History log of /opensolaris-onvv-gate/usr/src/uts/common/vm/page.h
Revision Date Author Comments
# 13076:e0735b974926 11-Aug-2010 Jonathan Adams <Jonathan.Adams@Sun.COM>

6973928 mdb_page_lookup() broken by 6778289; breaks ::pmap


# 13035:0847ab4cffd0 06-Aug-2010 Ondrej Kubecka <Ondrej.Kubecka@Sun.COM>

backout 1225025: needs more work


# 12963:6bd5885dca7c 29-Jul-2010 Stan Studzinski <Stan.Studzinski@Sun.COM>

1225025 mlock:ed anonymous memory remains backed by swap


# 12293:2286b8b6d071 03-May-2010 jmcp <James.McPherson@Sun.COM>

backout 6535949: needs more work


# 12284:3d1135425dbe 29-Apr-2010 Amritpal Sandhu <Paul.Sandhu@Sun.COM>

6535949 availability of 2M pages degrades over time on Solaris/x64


# 12268:1fe5cddd1eb7 27-Apr-2010 Peter Rival <Frank.Rival@oracle.com>

6778289 vm locks need to scale with the size of system (strands/memory size) (fix debug panic)


# 12230:12df01648ca5 23-Apr-2010 Peter Rival <Frank.Rival@oracle.com>

6778289 vm locks need to scale with the size of system (strands/memory size)


# 12156:3c537b2a7425 13-Apr-2010 Stan Studzinski <Stan.Studzinski@Sun.COM>

6675738 KM_NOSLEEP may still try too hard for some allocations


# 12117:79b43747e259 09-Apr-2010 Stan Studzinski <Stan.Studzinski@Sun.COM>

6885904 active segkp pages can not be retired


# 12004:93f274d4a367 25-Mar-2010 Gerry Liu <jiang.liu@intel.com>

PSARC/2009/104 Hot-Plug Support for ACPI-based Systems
PSARC/2009/550 PSMI Extensions for CPU Hotplug
PSARC/2009/551 acpihpd ACPI Hotplug Daemon
PSARC/2009/591 Attachment Points for Hotpluggable x86 Systems
6862510 provide support for cpu hot add on x86
6883891 cmi interface needs to support dynamic reconfiguration
6884154 x2APIC and kmdb may not function properly during CPU hotplug event.
6904971 low priority acpi nexus code review feedback
6877301 lgrp should support memory hotplug flag in SRAT table


# 11185:f0c31008e395 23-Nov-2009 Sean McEnroe <Sean.McEnroe@Sun.COM>

6887924 PP_ISKAS needs to be defined in terms of VN_ISKAS for vnodes
6871169 kphysm_add_memory_dynamic panics testing vmstress + DR
6877160 guest domain panic at platsvc:mdeg_notify_client+10c
6874763 memseg_alloc_meta() incorrectly maps page_t pages
6873569 multiple calls to memlist_read_lock() can cause deadlock hazard
6886354 DR failure with "memory span duplication" error
6886782 panic after pagefault in seg_kpm after LDom add-mem on primary
6887644 domain hang/deadlock during ldom mem DR when trying to grab a write lock


# 10271:7c80b70bb8de 06-Aug-2009 Jason Beloro <Jason.Beloro@Sun.COM>

6858457 Remove Solaris support for UltraSPARC-AT10 processor


# 10106:b235491976d3 16-Jul-2009 Jason Beloro <Jason.Beloro@Sun.COM>

FWARC 2008/540 Memory DR Domain Service
FWARC 2009/300 CPU MD node property for real address bits
6720954 add memory dr feature to ldoms
6742779 fsflush_do_pages() may incorrectly skip constituent file large pages


# 9544:5bcaa92746ac 05-May-2009 Christopher Baumbauer - Sun Microsystems - San Diego United States <Christopher.Baumbauer@Sun.COM>

6587140 page_retire()/page_trycapture should not try to retire non relocatable kernel pages
PSARC 2009/181 Page Retirement Kernel KStat Addition
6819876 pages slated for capture under x86 are being ignored


