{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":336309880,"defaultBranch":"main","name":"m1n1","ownerLogin":"jannau","currentUserCanPush":false,"isFork":true,"isEmpty":false,"createdAt":"2021-02-05T15:24:29.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/486189?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1723326462.0","currentOid":""},"activityList":{"items":[{"before":null,"after":"1c5a864d796d4ef17a95a428031a87c8d8025767","ref":"refs/heads/mtd-phram-adt-log","pushedAt":"2024-08-10T21:47:42.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"kboot: Expose m1n1 console log in phram MTD reserved memory node\n\nThis registers a log buffer IODEV similar to the frame buffer IODEV. It\nis using a console IODEV to allow writing to the buffer until m1n1 jumps\nto the next stage. This allows also to increase the buffer size without\nincreasing m1n1's size. The current CONSOLE_BUFFER_SIZE of 8k is not\nlarge enough to hold the entire log of executing m1n1 under the\nhypervisor on the M1 Ultra Mac Studio.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"kboot: Expose m1n1 console log in phram MTD reserved memory node"}},{"before":null,"after":"6c9d1c15c1353bd881ac32a0da1225988a461113","ref":"refs/heads/dcp_asc-dram_mask","pushedAt":"2024-05-17T22:20:54.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"dcp: Fix firmware mapping check on t8103/t600x\n\nt8103 and t600x use \"asc-dram-mask\" in iop-dcp-nub to mask bits out of\nDMA addresses. This needs to be used in the firmware mappings\ncheck/remap since the segments have maskable bits sets.\n\nFixes: 8332e24 (\"dcp: Undo carnage from bad stage1 DART code\")\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"dcp: Fix firmware mapping check on t8103/t600x"}},{"before":null,"after":"ef8332ec4aaf80c15412198918ea577115f21db1","ref":"refs/heads/dcpext0_fb_iova_14.5_fix","pushedAt":"2024-05-16T20:19:38.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"display: Fix display init on t602x dcpext0\n\ndcpext0 rejects SetSurfaceRequest if the io address for plane[0] is 0x0.\nStart searching for free space with an offset of a single page.","shortMessageHtmlLink":"display: Fix display init on t602x dcpext0"}},{"before":null,"after":"c33c17a36ebb4531acbb33510685ef5add567422","ref":"refs/heads/sio_die1","pushedAt":"2024-04-27T11:16:47.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"sio: Rework sio init for dual die SoCs\n\nUse a struct instead of global variable to hold the necessary setup\ninformation. This allows executing the setup for both sio instances.\nIn addition change the status to \"okay\" so that sio can remain disabled\nif the setup fails.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"sio: Rework sio init for dual die SoCs"}},{"before":null,"after":"7c7bb62622380c2b186213a1b052eabc0869472f","ref":"refs/heads/t6022_dcpext4_pmgr","pushedAt":"2023-12-16T08:15:01.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"display: Fix pmgr device id for t6022 dcpext4\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"display: Fix pmgr device id for t6022 dcpext4"}},{"before":"d39d1a7ff751698902a159a508f68f9484240ac3","after":"a21d7f61f830bbf35c5dcf56490bb61d07b71147","ref":"refs/heads/t6022_use_dcpext5","pushedAt":"2023-12-12T07:53:23.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"display: t6022: Use dcpext5 for HDMI out\n\ndcpext4 (dcpext0 on die 1) seems to behave as dcp0 on t8112/t6020 mac\nminis. It doesn't wake up after display init and sleep with active boot\nframebuffer. Use dcpext5 instead which doesn't show the same behavior.\nIt's unclear why the dcp instances behave differently. Special\ninitialization by iboot for the \"first\" dcp is one possible\nexplanation.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"display: t6022: Use dcpext5 for HDMI out"}},{"before":null,"after":"ddea8baf47bfb99e1bcb8ad59ddbf542706bcab9","ref":"refs/heads/t6022_nvme","pushedAt":"2023-12-11T22:44:16.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"nvme: t6022: Derive die from nvme_base address\n\nFor some reason the ans node for t602x devices has an empty\n\"clock-gates\" property. Use the MMIO address instead to determine on\nwhich die the device is.