€•ø7      Œsphinx.addnodes”Œdocument”“”)”}”(Œ	rawsource”Œ ”Œchildren”]”(Œtranslations”ŒLanguagesNode”“”)”}”(hhh]”(h Œpending_xref”“”)”}”(hhh]”Œdocutils.nodes”ŒText”“”ŒChinese (Simplified)”…””}”Œparent”hsbaŒ
attributes”}”(Œids”]”Œclasses”]”Œnames”]”Œdupnames”]”Œbackrefs”]”Œ	refdomain”Œstd”Œreftype”Œdoc”Œ	reftarget”Œ,/translations/zh_CN/arch/powerpc/dawr-power9”Œmodname”NŒ	classname”NŒrefexplicit”ˆuŒtagname”hhhubh)”}”(hhh]”hŒChinese (Traditional)”…””}”hh2sbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ,/translations/zh_TW/arch/powerpc/dawr-power9”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒItalian”…””}”hhFsbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ,/translations/it_IT/arch/powerpc/dawr-power9”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒJapanese”…””}”hhZsbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ,/translations/ja_JP/arch/powerpc/dawr-power9”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒKorean”…””}”hhnsbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ,/translations/ko_KR/arch/powerpc/dawr-power9”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒSpanish”…””}”hh‚sbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ,/translations/sp_SP/arch/powerpc/dawr-power9”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubeh}”(h]”h ]”h"]”h$]”h&]”Œcurrent_language”ŒEnglish”uh1h
hhŒ	_document”hŒsource”NŒline”NubhŒsection”“”)”}”(hhh]”(hŒtitle”“”)”}”(hŒDAWR issues on POWER9”h]”hŒDAWR issues on POWER9”…””}”(hh¨hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h¦hh£hžhhŸŒF/var/lib/git/docbuild/linux/Documentation/arch/powerpc/dawr-power9.rst”h KubhŒ	paragraph”“”)”}”(hX  On older POWER9 processors, the Data Address Watchpoint Register (DAWR) can
cause a checkstop if it points to cache inhibited (CI) memory. Currently Linux
has no way to distinguish CI memory when configuring the DAWR, so on affected
systems, the DAWR is disabled.”h]”hX  On older POWER9 processors, the Data Address Watchpoint Register (DAWR) can
cause a checkstop if it points to cache inhibited (CI) memory. Currently Linux
has no way to distinguish CI memory when configuring the DAWR, so on affected
systems, the DAWR is disabled.”…””}”(hh¹hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h Khh£hžhubh¢)”}”(hhh]”(h§)”}”(hŒAffected processor revisions”h]”hŒAffected processor revisions”…””}”(hhÊhžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h¦hhÇhžhhŸh¶h Kubh¸)”}”(hŒdThis issue is only present on processors prior to v2.3. The revision can be
found in /proc/cpuinfo::”h]”hŒcThis issue is only present on processors prior to v2.3. The revision can be
found in /proc/cpuinfo:”…””}”(hhØhžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h KhhÇhžhubhŒliteral_block”“”)”}”(hŒ†processor       : 0
cpu             : POWER9, altivec supported
clock           : 3800.000000MHz
revision        : 2.3 (pvr 004e 1203)”h]”hŒ†processor       : 0
cpu             : POWER9, altivec supported
clock           : 3800.000000MHz
revision        : 2.3 (pvr 004e 1203)”…””}”hhèsbah}”(h]”h ]”h"]”h$]”h&]”Œ	xml:space”Œpreserve”uh1hæhŸh¶h KhhÇhžhubh¸)”}”(hŒCOn a system with the issue, the DAWR is disabled as detailed below.”h]”hŒCOn a system with the issue, the DAWR is disabled as detailed below.”…””}”(hhøhžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h KhhÇhžhubeh}”(h]”Œaffected-processor-revisions”ah ]”h"]”Œaffected processor revisions”ah$]”h&]”uh1h¡hh£hžhhŸh¶h Kubh¢)”}”(hhh]”(h§)”}”(hŒTechnical Details:”h]”hŒTechnical Details:”…””}”(hj  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h¦hj  hžhhŸh¶h Kubh¸)”}”(hŒuDAWR has 6 different ways of being set.