# 7718:555eee33ff6e 26-Sep-2008 Jason Beloro <Jason.Beloro@Sun.COM>

PSARC/2007/003 Solaris Support for the Rock Processor
6368478 Solaris needs to support the Rock processor
6440653 stores to sun4v error queue head registers are missing a membar #Sync
6639717 FLUSH instruction used more than necessary in sun4v


# 7393:8a8d73648b44 22-Aug-2008 Donghai Qiao <Donghai.Qiao@Sun.COM>

6672470 Looping within uiomove()/xcopyout()/copyout_move()
6707987 kpm has some VAC handling issues
6388567 VAC flushing is broken on US III-IV+ for large pages


# 6880:0a9c7c25ebae 16-Jun-2008 dv142724

6658818 The macro PCF_INDEX is incorrect
6666035 Lock contention for pcf_buckets impacting ISM creation times
6666045 The data structures pcc_info_t and hw_page_map_t need to be cached aligned.


# 5466:8e6f5dfe7459 12-Nov-2007 kchow

6615393 1000x + penalty running with gigantic pages when they are not available.
6623149 trimkcage() may return out of bound pfn range and break large page allocation throttling logic
6625279 new panic on UltraSPARC-IIe during snv_77 (panic[cpu0]/thread=3000cefcc20: BAD TRAP)
6627247 gigantic page processes can hang unless freemem has a padding of the gigantic page size


# 5349:01422ec04372 28-Oct-2007 kchow

6453272 ctfmerge uses the largest pagesize from getpagesizes() which can be bad on systems with giant pages
6543997 divide by 0 panic in page_geti_contig_pages during 1g page testing
6587615 1g pagesize support on AMD family 0x10 processors
6588824 throttling large page coalescing needs to be revisited for gigantic pages
6613824 legacy applications (java) should be prevented from inadvertantly using 1g pagesize


# 4426:1b5c93035912 08-Jun-2007 aguzovsk

6544121 Implement text replication


# 4325:bf0932baf75c 25-May-2007 qiao

6466149 Long ph_mutex[] holdtimes in page_hashout() mediated by vph_mutex holds in vpn_vplist_dirty


# 3480:fa45699c47ff 24-Jan-2007 jfrank

6436999 Page Retirement feature cannot retire ISM pages with CE errors and induces high sys time.
6508988 page retire kstat pr_pending no longer functional


# 3446:5903aece022d 19-Jan-2007 mrj

PSARC 2006/469 EOF and removal of eeprom -I
PSARC 2006/568 direct boot (dboot) for x86
6219282 interrupt service routine called twice
6223946 potential infinite loop in fbt.c
6228819 tsc_gethrtimeunscaled resets to 0 on suspend
6342201 hat_unload() and cross call usage are needlessly slow on x86, x64
6378723 ctfconvert can't handle GCC empty struct C extension
6379124 ctfconvert can't handle C99 flexible array members
6401187 merging ip's CTF data into genunix introduces a race
6437553 shmat(2) fails on platforms that don't support large pages
6449286 eeprom -I should be torched
6464072 need support for firmware properties
6465816 need a debug record page
6467491 64-bit processes must point %fs or %gs to null selector to utilize [fs|gs]base
6475880 vestiges of old boot code on i386/amd64 should be purged
6475956 Interrupt handling code on x86 platforms should be easier to understand
6477828 pcplusmp psm module should be rearranged to allow multiple platforms to share the same apic code
6477867 x86 KDI should belong in the kernel
6477871 fix for 6232859 also needed for kmdb
6477872 MDB MMU commands need improvements
6477873 cpr debugging can be improved
6477877 mstate accounting should be reset when gethrtimef() is changed
6477915 32-bit x86 kernel should use interrupt gates for all traps
6477963 _update_sregs should be written in C
6477976 no longer need to capture [fs,gs]base in struct regs on every exception
6478642 Solaris needs to support multiple x86 platforms
6478648 ON could use an new & unencumbered xsvc driver
6478734 kernel lint not in C99 mode
6478826 swrand should have framework to lock individual pages
6480763 if interrupts can't be disabled, re-onlining an offline cpu fails
6481824 /dev/fb is not created if installation is done over tty
6483747 clock-tick processing should re-include threads waiting for I/O
6485872 use PTOU macro to access user area instead of old "u" from user.h
6486263 need way to extend kernel core dumps with pages that don't have page_t's
6486435 eeprom, prtdiag should only be platform specific where necessary
6486436 genunix mdb module makefiles duplicate list of files
6486437 intr_common.c should be common
6486445 /dev/mem needs support for page_t-less pages
6486451 cpu_pause() routine should be mach-specific
6486456 SIMULATOR_SUPPORT should die
6486832 x86 platform will need memlist/memseg locking
6486911 amd64 port not finished
6486972 platform TOD setting code may not have real TOD hdw to set
6492647 Better trap trace support for x-calls
6493613 apic_disable_intr() needs round robin fix
6497633 ctfmerge could allow for no ctf sections
6500637 floating point context switching is needlessly slow
6500656 move floating point configuration to C
6500666 need a means to override the setting of uname -m in the kernel build
6500669 update x86 defines
6502790 Files in the boot archive should be individually compressed
6503792 live upgrade environment + bfu + alternate root == incorrect menu.lst file
6504373 kmdb promif shouldn't needlessly busy-wait
6504374 ::interrupts could show level/edge
6506305 opteron erratum 123 should be enabled
6507629 porting x86 platform code to Xen
6510847 kobj.c is unaware of sparc hole between nucleus text and data (from dtrace)


