From patchwork Sat Sep 10 14:20:06 2011 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Benjamin Herrenschmidt X-Patchwork-Id: 114183 Return-Path: X-Original-To: patchwork-incoming@ozlabs.org Delivered-To: patchwork-incoming@ozlabs.org Received: from ozlabs.org (localhost [IPv6:::1]) by ozlabs.org (Postfix) with ESMTP id 953AB10214B for ; Sun, 11 Sep 2011 00:23:37 +1000 (EST) Received: by ozlabs.org (Postfix) id E5534B7D5E; Sun, 11 Sep 2011 00:21:00 +1000 (EST) Delivered-To: linuxppc-dev@ozlabs.org Received: from gate.crashing.org (gate.crashing.org [63.228.1.57]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTPS id CC133B7CEB for ; Sun, 11 Sep 2011 00:20:59 +1000 (EST) Received: from pasglop.au.ibm.com (localhost.localdomain [127.0.0.1]) by gate.crashing.org (8.14.1/8.13.8) with ESMTP id p8AEKAkZ007133; Sat, 10 Sep 2011 09:20:52 -0500 From: Benjamin Herrenschmidt To: linuxppc-dev@ozlabs.org Subject: [PATCH 19/21] powerpc/powernv: Add support for p5ioc2 PCI-X and PCIe Date: Sat, 10 Sep 2011 11:20:06 -0300 Message-Id: <1315664408-16797-19-git-send-email-benh@kernel.crashing.org> X-Mailer: git-send-email 1.7.4.1 In-Reply-To: <1315664408-16797-1-git-send-email-benh@kernel.crashing.org> References: <1315664408-16797-1-git-send-email-benh@kernel.crashing.org> X-BeenThere: linuxppc-dev@lists.ozlabs.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Errors-To: linuxppc-dev-bounces+patchwork-incoming=ozlabs.org@lists.ozlabs.org Sender: linuxppc-dev-bounces+patchwork-incoming=ozlabs.org@lists.ozlabs.org This adds support for PCI-X and PCIe on the p5ioc2 IO hub using OPAL. This includes allocating & setting up TCE tables and config space access routines. This also supports fallbacks via RTAS when OPAL is absent, using legacy TCE format pre-allocated via the device-tree (BML style) Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/powernv/Makefile | 2 + arch/powerpc/platforms/powernv/pci-p5ioc2.c | 185 +++++++++++++++++ arch/powerpc/platforms/powernv/pci.c | 286 +++++++++++++++++++++++++++ arch/powerpc/platforms/powernv/pci.h | 38 ++++ arch/powerpc/platforms/powernv/powernv.h | 6 + arch/powerpc/platforms/powernv/setup.c | 3 +- 6 files changed, 519 insertions(+), 1 deletions(-) create mode 100644 arch/powerpc/platforms/powernv/pci-p5ioc2.c create mode 100644 arch/powerpc/platforms/powernv/pci.c create mode 100644 arch/powerpc/platforms/powernv/pci.h diff --git a/arch/powerpc/platforms/powernv/Makefile b/arch/powerpc/platforms/powernv/Makefile index 618ad83..cf9dec8 100644 --- a/arch/powerpc/platforms/powernv/Makefile +++ b/arch/powerpc/platforms/powernv/Makefile @@ -2,3 +2,5 @@ obj-y += setup.o opal-takeover.o opal-wrappers.o opal.o obj-y += opal-rtc.o opal-nvram.o obj-$(CONFIG_SMP) += smp.o +obj-$(CONFIG_PCI) += pci.o pci-p5ioc2.o + diff --git a/arch/powerpc/platforms/powernv/pci-p5ioc2.c b/arch/powerpc/platforms/powernv/pci-p5ioc2.c new file mode 100644 index 0000000..afabc2b --- /dev/null +++ b/arch/powerpc/platforms/powernv/pci-p5ioc2.c @@ -0,0 +1,185 @@ +/* + * Support PCI/PCIe on PowerNV platforms + * + * Currently supports only P5IOC2 + * + * Copyright 2011 Benjamin Herrenschmidt, IBM Corp. + * + * 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. