)]}'
{
  "log": [
    {
      "commit": "097a423f1b40bd4ead0d41219b59b204ebbdcb54",
      "tree": "81a7593655f1e0fa026d5fcdeb33ebe2c9428057",
      "parents": [
        "9e222360dbfc4a356a8cec6ebb5d2dfb3b581ac6"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Oct 28 19:12:38 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Oct 28 19:12:38 2026 +0100"
      },
      "message": "parisc: SEABIOS_HPPA_VERSION 27\n\nMostly changes to improve support for 715 machine and FPU:\n\n- Enable FPU at CPU startup\n- Improve detection of monarch CPU with new CPU_HPA at bootup\n- Initialize LASI IAR, DINO IAR0 \u0026 IAR1 and TOC_ADDR registers with CPU HPA\n- Enable Harmony audio chip on LASI on B160L and 715/64\n- Add ibgcc replacement for 32-bit firmware with arith64 library\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "9e222360dbfc4a356a8cec6ebb5d2dfb3b581ac6",
      "tree": "d0bec7690aee9e406aaac9f49929cbe899a666ee",
      "parents": [
        "6406bceeb9aab63e00528e716e6c9423b457d18e"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Oct 27 22:56:43 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Oct 27 22:56:43 2026 +0100"
      },
      "message": "parisc/B160L: enable Harmony audio chip on LASI\n\nEnable the \"harmony\" sound chip integrated in LASI.\n\nSigned-off-by: Keith Monahan \u003ckeith@techtravels.org\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "6406bceeb9aab63e00528e716e6c9423b457d18e",
      "tree": "adf85ce783fffa6289cc50bb474c2da996f5f5a0",
      "parents": [
        "27e542649c1687724da00563e019b38a61ee2af5"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Oct 27 21:54:22 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Oct 27 21:54:22 2026 +0100"
      },
      "message": "parisc: Improve detection of monarch CPU at bootup\n\nWe allow various CPU HPA addresses, but additionally we need\nto improve the detection of the monarch CPU.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "27e542649c1687724da00563e019b38a61ee2af5",
      "tree": "71c5a5e8c7f97218571776c2b0467117e675e966",
      "parents": [
        "79db301ea8bc8a850996061186c925104c859913"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Oct 27 21:24:44 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Oct 27 21:24:44 2026 +0100"
      },
      "message": "parisc: Initialize DINO IAR0, IAR1 and TOC_ADDR with CPU HPA\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "79db301ea8bc8a850996061186c925104c859913",
      "tree": "cb26d967e383e7b10fc30852a05186dd915e8f36",
      "parents": [
        "2d2ef59c5d3c2f307a5de30848837b85b0f127c9"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Oct 27 21:06:38 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Oct 27 21:06:38 2026 +0100"
      },
      "message": "parisc: Initialize LASI IAR register\n\nStore CPU HPA IAR addresss into LASI IAR register.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "2d2ef59c5d3c2f307a5de30848837b85b0f127c9",
      "tree": "8017dff3ec7c7fdf1176cb3cf239a97f43894f35",
      "parents": [
        "c9bccb11451a97f0821130c4551a1d79e62d88d3"
      ],
      "author": {
        "name": "Keith Monahan",
        "email": "keith@techtravels.org",
        "time": "Sun Oct 25 17:46:43 2026 +0000"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Mon Oct 26 23:29:33 2026 +0100"
      },
      "message": "parisc/715: enable Mirage Jr Audio [2:0:8]\n\nHP-UX 9.05 beeps through the audio chip, and it panics if there isn\u0027t one,\nbecause the 715/64 has no HIL keyboard to beep through instead.\nChange firmware to report it to the OS. QEMU emulation needs to answer\nthe chip\u0027s registers and report a running audio clock.\n\nSigned-off-by: Keith Monahan \u003ckeith@techtravels.org\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "c9bccb11451a97f0821130c4551a1d79e62d88d3",
      "tree": "9f6dd0659ef9b5efb69edfe4a42173e4dc10fdce",
      "parents": [
        "d12e1a61130cbfa6d420b49da33b6486a36d85e3"
      ],
      "author": {
        "name": "Kirill A. Korinsky",
        "email": "kirill@korins.ky",
        "time": "Mon Oct 26 22:46:15 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Mon Oct 26 22:51:14 2026 +0100"
      },
      "message": "parisc: Enable FPU at CPU startup\n\nIn the firmware code we should not use FPU instructions.\nBut we still link in libgcc runtime functions which use the FPU, so\nwe need to enable the FPU here.\n\nSigned-off-by: Kirill A. Korinsky \u003ckirill@korins.ky\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "d12e1a61130cbfa6d420b49da33b6486a36d85e3",
      "tree": "aa82a2a91a45cf2065404e719b31af4aedb33dde",
      "parents": [
        "d33a19dfed266116cb29a895aefbbcc901d18a0e"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Mon Oct 26 22:34:42 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Mon Oct 26 22:38:36 2026 +0100"
      },
      "message": "parisc: Add built-in libgcc replacement for 32-bit firmware\n\nlibgcc still uses FPU instructions which we want to avoid.\nThe new file arith64.c provides most instructions and __muldi3 is the\nonly function left (on 32-bit) which we would need to add.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "d33a19dfed266116cb29a895aefbbcc901d18a0e",
      "tree": "14ed5d9cdce35f2f4ed5a35d9458946960971ae4",
      "parents": [
        "fe8137320cb8ee3f1f33d60a0ecf480f27ca4959"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Thu Oct 01 22:41:56 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Oct 02 20:09:34 2026 +0200"
      },
      "message": "parisc: SEABIOS_HPPA_VERSION 26\n\nNew SeaBIOS-hppa release with many fixes and updates for QEMU v11.2:\n- Allow QEMU to set monarch CPU_HPA at runtime, independend of CPU_HPA\n  value which was used at compile-time of SeaBIOS\n- Avoid potential deadlock of serial output at bootup\n- Add support for serial ports of SuperIO chip\n- Fix PDC_COPROC_CFG when called from a 32-bit process on a 64-bit PDC\n- Fix computation of IPL checksum over the loaded image\n- Assert at link time that the firmware fits below FIRMWARE_END\n- Fix various narrow PDC calls like PDC_CACHE_INFO, PDC_IODC_READ,\n  PDC_TOD_READ\n- Return PDC_TOD calibration value in the PDC 2.0 layout\n- Read the CPU clock rate from fw_cfg /etc/hppa/cpu-mhz\n- Various Elroy PCI IRQ fixes for A400 machine\n- Make lsi-scsi more robust against bus setup changes\n- Allow lsi-sci to drive the 53C810 SCSI chip\n- Various fixes to allow to boot from SCSI tape\n- Implement some PDC_PAT_MEM cell PDT functions\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "fe8137320cb8ee3f1f33d60a0ecf480f27ca4959",
      "tree": "8dd5c7790a9b6ea36f40715e803560d595d75bfc",
      "parents": [
        "010300faf08ce1916082183eb2a3bdfbc7f1a484"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:27:16 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Oct 02 20:09:34 2026 +0200"
      },
      "message": "parisc: Allow for different CPU_HPA addresses\n\nUntil now the CPU_HPA address had to be in sync between SeaBIOS and QEMU. With\nthis patch, SeaBIOS will detect if the first (monarch) CPU has another CPU_HPA\nvalue, and adjust all inventory tables accordingly.\nThe HPA is stored in the hpa0_cpu variable, and since it has to survive a\nSeaBIOS reset where SeaBIOS zeroes all of the BSS section, we need to put this\nvariable into the \"data\" section.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "010300faf08ce1916082183eb2a3bdfbc7f1a484",
      "tree": "be3325123d3582411ed50c15491e9451f3a315de",
      "parents": [
        "22055e1085e230261bf4a8bcda3a7924f5b7f09c"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Sep 29 21:39:10 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Add PDC_EEPROM function for ODE mapper on C3700\n\nThe ODE mapper tool tries to find the memory sizes on the C3700 (with\n64-bit PDC but called in narrow mode).\nIn the end the mapper tool calls the EEPROM functions.\nImplement the basic code, but return errors for now.\nODE mapper will the error out.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "22055e1085e230261bf4a8bcda3a7924f5b7f09c",
      "tree": "6df8a46f4736d0e27f9fa406a0031ca445972eee",
      "parents": [
        "a0cc0f199c291106eef16783f5288f8c53fc0a9d"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Sep 29 21:07:09 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Fix ODE to detect coprocessor in narrow mode\n\nThis fixes ODE to detect the FPU on 64-bit C3700 and A400 machines.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "a0cc0f199c291106eef16783f5288f8c53fc0a9d",
      "tree": "8e6a48652b015358328d94c73ce2115df9549aa9",
      "parents": [
