summaryrefslogtreecommitdiffstats
path: root/src/usr/hwas/common
Commit message (Collapse)AuthorAgeFilesLines
* Enhancements to default console outputDan Crowell2020-01-071-2/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Added the following pieces of extra information to the default console output: -- Attribute Overrides -- 3.49165|**Found 24 attribute overrides in Tank FAPI(1) 3.49169|- PEC:n0:p0:c0 3.49171| ATTR 0EB604DF [1] = 00 3.49174| ATTR 085F1C7E [1] = 00 3.49177| ATTR 048A8902 [1] = 01 3.49180| ATTR 04D8DF8D [2] = 0000 3.49182|- PEC:n0:p0:c1 3.49185| ATTR 0EB604DF [1] = 00 3.49188| ATTR 085F1C7E [1] = 00 3.49191| ATTR 048A8902 [1] = 03 -- New part detection -- 4.97449|Detected new part : 00030002 (Physical:/Sys0/Node0/DIMM2) -- Applying GARD records -- 8.78767|HWAS|Applying GARD record for HUID=0x000E0000 (Physical:/Sys0/Node0/Pr oc0/XBUS0) due to 0x90000135 8.78912|HWAS|Deconfig HUID 0x000E0000, Physical:/Sys0/Node0/Proc0/XBUS0 Change-Id: Ic2ae04515453dbb49e359abff952c87c83e23b72 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/88388 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Reviewed-by: Nicholas E Bofferding <bofferdn@us.ibm.com> Reviewed-by: Roland Veloz <rveloz@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: William G Hoffa <wghoffa@us.ibm.com>
* Automatically include config.hDan Crowell2019-12-062-2/+0
| | | | | | | | | | | | | | | | | | Rather than having to remember to include config.h anywhere we reference a CONFIG variable (and usually forgetting), this adds it to the default compiler flags so that it gets included in every source file we build. Change-Id: I53622ab4d46c55d942e98cae6ec03049fd5b3d08 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/87475 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Zachary Clark <zach@ibm.com> Reviewed-by: Roland Veloz <rveloz@us.ibm.com> Reviewed-by: Christian R Geddes <crgeddes@us.ibm.com> Reviewed-by: Nicholas E Bofferding <bofferdn@us.ibm.com>
* Allow dynmic i2c device addresses and set up PMIC targets to do thisChristian Geddes2019-11-011-2/+67
| | | | | | | | | | | | | | | | | | | | | | Depending on which vendor made a given OCMB the i2c device address of the PMIC targets on the OCMB will be different. To account for this we have added a new DYNAMIC_DEVICE_ADDRESS attribute. This attribute is filled out on the PMIC target by looking at the SPD on parent OCMB chip. This means that we must do presence detection on the OCMB prior to the the PMIC targets. While doing i2c operations if a given target has the DYNAMIC_DEVICE_ADDRESS we will use that over the devAddr in the any complex i2c attribute for that target. Change-Id: I22a185a65c064a1514751dd5828547c57af98df1 RTC: 209714 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/85394 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M Crowell <dcrowell@us.ibm.com> Reviewed-by: Roland Veloz <rveloz@us.ibm.com> Reviewed-by: William G Hoffa <wghoffa@us.ibm.com>
* Fix NVDIMM update error log commentsCorey Swenson2019-08-201-3/+3
| | | | | | | | | | | | | | First line of error log comment must include /*@ or error log will not be documented. Also fixed in other HB files. Change-Id: Ifa5eba6d6abd7f8565a4cc8d62a25a2b833725f2 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/82497 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Nicholas E Bofferding <bofferdn@us.ibm.com> Reviewed-by: Matt Derksen <mderkse1@us.ibm.com> Reviewed-by: Daniel M Crowell <dcrowell@us.ibm.com>
* Document Axone NPU configurationDan Crowell2019-05-291-6/+16
| | | | | | | | | | | | | | | | | | | - Updated simics_AXONE.system.xml with the valid target configuration that we should be using for NPUs in Axone. - Updated target xml files as well - Corrected 1 PG rule that no longer applies - Also modified the OBUS_BRICK layout in simics_AXONE as well Change-Id: I05c68be027cd4da39afabee04fefbb266b87c5fb RTC: 208518 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/76510 Reviewed-by: Christian R. Geddes <crgeddes@us.ibm.com> Reviewed-by: Glenn Miles <milesg@ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Force Axone simics to read all VPD from HW with config flagsChristian Geddes2019-05-142-5/+13