# 3290:256464cbb73c 20-Dec-2006 johansen

4894692 caching data in heap inflates crash dump
6499454 time to increase size of kmem default allocation caches
6499459 vm should stop checking kvp directly


# 3253:c929f34b62c5 14-Dec-2006 mec

PSARC 2006/360 Page retire and caged memory kstats
PSARC 2006/401 /dev/physmem
6385792 physical memory driver needed for memory testing


# 2999:3caa9d682245 27-Oct-2006 stans

6393251 data corruption caused by VMODSORT problem


# 2961:8b33bed4151e 21-Oct-2006 dp78419

PSARC 2006/236 Hashed Cache index support
6409758 Integrate the support for Hashed Cache index mode
6249718 page_freelist_fill() can return a page of the wrong mtype
6478353 page_get_mnode_freelist(szc > 0) should call page_freelist_coalesce() for every bin it tries
6478363 page_get_mnode_cachelist() may ignore PG_MATCH_COLOR flag
6470374 PLCNT_SZ() macro in sun4/vm/vm_dep.h has incorrect loop condition


# 2414:4349245e130c 21-Jul-2006 aguzovsk

4614772 MPSS to be extended to shared memory
6413095 madvise() freeing up whole pages even when passed sizes smaller than a page
6416714 anon_map_getpages panic during stress test
6428347 multilple pagelocks/softlocks on the same large page should decrement availrmem just once


# 2048:8ceabdf91507 24-May-2006 stans

4034947 anon_swap_adjust() should call kmem_reap() if availrmem is low.


# 1841:9d7ebafcda38 19-Apr-2006 praks

6256083 Need a lightweight file page mapping mechanism to substitute segmap
6387639 segkpm segment set to incorrect size for amd64


# 1373:21e3b2045b46 01-Feb-2006 kchow

6351310 assertion failed: (plcnt[mnode][mtype].plc_mt_clpgcnt...) vm_pagelist.c, line: 2239]


# 973:d16e094eee6e 22-Nov-2005 elowe

6272608 gcc and pcfs don't get along
6343344 after an L2 instruction write-back UE error the reboot command panics
6350643 gcc and page_retire.c don't get along
6351973 New page retire gets carried away with UEs


# 917:a3d1e8247fa0 14-Nov-2005 elowe

4908967 toxic page handling on shutdown takes disproportionate amount of time
6330830 Aggressive page retire v3


# 749:d7f9da43aeb7 24-Oct-2005 susans

6310079 contention for pidlock is a major bottleneck for the kenbus benchmark
6337073 per process slew must be applied for coloring large anonymous pages


# 414:c62c3f13a640 24-Aug-2005 kchow

6286816 page_numtopp_nolock is inefficent
6288107 Missing comment terminator in i86pc/ml/locore.s
6290613 high page_get_cachelist failure rate on jurassic
6297005 performance issues with page_nextn()


