korg/linux
0
0
Fork 0
mirror of https://mirrors.bfsu.edu.cn/git/linux.git synced 2024-11-17 17:24:17 +08:00
Code Issues Packages Projects Releases Wiki Activity
b5cb9ab1a0
linux/arch/powerpc/platforms/powernv/pci.h
Alexey Kardashevskiy b5cb9ab1a0 powerpc/powernv/npu: Enable NVLink pass through 
IBM POWER8 NVlink systems come with Tesla K40-ish GPUs each of which
also has a couple of fast speed links (NVLink). The interface to links
is exposed as an emulated PCI bridge which is included into the same
IOMMU group as the corresponding GPU.

In the kernel, NPUs get a separate PHB of the PNV_PHB_NPU type and a PE
which behave pretty much as the standard IODA2 PHB except NPU PHB has
just a single TVE in the hardware which means it can have either
32bit window or 64bit window or DMA bypass but never two of these.

In order to make these links work when GPU is passed to the guest,
these bridges need to be passed as well; otherwise performance will
degrade.

This implements and exports API to manage NPU state in regard to VFIO;
it replicates iommu_table_group_ops.

This defines a new pnv_pci_ioda2_npu_ops which is assigned to
the IODA2 bridge if there are NPUs for a GPU on the bridge.
The new callbacks call the default IODA2 callbacks plus new NPU API.
This adds a gpe_table_group_to_npe() helper to find NPU PE for the IODA2
table_group, it is not expected to fail as the helper is only called
from the pnv_pci_ioda2_npu_ops.

This does not define NPU-specific .release_ownership() so after
VFIO is finished, DMA on NPU is disabled which is ok as the nvidia
driver sets DMA mask when probing which enable 32 or 64bit DMA on NPU.

This adds a pnv_pci_npu_setup_iommu() helper which adds NPUs to
the GPU group if any found. The helper uses helpers to look for
the "ibm,gpu" property in the device tree which is a phandle of
the corresponding GPU.

This adds an additional loop over PEs in pnv_ioda_setup_dma() as the main
loop skips NPU PEs as they do not have 32bit DMA segments.

As pnv_npu_set_window() and pnv_npu_unset_window() are started being used
by the new IODA2-NPU IOMMU group, this makes the helpers public and
adds the DMA window number parameter.

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-By: Alistair Popple <alistair@popple.id.au>
[mpe: Add pnv_pci_ioda_setup_iommu_api() to fix build with IOMMU_API=n]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2016-05-11 21:54:31 +10:00