        "4aa50fd7c541f603015e8f1738e0713f9c3807eb"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Sep 29 20:57:33 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Fix memory detection on C3700 with ODE\n\nODE calls the old PDC_MEM memory functions. Report back what ODE\nexpects to successfully continue.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "4aa50fd7c541f603015e8f1738e0713f9c3807eb",
      "tree": "9b68cd82265acbf1b76b15f49bfb33b5deb509ac",
      "parents": [
        "6ffb13a0da4854ac5814fe24a75f1d602a2ea7e6"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 09:05:03 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc/a400: Number CPU modules from 160 on the A400\n\nThe module number of a CPU is part of its hardware path, which the OS\nshows (e.g. ioscan on HP-UX, the device tree on Linux).  All Astro\nbased machines numbered their CPUs from 32, which is right for the\nC3700 (\"Allegro W2 [32]\" in c3700.h), but on the A400 and A500 the\nfirst CPU is module 160 (\"Crescendo DC- 440 [160]\" in a400.h,\n\"Crescendo 650 W2 [160]\" in a500.h).\n\nUse 160 as base on the A400.  The C3700 stays at 32.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "6ffb13a0da4854ac5814fe24a75f1d602a2ea7e6",
      "tree": "21b72190eaf669abfd27902bf5971ef2e50d4163",
      "parents": [
        "fc08078d8e75e23b6f8eb0738bf2ad4535eeaf07"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 09:04:40 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc/a400: Second Elroy is at 0xfed34000 on rope 2\n\nOn the A400 (and A500) the second Elroy PCI host bridge is connected to\nAstro rope 2 at HPA 0xfed34000, as a400.h\u0027s own device list header\n(\"Elroy PCI Bridge [0:2] at 0xfffffffffed34000\") and the generated\na500.h show.  The firmware used the C3700 values instead: the device\nlist entry pointed at ELROY2_HPA (0xfed32000, marked \"XXX\") and its\nmodule path said rope 1.\n\nAdd ELROY4_HPA and use it and rope 2 for the A400: in the device list,\nfor the PCI config space access of bus 1 (elroy_offset()) and for the\nrope-to-bus mapping of the PAT PCI tables (pathnum_to_pcibus()).  The\nC3700 is unchanged.\n\nThis pairs with the QEMU change \"hw/hppa: Place the second Elroy of the\nA400 on rope 2\", which moves the emulated Elroy to the same address.\nWith only one side applied, the second Elroy of an A400 is reported at\nan address where the emulator has none (no PCI devices sit behind it by\ndefault).\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "fc08078d8e75e23b6f8eb0738bf2ad4535eeaf07",
      "tree": "dcef3f197bc150e65888eb328ed4f68337ee359b",
      "parents": [
        "f4f0b3f702a38226c54fe46abefa0ccfb196df41"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 09:01:49 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Honour fw_cfg memsplit-addr for the PAT low-memory split\n\nOn PAT-only machines the firmware hard-coded the end of the low memory\nregion to 1 GB, while QEMU already publishes where it actually splits\nmemory in /etc/hppa/memsplit-addr.  Read that value first and use it\nfor the split, falling back to 1 GB if it is not provided.\n\nNo functional change with current QEMU, which publishes 1 GB for the\nA400, but the firmware no longer silently disagrees with the machine\nmodel if QEMU ever moves the split.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "f4f0b3f702a38226c54fe46abefa0ccfb196df41",
      "tree": "878abc256143ac9d2e6a994b91d1833acbb4a344",
      "parents": [
        "b6ad0abda55f7a59f8edb10cb0638740edf30b18"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 09:01:19 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Refuse PDC_SYSTEM_MAP on PAT-only machines\n\nThe PDC PAT specification lists PDC_SYSTEM_MAP in \"Table 1-13. PA-RISC\n2.0 Procedures Obsolete on Single Cell PDC PAT Platforms\"; the module\ninventory is obtained with PDC_PAT_CELL instead.  Return PDC_BAD_PROC\nfor it on PAT-only machines (the A400), as for the other procedures in\nthat table which are already refused there.\n\nPDC_NVOLATILE is only listed as obsolete on multi-cell platforms\n(Table 1-14), so it stays as it is.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "b6ad0abda55f7a59f8edb10cb0638740edf30b18",
      "tree": "0cdf28bcfddf55feec7dd33ad023500408b21273",
      "parents": [
        "34ea8382cc750a380e9181a693f4edd5f0ad3c58"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 09:00:39 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Implement PDC_PAT_MEM cell PDT info and set good memory\n\nThe PAT specification lists PDC_PAT_MEM option 5 (Get Cell PDT Info)\nfor single- and multi-cell platforms, and option 9 (Set Good Memory) as\noptional.  Both returned PDC_BAD_OPTION.\n\nImplement option 5 with an empty page deallocation table (the emulated\nmemory has no errors): no current and no maximum entries, good_mem as\nset via PDC_MEM, and no double bit error location.  Accept option 9;\nall emulated memory is good, so there is nothing to do.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "34ea8382cc750a380e9181a693f4edd5f0ad3c58",
      "tree": "7e85e3612aa3fcbd17577783af2d7c658c88c704",
      "parents": [
        "f4872230d107aa286d6ab2730cee7b7c6a1c55ff"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:58:27 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: PDC_PAT_IO unaligned config reads and GET_PCI_CONFIG_FROM_HW\n\nTwo fixes to PDC_PAT_IO:\n\n- PCI_CONFIG_READ with a length of 4 at an offset which is not 32-bit\n  aligned (e.g. offset 0x09 to read the class code plus the following\n  byte) returned the aligned dword instead of the four bytes starting\n  at the offset.  Assemble the value from single byte reads in that\n  case.  Aligned reads are unchanged.\n\n- Implement option 11, \"Get PCI Configuration Address from Hardware\n  Path\", the inverse of option 12 which is implemented already.  It\n  returns the PCI config address (bdf \u003c\u003c 8) of the device at the given\n  hardware path, or PDC_INVALID_ARG if no such PCI device exists.  Both\n  options are listed for single- and multi-cell platforms in the PAT\n  specification.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "f4872230d107aa286d6ab2730cee7b7c6a1c55ff",
      "tree": "dec5fd80bab0d6dbe483701d4f6d5b3612957230",
      "parents": [
        "ef3c1e8ab331fd88d573ef2f2a77b36cdae951e3"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:55:52 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Validate and clear PAT return buffers\n\nThe PDC return buffer (R_addr) is \"a doubleword-aligned block of 32\ndoublewords allocated by the caller\" (PDC 2.0 specification).  For the\nPAT procedures which store into it:\n\n- refuse a misaligned R_addr with PDC_INVALID_ARG,\n- clear all 32 doublewords before filling in the results, so that\n  return values which the emulation does not set read as zero instead\n  of whatever the caller had in the buffer.\n\nThis is done for PDC_PAT_CELL GET_NUMBER and MODULE, PDC_PAT_CPU\nGET_NUMBER, PDC_PAT_EVENT GET_CAPS and SCAN, and PDC_PAT_PD\nGET_ADDR_MAP.  GET_ADDR_MAP additionally refuses a misaligned address\nmap buffer.\n\nPDC_PAT_CPU GET_NUMBER returned -1 as CPU number for an HPA which does\nnot belong to a CPU; return PDC_INVALID_ARG instead.\n\nPAT is only enabled on 64-bit machines which do not allow a 32-bit OS,\nso callers are normally wide.  A narrow caller\u0027s buffer holds 32 words\nonly, so it is not cleared.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "ef3c1e8ab331fd88d573ef2f2a77b36cdae951e3",
      "tree": "6de765be0a92d210c034bbc0492fe1db0ad0d881",
      "parents": [
        "e1267bba96aaab18cfaf8df1c613a1fe96f08bf0"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 11:41:51 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Report tape boot devices as sequential\n\nThe boot device class was always reported as CL_RANDOM, both in PAGE0\nmem_boot and by IODC ENTRY_INIT.  Report CL_SEQU when booting from a\nsequential device, and in that case let ENTRY_INIT SRCH_FRST return\nthe target and LUN of the tape instead of those of the first hard\ndisk.  ENTRY_INIT reports the class of the boot drive for the storage\ncontroller, not of each individual drive behind it.\n\nDisk and CD-ROM boots are not affected.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "e1267bba96aaab18cfaf8df1c613a1fe96f08bf0",
      "tree": "e0f74eb93c23a83d61b78b94ea2c38d0b5a18011",
      "parents": [