# 73:d6a2308c356e 24-Jun-2005 mec

6288365 memcntl panics on debug sun4v kernels for 4M pages


# 0:68f95e015346 14-Jun-2005 stevel@tonic-gate

OpenSolaris Launch


# 13076:e0735b974926 11-Aug-2010 Jonathan Adams <Jonathan.Adams@Sun.COM>

6973928 mdb_page_lookup() broken by 6778289; breaks ::pmap


# 13035:0847ab4cffd0 06-Aug-2010 Ondrej Kubecka <Ondrej.Kubecka@Sun.COM>

backout 1225025: needs more work


# 12963:6bd5885dca7c 29-Jul-2010 Stan Studzinski <Stan.Studzinski@Sun.COM>

1225025 mlock:ed anonymous memory remains backed by swap


# 12293:2286b8b6d071 03-May-2010 jmcp <James.McPherson@Sun.COM>

backout 6535949: needs more work


# 12284:3d1135425dbe 29-Apr-2010 Amritpal Sandhu <Paul.Sandhu@Sun.COM>

6535949 availability of 2M pages degrades over time on Solaris/x64


# 12268:1fe5cddd1eb7 27-Apr-2010 Peter Rival <Frank.Rival@oracle.com>

6778289 vm locks need to scale with the size of system (strands/memory size) (fix debug panic)


# 12230:12df01648ca5 23-Apr-2010 Peter Rival <Frank.Rival@oracle.com>

6778289 vm locks need to scale with the size of system (strands/memory size)


# 12156:3c537b2a7425 13-Apr-2010 Stan Studzinski <Stan.Studzinski@Sun.COM>

6675738 KM_NOSLEEP may still try too hard for some allocations


# 12117:79b43747e259 09-Apr-2010 Stan Studzinski <Stan.Studzinski@Sun.COM>

6885904 active segkp pages can not be retired


# 12004:93f274d4a367 25-Mar-2010 Gerry Liu <jiang.liu@intel.com>

PSARC/2009/104 Hot-Plug Support for ACPI-based Systems
PSARC/2009/550 PSMI Extensions for CPU Hotplug
PSARC/2009/551 acpihpd ACPI Hotplug Daemon
PSARC/2009/591 Attachment Points for Hotpluggable x86 Systems
6862510 provide support for cpu hot add on x86
6883891 cmi interface needs to support dynamic reconfiguration
6884154 x2APIC and kmdb may not function properly during CPU hotplug event.
6904971 low priority acpi nexus code review feedback
6877301 lgrp should support memory hotplug flag in SRAT table


# 11185:f0c31008e395 23-Nov-2009 Sean McEnroe <Sean.McEnroe@Sun.COM>

6887924 PP_ISKAS needs to be defined in terms of VN_ISKAS for vnodes
6871169 kphysm_add_memory_dynamic panics testing vmstress + DR
6877160 guest domain panic at platsvc:mdeg_notify_client+10c
6874763 memseg_alloc_meta() incorrectly maps page_t pages
6873569 multiple calls to memlist_read_lock() can cause deadlock hazard
6886354 DR failure with "memory span duplication" error
6886782 panic after pagefault in seg_kpm after LDom add-mem on primary
6887644 domain hang/deadlock during ldom mem DR when trying to grab a write lock


# 10271:7c80b70bb8de 06-Aug-2009 Jason Beloro <Jason.Beloro@Sun.COM>

6858457 Remove Solaris support for UltraSPARC-AT10 processor


# 10106:b235491976d3 16-Jul-2009 Jason Beloro <Jason.Beloro@Sun.COM>

FWARC 2008/540 Memory DR Domain Service
FWARC 2009/300 CPU MD node property for real address bits
6720954 add memory dr feature to ldoms
6742779 fsflush_do_pages() may incorrectly skip constituent file large pages


# 9544:5bcaa92746ac 05-May-2009 Christopher Baumbauer - Sun Microsystems - San Diego United States <Christopher.Baumbauer@Sun.COM>

6587140 page_retire()/page_trycapture should not try to retire non relocatable kernel pages
PSARC 2009/181 Page Retirement Kernel KStat Addition
6819876 pages slated for capture under x86 are being ignored


# 7718:555eee33ff6e 26-Sep-2008 Jason Beloro <Jason.Beloro@Sun.COM>

PSARC/2007/003 Solaris Support for the Rock Processor
6368478 Solaris needs to support the Rock processor
6440653 stores to sun4v error queue head registers are missing a membar #Sync
6639717 FLUSH instruction used more than necessary in sun4v


