History log of /opensolaris-onvv-gate/usr/src/uts/intel/sys/mc_amd.h
Revision Date Author Comments
# 11947:e9d33e5d3842 17-Mar-2010 Srihari Venkatesan <srihari.venkatesan@oracle.com>

6850096 Error reports should include identification for x8 ECC Symbolsize in syndrome-type
Contributed by Hans.Rosenfeld@amd.com


# 10026:2659dc230676 02-Jul-2009 Kuriakose Kuruvilla <Kuriakose.Kuruvilla@Sun.COM>

6834787 Handle minor cpuid/revision changes for AMD processors
6839150 AMD Socket Identifications impact FMA on AMD Family 0xF topologies


# 5327:7341f36188f5 24-Oct-2007 gavinm

6619031 avoid reading MCi_MISC during init until gp fault handling is improved
6620907 AMD family 0xf NB watchdog policy does not force enable
6620917 Perform additional NB MCA Config for AMD family 0x10
6620921 avoid all cores polling NB bank for AMD family 0x10
6620982 poller starts even if mca init declines


# 5254:38162db71c7d 14-Oct-2007 gavinm

PSARC 2007/591 Generic x86 Machine Check Architecture
PSARC 2007/594 Intel CPU and 5000/7300 Series Chipset FMA
6443855 x86 modinfo slots being wasted during cpu module path search
6475380 cmi_load() should unload unused CPU module
6520280 MCA status bits not properly being cleared for intel processors
6558878 topo_node_hash returns out of range hash
6567218 FMA for Intel processors and 5000/7300 series memory controller
6567634 generic x86 MCA support needs to be improved
6607616 prepare FMA/x86 for xVM
6607626 eversholt Makefile.com lacks a default target
6607637 Add number of cpus and cores per chip to cpu_info kstat
6607643 x86 on_trap handler jumps from frying pan into the fire
6616180 topo should translate slashes in authority fields


# 3766:0e6dc235b6f9 05-Mar-2007 gavinm

6504327 make with no target in i86pc fails
6521185 Solaris trusts BIOS too much wrt AMD erratum 172
6528027 fallback to cpu.generic if number of mca banks not as expected


# 3323:3cff27a475a5 22-Dec-2006 cindi

PSARC 2006/679 Fault Event Extension
6279682 fmadm -q rotate errlog hangs while fmd report an error into the log
6326650 fma needs more dcmds for getting errors from crash dumps
6372709 fltlog can grow without bound; needs to be managed by logadm
6400607 fmd dumped core in zfs-diagnosis
6464740 fmdump -v output should display FRU location (label) information separate from the FRU
6485057 mc-amd driver should provide access to HT routing tables
6493219 libtopo needs to allow re-enumeration according static .xml props
6495169 PCI topo enumerator should maintain a 'module' property
6496757 Incorrect initializer for topo_builtin.c:_topo_builtins[]
6496850 Need mdb support for libtopo
6498181 .po files inconsistencies wrt the event registry content
6498187 SCA1000 and SCA500 .po dictionary files are out of date
6498195 SMF dictionary .po files are inconsistent
6498201 ZFS dictionary .po files are out of date
6502660 eft fills in ASRU and FRU from libtopo when none are defined in a fault event specification
6502949 eft memory leak in platform_fault2ipath()
6506108 topo_fmri_nvl2str on mem scheme fails to lookup offset and physaddr
6506622 missing entry in PCIEX dict/po files


# 3164:475efbf127ed 24-Nov-2006 gavinm

6474853 Solaris should be cognizant of the platform while attempting to turn off smi mc polling
6489753 mc_amd.h not quite ready for revision G
6491720 DC/IC/BU system infill only capture physical <39:6>
6495485 i86pc/cpu/scripts do not get clobbered


