summaryrefslogtreecommitdiffstats
path: root/pk
Commit message (Collapse)AuthorAgeFilesLines
* Control DeadMan Loop SupportRaja Das2016-03-231-1/+1
| | | | | | | | | Change-Id: Icee15457cd7abb8779e1134fc88063b2437e3281 RTC:134270 Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/21038 Tested-by: Jenkins Server Reviewed-by: Sachin Gupta <sgupta2m@in.ibm.com> Reviewed-by: AMIT J. TENDOLKAR <amit.tendolkar@in.ibm.com>
* Added application level PK_PANIC offset for the various PPE enginesClaus Michael Olsen2015-10-281-0/+12
| | | | | | | | | | | | | to support easier debugging of application level halts and to prevent accidental reuse of same PK_PANIC codes as is currently used by the kernel, i.e. all the 0x00779xxx codes. Change-Id: I327d72994e5a6f6065681c81a50a50b9e298733d Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/21540 Tested-by: Jenkins Server Reviewed-by: Claus Michael Olsen <cmolsen@us.ibm.com> Reviewed-by: Douglas R. Gilbert <dgilbert@us.ibm.com> Reviewed-by: Gregory S. Still <stillgs@us.ibm.com>
* Added PK_TRACE_TIMER_OUTPUT flag to enable disabling of the 2 secClaus Michael Olsen2015-10-262-8/+13
| | | | | | | | | | | | | | | | | periodic PK trace timer dump by defining PK_TRACE_TIMER_OUTPUT to 0 in, say, global_app_cfg.h in the application dir. - Further, updated PK_{MILLI,MICRO,NANO}SECONDS() macros to use more appropriately sized constants because currently it's much too easy to cause overrun for even small numbers passed to these macros. Also, ditched division ops all together in MILLI and MICRO macros and now using shift operators instead of division in the NANO macro. This saves code space. Change-Id: I47758ac1f735bf4918413a6de28390c9e46d38dc Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/21029 Tested-by: Jenkins Server Reviewed-by: Douglas R. Gilbert <dgilbert@us.ibm.com> Reviewed-by: Gregory S. Still <stillgs@us.ibm.com>
* fix getscom_abs and _getscom codeWael El-Essawy2015-10-032-10/+34
| | | | | | | | | | | | | | | | | | | | | | | | Instead of reading the scom registers, and storing the results to a memory address specified by o_data, the two routines were loading data into a vector register, without storing it to the o_data pointer, basically losing it. There is no memory to memory PPC instruction, so the minimum number of instructions to execute the code is two: a vector load and a vector store. - Had to modify the ppe42_scom.h to make the _m_data parameter to the PPE_LVD and PPE_LVDX macros an output only. Otherwise, the compiler initializes the temp variable, adding two (useless) instructions. - make the PPE_LVDX macro have the same flavor as PPE_LVD by eliminating the pointer to m_data from both the _getscom function side and the macro implementation side. Change-Id: I67a9eef16ed1e0f95f02ff8ba0336558cc8b23e3 Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/20716 Tested-by: Jenkins Server Reviewed-by: Claus Michael Olsen <cmolsen@us.ibm.com> Reviewed-by: Gregory S. Still <stillgs@us.ibm.com>
* Remove unsupported sc instruction in PPE-KernelDoug Gilbert2015-09-092-22/+22
| | | | | | | | | | Change-Id: Icc11444905cd1333701980b49980139f96656bd7 CQ: HW326920 Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/20173 Tested-by: Jenkins Server Reviewed-by: YUE DU <daviddu@us.ibm.com> Reviewed-by: Claus Michael Olsen <cmolsen@us.ibm.com> Reviewed-by: Gregory S. Still <stillgs@us.ibm.com>
* Round 3 PPE FAPI2 Sync with current FAPI2 baseGreg Still2015-06-161-0/+2
| | | | | | | | | | | | | | | | | | | | | | | Change-Id: I16fe661deece510266e71bebea87ca546db2d252 Delete return_code.H in deference to base Change-Id: Ib8f11bc5eacbc64399de299d90f1698262967216 Remove plat_utils.C from fapi2ppefiles.mk Change-Id: Ie921e66b1d4b6972d5ad1d0ac2278c055d82ce19 Add include of pk_app_cfg.h to std_irq_config.h Change-Id: I488e43ea14a0ef1da927bcf010ff95c9349b08dd Image building updates to be more realistic Change-Id: I67084b96bb9e83b71fc732887955f115e7ed00db Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/18006 Reviewed-by: Derk Rembold <rembold@de.ibm.com> Tested-by: Derk Rembold <rembold@de.ibm.com>
* PPE getscom fix for procedure call versionGreg Still2015-06-102-19/+19
| | | | | | | | - Properly handle referenced return data - Needed the SBE FW - Future note: generated code has optimization opportunities Change-Id: I2481bd7b8cdebd89c629fd792f6c183a8e9f3c19
* Initial PPE FAPI2 platform implementation - targets+get/putscom; no AttributesGreg Still2015-05-213-6/+7
| | | | | | | | Change-Id: I805dd3286b1b9d33b585d903d38ba7555bf40c21 Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/16869 Reviewed-by: Glenn R. Miles <milesg@us.ibm.com> Reviewed-by: Derk Rembold <rembold@de.ibm.com> Tested-by: Derk Rembold <rembold@de.ibm.com>
* Add common ppe_compiler.mk and proc_hcd_common.HGreg Still2015-05-212-224/+109
