2021 Commits

Author SHA1 Message Date
Antonios Printezis
1a9d9b84f8 7087717: G1: make the G1PrintRegionLivenessInfo parameter diagnostic
Reviewed-by: brutisso, ysr
2011-09-09 05:20:58 -04:00
Bengt Rutisson
0be11c915c 6929868: G1: introduce min / max young gen size bounds
Make G1 handle young gen size command line flags more consistently

Reviewed-by: tonyp, jwilhelm
2011-09-08 16:29:41 +02:00
Antonios Printezis
d96de580b6 7084509: G1: fix inconsistencies and mistakes in the young list target length calculations
Fixed inconsistencies and mistakes in the young list target length calculations so that a) the calculated target length is optimal (before, it was not), b) other parameters like max survivor size and max gc locker eden expansion are always consistent with the calculated target length (before, they were not always), and c) the resulting target length was always bound by desired min and max values (before, it was not).

Reviewed-by: brutisso, johnc
2011-09-08 05:16:49 -04:00
Igor Veresov
1940a13d31 7086226: UseNUMA fails on old versions of windows
Return correct answers from os::numa_*() for UMA machines or if NUMA API is not supported

Reviewed-by: johnc
2011-09-07 18:58:33 -07:00
Y. Srinivas Ramakrishna
ed72e31727 4965777: GC changes to support use of discovered field for pending references
If and when the reference handler thread is able to use the discovered field to link reference objects in its pending list, so will GC. In that case, GC will scan through this field once a reference object has been placed on the pending list, but not scan that field before that stage, as the field is used by the concurrent GC thread to link discovered objects. When ReferenceHandleR thread does not use the discovered field for the purpose of linking the elements in the pending list, as would be the case in older JDKs, the JVM will fall back to the old behaviour of using the next field for that purpose.

Reviewed-by: jcoomes, mchung, stefank
2011-09-07 13:55:42 -07:00
Y. Srinivas Ramakrishna
ec13301b32 Merge 2011-09-07 15:00:13 -07:00
Antonios Printezis
24eb07061e 7050392: G1: Introduce flag to generate a log of the G1 ergonomic decisions
It introduces ergonomic decision logging in G1 for the following heuristics: heap sizing, collection set construction, concurrent cycle initiation, and partially-young GC start/end. The code has a bit of refactoring in a few places to make the decision logging possible. It also replaces alternative ad-hoc logging that we have under different parameters and switches (G1_DEBUG, G1PolicyVerbose).

Reviewed-by: johnc, ysr
2011-09-07 12:21:23 -04:00
Igor Veresov
a9cd50a1c8 7087583: Hotspot fails to allocate heap with mmap(MAP_HUGETLB)
Try using small pages when transparent huge pages allocation fails

Reviewed-by: ysr
2011-09-06 21:03:51 -07:00
Stefan Karlsson
5b6ba4e611 7085906: Replace the permgen allocated sentinelRef with a self-looped end
Remove the sentinelRef and let the last Reference in a discovered chain point back to itself.

Reviewed-by: ysr, jmasa
2011-09-01 16:18:17 +02:00
Y. Srinivas Ramakrishna
74b5a420d5 Merge 2011-08-31 23:55:58 -07:00
John Cuthbertson
6c8c4fb321 7066841: remove MacroAssembler::br_on_reg_cond() on sparc
Remove the macro assembler routine br_on_reg_cond() and replace the remaining calls to that routine with an equivalent.

Reviewed-by: kvn, iveresov
2011-08-31 10:16:02 -07:00
Igor Veresov
d8fd276e35 7082645: Hotspot doesn't compile on old linuxes after 7060836
Move syscall ids definitions into os_linux.cpp

Reviewed-by: johnc
2011-08-29 17:42:39 -07:00
John Cuthbertson
1bec3a1bd3 7080389: G1: refactor marking code in evacuation pause copy closures
Refactor code marking code in the evacuation pause copy closures so that an evacuated object is only marked by the thread that successfully copies it.

Reviewed-by: stefank, brutisso, tonyp
2011-08-29 10:13:06 -07:00
Vladimir Kozlov
fa7cc07ca0 Merge 2011-08-25 18:56:52 -07:00
Tom Deneau
06d05b7c47 7082969: NUMA interleaving
Support interleaving on NUMA systems for collectors that don't have NUMA-awareness.

