{sphinx.addnodesdocument)}( rawsourcechildren]( translations LanguagesNode)}(hhh](h pending_xref)}(hhh]docutils.nodesTextChinese (Simplified)}parenthsba attributes}(ids]classes]names]dupnames]backrefs] refdomainstdreftypedoc reftarget*/translations/zh_CN/gpu/rfc/i915_small_barmodnameN classnameN refexplicitutagnamehhh ubh)}(hhh]hChinese (Traditional)}hh2sbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget*/translations/zh_TW/gpu/rfc/i915_small_barmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hItalian}hhFsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget*/translations/it_IT/gpu/rfc/i915_small_barmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hJapanese}hhZsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget*/translations/ja_JP/gpu/rfc/i915_small_barmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hKorean}hhnsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget*/translations/ko_KR/gpu/rfc/i915_small_barmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hSpanish}hhsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget*/translations/sp_SP/gpu/rfc/i915_small_barmodnameN classnameN refexplicituh1hhh ubeh}(h]h ]h"]h$]h&]current_languageEnglishuh1h hh _documenthsourceNlineNubhsection)}(hhh](htitle)}(hI915 Small BAR RFC Sectionh]hI915 Small BAR RFC Section}(hhhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhhhD/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar.rsthKubh paragraph)}(hXgStarting from DG2 we will have resizable BAR support for device local-memory(i.e I915_MEMORY_CLASS_DEVICE), but in some cases the final BAR size might still be smaller than the total probed_size. In such cases, only some subset of I915_MEMORY_CLASS_DEVICE will be CPU accessible(for example the first 256M), while the remainder is only accessible via the GPU.h]hXgStarting from DG2 we will have resizable BAR support for device local-memory(i.e I915_MEMORY_CLASS_DEVICE), but in some cases the final BAR size might still be smaller than the total probed_size. In such cases, only some subset of I915_MEMORY_CLASS_DEVICE will be CPU accessible(for example the first 256M), while the remainder is only accessible via the GPU.}(hhhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhhhhubh)}(hhh](h)}(h.I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS flagh]h.I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS flag}(hhhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhhhhhK ubh)}(hXNew gem_create_ext flag to tell the kernel that a BO will require CPU access. This becomes important when placing an object in I915_MEMORY_CLASS_DEVICE, where underneath the device has a small BAR, meaning only some portion of it is CPU accessible. Without this flag the kernel will assume that CPU access is not required, and prioritize using the non-CPU visible portion of I915_MEMORY_CLASS_DEVICE.h]hXNew gem_create_ext flag to tell the kernel that a BO will require CPU access. This becomes important when placing an object in I915_MEMORY_CLASS_DEVICE, where underneath the device has a small BAR, meaning only some portion of it is CPU accessible. Without this flag the kernel will assume that CPU access is not required, and prioritize using the non-CPU visible portion of I915_MEMORY_CLASS_DEVICE.}(hhhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK hhhhubhindex)}(hhh]h}(h]h ]h"]h$]h&]entries](single$__drm_i915_gem_create_ext (C struct)c.__drm_i915_gem_create_exthNtauh1hhhhhhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhNubhdesc)}(hhh](hdesc_signature)}(h__drm_i915_gem_create_exth]hdesc_signature_line)}(h struct __drm_i915_gem_create_exth](hdesc_sig_keyword)}(hstructh]hstruct}(hj hhhNhNubah}(h]h ]kah"]h$]h&]uh1j hjhhhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKubhdesc_sig_space)}(h h]h }(hjhhhNhNubah}(h]h ]wah"]h$]h&]uh1jhjhhhjhKubh desc_name)}(h__drm_i915_gem_create_exth]h desc_sig_name)}(hjh]h__drm_i915_gem_create_ext}(hj4hhhNhNubah}(h]h ]nah"]h$]h&]uh1j2hj.ubah}(h]h ](sig-namedescnameeh"]h$]h&] xml:spacepreserveuh1j,hjhhhjhKubeh}(h]h ]h"]h$]h&]jJjK add_permalinkuh1jsphinx_line_type declaratorhhhhhjhKubah}(h]hah ](sig sig-objecteh"]h$]h&] is_multiline _toc_parts) _toc_namehuh1hhjhKhhhhubh desc_content)}(hhh]h)}(h]Existing gem_create behaviour, with added extension support using struct i915_user_extension.h]h]Existing gem_create behaviour, with added extension support using struct i915_user_extension.}(hjehhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhK^hjbhhubah}(h]h ]h"]h$]h&]uh1j`hhhhhjhKubeh}(h]h ](cstructeh"]h$]h&]domainj}objtypej~desctypej~noindex noindexentrynocontentsentryuh1hhhhhhhhNubh container)}(hX **Definition**:: struct __drm_i915_gem_create_ext { __u64 size; __u32 handle; #define I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS (1 << 0); __u32 flags; #define I915_GEM_CREATE_EXT_MEMORY_REGIONS 0; #define I915_GEM_CREATE_EXT_PROTECTED_CONTENT 1; __u64 extensions; }; **Members** ``size`` Requested size for the object. The (page-aligned) allocated size for the object will be returned. Note that for some devices we have might have further minimum page-size restrictions (larger than 4K), like for device local-memory. However in general the final size here should always reflect any rounding up, if for example using the I915_GEM_CREATE_EXT_MEMORY_REGIONS extension to place the object in device local-memory. The kernel will always select the largest minimum page-size for the set of possible placements as the