        "d420240852860d747832831d3f75eccc70f54979"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 11:41:23 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Boot from tape in variable-block mode\n\nBooting from a tape did not work:\n\n - The IPL and the boot sector were read with READ(10), which tape\n   drives do not implement (SCSI-2 defines READ(6) for sequential-\n   access devices).\n - IODC ENTRY_IO BOOTIN turns the medium byte address in ARG5 into an\n   LBA.  A tape has none; it reads from its current position, so a\n   boot loader which reads the medium at several addresses got the\n   wrong data.\n\nRead tapes one record at a time with cdb_read_tape_record(), i.e.\nREAD(6) in variable-block mode, into a 64 KiB record buffer, and treat\nthe boot medium as the concatenation of the records.  tape_read()\nreturns the bytes at a medium address: it rewinds the tape if the\naddress lies before the buffered record and reads forward to it\notherwise.  The boot menu reads the LIF header and the IPL, and\nENTRY_IO BOOTIN reads the boot device, through it.\n\nReading records in variable-block mode works for tapes written with\nany record length up to 64 KiB (a longer record stops the boot with a\nmessage), not only for records of the drive\u0027s block size.  The\ndrive\u0027s mode parameters are neither needed nor changed: a READ(6) with\nthe FIXED bit clear transfers one variable-length record whatever\nblock length the drive is set to (SCSI-2, 10.2.4).\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "d420240852860d747832831d3f75eccc70f54979",
      "tree": "2ccec948669f2e358e4c89afee9606753eb712d0",
      "parents": [
        "c86dd571d2d7f915e9159cbcda0ce31762b0f759"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 11:25:36 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "blockcmd: Add a function which reads one tape record\n\nAdd cdb_read_tape_record() for the tape boot of the parisc firmware,\nand make cdb_rewind_tape() available to it.\n\nIt reads the next record with READ(6) and the FIXED bit clear, i.e. in\nvariable-block mode: one record of up to the given buffer size is\ntransferred, whatever block length the drive is set to, so the\ndrive\u0027s mode parameters need not be changed.  SILI is left clear, so\na record of another length than requested ends in CHECK CONDITION with\nILI set and the INFORMATION field holding the requested length minus\nthe actual length (SCSI-2, 10.2.4).  REQUEST SENSE then gives the\nlength of a short record.  (With SILI set, a short record would end\nwith GOOD status, and \"the residue information normally provided in\nthe information field of the sense data may not be available\", SCSI-2\nnote 140.)\n\nA filemark, the end of the recorded data, an error, or a record longer\nthan the buffer make the function fail.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "c86dd571d2d7f915e9159cbcda0ce31762b0f759",
      "tree": "3d1601d62d3239c3910c4802130af70ca0ddc255",
      "parents": [
        "d71a38787ee939ce9d4c7cbfbd6fbf581589acec"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 11:24:20 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "lsi-scsi: Complete a request whose data phase ends early\n\nThe script moves op-\u003ecount * blocksize bytes and then reads the status\nbyte.  If the target transfers less and goes to the status phase\nearly, e.g. when a tape record is shorter than the buffer, the chip\nstops the script with a phase mismatch interrupt.  The driver then\ngave up with the command still in progress on the bus, so the next\nrequest could not select the target.\n\nOn a phase mismatch in the status phase, continue the script at the\nstatus instruction once, so that the status and message bytes are\nread and the bus is released.  The request still fails unless the\nstatus is GOOD; a caller which needs the residue gets it from the\nsense data.\n\nTested with QEMU\u0027s LSI model only.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "d71a38787ee939ce9d4c7cbfbd6fbf581589acec",
      "tree": "7762389fd00155cc9a6c987345c51349706ead8e",
      "parents": [
        "7375166a27c69a58bf6c5aaccaafc24acb947e20"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:52:41 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "lsi-scsi: Also drive the 53C810\n\nThe driver only binds to the 53C895A.  QEMU\u0027s LSI model also provides\nthe 53C810 (\"-device lsi53c810\"), which therefore could not be used as\na boot controller.  Accept the 53C810 as well.  The driver uses only\nbasic SCRIPTS (select, block move, jump on phase, interrupt) and a few\nstatus registers.\n\nTested with QEMU\u0027s lsi53c810 model only.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "7375166a27c69a58bf6c5aaccaafc24acb947e20",
      "tree": "687eb23f7b1c2815cc8424b494b2c3666164b805",
      "parents": [
        "79d9862abaece62041a3f7b397be7adfe1205bcd"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:52:25 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "lsi-scsi: Revalidate the I/O BAR before each request\n\nThe driver caches the I/O base of the controller at init time.  An OS\nor an offline diagnostic which owns the bus may re-size and re-assign\nthe PCI BARs, or temporarily clear PCI_COMMAND_IO, and later call the\nfirmware for boot I/O again (IODC ENTRY_IO).  With I/O decoding off,\nthe port accesses go nowhere, reads return 0xff and the request fails\nwithout ever reaching the chip; with a moved BAR the firmware talks to\nthe old address.\n\nRe-read BAR0 before each request and re-enable I/O space and bus\nmastering if they are off.  The BAR is only read, never rewritten, so\nif the guest moved it, the driver follows.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "79d9862abaece62041a3f7b397be7adfe1205bcd",
      "tree": "ec5de89f62f5d519712e4fafe8c4e4575ef3d01b",
      "parents": [
        "fb99ce837f9412f30e10e5e0ae761b197e393f45"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:47:36 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Rotate IRT interrupt lines by PCI pin on the A400\n\nOn the A400, spread the four interrupt pins of a PCI slot over the\nIOSAPIC inputs: INTA of slot n goes to input n, INTB to n+1, and so on\n(modulo ELROY_IRQS), so that the functions of a multi-function device\ndo not all share one line.\n\nAdd is_a400_machine() for this, which other A400 specific changes will\nuse as well.\n\nThis must be paired with the QEMU change \"hw/hppa: Rotate Elroy PCI\ninterrupts by pin on the A400\", which makes elroy_pci_map_irq() do the\nsame rotation for the A400.  The C3700 is unchanged on both sides.\nWith only one side applied, functions using INTB..INTD on an A400 get\nan IRT entry for an IOSAPIC input which never fires.\n\nNothing in this rotation is specific to the A400 hardware; it is\nlimited to the A400 only to leave the C3700 as it is.  No OS which\nuses these interrupt lines could be booted on either machine here, so\nthe choice is yours if it should apply to all Astro machines.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "fb99ce837f9412f30e10e5e0ae761b197e393f45",
      "tree": "343fdfcaa36687865e9d212d2f31f8729e55fdbe",
      "parents": [
        "c4c3f7c4d959e4363b5ffb1e74b006f09efebbc1"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:45:02 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Report the PCI slot as IOSAPIC input in the IRT\n\nQEMU\u0027s Elroy model routes the interrupt of a PCI device to the IOSAPIC\ninput which equals its slot number (elroy_pci_map_irq() returns\n\"slot \u0026 (ELROY_IRQS - 1)\").  The interrupt routing table which the\nfirmware builds for the OS instead numbered the IOSAPIC inputs with an\nenumeration counter.\n\nBoth only agree while the n-th PCI function found sits in slot n.  For\na multi-function device, or a gap in the slot numbers, the OS\nprogrammed an IOSAPIC line which never fires.  Use the slot number,\nas the emulated hardware does.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "c4c3f7c4d959e4363b5ffb1e74b006f09efebbc1",
      "tree": "aae8fb0fd7364ef9e901b5ed53ed6b3e75142ea3",
      "parents": [
        "443f7ee57e4ee9f6b5538ece3c3da07c45b200d5"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:43:30 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Read the CPU clock rate from fw_cfg /etc/hppa/cpu-mhz\n\nThe firmware reports a fixed 250 MHz CPU (CPU_CLOCK_MHZ) in several\nplaces: PDC_TOD timer calibration, PAGE0 mem_10msec (the interval timer\nticks per 10 ms), the PAT cpu_speed field and the boot banner.  If the\nemulator runs the interval timer at a different rate, operating\nsystems which trust these values keep time incorrectly.\n\nAllow QEMU to publish the rate as the 4-byte fw_cfg file\n/etc/hppa/cpu-mhz and use it for all of the above.  Without the file,\ne.g. with current QEMU, everything stays at 250 MHz.  Values outside\n1..4000 MHz are ignored, so that the rate in Hz still fits into 32 bits.\n\nThe calibration value is an IEEE double; build it from the integer\nrate with integer operations, because the 32-bit firmware is not linked\nwith floating point support.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "443f7ee57e4ee9f6b5538ece3c3da07c45b200d5",