Reviewed-by: iveresov, ysr
2011-08-25 02:57:46 -07:00
Bengt Rutisson
0c6ad2c298 7082220: Visual Studio projects broken after change 7016797: Hotspot: securely/restrictive load dlls and new
Add the psapi.lib library to Visual Studio projects

Reviewed-by: jwilhelm, poonam, kamg
2011-08-23 11:06:30 +02:00
Y. Srinivas Ramakrishna
4df87be52f Merge 2011-08-22 23:57:10 -07:00
Y. Srinivas Ramakrishna
3af2ea16d6 6810861: G1: support -XX:+{PrintClassHistogram,HeapDump}{Before,After}FullGC
Call {pre,post}_full_gc_dump() before and after a STW full gc of G1CollectedHeap. Also adjusted the prefix message, including the addition of missing whitespace.

Reviewed-by: brutisso, tonyp
2011-08-22 12:30:06 -07:00
Vladimir Kozlov
85b4875e2e 7081926: assert(VM_Version::supports_sse2()) failed: must support
Fix assert, prefetchnta is supported since SSE not SSE2.

Reviewed-by: never
2011-08-22 11:00:39 -07:00
Antonios Printezis
2cbacf440a 7081064: G1: remove develop params G1FixedSurvivorSpaceSize, G1FixedTenuringThreshold, and G1FixedEdenSize
Remove three develop parameters we don't use.

Reviewed-by: brutisso, jwilhelm
2011-08-22 10:16:33 -04:00
John Coomes
af31d0cf27 Merge 2011-08-19 14:08:18 -07:00
Vladimir Kozlov
2cfb876ce7 Merge 2011-08-19 08:55:53 -07:00
Bengt Rutisson
6fdb1414fa 6814390: G1: remove the concept of non-generational G1
Removed the possibility to turn off generational mode for G1.

Reviewed-by: johnc, ysr, tonyp
2011-08-19 09:30:59 +02:00
Vladimir Kozlov
ff4faaab4b 7080431: VM asserts if specified size(x) in .ad is larger than emitted size
Move code from finalize_offsets_and_shorten() to fill_buffer() to restore previous behavior.

Reviewed-by: never
2011-08-18 11:49:33 -07:00
Christian Thalinger
db338313bf 7079769: JSR 292: incorrect size() for CallStaticJavaHandle on sparc
Reviewed-by: never, kvn
2011-08-17 11:52:22 -07:00
Christian Thalinger
fc31a8e0b9 7079626: x64 emits unnecessary REX prefix
Reviewed-by: kvn, iveresov, never
2011-08-17 05:14:43 -07:00
Gary Collins
d164c986b5 Merge 2011-08-17 07:05:42 -04:00
Vladimir Kozlov
90651b2666 7079329: Adjust allocation prefetching for T4
On T4 2 BIS instructions should be issued to prefetch 64 bytes

Reviewed-by: iveresov, phh, twisti
2011-08-16 16:59:46 -07:00
Vladimir Kozlov
080f790edc 7079317: Incorrect branch's destination block in PrintoOptoAssembly output
Save/restore label and block in scratch_emit_size()

Reviewed-by: never
2011-08-16 11:53:57 -07:00
Zhengyu Gu
1b7f659a43 Merge 2011-08-16 09:18:47 -04:00
Christian Thalinger
134c40b4db 7071653: JSR 292: call site change notification should be pushed not pulled
Reviewed-by: kvn, never, bdelsart
2011-08-16 04:14:05 -07:00
Antonios Printezis
f482091133 7039627: G1: avoid BOT updates for survivor allocations and dirty survivor regions incrementally
Refactor the allocation code during GC to use the G1AllocRegion abstraction. Use separate subclasses of G1AllocRegion for survivor and old regions. Avoid BOT updates and dirty survivor cards incrementally for the former.

Reviewed-by: brutisso, johnc, ysr
2011-08-12 11:31:06 -04:00
Vladimir Kozlov
ac99f413d7 7063629: use cbcond in C2 generated code on T4
Use new short branch instruction in C2 generated code.