# 7393:8a8d73648b44 22-Aug-2008 Donghai Qiao <Donghai.Qiao@Sun.COM>

6672470 Looping within uiomove()/xcopyout()/copyout_move()
6707987 kpm has some VAC handling issues
6388567 VAC flushing is broken on US III-IV+ for large pages


# 6880:0a9c7c25ebae 16-Jun-2008 dv142724

6658818 The macro PCF_INDEX is incorrect
6666035 Lock contention for pcf_buckets impacting ISM creation times
6666045 The data structures pcc_info_t and hw_page_map_t need to be cached aligned.


# 5466:8e6f5dfe7459 12-Nov-2007 kchow

6615393 1000x + penalty running with gigantic pages when they are not available.
6623149 trimkcage() may return out of bound pfn range and break large page allocation throttling logic
6625279 new panic on UltraSPARC-IIe during snv_77 (panic[cpu0]/thread=3000cefcc20: BAD TRAP)
6627247 gigantic page processes can hang unless freemem has a padding of the gigantic page size


# 5349:01422ec04372 28-Oct-2007 kchow

6453272 ctfmerge uses the largest pagesize from getpagesizes() which can be bad on systems with giant pages
6543997 divide by 0 panic in page_geti_contig_pages during 1g page testing
6587615 1g pagesize support on AMD family 0x10 processors
6588824 throttling large page coalescing needs to be revisited for gigantic pages
6613824 legacy applications (java) should be prevented from inadvertantly using 1g pagesize


# 4426:1b5c93035912 08-Jun-2007 aguzovsk

6544121 Implement text replication


# 4325:bf0932baf75c 25-May-2007 qiao

6466149 Long ph_mutex[] holdtimes in page_hashout() mediated by vph_mutex holds in vpn_vplist_dirty


# 3480:fa45699c47ff 24-Jan-2007 jfrank

6436999 Page Retirement feature cannot retire ISM pages with CE errors and induces high sys time.
6508988 page retire kstat pr_pending no longer functional


# 3446:5903aece022d 19-Jan-2007 mrj

PSARC 2006/469 EOF and removal of eeprom -I
PSARC 2006/568 direct boot (dboot) for x86
6219282 interrupt service routine called twice
6223946 potential infinite loop in fbt.c
6228819 tsc_gethrtimeunscaled resets to 0 on suspend
6342201 hat_unload() and cross call usage are needlessly slow on x86, x64
6378723 ctfconvert can't handle GCC empty struct C extension
6379124 ctfconvert can't handle C99 flexible array members
6401187 merging ip's CTF data into genunix introduces a race
6437553 shmat(2) fails on platforms that don't support large pages
6449286 eeprom -I should be torched
6464072 need support for firmware properties
6465816 need a debug record page
6467491 64-bit processes must point %fs or %gs to null selector to utilize [fs|gs]base
6475880 vestiges of old boot code on i386/amd64 should be purged
6475956 Interrupt handling code on x86 platforms should be easier to understand
6477828 pcplusmp psm module should be rearranged to allow multiple platforms to share the same apic code
6477867 x86 KDI should belong in the kernel
6477871 fix for 6232859 also needed for kmdb
6477872 MDB MMU commands need improvements
6477873 cpr debugging can be improved
6477877 mstate accounting should be reset when gethrtimef() is changed
6477915 32-bit x86 kernel should use interrupt gates for all traps
6477963 _update_sregs should be written in C
6477976 no longer need to capture [fs,gs]base in struct regs on every exception
6478642 Solaris needs to support multiple x86 platforms
6478648 ON could use an new & unencumbered xsvc driver
6478734 kernel lint not in C99 mode
6478826 swrand should have framework to lock individual pages
6480763 if interrupts can't be disabled, re-onlining an offline cpu fails
6481824 /dev/fb is not created if installation is done over tty
6483747 clock-tick processing should re-include threads waiting for I/O
6485872 use PTOU macro to access user area instead of old "u" from user.h
6486263 need way to extend kernel core dumps with pages that don't have page_t's
6486435 eeprom, prtdiag should only be platform specific where necessary
6486436 genunix mdb module makefiles duplicate list of files
6486437 intr_common.c should be common
6486445 /dev/mem needs support for page_t-less pages
6486451 cpu_pause() routine should be mach-specific
6486456 SIMULATOR_SUPPORT should die
6486832 x86 platform will need memlist/memseg locking
6486911 amd64 port not finished
6486972 platform TOD setting code may not have real TOD hdw to set
6492647 Better trap trace support for x-calls
6493613 apic_disable_intr() needs round robin fix
6497633 ctfmerge could allow for no ctf sections
6500637 floating point context switching is needlessly slow
6500656 move floating point configuration to C
6500666 need a means to override the setting of uname -m in the kernel build
6500669 update x86 defines
6502790 Files in the boot archive should be individually compressed
6503792 live upgrade environment + bfu + alternate root == incorrect menu.lst file
6504373 kmdb promif shouldn't needlessly busy-wait
6504374 ::interrupts could show level/edge
6506305 opteron erratum 123 should be enabled
6507629 porting x86 platform code to Xen
6510847 kobj.c is unaware of sparc hole between nucleus text and data (from dtrace)


