<feed xmlns='http://www.w3.org/2005/Atom'>
<title>talos-hostboot/src/usr/xscom, branch master</title>
<subtitle>Talos™ II hostboot sources</subtitle>
<id>https://git.raptorcs.com/git/talos-hostboot/atom?h=master</id>
<link rel='self' href='https://git.raptorcs.com/git/talos-hostboot/atom?h=master'/>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/talos-hostboot/'/>
<updated>2018-04-18T21:24:54+00:00</updated>
<entry>
<title>Set the runtime scom trace buffers with constants</title>
<updated>2018-04-18T21:24:54+00:00</updated>
<author>
<name>Corey Swenson</name>
<email>cswenson@us.ibm.com</email>
</author>
<published>2018-04-04T19:16:30+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/talos-hostboot/commit/?id=e86221ce862640b1bc72b26182ef166a75a8ef1e'/>
<id>urn:sha1:e86221ce862640b1bc72b26182ef166a75a8ef1e</id>
<content type='text'>
Change-Id: I6c5e0ce2dde28210125cf698cc4c27b7fe3bdb1e
Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/56757
Tested-by: Jenkins Server &lt;pfd-jenkins+hostboot@us.ibm.com&gt;
Tested-by: Jenkins OP Build CI &lt;op-jenkins+hostboot@us.ibm.com&gt;
Tested-by: Jenkins OP HW &lt;op-hw-jenkins+hostboot@us.ibm.com&gt;
Tested-by: FSP CI Jenkins &lt;fsp-CI-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: Martin Gloff &lt;mgloff@us.ibm.com&gt;
Reviewed-by: Christian R. Geddes &lt;crgeddes@us.ibm.com&gt;
Reviewed-by: Daniel M. Crowell &lt;dcrowell@us.ibm.com&gt;
</content>
</entry>
<entry>
<title>Add ibscom runtime support</title>
<updated>2018-04-02T17:44:41+00:00</updated>
<author>
<name>Corey Swenson</name>
<email>cswenson@us.ibm.com</email>
</author>
<published>2018-03-23T15:49:35+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/talos-hostboot/commit/?id=b51298075aee402dbcef485088cfa71a6ca61725'/>
<id>urn:sha1:b51298075aee402dbcef485088cfa71a6ca61725</id>
<content type='text'>
 - Add ibscom runtime files
 - Make ibscom multicast workaround common
 - Call from ibscom rt_ibscom
 - Make scom sanity check common
 - Call from xscom rt_xscom ibscom rt_ibscom

RTC:190144
Change-Id: Ia52feac596ff6b6c8fb9f0db4493f35e4ee837af
Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/56215
Tested-by: Jenkins Server &lt;pfd-jenkins+hostboot@us.ibm.com&gt;
Tested-by: Jenkins OP Build CI &lt;op-jenkins+hostboot@us.ibm.com&gt;
Tested-by: Jenkins OP HW &lt;op-hw-jenkins+hostboot@us.ibm.com&gt;
Tested-by: FSP CI Jenkins &lt;fsp-CI-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: Martin Gloff &lt;mgloff@us.ibm.com&gt;
Reviewed-by: Daniel M. Crowell &lt;dcrowell@us.ibm.com&gt;
</content>
</entry>
<entry>
<title>Force checkstops for unhandled machine checks</title>
<updated>2018-03-29T16:03:50+00:00</updated>
<author>
<name>Dan Crowell</name>
<email>dcrowell@us.ibm.com</email>
</author>
<published>2017-10-01T21:09:56+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/talos-hostboot/commit/?id=90eaed6f430c88eb0127ce47671bd80b21f35433'/>
<id>urn:sha1:90eaed6f430c88eb0127ce47671bd80b21f35433</id>
<content type='text'>
Default MSR[ME]=0 during initial boot for bootloader and
 hostboot kernel
Once the xscom address range has been mapped in, enable the
 machine check handler to force a checkstop and set MSR[ME]=1
 to allow regular machine check handling