1) ptrace
2) h_set_mode(DAWR)
3) h_set_dabr()
4) kvmppc_set_one_reg()
5) xmon”h]”hŒuDAWR has 6 different ways of being set.
1) ptrace
2) h_set_mode(DAWR)
3) h_set_dabr()
4) kvmppc_set_one_reg()
5) xmon”…””}”(hj  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h Khj  hžhubh¸)”}”(hŒ·For ptrace, we now advertise zero breakpoints on POWER9 via the
PPC_PTRACE_GETHWDBGINFO call. This results in GDB falling back to
software emulation of the watchpoint (which is slow).”h]”hŒ·For ptrace, we now advertise zero breakpoints on POWER9 via the
PPC_PTRACE_GETHWDBGINFO call. This results in GDB falling back to
software emulation of the watchpoint (which is slow).”…””}”(hj-  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h K!hj  hžhubh¸)”}”(hŒ©h_set_mode(DAWR) and h_set_dabr() will now return an error to the
guest on a POWER9 host. Current Linux guests ignore this error, so
they will silently not get the DAWR.”h]”hŒ©h_set_mode(DAWR) and h_set_dabr() will now return an error to the
guest on a POWER9 host. Current Linux guests ignore this error, so
they will silently not get the DAWR.”…””}”(hj;  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h K%hj  hžhubh¸)”}”(hŒÝkvmppc_set_one_reg() will store the value in the vcpu but won't
actually set it on POWER9 hardware. This is done so we don't break
migration from POWER8 to POWER9, at the cost of silently losing the
DAWR on the migration.”h]”hŒákvmppc_set_one_reg() will store the value in the vcpu but wonâ€™t
actually set it on POWER9 hardware. This is done so we donâ€™t break
migration from POWER8 to POWER9, at the cost of silently losing the
DAWR on the migration.”…””}”(hjI  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h K)hj  hžhubh¸)”}”(hŒ6For xmon, the 'bd' command will return an error on P9.”h]”hŒ:For xmon, the â€˜bdâ€™ command will return an error on P9.”…””}”(hjW  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h K.hj  hžhubeh}”(h]”Œtechnical-details”ah ]”h"]”Œtechnical details:”ah$]”h&]”uh1h¡hh£hžhhŸh¶h Kubh¢)”}”(hhh]”(h§)”}”(hŒConsequences for users”h]”hŒConsequences for users”…””}”(hjp  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h¦hjm  hžhhŸh¶h K1ubh¸)”}”(hŒåFor GDB watchpoints (ie 'watch' command) on POWER9 bare metal , GDB
will accept the command. Unfortunately since there is no hardware
support for the watchpoint, GDB will software emulate the watchpoint
making it run very slowly.”h]”hŒéFor GDB watchpoints (ie â€˜watchâ€™ command) on POWER9 bare metal , GDB
will accept the command. Unfortunately since there is no hardware
support for the watchpoint, GDB will software emulate the watchpoint
making it run very slowly.”…””}”(hj~  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h K3hjm  hžhubh¸)”}”(hŒŠThe same will also be true for any guests started on a POWER9
host. The watchpoint will fail and GDB will fall back to software
emulation.”h]”hŒŠThe same will also be true for any guests started on a POWER9
host. The watchpoint will fail and GDB will fall back to software
emulation.”…””}”(hjŒ  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h K8hjm  hžhubh¸)”}”(hXÕ  If a guest is started on a POWER8 host, GDB will accept the watchpoint
and configure the hardware to use the DAWR. This will run at full
speed since it can use the hardware emulation. Unfortunately if this
guest is migrated to a POWER9 host, the watchpoint will be lost on the
