From patchwork Tue Oct 18 12:13:16 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jonathan Cameron X-Patchwork-Id: 13010434 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 6F85AC4332F for ; Tue, 18 Oct 2022 12:14:49 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229606AbiJRMOs (ORCPT ); Tue, 18 Oct 2022 08:14:48 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:46822 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229525AbiJRMOr (ORCPT ); Tue, 18 Oct 2022 08:14:47 -0400 Received: from frasgout.his.huawei.com (frasgout.his.huawei.com [185.176.79.56]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 70D8920BD0; Tue, 18 Oct 2022 05:14:46 -0700 (PDT) Received: from fraeml713-chm.china.huawei.com (unknown [172.18.147.226]) by frasgout.his.huawei.com (SkyGuard) with ESMTP id 4MsCSl0r85z685b0; Tue, 18 Oct 2022 20:13:43 +0800 (CST) Received: from lhrpeml500005.china.huawei.com (7.191.163.240) by fraeml713-chm.china.huawei.com (10.206.15.32) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.31; Tue, 18 Oct 2022 14:14:44 +0200 Received: from SecurePC-101-06.china.huawei.com (10.122.247.231) by lhrpeml500005.china.huawei.com (7.191.163.240) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.31; Tue, 18 Oct 2022 13:14:43 +0100 From: Jonathan Cameron To: , Dave Jiang CC: , Dan Williams , "Alison Schofield" , Vishal Verma , Ira Weiny , Ben Widawsky , , Will Deacon , Mark Rutland , Davidlohr Bueso Subject: [RFC PATCH v3 3/5] cxl/pci: Find and register CXL PMU devices Date: Tue, 18 Oct 2022 13:13:16 +0100 Message-ID: <20221018121318.22385-4-Jonathan.Cameron@huawei.com> X-Mailer: git-send-email 2.37.2 In-Reply-To: <20221018121318.22385-1-Jonathan.Cameron@huawei.com> References: <20221018121318.22385-1-Jonathan.Cameron@huawei.com> MIME-Version: 1.0 X-Originating-IP: [10.122.247.231] X-ClientProxiedBy: lhrpeml100001.china.huawei.com (7.191.160.183) To lhrpeml500005.china.huawei.com (7.191.163.240) X-CFilter-Loop: Reflected Precedence: bulk List-ID: X-Mailing-List: linux-cxl@vger.kernel.org CXL PMU devices can be found from entries in the Register Locator DVSEC. In order to register the minimum number of IRQ vectors necessary to support all CPMUs found, separate the registration into two steps. First find the devices, and query the IRQs used and then register the devices. Between these two steps, request the IRQ vectors necessary and enable bus master support. Future IRQ users for CXL type 3 devices (e.g. DOEs) will need to follow a similar pattern the number of vectors necessary is known before any parts of the driver stack rely on their availability. Signed-off-by: Jonathan Cameron --- Chances since v2: - 'use' Davidlohr's generic MSI/MSIX handling in the cxl/pci driver. There are various ways that could have been done. To keep the layering entirely intact I could have repeated the finding, register block identification twice. 