      "tree": "feb0efc3f81736e468b8db10d9287a4a93e0c254",
      "parents": [
        "e386e2249a51d5da2cda5ca81c198732b34b0047"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:41:29 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Return PDC_TOD calibration value in the PDC 2.0 layout\n\nPDC_TOD option 2 (calibrate timers) returns the Interval Timer\nfrequency as an IEEE double split into two return parameters.  The PDC\n2.0 specification says: \"The first 32-bits ... are contributed by\ncalib_0, the remaining 32-bits of the fraction field are defined by\ncalib_1.  Each of these values are returned in the low order (bits\n32:63) bits of their return parameter.\"\n\nThe 64-bit firmware stored the whole double into RET[0] instead, so a\nwide caller found the lower half of the double in calib_0 and nothing\nin calib_1, and after the 64-to-32-bit result conversion a narrow\ncaller got the lower half as calib_0 as well.\n\nStore the upper and lower 32 bits into RET[0] and RET[1].  On 32-bit\nfirmware the result is unchanged, because there the double already\nspanned RET[0] and RET[1] in this order.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "e386e2249a51d5da2cda5ca81c198732b34b0047",
      "tree": "47d3c202cdbace90c6b20301c4e59c655f23d1dc",
      "parents": [
        "4d1b3f9a892aefbbd1edafe57b5bd7c485902d24"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:40:49 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Fix the comment of PDC_PSW_ENDIAN_BIT\n\nPDC_PSW_ENDIAN_BIT is the \"e\" field of the PDC_PSW default mask.  It\nmirrors the PSW E-bit, for which the PA-RISC 2.0 architecture says:\n\"When 0, all memory references are big endian.  When 1, all memory\nreferences are little endian.\"  The comment claimed the opposite.\n\nThe code in pdc_psw() already uses the bit correctly (it refuses little\nendian mode), so this is a comment fix only.  It avoids someone\n\"fixing\" that BUG_ON() the wrong way around.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "4d1b3f9a892aefbbd1edafe57b5bd7c485902d24",
      "tree": "0563191c55b681d8e5cce1900c68e16e531dc146",
      "parents": [
        "b5b5872462aa141b0124bc0b94ba27692b542064"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:39:53 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: PDC_ADD_VALID accepts RAM above the low-memory split\n\nOn 64-bit machines with more memory than fits below the low-memory\nsplit, the remaining RAM is mapped at RAM_MAP_HIGH2 and RAM_MAP_HIGH1,\nand PDC_MEM_TABLE / the PAT address map report it there.\n\nPDC_ADD_VALID only knew about the low region, so for an address in the\nhigh windows (which are above 4 GB) it fell through to the device\nchecks and returned a requestor bus error, although the memory exists.\nAccept addresses inside the mid and high RAM windows.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "b5b5872462aa141b0124bc0b94ba27692b542064",
      "tree": "ba8ccadad2e4982cc60f0bf88532b8bb176f22ff",
      "parents": [
        "635d88922ce4220a9d533fe48a32b954867ff0d7"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:38:29 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Answer PDC_PIM HPMC and LPMC requests with \"invalid PIM contents\"\n\nThe PDC specification requires every processor to provide PDC_PIM\noption 0 (transfer HPMC data), and processors which support LPMCs to\nprovide option 2 (transfer LPMC data).  If no such event was recorded,\nthe call returns status -5 (invalid PIM contents).\n\nSeaBIOS returned PDC_BAD_OPTION for both, i.e. claimed the options do\nnot exist.  Since machine checks are not emulated there is never any\nHPMC or LPMC data, so return PDC_NE_MOD (-5) with actcnt 0 instead.\nAn OS asking for the PIM of a previous machine check at boot now gets\nthe architected \"nothing recorded\" answer.\n\nThe TOC and soft boot options are unchanged.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "635d88922ce4220a9d533fe48a32b954867ff0d7",
      "tree": "445ef7339d0d20ac08ac80505de419a681e84325",
      "parents": [
        "a9d2f958d2874f2f7151eb14e30312d09a39db3e"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:37:15 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Fix narrow PDC results that the generic conversion cannot handle\n\nWhen a 32-bit (narrow) OS calls the 64-bit PDC, the entry code converts\nthe 64-bit result buffer to 32-bit words after the call returns.  That\nworks for most procedures, but not in these cases:\n\n- PDC_CACHE_INFO copied the whole wide struct pdc_cache_info (240\n  bytes) into the caller\u0027s buffer, which is only 32 words (128 bytes)\n  for a narrow caller.  Store the 32-bit values directly.\n\n- PDC_IODC_READ returning PDC_IODC_COUNT: the conversion is skipped\n  for negative return values, so a narrow caller read the upper half\n  of the 64-bit entry size, i.e. 0.  Store 512 as a 32-bit word.\n\n- PDC_TOD_READ with a result buffer below MEM_PDC_ENTRY: the\n  conversion code halts the machine for such buffers.  Store tod_sec\n  and tod_usec as 32-bit words directly.\n\nFor this, pass narrow_mode to pdc_cache(), pdc_hpa(), pdc_iodc() and\npdc_tod().\n\nAlso add pdc_check_raddr() and let PDC_HPA_PROCESSOR refuse a\nmisaligned return buffer with PDC_INVALID_ARG (doubleword alignment for\nwide callers, word alignment for narrow callers) before storing into\nit.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "a9d2f958d2874f2f7151eb14e30312d09a39db3e",
      "tree": "5976201537dceb7f1fd22e7b16d576b1ddadcfa6",
      "parents": [
        "9a61ece2a446fa829a7644fbf15f61fff2ca7553"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:30:59 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Match device HPAs at 32-bit width\n\nfind_hpa_device() sign-extends the HPA it is given (COMPAT_VAL), while\nsome device table entries are built from plain 32-bit constants, e.g.\n\".hpa \u003d ELROY2_HPA\" in c3700.h and a400.h, which the 64-bit firmware\nstores zero-extended.  find_hppa_device_by_hpa() compares without any\nextension.  So depending on how the table entry was built and how the\ncaller passes the HPA, the lookup of an existing device can fail.\n\nHPAs are 32-bit I/O addresses, so compare the low 32 bits.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "9a61ece2a446fa829a7644fbf15f61fff2ca7553",
      "tree": "f6ebb8605b7ad95d98aea2c123e0a098d3d9c33c",
      "parents": [
        "6523b1ee38ac23395a9d09d9208fd499ae499e85"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:29:16 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Assert at link time that the firmware fits below FIRMWARE_END\n\nQEMU maps the firmware image into the FIRMWARE_START..FIRMWARE_END\nwindow, but nothing in the linker script checked that _firmware_end\nstays inside it.  An image which grows past the window links cleanly\nand only fails (or silently misbehaves) at boot.\n\nAdd an ASSERT() for both the 32-bit and the 64-bit layout so that such\na build fails at link time instead.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "6523b1ee38ac23395a9d09d9208fd499ae499e85",
      "tree": "42cb649298db796e99fd7f2fe0cac7ddc9f299f5",
      "parents": [
        "ba5af96f150c74f0e53a064826ced17c27854b03"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:28:04 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Do not halt on unimplemented PDC calls or unknown IODC HPAs\n\nMake two debug messages usable:\n- the PAT PDC_FIND_MODULE trace was disabled with \"if (0)\"; enable it\n  with DEBUG_PDC.\n- \"CHASSIS READ LOG NOT IMPLEMENTED\" was printed unconditionally on\n  every call, which floods the console when an OS polls the chassis\n  log.  Print it only with DEBUG_CHASSIS.  The return code is unchanged.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "ba5af96f150c74f0e53a064826ced17c27854b03",
      "tree": "7976d88dae0411c373e4fd672d019cd9d93a9f28",
      "parents": [
        "c5ded0b7f8cc552015a48fa30fbb1404698bd94b"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:26:58 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Fix LUN offset of the alternate boot path in stable storage\n\nThe alternate boot path in stable storage is a struct pdc_module_path at\noffset 0x80.  Its layers[] are 32-bit big-endian words starting at\nbyte 8, so the target (layers[0]) is the low byte at +11 and the LUN\n(layers[1]) is the low byte at +15.\n\nThe code stored the LUN at +12, the most significant byte of\nlayers[1].  Nothing noticed as long as the CD-ROM sits at LUN 0, but a\nnon-zero LUN produced a bogus alternate boot path for anyone reading\nit back via PDC_STABLE (ISL, the OS).\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "c5ded0b7f8cc552015a48fa30fbb1404698bd94b",