value to use when rounding up the **size**. ``handle`` Returned handle for the object. Object handles are nonzero. ``flags`` Optional flags. Supported values: I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS - Signal to the kernel that the object will need to be accessed via the CPU. Only valid when placing objects in I915_MEMORY_CLASS_DEVICE, and only strictly required on configurations where some subset of the device memory is directly visible/mappable through the CPU (which we also call small BAR), like on some DG2+ systems. Note that this is quite undesirable, but due to various factors like the client CPU, BIOS etc it's something we can expect to see in the wild. See :c:type:`__drm_i915_memory_region_info.probed_cpu_visible_size <__drm_i915_memory_region_info>` for how to determine if this system applies. Note that one of the placements MUST be I915_MEMORY_CLASS_SYSTEM, to ensure the kernel can always spill the allocation to system memory, if the object can't be allocated in the mappable part of I915_MEMORY_CLASS_DEVICE. Also note that since the kernel only supports flat-CCS on objects that can *only* be placed in I915_MEMORY_CLASS_DEVICE, we therefore don't support I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS together with flat-CCS. Without this hint, the kernel will assume that non-mappable I915_MEMORY_CLASS_DEVICE is preferred for this object. Note that the kernel can still migrate the object to the mappable part, as a last resort, if userspace ever CPU faults this object, but this might be expensive, and so ideally should be avoided. On older kernels which lack the relevant small-bar uAPI support (see also :c:type:`__drm_i915_memory_region_info.probed_cpu_visible_size <__drm_i915_memory_region_info>`), usage of the flag will result in an error, but it should NEVER be possible to end up with a small BAR configuration, assuming we can also successfully load the i915 kernel module. In such cases the entire I915_MEMORY_CLASS_DEVICE region will be CPU accessible, and as such there are zero restrictions on where the object can be placed. ``extensions`` The chain of extensions to apply to this object. This will be useful in the future when we need to support several different extensions, and we need to apply more than one when creating the object. See struct i915_user_extension. If we don't supply any extensions then we get the same old gem_create behaviour. For I915_GEM_CREATE_EXT_MEMORY_REGIONS usage see struct drm_i915_gem_create_ext_memory_regions. For I915_GEM_CREATE_EXT_PROTECTED_CONTENT usage see struct drm_i915_gem_create_ext_protected_content.h](h)}(h**Definition**::h](hstrong)}(h**Definition**h]h Definition}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKbhjubh literal_block)}(hX struct __drm_i915_gem_create_ext { __u64 size; __u32 handle; #define I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS (1 << 0); __u32 flags; #define I915_GEM_CREATE_EXT_MEMORY_REGIONS 0; #define I915_GEM_CREATE_EXT_PROTECTED_CONTENT 1; __u64 extensions; };h]hX struct __drm_i915_gem_create_ext { __u64 size; __u32 handle; #define I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS (1 << 0); __u32 flags; #define I915_GEM_CREATE_EXT_MEMORY_REGIONS 0; #define I915_GEM_CREATE_EXT_PROTECTED_CONTENT 1; __u64 extensions; };}hjsbah}(h]h ]h"]h$]h&]jJjKuh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKdhjubh)}(h **Members**h]j)}(hjh]hMembers}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKnhjubhdefinition_list)}(hhh](hdefinition_list_item)}(hXD``size`` Requested size for the object. The (page-aligned) allocated size for the object will be returned. Note that for some devices we have might have further minimum page-size restrictions (larger than 4K), like for device local-memory. However in general the final size here should always reflect any rounding up, if for example using the I915_GEM_CREATE_EXT_MEMORY_REGIONS extension to place the object in device local-memory. The kernel will always select the largest minimum page-size for the set of possible placements as the value to use when rounding up the **size**. h](hterm)}(h``size``h]hliteral)}(hjh]hsize}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]uh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKthjubh definition)}(hhh](h)}(hRequested size for the object.h]hRequested size for the object.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKjhjubh)}(hBThe (page-aligned) allocated size for the object will be returned.h]hBThe (page-aligned) allocated size for the object will be returned.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKlhjubh)}(hXNote that for some devices we have might have further minimum page-size restrictions (larger than 4K), like for device local-memory. However in general the final size here should always reflect any rounding up, if for example using the I915_GEM_CREATE_EXT_MEMORY_REGIONS extension to place the object in device local-memory. The kernel will always select the largest minimum page-size for the set of possible placements as the value to use when rounding up the **size**.h](hXNote that for some devices we have might have further minimum page-size restrictions (larger than 4K), like for device local-memory. However in general the final size here should always reflect any rounding up, if for example using the I915_GEM_CREATE_EXT_MEMORY_REGIONS extension to place the object in device local-memory. The kernel will always select the largest minimum page-size for the set of possible placements as the value to use when rounding up the }(hj hhhNhNubj)}(h**size**h]hsize}(hj(hhhNhNubah}(h]h ]h"]h$]h&]uh1jhj ubh.