diff options
36 files changed, 1147 insertions, 175 deletions
diff --git a/tools/arch/x86/include/asm/amd-ibs.h b/tools/arch/x86/include/asm/amd-ibs.h new file mode 100644 index 000000000000..174e7d83fcbd --- /dev/null +++ b/tools/arch/x86/include/asm/amd-ibs.h @@ -0,0 +1,132 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * From PPR Vol 1 for AMD Family 19h Model 01h B1 + * 55898 Rev 0.35 - Feb 5, 2021 + */ + +#include "msr-index.h" + +/* + * IBS Hardware MSRs + */ + +/* MSR 0xc0011030: IBS Fetch Control */ +union ibs_fetch_ctl { + __u64 val; + struct { + __u64 fetch_maxcnt:16,/* 0-15: instruction fetch max. count */ + fetch_cnt:16, /* 16-31: instruction fetch count */ + fetch_lat:16, /* 32-47: instruction fetch latency */ + fetch_en:1, /* 48: instruction fetch enable */ + fetch_val:1, /* 49: instruction fetch valid */ + fetch_comp:1, /* 50: instruction fetch complete */ + ic_miss:1, /* 51: i-cache miss */ + phy_addr_valid:1,/* 52: physical address valid */ + l1tlb_pgsz:2, /* 53-54: i-cache L1TLB page size + * (needs IbsPhyAddrValid) */ + l1tlb_miss:1, /* 55: i-cache fetch missed in L1TLB */ + l2tlb_miss:1, /* 56: i-cache fetch missed in L2TLB */ + rand_en:1, /* 57: random tagging enable */ + fetch_l2_miss:1,/* 58: L2 miss for sampled fetch + * (needs IbsFetchComp) */ + reserved:5; /* 59-63: reserved */ + }; +}; + +/* MSR 0xc0011033: IBS Execution Control */ +union ibs_op_ctl { + __u64 val; + struct { + __u64 opmaxcnt:16, /* 0-15: periodic op max. count */ + reserved0:1, /* 16: reserved */ + op_en:1, /* 17: op sampling enable */ + op_val:1, /* 18: op sample valid */ + cnt_ctl:1, /* 19: periodic op counter control */ + opmaxcnt_ext:7, /* 20-26: upper 7 bits of periodic op maximum count */ + reserved1:5, /* 27-31: reserved */ + opcurcnt:27, /* 32-58: periodic op counter current count */ + reserved2:5; /* 59-63: reserved */ + }; +}; + +/* MSR 0xc0011035: IBS Op Data 2 */ +union ibs_op_data { + __u64 val; + struct { + __u64 comp_to_ret_ctr:16, /* 0-15: op completion to retire count */ + tag_to_ret_ctr:16, /* 15-31: op tag to retire count */ + reserved1:2, /* 32-33: reserved */ + op_return:1, /* 34: return op */ + op_brn_taken:1, /* 35: taken branch op */ + op_brn_misp:1, /* 36: mispredicted branch op */ + op_brn_ret:1, /* 37: branch op retired */ + op_rip_invalid:1, /* 38: RIP is invalid */ + op_brn_fuse:1, /* 39: fused branch op */ + op_microcode:1, /* 40: microcode op */ + reserved2:23; /* 41-63: reserved */ + }; +}; + +/* MSR 0xc0011036: IBS Op Data 2 */ +union ibs_op_data2 { + __u64 val; + struct { + __u64 data_src:3, /* 0-2: data source */ + reserved0:1, /* 3: reserved */ + rmt_node:1, /* 4: destination node */ + cache_hit_st:1, /* 5: cache hit state */ + reserved1:57; /* 5-63: reserved */ + }; +}; + +/* MSR 0xc0011037: IBS Op Data 3 */ +union ibs_op_data3 { + __u64 val; + struct { + __u64 ld_op:1, /* 0: load op */ + st_op:1, /* 1: store op */ + dc_l1tlb_miss:1, /* 2: data cache L1TLB miss */ + dc_l2tlb_miss:1, /* 3: data cache L2TLB hit in 2M page */ + dc_l1tlb_hit_2m:1, /* 4: data cache L1TLB hit in 2M page */ + dc_l1tlb_hit_1g:1, /* 5: data cache L1TLB hit in 1G page */ + dc_l2tlb_hit_2m:1, /* 6: data cache L2TLB hit in 2M page */ + dc_miss:1, /* 7: data cache miss */ + dc_mis_acc:1, /* 8: misaligned access */ + reserved:4, /* 9-12: reserved */ + dc_wc_mem_acc:1, /* 13: write combining memory access */ + dc_uc_mem_acc:1, /* 14: uncacheable memory access */ + dc_locked_op:1, /* 15: locked operation */ + dc_miss_no_mab_alloc:1, /* 16: DC miss with no MAB allocated */ + dc_lin_addr_valid:1, /* 17: data cache linear address valid */ + dc_phy_addr_valid:1, /* 18: data cache physical address valid */ + dc_l2_tlb_hit_1g:1, /* 19: data cache L2 hit in 1GB page */ + l2_miss:1, /* 20: L2 cache miss */ + sw_pf:1, /* 21: software prefetch */ + op_mem_width:4, /* 22-25: load/store size in bytes */ + op_dc_miss_open_mem_reqs:6, /* 26-31: outstanding mem reqs on DC fill */ + dc_miss_lat:16, /* 32-47: data cache miss latency */ + tlb_refill_lat:16; /* 48-63: L1 TLB refill latency */ + }; +}; + +/* MSR 0xc001103c: IBS Fetch Control Extended */ +union ic_ibs_extd_ctl { + __u64 val; + struct { + __u64 itlb_refill_lat:16, /* 0-15: ITLB Refill latency for sampled fetch */ + reserved:48; /* 16-63: reserved */ + }; +}; + +/* + * IBS driver related + */ + +struct perf_ibs_data { + u32 size; + union { + u32 data[0]; /* data buffer starts here */ + u32 caps; + }; + u64 regs[MSR_AMD64_IBS_REG_COUNT_MAX]; +}; diff --git a/tools/arch/x86/include/uapi/asm/kvm.h b/tools/arch/x86/include/uapi/asm/kvm.h index a6c327f8ad9e..2ef1f6513c68 100644 --- a/tools/arch/x86/include/uapi/asm/kvm.h +++ b/tools/arch/x86/include/uapi/asm/kvm.h @@ -295,6 +295,7 @@ struct kvm_debug_exit_arch { #define KVM_GUESTDBG_USE_HW_BP 0x00020000 #define KVM_GUESTDBG_INJECT_DB 0x00040000 #define KVM_GUESTDBG_INJECT_BP 0x00080000 +#define KVM_GUESTDBG_BLOCKIRQ 0x00100000 /* for KVM_SET_GUEST_DEBUG */ struct kvm_guest_debug_arch { diff --git a/tools/include/uapi/asm-generic/unistd.h b/tools/include/uapi/asm-generic/unistd.h index a9d6fcd95f42..1c5fb86d455a 100644 --- a/tools/include/uapi/asm-generic/unistd.h +++ b/tools/include/uapi/asm-generic/unistd.h @@ -673,15 +673,15 @@ __SYSCALL(__NR_madvise, sys_madvise) #define __NR_remap_file_pages 234 __SYSCALL(__NR_remap_file_pages, sys_remap_file_pages) #define __NR_mbind 235 -__SC_COMP(__NR_mbind, sys_mbind, compat_sys_mbind) +__SYSCALL(__NR_mbind, sys_mbind) #define __NR_get_mempolicy 236 -__SC_COMP(__NR_get_mempolicy, sys_get_mempolicy, compat_sys_get_mempolicy) +__SYSCALL(__NR_get_mempolicy, sys_get_mempolicy) #define __NR_set_mempolicy 237 -__SC_COMP(__NR_set_mempolicy, sys_set_mempolicy, compat_sys_set_mempolicy) +__SYSCALL(__NR_set_mempolicy, sys_set_mempolicy) #define __NR_migrate_pages 238 -__SC_COMP(__NR_migrate_pages, sys_migrate_pages, compat_sys_migrate_pages) +__SYSCALL(__NR_migrate_pages, sys_migrate_pages) #define __NR_move_pages 239 -__SC_COMP(__NR_move_pages, sys_move_pages, compat_sys_move_pages) +__SYSCALL(__NR_move_pages, sys_move_pages) #endif #define __NR_rt_tgsigqueueinfo 240 @@ -877,9 +877,11 @@ __SYSCALL(__NR_landlock_restrict_self, sys_landlock_restrict_self) #define __NR_memfd_secret 447 __SYSCALL(__NR_memfd_secret, sys_memfd_secret) #endif +#define __NR_process_mrelease 448 +__SYSCALL(__NR_process_mrelease, sys_process_mrelease) #undef __NR_syscalls -#define __NR_syscalls 448 +#define __NR_syscalls 449 /* * 32 bit systems traditionally used different diff --git a/tools/include/uapi/drm/drm.h b/tools/include/uapi/drm/drm.h index d043752a74cf..3b810b53ba8b 100644 --- a/tools/include/uapi/drm/drm.h +++ b/tools/include/uapi/drm/drm.h @@ -635,8 +635,8 @@ struct drm_gem_open { /** * DRM_CAP_VBLANK_HIGH_CRTC * - * If set to 