CQ: SW401402
Change-Id: I104e39465e61b3b19d5c073e71271102711ae54f
Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/47179
Reviewed-by: Christian R. Geddes &lt;crgeddes@us.ibm.com&gt;
Tested-by: Jenkins Server &lt;pfd-jenkins+hostboot@us.ibm.com&gt;
Tested-by: Jenkins OP Build CI &lt;op-jenkins+hostboot@us.ibm.com&gt;
Tested-by: Jenkins OP HW &lt;op-hw-jenkins+hostboot@us.ibm.com&gt;
Tested-by: FSP CI Jenkins &lt;fsp-CI-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: William G. Hoffa &lt;wghoffa@us.ibm.com&gt;
Reviewed-by: Daniel M. Crowell &lt;dcrowell@us.ibm.com&gt;
</content>
</entry>
<entry>
<title>Move runtime scoms to common function</title>
<updated>2018-03-27T13:50:58+00:00</updated>
<author>
<name>Corey Swenson</name>
<email>cswenson@us.ibm.com</email>
</author>
<published>2018-03-19T21:46:58+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/talos-hostboot/commit/?id=8a783ea89563b0e452caab5c5900add8cc99e358'/>
<id>urn:sha1:8a783ea89563b0e452caab5c5900add8cc99e358</id>
<content type='text'>
  - Calls hypervisor scom interfaces
  - Move from xscom/runtime to scom/runtime

RTC:190143
Change-Id: I41e0788f03fa27ae87c31a48dd041dbd66ca09c2
Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/56042
CI-Ready: Thi N. Tran &lt;thi@us.ibm.com&gt;
Reviewed-by: Christian R. Geddes &lt;crgeddes@us.ibm.com&gt;
Tested-by: Jenkins Server &lt;pfd-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: Martin Gloff &lt;mgloff@us.ibm.com&gt;
Tested-by: Jenkins OP Build CI &lt;op-jenkins+hostboot@us.ibm.com&gt;
Tested-by: Jenkins OP HW &lt;op-hw-jenkins+hostboot@us.ibm.com&gt;
Tested-by: FSP CI Jenkins &lt;fsp-CI-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: Daniel M. Crowell &lt;dcrowell@us.ibm.com&gt;
</content>
</entry>
<entry>
<title>Add common XSCOM error literals to HBRT</title>
<updated>2018-02-27T20:01:01+00:00</updated>
<author>
<name>Brian Bakke</name>
<email>bbakke@us.ibm.com</email>
</author>
<published>2018-02-19T19:43:11+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/talos-hostboot/commit/?id=bb0dc7d71263f7b67f17b297cf16d4535fce5a76'/>
<id>urn:sha1:bb0dc7d71263f7b67f17b297cf16d4535fce5a76</id>
<content type='text'>
Add a set of common XSCOM error literals that can be used
by all platforms.  Add mechanisms to translate this common
set of literals into XSCOM error callouts.  Preserve legacy
handling/translation.

Change-Id: I0f9ce25604efa0caf7d49e4ec7b5f2176a3c3a91
RTC: 86782
Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/54650
Tested-by: Jenkins Server &lt;pfd-jenkins+hostboot@us.ibm.com&gt;
Tested-by: Jenkins OP Build CI &lt;op-jenkins+hostboot@us.ibm.com&gt;
Tested-by: Jenkins OP HW &lt;op-hw-jenkins+hostboot@us.ibm.com&gt;
Tested-by: FSP CI Jenkins &lt;fsp-CI-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: Martin Gloff &lt;mgloff@us.ibm.com&gt;
Reviewed-by: Corey V. Swenson &lt;cswenson@us.ibm.com&gt;
Reviewed-by: Daniel M. Crowell &lt;dcrowell@us.ibm.com&gt;
</content>
</entry>
<entry>
<title>Tweaks to multicast scom workaround</title>
<updated>2018-01-09T15:37:55+00:00</updated>
<author>
<name>Dan Crowell</name>
<email>dcrowell@us.ibm.com</email>
</author>
<published>2017-11-21T21:45:35+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/talos-hostboot/commit/?id=7f8aa4b7a01721e52c0c0b4d4d54c6ba8738c11c'/>
<id>urn:sha1:7f8aa4b7a01721e52c0c0b4d4d54c6ba8738c11c</id>
<content type='text'>
Added a mode to avoid a pile of FFDC collection for the case
where we expect errors could happen.

