summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
...
| * st/glsl_to_tgsi: fix dvec[34] loads from SSBONicolai Hähnle2016-11-091-6/+4
| | | | | | | | | | | | | | | | | | | | | | When splitting up loads, we have to add 16 bytes to the offset for the high components, just like already happens for stores. Fixes arb_gpu_shader_fp64@shader_storage@layout-std140-fp64-shader. Cc: 13.0 <mesa-stable@lists.freedesktop.org> Reviewed-by: Marek Olšák <marek.olsak@amd.com> (cherry picked from commit e4b378800eff13752dcfe1f5c6b640444208d543)
| * nir: Flip gl_SamplePosition in nir_lower_wpos_ytransform().Francisco Jerez2016-11-091-0/+24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Assuming the hardware is set up to use a screen coordinate system flipped vertically with respect to the GL's window coordinate system, the SYSTEM_VALUE_SAMPLE_POS vector will also be flipped vertically with respect to the value expected by the GL, so we need to give it the same treatment as gl_FragCoord. Fixes the following CTS tests on i965: ES31-CTS.functional.shaders.multisample_interpolation.interpolate_at_offset.at_sample_position.default_framebuffer ES31-CTS.functional.shaders.sample_variables.sample_pos.correctness.default_framebuffer when run with any multisample configuration, e.g. rgba8888d24s8ms4. Cc: <mesa-stable@lists.freedesktop.org> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com> (cherry picked from commit f3d387867f74ae758b41168f23992671f7dce254)
| * glx/windows: Add wgl.h to the sources listAndreas Boll2016-11-091-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | Otherwise it won't be picked in the tarball and the build will fail. Fixes: 533b3530c12 ("direct-to-native-GL for GLX clients on Cygwin ("Windows-DRI")") Cc: "13.0" <mesa-stable@lists.freedesktop.org> Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com> Reviewed-by: Jon Turney <jon.turney@dronecode.org.uk> (cherry picked from commit f792f0687f60e789fe9e0e2550c4eded95253fa1)
| * glsl: fix lowering of UBO references of named blocksNicolai Hähnle2016-11-091-5/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When a UBO reference has the form block_name.foo where block_name refers to a block where the first member has a non-zero offset, the base offset was incorrectly added to the reference. Fixes an assertion triggered in debug builds by GL45-CTS.enhanced_layouts.uniform_block_layout_qualifier_conflict. That test doesn't properly check for correct execution in this case, so I am also going to send out a piglit test. Cc: 13.0 <mesa-stable@lists.freedesktop.org> Reviewed-by: Iago Toral Quiroga <itoral@igalia.com> (cherry picked from commit 37d646c1b3626ad54ed93a784824af7b5abe8a99)
| * glsl: Update deref types when resizing implicitly sized arrays.Kenneth Graunke2016-11-091-23/+47
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | At link time, we resolve the size of implicitly sized arrays. When doing so, we update the type of the ir_variables. However, we neglected to update the type of ir_dereference nodes which reference those variables. It turns out array_resize_visitor (for GS/TCS/TES interface array handling) already did 2/3 of the cases for this, so we can simply refactor the code and reuse it. This fixes: GL45-CTS.shader_storage_buffer_object.basic-syntax GL45-CTS.shader_storage_buffer_object.basic-syntaxSSO which have an SSBO containing an implicitly sized array, followed by some other members. setup_buffer_access uses the dereference types to compute offsets to fields, and it had a stale type where the implicitly sized array's length was still 0 instead of the actual length. While we're here, we can also fix update_array_sizes to properly update deref types as well, fixing a FINISHME from 2010. Cc: mesa-stable@lists.freedesktop.org Signed-off-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> (cherry picked from commit 8df4aebc94337983194cc72c817c08ee938117a1)
| * mesa/glsl: delete previously linked shaders earlier when linkingTimothy Arceri2016-11-097-26/+25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This moves the delete linked shaders call to _mesa_clear_shader_program_data() which makes sure we delete them before returning due to any validation problems. It also reduces some code duplication. From the OpenGL 4.5 Core spec: "If LinkProgram failed, any information about a previous link of that program object is lost. Thus, a failed link does not restore the old state of program. ... If one of these commands is called with a program for which LinkProgram failed, no error is generated unless otherwise noted. Implementations may return information on variables and interface blocks that