1, the kernel supports specifying a CRTC index in the high bits of - * &drm_wait_vblank_request.type. + * If set to 1, the kernel supports specifying a :ref:`CRTC index<crtc_index>` + * in the high bits of &drm_wait_vblank_request.type. * * Starting kernel version 2.6.39, this capability is always set to 1. */ @@ -1050,6 +1050,16 @@ extern "C" { #define DRM_IOCTL_MODE_GETPROPBLOB DRM_IOWR(0xAC, struct drm_mode_get_blob) #define DRM_IOCTL_MODE_GETFB DRM_IOWR(0xAD, struct drm_mode_fb_cmd) #define DRM_IOCTL_MODE_ADDFB DRM_IOWR(0xAE, struct drm_mode_fb_cmd) +/** + * DRM_IOCTL_MODE_RMFB - Remove a framebuffer. + * + * This removes a framebuffer previously added via ADDFB/ADDFB2. The IOCTL + * argument is a framebuffer object ID. + * + * Warning: removing a framebuffer currently in-use on an enabled plane will + * disable that plane. The CRTC the plane is linked to may also be disabled + * (depending on driver capabilities). + */ #define DRM_IOCTL_MODE_RMFB DRM_IOWR(0xAF, unsigned int) #define DRM_IOCTL_MODE_PAGE_FLIP DRM_IOWR(0xB0, struct drm_mode_crtc_page_flip) #define DRM_IOCTL_MODE_DIRTYFB DRM_IOWR(0xB1, struct drm_mode_fb_dirty_cmd) diff --git a/tools/include/uapi/drm/i915_drm.h b/tools/include/uapi/drm/i915_drm.h index c2c7759b7d2e..bde5860b3686 100644 --- a/tools/include/uapi/drm/i915_drm.h +++ b/tools/include/uapi/drm/i915_drm.h @@ -572,6 +572,15 @@ typedef struct drm_i915_irq_wait { #define I915_SCHEDULER_CAP_PREEMPTION (1ul << 2) #define I915_SCHEDULER_CAP_SEMAPHORES (1ul << 3) #define I915_SCHEDULER_CAP_ENGINE_BUSY_STATS (1ul << 4) +/* + * Indicates the 2k user priority levels are statically mapped into 3 buckets as + * follows: + * + * -1k to -1 Low priority + * 0 Normal priority + * 1 to 1k Highest priority + */ +#define I915_SCHEDULER_CAP_STATIC_PRIORITY_MAP (1ul << 5) #define I915_PARAM_HUC_STATUS 42 @@ -674,6 +683,9 @@ typedef struct drm_i915_irq_wait { */ #define I915_PARAM_HAS_EXEC_TIMELINE_FENCES 55 +/* Query if the kernel supports the I915_USERPTR_PROBE flag. */ +#define I915_PARAM_HAS_USERPTR_PROBE 56 + /* Must be kept compact -- no holes and well documented */ typedef struct drm_i915_getparam { @@ -849,45 +861,113 @@ struct drm_i915_gem_mmap_gtt { __u64 offset; }; +/** + * struct drm_i915_gem_mmap_offset - Retrieve an offset so we can mmap this buffer object. + * + * This struct is passed as argument to the `DRM_IOCTL_I915_GEM_MMAP_OFFSET` ioctl, + * and is used to retrieve the fake offset to mmap an object specified by &handle. + * + * The legacy way of using `DRM_IOCTL_I915_GEM_MMAP` is removed on gen12+. + * `DRM_IOCTL_I915_GEM_MMAP_GTT` is an older supported alias to this struct, but will behave + * as setting the &extensions to 0, and &flags to `I915_MMAP_OFFSET_GTT`. + */ struct drm_i915_gem_mmap_offset { - /** Handle for the object being mapped. */ + /** @handle: Handle for the object being mapped. */ __u32 handle; + /** @pad: Must be zero */ __u32 pad; /** - * Fake offset to use for subsequent mmap call + * @offset: The fake offset to use for subsequent mmap call * * This is a fixed-size type for 32/64 compatibility. */ __u64 offset; /** - * Flags for extended behaviour. + * @flags: Flags for extended behaviour. + * + * It is mandatory that one of the `MMAP_OFFSET` types + * should be included: * - * It is mandatory that one of the MMAP_OFFSET types - * (GTT, WC, WB, UC, etc) should be