| | | | | | | | | | | | | | | | | | This commit will set the config flags to always read from HW rather than the old VPD cache in PNOR. Until this point in Axone we were still using an old copy of MVPD that we write into PNOR during the startup simics scripts. From this commit onward we will use the actual VPD simics provides. To handle this, some updates we needed to the PG rules for Axone. Change-Id: Ie06cefe1aec37edfc4c379ee1173bc51fc6bbe1f Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/76519 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Matt Derksen <mderkse1@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Add Axone to RISK_LEVEL logicDan Crowell2019-05-081-0/+33
| | | | | | | | | | | | | | | | | | Axone reuses the RISK_LEVEL settings for Nimbus DD2.3 so changes are made to reflect that. This is primarily a documentation exercise but there is logic now to normalize the RISK_LEVEl up to 4,5 versus 0,1 just to stay sane and reduce the test matrix. Change-Id: I5410d1bf7b12fc7f771e2c9826fcd086b2520091 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/76757 Reviewed-by: Matt Derksen <mderkse1@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Matthew Raybuck <matthew.raybuck@ibm.com> Reviewed-by: Jayashankar Padath <jayashankar.padath@in.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Gemini vs Explorer Presence Detection via SPDMatthew Raybuck2019-05-011-2/+1
| | | | | | | | | | | | | | | | | | | | | | | Since the OCMB chip is held in reset until after presence detection the IDEC register cannot be read to differentiate between Gemini and Explorer chip types. To work around this issue, during the early part of IPL when presence detection is occurring the OCMB IDEC function will instead read the SPD and populate the necessary attributes with what is found there. That will be used to determine the difference between Gemini and Explorer until later when the OCMB IDEC register can be read from. At that point the IDEC read will be executed again and the data read from the OCMB IDEC register will be used to cross-check the data read from the SPD. Any discrepancies will be handled with predictive error logs. Change-Id: Ica664b06ff3488f48253d3ef02eff2d49c5d240d RTC: 208696 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/76108 Reviewed-by: Christian R. Geddes <crgeddes@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Collect IDEC for Explorer chipMatthew Raybuck2019-04-301-10/+15
| | | | | | | | | | | | | | | | | | | The OCMB Explorer Chip doesn't read for IDEC but instead assumes hardcoded values. Since the Explorer chip is held in reset until iStep 10.4, this commit will prevent IDEC reads during discoverTargets and instead perform the read when exp_check_for_ready() is successful. Change-Id: I4ef5a01badb195acca0c2187ef76ea55f58eafe4 RTC:201996 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/75881 Reviewed-by: Matt Derksen <mderkse1@us.ibm.com> Reviewed-by: Christian R. Geddes <crgeddes@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com> Tested-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Add new PMIC target for AxoneMatt Derksen2019-04-181-1/+3
| | | | | | | | | | | | | | | | PMIC is a voltage regulator for the DDIMM. It supplies power to the OCMB and DIMM targets. Change-Id: I10c1b03169f53b070f521ec9cd60cdbd15c4a268 RTC:206184 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/75136 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Glenn Miles <milesg@ibm.com> Reviewed-by: Christian R. Geddes <crgeddes@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Flush Errlmanager Queue Before Fetching Gard RecordsIlya Smirnov2019-04-101-0/+6
| | | | | | | | | | | | | | | | | | | | | | When an error occurs or is injected into an EX target, the gard record for the target does not make it into the GUARD partition in time for the gard processing in istep 6.9 because the associated error log is sent asynchronously. This change flushes the errlmanager message queue right before the processing of the gard records to insure all gard records make it into GUARD partition. Change-Id: I06625672df7e926d6d080673241ad2ee4d87a7c9 CQ: SW461565 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/75684 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Michael Baiocchi <mbaiocch@us.ibm.com> Reviewed-by: Roland Veloz <rveloz@us.ibm.com> Reviewed-by: William G. Hoffa <wghoffa@us.ibm.com>
* Improve performance of general PG algorithmMatthew Raybuck2019-03-182-248/+380
