0, %[offsetof_port_phys]] \n" +" mov r3, #"__stringify(CCI_ENABLE_REQ)" \n" +" str r3, [r0, #"__stringify(CCI_PORT_CTRL)"] \n" + + /* poll the status reg for completion */ +" adr r1, 7f \n" +" ldr r0, [r1] \n" +" ldr r0, [r0, r1] @ cci_ctrl_base \n" +"4: ldr r1, [r0, #"__stringify(CCI_CTRL_STATUS)"] \n" +" tst r1, #1 \n" +" bne 4b \n" + +" mov r0, #0 \n" +" bx lr \n" + +" .align 2 \n" +"5: .word cpu_port - . \n" +"6: .word . \n" +" .word ports - 6b \n" +"7: .word cci_ctrl_phys - . \n" + : : + [sizeof_cpu_port] "i" (sizeof(cpu_port)), +#ifndef __ARMEB__ + [offsetof_cpu_port_mpidr_lsb] "i" (offsetof(struct cpu_port, mpidr)), +#else + [offsetof_cpu_port_mpidr_lsb] "i" (offsetof(struct cpu_port, mpidr)+4), +#endif + [offsetof_cpu_port_port] "i" (offsetof(struct cpu_port, port)), + [sizeof_struct_cpu_port] "i" (sizeof(struct cpu_port)), + [sizeof_struct_ace_port] "i" (sizeof(struct cci_ace_port)), + [offsetof_port_phys] "i" (offsetof(struct cci_ace_port, phys)) ); + + unreachable(); +} + /** * __cci_control_port_by_device() - function to control a CCI port by device * reference @@ -306,6 +405,7 @@ static int __init cci_probe(void) struct cci_nb_ports const *cci_config; int ret, i, nb_ace = 0, nb_ace_lite = 0; struct device_node *np, *cp; + struct resource res; const char *match_str; bool is_ace; @@ -323,9 +423,12 @@ static int __init cci_probe(void) if (!ports) return -ENOMEM; - cci_ctrl_base = of_iomap(np, 0); - - if (!cci_ctrl_base) { + ret = of_address_to_resource(np, 0, &res); + if (!ret) { + cci_ctrl_base = ioremap(res.start, resource_size(&res)); + cci_ctrl_phys = res.start; + } + if (ret || !cci_ctrl_base) { WARN(1, "unable to ioremap CCI ctrl\n"); ret = -ENXIO; goto memalloc_err; @@ -353,9 +456,12 @@ static int __init cci_probe(void) continue; } - ports[i].base = of_iomap(cp, 0); - - if (!ports[i].base) { + ret = of_address_to_resource(cp, 0, &res); + if (!ret) { + ports[i].base = ioremap(res.start, resource_size(&res)); + ports[i].phys = res.start; + } + if (ret || !ports[i].base) { WARN(1, "unable to ioremap CCI port %d\n", i); continue; } @@ -382,6 +488,7 @@ static int __init cci_probe(void) * cluster power-up/power-down. Make sure it reaches main memory. */ sync_cache_w(&cci_ctrl_base); + sync_cache_w(&cci_ctrl_phys); sync_cache_w(&ports); sync_cache_w(&cpu_port); __sync_cache_range_w(ports, sizeof(*ports) * nb_cci_ports); -- cgit v1.2.3 From 1d36a70525f2876794e1b3e4bc8b82c7d43682b4 Mon Sep 17 00:00:00 2001 From: Jon Medhurst Date: Fri, 10 May 2013 14:37:37 +0100 Subject: ARM: vexpress: Add CCI nodes to TC2 device-tree Signed-off-by: Jon Medhurst --- arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) diff --git a/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts b/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts index d2803be4e1a8..12bd4ea07eac 100644 --- a/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts +++ b/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts @@ -37,30 +37,35 @@ device_type = "cpu"; compatible = "arm,cortex-a15"; reg = <0>; + cci-control-port = <&cci_control1>; }; cpu1: cpu@1 { device_type = "cpu"; compatible = "arm,cortex-a15"; reg = <1>; + cci-control-port = <&cci_control1>; }; cpu2: cpu@2 { device_type = "cpu"; compatible = "arm,cortex-a7"; reg = <0x100>; + cci-control-port = <&cci_control2>; }; cpu3: cpu@3 { device_type = "cpu"; compatible = "arm,cortex-a7"; reg = <0x101>; + cci-control-port = <&cci_control2>; }; cpu4: cpu@4 { device_type = "cpu"; compatible = "arm,cortex-a7"; reg = <0x102>; + cci-control-port = <&cci_control2>; }; }; @@ -104,6 +109,26 @@ interrupts = <1 9 0xf04>; }; + cci@2c090000 { + compatible = "arm,cci-400"; + #address-cells = <1>; + #size-cells = <1>; + reg = <0 0x2c090000 0 0x1000>; + ranges = <0x0 0x0 0x2c090000 0x10000>; + + cci_control1: slave-if@4000 { + compatible = "arm,cci-400-ctrl-if"; + interface-type = "ace"; + reg = <0x4000 0x1000>; + }; + + cci_control2: slave-if@5000 { + compatible = "arm,cci-400-ctrl-if"; + interface-type = "ace"; + reg = <0x5000 0x1000>; + }; + }; + memory-controller@7ffd0000 { compatible = "arm,pl354", "arm,primecell"; reg = <0 0x7ffd0000 0 0x1000>; -- cgit v1.2.3 From 0b9bd1cc8d48a4dcb61aa01a4a80d690f99f24a1 Mon Sep 17 00:00:00 2001 From: Jon Medhurst Date: Fri, 10 May 2013 14:47:27 +0100 Subject: drivers: ARM CCI: Add a platform driver stub for the PMU This is a hack to enable the old CCI PMU patches to be used with the new CCI driver. The CCI PMU is (mis)represented by a separate node in TC2 device-tree. Signed-off-by: Jon Medhurst --- arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts | 10 ++++++ drivers/bus/arm-cci.c | 51 ++++++++++++++++++++++++++++++ 2 files changed, 61 insertions(+) diff --git a/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts b/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts index 12bd4ea07eac..9d802fc71fcf 100644 --- a/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts +++ b/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts @@ -129,6 +129,16 @@ }; }; + cci-pmu@2c099000 { + compatible = "arm,cci-400-pmu"; + reg = <0 0x2c099000 0 0x6000>; + interrupts = <0 101 4>, + <0 102 4>, + <0 103 4>, + <0 104 4>, + <0 105 4>; + }; + memory-controller@7ffd0000 { compatible = "arm,pl354", "arm,primecell"; reg = <0 0x7ffd0000 0 0x1000>; diff --git a/drivers/bus/arm-cci.c b/drivers/bus/arm-cci.c index 380319ebf729..a621d3d8b90f 100644 --- a/drivers/bus/arm-cci.c +++ b/drivers/bus/arm-cci.c @@ -17,12 +17,15 @@ #include #include #include +#include #include #include #include #include +#define DRIVER_NAME "CCI" + #define CCI_PORT_CTRL 0x0 #define CCI_CTRL_STATUS 0xc @@ -54,6 +57,53 @@ static unsigned int nb_cci_ports; static void __iomem *cci_ctrl_base; static unsigned long cci_ctrl_phys; +#ifdef CONFIG_HW_PERF_EVENTS + +static void __iomem *cci_pmu_base; + +static int cci_pmu_probe(struct platform_device *pdev) +{ + struct resource *res; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + cci_pmu_base = devm_ioremap_resource(&pdev->dev, res); + if (IS_ERR(cci_pmu_base)) + return PTR_ERR(cci_pmu_base); + + return 0; +} + +static const struct of_device_id arm_cci_pmu_matches[] = { + {.compatible = "arm,cci-400-pmu"}, + {}, +}; + +static