included. + * - `I915_MMAP_OFFSET_GTT`: Use mmap with the object bound to GTT. (Write-Combined) + * - `I915_MMAP_OFFSET_WC`: Use Write-Combined caching. + * - `I915_MMAP_OFFSET_WB`: Use Write-Back caching. + * - `I915_MMAP_OFFSET_FIXED`: Use object placement to determine caching. + * + * On devices with local memory `I915_MMAP_OFFSET_FIXED` is the only valid + * type. On devices without local memory, this caching mode is invalid. + * + * As caching mode when specifying `I915_MMAP_OFFSET_FIXED`, WC or WB will + * be used, depending on the object placement on creation. WB will be used + * when the object can only exist in system memory, WC otherwise. */ __u64 flags; -#define I915_MMAP_OFFSET_GTT 0 -#define I915_MMAP_OFFSET_WC 1 -#define I915_MMAP_OFFSET_WB 2 -#define I915_MMAP_OFFSET_UC 3 - /* - * Zero-terminated chain of extensions. +#define I915_MMAP_OFFSET_GTT 0 +#define I915_MMAP_OFFSET_WC 1 +#define I915_MMAP_OFFSET_WB 2 +#define I915_MMAP_OFFSET_UC 3 +#define I915_MMAP_OFFSET_FIXED 4 + + /** + * @extensions: Zero-terminated chain of extensions. * * No current extensions defined; mbz. */ __u64 extensions; }; +/** + * struct drm_i915_gem_set_domain - Adjust the objects write or read domain, in + * preparation for accessing the pages via some CPU domain. + * + * Specifying a new write or read domain will flush the object out of the + * previous domain(if required), before then updating the objects domain + * tracking with the new domain. + * + * Note this might involve waiting for the object first if it is still active on + * the GPU. + * + * Supported values for @read_domains and @write_domain: + * + * - I915_GEM_DOMAIN_WC: Uncached write-combined domain + * - I915_GEM_DOMAIN_CPU: CPU cache domain + * - I915_GEM_DOMAIN_GTT: Mappable aperture domain + * + * All other domains are rejected. + * + * Note that for discrete, starting from DG1, this is no longer supported, and + * is instead rejected. On such platforms the CPU domain is effectively static, + * where we also only support a single &drm_i915_gem_mmap_offset cache mode, + * which can't be set explicitly and instead depends on the object placements, + * as per the below. + * + * Implicit caching rules, starting from DG1: + * + * - If any of the object placements (see &drm_i915_gem_create_ext_memory_regions) + * contain I915_MEMORY_CLASS_DEVICE then the object will be allocated and + * mapped as write-combined only. + * + * - Everything else is always allocated and mapped as write-back, with the + * guarantee that everything is also coherent with the GPU. + * + * Note that this is likely to change in the future again, where we might need + * more flexibility on future devices, so making this all explicit as part of a + * new &drm_i915_gem_create_ext extension is probable. + */ struct drm_i915_gem_set_domain { - /** Handle for the object */ + /** @handle: Handle for the object. */ __u32 handle; - /** New read domains */ + /** @read_domains: New read domains. */ __u32 read_domains; - /** New write domain */ + /** + * @write_domain: New write domain. + * + * Note that having something in the write domain implies it's in the + * read domain, and only that read domain. + */ __u32 write_domain; }; @@ -1348,12 +1428,11 @@ struct drm_i915_gem_busy { * reading from the object simultaneously. * * The value of each engine class is the same as specified in the - * I915_CONTEXT_SET_ENGINES