aboutsummaryrefslogtreecommitdiffstats
path: root/erts
AgeCommit message (Collapse)Author
2012-12-14erts: Do not cache segments that are misalignedBjörn-Egil Dahlberg
* SBC may realloc carriers to misaligned addresses which is perfectly fine. However, those segments may not be cached because MBC allocations might find them and MBC's *needs* correct alignment.
2012-12-14erts: Add mseg cache for large sbc segmentsBjörn-Egil Dahlberg
* Not a power of two (unpowered) segements
2012-12-14erts: Reintroduce mseg options amcbf and rmcbfBjörn-Egil Dahlberg
Used with new sbc cache
2012-12-14erts: Optimize erl_alloc_util.c by substitute MBC_BLK_SZSverker Eriksson
with either MBC_ABLK_SZ or MBC_FBLK_SZ in all cases when we already know what kind of block it is.
2012-12-14erts: Fix bug when allocating size near sbc_thresholdSverker Eriksson
A block larger than sbc_threshold can be allocated in MBC if the subsequent "residue block" is smaller than min_block_size. Solved by lowering sbc_threshold to ("hard limit" - min_block_size).
2012-12-14erts: Make gc sizes fit into MB Carrier blocksBjörn-Egil Dahlberg
* Account for block header size in gc-sizes * Also slow down growth to 20% instead of 25% when size threshold is reached
2012-12-14erts: Force sbmbc to be disabled in a crude waySverker Eriksson
2012-12-14erts: Fix new header scheme for win64Sverker Eriksson
2012-12-14erts: Fix mseg cache. Malplaced NULL pointerBjörn-Egil Dahlberg
2012-12-14erts: Remove unused mseg options amcbf and rmcbfBjörn-Egil Dahlberg
2012-12-14erts: Use aligned bits as constant in mseg_allocBjörn-Egil Dahlberg
HAVE_SUPER_ALIGNED_MB_CARRIERS is always true with mmap and thus aligned bits is a constant and so is "page" size for mmap. Conflicts: erts/emulator/sys/common/erl_mseg.h
2012-11-21erts: Don't let zero be considered a power of twoBjörn-Egil Dahlberg
2012-11-21erts: Use MSEG_ALIGN_BITS and MSEG_ALIGNED_SIZEBjörn-Egil Dahlberg
* Don't redfine ALIGN_BITS or ALIGNED_SIZE, the global defined MSEG_* constants are just as good.
2012-11-21erts: Make carrier header sizes compile time constantsSverker Eriksson
No allocator strategy is using customized carrier headers anyway.
2012-11-19erts: Cleanup minor things in alloc_utilSverker Eriksson
2012-11-16erts: Fix compile warnings in alloc_utilSverker Eriksson
2012-11-15erts: Enable new header scheme without super alignment on 64-bitSverker Eriksson
2012-11-15erts: Fix alloc_SUITE:coalesce for new header schemeSverker Eriksson
2012-11-15erts: Ensure MBC limits due to super alignmentSverker Eriksson
2012-11-15erts: Add carrier pointer to header of free blockSverker Eriksson
2012-11-15erts: Refactor renaming a couple of macros in alloc_utilSverker Eriksson
To get a consistent naming scheme where FBLK = Free block ABLK = Allocated block
2012-11-15erts: Refactor removing dummy footer in carrier headerSverker Eriksson
2012-11-15erts: Save one word per block for thread preferred allocatorsSverker Eriksson
by making use of the new block header scheme to find the carrier header and thereby the allocator.
2012-11-15erts: Refactor erl_alloc_util.cSverker Eriksson
Make use of type FreeBlkFtr_t Rename GET_MB_CARRIER into BLK2MBC (like BLK2SBC) Remove some unused macros
2012-11-15erts: Change single carrier block header flagsSverker Eriksson
to allow realloc to determine block size (in MBC or SBC) without having to read the footer of the previous block that might be written to by concurrent thread.