| | | | | | | | | | | | | | | | | | | The previous PG algorithm had several places where targets would be rechecked or modified unnecessarily. This commit restructures the PG algorithm and pulls all related code out of discoverTargets and adds it to checkPartialGoodForDescendants which will be called by discoverTargets. Change-Id: I045f17f7ac22c673633cd3d951997b2371be1b8d RTC:206066 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/72732 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Nicholas E. Bofferding <bofferdn@us.ibm.com> Reviewed-by: Ilya Smirnov <ismirno@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Support UCD discoveryNick Bofferding2019-03-151-3/+6
| | | | | | | | | | | | | | | | | | - Added new attribute to indicate if target should be assumed present - Added that attribute to SP/BMC/UCD targets - Check for assumed present UCDs in discover targets - Iterate through every UCD for data flash update RTC: 201991 Change-Id: Ia535a58ea0355582621a23d3c1b50b2417ad362f Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/73047 Reviewed-by: Matthew Raybuck <matthew.raybuck@ibm.com> Reviewed-by: Michael Baiocchi <mbaiocch@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Improve performance of Partial Good logicMatthew Raybuck2019-03-011-8/+19
| | | | | | | | | | | | | | | | | | | | The function PARTIAL_GOOD::findRulesForTarget() would call masterProcChipTargetHandle() to figure out if a rule applied to a specific target based on the ATTR_MODEL of a system. Since finding the master proc that way is an expensive operation, findRulesForTarget() will now use an instance variable initialized once instead of querying for it repeatedly thereby increasing performance substantially. Change-Id: I6e9c01daf356fe21d76d920f5d3210717f5094c1 CQ:SW450113 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/72620 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Reviewed-by: Nicholas E. Bofferding <bofferdn@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Ilya Smirnov <ismirno@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* OpenPOWER support for native and compatibility mode for DD2.3Matt Derksen2019-02-282-11/+437
| | | | | | | | | | | | | | | | | | | | | | | Witherspoon (OpenPOWER) will run in native mode, which means all the processors have to be at the same level (all DD2.2 or all DD2.3). Other systems (ZZ) we support running in a mixed configuration where DD2.3 processors will be run in 2.2 compatibility mode. We also need to support arbitrary OP systems that might want to run in compatibility mode even without mixed parts. See src/usr/hwas/common/README.md for the rules and resulting mode. Change-Id: I80fb98e2687b945ba506f2d75b1533884443e10b RTC:201485 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/72214 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Reviewed-by: Christian R. Geddes <crgeddes@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Handle partial-bad MCS logic correctlyDan Crowell2019-02-191-0/+4
| | | | | | | | | | | | | | | | | One of the rewritten PG rules was incorrect for modules that have a MCS marked bad, e.g. Sforza. Change-Id: I30672b51f81ba74b51ece6e878e462106c090350 CQ: SW457231 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/72010 Reviewed-by: Matt Derksen <mderkse1@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Reviewed-by: Ilya Smirnov <ismirno@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Matthew Raybuck <matthew.raybuck@ibm.com> Reviewed-by: William G. Hoffa <wghoffa@us.ibm.com>
* Manually garded targets incorrectly being resource recoveredMatt Derksen2019-02-151-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Garded targets were not always setting GARD_APPLIED bit in ATTR_HWAS_STATE_CHANGED which was causing ATTR_BLOCK_SPEC_DECONFIG to always be set after a resource recovery. To recreate failure: 1) gard a critical target (gard --md 0x50000) 2) IPL 3) verify resource recoverd (gard --gc fabric) 4) verify ATTR_BLOCK_SPEC_DECONFIG set to 1 (hwsvCmd -y) 5) power down 6) clear previous gard (gard --clr all) 7) gard all XBUSes (gard --md 0xE0000 .. gard --md 0xE0005) 8) IPL 9) verify XBUSes were incorrectly resource recovered (gard --gc fabric) This SPEC_DECONFIG code was added in fips920. We don't believe the fix needs to go back because it requires a double failure and manual garding. The first failure must be a gard of a critical resource, which forces a reconfig. In the field, the failed parts would be replaced and rely on a different bit (RESRC_RECOV) and code path. Also, if this isn't fixed the targets will just be garded again later. This also only applies to Predictive garded parts. Change-Id: I183b74ef200a5034e5dba20c0f64ad8f75763167 CQ:SW454599 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/71929 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Roland Veloz <rveloz@us.ibm.com> Reviewed-by: Christian R. Geddes <crgeddes@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Better logging for Partial Good Issues (2/2)Matt Raybuck2019-02-152-40/+174