}(hj hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKnhjubeh}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjhKthjubj)}(hH``handle`` Returned handle for the object. Object handles are nonzero. h](j)}(h ``handle``h]j)}(hjSh]hhandle}(hjUhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjQubah}(h]h ]h"]h$]h&]uh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhK{hjMubj)}(hhh](h)}(hReturned handle for the object.h]hReturned handle for the object.}(hjlhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKyhjiubh)}(hObject handles are nonzero.h]hObject handles are nonzero.}(hj{hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhK{hjiubeh}(h]h ]h"]h$]h&]uh1jhjMubeh}(h]h ]h"]h$]h&]uh1jhjhhK{hjubj)}(hX``flags`` Optional flags. Supported values: I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS - Signal to the kernel that the object will need to be accessed via the CPU. Only valid when placing objects in I915_MEMORY_CLASS_DEVICE, and only strictly required on configurations where some subset of the device memory is directly visible/mappable through the CPU (which we also call small BAR), like on some DG2+ systems. Note that this is quite undesirable, but due to various factors like the client CPU, BIOS etc it's something we can expect to see in the wild. See :c:type:`__drm_i915_memory_region_info.probed_cpu_visible_size <__drm_i915_memory_region_info>` for how to determine if this system applies. Note that one of the placements MUST be I915_MEMORY_CLASS_SYSTEM, to ensure the kernel can always spill the allocation to system memory, if the object can't be allocated in the mappable part of I915_MEMORY_CLASS_DEVICE. Also note that since the kernel only supports flat-CCS on objects that can *only* be placed in I915_MEMORY_CLASS_DEVICE, we therefore don't support I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS together with flat-CCS. Without this hint, the kernel will assume that non-mappable I915_MEMORY_CLASS_DEVICE is preferred for this object. Note that the kernel can still migrate the object to the mappable part, as a last resort, if userspace ever CPU faults this object, but this might be expensive, and so ideally should be avoided. On older kernels which lack the relevant small-bar uAPI support (see also :c:type:`__drm_i915_memory_region_info.probed_cpu_visible_size <__drm_i915_memory_region_info>`), usage of the flag will result in an error, but it should NEVER be possible to end up with a small BAR configuration, assuming we can also successfully load the i915 kernel module. In such cases the entire I915_MEMORY_CLASS_DEVICE region will be CPU accessible, and as such there are zero restrictions on where the object can be placed. h](j)}(h ``flags``h]j)}(hjh]hflags}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]uh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubj)}(hhh](h)}(hOptional flags.h]hOptional flags.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh)}(hSupported values:h]hSupported values:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh)}(hvI915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS - Signal to the kernel that the object will need to be accessed via the CPU.h]hvI915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS - Signal to the kernel that the object will need to be accessed via the CPU.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh)}(hXOnly valid when placing objects in I915_MEMORY_CLASS_DEVICE, and only strictly required on configurations where some subset of the device memory is directly visible/mappable through the CPU (which we also call small BAR), like on some DG2+ systems. Note that this is quite undesirable, but due to various factors like the client CPU, BIOS etc it's something we can expect to see in the wild. See :c:type:`__drm_i915_memory_region_info.probed_cpu_visible_size <__drm_i915_memory_region_info>` for how to determine if this system applies.h](hXOnly valid when placing objects in I915_MEMORY_CLASS_DEVICE, and only strictly required on configurations where some subset of the device memory is directly visible/mappable through the CPU (which we also call small BAR), like on some DG2+ systems. Note that this is quite undesirable, but due to various factors like the client CPU, BIOS etc it’s something we can expect to see in the wild. See }(hjhhhNhNubh)}(h_:c:type:`__drm_i915_memory_region_info.probed_cpu_visible_size <__drm_i915_memory_region_info>`h]j)}(hjh]h5__drm_i915_memory_region_info.probed_cpu_visible_size}(hjhhhNhNubah}(h]h ](xrefj}c-typeeh"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]refdocgpu/rfc/i915_small_bar refdomainj}reftypetype refexplicitrefwarn c:parent_keysphinx.domains.c LookupKey)}data]sb reftarget__drm_i915_memory_region_infouh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh- for how to determine if this system applies.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhjhKhjubh)}(hNote that one of the placements MUST be I915_MEMORY_CLASS_SYSTEM, to ensure the kernel can always spill the allocation to system memory, if the object can't be allocated in the mappable part of I915_MEMORY_CLASS_DEVICE.h]hNote that one of the placements MUST be I915_MEMORY_CLASS_SYSTEM, to ensure the kernel can always spill the allocation to system memory, if the object can’t be allocated in the mappable part of I915_MEMORY_CLASS_DEVICE.