would have been active had the program been linked successfully. In cases where the link failed because the program required too many resources, these commands may help applications determine why limits were exceeded." Therefore it's expected that we shouldn't be able to query the program that failed to link and retrieve information about a previously successful link. Before this change the linker was doing validation before freeing the previously linked shaders and therefore could exit on failure before they were freed. This change also fixes an issue in compat profile where a program with no shaders attached is expect to fall back to fixed function but was instead trying to relink IR from a previous link. Reviewed-by: Tapani Pälli <tapani.palli@intel.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97715 Cc: "13.0" <mesa-stable@lists.freedesktop.org> (cherry picked from commit d2861d682a235993844989f7742c9539c3e10245)
| * radv: add support for anisotropic filtering on VI+Fredrik Höglund2016-11-091-3/+27
| | | | | | | | | | | | | | | | Ported from radeonsi. Cc: "13.0" <mesa-stable@lists.freedesktop.org> Signed-off-by: Dave Airlie <airlied@redhat.com> (cherry picked from commit e7b9c5eb74619dbda3f24928196ab8c8d9408c0b)
| * radv: fix dual source blendingDave Airlie2016-11-091-3/+9
| | | | | | | | | | | | | | | | | | | | | | | | Dolphin tried to use this, but we hadn't had any tests for it properly. All that is required is the shader output format needs to be set for 0 and 1 exports. Cc: "13.0" <mesa-stable@lists.freedesktop.org> Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl> Signed-off-by: Dave Airlie <airlied@redhat.com> (cherry picked from commit 73592b92844af3c1298ef5e66e3055793443546d)
| * glx/glvnd: Fix dispatch function names and indicesAdam Jackson2016-11-092-146/+144
| | | | | | | | | | | | | | | | | | | | As this array was not actually sorted, FindGLXFunction's binary search would only sometimes work. Cc: "13.0" <mesa-stable@lists.freedesktop.org> Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com> Signed-off-by: Adam Jackson <ajax@redhat.com> (cherry picked from commit 8bca8d89ef3babbecc4f3160cbbcb93d33b7a6ee)
| * glx/glvnd: Don't modify the dummy slot in the dispatch tableAdam Jackson2016-11-091-0/+3
| | | | | | | | | | | | | | Cc: "13.0" <mesa-stable@lists.freedesktop.org> Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com> Signed-off-by: Adam Jackson <ajax@redhat.com> (cherry picked from commit deb0eb1660f612862c77d225582e102e7ab717ee)
| * anv/pipeline: Properly cache prog_data::paramJason Ekstrand2016-11-093-14/+36
| | | | | | | | | | | | | | | | | | | | | | | | Before we were caching the prog data but we weren't doing anything with brw_stage_prog_data::param so anything with push constants wasn't getting cached properly. This commit fixes that. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98012 Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Cc: "13.0" <mesa-stable@lists.freedesktop.org> (cherry picked from commit 71cc1e188d52bebe86a1ac72fe90f4e2a7e76778)
| * anv/pipeline: Put actual pointers in anv_shader_binJason Ekstrand2016-11-095-50/+45
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | While we can simply calculate offsets to get to things such as the prog_data and the key, it's much more user-friendly if there are just pointers. Also, it's a bit more fool-proof. While we're at it, we rework the pipeline cache API to use the brw_stage_prog_data type directly. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98012 Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Cc: "13.0" <mesa-stable@lists.freedesktop.org> (cherry picked from commit ff3185e3ba85b3635a1f645e8e951954f4022afe)
| * intel/blorp: Pass a brw_stage_prog_data to upload_shaderJason Ekstrand2016-11-095-6/+9
| | | | | | | | | | | | | | | | | | Signed-off-by: Jason Ekstrand <jason@jlekstrand.net> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98012 Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Cc: "13.0" <mesa-stable@lists.freedesktop.org> (cherry picked from commit 4306c10a885bdd833328a2a4eeb0079aa4f4bae3)
| * intel/blorp: Use wm_prog_data instead of hand-rolling our ownJason Ekstrand2016-11-095-66/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Jason Ekstrand <jason@jlekstrand.net> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98012 Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com> Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com> Cc: "13.0" <mesa-stable@lists.freedesktop.org> (cherry picked from commit 058304f0819c96fd18eb17911022a410540aa295) [Emil Velikov: brw_compile_fs() has different signature] Signed-off-by: Emil Velikov <emil.velikov@collabora.com> Conflicts: src/intel/blorp/blorp.c