| | | | | | | | | | | | | | | | | | | PG issues were only ever displayed in traces. This commit creates a new error log user detail section that will display the PG vector and highlight which entries have issues. In addition, targets that are determined to be non-functional will be added to the log and a hardware callout will be added if a chip is found non-functional. Change-Id: I9b401bcd5686951303946789034671c1a395fb91 RTC:180365 CMVC-Prereq:1075293 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/70004 Reviewed-by: Nicholas E. Bofferding <bofferdn@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Reviewed-by: Michael Baiocchi <mbaiocch@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Fix deconfigure parent rollup policyMatt Derksen2019-01-311-24/+74
| | | | | | | | | | | | | | | | | | | Change to specifically designate if a target is allowed to be deconfigured by child rollup and if it should rollup to its parent. Need to cover the case where there are two different types of children and only one is allowed to rollup to the parent. Also prevent the deconfigure rollup from happening to the parent. Change-Id: I514876a46e9c8180e1fc99a969e0ca4247fbf2d9 CQ:SW454562 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/70759 Reviewed-by: Christian R. Geddes <crgeddes@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* remove unnecessary checks in checkMinimumHardwareMatt Raybuck2019-01-251-267/+0
| | | | | | | | | | | | | | | | | | There were a series of checks that were performed after we had determined the DIMMs were functional that would be implicitly true since the DIMMs are functional. Those were removed. Change-Id: I14feae0129ac96c2f701000b0f265bff36250371 RTC:196804 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/70560 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Ilya Smirnov <ismirno@us.ibm.com> Reviewed-by: Nicholas E. Bofferding <bofferdn@us.ibm.com> Reviewed-by: DHRUVARAJ SUBHASH CHANDRAN <dhruvaraj@in.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* deconfigByAssoc updates for Axone and OMCBMatt Raybuck2019-01-231-0/+33
| | | | | | | | | | | | | | | | | Updates the deconfigByAssoc functions to use the special OMIC_PARENT and OMI_CHILD relations when dealing with target hierarchy since OMIs have two parents. Change-Id: Ibc512777982e24de21ab7878d816a72fab1b19f7 RTC:196804 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/70510 Reviewed-by: Nicholas E. Bofferding <bofferdn@us.ibm.com> Reviewed-by: Ilya Smirnov <ismirno@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* presentByAssoc updates for Axone and OCMBMatt Raybuck2019-01-161-19/+217
| | | | | | | | | | | | | | | Added support for MC->OMIC->OMI path in presentByAssoc. Change-Id: I076194b44ac069b18afa5ff96af36c5fa859336e RTC:196804 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/70307 Reviewed-by: Nicholas E. Bofferding <bofferdn@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Ilya Smirnov <ismirno@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Deconfig updates for Axone and OCMBMatt Raybuck2019-01-161-30/+488
| | | | | | | | | | | | | | | | | | | Added support for MC->MI->MCC->OMI->OCMB path in presentByAssoc and made changes to checkMinimumHardware that will prevent all dimms from being incorrectly deconfigured so that AXONE simics is able to reach iStep 6.11. Change-Id: I0be474733a78097505c85826a1f34a854144f931 RTC:196804 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/70058 Reviewed-by: Nicholas E. Bofferding <bofferdn@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Ilya Smirnov <ismirno@us.ibm.com> Reviewed-by: Michael Baiocchi <mbaiocch@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* No parent rollup for OBUS childrenMatt Derksen2019-01-111-1/+34