2012-11-15erts: Make alloc_SUITE:bucket_mask handle larger carriersSverker Eriksson
2012-11-15erts: Use new mseg flag argument for carrier sizingSverker Eriksson
2012-11-15erts: Move carrier alignment define to erl_msg.hSverker Eriksson
2012-11-15erts: Add carrier offset to internal allocation headersSverker Eriksson
2012-11-13erts: Fix compiler warnings for halfword vmSverker Eriksson
2012-11-13erts: New mseg allocator cacheBjörn-Egil Dahlberg
* utilize the power of two
2012-11-13erts: Teach mseg alloc the value of halfword beamBjörn-Egil Dahlberg
2012-11-13erts: Let mseg allocate larger alignmentsBjörn-Egil Dahlberg
2012-10-31erts: Remove faked MSEG_ALLOCBjörn-Egil Dahlberg
* Not used except in valgrind but there mseg is disabled completely via Meamin.
2012-10-31Merge branch 'raimo/IPV6_V6ONLY/OTP-8928'Raimo Niskanen
* raimo/IPV6_V6ONLY/OTP-8928: kernel: Document socket option ipv6_v6only kernel: Add test cases for socket option ipv6_v6only erts,kernel: Implement socket option ipv6_v6only in erlang code erts: Implement socket option IPV6_V6ONLY erts: Add configure test for IPV6_V6ONLY
2012-10-31erts,kernel: Implement socket option ipv6_v6only in erlang codeRaimo Niskanen
2012-10-31erts: Implement socket option IPV6_V6ONLYRaimo Niskanen
2012-10-26Merge branch 'bjorn/erts/beam_makeops'Björn Gustavsson
* bjorn/erts/beam_makeops: beam_makeops: Turn on warnings and eliminate existing warnings beam_makeops: Eliminate a deprecation warning
2012-10-24Merge branch 'maint'Björn-Egil Dahlberg
Conflicts: erts/etc/common/heart.c
2012-10-24Merge branch 'egil/ensure-erl_crash.dump/OTP-10422' into maintBjörn-Egil Dahlberg
* egil/ensure-erl_crash.dump/OTP-10422: test: Relax timeouts for heart_SUITE erts: Fix crash dump write to port hack erts: Fix lock check assertion doc: Document ERL_CRASH_DUMP_SECONDS behaviour test: Add test for heart restart on crash test: Add test for heart restart on crash erts: Change ERL_CRASH_DUMP_SECONDS behaviour test: Refactor away ?line macros in heart_SUITE erts: Search for heart in ports that are alive heart: Refactor heart debugging erts, heart: Ensure erl_crash.dump is written
2012-10-23Merge branch 'sverk/valgrind-quoting'Sverker Eriksson
* sverk/valgrind-quoting: Make cerl -valgrind work with quoted spaces in command line
2012-10-23erts: Fix crash dump write to port hackBjörn-Egil Dahlberg
More future proof with R16
2012-10-23erts: Fix lock check assertionBjörn-Egil Dahlberg
2012-10-22beam_makeops: Turn on warnings and eliminate existing warningsBjörn Gustavsson
2012-10-18doc: Document ERL_CRASH_DUMP_SECONDS behaviourBjörn-Egil Dahlberg
* heart reboot behaviour * erl_crash.dump file write behaviour
2012-10-18Merge branch 'jf/fix_sctp_peeloff_active_true'Henrik Nord
* jf/fix_sctp_peeloff_active_true: Set new peeled off SCTP socket to nonblocking socket SCTP test case with socket active options once and true OTP-10491
2012-10-16erts: Change ERL_CRASH_DUMP_SECONDS behaviourBjörn-Egil Dahlberg
Not setting ERL_CRASH_DUMP_SECONDS will now terminate beam immediately on a crash without writing a crash dump file. Setting ERL_CRASH_DUMP_SECONDS to 0 will also terminate beam immediately on a crash without writing a crash dump file, i.e. same as not setting ERL_CRASH_DUMP_SECONDS environment variable. Setting ERL_CRASH_DUMP_SECONDS to a negative value will let the beam wait indefinitely on the crash dump file being written. Setting ERL_CRASH_DUMP_SECONDS to a positive value will let the beam wait that many seconds on the crash dump file being written. A positive value will set both an alarm in beam AND a heart timeout for restart if heart is running. This is due to the change of 'heart' behavior when 'heart' is listening for a crash.
2012-10-15erts: Search for heart in ports that are aliveBjörn-Egil Dahlberg
2012-10-15heart: Refactor heart debuggingBjörn-Egil Dahlberg
2012-10-15erts, heart: Ensure erl_crash.dump is writtenBjörn-Egil Dahlberg
When a crash dump is about to be written and we have heartbeat enabled on a system. We need time to write it before heart explicitly kills the beam.