Reviewed-by: never
2011-08-11 12:08:11 -07:00
John Cuthbertson
0da9dc64d2 7074579: G1: JVM crash with JDK7 running ATG CRMDemo Fusion App
Handlize MemoryUsage klass oop in createGCInfo routine

Reviewed-by: tonyp, fparain, ysr, jcoomes
2011-08-11 11:36:29 -07:00
Kevin Walls
e6603b37f2 Merge 2011-08-11 14:58:29 +01:00
Vladimir Kozlov
52f678435a 7077439: Possible reference through NULL in loopPredicate.cpp:726
Use cl->is_valid_counted_loop() check.

Reviewed-by: never
2011-08-10 14:06:57 -07:00
Andrew Haley
aefb3a004f 7073913: The fix for 7017193 causes segfaults
Buffer overflow in os::get_line_chars

Reviewed-by: coleenp, dholmes, dcubed
2011-08-10 15:04:21 +04:00
Andrew John Hughes
c4b791d93a 7060836: RHEL 5.5 and 5.6 should support UseNUMA
Add a wrapper for sched_getcpu() for systems where libc lacks it

Reviewed-by: ysr
2011-08-05 16:50:14 -07:00
Igor Veresov
427c404400 7060842: UseNUMA crash with UseHugreTLBFS running SPECjvm2008
Use mmap() instead of madvise(MADV_DONTNEED) to uncommit pages

Reviewed-by: ysr
2011-08-05 16:44:01 -07:00
John Cuthbertson
f7d3d2b63e 7069863: G1: SIGSEGV running SPECjbb2011 and -UseBiasedLocking
Align the reserved size of the heap and perm to the heap region size to get a preferred heap base that is aligned to the region size, and call the correct heap reservation constructor. Also add a check in the heap reservation code that the reserved space starts at the requested address (if any).

Reviewed-by: kvn, ysr
2011-08-02 12:13:13 -07:00
Roland Westrelin
b543a07b9a 7074017: Introduce MemBarAcquireLock/MemBarReleaseLock nodes for monitor enter/exit code paths
Replace MemBarAcquire/MemBarRelease nodes on the monitor enter/exit code paths with new MemBarAcquireLock/MemBarReleaseLock nodes

Reviewed-by: kvn, twisti
2011-08-02 18:36:40 +02:00
Tom Rodriguez
9cd1a00e3f 7075623: 6990212 broke raiseException in 64 bit
Reviewed-by: kvn, twisti
2011-08-08 13:19:46 -07:00
Chris Phillips
6a54c8a55f 7071823: Zero: zero/shark doesn't build after b147-fcs
Reviewed-by: gbenson, twisti
2011-08-08 05:49:04 -07:00
Krystal Mok
24e8905fdf 7072527: CMS: JMM GC counters overcount in some cases
Avoid overcounting when CMS has concurrent mode failure.

Reviewed-by: ysr
2011-08-02 14:37:35 +01:00
John Cuthbertson
144fb9dc25 7068240: G1: Long "parallel other time" and "ext root scanning" when running specific benchmark
In root processing, move the scanning of the reference processor's discovered lists to before RSet updating and scanning. When scanning the reference processor's discovered lists, use a buffering closure so that the time spent copying any reference object is correctly attributed. Also removed a couple of unused and irrelevant timers.

Reviewed-by: ysr, jmasa
2011-08-01 10:04:28 -07:00
Vladimir Kozlov
df8d608ac7 7068051: SIGSEGV in PhaseIdealLoop::build_loop_late_post
Removed predicate cloning from loop peeling optimization and from split fall-in paths.

Reviewed-by: never
2011-07-29 09:16:29 -07:00
Tom Rodriguez
e30a679033 7060619: C1 should respect inline and dontinline directives from CompilerOracle
Reviewed-by: kvn, iveresov
2011-07-28 13:03:39 -07:00
Xerxes Ranby
23b10ec618 7066143: JSR 292: Zero support after regressions from 7009923 and 7009309
Reviewed-by: jrose, twisti
2011-07-28 02:14:44 -07:00
Vladimir Kozlov
18329266ea 7069452: Cleanup NodeFlags
Remove flags which duplicate information in Node::NodeClasses.

Reviewed-by: never
2011-07-27 17:28:36 -07:00
Tom Rodriguez
c895368552 7071427: AdapterFingerPrint can hold 8 entries per int
Reviewed-by: kvn
2011-07-27 15:06:35 -07:00