| * anv: Better handle return codes from anv_physical_device_initJason Ekstrand2016-11-091-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | The case where we just want the loop to continue is INCOMPATIBLE_DRIVER because that simply means that whatever FD we opened isn't a supported Intel chip. Other error codes such as OUT_OF_HOST_MEMORY are actual errors and we should be returning early in that case. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net> Reviewed-by: Dave Airlie <airlied@redhat.com> Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com> Cc: "13.0" <mesa-stable@lists.freedesktop.org> (cherry picked from commit a5f8ff6ca18c38a7c4b5b37a4d5b14ca01e71b1e)
| * vulkan/wsi/x11: Clean up connections in finish_wsiJason Ekstrand2016-11-091-0/+4
| | | | | | | | | | | | | | | | Signed-off-by: Jason Ekstrand <jason@jlekstrand.net> Reviewed-by: Dave Airlie <airlied@redhat.com> Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com> Cc: "13.0" <mesa-stable@lists.freedesktop.org> (cherry picked from commit daeb21e47845795d0320811db1e202540fba356b)
| * vulkan/wsi/x11: Better handle wsi_x11_connection_create failureJason Ekstrand2016-11-091-0/+2
| | | | | | | | | | | | | | | | | | | | | | Without this fix, the function would still end up returning NULL but it would put that NULL connection in the hash table which would be bad. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net> Reviewed-by: Dave Airlie <airlied@redhat.com> Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com> Cc: "13.0" <mesa-stable@lists.freedesktop.org> (cherry picked from commit fc0e9e3e40e4b044ce1b62c1b757941f4ed4c820)
| * radeonsi: fix BFE/BFI lowering for GLSL semanticsNicolai Hähnle2016-11-091-3/+34
| | | | | | | | | | | | | | | | Fixes spec/arb_gpu_shader5/execution/built-in-functions/*-bitfield{Extract,Insert} Cc: 13.0 <mesa-stable@lists.freedesktop.org> Reviewed-by: Marek Olšák <marek.olsak@amd.com> (cherry picked from commit 5aef14932ac047dc5f1af311a26b7f41b140d79f)
| * radv: expose xlib platform extensionDave Airlie2016-11-091-0/+6
| | | | | | | | | | | | | | | | | | | | I missed this when I added the xlib code, this allows dolphin emu to start and crash later. Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl> Cc: "13.0" <mesa-stable@lists.freedesktop.org> Signed-off-by: Dave Airlie <airlied@redhat.com> (cherry picked from commit 9f0726f3e509c80c78ddb5e7411fa34f676de71d)
| * anv/device: Return DEVICE_LOST if execbuf2 failsJason Ekstrand2016-11-091-6/+4
| | | | | | | | | | | | | | | | | | | | | | | | This makes more sense than OUT_OF_HOST_MEMORY. Technically, you can recover from a failed execbuf2 but the batch you just submitted didn't fully execute so things are in an ill-defined state. The app doesn't want to continue from that point anyway. Signed-off-by: Jason Ekstrand <jason@jlekstrand.net> Cc: "13.0" <mesa-stable@lists.freedesktop.org> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> (cherry picked from commit c41ec1679f02d6705ceda49d9ad966d0d0833824)
* | HACK: i915: enable image external samplingMauro Rossi2016-11-011-0/+1
| | | | | | | | | | | | | | | | | | This hack is the alternative way of enabling GL_OES_EGL_image_external as opposed to defining mesa environment variable in init.rc As a reference this hack works on i915 also because of commit "android: a workaround for GL_INVALID_OPERATION (0x502) error" which allows for external image being mapped despited not being dma-buf
* | i915: add mock implementation of GL_OES_EGL_image_external (v2)Mauro Rossi2016-11-015-0/+6
| | | | | | | | | | | | | | | | | | | | | | This is similar to commit 7420c9dab4aaf87e6b840410226c296c4668a48f but for the i915 driver, as neeeded in android-x86. In version v2, as per i965 implementation, the extension GL_OES_EGL_image_external is not enabled by default. The by-the-book way to enable the extension is mesa environment variable MESA_EXTENSION_OVERRIDE="+GL_OES_EGL_image_external" to be set in init.rc
* | android: intel/genxml: generate aubinator xml headers (v2)Mauro Rossi2016-11-011-10/+35
| | | | | | | | | | | | | | | | | | | | | | | | | | | | New generated headers were introduced by commit 63a366a "intel: aubinator: generate a standalone binary" The new aubinator generated files are listed in Makefile.sources variable, rules are added to Android build system for future use. New macro xml-header-gen is basically a copy&paste of automake rules, but xxd is invoked directly, as Android build systems stops and shows the error in case of xxd missing. Existing macro header-gen is renamed to pack-header-gen as a refinement.
