Lines Matching full:can

68 Mark word depends on configuration and can have different sizes and layout. So here all possible co…
107 However, we can also support such version of the object header(Hash is stored just after the object…
228 …pport for pool extension and reduction(i.e., we can add another memory chunk to the allocator, and…
240 It is a region-based allocator, i.e., all allocated in region/arena objects can be efficiently deal…
249 So, if we can hold JIT memory usage at some reasonable level - Arena Allocator ideally fits JIT req…
256 This allocator can be tuned to provide more performance.
257 For example, if we have some callgraph info, we can use it and allocate code for connected methods …
288 Humongous objects can be allocated just by proxying requests to the OS(but keep reference to it som…
301 We can use profile information about allocation size for improving main allocator metrics.
302 If we see a very popular allocation size in profile, we can add it as an explicit segregated size a…
309 There is no universal approach in this paper, but we can try to mix approach from this paper 
314 All used memory is divided in chunks. Main allocator can extend his pool with these chunks.
316 For the cases when we can get memory shortage we should have some preallocated buffer which allow R…
347 Garbage collector(GC) automatically recycles memory that it can prove will never be used again.
366 …sion that Runtime has GC. I.e., it supports all required GC interfaces and can be integrated into …
368 Epsilon GC should be used only for debug and profiling purposes. I.e., we can disable GC and measur…
404 So we can support it optionally for experiments. As alternative we can introduce some average age m…
412 Note #2: we can tune minor GC by trying make concurrent marking and re-mark, but it will require co…
424 Note: If we don't have Survivor spaces we can implement non-moving generational GC. 
434 Since typical heap size for mobile applications is small - this GC can be considered as good choice…
436 …ll heap consists of memory regions with fixed size(it can vary, i.e. size of memory region #K+1 ca…
468 Note: RSets optionally can be refined with special threads
488 Note: compaction is target and mode dependent, so for low-memory devices we can consider [semi-spac…
489 For straight-forward approach we can consider some support from OS to minimize overlapping of semi-…
500 For example, it can be something like this: `LDR R12, [RVState, #SPaddrOffset]`
516 For each safepoint, we should have a method that can provide GC with information about objects on t…
518 But for JIT/compiler, it looks like we need some generated(by JIT/compiler) method that can get all…
519 This method can actually be just some code without prologue and epilogue.
544 Heap inconsistency can happen when GC reclaim alive/reachable object.
549 Besides addressing of heap inconsistency problem, write barrier can be used for maintaining incomin…
551 So we can solve these issues with GC WRB(write barrier). GC WRB can be _pre_(inserted before the st…
564 _Post_ barrier can be used to solve issue with tracking references from tenured generation to the y…
578 At some moment we can have two instance of the one object.
580 1. All writes happen into `to-space` instance of the object, but reads can happen from both `from-s…
591 Note: for performance, we can put into ExecState address of conditional flag for conditional barrie…
595 …1. To describe barrier use some meta-language or IR which can be interpreted/encoded by all compil…