\n\nFixes: 34f49a5 (\"nvme: assume die 0 if clock-gates not set\")\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"nvme: t6022: Derive die from nvme_base address"}},{"before":"cab8c863ffeca97e5ef0bb5d7d2f10f443995190","after":"d39d1a7ff751698902a159a508f68f9484240ac3","ref":"refs/heads/t6022_use_dcpext5","pushedAt":"2023-12-11T00:35:35.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"display: t6022: Use dcpext5 for HDMI out\n\ndcpext4 (dcpext0 on die 1) seems to behave as dcp0 on t8112/t6020 mac\nminis. It doesn't wake up after display init and sleep with active boot\nframebuffer. Use dcpext5 instead which doesn't show the same behavior.\nIt's unclear why the dcp instances behave differently. Special\ninitialization by iboot for the \"first\" dcp is one possible\nexplanation.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"display: t6022: Use dcpext5 for HDMI out"}},{"before":null,"after":"cab8c863ffeca97e5ef0bb5d7d2f10f443995190","ref":"refs/heads/t6022_use_dcpext5","pushedAt":"2023-12-10T23:06:33.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"display: t6022: Use dcpext5 for HDMI out\n\ndcpext4 (dcpext0 on die 1) seems to behave as dcp0 on t8112/t6020 mac\nminis. It doesn't wake up after display init and sleep with active boot\nframebuffer. Use dcpext5 instead which doesn't show the same behavior.\nIt's unclear why the dcp instances behave differently. Special\ninitialization by iboot for the \"first\" dcp is one possible\nexplanation.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"display: t6022: Use dcpext5 for HDMI out"}},{"before":null,"after":"8d2856619625f15a7a7eecab600bb5989294c51f","ref":"refs/heads/m2_desktop_use_dcpext","pushedAt":"2023-11-27T00:40:44.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"m1n1.fw.dcp.ipc: Add D592 - abort_swap_ap_gated\n\nAlso fixes a type for set_idle_caching_state_ap.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"m1n1.fw.dcp.ipc: Add D592 - abort_swap_ap_gated"}},{"before":null,"after":"bc2b152458115a147434e98f7b05342f4127038e","ref":"refs/heads/firmware_V13_6_2","pushedAt":"2023-11-14T23:22:56.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"kboot_gpu: Declare GPU firmware 13.6.2 as 13.5 compatible\n\nTested on t6001-j314c.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"kboot_gpu: Declare GPU firmware 13.6.2 as 13.5 compatible"}},{"before":"49c6766ff6d1ea27e957e37f5912e769a0622d98","after":"893a2955865a360f27bde00fad48dfd28b1a1744","ref":"refs/heads/t602x_dcpext","pushedAt":"2023-11-07T20:13:29.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"kboot: dcp: Enable dcpext on t600x/t8103 as well\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"kboot: dcp: Enable dcpext on t600x/t8103 as well"}},{"before":"2eea7b26ab5485d765ca37df28198e3d695febf4","after":"49c6766ff6d1ea27e957e37f5912e769a0622d98","ref":"refs/heads/t602x_dcpext","pushedAt":"2023-11-06T18:16:26.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"clk: Add missing '\\n' in printf message\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"clk: Add missing '\\n' in printf message"}},{"before":null,"after":"2eea7b26ab5485d765ca37df28198e3d695febf4","ref":"refs/heads/t602x_dcpext","pushedAt":"2023-11-06T18:12:34.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"clk: Add missing '\\n' in printf message\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"clk: Add missing '\\n' in printf message"}},{"before":null,"after":"85e90d7efdedbec6c4b1bb2a85a952c7a50f4071","ref":"refs/heads/m2_dptx_fixes","pushedAt":"2023-10-12T20:49:56.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"dcp/afk: Retry harder to init afk endpoints/services\n\nExtend start_interface() with the number of expected services so it can\nstop when all services are initialized.\n\nFixes failed display init on some devices in stage 1 due to a too slow\nIOP which does not manage to send service announce messages in 20\nmailbox / ringbuffer query operations.\n\nProper fix for this is to let the endpoint implementations drive the\ninitialization via afk_epic_work() until the expected services are\nfound.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"dcp/afk: Retry harder to init afk endpoints/services"}},{"before":"68c05e621415a6a31493a95169a25d9418a6e80d","after":"2b942b34ba23921286cb784ec04650e87fc18795","ref":"refs/heads/dptx_1st_stage_delay","pushedAt":"2023-10-06T06:52:57.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"dcp: Wait 25ms after enabling dcp/dptx-phy power-domains\n\nFixes display initialization from stage 1 m1n1. 