struct platform_driver cci_pmu_platform_driver = { + .driver = { + .name = DRIVER_NAME, + .of_match_table = arm_cci_pmu_matches, + }, + .probe = cci_pmu_probe, +}; + +static int __init cci_pmu_init(void) +{ + if (platform_driver_register(&cci_pmu_platform_driver)) + WARN(1, "unable to register CCI platform driver\n"); + return 0; +} + +#else + +static int __init cci_pmu_init(void) +{ + return 0; +} + +static void cci_pmu_destroy(void) { } + +#endif /* CONFIG_HW_PERF_EVENTS */ + struct cpu_port { u64 mpidr; u32 port; @@ -529,5 +579,6 @@ bool __init cci_probed(void) EXPORT_SYMBOL_GPL(cci_probed); early_initcall(cci_init); +core_initcall(cci_pmu_init); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("ARM CCI support"); -- cgit v1.2.3 From 4311d5a4b599317bc321b1d8b9c0d1f4e49e7df5 Mon Sep 17 00:00:00 2001 From: Punit Agrawal Date: Thu, 20 Sep 2012 16:39:25 +0100 Subject: Support CCI PMU in perf CCI400 has a set of counters that can be used to profile different transations at CCI master and slave interfaces. These counters can observe different kinds of transations passing through the CCI and provide a system-level view of activity. This patch adds support for CCI PMU by extending the existing CCI driver. Signed-off-by: Punit Agrawal Signed-off-by: Jon Medhurst --- drivers/bus/arm-cci.c | 365 +++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 364 insertions(+), 1 deletion(-) diff --git a/drivers/bus/arm-cci.c b/drivers/bus/arm-cci.c index a621d3d8b90f..ed284ab0041b 100644 --- a/drivers/bus/arm-cci.c +++ b/drivers/bus/arm-cci.c @@ -22,6 +22,8 @@ #include #include +#include +#include #include #define DRIVER_NAME "CCI" @@ -61,6 +63,362 @@ static unsigned long cci_ctrl_phys; static void __iomem *cci_pmu_base; +#define CCI400_PMCR 0x0100 + +#define CCI400_PMU_CYCLE_CNTR_BASE 0x0000 +#define CCI400_PMU_CNTR_BASE(idx) (CCI400_PMU_CYCLE_CNTR_BASE + (idx) * 0x1000) + +#define CCI400_PMCR_CEN 0x00000001 +#define CCI400_PMCR_RST 0x00000002 +#define CCI400_PMCR_CCR 0x00000004 +#define CCI400_PMCR_CCD 0x00000008 +#define CCI400_PMCR_EX 0x00000010 +#define CCI400_PMCR_DP 0x00000020 +#define CCI400_PMCR_NCNT_MASK 0x0000F800 +#define CCI400_PMCR_NCNT_SHIFT 11 + +#define CCI400_PMU_EVT_SEL 0x000 +#define CCI400_PMU_CNTR 0x004 +#define CCI400_PMU_CNTR_CTRL 0x008 +#define CCI400_PMU_OVERFLOW 0x00C + +#define CCI400_PMU_OVERFLOW_FLAG 1 + +enum cci400_perf_events { + CCI400_PMU_CYCLES = 0xFF +}; + +#define CCI400_PMU_EVENT_MASK 0xff +#define CCI400_PMU_EVENT_SOURCE(event) ((event >> 5) & 0x7) +#define CCI400_PMU_EVENT_CODE(event) (event & 0x1f) + +#define CCI400_PMU_EVENT_SOURCE_S0 0 +#define CCI400_PMU_EVENT_SOURCE_S4 4 +#define CCI400_PMU_EVENT_SOURCE_M0 5 +#define CCI400_PMU_EVENT_SOURCE_M2 7 + +#define CCI400_PMU_EVENT_SLAVE_MIN 0x0 +#define CCI400_PMU_EVENT_SLAVE_MAX 0x13 + +#define CCI400_PMU_EVENT_MASTER_MIN 0x14 +#define CCI400_PMU_EVENT_MASTER_MAX 0x1A + +#define CCI400_PMU_MAX_HW_EVENTS 5 /* CCI PMU has 4 counters + 1 cycle counter */ + +#define CCI400_PMU_CYCLE_COUNTER_IDX 0 +#define CCI400_PMU_COUNTER0_IDX 1 +#define