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh)}(hAlso note that since the kernel only supports flat-CCS on objects that can *only* be placed in I915_MEMORY_CLASS_DEVICE, we therefore don't support I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS together with flat-CCS.h](hKAlso note that since the kernel only supports flat-CCS on objects that can }(hj-hhhNhNubhemphasis)}(h*only*h]honly}(hj7hhhNhNubah}(h]h ]h"]h$]h&]uh1j5hj-ubh be placed in I915_MEMORY_CLASS_DEVICE, we therefore don’t support I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS together with flat-CCS.}(hj-hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh)}(hX5Without this hint, the kernel will assume that non-mappable I915_MEMORY_CLASS_DEVICE is preferred for this object. Note that the kernel can still migrate the object to the mappable part, as a last resort, if userspace ever CPU faults this object, but this might be expensive, and so ideally should be avoided.h]hX5Without this hint, the kernel will assume that non-mappable I915_MEMORY_CLASS_DEVICE is preferred for this object. Note that the kernel can still migrate the object to the mappable part, as a last resort, if userspace ever CPU faults this object, but this might be expensive, and so ideally should be avoided.}(hjPhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh)}(hXOn older kernels which lack the relevant small-bar uAPI support (see also :c:type:`__drm_i915_memory_region_info.probed_cpu_visible_size <__drm_i915_memory_region_info>`), usage of the flag will result in an error, but it should NEVER be possible to end up with a small BAR configuration, assuming we can also successfully load the i915 kernel module. In such cases the entire I915_MEMORY_CLASS_DEVICE region will be CPU accessible, and as such there are zero restrictions on where the object can be placed.h](hJOn older kernels which lack the relevant small-bar uAPI support (see also }(hj_hhhNhNubh)}(h_:c:type:`__drm_i915_memory_region_info.probed_cpu_visible_size <__drm_i915_memory_region_info>`h]j)}(hjih]h5__drm_i915_memory_region_info.probed_cpu_visible_size}(hjkhhhNhNubah}(h]h ](jj}c-typeeh"]h$]h&]uh1jhjgubah}(h]h ]h"]h$]h&]refdocj refdomainj}reftypetype refexplicitrefwarnj j j__drm_i915_memory_region_infouh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhj_ubhXR), usage of the flag will result in an error, but it should NEVER be possible to end up with a small BAR configuration, assuming we can also successfully load the i915 kernel module. In such cases the entire I915_MEMORY_CLASS_DEVICE region will be CPU accessible, and as such there are zero restrictions on where the object can be placed.}(hj_hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhjhKhjubeh}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjhKhjubj)}(hX``extensions`` The chain of extensions to apply to this object. This will be useful in the future when we need to support several different extensions, and we need to apply more than one when creating the object. See struct i915_user_extension. If we don't supply any extensions then we get the same old gem_create behaviour. For I915_GEM_CREATE_EXT_MEMORY_REGIONS usage see struct drm_i915_gem_create_ext_memory_regions. For I915_GEM_CREATE_EXT_PROTECTED_CONTENT usage see struct drm_i915_gem_create_ext_protected_content.h](j)}(h``extensions``h]j)}(hjh]h extensions}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]uh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubj)}(hhh](h)}(h0The chain of extensions to apply to this object.h]h0The chain of extensions to apply to this object.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh)}(hThis will be useful in the future when we need to support several different extensions, and we need to apply more than one when creating the object. See struct i915_user_extension.h]hThis will be useful in the future when we need to support several different extensions, and we need to apply more than one when creating the object. See struct i915_user_extension.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh)}(hPIf we don't supply any extensions then we get the same old gem_create behaviour.h]hRIf we don’t supply any extensions then we get the same old gem_create behaviour.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh)}(h_For I915_GEM_CREATE_EXT_MEMORY_REGIONS usage see struct drm_i915_gem_create_ext_memory_regions.h]h_For I915_GEM_CREATE_EXT_MEMORY_REGIONS usage see struct drm_i915_gem_create_ext_memory_regions.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh)}(heFor I915_GEM_CREATE_EXT_PROTECTED_CONTENT usage see struct drm_i915_gem_create_ext_protected_content.h]heFor I915_GEM_CREATE_EXT_PROTECTED_CONTENT usage see struct drm_i915_gem_create_ext_protected_content.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhKhjubeh}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjhKhjubeh}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ] kernelindentah"]h$]h&]uh1jhhhhhhhNubh)}(h**Description**h]j)}(hj!h]h Description}(hj#hhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhKhhhhubh)}(hXcNote that new buffer flags should be added here, at least for the stuff that is immutable. Previously we would have two ioctls, one to create the object with gem_create, and another to apply various parameters, however this creates some ambiguity for the params which are considered immutable. Also in general we're phasing out the various SET/GET ioctls.h]hXeNote that new buffer flags should be added here, at least for the stuff that is immutable. Previously we would have two ioctls, one to create the object with gem_create, and another to apply various parameters, however this creates some ambiguity for the params which are considered immutable. Also in general we’re phasing out the various SET/GET ioctls.