      "tree": "8d44b473164c504dad616fea203786b04f7030bc",
      "parents": [
        "1fb980fc1004be6e742ddb781ca96af4f37cd0b5"
      ],
      "author": {
        "name": "Craig Lalley",
        "email": "mr_lalley@yahoo.com",
        "time": "Mon Sep 28 08:26:25 2026 -0700"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Sep 30 22:41:56 2026 +0200"
      },
      "message": "parisc: Compute the IPL checksum over the loaded image\n\nThe IPL checksum loop walked ipl_addr, which is the byte offset of the\nIPL on the boot medium, as if it were a pointer into memory.  The IPL\nhas just been read to \u0027target\u0027, so sum the words there instead.\n\nWith the old code the check read unrelated low memory and could print\na spurious \"IPL checksum is not 0\" warning for a valid boot image.\n\nSigned-off-by: Craig Lalley \u003cmr_lalley@yahoo.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "1fb980fc1004be6e742ddb781ca96af4f37cd0b5",
      "tree": "77678930adac6d307f25ef93a7d774b4e903ae1c",
      "parents": [
        "f146311e2d30334316ede7a3833bb7999a8f5b64"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Sep 29 17:19:21 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Sep 29 19:27:14 2026 +0200"
      },
      "message": "parisc: Copy compat results if target buffer \u003e\u003d PAGE_SIZE\n\n... instead of MEM_PDC_ENTRY as before.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "f146311e2d30334316ede7a3833bb7999a8f5b64",
      "tree": "1784b7a8fb7a67e24df5551a95148433d6603ecc",
      "parents": [
        "c41dfde8ea14cb40f24aa1553e0337b8ef671b3d"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Sep 12 00:02:33 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Sep 12 00:18:19 2026 +0200"
      },
      "message": "parisc: Drop unsused variable to avoid compiler warning\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "c41dfde8ea14cb40f24aa1553e0337b8ef671b3d",
      "tree": "903d067e4b645aed5801a76746decbd92bbd3118",
      "parents": [
        "d078cfcb6747d30de74ec7fb49df0e1f353fdec8"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Sep 12 00:01:45 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Sep 12 00:18:19 2026 +0200"
      },
      "message": "string: Make memset() function externally visible\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "d078cfcb6747d30de74ec7fb49df0e1f353fdec8",
      "tree": "b5bb6a37d10513b5023ef51b3fc09fb4952808cb",
      "parents": [
        "261bf85c96851cb5de92c422c5796097252de29e"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Sep 12 00:00:46 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Sep 12 00:18:19 2026 +0200"
      },
      "message": "ncr710: Avoid warning about unused poll_count variable\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "261bf85c96851cb5de92c422c5796097252de29e",
      "tree": "1d94ea51bac4e4a1b679619d6d7590e8c1926110",
      "parents": [
        "2af0ad566472abdb52bcb012ccef7291b1f2ada6"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Aug 22 11:06:12 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Aug 22 11:06:12 2026 +0200"
      },
      "message": "parisc: Add SuperIO PCI serial port to BCH output\n\nIf a PCI serial device is used for the boot console, show it in the BCH\nhandler. This improves output for the C3700 machine with serial port on\nSuperIO chip.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "2af0ad566472abdb52bcb012ccef7291b1f2ada6",
      "tree": "775cfd8111d2fa7c24a9440630b9f011639e1292",
      "parents": [
        "4426c1ddfc292fb98e6edc47dd7cadcc5fccf15f"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Aug 22 10:54:06 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Aug 22 10:54:06 2026 +0200"
      },
      "message": "parisc: Fix PDC_COPROC_CFG when called from a 32-bit process on a 64-bit PDC\n\nWhen Linux starts the secondary CPUs, they call at startup the\nPDC_COPROC_CFG PDC function to enable the numerical coprocessor. For a\n32-bit Linux kernel on a 64-bit PDC (like for the C3700 machine), the\nmemset() in the PDC_COPROC_CFG PDC call initialized 32 64-bit values in\nthe caller-provided return buffer, and as such the double amount of\nbytes were written back to the caller than what a 32-bit process\nexpected. This finally crashed the 32-bit Linux kernel when booting with\nmultiple CPUs, since coincidentally the in-kernel \"root\" variable, which\nstores the in-kernel device tree, was then overwritten with zero and\nthus lead to random kernel crashes.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "4426c1ddfc292fb98e6edc47dd7cadcc5fccf15f",
      "tree": "3d5d1d0d5c4cf097f5b1060a9b0567ea795cec50",
      "parents": [
        "073c255758ce7211c78eb8af588c989c06e928af"
      ],
      "author": {
        "name": "Abizer Lokhandwala",
        "email": "abizerlokhandwalastd10@gmail.com",
        "time": "Sat Jul 04 11:26:43 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Jul 18 20:07:44 2026 +0200"
      },
      "message": "parisc: Add support for SuperIO chip\n\nAllow PDC to detect the NS SuperIO chip as built-in on various C-class\nmachines, like C3700. In SeaBIOS we want to allow booting from the IDE\nchip, and output to the serial ports.\n\nSigned-off-by: Abizer Lokhandwala \u003cabizerlokhandwalastd10@gmail.com\u003e\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "073c255758ce7211c78eb8af588c989c06e928af",
      "tree": "5ebcc0252ff12965ff34af449d8bf4624564b73f",
      "parents": [
        "6d6a72c218f6bf6b0e0b6a8bd26c5c67ba3ad2e0"
      ],
      "author": {
        "name": "mrlalley",
        "email": "70232803+mrlalley@users.noreply.github.com",
        "time": "Sat Jul 18 02:53:57 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sat Jul 18 11:53:57 2026 +0200"
      },
      "message": "parisc: bound the LSR poll in parisc_serial_out() to avoid a firmware deadlock (#7)\n\nIf the guest OS transiently unmaps or resizes the console UART\u0027s PCI\nBAR (e.g. the standard 0xffffffff BAR-sizing probe issued through\nPDC_PAT_IO), any firmware print in that window polls the LSR at a\ncached MMIO address that no longer decodes.  Reads of unbacked\naddresses return 0, the ready test ((lsr \u0026 0x60) \u003d\u003d 0x60) can never\nbe satisfied, and SeaBIOS spins forever INSIDE the PDC call -- so the\nguest never regains control to restore the BAR.  Measured on QEMU\nhppa (A400/C3700 family) booting MPE/iX: 4,114,337 rejected reads\nfrom a single print, byte-identically reproducible.\n\nBound the poll: after SERIAL_OUT_POLL_LIMIT dry iterations, drop the\ncharacter and return.  A transiently-unmapped console now costs a few\ndropped characters instead of a permanent machine hang.\n\nSigned-off-by: Mr Lalley \u003cmr_lalley@yahoo.com\u003e"
    },
    {
      "commit": "6d6a72c218f6bf6b0e0b6a8bd26c5c67ba3ad2e0",
      "tree": "d5d64bb1a251c4d92efe26b22fb8b00e95b22bfb",
      "parents": [
        "96d9f4b79aa5f196f8d9cb8561b0f26a27af09a3"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Jul 03 20:43:34 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Jul 03 20:43:34 2026 +0200"
      },
      "message": "parisc: SEABIOS_HPPA_VERSION 25\n\nNew SeaBIOS-hppa release with various fixes and updates for qemu-v11.1:\n- Work around SCSI transfer limit on 715\n- Fix PDC_MODEL call to not copy PSW default word on 32-bit machines\n- Fix PDC_SYSTEM_MAP call to prevent boot issues on C3700 machine\n- Fix cc_line in pdc_cache to allow BSDs to boot\n- Initial support to boot from SCSI TAPE drives\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "96d9f4b79aa5f196f8d9cb8561b0f26a27af09a3",
      "tree": "c482fd1a287a70a3b82f7206cfa52bb06657765b",
      "parents": [
        "65cb731ff79ca5deef4c72d20e22cfcbd62fe15f"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Jul 03 20:26:43 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Jul 03 20:26:43 2026 +0200"
      },
      "message": "parisc: Wire up SCSI TAPE boot\n\nAllow IPL to detect and boot from SCSI TAPE.\nNecessary code in QEMU is still pending.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "65cb731ff79ca5deef4c72d20e22cfcbd62fe15f",
      "tree": "aabc7890c36dec0455ca5d3cfbc3f62acb328675",
      "parents": [
        "f13b4fb9fc02dea124fd97627c5712d327d84a45"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Jul 03 20:23:15 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Jul 03 20:23:15 2026 +0200"
      },
      "message": "parisc/block: Add block functions for SCSI TAPE support\n\nAdd code that block layer can read from SCSI TAPE.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "f13b4fb9fc02dea124fd97627c5712d327d84a45",