| | | | | | | | | | | | | | | HWSV code is deconfiguring either SMPGROUP or OBUS_BRICKS under the OBUS. Need to prevent the deconfigures from also deconfiguring the parent OBUS. Change-Id: Ibca2766009b63582fcfa537139a078bef3a58f47 CQ:SW454103 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/70322 Reviewed-by: Christian R. Geddes <crgeddes@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Add presence detection for i2c mux targetsChristian Geddes2019-01-111-4/+6
| | | | | | | | | | | | | | | | | | | | | | | This commit removes the ddimm.C file that had the deviceFramework routing for OCMB presence detection and replaces it with a new file in src/usr/i2c/i2cTargetPres.C that is more generic for any target that has the FAPI_I2C_CONTROLLER_INFO attribute. The i2c_mux target also now uses this same code for its presence detection. As a result of this change the src/usr/i2c/mux_i2c.* files have also been removed. When getting rid of the ddimm.C file I had to put the IDEC device routing somewhere else so I moved it to the hwasPlat code where the other IDEC device routes are registered. RTC: 196805 Change-Id: I27e5e3e8d0fe107c3d44a450e20efa6f50fa0c5f Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/69944 Reviewed-by: Matt Derksen <mderkse1@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Roland Veloz <rveloz@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Deconfigure parent if no more functioning children of child's typeMatt Derksen2019-01-081-3/+68
| | | | | | | | | | | | | | | | | | Defect discovered that a DIMM deconfig did not rollup to the membuf because the membuf also has an l4_centaur child. The fix is to check that the parent has no more children like the one being deconfigured. Also make sure that deconfiguring l4_centaur does not deconfigure its parent membuf. Change-Id: Ic499092d0da926dcd3e0c08f5205e491e04f9680 CQ:SW451955 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/70093 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Christian R. Geddes <crgeddes@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Use new Bus callout InterfaceAlpana Kumari2018-12-071-10/+14
| | | | | | | | | | | | | | | | HWSV has supported new interface of BUS Callout with flag parameter, HB has to use this New interface Change-Id: I2ceb6a086938fdb943a93b65f98158dfaec83a1d CQ: SW448379 CMVC-Prereq: 1073384 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/69335 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Sachin Gupta <sgupta2m@in.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Driver changes to support i2c muxRoland Veloz2018-12-061-5/+69
| | | | | | | | | | | | | | | | | | | | | | | | | - Updated data structures gpioAddr_t, eeprom_addr_t, misc_args_t, nvdimm_addr_t and tpm_info_t with I2C MUX data members. Also added constructors to these structures to default there data members with the correct default info. - Updated macros DEVICE_I2C_PARMS, DEVICE_I2C_ADDRESS and DEVICE_I2C_ADDRESS_OFFSET to take the I2C MUX bus selector parameter and the I2C MUX entity path. - Added method i2cAccessMux to file i2c.H/.C that will setup the call for the I2C MUX. Method i2cCommonOP calls i2cAccessMux which then calls i2cCommonOp with appropriate parameters for the I2C MUX: i2cCommonOP -> i2cAccessMux -> i2cCommonOP. - Updated i2ctest.H with new I2C MUX params to get it to pass. RTC:191352 Change-Id: I6a70860eb2286bbd23d6157d72351b8adfa21aac Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/66651 Reviewed-by: Ilya Smirnov <ismirno@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Reviewed-by: Nicholas E. Bofferding <bofferdn@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Remove hardcoding of partial good logic (2/2)Matt Raybuck2018-12-032-616/+651
| | | | | | | | | | | | | | | | | The original pg logic was a hard-coded criss-crossed jumble of logic that was difficult to understand. It has now been generalized to an algorithm that applies the correct logic based on a set of rules kept in a table. This should make it easier to understand and maintain going forward. Change-Id: I51435cc2ca6bbfa9ebc8a3d52d4ebf23b5bd2730 RTC:193270 CMVC-Prereq: 1072559 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/68232 Reviewed-by: Ilya Smirnov <ismirno@us.ibm.com> Reviewed-by: Nicholas E. Bofferding <bofferdn@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Remove hardcoding of partial good logic (1/2)Matt Raybuck2018-11-141-0/+24