parameter and via perf, i.e. + * I915_CONTEXT_PARAM_ENGINES context parameter and via perf, i.e. * I915_ENGINE_CLASS_RENDER, I915_ENGINE_CLASS_COPY, etc. - * reported as active itself. Some hardware may have parallel - * execution engines, e.g. multiple media engines, which are - * mapped to the same class identifier and so are not separately - * reported for busyness. + * Some hardware may have parallel execution engines, e.g. multiple + * media engines, which are mapped to the same class identifier and so + * are not separately reported for busyness. * * Caveat emptor: * Only the boolean result of this query is reliable; that is whether @@ -1364,43 +1443,79 @@ struct drm_i915_gem_busy { }; /** - * I915_CACHING_NONE - * - * GPU access is not coherent with cpu caches. Default for machines without an - * LLC. - */ -#define I915_CACHING_NONE 0 -/** - * I915_CACHING_CACHED - * - * GPU access is coherent with cpu caches and furthermore the data is cached in - * last-level caches shared between cpu cores and the gpu GT. Default on - * machines with HAS_LLC. + * struct drm_i915_gem_caching - Set or get the caching for given object + * handle. + * + * Allow userspace to control the GTT caching bits for a given object when the + * object is later mapped through the ppGTT(or GGTT on older platforms lacking + * ppGTT support, or if the object is used for scanout). Note that this might + * require unbinding the object from the GTT first, if its current caching value + * doesn't match. + * + * Note that this all changes on discrete platforms, starting from DG1, the + * set/get caching is no longer supported, and is now rejected. Instead the CPU + * caching attributes(WB vs WC) will become an immutable creation time property + * for the object, along with the GTT caching level. For now we don't expose any + * new uAPI for this, instead on DG1 this is all implicit, although this largely + * shouldn't matter since DG1 is coherent by default(without any way of + * controlling it). + * + * Implicit caching rules, starting from DG1: + * + * - If any of the object placements (see &drm_i915_gem_create_ext_memory_regions) + * contain I915_MEMORY_CLASS_DEVICE then the object will be allocated and + * mapped as write-combined only. + * + * - Everything else is always allocated and mapped as write-back, with the + * guarantee that everything is also coherent with the GPU. + * + * Note that this is likely to change in the future again, where we might need + * more flexibility on future devices, so making this all explicit as part of a + * new &drm_i915_gem_create_ext extension is probable. + * + * Side note: Part of the reason for this is that changing the at-allocation-time CPU + * caching attributes for the pages might be required(and is expensive) if we + * need to then CPU map the pages later with different caching attributes. This + * inconsistent caching behaviour, while supported on x86, is not universally + * supported on other architectures. So for simplicity we opt for setting + * everything at creation time, whilst also making it immutable, on discrete + * platforms. */ -#define I915_CACHING_CACHED 1 -/** - * I915_CACHING_DISPLAY - * - * Special GPU caching mode which is coherent with the scanout engines. - * Transparently falls back to I915_CACHING_NONE on platforms where no special - * cache mode (like write-through or gfdt flushing) is