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "powernv.h" +#include "pci.h" + +/* For now, use a fixed amount of TCE memory for each p5ioc2 + * hub, 16M will do + */ +#define P5IOC2_TCE_MEMORY 0x01000000 + +static void __devinit pnv_pci_p5ioc2_dma_dev_setup(struct pnv_phb *phb, + struct pci_dev *pdev) +{ + if (phb->p5ioc2.iommu_table.it_map == NULL) + iommu_init_table(&phb->p5ioc2.iommu_table, phb->hose->node); + + set_iommu_table_base(&pdev->dev, &phb->p5ioc2.iommu_table); +} + +static void __init pnv_pci_init_p5ioc2_phb(struct device_node *np, + void *tce_mem, u64 tce_size) +{ + struct pnv_phb *phb; + const u64 *prop64; + u64 phb_id; + int64_t rc; + static int primary = 1; + + pr_info(" Initializing p5ioc2 PHB %s\n", np->full_name); + + prop64 = of_get_property(np, "ibm,opal-phbid", NULL); + if (!prop64) { + pr_err(" Missing \"ibm,opal-phbid\" property !\n"); + return; + } + phb_id = be64_to_cpup(prop64); + pr_devel(" PHB-ID : 0x%016llx\n", phb_id); + pr_devel(" TCE AT : 0x%016lx\n", __pa(tce_mem)); + pr_devel(" TCE SZ : 0x%016llx\n", tce_size); + + rc = opal_pci_set_phb_tce_memory(phb_id, __pa(tce_mem), tce_size); + if (rc != OPAL_SUCCESS) { + pr_err(" Failed to set TCE memory, OPAL error %lld\n", rc); + return; + } + + phb = alloc_bootmem(sizeof(struct pnv_phb)); + if (phb) { + memset(phb, 0, sizeof(struct pnv_phb)); + phb->hose = pcibios_alloc_controller(np); + } + if (!phb || !phb->hose) { + pr_err(" Failed to allocate PCI controller\n"); + return; + } + + spin_lock_init(&phb->lock); + phb->hose->first_busno = 0; + phb->hose->last_busno = 0xff; + phb->hose->private_data = phb; + phb->opal_id = phb_id; + phb->type = PNV_PHB_P5IOC2; + + phb->regs = of_iomap(np, 0); + + if (phb->regs == NULL) + pr_err(" Failed to map registers !\n"); + else { + pr_devel(" P_BUID = 0x%08x\n", in_be32(phb->regs + 0x100)); + pr_devel(" P_IOSZ = 0x%08x\n", in_be32(phb->regs + 0x1b0)); + pr_devel(" P_IO_ST = 0x%08x\n", in_be32(phb->regs + 0x1e0)); + pr_devel(" P_MEM1_H = 0x%08x\n", in_be32(phb->regs + 0x1a0)); + pr_devel(" P_MEM1_L = 0x%08x\n", in_be32(phb->regs + 0x190)); + pr_devel(" P_MSZ1_L = 0x%08x\n", in_be32(phb->regs + 0x1c0)); + pr_devel(" P_MEM_ST = 0x%08x\n", in_be32(phb->regs + 0x1d0)); + pr_devel(" P_MEM2_H = 0x%08x\n", in_be32(phb->regs + 0x2c0)); + pr_devel(" P_MEM2_L = 0x%08x\n", in_be32(phb->regs + 0x2b0)); + pr_devel(" P_MSZ2_H = 0x%08x\n", in_be32(phb->regs + 0x2d0)); + pr_devel(" P_MSZ2_L = 0x%08x\n", in_be32(phb->regs + 0x2e0)); + } + + /* Interpret the "ranges" property */ + /* This also maps the I/O region and sets isa_io/mem_base */ + pci_process_bridge_OF_ranges(phb->hose, np, primary); + primary = 0; + + phb->hose->ops = &pnv_pci_ops; + + /* Setup TCEs */ + phb->dma_dev_setup = pnv_pci_p5ioc2_dma_dev_setup; + pnv_pci_setup_iommu_table(&phb->p5ioc2.iommu_table, + tce_mem, tce_size, 0); +} + +void __init pnv_pci_init_p5ioc2_hub(struct device_node *np) +{ + struct device_node *phbn; + const u64 *prop64; + u64 hub_id; + void *tce_mem; + uint64_t tce_per_phb; + int64_t rc; + int phb_count = 0; + + pr_info("Probing p5ioc2 IO-Hub %s\n", np->full_name); + + prop64 = of_get_property(np, "ibm,opal-hubid", NULL); + if (!prop64) { + pr_err(" Missing \"ibm,opal-hubid\" property !