| | | | | | | | | | | | | | | | | | | This is the first commit of two. This commit will allow hwsv project to be able to add necessary changes to fips930/src/hwsv/server/makefile and still compile and pass CI before the rest of the code for this story is added. Change-Id: I3547f00e054fc593e31bbe2f2281cff95d385a17 RTC:193270 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/68702 Reviewed-by: Roland Veloz <rveloz@us.ibm.com> Reviewed-by: Ilya Smirnov <ismirno@us.ibm.com> Reviewed-by: Nicholas E. Bofferding <bofferdn@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Support openpower-specific I2C device calloutsJaymes Wilks2018-11-121-1/+1
| | | | | | | | | | | | | | | | | | | | | This change supports the openpower path for adding I2C device callouts to error logs. The process works as follows: - Create an I2C device lookup table on first use of I2C callout - Use that table to map I2C info to the actual device to callout - Callout any other I2C devices on the bus with lesser priority - If no I2C match found, callout the I2C master instead - If an I2C device was found, callout the I2C master as low Change-Id: Ib7b248ae60e7e834d6165bbdf4bd9b776ea2421b RTC:94872 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/64833 Reviewed-by: Michael Baiocchi <mbaiocch@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Reviewed-by: Nicholas E. Bofferding <bofferdn@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Deconfig parent rollup simplifiedMatt Derksen2018-11-092-326/+138
| | | | | | | | | | | | | | | Create a common deconfiguration rollup to parent. A new attribute was created to indicate when to stop rolling up the parent branch Change-Id: I80954081527f29bc466622d30644115a96be385f RTC:193268 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/66444 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Christian R. Geddes <crgeddes@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* fix Skip resource recovery for node and power gard.Dhruvaraj Subhashchandran2018-10-221-5/+5
| | | | | | | | | | | | | Change-Id: Ia962954dd85fe0ead36245016be3ac8380216429 Signed-off-by: Dhruvaraj Subhashchandran <dhruvaraj@in.ibm.com> CQ: SW448863 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/67535 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Reviewed-by: Matt Derksen <mderkse1@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Skip resource recovery for node and power gard.Dhruvaraj Subhashchandran2018-10-091-9/+13
| | | | | | | | | | | | | | | | | | | | After the node is deconfigured with non-fatal or power error running a resource recovery will always recover the node since all vital parts are deconfigured in the node. so skipping the resource recovery for node. Power gard need be considered as fatal so skipping resource recovery for power gard too. Change-Id: I246733b59d908a56401a0c602897a5d12522406b Signed-off-by: Dhruvaraj Subhashchandran <dhruvaraj@in.ibm.com> CQ: SW444154 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/66981 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Matt Derksen <mderkse1@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Update PG detection for new Axone memory targetsChristian Geddes2018-09-051-11/+34
| | | | | | | | | | | | | | | Add support for presence detection of new OMI, OMIC, and MCC targets. Detection is done by looking at the MC01, and MC23 PG data. Change-Id: I2706bb28f64df653546ce22451e690507be09d8e RTC: 197541 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/64966 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Matt Derksen <mderkse1@us.ibm.com> Reviewed-by: William G. Hoffa <wghoffa@us.ibm.com>
* Deconfig EC/EX/EQ at runtimeRoland Veloz2018-08-302-10/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Applied the normal parent/child rules for EC/EX/EQ logic to be done in the runtime deconfig logic. After applying the deconfig logic, then reset the PM for the parent proc of the EX/EX/EQ. To facilitate the code changes, modified the return value of the TARGETING::getParent(...) call. It no longer returns a const * Target, considering 5 times out of 7 of it's usage, users were just staic_cast(ing) away the const. Clear out the HB reset attribute in the mpipl path. Modify PRD logic to use runtime deconfig call in malfunction alert path. CMVC-Prereq: 1066157 Change-Id: If81f2f15c6e84bf2adb6fd66c1321b8f1ddf1a86 CQ: SW443157 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/64873 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Roland Veloz <rveloz@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Account for mixed procs in compatibility modeDean Sanner2018-08-211-1/+54