# 3290:256464cbb73c 20-Dec-2006 johansen

4894692 caching data in heap inflates crash dump
6499454 time to increase size of kmem default allocation caches
6499459 vm should stop checking kvp directly


# 3253:c929f34b62c5 14-Dec-2006 mec

PSARC 2006/360 Page retire and caged memory kstats
PSARC 2006/401 /dev/physmem
6385792 physical memory driver needed for memory testing


# 2999:3caa9d682245 27-Oct-2006 stans

6393251 data corruption caused by VMODSORT problem


# 2961:8b33bed4151e 21-Oct-2006 dp78419

PSARC 2006/236 Hashed Cache index support
6409758 Integrate the support for Hashed Cache index mode
6249718 page_freelist_fill() can return a page of the wrong mtype
6478353 page_get_mnode_freelist(szc > 0) should call page_freelist_coalesce() for every bin it tries
6478363 page_get_mnode_cachelist() may ignore PG_MATCH_COLOR flag
6470374 PLCNT_SZ() macro in sun4/vm/vm_dep.h has incorrect loop condition


# 2414:4349245e130c 21-Jul-2006 aguzovsk

4614772 MPSS to be extended to shared memory
6413095 madvise() freeing up whole pages even when passed sizes smaller than a page
6416714 anon_map_getpages panic during stress test
6428347 multilple pagelocks/softlocks on the same large page should decrement availrmem just once


# 2048:8ceabdf91507 24-May-2006 stans

4034947 anon_swap_adjust() should call kmem_reap() if availrmem is low.


# 1841:9d7ebafcda38 19-Apr-2006 praks

6256083 Need a lightweight file page mapping mechanism to substitute segmap
6387639 segkpm segment set to incorrect size for amd64


# 1373:21e3b2045b46 01-Feb-2006 kchow

6351310 assertion failed: (plcnt[mnode][mtype].plc_mt_clpgcnt...) vm_pagelist.c, line: 2239]


# 973:d16e094eee6e 22-Nov-2005 elowe

6272608 gcc and pcfs don't get along
6343344 after an L2 instruction write-back UE error the reboot command panics
6350643 gcc and page_retire.c don't get along
6351973 New page retire gets carried away with UEs


# 917:a3d1e8247fa0 14-Nov-2005 elowe

4908967 toxic page handling on shutdown takes disproportionate amount of time
6330830 Aggressive page retire v3


# 749:d7f9da43aeb7 24-Oct-2005 susans

6310079 contention for pidlock is a major bottleneck for the kenbus benchmark
6337073 per process slew must be applied for coloring large anonymous pages


# 414:c62c3f13a640 24-Aug-2005 kchow

6286816 page_numtopp_nolock is inefficent
6288107 Missing comment terminator in i86pc/ml/locore.s
6290613 high page_get_cachelist failure rate on jurassic
6297005 performance issues with page_nextn()


# 73:d6a2308c356e 24-Jun-2005 mec

6288365 memcntl panics on debug sun4v kernels for 4M pages


# 0:68f95e015346 14-Jun-2005 stevel@tonic-gate

OpenSolaris Launch