Also added a couple new flags to control which chiplets get
accessed as part of the workaround, one to avoid access to
slave cores before they are running and the other to avoid
access to the memory chiplets before they have clocks enabled.

Without these changes, the code is still functionally correct
but the traces get filled with pointless non-errors which makes
real problems harder to find and also increases boot time.

Change-Id: I834781261d16748cbbfd8511d9649ec19de53d81
Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/50503
Tested-by: Jenkins Server &lt;pfd-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: Christian R. Geddes &lt;crgeddes@us.ibm.com&gt;
Tested-by: Jenkins OP Build CI &lt;op-jenkins+hostboot@us.ibm.com&gt;
Tested-by: Jenkins OP HW &lt;op-hw-jenkins+hostboot@us.ibm.com&gt;
Tested-by: FSP CI Jenkins &lt;fsp-CI-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: Corey V. Swenson &lt;cswenson@us.ibm.com&gt;
Reviewed-by: Prachi Gupta &lt;pragupta@us.ibm.com&gt;
Reviewed-by: Daniel M. Crowell &lt;dcrowell@us.ibm.com&gt;
</content>
</entry>
<entry>
<title>Set variables to nullptr after they are deleted</title>
<updated>2017-09-25T15:54:57+00:00</updated>
<author>
<name>Christian Geddes</name>
<email>crgeddes@us.ibm.com</email>
</author>
<published>2017-09-08T15:22:52+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/talos-hostboot/commit/?id=1b3b999258db544d6af1dbfb7ee3437efd17581b'/>
<id>urn:sha1:1b3b999258db544d6af1dbfb7ee3437efd17581b</id>
<content type='text'>
While updating some LPC logic a developer noticed a bug where
an errlog handle wasnt getting set to nullptr after it was deleted
this caused a confusing bug that took awhile to track. I noticed
that this bug was all over our code so this commit fixes a lot of
cases where variables get deleted but not set to nullptr

Change-Id: I103b5c71b93686c3c89b04d1d565a24d4de74e6f
Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/45890
Tested-by: Jenkins Server &lt;pfd-jenkins+hostboot@us.ibm.com&gt;
Tested-by: FSP CI Jenkins &lt;fsp-CI-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: William G. Hoffa &lt;wghoffa@us.ibm.com&gt;
Tested-by: Jenkins OP Build CI &lt;op-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: Matt Derksen &lt;mderkse1@us.ibm.com&gt;
Tested-by: Jenkins OP HW &lt;op-hw-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: Daniel M. Crowell &lt;dcrowell@us.ibm.com&gt;
</content>
</entry>
<entry>
<title>Apply multicast offline workaround to hbrt path under opal</title>
<updated>2017-09-08T14:44:11+00:00</updated>
<author>
<name>Dan Crowell</name>
<email>dcrowell@us.ibm.com</email>
</author>
<published>2017-08-31T20:53:08+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/talos-hostboot/commit/?id=4795940579240c074c9cd486a957b4c61a04367c'/>
<id>urn:sha1:4795940579240c074c9cd486a957b4c61a04367c</id>
<content type='text'>
Multicast operations to non-pcbslave registers will return a
piberr of 'chiplet offline' if any chiplets (e.g. cores) are
offline.  To handle this we already ignore this scenario during
IPL.  This change adds some decode logic to convert opal
return codes into their equivalent piberr values.  With this
logic the workaround should trigger and avoid scom errors.