# 2869:324151eecd58 05-Oct-2006 gavinm

PSARC 2006/564 FMA for Athlon 64 and Opteron Rev F/G Processors
PSARC 2006/566 eversholt language enhancements: confprop_defined
6362846 eversholt doesn't allow dashes in pathname components
6391591 AMD NB config should not set NbMcaToMstCpuEn
6391605 AMD DRAM scrubber should be disabled when errata #99 applies
6398506 memory controller driver should not bother to attach at all on rev F
6424822 FMA needs to support AMD family 0xf revs F and G
6443847 FMA x64 multibit ChipKill rules need to follow MQSC guidelines
6443849 Accrue inf_sys and s_ecc ECC errors against memory
6443858 mc-amd can free unitsrtr before usage in subsequent error path
6443891 mc-amd does not recognise mismatched dimm support
6455363 x86 error injector should allow addr option for most errors
6455370 Opteron erratum 101 only applies on revs D and earlier
6455373 Identify chip-select lines used on a dimm
6455377 improve x64 quadrank dimm support
6455382 add generic interfaces for amd chip revision and package/socket type
6468723 mem scheme fmri containment test for hc scheme is busted
6473807 eversholt could use some mdb support
6473811 eversholt needs a confprop_defined function
6473819 eversholt should show version of rules active in DE
6475302 ::nvlist broken by some runtime link ordering changes


# 1414:b4126407ac5b 11-Feb-2006 cindi

PSARC 2006/020 FMA for Athlon 64 and Opteron Processors
PSARC 2006/028 eversholt language enhancements
6181364 Eversholt needs method to revise value of a fault's property
6183842 eft can construct extra propagations in the instance tree
6187143 eversholt needs to use fmd_case_add_serd() to add counted ereports against open case
6232253 wildcarding may not pick up matches buried in config path
6284455 eversholt wildcarding and vertical expansion have trouble working together
6298484 properties are not auto-converting to integers in eversholt constraints
6298972 eversholt should be able to mark faults as no-message like the cpumem DE
6298974 nested SERD engines don't work
6298981 eft memory usage could improve by caching common constraint expressions
6323319 call() is not allowing string-valued returns
6323322 a global variable should be allowed as the RHS of an nvpair
6323393 eversholt caches a little too much info when caching constraints
6323554 eversholt type conversion can cause core dump
6328144 libexacct leaks like a really big sieve when faced with non-exacct input
6331093 payloadprop should be able to read and interpret hc scheme fmris
6332245 payloadprop() returns cached value from existing FME when not appropriate
6333617 eversholt should have way to check if a global is defined
6346926 eversholt needs a way to maintain diagnosis statistics
6359264 Provide FMA support for AMD64 processors
6363503 Can not register error handler callbacks for root node
6366821 cpu scheme serial number should be a string
6367031 eft.so leaks memory
6370284 cpumem-diagnosis checks the asru version against FM_EREPORT_VERSION instead of FM_CPU_SCHEME_VERSION
6377319 eft could close cases for resources already in the faulty state
6379498 fmd dies on assertion failure when repairing an fmd module
6381022 fmd_case_insert_event() should reject duplicates and save memory


# 11947:e9d33e5d3842 17-Mar-2010 Srihari Venkatesan <srihari.venkatesan@oracle.com>

6850096 Error reports should include identification for x8 ECC Symbolsize in syndrome-type
Contributed by Hans.Rosenfeld@amd.com


# 10026:2659dc230676 02-Jul-2009 Kuriakose Kuruvilla <Kuriakose.Kuruvilla@Sun.COM>

6834787 Handle minor cpuid/revision changes for AMD processors
6839150 AMD Socket Identifications impact FMA on AMD Family 0xF topologies


# 5327:7341f36188f5 24-Oct-2007 gavinm

6619031 avoid reading MCi_MISC during init until gp fault handling is improved
6620907 AMD family 0xf NB watchdog policy does not force enable
6620917 Perform additional NB MCA Config for AMD family 0x10
6620921 avoid all cores polling NB bank for AMD family 0x10
6620982 poller starts even if mca init declines


# 5254:38162db71c7d 14-Oct-2007 gavinm

PSARC 2007/591 Generic x86 Machine Check Architecture
PSARC 2007/594 Intel CPU and 5000/7300 Series Chipset FMA
6443855 x86 modinfo slots being wasted during cpu module path search
6475380 cmi_load() should unload unused CPU module
6520280 MCA status bits not properly being cleared for intel processors
6558878 topo_node_hash returns out of range hash
6567218 FMA for Intel processors and 5000/7300 series memory controller
6567634 generic x86 MCA support needs to be improved
6607616 prepare FMA/x86 for xVM
6607626 eversholt Makefile.com lacks a default target
6607637 Add number of cpus and cores per chip to cpu_info kstat
6607643 x86 on_trap handler jumps from frying pan into the fire
6616180 topo should translate slashes in authority fields