}(hj7hhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:19: ./Documentation/gpu/rfc/i915_small_bar.hhK_hhhhubeh}(h].i915-gem-create-ext-flag-needs-cpu-access-flagah ]h"].i915_gem_create_ext_flag_needs_cpu_access flagah$]h&]uh1hhhhhhhhK ubh)}(hhh](h)}(h!probed_cpu_visible_size attributeh]h!probed_cpu_visible_size attribute}(hjQhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjNhhhhhKubh)}(hXNew struct__drm_i915_memory_region attribute which returns the total size of the CPU accessible portion, for the particular region. This should only be applicable for I915_MEMORY_CLASS_DEVICE. We also report the unallocated_cpu_visible_size, alongside the unallocated_size.h]hXNew struct__drm_i915_memory_region attribute which returns the total size of the CPU accessible portion, for the particular region. This should only be applicable for I915_MEMORY_CLASS_DEVICE. We also report the unallocated_cpu_visible_size, alongside the unallocated_size.}(hj_hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjNhhubh)}(hX+Vulkan will need this as part of creating a separate VkMemoryHeap with the VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT set, to represent the CPU visible portion, where the total size of the heap needs to be known. It also wants to be able to give a rough estimate of how memory can potentially be allocated.h]hX+Vulkan will need this as part of creating a separate VkMemoryHeap with the VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT set, to represent the CPU visible portion, where the total size of the heap needs to be known. It also wants to be able to give a rough estimate of how memory can potentially be allocated.}(hjmhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjNhhubh)}(hhh]h}(h]h ]h"]h$]h&]entries](h(__drm_i915_memory_region_info (C struct)c.__drm_i915_memory_region_infohNtauh1hhjNhhhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhNubh)}(hhh](h)}(h__drm_i915_memory_region_infoh]j)}(h$struct __drm_i915_memory_region_infoh](j )}(hj h]hstruct}(hjhhhNhNubah}(h]h ]jah"]h$]h&]uh1j hjhhhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKubj)}(h h]h }(hjhhhNhNubah}(h]h ]j(ah"]h$]h&]uh1jhjhhhjhKubj-)}(h__drm_i915_memory_region_infoh]j3)}(hjh]h__drm_i915_memory_region_info}(hjhhhNhNubah}(h]h ]j>ah"]h$]h&]uh1j2hjubah}(h]h ](jEjFeh"]h$]h&]jJjKuh1j,hjhhhjhKubeh}(h]h ]h"]h$]h&]jJjKjRuh1jjSjThjhhhjhKubah}(h]jah ](jXjYeh"]h$]h&]j]j^)j_huh1hhjhKhjhhubja)}(hhh]h)}(h,Describes one region as known to the driver.h]h,Describes one region as known to the driver.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKhjhhubah}(h]h ]h"]h$]h&]uh1j`hjhhhjhKubeh}(h]h ](j}structeh"]h$]h&]jj}jjjjjjjuh1hhhhjNhjhNubj)}(hX **Definition**:: struct __drm_i915_memory_region_info { struct drm_i915_gem_memory_class_instance region; __u32 rsvd0; __u64 probed_size; __u64 unallocated_size; union { __u64 rsvd1[8]; struct { __u64 probed_cpu_visible_size; __u64 unallocated_cpu_visible_size; }; }; }; **Members** ``region`` The class:instance pair encoding ``rsvd0`` MBZ ``probed_size`` Memory probed by the driver Note that it should not be possible to ever encounter a zero value here, also note that no current region type will ever return -1 here. Although for future region types, this might be a possibility. The same applies to the other size fields. ``unallocated_size`` Estimate of memory remaining Requires CAP_PERFMON or CAP_SYS_ADMIN to get reliable accounting. Without this (or if this is an older kernel) the value here will always equal the **probed_size**. Note this is only currently tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will always equal the **probed_size**). ``{unnamed_union}`` anonymous ``rsvd1`` MBZ ``{unnamed_struct}`` anonymous ``probed_cpu_visible_size`` Memory probed by the driver that is CPU accessible. This will be always be <= **probed_size**, and the remainder (if there is any) will not be CPU accessible. On systems without small BAR, the **probed_size** will always equal the **probed_cpu_visible_size**, since all of it will be CPU accessible. Note this is only tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will always equal the **probed_size**). Note that if the value returned here is zero, then this must be an old kernel which lacks the relevant small-bar uAPI support (including I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS), but on such systems we should never actually end up with a small BAR configuration, assuming we are able to load the kernel module. Hence it should be safe to treat this the same as when **probed_cpu_visible_size** == **probed_size**. ``unallocated_cpu_visible_size`` Estimate of CPU visible memory remaining Note this is only tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will always equal the **probed_cpu_visible_size**). Requires CAP_PERFMON or CAP_SYS_ADMIN to get reliable accounting. Without this the value here will always equal the **probed_cpu_visible_size**. Note this is only currently tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will also always equal the **probed_cpu_visible_size**). If this is an older kernel the value here will be zero, see also **probed_cpu_visible_size**.h](h)}(h**Definition**::h](j)}(h**Definition**h]h Definition}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubj)}(hXEstruct __drm_i915_memory_region_info { struct drm_i915_gem_memory_class_instance region; __u32 rsvd0; __u64 probed_size; __u64 unallocated_size; union { __u64 rsvd1[8]; struct { __u64 probed_cpu_visible_size; __u64 unallocated_cpu_visible_size; }; }; };h]hXEstruct __drm_i915_memory_region_info { struct drm_i915_gem_memory_class_instance region; __u32 rsvd0; __u64 probed_size; __u64 unallocated_size; union { __u64 rsvd1[8]; struct { __u64 probed_cpu_visible_size; __u64 unallocated_cpu_visible_size; }; }; };}hjsbah}(h]h ]h"]h$]h&]jJjKuh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh)}(h **Members**h]j)}(hj%h]hMembers}(hj'hhhNhNubah}(h]h ]h"]h$]h&]uh1jhj#ubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubj)}(hhh](j)}(h,``region`` The class:instance pair encoding h](j)}(h ``region``h]j)}(hjDh]hregion}(hjFhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjBubah}(h]h ]h"]h$]h&]uh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhK hj>ubj)}(hhh]h)}(h The class:instance pair encodingh]h The class:instance pair encoding}(hj]hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjYhK hjZubah}(h]h ]h"]h$]h&]uh1jhj>ubeh}(h]h ]h"]h$]h&]uh1jhjYhK hj;ubj)}(h``rsvd0`` MBZ h](j)}(h ``rsvd0``h]j)}(hj}h]hrsvd0}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhj{ubah}(h]h ]h"]h$]h&]uh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKhjwubj)}(hhh]h)}(hMBZh]hMBZ}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhKhjubah}(h]h ]h"]h$]h&]uh1jhjwubeh}(h]h ]h"]h$]h&]uh1jhjhKhj;ubj)}(hX ``probed_size`` Memory probed by the driver Note that it should not be possible to ever encounter a zero value here, also note that no current region type will ever return -1 here. Although for future region types, this might be a possibility. The same applies to the other size fields. h](j)}(h``probed_size``h]j)}(hjh]h probed_size}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]uh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubj)}(hhh](h)}(hMemory probed by the driverh]hMemory probed by the driver}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh)}(hNote that it should not be possible to ever encounter a zero value here, also note that no current region type will ever return -1 here. Although for future region types, this might be a possibility. The same applies to the other size fields.h]hNote that it should not be possible to ever encounter a zero value here, also note that no current region type will ever return -1 here. Although for future region types, this might be a possibility. The same applies to the other size fields.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubeh}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjhKhj;ubj)}(hXi``unallocated_size`` Estimate of memory remaining Requires CAP_PERFMON or CAP_SYS_ADMIN to get reliable accounting. Without this (or if this is an older kernel) the value here will always equal the **probed_size**. Note this is only currently tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will always equal the **probed_size**). h](j)}(h``unallocated_size``h]j)}(hjh]hunallocated_size}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]uh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhK"hjubj)}(hhh](h)}(hEstimate of memory remainingh]hEstimate of memory remaining}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh)}(hX5Requires CAP_PERFMON or CAP_SYS_ADMIN to get reliable accounting. Without this (or if this is an older kernel) the value here will always equal the **probed_size**. Note this is only currently tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will always equal the **probed_size**).h](hRequires CAP_PERFMON or CAP_SYS_ADMIN to get reliable accounting. Without this (or if this is an older kernel) the value here will always equal the }(hj'hhhNhNubj)}(h**probed_size**h]h probed_size}(hj/hhhNhNubah}(h]h ]h"]h$]h&]uh1jhj'ubh. Note this is only currently tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will always equal the }(hj'hhhNhNubj)}(h**probed_size**h]h probed_size}(hjAhhhNhNubah}(h]h ]h"]h$]h&]uh1jhj'ubh).}(hj'hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubeh}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjhK"hj;ubj)}(h``{unnamed_union}`` anonymous h](j)}(h``{unnamed_union}``h]j)}(hjlh]h{unnamed_union}}(hjnhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjjubah}(h]h ]h"]h$]h&]uh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhK%hjfubj)}(hhh]h)}(h anonymoush]h anonymous}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhK%hjubah}(h]h ]h"]h$]h&]uh1jhjfubeh}(h]h ]h"]h$]h&]uh1jhjhK%hj;ubj)}(h``rsvd1`` MBZ h](j)}(h ``rsvd1``h]j)}(hjh]hrsvd1}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]uh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubj)}(hhh]h)}(hMBZh]hMBZ}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhKhjubah}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjhKhj;ubj)}(h``{unnamed_struct}`` anonymous h](j)}(h``{unnamed_struct}``h]j)}(hjh]h{unnamed_struct}}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]uh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubj)}(hhh]h)}(h anonymoush]h anonymous}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhKhjubah}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjhKhj;ubj)}(hXs``probed_cpu_visible_size`` Memory