| | | | | | | | | | | | | | | | Current code enforces the same EC level for all processors in the system. However, this breaks compatibility mode for P9C 1.2/1.3 and P9N 2.2/2.3. Added code to allow these to pass the EC check when mixing parts Change-Id: Ie013d94dfdee3627911f4afbec5f65d5ce984227 CQ:SW442691 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/64823 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Add flag to HWAS Callout for SMP repair indicatorDan Crowell2018-08-191-0/+3
| | | | | | | | | | | | | Used one of the reserved bytes of the callout structure to hold a flag that will indicate if the SMP bus being called out is completely down, and thus eligible for concurrent repair. Change-Id: I299e3393529e811aa006a087de51165b79038201 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/63946 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Implement generic i2c device callouts for FSPJaymes Wilks2018-08-131-0/+29
| | | | | | | | | | | | | | | | | | | Creates a new type of callout for i2c devices which populates the error log user details section with the address, port, engine, and i2c master of the device being called out. The change only affects the FSP path. Change-Id: I92acc092f105ee21346594f5e3733957e15ad768 RTC:94872 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/63223 Reviewed-by: Michael Baiocchi <mbaiocch@us.ibm.com> Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Nicholas E. Bofferding <bofferdn@us.ibm.com> Reviewed-by: MURULIDHAR NATARAJU <murulidhar@in.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Add recovered gard log only for garded targets.Dhruvaraj Subhashchandran2018-08-031-7/+17
| | | | | | | | | | | | | | | Adding recovered gard log for recovery of associtive deconfig targets was flooding the gard trace. Change-Id: I70341864103c447fd01f4c68d86a4d637a7d9c54 CQ: SW435589 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/63270 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Sachin Gupta <sgupta2m@in.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Add deconfigure SMPGROUP peer targets rule for SMPGROUParavnair-in2018-07-121-0/+13
| | | | | | | | | | | | | Make sure the peer target of an SMPGROUP is deconfigured when it is deconfigured. Change-Id: I74e8ba63124a1607b7cc9f639bf38f6fd122bb8a Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/62224 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* EC level match for only functional master Proc per NodeAlpana Kumari2018-06-281-1/+2
| | | | | | | | | | | Change-Id: I0c2ecba2450faf8437bd286fa0d1d4f2d7e6b171 CQ: SW433870 Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/61492 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Ensure hwas state reflects resource recovery actionsAndrew Geissler2018-06-191-2/+81
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Once resource recovery is run to recover hardware resources to allow the system to boot, it will not be run again until the configuration of the hardware changes. An issue with that design is that the HWAS state, deconfiguredByEid, will not be properly updated in situations where resource recovery has already run. This state is used by the gard command line tool to inform the user on the state of their garded hardware. This commit ensure the deconfiguredByEid is properly updated for targets that are guarded, but have been resource recovered to allow system boot. Change-Id: Ib15c1e8402e7c13b6497915c5138831e5e591bbe CQ: SW432846 Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/60486 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Christian R. Geddes <crgeddes@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: DHRUVARAJ SUBHASH CHANDRAN <dhruvaraj@in.ibm.com> Reviewed-by: ARAVIND T. NAIR <aravindnair@in.ibm.com> Reviewed-by: William G. Hoffa <wghoffa@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* SW419349: Handle override of deconfig by Error vs FCO reasons by associationAmit Tendolkar2018-06-181-27/+87
| | | | | | | | | | | | | | | | | | | | | | 1. Override the deconfig reason of a parent .. a) getting deconfig by association due to error on its child && b) has at least 1 other child already deconfig by FCO .. to deconfig by FCO 2. Override the deconfig reason of a child .. c) whose parent is getting deconfig by an error && d) that is itself already deconfig by FCO .. to deconfig by that error Change-Id: I9c73fe35aedb832f9a4b33bd299bb30fe589095c CQ: SW419349 Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/59934 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Reviewed-by: William G. Hoffa <wghoffa@us.ibm.com> Reviewed-by: Prachi Gupta <pragupta@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Fix unbalance FCO distribution between procsThi Tran2018-06-141-21/+70