\n"); + return; + } + hub_id = be64_to_cpup(prop64); + pr_info(" HUB-ID : 0x%016llx\n", hub_id); + + /* Currently allocate 16M of TCE memory for every Hub + * + * XXX TODO: Make it chip local if possible + */ + tce_mem = __alloc_bootmem(P5IOC2_TCE_MEMORY, P5IOC2_TCE_MEMORY, + __pa(MAX_DMA_ADDRESS)); + if (!tce_mem) { + pr_err(" Failed to allocate TCE Memory !\n"); + return; + } + pr_debug(" TCE : 0x%016lx..0x%016lx\n", + __pa(tce_mem), __pa(tce_mem) + P5IOC2_TCE_MEMORY - 1); + rc = opal_pci_set_hub_tce_memory(hub_id, __pa(tce_mem), + P5IOC2_TCE_MEMORY); + if (rc != OPAL_SUCCESS) { + pr_err(" Failed to allocate TCE memory, OPAL error %lld\n", rc); + return; + } + + /* Count child PHBs */ + for_each_child_of_node(np, phbn) { + if (of_device_is_compatible(phbn, "ibm,p5ioc2-pcix") || + of_device_is_compatible(phbn, "ibm,p5ioc2-pciex")) + phb_count++; + } + + /* Calculate how much TCE space we can give per PHB */ + tce_per_phb = __rounddown_pow_of_two(P5IOC2_TCE_MEMORY / phb_count); + pr_info(" Allocating %lld MB of TCE memory per PHB\n", + tce_per_phb >> 20); + + /* Initialize PHBs */ + for_each_child_of_node(np, phbn) { + if (of_device_is_compatible(phbn, "ibm,p5ioc2-pcix") || + of_device_is_compatible(phbn, "ibm,p5ioc2-pciex")) { + pnv_pci_init_p5ioc2_phb(phbn, tce_mem, tce_per_phb); + tce_mem += tce_per_phb; + } + } +} diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c new file mode 100644 index 0000000..b512489 --- /dev/null +++ b/arch/powerpc/platforms/powernv/pci.c @@ -0,0 +1,286 @@ +/* + * Support PCI/PCIe on PowerNV platforms + * + * Currently supports only P5IOC2 + * + * Copyright 2011 Benjamin Herrenschmidt, IBM Corp. + * + * 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. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "powernv.h" +#include "pci.h" + + +#define cfg_dbg(fmt...) do { } while(0) +//#define cfg_dbg(fmt...) printk(fmt) + + +static void pnv_pci_config_check_eeh(struct pnv_phb *phb, struct pci_bus *bus, + u32 bdfn) +{ + s64 rc; + u8 fstate; + u16 pcierr; + u32 pe_no; + + /* Get PE# if we support IODA */ + pe_no = phb->bdfn_to_pe ? phb->bdfn_to_pe(phb, bus, bdfn & 0xff) : 0; + + /* Read freeze status */ + rc = opal_pci_eeh_freeze_status(phb->opal_id, pe_no, &fstate, &pcierr, + NULL); + if (rc) { + pr_warning("PCI %d: Failed to read EEH status for PE#%d," + " err %lld\n", phb->hose->global_number, pe_no, rc); + return; + } + cfg_dbg(" -> EEH check, bdfn=%04x PE%d fstate=%x\n", + bdfn, pe_no, fstate); + if (fstate != 0) { + rc = opal_pci_eeh_freeze_clear(phb->opal_id, pe_no, + OPAL_EEH_ACTION_CLEAR_FREEZE_ALL); + if (rc) { + pr_warning("PCI %d: Failed to clear EEH freeze state" + " for PE#%d, err %lld\n", + phb->hose->global_number, pe_no, rc); + } + } +} + +static int pnv_pci_read_config(struct pci_bus *bus, + unsigned int devfn, + int where, int size, u32 *val) +{ + struct pci_controller *hose = pci_bus_to_host(bus); + struct pnv_phb *phb = hose->private_data; + u32 bdfn = (((uint64_t)bus->number) << 8) | devfn; + s64 rc; + + if (hose == NULL) + return PCIBIOS_DEVICE_NOT_FOUND; + + switch (size) { + case 1: { + u8 v8; + rc = opal_pci_config_read_byte(phb->opal_id, bdfn, where, &v8); + *val = (rc == OPAL_SUCCESS) ? v8 : 0xff; + break; + } + case 2: { + u16 v16; + rc = opal_pci_config_read_half_word(phb->opal_id, bdfn, where, + &v16); + *val = (rc == OPAL_SUCCESS) ? v16 : 0xffff; + break; + } + case 4: { + u32 v32; + rc = opal_pci_config_read_word(phb->opal_id, bdfn, where, &v32); + *val = (rc == OPAL_SUCCESS) ? v32 : 0xffffffff; + break; + } + default: + return PCIBIOS_FUNC_NOT_SUPPORTED; + } + cfg_dbg("pnv_pci_read_config bus: %x devfn: %x +%x/%x -> %08x\n", + bus->number, devfn, where, size, *val); + + /* Check if the PHB got frozen due to an error (no response) */ + pnv_pci_config_check_eeh(phb, bus, bdfn); + + return PCIBIOS_SUCCESSFUL; +} + +static int pnv_pci_write_config(struct pci_bus *bus, + unsigned int devfn, + int where, int size, u32 val) +{ + struct pci_controller *hose = pci_bus_to_host(bus); + struct pnv_phb *phb = hose->private_data; + u32 bdfn = (((uint64_t)bus->number) << 8) | devfn; + + if (hose == NULL) + return PCIBIOS_DEVICE_NOT_FOUND; + + cfg_dbg("pnv_pci_write_config bus: %x devfn: %x +%x/%x -> %08x\n", + bus->number, devfn, where, size, val); + switch (size) { + case 1: + opal_pci_config_write_byte(phb->opal_id, bdfn, where, val); + break; + case 2: + opal_pci_config_write_half_word(phb->opal_id, bdfn, where, val); + break; + case 4: + opal_pci_config_write_word(phb->opal_id, bdfn, where, val); + break; + default: + return PCIBIOS_FUNC_NOT_SUPPORTED; + } + /* Check if the PHB got frozen due to an error (no response) */ + pnv_pci_config_check_eeh(phb, bus, bdfn); + + return PCIBIOS_SUCCESSFUL; +} + +struct pci_ops pnv_pci_ops = { + .read = pnv_pci_read_config, + .write = pnv_pci_write_config, +}; + +static int pnv_tce_build(struct iommu_table *tbl, long index, long npages, + unsigned long uaddr, enum dma_data_direction direction, + struct dma_attrs *attrs) +{ + u64 proto_tce; + u64 *tcep; + u64 rpn; + + proto_tce = TCE_PCI_READ; // Read allowed + + if (direction != DMA_TO_DEVICE) + proto_tce |= TCE_PCI_WRITE; + + tcep = ((u64 *)tbl->it_base) + index; + + while (npages--) { + /* can't move this out since we might cross LMB boundary */ + rpn = (virt_to_abs(uaddr)) >> TCE_SHIFT; + *tcep = proto_tce | (rpn & TCE_RPN_MASK) << TCE_RPN_SHIFT; + + uaddr += TCE_PAGE_SIZE; + tcep++; + } + return 0; +} + +static void pnv_tce_free(struct iommu_table *tbl, long index, long npages) +{ + u64 *tcep = ((u64 *)tbl->it_base) + index; + + while (npages--) + *(tcep++) = 0; +} + +void pnv_pci_setup_iommu_table(struct iommu_table *tbl, + void *tce_mem, u64 tce_size, + u64 dma_offset) +{ + tbl->it_blocksize = 16; + tbl->it_base = (unsigned long)tce_mem; + tbl->it_offset = dma_offset >> IOMMU_PAGE_SHIFT; + tbl->it_index = 0; + tbl->it_size = tce_size >> 3; + tbl->it_busno = 0; + tbl->it_type = TCE_PCI; +} + +static struct iommu_table * __devinit +pnv_pci_setup_bml_iommu(struct pci_controller *hose) +{ + struct iommu_table *tbl; + const __be64 *basep; + const __be32 *sizep; + + basep = of_get_property(hose->dn, "linux,tce-base", NULL); + sizep = of_get_property(hose->dn, "linux,tce-size", NULL); + if (basep == NULL || sizep == NULL) { + pr_err("PCI: %s has missing tce entries !