| | | | | | | | | | | | | | | | | | - ppe_compiler.mk for pointing to the common PPE compiler across platforms - proc_hcd_common.H for things common across PPE HCD procedures Change-Id: Ic8883a655eacfcd860e0036c516555ab12984d08 New PPE getscom routines leveragin upgraded compiler - Pass pointers to data elements to use indexed lvdx/stvdx instructions Change-Id: I7b99393919bc533bac5743001ee2c50b778eb2b2 Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/17172 Reviewed-by: Glenn R. Miles <milesg@us.ibm.com> Reviewed-by: Brian Silver <bsilver@us.ibm.com> Reviewed-by: Derk Rembold <rembold@de.ibm.com> Tested-by: Derk Rembold <rembold@de.ibm.com>
* Update ppe42_msr.h with SEM bit position definesGreg Still2015-05-211-0/+7
| | | | | | | | | Change-Id: I242cfc3849d75546143f03559d7b5f77d45688ff Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/17173 Reviewed-by: Glenn R. Miles <milesg@us.ibm.com> Reviewed-by: Sachin Gupta <sgupta2m@in.ibm.com> Reviewed-by: Derk Rembold <rembold@de.ibm.com> Tested-by: Derk Rembold <rembold@de.ibm.com>
* ppetracepp tool: support coexisting with tracepp toolGlenn Miles2015-05-132-5/+5
| | | | | | | | Change-Id: Ia62df48af0e0722644cce8ce93c9b7473f747ae5 Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/17562 Reviewed-by: Richard J. Knight <rjknight@us.ibm.com> Reviewed-by: Thi N. Tran <thi@us.ibm.com> Tested-by: Thi N. Tran <thi@us.ibm.com>
* Support for kernel debug pointersGlenn Miles2015-04-245-8/+110
| | | | | | | | | Change-Id: I5d41ff0390ec0b2c453bd7072777e58c81a0eeb5 Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/17336 Reviewed-by: Richard J. Knight <rjknight@us.ibm.com> Reviewed-by: Glenn R. Miles <milesg@us.ibm.com> Reviewed-by: Thi N. Tran <thi@us.ibm.com> Tested-by: Thi N. Tran <thi@us.ibm.com>
* Minor changes to kernel APIGlenn Miles2015-04-2129-1577/+971
| | | | | | | | | | | | | | | | | | | | | -Removed support for fast irq handlers -Removed sleep_absolute interface -Removed pk_timer_create_nonpreemptible interface -Removed pk_interrupt_preemption_enable/disable interfaces -Removed references to critical/noncritical irqs -Added application context interfaces -Added bottom-half support -Moved timer handler from interrupt to bottom-half -Moved thread utility functions into a separate file Change-Id: Ie39fe0363ef52b195a808a8390cc12c2c7478674 Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/16898 Reviewed-by: Glenn R. Miles <milesg@us.ibm.com> Reviewed-by: Thi N. Tran <thi@us.ibm.com> Reviewed-by: Richard J. Knight <rjknight@us.ibm.com> Reviewed-by: Gregory S. Still <stillgs@us.ibm.com> Reviewed-by: Derk Rembold <rembold@de.ibm.com> Tested-by: Derk Rembold <rembold@de.ibm.com>
* Add efficient time scaling supportGlenn Miles2015-03-237-99/+77
| | | | | | | | | Also removed automatic rescheduling of periodic timers Change-Id: I6427a8d8ed5ca4b75389c3610a16dba10783a8ae Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/16537 Reviewed-by: Glenn R. Miles <milesg@us.ibm.com> Tested-by: Glenn R. Miles <milesg@us.ibm.com>
* More changes for external timebase supportGlenn Miles2015-03-206-23/+15
| | | | | | | Change-Id: I6e643da771e8106dc56f831147cb078350ba9dde Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/16503 Reviewed-by: Glenn R. Miles <milesg@us.ibm.com> Tested-by: Glenn R. Miles <milesg@us.ibm.com>
* Support for using an external timebase registerGlenn Miles2015-03-1813-135/+243
| | | | | | | Change-Id: I2c40fdf88f51315b19e8d639f874fb79717419ce Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/16369 Reviewed-by: Glenn R. Miles <milesg@us.ibm.com> Tested-by: Glenn R. Miles <milesg@us.ibm.com>
* Add external interrupt support for standard PPE'sGlenn Miles2015-03-0918-148/+772
| | | | | | | Change-Id: I5ac49d4a9aace2dfbe3cdebf0dd7942c1e22da4e Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/16240 Reviewed-by: Glenn R. Miles <milesg@us.ibm.com> Tested-by: Glenn R. Miles <milesg@us.ibm.com>
* Added SCOM proxy server support for GPE'sGlenn Miles2015-03-043-1/+99
| | | | | | | Change-Id: I1695bffc423955e72ff1571fdd10ef04c853f625 Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/16056 Tested-by: Glenn R. Miles <milesg@us.ibm.com> Reviewed-by: Glenn R. Miles <milesg@us.ibm.com>
* Seed the files in the pk directoryGlenn Miles2015-02-1963-0/+10396
| | | | | | | | Change-Id: I03398098e6625f0e06e4a96769b03002a1c71d35 Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/15813 Reviewed-by: Glenn R. Miles <milesg@us.ibm.com> Reviewed-by: Derk Rembold <rembold@de.ibm.com> Tested-by: Derk Rembold <rembold@de.ibm.com>
* Add directory structureStephen Cprek2015-02-123-0/+0
Change-Id: I9d2fdc9757d060cc07ee3ea5d3622d0fd9d3d2ba Reviewed-on: http://gfw160.aus.stglabs.ibm.com:8080/gerrit/15479 Reviewed-by: Derk Rembold <rembold@de.ibm.com> Tested-by: Derk Rembold <rembold@de.ibm.com> Reviewed-by: A. Patrick Williams III <iawillia@us.ibm.com>
OpenPOWER on IntegriCloud