/*
* CDDL HEADER START
*
* The contents of this file are subject to the terms of the
* Common Development and Distribution License (the "License").
* You may not use this file except in compliance with the License.
*
* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
* See the License for the specific language governing permissions
* and limitations under the License.
*
* When distributing Covered Code, include this CDDL HEADER in each
* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
* If applicable, add the following below this CDDL HEADER, with the
* fields enclosed by brackets "[]" replaced with your own identifying
* information: Portions Copyright [yyyy] [name of copyright owner]
*
* CDDL HEADER END
*/
/*
*/
#ifndef _SYS_IB_ADAPTERS_TAVOR_MR_H
#define _SYS_IB_ADAPTERS_TAVOR_MR_H
/*
* Contains all of the prototypes, #defines, and structures necessary
* Specifically it contains #defines, macros, and prototypes for each of
* the IBTF's CI interfaces. In particular each of the prototypes defined
* below is called from a corresponding CI interface routine (as specified
* in the tavor_ci.c file).
*/
#ifdef __cplusplus
extern "C" {
#endif
/*
* The following defines specify the default number of MPT entries and their
* individual entry size. Settings exist for the supported DDR DIMM sizes of
* 128MB and 256MB. If a DIMM greater than 256 is found, then the 256MB
* profile is used. See tavor_cfg.c for more discussion on config profiles.
*
* For manual configuration (not using config profiles), this value is
* controllable through the "tavor_log_num_mpt" configuration variable. To
* override config profile settings the 'tavor_alt_config_enable' configuration
* variable must first be set.
*/
/*
* Minimal configuration value.
*/
/*
* The following defines specify the size of each individual MTT entry and
* the number of MTT entries that make up an MTT segment (TAVOR_MTTSEG_SIZE)
*/
/*
* These define the total number of MTT segments. By default we are setting
* this number of MTT segments (the MTT table size) to 2M segments. This
* default value is used to initialize the "tavor_log_num_mttseg" config
* variable.
* Note: Each segment is currently set to 8 MTT entries (TAVOR_MTTSEG_SIZE).
* This means that we can support up to 16M MTT entries (i.e. "pages").
*/
/*
* Minimal configuration value.
*/
/*
* Macro to round a number of MTT entries to the number of MTT segments.
*/
/*
* This define is used to specify the "MTT page walk version" in the Tavor
* INIT_HCA command.
*/
#define TAVOR_MTT_PG_WALK_VER 0
/*
* This define is the maximum size of a memory region or window (log 2). It is
* set depending on size of the DDR being either 128MB or 256MB. These defines
* are used to initialize the "tavor_log_max_mrw_sz" configuration variable,
* and are proportional to the max MPT size set above.
*/
/*
* Minimal configuration value.
*/
/*
* a given memory regions resources should be freed up. TAVOR_MR_DEREG_ALL
* says what it means, free up all the resources associated with the region.
* TAVOR_MR_DEREG_NO_HW2SW_MPT indicates that it is unnecessary to attempt
* the ownership transfer (from hardware to software) for the given MPT entry.
* And TAVOR_MR_DEREG_NO_HW2SW_MPT_OR_UNBIND indicates that it is not only
* unnecessary to attempt the ownership transfer for MPT, but it is also
* unnecessary to attempt to unbind the memory.
* In general, these last two are specified when tavor_mr_deregister() is
* called from tavor_mr_reregister(), where the MPT ownership transfer or
* memory unbinding may have already been successfully performed.
*/
/*
* The following define is used by tavor_mr_rereg_xlat_helper() to determine
* whether or not a given DMA handle can be reused. If the DMA handle was
* previously initialized for IOMMU bypass mapping, then it can not be reused
* to reregister a region for DDI_DMA_STREAMING access.
*/
!((flags) & IBT_MR_NONCOHERENT))
/*
* The tavor_sw_refcnt_t structure is used internally by the Tavor driver to
* track all the information necessary to manage shared memory regions. Since
* a shared memory region _will_ have its own distinct MPT entry, but will
* _share_ its MTT entries with another region, it is necessary to track the
* number of times a given MTT structure is shared. This ensures that it will
* not be prematurely freed up and that can be destroyed only when it is
* appropriate to do so.
*
* Each tavor_sw_refcnt_t structure contains a lock and a reference count
* variable which are used to track the necessary information.