| | | | | | | | | | | | | | | | | | | | | Current code doesn't take into account the case where an EX has no core under. In this scenario, the code skips the current proc and takes out the cores in the next proc. This leads to more cores taken out from the next proc if current proc has more EXs with no core. The fix is to look and take out the next available cores in current proc until no core available before move on to the next proc. Change-Id: I0e8650766c9f13c5616141b87e05967950d6fb56 CQ:SW424855 Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/58460 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Richard J. Knight <rjknight@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Deconfig by association rule for SMPGROUP targetsaravnair-in2018-06-081-0/+22
| | | | | | | | | | | | | | If the two SMPGROUP targets under an OBUS target is deconfigured, the parent OBUS target also needs to be deconfigured by association. Change-Id: Ifb0fcf46a8c65a2655eade04ae7035d228bca92b Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/60093 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Look for any parent on deconfigureAndrew Geissler2018-06-071-11/+35
| | | | | | | | | | | | | | | | | | | | | | | | | There are error scenarios where a unit can get deconfigured and its parent is not considered present. This can lead to errors in the deconfig by association logic because a unit should always have a present parent. This change has the deconfig logic just find the parent. If it's considered not present or not functional then the code can handle that, it just needs the parent target. This commit also fixes a bug found in testing in the DIMM deconfig by association path. Change-Id: I98c4185192a0944543cc545277504b50bd951c3e CQ: SW431931 Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/60053 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Prachi Gupta <pragupta@us.ibm.com> Reviewed-by: Richard J. Knight <rjknight@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* missing memory: istep 7 and 14 changesPrachi Gupta2018-06-061-3/+69
| | | | | | | | | | | | | | | | | | | | | There are two cases where hostboot's attention is required in istep7: - If HRMOR we booted with doesn't fall in the range of proc_mem_to_use's memory, then the SBE is old. HB will do an sbe update and request re-ipl - If HB deconfigured a bunch of dimms in istep7 and ran out of memory, then we will request a reconfig loop Then, in istep14, we added another sanity check to make sure we have memory as expected to prevent unexpected failure after exiting cache contained mode. Change-Id: I018f4ce862cc79b5d7bacbe01cc28d1d2b4fc788 CQ:SW430015 Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/59696 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>
* Add RESOURCE_RECOVERED event for all TargetsDan Crowell2018-06-051-1/+1
| | | | | | | | | | | | | | Lost an event during the last refactor, putting it back. Change-Id: Ica5815c7c06cd981216fecbd927e1c1aa527519a CQ: SW431492 Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/59844 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Prachi Gupta <pragupta@us.ibm.com> Reviewed-by: Matt Derksen <mderkse1@us.ibm.com> Reviewed-by: William G. Hoffa <wghoffa@us.ibm.com>
* Always deconfigure the parent of any deconfigured DIMMDan Crowell2018-06-051-83/+18
| | | | | | | | | | | | | | | | | Whenever a DIMM is deconfigured, we will also deconfigure the immediate parent target (e.g. MCA, MBA, etc) to ensure there is not an unbalanced load on the ports. Change-Id: I30b0b5a4597abfc2f7d429a69456756d468e4643 CQ: SW425876 Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/59507 Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com> Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com> Reviewed-by: Martin Gloff <mgloff@us.ibm.com> Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com> Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com> Reviewed-by: Richard J. Knight <rjknight@us.ibm.com> Reviewed-by: William G. Hoffa <wghoffa@us.ibm.com>
OpenPOWER on IntegriCloud