talos-hostboot/src/include/usr/ipmi, branch 07-25-2019

ipmidcmi: Get DCMI capabilities info to check PM support

2019-07-23T18:54:41+00:00

According to section 6.6 Power management of DCMI spec, the DCMI Get Power Limit command isn`t supported if Get DCMI Capabilities Info command indicates no Power Management support. Commit adds code to get DCMI capabilities information from BMC to check this Tested on the Vesnin P8 server Change-Id: I3c4216e2ee4f5fa98ffbcfa95c4af6771f81b43c RTC: 123256 Resolves #176 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/80681 Tested-by: Jenkins Server Tested-by: Jenkins OP Build CI Tested-by: Jenkins OP HW Tested-by: FSP CI Jenkins Reviewed-by: Martha Broyles Reviewed-by: Daniel M Crowell

Reduce ipmi trace spam for pnor hiomap messages

2018-11-27T22:28:17+00:00

Enable IPMI errl after targeting is initialized

2018-11-05T17:51:47+00:00

A recent change split IPMI into base and extended. Move the enable of IPMI errl from base to extended to avoid targeting errors. Change-Id: Ia833dd8178ff407dff1fa514b32468c5fc70e945 CQ:SW450286 Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/68265 Tested-by: Jenkins Server Tested-by: Jenkins OP Build CI Tested-by: Jenkins OP HW Tested-by: FSP CI Jenkins Reviewed-by: ANDREW R. JEFFERY Reviewed-by: Michael Baiocchi Reviewed-by: Daniel M. Crowell

pnor: Introduce an IPMI-based PNOR driver implementation

2018-10-10T18:45:29+00:00

Similar to the AST MBOX implementation, the IPMI PNOR implementation negotiates the layout of the LPC FW space with the BMC, but using IPMI rather than the AST mailbox as a protocol transport. The same protocol is still used and has simply been adapted to the new interface. Note that currently the change of transport has had a 2-3x impact on boot performance. Optimisation is an ongoing effort. Change-Id: I7f838f5b5e88ac877a725386a33df58ee5e7213c Signed-off-by: Andrew Jeffery Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/65942 Tested-by: Jenkins Server Tested-by: Jenkins OP HW Tested-by: FSP CI Jenkins Reviewed-by: Daniel M. Crowell

ipmi: Terminate SEL task via shutdown event

2018-10-09T15:07:42+00:00

The IpmiSEL task will become part of the extended image once the IPMI module is split in two. Once split, if we need to handle an early shutdown we cannot be referencing code from HBI in HBB. To that end, ensure we don't instantiate an IpmiSEL via Singleton in an attempt to shutdown its event loop when it hasn't been loaded, let alone started. Instead, have IpmiSEL register itself in the shutdown handler, and shut down the SEL task before IpmiRP. Change-Id: I358f6cb1f5528a4ad72c93477ad883cad19e2bf6 Signed-off-by: Andrew Jeffery Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/67076 Tested-by: Jenkins Server Tested-by: Jenkins OP Build CI Tested-by: Jenkins OP HW Tested-by: FSP CI Jenkins Reviewed-by: Corey V. Swenson Reviewed-by: Daniel M. Crowell

ipmi: Introduce register_for_event() interface

2018-10-08T20:29:30+00:00

register_for_event() allows the IPMI PNOR implementation to indicate interest in SELs from the hiomap protocol used for managing the LPC FW space. Change-Id: I3bf6cb7f860d41a0c46755e23fd54276ae2258ff Signed-off-by: Andrew Jeffery Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/65938 Tested-by: Jenkins Server Tested-by: Jenkins OP Build CI Tested-by: Jenkins OP HW Tested-by: FSP CI Jenkins Reviewed-by: Daniel M. Crowell

Send errors from previous boots as callhome type eSELs

2018-07-17T21:18:04+00:00

During early boot, Hostboot attempts to resend unacknowledged error logs from prior boots as eSELS, without correponding SELs. BMCs typically require both in order to expose a given error log to a customer. This change morphs errors from prior boots into callhome type logs, so that a simple eSEL will be enough to get the error propagated. Change-Id: If499defe8a39b9254f08392b264d72047b7e5f7c CQ: SW426731 RTC: 193265 Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/62079 Tested-by: Jenkins Server Tested-by: Jenkins OP Build CI Tested-by: Jenkins OP HW Tested-by: FSP CI Jenkins Reviewed-by: William G. Hoffa

Increase default watchdog timeout to 10 minutes

2018-05-29T14:33:43+00:00

We have discovered some error cases within the memory diagnostics step where a single hardware operation can take up to 6 minutes to complete. The code sends a watchdog ping before issuing any command, but the current timeout (2 minutes) is insufficient to handle these long commands. Resolves boston-openpower/#1240 Change-Id: I4e27a45c927c014b5fcc28d4510c5996a7009953 Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/59254 Tested-by: Jenkins Server Tested-by: Jenkins OP Build CI Tested-by: Jenkins OP HW Tested-by: FSP CI Jenkins Reviewed-by: William G. Hoffa Reviewed-by: Daniel M. Crowell

Get SN from BMC and update into PVPD EEPROM

2018-05-19T21:09:16+00:00

Resolves #135 Change-Id: I46b5c2208ed28822c6001a955f4056d976e79339 Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/58010 Tested-by: Jenkins Server Tested-by: Jenkins OP Build CI Tested-by: Jenkins OP HW Tested-by: FSP CI Jenkins Reviewed-by: Daniel M. Crowell

Enable super-long watchdog timer when console traces are enabled

2018-05-02T17:53:54+00:00

When console traces are enabled, the boot takes significantly longer than usual. Our standard watchdog timeout of 2 minutes is not sufficient. Made the following changes: - Tied the enablement of the long timeout to the CONFIG flags for console traces - Increased the timeout to 1 hour (from 40 minutes) since this is lab only and false positives are the worst... - Added a couple clarifying traces to make future watchdog debug simpler Change-Id: I92bb0b07cb77a13cb390450999b285f2ad74168d CQ: SW425407 Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/57580 Reviewed-by: Martin Gloff Reviewed-by: Richard J. Knight Tested-by: Jenkins Server Tested-by: Jenkins OP Build CI Tested-by: FSP CI Jenkins Reviewed-by: Christian R. Geddes Tested-by: Jenkins OP HW Reviewed-by: William G. Hoffa