* | Add the RGBA8888 and RGBX8888 back to Intel screen configXu, Randy2016-11-011-0/+2
| | | | | | | | | | | | Otherwise the OGLES native app may cause primary screen format mismatch Signed-off-by: Xu, Randy <randy.xu@intel.com>
* | st/dri: remove trailing whitespaceMauro Rossi2016-11-011-1/+1
| |
* | android: st/mesa: fix building error in libmesa_st_mesaMauro Rossi2016-11-011-1/+3
| | | | | | | | Fixes building error due to dependency on nir generated headers
* | egl/android: update swrast loader and extensions storageMauro Rossi2016-11-011-10/+16
| | | | | | | | | | | | | | | | | | | | | | | | Changes are applied to swrast loader and extensions storage in the same way as per commit 6a8fe32 "egl/android: don't populate dri2_dpy->dri2_loader_extension" and commit f871946 "egl/dri2: rework dri2_egl_display::extensions storage" For simplification, primitives in droid_swrast_loader_extensions are listed in the same order as per commit 2dbe14a "egl/wayland: don't populate dri2_dpy->swrast_loader_extension"
* | android: fix building errors on Android 7.0Chih-Wei Huang2016-11-013-9/+3
| |
* | nv30 locking fixesIlia Mirkin2016-11-012-2/+22
| |
* | nouveau: more locking - make sure that fence work is always done withIlia Mirkin2016-11-014-4/+17
| | | | | | | | the push mutex acquired
* | WIP nouveau: add lockingIlia Mirkin2016-11-0130-45/+372
| |
* | android: avoid using fallback functionPaulo Sergio Travaglia2016-11-011-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | When a target different than EGL_NATIVE_BUFFER_ANDROID is used with eglCreateImageKHR on sw rendering, a fallback funcion is used. As this returns NULL, EGL_NO_IMAGE_KHR is returned, which crashes many apps Using a function implemented for EGL/DRI2 in this cases allows eglCreateImageKHR usage for other targets (like EGL_GL_TEXTURE_2D_KHR)
* | android: more fixes for llvmpipe software renderingChih-Wei Huang2016-11-012-2/+11
| | | | | | | | | | * add dri2_create_from_texture to driswImageExtension * add dri2FenceExtension to drisw_screen_extensions
* | android: add a fast path for swrastWuZhen2016-11-011-8/+16
| | | | | | | | | | | | | | | | | | | | Observed increased antutu score on 1024x768, should see more with larger resolution NO_REF_TASK Tested: local run Change-Id: Ia05e0551a65ae7678561b167c1f5e75509ea4e84
* | virgl: fix null pointer exceptionsChih-Wei Huang2016-11-011-0/+2
| |
* | android: support swrastWuZhen2016-11-0113-22/+545
| | | | | | | | | | | | | | | | | | | | System boots up with gles_mesa/softpipe/llvmpipe. NO_REF_TASK Tested: local run Change-Id: I629ed0ca9fad12e32270eb8e8bfa9f7681b68474 Signed-off-by: Chih-Wei Huang <cwhuang@linux.org.tw>
* | android: enable x86 asm and sse4 for x86 and x86_64WuZhen2016-11-017-36/+47
| | | | | | | | | | | | | | | | | | | | Support multilib compilation use runtime feature selection. NO_REF_TASK Tested: local run Change-Id: Iee7961effdecde09cfbdaf09455bfb0912699ae3 Signed-off-by: Chih-Wei Huang <cwhuang@linux.org.tw>
* | android: print debug info to logcatWuZhen2016-11-014-3/+20
| | | | | | | | | | | | | | | | | | | | Redirect logs printed to stderr to logcat. NO_REF_TASK Tested: local run Change-Id: I58e3966a608af361b86c54b4c95a92561b711968 Signed-off-by: Chih-Wei Huang <cwhuang@linux.org.tw>
* | st/dri: fix double free of dri_drawableWuZhen2016-11-011-2/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In the callchain destroy_surface->destroyDrawable->dri_put_drawable-> dri_put_drawable->DestroyBuffer By the semantic of it, dri_destroy_buffer should not free drawable struct, all vendor specific and legacy swrast version of the function do not. wonder why no body else ran into this. NO_REF_TASK Tested: local run Change-Id: Ibe82d82d2e34b162e64bf0b8805f8a4553d362d5 Signed-off-by: Chih-Wei Huang <cwhuang@linux.org.tw>
* | android: change some PIPE to SVGA3D format mappingsChih-Wei Huang2016-11-011-0/+6
| | | | | | | | | | | | | | | | This is a try-and-error patch which fixes the Android-x86 black screen issue of VMware on Linux host. Tested OK on VMware Workstation 12 Player. But the red and blue colors are exchanged. Note it doesn't affect VMware on Windows host.