POWER9. Loads and stores to the watchpoint locations will not be
trapped in GDB. The watchpoint is remembered, so if the guest is
migrated back to the POWER8 host, it will start working again.”h]”hXÕ  If a guest is started on a POWER8 host, GDB will accept the watchpoint
and configure the hardware to use the DAWR. This will run at full
speed since it can use the hardware emulation. Unfortunately if this
guest is migrated to a POWER9 host, the watchpoint will be lost on the
POWER9. Loads and stores to the watchpoint locations will not be
trapped in GDB. The watchpoint is remembered, so if the guest is
migrated back to the POWER8 host, it will start working again.”…””}”(hjš  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h K<hjm  hžhubeh}”(h]”Œconsequences-for-users”ah ]”h"]”Œconsequences for users”ah$]”h&]”uh1h¡hh£hžhhŸh¶h K1ubh¢)”}”(hhh]”(h§)”}”(hŒForce enabling the DAWR”h]”hŒForce enabling the DAWR”…””}”(hj³  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h¦hj°  hžhhŸh¶h KEubh¸)”}”(hŒCKernels (since ~v5.2) have an option to force enable the DAWR via::”h]”hŒBKernels (since ~v5.2) have an option to force enable the DAWR via:”…””}”(hjÁ  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h KFhj°  hžhubhç)”}”(hŒ8echo Y > /sys/kernel/debug/powerpc/dawr_enable_dangerous”h]”hŒ8echo Y > /sys/kernel/debug/powerpc/dawr_enable_dangerous”…””}”hjÏ  sbah}”(h]”h ]”h"]”h$]”h&]”höh÷uh1hæhŸh¶h KHhj°  hžhubh¸)”}”(hŒ%This enables the DAWR even on POWER9.”h]”hŒ%This enables the DAWR even on POWER9.”…””}”(hjÝ  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h KJhj°  hžhubh¸)”}”(hŒ2This is a dangerous setting, USE AT YOUR OWN RISK.”h]”hŒ2This is a dangerous setting, USE AT YOUR OWN RISK.”…””}”(hjë  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h KLhj°  hžhubh¸)”}”(hŒŸSome users may not care about a bad user crashing their box
(ie. single user/desktop systems) and really want the DAWR.  This
allows them to force enable DAWR.”h]”hŒŸSome users may not care about a bad user crashing their box
(ie. single user/desktop systems) and really want the DAWR.  This
allows them to force enable DAWR.”…””}”(hjù  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h KNhj°  hžhubh¸)”}”(hŒ¦This flag can also be used to disable DAWR access. Once this is
cleared, all DAWR access should be cleared immediately and your
machine once again safe from crashing.”h]”hŒ¦This flag can also be used to disable DAWR access. Once this is
cleared, all DAWR access should be cleared immediately and your
machine once again safe from crashing.”…””}”(hj  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h KRhj°  hžhubh¸)”}”(hX  Userspace may get confused by toggling this. If DAWR is force
enabled/disabled between getting the number of breakpoints (via
PTRACE_GETHWDBGINFO) and setting the breakpoint, userspace will get an
inconsistent view of what's available. Similarly for guests.”h]”hX  Userspace may get confused by toggling this. If DAWR is force
enabled/disabled between getting the number of breakpoints (via
PTRACE_GETHWDBGINFO) and setting the breakpoint, userspace will get an
inconsistent view of whatâ€™s available. Similarly for guests.”…””}”(hj  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h KVhj°  hžhubh¸)”}”(hX)  For the DAWR to be enabled in a KVM guest, the DAWR needs to be force
enabled in the host AND the guest. For this reason, this won't work on
POWERVM as it doesn't allow the HCALL to work. Writes of 'Y' to the