236 lines
7.0 KiB
C
Raw Blame History

#ifndef __POWERNV_PCI_H
#define __POWERNV_PCI_H
struct pci_dn;
enum pnv_phb_type {
PNV_PHB_IODA1 = 0,
PNV_PHB_IODA2 = 1,
PNV_PHB_NPU = 2,
};
/* Precise PHB model for error management */
enum pnv_phb_model {
PNV_PHB_MODEL_UNKNOWN,
PNV_PHB_MODEL_P7IOC,
PNV_PHB_MODEL_PHB3,
PNV_PHB_MODEL_NPU,
};
#define PNV_PCI_DIAG_BUF_SIZE 8192
#define PNV_IODA_PE_DEV (1 << 0) /* PE has single PCI device */
#define PNV_IODA_PE_BUS (1 << 1) /* PE has primary PCI bus */
#define PNV_IODA_PE_BUS_ALL (1 << 2) /* PE has subordinate buses */
#define PNV_IODA_PE_MASTER (1 << 3) /* Master PE in compound case */
#define PNV_IODA_PE_SLAVE (1 << 4) /* Slave PE in compound case */
#define PNV_IODA_PE_VF (1 << 5) /* PE for one VF */
/* Data associated with a PE, including IOMMU tracking etc.. */
struct pnv_phb;
struct pnv_ioda_pe {
unsigned long flags;
struct pnv_phb *phb;
/* A PE can be associated with a single device or an
* entire bus (& children). In the former case, pdev
* is populated, in the later case, pbus is.
*/
#ifdef CONFIG_PCI_IOV
struct pci_dev *parent_dev;
#endif
struct pci_dev *pdev;
struct pci_bus *pbus;
/* Effective RID (device RID for a device PE and base bus
* RID with devfn 0 for a bus PE)
*/
unsigned int rid;
/* PE number */
unsigned int pe_number;
/* "Base" iommu table, ie, 4K TCEs, 32-bit DMA */
struct iommu_table_group table_group;
/* 64-bit TCE bypass region */
bool tce_bypass_enabled;
uint64_t tce_bypass_base;
/* MSIs. MVE index is identical for for 32 and 64 bit MSI
* and -1 if not supported. (It's actually identical to the
* PE number)
*/
int mve_number;
/* PEs in compound case */
struct pnv_ioda_pe *master;
struct list_head slaves;
/* Link in list of PE#s */
struct list_head list;
};
#define PNV_PHB_FLAG_EEH (1 << 0)
struct pnv_phb {
struct pci_controller *hose;
enum pnv_phb_type type;
enum pnv_phb_model model;
u64 hub_id;
u64 opal_id;
int flags;
void __iomem *regs;
int initialized;
spinlock_t lock;
#ifdef CONFIG_DEBUG_FS
int has_dbgfs;
struct dentry *dbgfs;
#endif
#ifdef CONFIG_PCI_MSI
unsigned int msi_base;
unsigned int msi32_support;
struct msi_bitmap msi_bmp;
#endif
int (*msi_setup)(struct pnv_phb *phb, struct pci_dev *dev,
unsigned int hwirq, unsigned int virq,
unsigned int is_64, struct msi_msg *msg);
void (*dma_dev_setup)(struct pnv_phb *phb, struct pci_dev *pdev);
void (*fixup_phb)(struct pci_controller *hose);
int (*init_m64)(struct pnv_phb *phb);
void (*reserve_m64_pe)(struct pci_bus *bus,
unsigned long *pe_bitmap, bool all);
struct pnv_ioda_pe *(*pick_m64_pe)(struct pci_bus *bus, bool all);
int (*get_pe_state)(struct pnv_phb *phb, int pe_no);
void (*freeze_pe)(struct pnv_phb *phb, int pe_no);
int (*unfreeze_pe)(struct pnv_phb *phb, int pe_no, int opt);
struct {
/* Global bridge info */
unsigned int total_pe_num;
unsigned int reserved_pe_idx;
/* 32-bit MMIO window */
unsigned int m32_size;
unsigned int m32_segsize;
unsigned int m32_pci_base;
/* 64-bit MMIO window */
unsigned int m64_bar_idx;
unsigned long m64_size;
unsigned long m64_segsize;
unsigned long m64_base;
unsigned long m64_bar_alloc;
/* IO ports */
unsigned int io_size;
unsigned int io_segsize;
unsigned int io_pci_base;
/* PE allocation */
struct mutex pe_alloc_mutex;
unsigned long *pe_alloc;
struct pnv_ioda_pe *pe_array;
/* M32 & IO segment maps */
unsigned int *m64_segmap;
unsigned int *m32_segmap;
unsigned int *io_segmap;
/* DMA32 segment maps - IODA1 only */
unsigned int dma32_count;