1st time to get the max irq, then start again to find the devices. That seemed rather excessive. So instead I just stashed the max irq number in a pass that was identifying all the necessary info about the CPMU devices we are going to register later. Other open questions 1) Does hanging the CPMU off the PCI device make sense. These can occur in Switch upstream ports as well. 2) Naming. It would be nice if the device naming indicated which EP these were associated with, but that wouldn't be inline with the rest of the CXL bus device naming. What is best option here? --- drivers/cxl/core/Makefile | 1 + drivers/cxl/core/core.h | 3 ++ drivers/cxl/core/cpmu.c | 69 +++++++++++++++++++++++++++++++++++++++ drivers/cxl/core/port.c | 2 ++ drivers/cxl/core/regs.c | 29 ++++++++++++++++ drivers/cxl/cpmu.h | 54 ++++++++++++++++++++++++++++++ drivers/cxl/cxl.h | 15 +++++++++ drivers/cxl/cxlmem.h | 2 ++ drivers/cxl/cxlpci.h | 1 + drivers/cxl/pci.c | 53 ++++++++++++++++++++++++++++-- 10 files changed, 227 insertions(+), 2 deletions(-) diff --git a/drivers/cxl/core/Makefile b/drivers/cxl/core/Makefile index 79c7257f4107..1318e8a6830f 100644 --- a/drivers/cxl/core/Makefile +++ b/drivers/cxl/core/Makefile @@ -10,4 +10,5 @@ cxl_core-y += memdev.o cxl_core-y += mbox.o cxl_core-y += pci.o cxl_core-y += hdm.o +cxl_core-y += cpmu.o cxl_core-$(CONFIG_CXL_REGION) += region.o diff --git a/drivers/cxl/core/core.h b/drivers/cxl/core/core.h index 1d8f87be283f..d2b12cdfd61f 100644 --- a/drivers/cxl/core/core.h +++ b/drivers/cxl/core/core.h @@ -14,12 +14,14 @@ extern struct device_attribute dev_attr_create_pmem_region; extern struct device_attribute dev_attr_delete_region; extern struct device_attribute dev_attr_region; extern const struct device_type cxl_pmem_region_type; +extern const struct device_type cxl_cpmu_type; extern const struct device_type cxl_region_type; void cxl_decoder_kill_region(struct cxl_endpoint_decoder *cxled); #define CXL_REGION_ATTR(x) (&dev_attr_##x.attr) #define CXL_REGION_TYPE(x) (&cxl_region_type) #define SET_CXL_REGION_ATTR(x) (&dev_attr_##x.attr), #define CXL_PMEM_REGION_TYPE(x) (&cxl_pmem_region_type) +#define CXL_CPMU_TYPE(x) (&cxl_cpmu_region_type) int cxl_region_init(void); void cxl_region_exit(void); #else @@ -37,6 +39,7 @@ static inline void cxl_region_exit(void) #define CXL_REGION_TYPE(x) NULL #define SET_CXL_REGION_ATTR(x) #define CXL_PMEM_REGION_TYPE(x) NULL +#define CXL_CPMU_TYPE(x) NULL #endif struct cxl_send_command; diff --git a/drivers/cxl/core/cpmu.c b/drivers/cxl/core/cpmu.c new file mode 100644 index 000000000000..cad02f3d43c3 --- /dev/null +++ b/drivers/cxl/core/cpmu.c @@ -0,0 +1,69 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright(c) 2022 Huawei. All rights reserved. */ + +#include +#include +#include +#include +#include +#include +#include "core.h" + +static DEFINE_IDA(cpmu_ida); + +static void cxl_cpmu_release(struct device *dev) +{ + struct cxl_cpmu *cpmu = container_of(dev, struct cxl_cpmu, dev); + + ida_free(&cpmu_ida, cpmu->id); + kfree(cpmu); +} + +const struct device_type cxl_cpmu_type = { + .name = "cxl_cpmu", + .release = cxl_cpmu_release, +}; + +static void remove_dev(void *dev) +{ + device_del(dev); +} + +int devm_cxl_cpmu_add(struct device *parent, struct cxl_cpmu_regs *regs, int index) +{ + struct cxl_cpmu *cpmu; + struct device *dev; + int rc; + + cpmu = kzalloc(sizeof(*cpmu), GFP_KERNEL); + if (!cpmu) + return -ENOMEM; + + cpmu->base = regs->cpmu; + dev = &cpmu->dev; + device_initialize(dev); + device_set_pm_not_required(dev); + dev->parent = parent; + dev->bus = &cxl_bus_type; + dev->type = &cxl_cpmu_type; + rc = ida_alloc(&cpmu_ida, GFP_KERNEL); + if (rc < 0) + goto err; + cpmu->id = rc; + + rc = dev_set_name(dev, "cpmu%d", cpmu->id); + if (rc) + goto err; + + rc = device_add(dev); + if (rc) + goto err; + + return devm_add_action_or_reset(parent, remove_dev, dev); + +err: + put_device(&cpmu->dev); + return rc; +} +EXPORT_SYMBOL_NS_GPL(devm_cxl_cpmu_add, CXL); + diff --git a/drivers/cxl/core/port.c b/drivers/cxl/core/port.c index 1629c7a4033f..66bde8e2edb8 100644 --- a/drivers/cxl/core/port.c +++ b/drivers/cxl/core/port.c @@ -55,6 +55,8 @@ static int cxl_device_id(struct device *dev) return CXL_DEVICE_MEMORY_EXPANDER; if (dev->type == CXL_REGION_TYPE()) return CXL_DEVICE_REGION; + if (dev->type == &cxl_cpmu_type) + return CXL_DEVICE_CPMU; return 0; } diff --git a/drivers/cxl/core/regs.c b/drivers/cxl/core/regs.c index 2f651211d120..1ba1a77ecbf1 100644 --- a/drivers/cxl/core/regs.c +++ b/drivers/cxl/core/regs.c @@ -6,6 +6,7 @@ #include #include #include +#include /** * DOC: cxl registers @@ -331,3 +332,31 @@ int cxl_count_regblock(struct pci_dev *pdev, enum cxl_regloc_type type) } EXPORT_SYMBOL_NS_GPL(cxl_count_regblock, CXL); +int cxl_map_cpmu_regs(struct pci_dev *pdev, + struct cxl_cpmu_regs *regs, + struct cxl_register_map *map) +{ + struct device *dev = &pdev->dev; + resource_size_t phys_addr; + + phys_addr = pci_resource_start(pdev, map->barno) + map->block_offset; + regs->cpmu = devm_cxl_iomap_block(dev, phys_addr, CPMU_REGMAP_SIZE); + if (!regs->cpmu) + return -ENOMEM; + + return 0; +} +EXPORT_SYMBOL_NS_GPL(cxl_map_cpmu_regs, CXL); + +int cxl_cpmu_get_irq(struct cxl_cpmu_regs *regs) +{ + int irq = -1; + u64 val; + + val = readq(regs->cpmu + CPMU_CAP_REG); + if (FIELD_GET(CPMU_CAP_INT, val)) + irq = FIELD_GET(CPMU_CAP_MSI_N_MSK, val); + + return irq; +} +EXPORT_SYMBOL_NS_GPL(cxl_cpmu_get_irq, CXL); diff --git a/drivers/cxl/cpmu.h b/drivers/cxl/cpmu.h new file mode 100644 index 000000000000..2db885524330 --- /dev/null +++ b/drivers/cxl/cpmu.h @@ -0,0 +1,54 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright(c) 2022 Huawei + * CXL Specification rev 3.0 Setion 8.2.7 (CPMU Register Interface) + */ +#ifndef CXL_CPMU_H +#define CXL_CPMU_H + +#define CPMU_CAP_REG 0x0 +#define CPMU_CAP_NUM_COUNTERS_MSK GENMASK_ULL(4, 0) +#define CPMU_CAP_COUNTER_WIDTH_MSK GENMASK_ULL(15, 8) +#define CPMU_CAP_NUM_EVN_CAP_REG_SUP_MSK GENMASK_ULL(24, 20) +#define CPMU_CAP_FILTERS_SUP_MSK GENMASK_ULL(39, 32) +#define CPMU_CAP_MSI_N_MSK GENMASK_ULL(47, 44) +#define CPMU_CAP_WRITEABLE_WHEN_FROZEN BIT_ULL(48) +#define CPMU_CAP_FREEZE BIT_ULL(49) +#define CPMU_CAP_INT BIT_ULL(50) +#define CPMU_CAP_VERSION_MSK GENMASK_ULL(63, 60) + +#define CPMU_OVERFLOW_REG 0x10 +#define CPMU_FREEZE_REG 0x18 +#define CPMU_EVENT_CAP_REG(n) (0x100 + 8 * (n)) +#define CPMU_EVENT_CAP_SUPPORTED_EVENTS_MSK GENMASK_ULL(31, 0) +#define CPMU_EVENT_CAP_GROUP_ID_MSK GENMASK_ULL(47, 32) +#define CPMU_EVENT_CAP_VENDOR_ID_MSK GENMASK_ULL(63, 48) + +#define CPMU_COUNTER_CFG_REG(n) (0x200 + 8 * (n)) +#define CPMU_COUNTER_CFG_TYPE_MSK GENMASK_ULL(1, 0) +#define CPMU_COUNTER_CFG_TYPE_FREE_RUN 