      "tree": "3024a2e5d434a0005780744b18fa6d3c109f0276",
      "parents": [
        "b7951b5564a71f85a4f7668ff4a60a8590f96869"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Jul 03 20:18:02 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Jul 03 20:18:02 2026 +0200"
      },
      "message": "parisc/block: Add constants for TAPE support\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "b7951b5564a71f85a4f7668ff4a60a8590f96869",
      "tree": "23d7704af2b4411d6605c3c0849fb5a7c0e9a406",
      "parents": [
        "c745e4ef436cfdc205f4dc72b4e9e6507ea9f5d5"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Jun 07 19:19:08 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Jul 03 20:15:40 2026 +0200"
      },
      "message": "parisc: Fix cc_line in pdc_cache for BSD\n\nIt was reported that NetBSD needed a value of 2 for cc_line.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "c745e4ef436cfdc205f4dc72b4e9e6507ea9f5d5",
      "tree": "bd84f481c6e6c89634894c2229ea3d7f07f42172",
      "parents": [
        "e0dae63d75d99fa27d4c05a6b79a3c0b225740ad"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Jul 03 14:35:26 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Jul 03 20:12:45 2026 +0200"
      },
      "message": "parisc: Fix PDC_MODEL call to not copy PSW default word\n\nThe PSW default word is used by 64-bit machines, so avoid copying this\nfield on 32-bit only machines.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "e0dae63d75d99fa27d4c05a6b79a3c0b225740ad",
      "tree": "2816b16aa4dca0b6389727311c675f1dd0e41d25",
      "parents": [
        "2f1f907c1452a1dc72ab2adbc035f8f08c4b8151"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue May 19 22:55:49 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue May 19 22:55:49 2026 +0200"
      },
      "message": "parisc: Fix memory overwrite of PDC_SYSTEM_MAP PDC_FIND_MODULE PDC call\n\nFix pdc_call(PDC_SYSTEM_MAP, PDC_FIND_MODULE, ...) to not overwrite\nthe ARG2 result struct. This fixes a boot problem on emulated C3700.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "2f1f907c1452a1dc72ab2adbc035f8f08c4b8151",
      "tree": "0a858bac9fe9aba49f36bc7ee8f39f2fca4aa935",
      "parents": [
        "d9560852a34f156155b3777745baa0d96d553f22"
      ],
      "author": {
        "name": "Michael Tokarev",
        "email": "mjt@tls.msk.ru",
        "time": "Wed Mar 25 12:32:51 2026 +0300"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Thu Apr 02 15:36:35 2026 +0200"
      },
      "message": "parisc: Include missing \u003cstdbool.h\u003e\n\nThe parisc.c file uses boolean type and true/false.  Normally, this\nneeds a header file \u003cstdbool.h\u003e to be included - unless you use compiler\nin C23 mode, which pre-defines these.\n\nSigned-off-by: Michael Tokarev \u003cmjt@tls.msk.ru\u003e\n"
    },
    {
      "commit": "d9560852a34f156155b3777745baa0d96d553f22",
      "tree": "adc7fabd61897a7d5b879469c75bd86b3e5695ff",
      "parents": [
        "be4e01b29dfea726e66f9d744c2d876e7e7e05e1"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 31 20:16:05 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 31 20:16:05 2026 +0200"
      },
      "message": "parisc: Work around SCSI transfer limit on 715\n\nThe SCSI controller of the 715 machine has a transfer limit of 64k.\nAllow loading IPL (bootloader) files bigger than that by doing multiple\nreads with each less than 64k. This allows loading the bootloader of\nHP-UX 11iv1 on the 715.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "be4e01b29dfea726e66f9d744c2d876e7e7e05e1",
      "tree": "47bde9d17e7931c1a3d4cad94afaa0ec11b1d9e3",
      "parents": [
        "007b04e54f017ad1fec449a8a92377ecf590a308"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Mon Mar 30 21:11:58 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Mon Mar 30 21:11:58 2026 +0200"
      },
      "message": "parisc: SEABIOS_HPPA_VERSION 24\n\nNew SeaBIOS-hppa release with various fixes for qemu-v11:\n- Initialize Astro chip with relevant LMMIO, GMMIO and others\n- Disable Artist on 64-bit machines\n- Fully implement parisc memory ranges\n- Change inventory to list CPU first for HP-UX\n- Allow 715 to boot from HP-UX 11iv1 CD-ROM\n- Prepare for MULTICELL machines\n- Initialize PCI_CACHE_LINE_SIZE PCI value\n- Call TOC/HPMC handler code from OS if installed\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "007b04e54f017ad1fec449a8a92377ecf590a308",
      "tree": "1a5f02d9b852f2e47cb0d54366a1b9de27216fd4",
      "parents": [
        "22dc95b23a86dad3f93629fc2916544860939eab"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Mon Mar 30 00:09:51 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Mon Mar 30 21:09:18 2026 +0200"
      },
      "message": "parisc: Fully implement parisc memory ranges\n\nOn PARISC there are up to 3 memory ranges on 64-bit machines.  PAT machines\nsplit memory at 1GB, older 64-bit machines keep up to 3.75 GB in low memory,\ne.g. on C3700 this will be mapped to support 32-bit applications.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "22dc95b23a86dad3f93629fc2916544860939eab",
      "tree": "42c840459853a50c43aa0b61e90399e6566ed7a2",
      "parents": [
        "618c2e3ff5066469a52d23a7c8b382409d5e7eb6"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:14:54 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:57:54 2026 +0200"
      },
      "message": "parisc: Initialize Astro chip with relevant LMMIO, GMMIO and others\n\nQemu now supports the relevant Astro registers, so initialize them\nto sane defaults.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "618c2e3ff5066469a52d23a7c8b382409d5e7eb6",
      "tree": "ecaf4dc46eb33ae30e47a0f4ee64cd35011f812c",
      "parents": [
        "9f285de5f657b7b41731c3cf41524986bf56ed9c"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:10:57 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:57:54 2026 +0200"
      },
      "message": "parisc: Adjust flags of hwpath to include cell node nr.\n\nOn PAT machines the cell number of the drive is encoded in the lower\nbits of the flag variable. Set it to zero.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "9f285de5f657b7b41731c3cf41524986bf56ed9c",
      "tree": "dc825d749c57a768de7ba9455ff93c715f7a60d0",
      "parents": [
        "6c85dc372ce9681130f5436ae533446d43fbacfc"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:06:51 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:57:54 2026 +0200"
      },
      "message": "parisc: Fix HPA_is_astro_mmio() to detect PCI range\n\nUse pcimem_start and pcimem_end to detect range.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "6c85dc372ce9681130f5436ae533446d43fbacfc",
      "tree": "a493816978333e4ec33efee3e5b261dfff2d63b3",
      "parents": [
        "c934b8bca5b50eb5c57897971307f5de4e0c6822"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:04:29 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:57:54 2026 +0200"
      },
      "message": "parisc: Provide writeq() function\n\nThis will be needed to write LMMIO and other registers from Astro.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "c934b8bca5b50eb5c57897971307f5de4e0c6822",
      "tree": "4a794d91a5fdcfdb1cedb91b2fd803902d776ccc",
      "parents": [
        "6dbf902739acae88089464d2d9281a42c9a497fa"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:02:46 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:57:54 2026 +0200"
      },
      "message": "pciinit: Initialize PCI_CACHE_LINE_SIZE entry\n\nThis avoid a warning when starting HP-UX 11.23:\nSCSI: PCI command register PCI_CS_CACHE_LINE_SIZE not set for bus: -1\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "6dbf902739acae88089464d2d9281a42c9a497fa",
      "tree": "cb9fba4a8be1197830e71958ab7a1b913a07e7f6",
      "parents": [
        "deaf936b280e5b94115a883482267b70cd1fbbb2"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:00:39 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:57:54 2026 +0200"
      },
      "message": "parisc: Disable artist on 64-bit machines\n\nThe Artist graphics card can not be used on 64-bit machines,\nso disable it by default on 64-bit ROMs.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "deaf936b280e5b94115a883482267b70cd1fbbb2",
      "tree": "fdffb8baf868038ac7d42f470b2deb77183a2d45",
      "parents": [
        "ad3922ae111355b50ab9389cb3a5efeb378b220d"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 19:58:50 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:57:54 2026 +0200"
      },
      "message": "parisc: CPU must be first in list for HP-UX\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "ad3922ae111355b50ab9389cb3a5efeb378b220d",