probed by the driver that is CPU accessible. This will be always be <= **probed_size**, and the remainder (if there is any) will not be CPU accessible. On systems without small BAR, the **probed_size** will always equal the **probed_cpu_visible_size**, since all of it will be CPU accessible. Note this is only tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will always equal the **probed_size**). Note that if the value returned here is zero, then this must be an old kernel which lacks the relevant small-bar uAPI support (including I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS), but on such systems we should never actually end up with a small BAR configuration, assuming we are able to load the kernel module. Hence it should be safe to treat this the same as when **probed_cpu_visible_size** == **probed_size**. h](j)}(h``probed_cpu_visible_size``h]j)}(hjh]hprobed_cpu_visible_size}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]uh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKBhjubj)}(hhh](h)}(h3Memory probed by the driver that is CPU accessible.h]h3Memory probed by the driver that is CPU accessible.}(hj0hhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhK+hj-ubh)}(hjThis will be always be <= **probed_size**, and the remainder (if there is any) will not be CPU accessible.h](hThis will be always be <= }(hj?hhhNhNubj)}(h**probed_size**h]h probed_size}(hjGhhhNhNubah}(h]h ]h"]h$]h&]uh1jhj?ubhA, and the remainder (if there is any) will not be CPU accessible.}(hj?hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhK.hj-ubh)}(hOn systems without small BAR, the **probed_size** will always equal the **probed_cpu_visible_size**, since all of it will be CPU accessible.h](h"On systems without small BAR, the }(hj`hhhNhNubj)}(h**probed_size**h]h probed_size}(hjhhhhNhNubah}(h]h ]h"]h$]h&]uh1jhj`ubh will always equal the }(hj`hhhNhNubj)}(h**probed_cpu_visible_size**h]hprobed_cpu_visible_size}(hjzhhhNhNubah}(h]h ]h"]h$]h&]uh1jhj`ubh), since all of it will be CPU accessible.}(hj`hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhK2hj-ubh)}(hNote this is only tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will always equal the **probed_size**).h](huNote this is only tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will always equal the }(hjhhhNhNubj)}(h**probed_size**h]h probed_size}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh).}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhK6hj-ubh)}(hXNote that if the value returned here is zero, then this must be an old kernel which lacks the relevant small-bar uAPI support (including I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS), but on such systems we should never actually end up with a small BAR configuration, assuming we are able to load the kernel module. Hence it should be safe to treat this the same as when **probed_cpu_visible_size** == **probed_size**.h](hXpNote that if the value returned here is zero, then this must be an old kernel which lacks the relevant small-bar uAPI support (including I915_GEM_CREATE_EXT_FLAG_NEEDS_CPU_ACCESS), but on such systems we should never actually end up with a small BAR configuration, assuming we are able to load the kernel module. Hence it should be safe to treat this the same as when }(hjhhhNhNubj)}(h**probed_cpu_visible_size**h]hprobed_cpu_visible_size}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh == }(hjhhhNhNubj)}(h**probed_size**h]h probed_size}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhK:hj-ubeh}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhj,hKBhj;ubj)}(hXq``unallocated_cpu_visible_size`` Estimate of CPU visible memory remaining Note this is only tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will always equal the **probed_cpu_visible_size**). Requires CAP_PERFMON or CAP_SYS_ADMIN to get reliable accounting. Without this the value here will always equal the **probed_cpu_visible_size**. Note this is only currently tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will also always equal the **probed_cpu_visible_size**). If this is an older kernel the value here will be zero, see also **probed_cpu_visible_size**.h](j)}(h ``unallocated_cpu_visible_size``h]j)}(hjh]hunallocated_cpu_visible_size}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]uh1jhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKVhjubj)}(hhh](h)}(h(Estimate of CPU visible memory remainingh]h(Estimate of CPU visible memory remaining}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKGhjubh)}(hNote this is only tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will always equal the **probed_cpu_visible_size**).h](huNote this is only tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will always equal the }(hj!hhhNhNubj)}(h**probed_cpu_visible_size**h]hprobed_cpu_visible_size}(hj)hhhNhNubah}(h]h ]h"]h$]h&]uh1jhj!ubh).}(hj!hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKJhjubh)}(hX3Requires CAP_PERFMON or CAP_SYS_ADMIN to get reliable accounting. Without this the value here will always equal the **probed_cpu_visible_size**. Note this is only currently tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will also always equal the **probed_cpu_visible_size**).h](huRequires CAP_PERFMON or CAP_SYS_ADMIN to get reliable accounting. Without this the value here will always equal the }(hjBhhhNhNubj)}(h**probed_cpu_visible_size**h]hprobed_cpu_visible_size}(hjJhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjBubh. Note this is only currently tracked for I915_MEMORY_CLASS_DEVICE regions (for other types the value here will also always equal the }(hjBhhhNhNubj)}(h**probed_cpu_visible_size**h]hprobed_cpu_visible_size}(hj\hhhNhNubah}(h]h ]h"]h$]h&]uh1jhjBubh).