20ms looked reliable on\nj473 but let's add 25% safety margin. Someone with more motivation (and\nHW serial) to do repeated `kmutil configure-boot ...` can try to\nminimize this and test on t602x.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"dcp: Wait 25ms after enabling dcp/dptx-phy power-domains"}},{"before":null,"after":"68c05e621415a6a31493a95169a25d9418a6e80d","ref":"refs/heads/dptx_1st_stage_delay","pushedAt":"2023-10-06T06:42:01.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"dcp: Wait 25ms after enabling dcp/dptx-phy power-domains\n\nFixes display initialization from stage 1 m1n1. 20ms looked reliable on\nj473 but let's add 25% safety margin. Someone with more motivation (and\nHW serial) to do repeated `kmutil configure-boot ...` can try to\nminimize this and test on t602x.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"dcp: Wait 25ms after enabling dcp/dptx-phy power-domains"}},{"before":"fc4a2ecbb3689a631b88c56f93e2d26cc7b10139","after":"216939aa59d5a20cdeed83d6098bc7996f129075","ref":"refs/heads/agx_resv_mem_no_map","pushedAt":"2023-10-04T18:23:16.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"kboot/gpu: Add \"no-map\" property for reserved memory nodes\n\nWithout this property u-boot will add reserved memory regions to the\nEFI memory map with the wrong type. This results in Linux mapping it\nand thus breaking the intended use in the asahi driver.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"kboot/gpu: Add \"no-map\" property for reserved memory nodes"}},{"before":"f1271b4ddce300ec586782e32282ad6714c74477","after":"ce8b852275d8a35f749f28be4b011812844f4065","ref":"refs/heads/nvram_dynamic_size","pushedAt":"2023-09-30T19:19:59.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"kboot: Parse \"anvram\" partition start and size from ADT\n\n\"nvram\" size depends on the iboot version. Monterey (12.3) and earlier\nused 128k nvram partition but Ventura since at least 13.2 increased the\nsize to 1M. Parse the values from the ADT and fill and enable the node\nin the device tree.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"kboot: Parse \"anvram\" partition start and size from ADT"}},{"before":null,"after":"f1271b4ddce300ec586782e32282ad6714c74477","ref":"refs/heads/nvram_dynamic_size","pushedAt":"2023-09-30T13:12:35.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"kboot: Parse \"anvram\" partition start and size from ADT\n\n\"nvram\" size depends on the iboot version. Monterey (12.3) and earlier\nused 128k nvram partition but Ventura since at least 13.2 increased the\nsize to 1M. Parse the values from the ADT and fill and enable the node\nin the device tree.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"kboot: Parse \"anvram\" partition start and size from ADT"}},{"before":null,"after":"fc4a2ecbb3689a631b88c56f93e2d26cc7b10139","ref":"refs/heads/agx_resv_mem_no_map","pushedAt":"2023-09-23T14:53:38.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"kboot/gpu: Add \"no-map\" property for reserved memory nodes\n\nWithout this property u-boot will add reserved memory regions to the\nEFI memory map with the wrong type. This results in Linux mapping it\nand thus breaking the intended use in the asahi driver.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"kboot/gpu: Add \"no-map\" property for reserved memory nodes"}},{"before":"ff1a4dde5d699a669a4fe992d136fbba3f835a1d","after":"ae41be12fab4fcb3b2a3fd33d0ab1f3e51d49571","ref":"refs/heads/m2_dptx_v13_5","pushedAt":"2023-09-22T18:45:03.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"display: Support the HDMI output on M2* desktop systems\n\nOn M2* desktop systems the HDMI output is not managed by dcp/dcp0. This\nallows more flexibility, for example dcp on the M2 Mac mini can be\nrouted to the USB-C/ThunderBolt ports.\nFor m1n1 this adds quite ab bit of complexity. In addition to the DCP\niboot endpoint/protocol two additional endpoints have to be started.\nOnly the dptx-port endpoint has a meaningful AP <=> DCP communication.\nIn addition we start the system endpoint to optionally enable more\nverbose syslog messages (if RTKIT_SYSLOG is defined).\nIn addition the dptx-phy or lpdptx-phy has to be programmed. Those two\nphy look mostly identical. MacOS seems to use a static DP configuration\n(4 lanes, HBR3*) regardless of the connected display.