      "tree": "1789c494dffe6d49dda2521a717d7c154f44e2cc",
      "parents": [
        "b3cf5fb313c95a4349ca500368a31362160d72f9"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 19:39:35 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:57:54 2026 +0200"
      },
      "message": "parisc: Call TOC/HPMC handler code from OS if installed\n\nUse SeaBIOS-hppa own TOC handler as long as we start.\nBefore jumping into IPL or kernel code, install a toc handler which\nsimply runs the \"break\" assembly command. That way we may get\na long syslog output if the HP-UX kernel breaks.\nThis simplyfies HP-UX debugging.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "b3cf5fb313c95a4349ca500368a31362160d72f9",
      "tree": "9c4ff1507478c9fe880a5d6ccf3af750caa0375b",
      "parents": [
        "cff11f784f1331d3bbe0f0cddb104f29342c3c52"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 19:13:54 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:57:54 2026 +0200"
      },
      "message": "parisc: Introduce MULTICELL variable\n\nand disable specific funtions on multicell machines, according to the\nPAT2.5 documentation.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "cff11f784f1331d3bbe0f0cddb104f29342c3c52",
      "tree": "2bfddb049355a9fd663e434d52097eec8bd1a4c4",
      "parents": [
        "08b0821eece27b1e64da3130bb286c1f396a5eaf"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 19:09:37 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:57:54 2026 +0200"
      },
      "message": "parisc: Adjust space address to get disassmbly from qemu\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "08b0821eece27b1e64da3130bb286c1f396a5eaf",
      "tree": "3bba762d319d3480446038d54a8f0a0a4835cce7",
      "parents": [
        "413096b6d6bea3af6185ace599413c10eb270ab0"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 19:07:08 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:57:54 2026 +0200"
      },
      "message": "parisc: Fix pim ptr check\n\nThe address of pim11 and pim20 can be at 0x30100fff........\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "413096b6d6bea3af6185ace599413c10eb270ab0",
      "tree": "861fdf43a7406d8d11c426e4d0849760ad864064",
      "parents": [
        "cf3a472f4ddea30d2d483a8ba4b005505f0dabb3"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 19:04:52 2026 +0200"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 29 20:57:54 2026 +0200"
      },
      "message": "parisc: Allow 715 to boot from HP-UX 11iv1 CD-ROM\n\nignore the per-SCSI max_bytes_transfer for now.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "cf3a472f4ddea30d2d483a8ba4b005505f0dabb3",
      "tree": "e6d28ef377644c28e321dd227a054b4d7adc84ad",
      "parents": [
        "8dda10248c6c371d7ffde25eb0d64c3776cce5e4"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 20:01:44 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 20:25:41 2026 +0100"
      },
      "message": "parisc: SEABIOS_HPPA_VERSION 23\n\nNew SeaBIOS-hppa release with various fixes for qemu-v11:\n- Various fixes to support CPUs with 40 and 44 bits physical address\n- Fix PAT_CPU call when asking for current CPU\n- Add function to stop CPU in PDC_PROC\n- Prevent execution of some functions when running as PAT firmware\n- Add HP rp2470 (A500-6X) machine description\n- Tune cache parameters to speed up cache flushes in operating systems\n- Use Wide SCSI mode on 64-bit\n- Allow callling PDC_MODEL_GET_PLATFORM_INFO\n- Revert resetting LSI SCSI with ODE on HP3000\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "8dda10248c6c371d7ffde25eb0d64c3776cce5e4",
      "tree": "18571d0b6b827ef0c810711e24b1f28b0a22d98c",
      "parents": [
        "869779bd9c9fd5a9ab7716755d23b84c73815c1c"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 19:53:00 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 20:25:41 2026 +0100"
      },
      "message": "Revert \"parisc: ODE for HP3000 needs SCSI reset\"\n\nThis reverts commit ec7704ea8bd8d89feedde4c491cee8bc718db5c9.\nIt seems we had a memory overwrite in PAT_CPU (fix in \"parisc: Fix\nPAT_CPU call wrt calling CPU\") due to which we had to reset the SCSI\nbus. This is now not needed any longer.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "869779bd9c9fd5a9ab7716755d23b84c73815c1c",
      "tree": "14b404de362adc7c92d561150c5b0475979e8175",
      "parents": [
        "ec6f6ed42dc415539a517da2ac1c1c65844d414e"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 15 22:27:04 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 20:25:41 2026 +0100"
      },
      "message": "parisc: Use Wide SCSI mode on 64-bit and allow PDC_MODEL_GET_PLATFORM_INFO\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "ec6f6ed42dc415539a517da2ac1c1c65844d414e",
      "tree": "cadd5c3f03ce2442cc470f893fe8793d841d3152",
      "parents": [
        "5bf044e21e556d2f354233d8dac1512103dc512b"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sun Mar 15 18:13:30 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 20:25:41 2026 +0100"
      },
      "message": "parisc: Enhance comments wrt dest_lid and id_eid\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "5bf044e21e556d2f354233d8dac1512103dc512b",
      "tree": "5949380f0759d72a8b9e962922f69c97a167df35",
      "parents": [
        "3a73b596ce907f6e2e7201e820e1e2366f619d4d"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Mar 14 23:35:15 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 20:25:41 2026 +0100"
      },
      "message": "parisc: Tune cache settings reported by PDC\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "3a73b596ce907f6e2e7201e820e1e2366f619d4d",
      "tree": "805cc15c1b6cca63f6f7e6edfd2976c05d58e6dc",
      "parents": [
        "0a92317895cecded568fbe573a6a9b3f769002dc"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Mar 14 22:34:18 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 20:25:41 2026 +0100"
      },
      "message": "parisc: Add function to stop CPU in PDC_PROC\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "0a92317895cecded568fbe573a6a9b3f769002dc",
      "tree": "4d2d6d61e8f16f6300c088e5a4646f311d5995a1",
      "parents": [
        "87e275361b83936b0983c9b2f455a79e5e16fa65"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Mar 14 00:05:36 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 20:25:41 2026 +0100"
      },
      "message": "parisc: Fix PAT_CPU call wrt calling CPU\n\nARG3\u003d-1 means to return HPA or number of that CPU which\nis calling the PDC function.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "87e275361b83936b0983c9b2f455a79e5e16fa65",
      "tree": "f8f4eed10768e2653aefc83e95e7842788f64458",
      "parents": [
        "c85c3470fc4d21d6163c419d8aa0cb467414c937"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Mar 13 19:20:54 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 20:25:41 2026 +0100"
      },
      "message": "parisc: Fix boot on CPUs with 44 bits physical addresses\n\nAdjust ASTRO_BASE_HPA and ELROY0_BASE_HPA to the 44-bit physical\naddress space. Prefer F_EXTEND() macro to extend the addresses\ninto full 64-bit address space.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "c85c3470fc4d21d6163c419d8aa0cb467414c937",
      "tree": "ce4638833a7bc572a50c113090e4db9d12c7eb5d",
      "parents": [
        "27716df271c6c944a925c0f6ce1a2035e34d53f8"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Mar 13 18:32:54 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 20:25:41 2026 +0100"
      },
      "message": "parisc: Use FIRMWARE_HIGH constant\n"
    },
    {
      "commit": "27716df271c6c944a925c0f6ce1a2035e34d53f8",
      "tree": "5906a0ec3c3d43a683290cf6d54ef332355e9b2e",
      "parents": [
        "9ac5070becb464cebceeda1aca6b24866d7d603d"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Mar 13 17:43:47 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 20:25:41 2026 +0100"
      },
      "message": "parisc: Add HP rp2470 (A500-6X) machine description\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "9ac5070becb464cebceeda1aca6b24866d7d603d",
      "tree": "cc017ba45f6f38999d63853cef189888716f6025",
      "parents": [
        "0f3b5bc06d1229ef45c89c9f5d3b33eff6d25ab2"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Feb 27 11:57:21 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 20:25:41 2026 +0100"
      },
      "message": "parisc: Use hppa seabios version as IODC version\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "0f3b5bc06d1229ef45c89c9f5d3b33eff6d25ab2",
      "tree": "2ad7132379cae728fd55897a9cba6378d23fb790",
      "parents": [
        "1a8ada1fb70643172e251aacbac673c9ecda99e9"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Feb 27 11:50:19 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Tue Mar 17 20:25:41 2026 +0100"