}(hjBhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKOhjubh)}(h]If this is an older kernel the value here will be zero, see also **probed_cpu_visible_size**.h](hAIf this is an older kernel the value here will be zero, see also }(hjuhhhNhNubj)}(h**probed_cpu_visible_size**h]hprobed_cpu_visible_size}(hj}hhhNhNubah}(h]h ]h"]h$]h&]uh1jhjuubh.}(hjuhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhjhKVhjubeh}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ]h"]h$]h&]uh1jhjhKVhj;ubeh}(h]h ]h"]h$]h&]uh1jhjubeh}(h]h ] kernelindentah"]h$]h&]uh1jhjNhhhjhNubh)}(h**Description**h]j)}(hjh]h Description}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]uh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKZhjNhhubh)}(hNote this is using both struct drm_i915_query_item and struct drm_i915_query. For this new query we are adding the new query id DRM_I915_QUERY_MEMORY_REGIONS at :c:type:`drm_i915_query_item.query_id `.h](hNote this is using both struct drm_i915_query_item and struct drm_i915_query. For this new query we are adding the new query id DRM_I915_QUERY_MEMORY_REGIONS at }(hjhhhNhNubh)}(h<:c:type:`drm_i915_query_item.query_id `h]j)}(hjh]hdrm_i915_query_item.query_id}(hjhhhNhNubah}(h]h ](jj}c-typeeh"]h$]h&]uh1jhjubah}(h]h ]h"]h$]h&]refdocj refdomainj}reftypetype refexplicitrefwarnj j jdrm_i915_query_itemuh1hhm/var/lib/git/docbuild/linux/Documentation/gpu/rfc/i915_small_bar:34: ./Documentation/gpu/rfc/i915_small_bar.hhKhjubh.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhjhKhjNhhubeh}(h]!probed-cpu-visible-size-attributeah ]h"]!probed_cpu_visible_size attributeah$]h&]uh1hhhhhhhhKubh)}(hhh](h)}(hError Capture restrictionsh]hError Capture restrictions}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhK&ubh)}(h0With error capture we have two new restrictions:h]h0With error capture we have two new restrictions:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK'hj hhubh block_quote)}(hX1) Error capture is best effort on small BAR systems; if the pages are not CPU accessible, at the time of capture, then the kernel is free to skip trying to capture them. 2) On discrete and newer integrated platforms we now reject error capture on recoverable contexts. In the future the kernel may want to blit during error capture, when for example something is not currently CPU accessible.h](h)}(h1) Error capture is best effort on small BAR systems; if the pages are not CPU accessible, at the time of capture, then the kernel is free to skip trying to capture them.h]h1) Error capture is best effort on small BAR systems; if the pages are not CPU accessible, at the time of capture, then the kernel is free to skip trying to capture them.}(hj% hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK)hj! ubh)}(h2) On discrete and newer integrated platforms we now reject error capture on recoverable contexts. In the future the kernel may want to blit during error capture, when for example something is not currently CPU accessible.h]h2) On discrete and newer integrated platforms we now reject error capture on recoverable contexts. In the future the kernel may want to blit during error capture, when for example something is not currently CPU accessible.}(hj3 hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK-hj! ubeh}(h]h ]h"]h$]h&]uh1j hhhK)hj hhubeh}(h]error-capture-restrictionsah ]h"]error capture restrictionsah$]h&]uh1hhhhhhhhK&ubeh}(h]i915-small-bar-rfc-sectionah ]h"]i915 small bar rfc sectionah$]h&]uh1hhhhhhhhKubeh}(h]h ]h"]h$]h&]sourcehuh1hcurrent_sourceN current_lineNsettingsdocutils.frontendValues)}(hN generatorN datestampN source_linkN source_urlN toc_backlinksentryfootnote_backlinksK sectnum_xformKstrip_commentsNstrip_elements_with_classesN strip_classesN report_levelK halt_levelKexit_status_levelKdebugNwarning_streamN tracebackinput_encoding utf-8-siginput_encoding_error_handlerstrictoutput_encodingutf-8output_encoding_error_handlerjz error_encodingutf-8error_encoding_error_handlerbackslashreplace language_codeenrecord_dependenciesNconfigN id_prefixhauto_id_prefixid dump_settingsNdump_internalsNdump_transformsNdump_pseudo_xmlNexpose_internalsNstrict_visitorN_disable_configN_sourceh _destinationN _config_files]7/var/lib/git/docbuild/linux/Documentation/docutils.confafile_insertion_enabled raw_enabledKline_length_limitM'pep_referencesN pep_base_urlhttps://peps.python.org/pep_file_url_templatepep-%04drfc_referencesN rfc_base_url&https://datatracker.ietf.org/doc/html/ tab_widthKtrim_footnote_reference_spacesyntax_highlightlong smart_quotessmartquotes_locales]character_level_inline_markupdoctitle_xform docinfo_xformKsectsubtitle_xform image_loadinglinkembed_stylesheetcloak_email_addressessection_self_linkenvNubreporterNindirect_targets]substitution_defs}substitution_names}refnames}refids}nameids}(jT jQ jKjHjjjL jI u nametypes}(jT jKjjL uh}(jQ hjHhhhjjNjjjI j u footnote_refs} citation_refs} autofootnotes]autofootnote_refs]symbol_footnotes]symbol_footnote_refs] footnotes] citations]autofootnote_startKsymbol_footnote_startK id_counter collectionsCounter}Rparse_messages]transform_messages] transformerN include_log] decorationNhhub.