*
* The following macros (below) are used to manipulate and query the MTT
* reference count parameters. TAVOR_MTT_REFCNT_INIT() is used to initialize
* a newly allocated tavor_sw_refcnt_t struct (setting the "swrc_refcnt" to 1).
* And the TAVOR_MTT_IS_NOT_SHARED() and TAVOR_MTT_IS_SHARED() macros are
* used to query the current status of tavor_sw_refcnt_t struct to determine
* if its "swrc_refcnt" is one or not.
*/
typedef struct tavor_sw_refcnt_s {
/*
* The tavor_bind_info_t structure is used internally by the Tavor driver to
* track all the information necessary to perform the DMA mappings necessary
* for memory registration. It is specifically passed into both the
* tavor_mr_mem_bind() and tavor_mr_mtt_write() functions which perform most
* of the necessary operations for Tavor memory registration.
*
* This structure is used to pass all the information necessary for a call
* to either ddi_dma_addr_bind_handle() or ddi_dma_buf_bind_handle(). Note:
* the fields which need to be valid for each type of binding are slightly
* different and that it indicated by the value in the "bi_type" field. The
* "bi_type" field may be set to either of the following defined values:
* TAVOR_BINDHDL_VADDR (to indicate an "addr" bind) or TAVOR_BINDHDL_BUF (to
* indicate a "buf" bind).
*
* Upon return from tavor_mr_mem_bind(), the tavor_bind_info_t struct will
* have its "bi_dmahdl", "bi_dmacookie", and "bi_cookiecnt" fields filled in.
* It is these values which are of particular interest to the
* tavor_mr_mtt_write() routine (they hold the PCI mapped addresses).
*
* Once initialized and used in this way, the tavor_bind_info_t will not to be
* modified in anyway until it is subsequently passed to tavor_mr_mem_unbind()
* where the memory and resources will be unbound and reclaimed. Note: the
* "bi_free_dmahdl" flag indicated whether the ddi_dma_handle_t should be
* freed as part of the tavor_mr_mem_unbind() operation or whether it will
* be freed later elsewhere.
*/
typedef struct tavor_bind_info_s {
#define TAVOR_BINDHDL_NONE 0
/*
* The tavor_sw_mr_s structure is also referred to using the "tavor_mrhdl_t"
* typedef (see tavor_typedef.h). It encodes all the information necessary
* to track the various resources needed to register, reregister, deregister,
* and perform all the myriad other operations on both memory regions _and_
* memory windows.
*
* A pointer to this structure is returned from many of the IBTF's CI verbs
* interfaces for memory registration.
*
* It contains pointers to the various resources allocated for a memory
* region, i.e. MPT resource, MTT resource, and MTT reference count resource.
* In addition it contains the tavor_bind_info_t struct used for the memory
* bind operation on a given memory region.
*
* It also has a pointers to the associated PD handle, placeholders for access
* flags, memory keys, and suggested page size for the region. It also has
* the necessary backpointer to the resource that corresponds to the structure
* itself. And lastly, it contains a placeholder for a callback which should
* be called on memory region unpinning.
*/
struct tavor_sw_mr_s {
void (*mr_umem_cbfunc)(void *, void *);
void *mr_umem_cbarg1;
void *mr_umem_cbarg2;
};
/*
* The tavor_mr_options_t structure is used in several of the Tavor memory
* registration routines to provide additional option functionality. When
* a NULL pointer is passed in place of a pointer to this struct, it is a
* way of specifying the "default" behavior. Using this structure, however,
* is a way of controlling any extended behavior.
*
* Currently, the only defined "extended" behaviors are for specifying whether
* a given memory region should bypass the PCI IOMMU (TAVOR_BINDMEM_BYPASS)
* or be mapped into the IOMMU (TAVOR_BINDMEM_NORMAL), for specifying whether
* a given ddi_dma_handle_t should be used in the bind operation, and for
* specifying whether a memory registration should attempt to return an IB
* vaddr which is "zero-based" (aids in alignment contraints for QPs).
*
* This defaults today to always bypassing the IOMMU (can be changed by using
* the "tavor_iommu_bypass" configuration variable), to always allocating
* a new dma handle, and to using the virtual address passed in (i.e. not
* "zero-based").
*/
typedef struct tavor_mr_options_s {
#define TAVOR_BINDMEM_BYPASS 0
tavor_rsrc_t *mtt);
#ifdef __cplusplus
}
#endif
#endif /* _SYS_IB_ADAPTERS_TAVOR_MR_H */