      },
      "message": "parisc: Block some functions in 64-bit PAT firmware\n\nBlock functions which are not implemented when running a 64-bit PAT\nfirmware, e.g. on HalfDome/W.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "1a8ada1fb70643172e251aacbac673c9ecda99e9",
      "tree": "c347835395bd6bbc9b7aa08fadfbb1284c87fbe0",
      "parents": [
        "b9ac10c1e803e09b5f143161bf419783b3219352"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Feb 13 18:53:27 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: SEABIOS_HPPA_VERSION 22\n\nAdd A400-44 machine with initial functional 64-bit PAT PDC support:\n- Linux 64-bit kernel runs nicely.\n- ODE 2006 works nicely on 715, B160L, and A400, but has problems on C3700.\n- MPE and 64-bit HP-UX11 need more work, although HP-UX11 boot up to a crash\n  where it reports where it had problems.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "b9ac10c1e803e09b5f143161bf419783b3219352",
      "tree": "28717ed1cb73ba34ac15b0dd632e7703837388cf",
      "parents": [
        "39595e283341f3e79b5a4f6589f6847edea9b548"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 16:41:35 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: Add optional code for PDC_RELOCATE\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "39595e283341f3e79b5a4f6589f6847edea9b548",
      "tree": "78f451df08aee654dbef486eba9f1d4f3f7b7b64",
      "parents": [
        "8b095f136ca91134f1e2e27c783ab83fe307204e"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 03:43:24 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: mention id_eid used for iommu IRQ on PAT machines\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "8b095f136ca91134f1e2e27c783ab83fe307204e",
      "tree": "4ef6b5b7417b6d44dea80767d1b469d15c10afc8",
      "parents": [
        "f0942f7c837fde37a8cda28f6c6d4af37f10367d"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 03:42:22 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: Implement pdc_pat_mem() and avoid overwrite on page0\n\nPAT_MEM() is used on A400.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "f0942f7c837fde37a8cda28f6c6d4af37f10367d",
      "tree": "23f577323efe2c718746bca7e3b98d6ed49f8123",
      "parents": [
        "54903e3d2986183a9449a4fd071b9d0e5e90bd73"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 03:40:46 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: Astro has busnum 0 on A400 machine\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "54903e3d2986183a9449a4fd071b9d0e5e90bd73",
      "tree": "3b0ed1431d4c8955178f6334bafd89a01a78edd1",
      "parents": [
        "e0ebc18b244da963e05f51c14b6aeb9d4584a5c9"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 03:39:24 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: minor cleanups and warning fix\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "e0ebc18b244da963e05f51c14b6aeb9d4584a5c9",
      "tree": "ce4bade43a813ad0841f229780b08572f6308f2b",
      "parents": [
        "58c1cf017ac8e1caca1107179c8ca781116c3a85"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 03:31:33 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: Add PDC_NVM firmware functions\n\nPDC_NVM is used on first-generation PAT machines (like A400)\nand is replaced by PAT_NVOLATILE later.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "58c1cf017ac8e1caca1107179c8ca781116c3a85",
      "tree": "c1496d7dd495f26b2563742435b6f3dc2915daac",
      "parents": [
        "024b6205dd68592f1a918b282decf4407a98f672"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 03:29:18 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: Fix PCI accesses in pat_io()\n\nThe mapper2 tool from ODE now correctly shows the PCI bus.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "024b6205dd68592f1a918b282decf4407a98f672",
      "tree": "d5a3c50874fa99d70fb9375fa688cebffce6beeb",
      "parents": [
        "fc7ad9cd4a2e6b25f0eac6ec33338f17e50905df"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 03:26:23 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: Mention obsolete PDC_BROADCAST entry\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "fc7ad9cd4a2e6b25f0eac6ec33338f17e50905df",
      "tree": "a5934e23b9552c2cd0bc1b4cf448677597c92eb9",
      "parents": [
        "fe2fb2f748f8c07050343f4d68a982dd14caf1ba"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 03:21:16 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: Improve output of hpa_name() function\n\nFix string parsing so that the human-readable name of an HPA actually is\nprinted.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "fe2fb2f748f8c07050343f4d68a982dd14caf1ba",
      "tree": "f930851cee0e3c6cff839bf4ccbe2deeb74b349a",
      "parents": [
        "e42912654432dd39154f14e9e6c56337163b6085"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Feb 14 23:23:56 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: Add PDC_PAT_CHASSIS_LOG\n\nAdd basic PAT chassis logging.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "e42912654432dd39154f14e9e6c56337163b6085",
      "tree": "6844194aa377945f07be30443a6bb68e2849b2d6",
      "parents": [
        "47841213a77a7893dd64cd374754e468903ac5de"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Feb 14 20:19:28 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: For PAT_PD/PDC_PAT_PD_GET_ADDR_MAP do not overwrite PAGE0\n\nBe careful to not overwrite PAGE0.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "47841213a77a7893dd64cd374754e468903ac5de",
      "tree": "8d0c00d8117cce221da92b23d295f8633d98e83b",
      "parents": [
        "0b5ec3478f0932ebea987b95de2e6de57e45e4a5"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Feb 14 20:03:47 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: Do not implement PDC_PAT_CELL_GET_INFO\n\nAccording to docs PDC_PAT_CELL_GET_INFO is only required for multi-cell\nmachines.\nHP-UX 11v3 boots further now.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "0b5ec3478f0932ebea987b95de2e6de57e45e4a5",
      "tree": "253ab68256a9e12047ce9f1eaf3f683886569186",
      "parents": [
        "76c5be10d86d013f3dbf9a128d11313abdb5113d"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Feb 14 19:33:45 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: Add default CPU location for PAT firmware\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "76c5be10d86d013f3dbf9a128d11313abdb5113d",
      "tree": "0c5844f9e9efd487363b7e6520a68514817f7095",
      "parents": [
        "f02c8bcd02675489f64208f6c18c8e73f7505146"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Sat Feb 14 19:15:01 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: return bad option for PDC_PAT_PD_GET_ALIVE_CELLS\n\nLet it be a single CELL machines for now.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "f02c8bcd02675489f64208f6c18c8e73f7505146",
      "tree": "7c739adc2cbe5fc56c22f5c65a025597976e8a1f",
      "parents": [
        "a2159ee7e1d7810495e3639581f418a6fd16d15e"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Feb 13 23:52:55 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Wed Feb 18 20:16:03 2026 +0100"
      },
      "message": "parisc: Disable PAT_COMPLEX to force sigle cell mode\n\nHP-UX will default to singled cell mode and reports:\npat_init_pd: will default to single cell\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "a2159ee7e1d7810495e3639581f418a6fd16d15e",
      "tree": "cd77a6d2a1bc2c3f4244998048a0e50095043f39",
      "parents": [
        "cde64ec569039d18de2bd7bf1c6376e9d006c6f5"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Feb 13 18:43:03 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Feb 13 23:15:06 2026 +0100"
      },
      "message": "parisc: Wire up A400 machine\n\nEverything is now in place. Wire up the A400 by adding the include\nfile, build up the machine description and fill the pat_info_block.\n\nThe machine is now able to boot Linux. HP-UX 11iv3 and MPE still\nneeds more debugging.\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    },
    {
      "commit": "cde64ec569039d18de2bd7bf1c6376e9d006c6f5",
      "tree": "0c1f9362b0c964adfb0b9ae058d1babc0875c316",
      "parents": [
        "3fb4c031079277d50679e0cfadaad93e47cecc4a"
      ],
      "author": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Feb 13 18:41:42 2026 +0100"
      },
      "committer": {
        "name": "Helge Deller",
        "email": "deller@gmx.de",
        "time": "Fri Feb 13 23:15:06 2026 +0100"
      },
      "message": "parisc: Add unknown PDC #27 call\n\nSigned-off-by: Helge Deller \u003cdeller@gmx.de\u003e\n"
    }
  ],
  "next": "3fb4c031079277d50679e0cfadaad93e47cecc4a"
}