available. The kernel - * automatically sets this mode when using a buffer as a scanout target. - * Userspace can manually set this mode to avoid a costly stall and clflush in - * the hotpath of drawing the first frame. - */ -#define I915_CACHING_DISPLAY 2 - struct drm_i915_gem_caching { /** - * Handle of the buffer to set/get the caching level of. */ + * @handle: Handle of the buffer to set/get the caching level. + */ __u32 handle; /** - * Cacheing level to apply or return value + * @caching: The GTT caching level to apply or possible return value. + * + * The supported @caching values: * - * bits0-15 are for generic caching control (i.e. the above defined - * values). bits16-31 are reserved for platform-specific variations - * (e.g. l3$ caching on gen7). */ + * I915_CACHING_NONE: + * + * GPU access is not coherent with CPU caches. Default for machines + * without an LLC. This means manual flushing might be needed, if we + * want GPU access to be coherent. + * + * I915_CACHING_CACHED: + * + * GPU access is coherent with CPU caches and furthermore the data is + * cached in last-level caches shared between CPU cores and the GPU GT. + * + * I915_CACHING_DISPLAY: + * + * Special GPU caching mode which is coherent with the scanout engines. + * Transparently falls back to I915_CACHING_NONE on platforms where no + * special cache mode (like write-through or gfdt flushing) is + * available. The kernel automatically sets this mode when using a + * buffer as a scanout target. Userspace can manually set this mode to + * avoid a costly stall and clflush in the hotpath of drawing the first + * frame. + */ +#define I915_CACHING_NONE 0 +#define I915_CACHING_CACHED 1 +#define I915_CACHING_DISPLAY 2 __u32 caching; }; @@ -1639,6 +1754,10 @@ struct drm_i915_gem_context_param { __u32 size; __u64 param; #define I915_CONTEXT_PARAM_BAN_PERIOD 0x1 +/* I915_CONTEXT_PARAM_NO_ZEROMAP has been removed. On the off chance + * someone somewhere has attempted to use it, never re-use this context + * param number. + */ #define I915_CONTEXT_PARAM_NO_ZEROMAP 0x2 #define I915_CONTEXT_PARAM_GTT_SIZE 0x3 #define I915_CONTEXT_PARAM_NO_ERROR_CAPTURE 0x4 @@ -1723,24 +1842,8 @@ struct drm_i915_gem_context_param { */ #define I915_CONTEXT_PARAM_PERSISTENCE 0xb -/* - * I915_CONTEXT_PARAM_RINGSIZE: - * - * Sets the size of the CS ringbuffer to use for logical ring contexts. This - * applies a limit of how many batches can be queued to HW before the caller - * is blocked due to lack of space for more commands. - * - * Only reliably possible to be set prior to first use, i.e. during - * construction. At any later point, the current execution must be flushed as - * the ring can only be changed while the context is idle. Note, the ringsize - * can be specified as a constructor property, see - * I915_CONTEXT_CREATE_EXT_SETPARAM, but can also be set later if required. - * - * Only applies to the current set of engine and lost when those engines - * are replaced by a new mapping (see I915_CONTEXT_PARAM_ENGINES). - * - * Must be between 4 - 512 KiB, in intervals of page size [4 KiB]. - * Default is 16 KiB. +/* This API has been removed. On the off chance someone somewhere has + * attempted to use it, never re-use this context param number. */ #define I915_CONTEXT_PARAM_RINGSIZE 0xc /* Must be kept compact -- no holes and well documented */ @@ -1807,6 +1910,69 @@ struct