1skiboot-5.3.0-rc1 2================= 3 4skiboot-5.3.0-rc1 was released on Monday July 25th, 2016 5 6skiboot-5.3.0-rc1 is the first release candidate of skiboot 5.3, which will 7become the new stable release of skiboot following the 5.2 release, first 8released March 16th 2016. 9 10skiboot-5.3.0-rc1 contains all bug fixes as of skiboot-5.1.16 11and skiboot-5.2.4 (the existing stable releases). 12 13For how the skiboot stable releases work, see :ref:`stable-rules`. 14 15The current plan is to release skiboot-5.3.0 August 1st 2016. 16 17Over skiboot-5.2, we have the following changes: 18 19OPAL API/Device Tree 20-------------------- 21 22- Reserve OPAL API numbers for XICS emulation for XIVE 23 Additionally, we put in some skeleton docs for what's coming, 24 key points being that this is for P9 and above, relies on a device 25 being present in the device tree and is modelled on the PAPR calls. 26- interrupts: Remove #interrupt-cells from ICP nodes 27- Stop adding legacy linux, phandle to device tree, just add phandle 28 No Linux kernel has ever existed for powernv that only knows linux,phandle. 29 30POWER9 31------ 32 33- Add base POWER9 support 34 In *NO WAY* is this geared towards real POWER9 hardware. 35 Suitable for use in simulators *only*, and even then, only if you 36 intensely know what you're doing. 37- Document changes in OPAL API for POWER9 38 Some things are going to change, we start documenting them. 39- cpu: supply ibm,dec-bits via devicetree 40- power9: Add example device tree for phb4 41- device-tree: Only advertise ibm, opal-v3 (not v2) on POWER9 and above 42 43CAPI 44---- 45 46- phb3: Test CAPI mode on both CAPP units on Naples 47- hmi: Recover both CAPP units on Naples after malfunction alert 48- chiptod: Sync timebase in both CAPP units on Naples 49- phb3: Set CAPI mode for both CAPP units on Naples 50- phb3: Load CAPP ucode to both CAPP units on Naples 51- phb3: Add support for CAPP DMA mode 52 The XSL used in the Mellanox CX4 card uses a DMA mode of CAPI, which 53 requires a few registers configured specially. This adds a new mode to 54 the OPAL_PCI_SET_PHB_CAPI_MODE API to enable CAPI in DMA mode. 55 56PCI 57--- 58 59- pci: Do a dummy config write to devices to establish bus number 60- phb: Work around XSL bug sending PTE updates with wrong scope 61- Support for PCI hotplug (if a platform supports it) 62 63Garrison 64-------- 65 66- NVLink/NPU support 67- Full garrison platform support. 68 69BMC based platforms 70------------------- 71 72- bt: use the maximum retry count returned by the BMC 73- SEL: Fix eSEL ID while logging eSEL event 74 Commit 127a7dac added eSEL ID to SEL event in reverse order (0700 instead 75 of 0007). This code fixes this issue by adding ID in proper order. 76 77Tests/Simulation 78---------------- 79 80- test/hello_world: always use shutdown type zero 81- make check: make test runs less noisy 82- boot-tests: force booting from primary (non-golden) side 83- mambo: Enable multicore configurations 84- mambo: Flatten device tree at the end 85- mambo: Increase memory to 4GB and change memory map 86- Timebase quirk for slow simulators like AWAN and SIMICS 87- chip: Add simics specific quirks 88- mambo: Flash driver using bogus disk 89- platform/mambo: Add a heartbeat time, making console more responsive 90- mambo: Fix bt command and add little endian support 91 92FSP platforms 93------------- 94 95- beginnings of support for SPIRA-S structure 96- Handle mbox response with bad status:0x24 during FSP termination 97- FSP: Validate fsp_msg response memory allocation 98- FSP/ELOG: Fix OPAL generated elog event notification 99- FSP/ELOG: Disable event notification during kexec 100 Possible crash if error log timing around kexec is unfortunate 101- fsp/console: Ignore data on unresponsive consoles 102 103 Linux kernels from v4.1 onwards will try to request an irq for each hvc 104 console using OPAL_EVENT_CONSOLE_INPUT, however because the IRQF_SHARED 105 flag is not set any console after the first will fail. If there is data 106 on one of these failed consoles OPAL will set OPAL_EVENT_CONSOLE_INPUT 107 every time fsp_console_read is called, leading to RCU stalls in the 108 kernel. 