\nThe added SMC gpio support is used to power the MCDP29xx DP to HDMI\nconverter on.\nConfig  is unfortunately a mess since Apple can't name the device tree\nnodes consistently. On M2 Ultra devices only the four dcpext*/dispext*\non each die are used. M2 Ultra support is untested.\n\n* link rate might differ on t602x based devices for 8k60 support\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"display: Support the HDMI output on M2* desktop systems"}},{"before":"340ea0d9ffcfd852d36717c49b0781f7b558613a","after":"ff1a4dde5d699a669a4fe992d136fbba3f835a1d","ref":"refs/heads/m2_dptx_v13_5","pushedAt":"2023-09-20T19:29:27.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"display: Support the HDMI output on M2* desktop systems\n\nOn M2* desktop systems the HDMI output is not managed by dcp/dcp0. This\nallows more flexibility, for example dcp on the M2 Mac mini can be\nrouted to the USB-C/ThunderBolt ports.\nFor m1n1 this adds quite ab bit of complexity. In addition to the DCP\niboot endpoint/protocol two additional endpoints have to be started.\nOnly the dptx-port endpoint has a meaningful AP <=> DCP communication.\nIn addition we start the system endpoint to optionally enable more\nverbose syslog messages.\nIn addition the dptx-phy or lpdptx-phy has to be programmed. Those two\nphy look mostly identical. MacOS seems to use a static DP configuration\n(4 lanes, HBR3*) regardless of the connected display.\nStill missing but prepared is the SMC gpio support required to power the\nMCDP29xx DP to HDMI converter on.\nConfig  is unfortunately a mess since Apple can't name the device tree\nnodes consistently. On M2 Ultra devices only the four dcpext*/dispext*\non each die are used. M2 Ultra support is untested and broken due to the\nwrong clock-gate.\n\n* link rate might differ on t602x based devices for 8k60 support\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"display: Support the HDMI output on M2* desktop systems"}},{"before":"d1b5013e2badfe9354c8d9ed2568c56c3a008cd5","after":"1ca7854100b7d6867216de2b28b96fa01f68b21e","ref":"refs/heads/touchbar_non_fatal_errors","pushedAt":"2023-09-18T20:59:40.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"kboot: disable touchbar on missing calibration data\n\nThere is at least on Macbook Pro (M1, 13-inch) with missing\n\"multi-touch-calibration\" in the ADT. Disabling the device as error\nhandling instead of refusing to boot the kernel.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"kboot: disable touchbar on missing calibration data"}},{"before":null,"after":"d1b5013e2badfe9354c8d9ed2568c56c3a008cd5","ref":"refs/heads/touchbar_non_fatal_errors","pushedAt":"2023-09-18T20:57:38.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"kboot: disable touchbar on missing calibration data\n\nThere is at least on Macbook Pro (M1, 13-inch) with missing\n\"multi-touch-calibration\" in the ADT. Disabling the device as error\nhandling instead of refusing to boot the kernel.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"kboot: disable touchbar on missing calibration data"}},{"before":"dc81c87ff14cbc084749e58627c5de710cd5d4d8","after":"340ea0d9ffcfd852d36717c49b0781f7b558613a","ref":"refs/heads/m2_dptx_v13_5","pushedAt":"2023-09-11T07:03:31.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"display: Support the HDMI output on M2* desktop systems\n\nOn M2* desktop systems the HDMI output is not managed by dcp/dcp0. This\nallows more flexibility, for example dcp on the M2 Mac mini can be\nrouted to the USB-C/ThunderBolt ports.\nFor m1n1 this adds quite ab bit of complexity. In addition to the DCP\niboot endpoint/protocol two additional endpoints have to be started.\nOnly the dptx-port endpoint has a meaningful AP <=> DCP communication.\nIn addition we start the system endpoint to optionally enable more\nverbose syslog messages.\nIn addition the dptx-phy or lpdptx-phy has to be programmed. Those two\nphy look mostly identical. MacOS seems to use a static DP configuration\n(4 lanes, HBR3*) regardless of the connected display.\nStill missing but prepared is the SMC gpio support required to power the\nMCDP29xx DP to HDMI converter on.\nConfig  is unfortunately a mess since Apple can't name the device tree\nnodes consistently. On M2 Ultra devices only the four dcpext*/dispext*\non each die are used. M2 Ultra support is untested and broken due to the\nwrong clock-gate.