drm_i915_gem_context_param_sseu { __u32 rsvd; }; +/** + * DOC: Virtual Engine uAPI + * + * Virtual engine is a concept where userspace is able to configure a set of + * physical engines, submit a batch buffer, and let the driver execute it on any + * engine from the set as it sees fit. + * + * This is primarily useful on parts which have multiple instances of a same + * class engine, like for example GT3+ Skylake parts with their two VCS engines. + * + * For instance userspace can enumerate all engines of a certain class using the + * previously described `Engine Discovery uAPI`_. After that userspace can + * create a GEM context with a placeholder slot for the virtual engine (using + * `I915_ENGINE_CLASS_INVALID` and `I915_ENGINE_CLASS_INVALID_NONE` for class + * and instance respectively) and finally using the + * `I915_CONTEXT_ENGINES_EXT_LOAD_BALANCE` extension place a virtual engine in + * the same reserved slot. + * + * Example of creating a virtual engine and submitting a batch buffer to it: + * + * .. code-block:: C + * + * I915_DEFINE_CONTEXT_ENGINES_LOAD_BALANCE(virtual, 2) = { + * .base.name = I915_CONTEXT_ENGINES_EXT_LOAD_BALANCE, + * .engine_index = 0, // Place this virtual engine into engine map slot 0 + * .num_siblings = 2, + * .engines = { { I915_ENGINE_CLASS_VIDEO, 0 }, + * { I915_ENGINE_CLASS_VIDEO, 1 }, }, + * }; + * I915_DEFINE_CONTEXT_PARAM_ENGINES(engines, 1) = { + * .engines = { { I915_ENGINE_CLASS_INVALID, + * I915_ENGINE_CLASS_INVALID_NONE } }, + * .extensions = to_user_pointer(&virtual), // Chains after load_balance extension + * }; + * struct drm_i915_gem_context_create_ext_setparam p_engines = { + * .base = { + * .name = I915_CONTEXT_CREATE_EXT_SETPARAM, + * }, + * .param = { + * .param = I915_CONTEXT_PARAM_ENGINES, + * .value = to_user_pointer(&engines), + * .size = sizeof(engines), + * }, + * }; + * struct drm_i915_gem_context_create_ext create = { + * .flags = I915_CONTEXT_CREATE_FLAGS_USE_EXTENSIONS, + * .extensions = to_user_pointer(&p_engines); + * }; + * + * ctx_id = gem_context_create_ext(drm_fd, &create); + * + * // Now we have created a GEM context with its engine map containing a + * // single virtual engine. Submissions to this slot can go either to + * // vcs0 or vcs1, depending on the load balancing algorithm used inside + * // the driver. The load balancing is dynamic from one batch buffer to + * // another and transparent to userspace. + * + * ... + * execbuf.rsvd1 = ctx_id; + * execbuf.flags = 0; // Submits to index 0 which is the virtual engine + * gem_execbuf(drm_fd, &execbuf); + */ + /* * i915_context_engines_load_balance: * @@ -1883,6 +2049,61 @@ struct i915_context_engines_bond { struct i915_engine_class_instance engines[N__]; \ } __attribute__((packed)) name__ +/** + * DOC: Context Engine Map uAPI + * + * Context engine map is a new way of addressing engines when submitting batch- + * buffers, replacing the existing way of using identifiers like `I915_EXEC_BLT` + * inside the flags field of `struct drm_i915_gem_execbuffer2`. + * + * To use it created GEM contexts need to be configured with a list of engines + * the user is intending to submit to. This is accomplished using the + * `I915_CONTEXT_PARAM_ENGINES` parameter and `struct + * i915_context_param_engines`. + * + * For such contexts the `I915_EXEC_RING_MASK` field becomes an index into the + * configured map. + * + * Example