* | gallium/radeon: define some prototypes of LLVMInitialize functionsChih-Wei Huang2016-11-012-8/+52
| |
* | android: re-generate git_sha1.h if git HEAD updatedChih-Wei Huang2016-11-011-1/+1
| | | | | | | | | | | | | | The git_sha1.h has to depend on the git HEAD otherwise it will never be updated. Signed-off-by: Chih-Wei Huang <cwhuang@linux.org.tw>
* | egl/android: fix missing gralloc_drm_get_gem_handle dependencyRob Herring2016-11-011-1/+3
| | | | | | | | | | | | | | | | | | gralloc_drm_get_gem_handle has been removed from AOSP drm_gralloc. Remove the dependency on it and just access the struct gralloc_drm_handle_t directly. Maybe there's a better way than accessing the gralloc private data. Signed-off-by: Rob Herring <robh@kernel.org>
* | HACK: glsl: disable error on non-constant array indexingRob Herring2016-11-011-1/+1
| | | | | | | | | | | | | | DRM HWC has a shader program that hits this error. Work-around it by ignoring the error. Signed-off-by: Rob Herring <robh@kernel.org>
* | gallium: introduce load_pipe_screen()Rob Herring2016-11-014-3/+14
| | | | | | | | | | | | | | | | | | | | | | | | Introduce load_pipe_screen() public entry point for other code which dlopen()'s gralloc_dri.so for purposes of loading a pipe_screen. This way drm_gralloc can avoid static linking of each gallium winsys and driver, and avoid duplicated logic to figure out which pipe driver to load. This is based on Rob Clark's work. I moved it into pipe_loader which seems to be a better spot. Signed-off-by: Rob Herring <robh@kernel.org>
* | Android: Export gallium_dri include filesRob Herring2016-11-011-0/+6
| | | | | | | | | | | | | | This doesn't work yet because the exported include files can't be picked up by the android build system unless the library has a 'lib' prefix. Signed-off-by: Rob Herring <robh@kernel.org>
* | android: add support for LLVM 3.7.0 for marshmallowMauro Rossi2016-11-011-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | The changes add support for LLVM 3.7.0 for marshmallow, while keeping support for LLVM 3.5.0 with lollipop. MESA_LLVM_VERSION_PATCH=0 is compatible with radeonsi build in lollipop-x86, since mesa 11.0 and newer do not check anymore for LLVM 3.5.2 This changes, combined with specific R600 patches for external/llvm, enable building gallium radeonsi driver in marshmallow-x86. The patch is applicable to 11.2.0devel, 11.1 and 11.0 branches.
* | android: a workaround for GL_INVALID_OPERATION (0x502) errorChih-Wei Huang2016-11-011-0/+2
| | | | | | | | | | | | | | | | Inspired by the nice work of Paulo Sergio (pstglia): "Bypass a check for external target enabled only for images created with EGL_EXT_image_dma_buf_import - this makes Android graphical components receive a GL_INVALID_OPERATION (0x502)."
* | i915: Support Android RGBX8888 format for EGL generated imagesSean V Kelley2016-11-013-0/+4
| | | | | | | | | | | | | | | | | | | | | | Enabled MESA_FORMAT_R8G8B8X8_UNORM (formerly MESA_FORMAT_RGBX8888_REV) for RGBX with i915. Android software requires RGBX8888 format to be supported for software rendering. Previously had done so for i965 but have requests for i915 support. Change-Id: I87c3a8e3f206d75d58b63182ed94851749740762 Signed-off-by: Sean V Kelley <sean.v.kelley@linux.intel.com>
* | i915: enable GL_APPLE_texture_2D_limited_npot in mesa 11.1Mauro Rossi2016-11-011-0/+1
| | | | | | | | | | | | | | | | | | | | | | This is the port of Chia-I Wu commit 79ac0e to mesa 11.1 branch, where the extensions went to a dedicated file extensions_table.h Original commit message: "Enable GL_ARB_texture_non_power_of_two for GLESv1 to advertise GL_APPLE_texture_2D_limited_npo. The former is actually a superset of the latter, but we really want to enable NPOT support in some way."
generated by cgit v1.1 at 2025-08-17 00:36:34 +0000