0 +#define CPMU_COUNTER_CFG_TYPE_FIXED_FUN 1 +#define CPMU_COUNTER_CFG_TYPE_CONFIGURABLE 2 +#define CPMU_COUNTER_CFG_ENABLE BIT_ULL(8) +#define CPMU_COUNTER_CFG_INT_ON_OVRFLW BIT_ULL(9) +#define CPMU_COUNTER_CFG_FREEZE_ON_OVRFLW BIT_ULL(10) +#define CPMU_COUNTER_CFG_EDGE BIT_ULL(11) +#define CPMU_COUNTER_CFG_INVERT BIT_ULL(12) +#define CPMU_COUNTER_CFG_THRESHOLD_MSK GENMASK_ULL(23, 16) +#define CPMU_COUNTER_CFG_EVENTS_MSK GENMASK_ULL(55, 24) +#define CPMU_COUNTER_CFG_EVENT_GRP_ID_IDX_MSK GENMASK_ULL(63, 59) + +#define CPMU_FILTER_CFG_REG(n, f) (0x400 + 4 * ((f) + (n) * 8)) +#define CPMU_FILTER_CFG_VALUE_MSK GENMASK(15, 0) + +#define CPMU_COUNTER_REG(n) (0xc00 + 8 * (n)) + +#define CPMU_REGMAP_SIZE 0xe00 /* Table 8-32 CXL 3.0 specification */ +struct cxl_cpmu { + struct device dev; + void __iomem *base; + int id; +}; + +#define to_cxl_cpmu(dev) container_of(dev, struct cxl_cpmu, dev) +#endif diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h index 5a1bcdbda654..2768e1677e2b 100644 --- a/drivers/cxl/cxl.h +++ b/drivers/cxl/cxl.h @@ -166,6 +166,10 @@ struct cxl_regs { struct_group_tagged(cxl_device_regs, device_regs, void __iomem *status, *mbox, *memdev; ); + + struct_group_tagged(cxl_cpmu_regs, cpmu_regs, + void __iomem *cpmu; + ); }; struct cxl_reg_map { @@ -184,6 +188,10 @@ struct cxl_device_reg_map { struct cxl_reg_map memdev; }; +struct cxl_cpmu_reg_map { + struct cxl_reg_map cpmu; +}; + /** * struct cxl_register_map - DVSEC harvested register block mapping parameters * @base: virtual base of the register-block-BAR + @block_offset @@ -201,6 +209,7 @@ struct cxl_register_map { union { struct cxl_component_reg_map component_map; struct cxl_device_reg_map device_map; + struct cxl_cpmu_reg_map cpmu_map; }; }; @@ -219,6 +228,10 @@ enum cxl_regloc_type; int cxl_count_regblock(struct pci_dev *pdev, enum cxl_regloc_type type); int cxl_find_regblock(struct pci_dev *pdev, enum cxl_regloc_type type, struct cxl_register_map *map, int index); +int cxl_map_cpmu_regs(struct pci_dev *pdev, + struct cxl_cpmu_regs *regs, + struct cxl_register_map *map); +int cxl_cpmu_get_irq(struct cxl_cpmu_regs *regs); void __iomem *devm_cxl_iomap_block(struct device *dev, resource_size_t addr, resource_size_t length); @@ -626,6 +639,7 @@ void cxl_driver_unregister(struct cxl_driver *cxl_drv); #define CXL_DEVICE_MEMORY_EXPANDER 5 #define CXL_DEVICE_REGION 6 #define CXL_DEVICE_PMEM_REGION 7 +#define CXL_DEVICE_CPMU 8 #define MODULE_ALIAS_CXL(type) MODULE_ALIAS("cxl:t" __stringify(type) "*") #define CXL_MODALIAS_FMT "cxl:t%d" @@ -653,6 +667,7 @@ static inline struct cxl_pmem_region *to_cxl_pmem_region(struct device *dev) } #endif +int devm_cxl_cpmu_add(struct device *parent, struct cxl_cpmu_regs *regs, int idx); /* * Unit test builds overrides this to __weak, find the 'strong' version * of these symbols in tools/testing/cxl/. diff --git a/drivers/cxl/cxlmem.h b/drivers/cxl/cxlmem.h index 72b69b003302..e2d5a4edd80d 100644 --- a/drivers/cxl/cxlmem.h +++ b/drivers/cxl/cxlmem.h @@ -212,6 +212,7 @@ struct cxl_endpoint_dvsec_info { * @serial: PCIe Device Serial Number * @doe_mbs: PCI DOE mailbox array * @has_irq: PCIe MSI-X/MSI support + * @cpmu_max_vector: Maximum MSI/MSIX