109 110 As a workaround for unpatched kernels, cease setting 111 OPAL_EVENT_CONSOLE_INPUT for consoles that we have noticed are not being 112 read. 113 114HMI 115--- 116 117- hmi: Fix a bug where partial hmi event was reported to host. 118- hmi: Add handling for NPU checkstops 119- hmi: Only raise a catchall HMI if no other components have 120- hmi: Rework HMI event handling of FIR read failure 121 122Tools 123----- 124 125- external: Add a getsram command 126 The getsram command reads the OCC SRAM. This is useful for debug. 127- bug fixes in flash utilities (pflash/gard) 128- pflash: Allow building under yocto. 129- external/opal-prd: Ensure that struct host_interfaces matches the thunk 130- external/pflash: Handle incorrect cmd-line options better 131- libflash: fix bug on reading truncated flash file 132- pflash: add support for manipulating file rather than flash 133- gard: fix compile error on ARM 134- libflash: Add sanity checks to ffs init code. 135- external: Add dynamically linked pflash 136 137Mambo 138----- 139 140- Test device tree for kernel location 141 This can reduce the boot time since the kernel no longer needs to 142 relocate itself when loaded directly at 0. 143 144Generic 145------- 146 147- hw/lpc: Log LPC SYNC errors as OPAL_PLATFORM_ERR_EVT errors 148- Explicitly disable the attn instruction on all CPUs on boot. 149- hw/xscom: Reset XSCOM engine after finite number of retries when busy 150- hw/xscom: Reset XSCOM engine after querying sleeping core FIR 151- core/timer: Add support for platform specific heartbeat 152- Fix GCOV_COUNTERS ifdef logic for GCC 6.0 153- core: Fix backtrace for gcc 6 154 fixes a compiler warning on GCC 6 and above 155- cpu: Don't call time_wait with lock held 156 Also make the locking around re-init safer, properly block the 157 OS from restarting a thread that was caught for re-init. 158- flash: Increase the maximum number of flash devices 159 160Contributors 161------------ 162 163Extending the analysis done for the last few releases, we can see our trends 164in code review across versions: 165 166======== ====== ======= ======= ====== ======== 167Release csets Ack Reviews Tested Reported 168======== ====== ======= ======= ====== ======== 1695.0 329 15 20 1 0 1705.1 372 13 38 1 4 1715.2-rc1 334 20 34 6 11 1725.3-rc1 302 36 53 4 5 173======== ====== ======= ======= ====== ======== 174 175An increase in reviews this cycle is great! 176 177Detailed statistics for 5.3.0-rc1 are below: 178 179Processed 302 csets from 31 developers 180A total of 20887 lines added, 4540 removed (delta 16347) 181 182Developers with the most changesets 183 184=========================== ============ 185=========================== ============ 186Stewart Smith 82 (27.2%) 187Gavin Shan 36 (11.9%) 188Benjamin Herrenschmidt 28 (9.3%) 189Michael Neuling 25 (8.3%) 190Vasant Hegde 24 (7.9%) 191Russell Currey 14 (4.6%) 192Brad Bishop 12 (4.0%) 193Vipin K Parashar 10 (3.3%) 194Cédric Le Goater 9 (3.0%) 195Shreyas B. Prabhu 8 (2.6%) 196Jeremy Kerr 7 (2.3%) 197Philippe Bergheaud 6 (2.0%) 198Cyril Bur 5 (1.7%) 199Mukesh Ojha 4 (1.3%) 200Alistair Popple 4 (1.3%) 201Ian Munsie 4 (1.3%) 202Oliver O'Halloran 3 (1.0%) 203Chris Smart 3 (1.0%) 204Sam Mendoza-Jonas 2 (0.7%) 205Joel Stanley 2 (0.7%) 206Dinar Valeev 2 (0.7%) 207Shilpasri G Bhat 2 (0.7%) 208Patrick Williams 2 (0.7%) 209Deb McLemore 1 (0.3%) 210Balbir Singh 1 (0.3%) 211Andrew Donnellan 1 (0.3%) 212Suraj Jitindar Singh 1 (0.3%) 213Frederic Bonnard 1 (0.3%) 214Kamalesh Babulal 1 (0.3%) 215Mamatha 1 (0.3%) 216Mahesh Salgaonkar 1 (0.3%) 217=========================== ============ 218 219Developers with the most changed lines 220 221========================= ============ 222========================= ============ 223Benjamin Herrenschmidt 7491 (34.4%) 224Gavin Shan 4821 (22.1%) 225Vasant Hegde 4740 (21.7%) 226Stewart Smith 1294 (5.9%) 227Michael Neuling 620 (2.8%) 228Cédric Le Goater 470 (2.2%) 229Jeremy Kerr 338 (1.6%) 230Shreyas B. Prabhu 330 (1.5%) 231Vipin K Parashar 305 (1.4%) 232Russell Currey 295 (1.4%) 233Alistair Popple 229 (1.1%) 234Philippe Bergheaud 170 (0.8%) 235Ian Munsie 133 (0.6%) 236Dinar Valeev 126 (0.6%) 237Brad Bishop 80 (0.4%) 238Oliver O'Halloran 80 (0.4%) 239Cyril Bur 62 (0.3%) 240Frederic Bonnard 61 (0.3%) 241Sam Mendoza-Jonas 32 (0.1%) 242Chris Smart 27 (0.1%) 243Shilpasri G Bhat 20 (0.1%) 244Patrick Williams 18 (0.1%) 245Suraj Jitindar Singh 17 (0.1%) 246Mamatha 15 (0.1%) 247Mukesh Ojha 8 (0.0%) 248Mahesh Salgaonkar 8 (0.0%) 249Joel Stanley 4 (0.0%) 250Balbir Singh 4 (0.0%) 251Kamalesh Babulal 2 (0.0%) 252Deb McLemore 1 (0.0%) 253Andrew Donnellan 1 (0.0%) 254========================= ============ 255 256Developers with the most lines removed 257 258========================= ============ 259========================= ============ 260Dinar Valeev 68 (1.5%) 261Patrick Williams 10 (0.2%) 262Mukesh Ojha 4 (0.1%) 263Kamalesh Babulal 1 (0.0%) 264========================= ============ 265 266Developers with the most signoffs (total 249) 267 268========================= ============ 269========================= ============ 270Stewart Smith 236 (94.8%) 271Vaidyanathan Srinivasan 6 (2.4%) 272Benjamin Herrenschmidt 3 (1.2%) 273Michael Neuling 2 (0.8%) 274Oliver O'Halloran 1 (0.4%) 275Vipin K Parashar 1 (0.4%) 276========================= ============ 277 278Developers with the most reviews (total 53) 279 280========================= ============ 281========================= ============ 282Andrew Donnellan 11 (20.8%) 283Russell Currey 9 (17.0%) 284Joel Stanley 7 (13.2%) 285Alistair Popple 7 (13.2%) 286Mukesh Ojha 5 (9.4%) 287Cyril Bur 3 (5.7%) 288Mahesh Salgaonkar 2 (3.8%) 289Gavin Shan 2 (3.8%) 290Vasant Hegde 2 (3.8%) 291Stewart Smith 1 (1.9%) 292Vaidyanathan Srinivasan 1 (1.9%) 293Vipin K Parashar 1 (1.9%) 294Frederic Barrat 1 (1.9%) 295Cédric Le Goater 1 (1.9%) 296========================= ============ 297 298Developers with the most test credits (total 4) 299 300========================= ============ 301========================= ============ 302Andrew Donnellan 2 (50.0%) 303Russell Currey 1 (25.0%) 304Vaibhav Jain 1 (25.0%) 305========================= ============ 306 307Developers who gave the most tested-by credits (total 4) 308 309========================= ============ 310========================= ============ 311Michael Neuling 3 (75.0%) 312Gavin Shan 1 (25.0%) 313========================= ============ 314 315Developers with the most report credits (total 5) 316 317========================= ============ 318========================= ============ 319Mukesh Ojha 2 (40.0%) 320Russell Currey 1 (20.0%) 321Pridhiviraj Paidipeddi 1 (20.0%) 322Balbir Singh 1 (20.0%) 323========================= ============ 324 325Developers who gave the most report credits (total 5) 326 327========================= ============ 328========================= ============ 329Gavin Shan 2 (40.0%) 330Stewart Smith 2 (40.0%) 331Vasant Hegde 1 (20.0%) 332========================= ============ 333