Resolves open-power/boston-openpower/#463

RTC: 86782
Change-Id: I5f587ff9bd1db08e3a5ad5131f8053e8495f45a2
Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/45526
Tested-by: Jenkins Server &lt;pfd-jenkins+hostboot@us.ibm.com&gt;
Tested-by: Jenkins OP Build CI &lt;op-jenkins+hostboot@us.ibm.com&gt;
Tested-by: FSP CI Jenkins &lt;fsp-CI-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: Corey V. Swenson &lt;cswenson@us.ibm.com&gt;
Reviewed-by: Christian R. Geddes &lt;crgeddes@us.ibm.com&gt;
Reviewed-by: Daniel M. Crowell &lt;dcrowell@us.ibm.com&gt;
</content>
</entry>
<entry>
<title>Use XSCOM BAR from Bootloader instead of hardcoded value</title>
<updated>2017-06-07T16:43:35+00:00</updated>
<author>
<name>Dan Crowell</name>
<email>dcrowell@us.ibm.com</email>
</author>
<published>2017-05-30T20:32:26+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/talos-hostboot/commit/?id=e9489e7f7bdf45edb5e274c6b771485532ea906c'/>
<id>urn:sha1:e9489e7f7bdf45edb5e274c6b771485532ea906c</id>
<content type='text'>
The memory map can be modified by the SBE such that the XSCOM
BAR is not in the default location.  The BAR value is passed up
through the bootloader into hostboot.

Change-Id: I469b7534d384bce4bb8c72f7cd78d0075ac04632
RTC: 173519
Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/41140
Reviewed-by: Martin Gloff &lt;mgloff@us.ibm.com&gt;
Tested-by: Jenkins Server &lt;pfd-jenkins+hostboot@us.ibm.com&gt;
Tested-by: FSP CI Jenkins &lt;fsp-CI-jenkins+hostboot@us.ibm.com&gt;
Tested-by: Jenkins OP Build CI &lt;op-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: Christian R. Geddes &lt;crgeddes@us.ibm.com&gt;
Reviewed-by: Daniel M. Crowell &lt;dcrowell@us.ibm.com&gt;
</content>
</entry>
<entry>
<title>Map BAR attributes based on data from Bootloader</title>
<updated>2017-06-02T13:23:59+00:00</updated>
<author>
<name>Dan Crowell</name>
<email>dcrowell@us.ibm.com</email>
</author>
<published>2017-05-10T20:55:40+00:00</published>
<link rel='alternate' type='text/html' href='https://git.raptorcs.com/git/talos-hostboot/commit/?id=6b508aaf86e1d397155ada70bd1678cf5efde9aa'/>
<id>urn:sha1:6b508aaf86e1d397155ada70bd1678cf5efde9aa</id>
<content type='text'>
If the master processor has no memory behind it the entire
memory map must be modified.  Each processor has its own statically
defined map that covers both memory and MMIOs.  If the master
has no memory, its memory map is swapped with another processor.
Each processor gets a new effective fabric id that is then used
to compute all of the BAR values for those processors.

The SBE boots with a certain memory map programmed into the master
processor.  That value is then passed up through the bootloader
into Hostboot.  This value is compared to the BAR values that
Hostboot assumes it is using.  Based on that comparison, various
attributes are computed to match the effective fabric positions.

Change-Id: I2b0d1959c303df8c9c28c8f0a5b5be1e77aa154f
RTC: 173528
Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/40359
Tested-by: Daniel M. Crowell &lt;dcrowell@us.ibm.com&gt;
Reviewed-by: Martin Gloff &lt;mgloff@us.ibm.com&gt;
Reviewed-by: Christian R. Geddes &lt;crgeddes@us.ibm.com&gt;
Tested-by: Jenkins Server &lt;pfd-jenkins+hostboot@us.ibm.com&gt;
Tested-by: FSP CI Jenkins &lt;fsp-CI-jenkins+hostboot@us.ibm.com&gt;
Reviewed-by: William G. Hoffa &lt;wghoffa@us.ibm.com&gt;
</content>
</entry>
</feed>