CCI400_PMU_COUNTER_LAST(cci_pmu) (CCI400_PMU_CYCLE_COUNTER_IDX + cci_pmu->num_events - 1) + + +static struct perf_event *events[CCI400_PMU_MAX_HW_EVENTS]; +static unsigned long used_mask[BITS_TO_LONGS(CCI400_PMU_MAX_HW_EVENTS)]; +static struct pmu_hw_events cci_hw_events = { + .events = events, + .used_mask = used_mask, +}; + +static int cci_pmu_validate_hw_event(u8 hw_event) +{ + u8 ev_source = CCI400_PMU_EVENT_SOURCE(hw_event); + u8 ev_code = CCI400_PMU_EVENT_CODE(hw_event); + + if (ev_source <= CCI400_PMU_EVENT_SOURCE_S4 && + ev_code <= CCI400_PMU_EVENT_SLAVE_MAX) + return hw_event; + else if (CCI400_PMU_EVENT_SOURCE_M0 <= ev_source && + ev_source <= CCI400_PMU_EVENT_SOURCE_M2 && + CCI400_PMU_EVENT_MASTER_MIN <= ev_code && + ev_code <= CCI400_PMU_EVENT_MASTER_MAX) + return hw_event; + + return -EINVAL; +} + +static inline int cci_pmu_counter_is_valid(struct arm_pmu *cci_pmu, int idx) +{ + return CCI400_PMU_CYCLE_COUNTER_IDX <= idx && + idx <= CCI400_PMU_COUNTER_LAST(cci_pmu); +} + +static inline u32 cci_pmu_read_register(int idx, unsigned int offset) +{ + return readl_relaxed(cci_pmu_base + CCI400_PMU_CNTR_BASE(idx) + offset); +} + +static inline void cci_pmu_write_register(u32 value, int idx, unsigned int offset) +{ + return writel_relaxed(value, cci_pmu_base + CCI400_PMU_CNTR_BASE(idx) + offset); +} + +static inline void cci_pmu_disable_counter(int idx) +{ + cci_pmu_write_register(0, idx, CCI400_PMU_CNTR_CTRL); +} + +static inline void cci_pmu_enable_counter(int idx) +{ + cci_pmu_write_register(1, idx, CCI400_PMU_CNTR_CTRL); +} + +static inline void cci_pmu_select_event(int idx, unsigned long event) +{ + event &= CCI400_PMU_EVENT_MASK; + cci_pmu_write_register(event, idx, CCI400_PMU_EVT_SEL); +} + +static u32 cci_pmu_get_max_counters(void) +{ + u32 n_cnts = (readl_relaxed(cci_ctrl_base + CCI400_PMCR) & + CCI400_PMCR_NCNT_MASK) >> CCI400_PMCR_NCNT_SHIFT; + + /* add 1 for cycle counter */ + return n_cnts + 1; +} + +static struct pmu_hw_events *cci_pmu_get_hw_events(void) +{ + return &cci_hw_events; +} + +static int cci_pmu_get_event_idx(struct pmu_hw_events *hw, struct perf_event *event) +{ + struct arm_pmu *cci_pmu = to_arm_pmu(event->pmu); + struct hw_perf_event *hw_event = &event->hw; + unsigned long cci_event = hw_event->config_base & CCI400_PMU_EVENT_MASK; + int idx; + + if (cci_event == CCI400_PMU_CYCLES) { + if (test_and_set_bit(CCI400_PMU_CYCLE_COUNTER_IDX, hw->used_mask)) + return -EAGAIN; + + return CCI400_PMU_CYCLE_COUNTER_IDX; + } + + for (idx = CCI400_PMU_COUNTER0_IDX; idx <= CCI400_PMU_COUNTER_LAST(cci_pmu); ++idx) { + if (!test_and_set_bit(idx, hw->used_mask)) + return idx; + } + + /* No counters available */ + return -EAGAIN; +} + +static int cci_pmu_map_event(struct perf_event *event) +{ + int mapping; + u8 config = event->attr.config & CCI400_PMU_EVENT_MASK; + + if (event->attr.type < PERF_TYPE_MAX) + return -ENOENT; + + /* 0xff is used to represent CCI Cycles */ + if (config == 0xff) + mapping = config; + else + mapping = cci_pmu_validate_hw_event(config); + + return mapping; +} + +static int cci_pmu_request_irq(struct arm_pmu *cci_pmu, irq_handler_t handler) +{ + int irq, err, i = 0; + struct platform_device *pmu_device = cci_pmu->plat_device; + + if (unlikely(!pmu_device)) + return -ENODEV; + + /* CCI exports 6 interrupts - 1 nERRORIRQ + 5 nEVNTCNTOVERFLOW (PMU) + nERRORIRQ will be handled by secure firmware on TC2. So we + assume that all CCI interrupts listed in the linux device + tree are PMU interrupts. + + The following code should then be able to handle different routing + of the CCI PMU interrupts. + */ + while ((irq = platform_get_irq(pmu_device, i)) > 0) { + err = request_irq(irq, handler, 0, "arm-cci-pmu", cci_pmu); + if (err) { + dev_err(&pmu_device->dev, "unable to request IRQ%d for ARM CCI PMU counters\n", + irq); + return err; + } + i++; + } + + return 0; +} + +static irqreturn_t cci_pmu_handle_irq(int irq_num, void *dev) +{ + struct arm_pmu *cci_pmu = (struct arm_pmu *)dev; + struct pmu_hw_events *events = cci_pmu->get_hw_events(); + struct perf_sample_data data; + struct pt_regs *regs; + int idx; + + regs = get_irq_regs(); + + /* Iterate over counters and update the corresponding perf events. + This should work regardless of whether we have per-counter overflow + interrupt or a combined overflow interrupt. */ + for (idx = CCI400_PMU_CYCLE_COUNTER_IDX; idx <= CCI400_PMU_COUNTER_LAST(cci_pmu); idx++) { + struct perf_event *event = events->events[idx]; + struct hw_perf_event *hw_counter; + + if (!event) + continue; + + hw_counter = &event->hw; + + /* Did this counter overflow? */ + if (!(cci_pmu_read_register(idx, CCI400_PMU_OVERFLOW) & CCI400_PMU_OVERFLOW_FLAG)) + continue; + cci_pmu_write_register(CCI400_PMU_OVERFLOW_FLAG, idx, CCI400_PMU_OVERFLOW); + + armpmu_event_update(event); + perf_sample_data_init(&data, 0, hw_counter->last_period); + if (!armpmu_event_set_period(event)) + continue; + + if (perf_event_overflow(event, &data, regs)) + cci_pmu->disable(event); + } + + irq_work_run(); + return IRQ_HANDLED; +} + +static void cci_pmu_free_irq(struct arm_pmu *cci_pmu) +{ + int irq, i = 0; + struct platform_device *pmu_device = cci_pmu->plat_device; + + while ((irq = platform_get_irq(pmu_device, i)) > 0) { + free_irq(irq, cci_pmu); + i++; + } +} + +static void cci_pmu_enable_event(struct perf_event *event) +{ + unsigned long flags; + struct arm_pmu *cci_pmu = to_arm_pmu(event->pmu); + struct pmu_hw_events *events = cci_pmu->get_hw_events(); + struct hw_perf_event *hw_counter = &event->hw; + int idx = hw_counter->idx; + + if (unlikely(!cci_pmu_counter_is_valid(cci_pmu, idx))) { + dev_err(&cci_pmu->plat_device->dev, "Invalid CCI PMU counter %d\n", idx); + return; + } + + raw_spin_lock_irqsave(&events->pmu_lock, flags); + + /* Configure the event to count, unless you are counting cycles */ + if (idx != CCI400_PMU_CYCLE_COUNTER_IDX) + cci_pmu_select_event(idx, hw_counter->config_base); + + cci_pmu_enable_counter(idx); + + raw_spin_unlock_irqrestore(&events->pmu_lock, flags); +} + +static void cci_pmu_disable_event(struct perf_event *event) +{ + unsigned long flags; + struct arm_pmu *cci_pmu = to_arm_pmu(event->pmu); + struct pmu_hw_events *events = cci_pmu->get_hw_events(); + struct hw_perf_event *hw_counter = &event->hw; + int idx = hw_counter->idx; + + if (unlikely(!cci_pmu_counter_is_valid(cci_pmu, idx))) { + dev_err(&cci_pmu->plat_device->dev, "Invalid CCI PMU counter %d\n", idx); + return; + } + + raw_spin_lock_irqsave(&events->pmu_lock, flags); + + cci_pmu_disable_counter(idx); + + raw_spin_unlock_irqrestore(&events->pmu_lock, flags); +} + +static void cci_pmu_start(struct arm_pmu *cci_pmu) +{ + u32 val; + unsigned long flags; + struct pmu_hw_events *events = cci_pmu->get_hw_events(); + + raw_spin_lock_irqsave(&events->pmu_lock, flags); + + /* Enable all the PMU counters. */ + val = readl(cci_ctrl_base + CCI400_PMCR) | CCI400_PMCR_CEN; + writel(val, cci_ctrl_base + CCI400_PMCR); + + raw_spin_unlock_irqrestore(&events->pmu_lock, flags); +} + +static void cci_pmu_stop(struct arm_pmu *cci_pmu) +{ + u32 val; + unsigned long flags; + struct pmu_hw_events *events = cci_pmu->get_hw_events(); + + raw_spin_lock_irqsave(&events->pmu_lock, flags); + + /* Disable all the PMU counters. */ + val = readl(cci_ctrl_base + CCI400_PMCR) & ~CCI400_PMCR_CEN; + writel(val, cci_ctrl_base + CCI400_PMCR); + + raw_spin_unlock_irqrestore(&events->pmu_lock, flags); +} + +static u32 cci_pmu_read_counter(struct perf_event *event) +{ + struct arm_pmu *cci_pmu = to_arm_pmu(event->pmu); + struct hw_perf_event *hw_counter = &event->hw; + int idx = hw_counter->idx; + u32 value; + + if (unlikely(!cci_pmu_counter_is_valid(cci_pmu, idx))) { + dev_err(&cci_pmu->plat_device->dev, "Invalid CCI PMU counter %d\n", idx); + return 0; + } + value = cci_pmu_read_register(idx, CCI400_PMU_CNTR); + + return value; +} + +static void cci_pmu_write_counter(struct perf_event *event, u32 value) +{ + struct arm_pmu *cci_pmu = to_arm_pmu(event->pmu); + struct hw_perf_event *hw_counter = &event->hw; + int idx = hw_counter->idx; + + if (unlikely(!cci_pmu_counter_is_valid(cci_pmu, idx))) + dev_err(&cci_pmu->plat_device->dev, "Invalid CCI PMU counter %d\n", idx); + else + cci_pmu_write_register(value, idx, CCI400_PMU_CNTR); +} + +static struct arm_pmu cci_pmu = { + .name = DRIVER_NAME, + .max_period = (1LLU << 32) - 1, + .get_hw_events = cci_pmu_get_hw_events, + .get_event_idx = cci_pmu_get_event_idx, + .map_event = cci_pmu_map_event, + .request_irq = cci_pmu_request_irq, + .handle_irq = cci_pmu_handle_irq, + .free_irq = cci_pmu_free_irq, + .enable = cci_pmu_enable_event, + .disable = cci_pmu_disable_event, + .start = cci_pmu_start, + .stop = cci_pmu_stop, + .read_counter = cci_pmu_read_counter, + .write_counter = cci_pmu_write_counter, +}; + static int cci_pmu_probe(struct platform_device *pdev) { struct resource *res; @@ -70,7 +428,12 @@ static int cci_pmu_probe(struct platform_device *pdev) if (IS_ERR(cci_pmu_base)) return PTR_ERR(cci_pmu_base); - return 0; + cci_pmu.plat_device = pdev; + cci_pmu.num_events = cci_pmu_get_max_counters(); + raw_spin_lock_init(&cci_hw_events.pmu_lock); + cpumask_setall(&cci_pmu.valid_cpus); + + return armpmu_register(&cci_pmu, -1); } static const struct of_device_id arm_cci_pmu_matches[] = { -- cgit v1.2.3