dawr_enable_dangerous file will fail if the hypervisor doesn't support
writing the DAWR.”h]”hX3  For the DAWR to be enabled in a KVM guest, the DAWR needs to be force
enabled in the host AND the guest. For this reason, this wonâ€™t work on
POWERVM as it doesnâ€™t allow the HCALL to work. Writes of â€˜Yâ€™ to the
dawr_enable_dangerous file will fail if the hypervisor doesnâ€™t support
writing the DAWR.”…””}”(hj#  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h K[hj°  hžhubh¸)”}”(hŒ>To double check the DAWR is working, run this kernel selftest:”h]”hŒ>To double check the DAWR is working, run this kernel selftest:”…””}”(hj1  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h Kahj°  hžhubhŒblock_quote”“”)”}”(hŒ8tools/testing/selftests/powerpc/ptrace/ptrace-hwbreak.c
”h]”h¸)”}”(hŒ7tools/testing/selftests/powerpc/ptrace/ptrace-hwbreak.c”h]”hŒ7tools/testing/selftests/powerpc/ptrace/ptrace-hwbreak.c”…””}”(hjE  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h KchjA  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j?  hŸh¶h Kchj°  hžhubh¸)”}”(hŒ2Any errors/failures/skips mean something is wrong.”h]”hŒ2Any errors/failures/skips mean something is wrong.”…””}”(hjY  hžhhŸNh Nubah}”(h]”h ]”h"]”h$]”h&]”uh1h·hŸh¶h Kehj°  hžhubeh}”(h]”Œforce-enabling-the-dawr”ah ]”h"]”Œforce enabling the dawr”ah$]”h&]”uh1h¡hh£hžhhŸh¶h KEubeh}”(h]”Œdawr-issues-on-power9”ah ]”h"]”Œdawr issues on power9”ah$]”h&]”uh1h¡hhhžhhŸh¶h Kubeh}”(h]”h ]”h"]”h$]”h&]”Œsource”h¶uh1hŒcurrent_source”NŒcurrent_line”NŒsettings”Œdocutils.frontend”ŒValues”“”)”}”(h¦NŒ	generator”NŒ	datestamp”NŒsource_link”NŒ
source_url”NŒtoc_backlinks”Œentry”Œfootnote_backlinks”KŒsectnum_xform”KŒstrip_comments”NŒstrip_elements_with_classes”NŒstrip_classes”NŒreport_level”KŒ
halt_level”KŒexit_status_level”KŒdebug”NŒwarning_stream”NŒ	traceback”ˆŒinput_encoding”Œ	utf-8-sig”Œinput_encoding_error_handler”Œstrict”Œoutput_encoding”Œutf-8”Œoutput_encoding_error_handler”jš  Œerror_encoding”Œutf-8”Œerror_encoding_error_handler”Œbackslashreplace”Œlanguage_code”Œen”Œrecord_dependencies”NŒconfig”NŒ	id_prefix”hŒauto_id_prefix”Œid”Œdump_settings”NŒdump_internals”NŒdump_transforms”NŒdump_pseudo_xml”NŒexpose_internals”NŒstrict_visitor”NŒ_disable_config”NŒ_source”h¶Œ_destination”NŒ_config_files”]”Œ7/var/lib/git/docbuild/linux/Documentation/docutils.conf”aŒfile_insertion_enabled”ˆŒraw_enabled”KŒline_length_limit”M'Œpep_references”NŒpep_base_url”Œhttps://peps.python.org/”Œpep_file_url_template”Œpep-%04d”Œrfc_references”NŒrfc_base_url”Œ&https://datatracker.ietf.org/doc/html/”Œ	tab_width”KŒtrim_footnote_reference_space”‰Œsyntax_highlight”Œlong”Œsmart_quotes”ˆŒsmartquotes_locales”]”Œcharacter_level_inline_markup”‰Œdoctitle_xform”‰Œdocinfo_xform”KŒsectsubtitle_xform”‰Œimage_loading”Œlink”Œembed_stylesheet”‰Œcloak_email_addresses”ˆŒsection_self_link”‰Œenv”NubŒreporter”NŒindirect_targets”]”Œsubstitution_defs”}”Œsubstitution_names”}”Œrefnames”}”Œrefids”}”Œnameids”}”(jt  jq  j  j  jj  jg  j­  jª  jl  ji  uŒ	nametypes”}”(jt  ‰j  ‰jj  ‰j­  ‰jl  ‰uh}”(jq  h£j  hÇjg  j  jª  jm  ji  j°  uŒfootnote_refs”}”Œcitation_refs”}”Œautofootnotes”]”Œautofootnote_refs”]”Œsymbol_footnotes”]”Œsymbol_footnote_refs”]”Œ	footnotes”]”Œ	citations”]”Œautofootnote_start”KŒsymbol_footnote_start”K Œ
id_counter”Œcollections”ŒCounter”“”}”…”R”Œparse_messages”]”Œtransform_messages”]”Œtransformer”NŒinclude_log”]”Œ
decoration”Nhžhub.