\n", hose->dn->full_name); + return NULL; + } + tbl = kzalloc_node(sizeof(struct iommu_table), GFP_KERNEL, hose->node); + if (WARN_ON(!tbl)) + return NULL; + pnv_pci_setup_iommu_table(tbl, __va(be64_to_cpup(basep)), + be32_to_cpup(sizep), 0); + iommu_init_table(tbl, hose->node); + return tbl; +} + +static void __devinit pnv_pci_dma_fallback_setup(struct pci_controller *hose, + struct pci_dev *pdev) +{ + struct device_node *np = pci_bus_to_OF_node(hose->bus); + struct pci_dn *pdn; + + if (np == NULL) + return; + pdn = PCI_DN(np); + if (!pdn->iommu_table) + pdn->iommu_table = pnv_pci_setup_bml_iommu(hose); + if (!pdn->iommu_table) + return; + set_iommu_table_base(&pdev->dev, pdn->iommu_table); +} + +static void __devinit pnv_pci_dma_dev_setup(struct pci_dev *pdev) +{ + struct pci_controller *hose = pci_bus_to_host(pdev->bus); + struct pnv_phb *phb = hose->private_data; + + /* If we have no phb structure, try to setup a fallback based on + * the device-tree (RTAS PCI for example) + */ + if (phb && phb->dma_dev_setup) + phb->dma_dev_setup(phb, pdev); + else + pnv_pci_dma_fallback_setup(hose, pdev); +} + +void __init pnv_pci_init(void) +{ + struct device_node *np; + + pci_set_flags(PCI_CAN_SKIP_ISA_ALIGN); + + /* We do not want to just probe */ + pci_probe_only = 0; + + /* OPAL absent, try POPAL first then RTAS detection of PHBs */ + if (!firmware_has_feature(FW_FEATURE_OPAL)) { +#ifdef CONFIG_PPC_POWERNV_RTAS + init_pci_config_tokens(); + find_and_init_phbs(); +#endif /* CONFIG_PPC_POWERNV_RTAS */ + } else { + /* OPAL is here, do our normal stuff */ + + /* Look for p5ioc2 IO-Hubs */ + for_each_compatible_node(np, NULL, "ibm,p5ioc2") + pnv_pci_init_p5ioc2_hub(np); + } + + /* Setup the linkage between OF nodes and PHBs */ + pci_devs_phb_init(); + + /* Configure IOMMU DMA hooks */ + ppc_md.pci_dma_dev_setup = pnv_pci_dma_dev_setup; + ppc_md.tce_build = pnv_tce_build; + ppc_md.tce_free = pnv_tce_free; + set_pci_dma_ops(&dma_iommu_ops); + +} diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h new file mode 100644 index 0000000..b8be721 --- /dev/null +++ b/arch/powerpc/platforms/powernv/pci.h @@ -0,0 +1,38 @@ +#ifndef __POWERNV_PCI_H +#define __POWERNV_PCI_H + +struct pci_dn; + +enum pnv_phb_type { + PNV_PHB_P5IOC2, + PNV_PHB_IODA1, + PNV_PHB_IODA2, +}; + +struct pnv_phb { + struct pci_controller *hose; + enum pnv_phb_type type; + u64 opal_id; + void __iomem *regs; + spinlock_t lock; + + void (*dma_dev_setup)(struct pnv_phb *phb, struct pci_dev *pdev); + void (*fixup_phb)(struct pci_controller *hose); + u32 (*bdfn_to_pe)(struct pnv_phb *phb, struct pci_bus *bus, u32 devfn); + + union { + struct { + struct iommu_table iommu_table; + } p5ioc2; + }; +}; + +extern struct pci_ops pnv_pci_ops; + +extern void pnv_pci_setup_iommu_table(struct iommu_table *tbl, + void *tce_mem, u64 tce_size, + u64 dma_offset); +extern void pnv_pci_init_p5ioc2_hub(struct device_node *np); + + +#endif /* __POWERNV_PCI_H */ diff --git a/arch/powerpc/platforms/powernv/powernv.h b/arch/powerpc/platforms/powernv/powernv.h index 35b7160..8a9df7f 100644 --- a/arch/powerpc/platforms/powernv/powernv.h +++ b/arch/powerpc/platforms/powernv/powernv.h @@ -7,4 +7,10 @@ extern void pnv_smp_init(void); static inline void pnv_smp_init(void) { } #endif +#ifdef CONFIG_PCI +extern void pnv_pci_init(void); +#else +static inline void pnv_pci_init(void) { } +#endif + #endif /* _POWERNV_H */ diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c index f0242f3..467bd4a 100644 --- a/arch/powerpc/platforms/powernv/setup.c +++ b/arch/powerpc/platforms/powernv/setup.c @@ -40,7 +40,8 @@ static void __init pnv_setup_arch(void) /* Initialize SMP */ pnv_smp_init(); - /* XXX PCI */ + /* Setup PCI */ + pnv_pci_init(); /* Setup RTC and NVRAM callbacks */ if (firmware_has_feature(FW_FEATURE_OPAL))