of creating such context and submitting against it: + * + * .. code-block:: C + * + * I915_DEFINE_CONTEXT_PARAM_ENGINES(engines, 2) = { + * .engines = { { I915_ENGINE_CLASS_RENDER, 0 }, + * { I915_ENGINE_CLASS_COPY, 0 } } + * }; + * struct drm_i915_gem_context_create_ext_setparam p_engines = { + * .base = { + * .name = I915_CONTEXT_CREATE_EXT_SETPARAM, + * }, + * .param = { + * .param = I915_CONTEXT_PARAM_ENGINES, + * .value = to_user_pointer(&engines), + * .size = sizeof(engines), + * }, + * }; + * struct drm_i915_gem_context_create_ext create = { + * .flags = I915_CONTEXT_CREATE_FLAGS_USE_EXTENSIONS, + * .extensions = to_user_pointer(&p_engines); + * }; + * + * ctx_id = gem_context_create_ext(drm_fd, &create); + * + * // We have now created a GEM context with two engines in the map: + * // Index 0 points to rcs0 while index 1 points to bcs0. Other engines + * // will not be accessible from this context. + * + * ... + * execbuf.rsvd1 = ctx_id; + * execbuf.flags = 0; // Submits to index 0, which is rcs0 for this context + * gem_execbuf(drm_fd, &execbuf); + * + * ... + * execbuf.rsvd1 = ctx_id; + * execbuf.flags = 1; // Submits to index 0, which is bcs0 for this context + * gem_execbuf(drm_fd, &execbuf); + */ + struct i915_context_param_engines { __u64 extensions; /* linked chain of extension blocks, 0 terminates */ #define I915_CONTEXT_ENGINES_EXT_LOAD_BALANCE 0 /* see i915_context_engines_load_balance */ @@ -1901,20 +2122,10 @@ struct drm_i915_gem_context_create_ext_setparam { struct drm_i915_gem_context_param param; }; -struct drm_i915_gem_context_create_ext_clone { +/* This API has been removed. On the off chance someone somewhere has + * attempted to use it, never re-use this extension number. + */ #define I915_CONTEXT_CREATE_EXT_CLONE 1 - struct i915_user_extension base; - __u32 clone_id; - __u32 flags; -#define I915_CONTEXT_CLONE_ENGINES (1u << 0) -#define I915_CONTEXT_CLONE_FLAGS (1u << 1) -#define I915_CONTEXT_CLONE_SCHEDATTR (1u << 2) -#define I915_CONTEXT_CLONE_SSEU (1u << 3) -#define I915_CONTEXT_CLONE_TIMELINE (1u << 4) -#define I915_CONTEXT_CLONE_VM (1u << 5) -#define I915_CONTEXT_CLONE_UNKNOWN -(I915_CONTEXT_CLONE_VM << 1) - __u64 rsvd; -}; struct drm_i915_gem_context_destroy { __u32 ctx_id; @@ -1986,14 +2197,69 @@ struct drm_i915_reset_stats { __u32 pad; }; +/** + * struct drm_i915_gem_userptr - Create GEM object from user allocated memory. + * + * Userptr objects have several restrictions on what ioctls can be used with the + * object handle. + */ struct drm_i915_gem_userptr { + /** + * @user_ptr: The pointer to the allocated memory. + * + * Needs to be aligned to PAGE_SIZE. + */ __u64 user_ptr; + + /** + * @user_size: + * + * The size in bytes for the allocated memory. This will also become the + * object size. + * + * Needs to be aligned to PAGE_SIZE, and should be at least PAGE_SIZE, + * or larger. + */ __u64 user_size; + + /** + * @flags: + * + * Supported flags: + * + * I915_USERPTR_READ_ONLY: + * + * Mark the object as readonly, this also means GPU access can only be + * readonly. This is only supported on HW which supports readonly access + * through the GTT. If the HW can't support readonly access, an error is + * returned. + * + * I915_USERPTR_PROBE: + * + * Probe the provided @user_ptr range and validate that the @user_ptr is + * indeed pointing to normal memory and that the range is also vali |