vector used for CPMU instances. * @mbox_send: @dev specific transport for transmitting mailbox commands * * See section 8.2.9.5.2 Capacity Configuration and Label Storage for @@ -249,6 +250,7 @@ struct cxl_dev_state { struct xarray doe_mbs; bool has_irq; + int cpmu_max_vector; int (*mbox_send)(struct cxl_dev_state *cxlds, struct cxl_mbox_cmd *cmd); }; diff --git a/drivers/cxl/cxlpci.h b/drivers/cxl/cxlpci.h index eec597dbe763..a842c3d1341a 100644 --- a/drivers/cxl/cxlpci.h +++ b/drivers/cxl/cxlpci.h @@ -59,6 +59,7 @@ enum cxl_regloc_type { CXL_REGLOC_RBI_COMPONENT, CXL_REGLOC_RBI_VIRT, CXL_REGLOC_RBI_MEMDEV, + CXL_REGLOC_RBI_CPMU, CXL_REGLOC_RBI_TYPES }; diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c index ef066e24d3a3..2e8361679431 100644 --- a/drivers/cxl/pci.c +++ b/drivers/cxl/pci.c @@ -428,6 +428,11 @@ static void devm_cxl_pci_create_doe(struct cxl_dev_state *cxlds) } } +static int cxl_cpmu_get_max_msgnum(struct cxl_dev_state *cxlds) +{ + return cxlds->cpmu_max_vector; +} + /** * struct cxl_irq_cap - CXL feature that is capable of receiving MSI-X/MSI irqs. * @@ -442,7 +447,7 @@ struct cxl_irq_cap { }; static const struct cxl_irq_cap cxl_irq_cap_table[] = { - NULL + { "CPMU", cxl_cpmu_get_max_msgnum }, }; static void cxl_pci_free_irq_vectors(void *data) @@ -497,10 +502,11 @@ static int cxl_pci_alloc_irq_vectors(struct cxl_dev_state *cxlds) static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) { + struct cxl_cpmu_regs *cpmu_regs_array = NULL; struct cxl_register_map map; struct cxl_memdev *cxlmd; struct cxl_dev_state *cxlds; - int rc; + int i, rc, cpmu_count; /* * Double check the anonymous union trickery in struct cxl_regs @@ -561,11 +567,54 @@ static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) if (rc) return rc; + /* + * Before registering sub devices that use interrupts, the maximum + * interrupt used on the device should be established to allow the correct + * number of irq vectors to be requested. Separate the registration process + * into before / after interrupt vector request. + */ + cpmu_count = cxl_count_regblock(pdev, CXL_REGLOC_RBI_CPMU); + if (cpmu_count) { + cpmu_regs_array = kmalloc_array(cpmu_count, sizeof(cpmu_regs_array), + GFP_KERNEL); + if (!cpmu_regs_array) + return -ENOMEM; + } + + /* Pass one - just enough to find out what interrupts are used */ + cxlds->cpmu_max_vector = -1; + for (i = 0; i < cpmu_count; i++) { + int irq; + + rc = cxl_find_regblock(pdev, CXL_REGLOC_RBI_CPMU, &map, i); + if (rc) { + kfree(cpmu_regs_array); + return rc; + } + + rc = cxl_map_cpmu_regs(pdev, &cpmu_regs_array[i], &map); + if (rc) { + kfree(cpmu_regs_array); + return rc; + } + + irq = cxl_cpmu_get_irq(&cpmu_regs_array[i]); + cxlds->cpmu_max_vector = max(cxlds->cpmu_max_vector, irq); + } + if (!cxl_pci_alloc_irq_vectors(cxlds)) { cxlds->has_irq = true; + pci_set_master(pdev); } else cxlds->has_irq = false; + /* Pass 2 - register the CPMU instances, currently no support without interrupts */ + if (cxlds->has_irq) { + for (i = 0; i < cpmu_count; i++) + devm_cxl_cpmu_add(cxlds->dev, &cpmu_regs_array[i], i); + } + kfree(cpmu_regs_array); + cxlmd = devm_cxl_add_memdev(cxlds); if (IS_ERR(cxlmd)) return PTR_ERR(cxlmd);