unsigned int *dma32_segmap;
/* IRQ chip */
int irq_chip_init;
struct irq_chip irq_chip;
/* Sorted list of used PE's based
* on the sequence of creation
*/
struct list_head pe_list;
struct mutex pe_list_mutex;
/* Reverse map of PEs, will have to extend if
* we are to support more than 256 PEs, indexed
* bus { bus, devfn }
*/
unsigned char pe_rmap[0x10000];
/* TCE cache invalidate registers (physical and
* remapped)
*/
phys_addr_t tce_inval_reg_phys;
__be64 __iomem *tce_inval_reg;
} ioda;
/* PHB and hub status structure */
union {
unsigned char blob[PNV_PCI_DIAG_BUF_SIZE];
struct OpalIoP7IOCPhbErrorData p7ioc;
struct OpalIoPhb3ErrorData phb3;
struct OpalIoP7IOCErrorData hub_diag;
} diag;
};
extern struct pci_ops pnv_pci_ops;
extern int pnv_tce_build(struct iommu_table *tbl, long index, long npages,
unsigned long uaddr, enum dma_data_direction direction,
struct dma_attrs *attrs);
extern void pnv_tce_free(struct iommu_table *tbl, long index, long npages);
extern int pnv_tce_xchg(struct iommu_table *tbl, long index,
unsigned long *hpa, enum dma_data_direction *direction);
extern unsigned long pnv_tce_get(struct iommu_table *tbl, long index);
void pnv_pci_dump_phb_diag_data(struct pci_controller *hose,
unsigned char *log_buff);
int pnv_pci_cfg_read(struct pci_dn *pdn,
int where, int size, u32 *val);
int pnv_pci_cfg_write(struct pci_dn *pdn,
int where, int size, u32 val);
extern struct iommu_table *pnv_pci_table_alloc(int nid);
extern long pnv_pci_link_table_and_group(int node, int num,
struct iommu_table *tbl,
struct iommu_table_group *table_group);
extern void pnv_pci_unlink_table_and_group(struct iommu_table *tbl,
struct iommu_table_group *table_group);
extern void pnv_pci_setup_iommu_table(struct iommu_table *tbl,
void *tce_mem, u64 tce_size,
u64 dma_offset, unsigned page_shift);
extern void pnv_pci_init_ioda_hub(struct device_node *np);
extern void pnv_pci_init_ioda2_phb(struct device_node *np);
extern void pnv_pci_init_npu_phb(struct device_node *np);
extern void pnv_pci_ioda_tce_invalidate(struct iommu_table *tbl,
__be64 *startp, __be64 *endp, bool rm);
extern void pnv_pci_reset_secondary_bus(struct pci_dev *dev);
extern int pnv_eeh_phb_reset(struct pci_controller *hose, int option);
extern void pnv_pci_dma_dev_setup(struct pci_dev *pdev);
extern void pnv_pci_dma_bus_setup(struct pci_bus *bus);
extern int pnv_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type);
extern void pnv_teardown_msi_irqs(struct pci_dev *pdev);
extern void pe_level_printk(const struct pnv_ioda_pe *pe, const char *level,
const char *fmt, ...);
#define pe_err(pe, fmt, ...) \
pe_level_printk(pe, KERN_ERR, fmt, ##__VA_ARGS__)
#define pe_warn(pe, fmt, ...) \
pe_level_printk(pe, KERN_WARNING, fmt, ##__VA_ARGS__)
#define pe_info(pe, fmt, ...) \
pe_level_printk(pe, KERN_INFO, fmt, ##__VA_ARGS__)
/* Nvlink functions */
extern void pnv_npu_try_dma_set_bypass(struct pci_dev *gpdev, bool bypass);
extern void pnv_pci_ioda2_tce_invalidate_entire(struct pnv_phb *phb, bool rm);
extern struct pnv_ioda_pe *pnv_pci_npu_setup_iommu(struct pnv_ioda_pe *npe);
extern long pnv_npu_set_window(struct pnv_ioda_pe *npe, int num,
struct iommu_table *tbl);
extern long pnv_npu_unset_window(struct pnv_ioda_pe *npe, int num);
extern void pnv_npu_take_ownership(struct pnv_ioda_pe *npe);
extern void pnv_npu_release_ownership(struct pnv_ioda_pe *npe);
#endif /* __POWERNV_PCI_H */
Reference in New Issue View Git Blame Copy Permalink