\n\n* link rate might differ on t602x based devices for 8k60 support\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"display: Support the HDMI output on M2* desktop systems"}},{"before":"63d7a4ddfcbdbb62ec1d6b05599e6095551540e9","after":"dc81c87ff14cbc084749e58627c5de710cd5d4d8","ref":"refs/heads/m2_dptx_v13_5","pushedAt":"2023-09-09T08:21:12.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"display: Support the HDMI output on M2* desktop systems\n\nOn M2* desktop systems the HDMI output is not managed by dcp/dcp0. This\nallows more flexibility, for example dcp on the M2 Mac mini can be\nrouted to the USB-C/ThunderBolt ports.\nFor m1n1 this adds quite ab bit of complexity. In addition to the DCP\niboot endpoint/protocol two additional endpoints have to be started.\nOnly the dptx-port endpoint has a meaningful AP <=> DCP communication.\nIn addition we start the system endpoint to optionally enable more\nverbose syslog messages.\nIn addition the dptx-phy or lpdptx-phy has to be programmed. Those two\nphy look mostly identical. MacOS seems to use a static DP configuration\n(4 lanes, HBR3*) regardless of the connected display.\nStill missing but prepared is the SMC gpio support required to power the\nMCDP29xx DP to HDMI converter on.\nConfig  is unfortunately a mess since Apple can't name the device tree\nnodes consistently. On M2 Ultra devices only the four dcpext*/dispext*\non each die are used. M2 Ultra support is untested and broken due to the\nwrong clock-gate.\n\n* link rate might differ on t602x based devices for 8k60 support\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"display: Support the HDMI output on M2* desktop systems"}},{"before":"1d5a6b0ba1b942e5617703daf507c2691ee0d1dd","after":"63d7a4ddfcbdbb62ec1d6b05599e6095551540e9","ref":"refs/heads/m2_dptx_v13_5","pushedAt":"2023-09-09T07:15:35.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"display: Support the HDMI output on M2* desktop systems\n\nOn M2* desktop systems the HDMI output is not managed by dcp/dcp0. This\nallows more flexibility, for example dcp on the M2 Mac mini can be\nrouted to the USB-C/ThunderBolt ports.\nFor m1n1 this adds quite ab bit of complexity. In addition to the DCP\niboot endpoint/protocol two additional endpoints have to be started.\nOnly the dptx-port endpoint has a meaningful AP <=> DCP communication.\nIn addition we start the system endpoint to optionally enable more\nverbose syslog messages.\nIn addition the dptx-phy or lpdptx-phy has to be programmed. Those two\nphy look mostly identical. MacOS seems to use a static DP configuration\n(4 lanes, HBR3*) regardless of the connected display.\nStill missing but prepared is the SMC gpio support required to power the\nMCDP29xx DP to HDMI converter on.\nConfig  is unfortunately a mess since Apple can't name the device tree\nnodes consistently. On M2 Ultra devices only the four dcpext*/dispext*\non each die are used. M2 Ultra support is untested and broken due to the\nwrong clock-gate.\n\n* link rate might differ on t602x based devices for 8k60 support\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"display: Support the HDMI output on M2* desktop systems"}},{"before":"d82b20cebdf19bf100444e978086746a3a9492c6","after":"1d5a6b0ba1b942e5617703daf507c2691ee0d1dd","ref":"refs/heads/m2_dptx_v13_5","pushedAt":"2023-09-06T06:04:55.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"experiments/dp2hdmi.py: Do not power dp2hdmi on\n\nm1n1 will handle this on its own.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"experiments/dp2hdmi.py: Do not power dp2hdmi on"}},{"before":"716e435f50d8a7460902f6ccf2fa2e63990688fd","after":"d82b20cebdf19bf100444e978086746a3a9492c6","ref":"refs/heads/m2_dptx_v13_5","pushedAt":"2023-09-05T06:58:46.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"jannau","name":"Janne Grunau","path":"/jannau","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/486189?s=80&v=4"},"commit":{"message":"proxy: experiments: Add dp2hdmi.py for manual display config\n\nSMC GPIO support is required for powering the dp2hdmi converter on.\nUses m1n1's C implementation otherwise.\n\nSigned-off-by: Janne Grunau <j@jannau.net>","shortMessageHtmlLink":"proxy: experiments: Add dp2hdmi.py for manual display config"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAElx_fEQA","startCursor":null,"endCursor":null}},"title":"Activity · jannau/m1n1"}