# 3766:0e6dc235b6f9 05-Mar-2007 gavinm

6504327 make with no target in i86pc fails
6521185 Solaris trusts BIOS too much wrt AMD erratum 172
6528027 fallback to cpu.generic if number of mca banks not as expected


# 3164:475efbf127ed 24-Nov-2006 gavinm

6474853 Solaris should be cognizant of the platform while attempting to turn off smi mc polling
6489753 mc_amd.h not quite ready for revision G
6491720 DC/IC/BU system infill only capture physical <39:6>
6495485 i86pc/cpu/scripts do not get clobbered


# 2869:324151eecd58 05-Oct-2006 gavinm

PSARC 2006/564 FMA for Athlon 64 and Opteron Rev F/G Processors
PSARC 2006/566 eversholt language enhancements: confprop_defined
6362846 eversholt doesn't allow dashes in pathname components
6391591 AMD NB config should not set NbMcaToMstCpuEn
6391605 AMD DRAM scrubber should be disabled when errata #99 applies
6398506 memory controller driver should not bother to attach at all on rev F
6424822 FMA needs to support AMD family 0xf revs F and G
6443847 FMA x64 multibit ChipKill rules need to follow MQSC guidelines
6443849 Accrue inf_sys and s_ecc ECC errors against memory
6443858 mc-amd can free unitsrtr before usage in subsequent error path
6443891 mc-amd does not recognise mismatched dimm support
6455363 x86 error injector should allow addr option for most errors
6455370 Opteron erratum 101 only applies on revs D and earlier
6455373 Identify chip-select lines used on a dimm
6455377 improve x64 quadrank dimm support
6455382 add generic interfaces for amd chip revision and package/socket type
6468723 mem scheme fmri containment test for hc scheme is busted
6473807 eversholt could use some mdb support
6473811 eversholt needs a confprop_defined function
6473819 eversholt should show version of rules active in DE
6475302 ::nvlist broken by some runtime link ordering changes


# 1414:b4126407ac5b 11-Feb-2006 cindi

PSARC 2006/020 FMA for Athlon 64 and Opteron Processors
PSARC 2006/028 eversholt language enhancements
6181364 Eversholt needs method to revise value of a fault's property
6183842 eft can construct extra propagations in the instance tree
6187143 eversholt needs to use fmd_case_add_serd() to add counted ereports against open case
6232253 wildcarding may not pick up matches buried in config path
6284455 eversholt wildcarding and vertical expansion have trouble working together
6298484 properties are not auto-converting to integers in eversholt constraints
6298972 eversholt should be able to mark faults as no-message like the cpumem DE
6298974 nested SERD engines don't work
6298981 eft memory usage could improve by caching common constraint expressions
6323319 call() is not allowing string-valued returns
6323322 a global variable should be allowed as the RHS of an nvpair
6323393 eversholt caches a little too much info when caching constraints
6323554 eversholt type conversion can cause core dump
6328144 libexacct leaks like a really big sieve when faced with non-exacct input
6331093 payloadprop should be able to read and interpret hc scheme fmris
6332245 payloadprop() returns cached value from existing FME when not appropriate
6333617 eversholt should have way to check if a global is defined
6346926 eversholt needs a way to maintain diagnosis statistics
6359264 Provide FMA support for AMD64 processors
6363503 Can not register error handler callbacks for root node
6366821 cpu scheme serial number should be a string
6367031 eft.so leaks memory
6370284 cpumem-diagnosis checks the asru version against FM_EREPORT_VERSION instead of FM_CPU_SCHEME_VERSION
6377319 eft could close cases for resources already in the faulty state
6379498 fmd dies on assertion failure when repairing an fmd module
6381022 fmd_case_insert_event() should reject duplicates and save memory