Searched refs:overhead (Results 1 – 16 of 16) sorted by relevance
| /system/extras/simpleperf/runtest/ |
| D | runtest.py | 58 def __init__(self, name, comm, overhead, children_overhead): argument
61 self.overhead = overhead
73 self.name, self.comm, self.overhead, self.children_overhead))
110 def check_overhead(self, overhead): argument
112 if self.min_overhead > overhead:
115 if self.max_overhead < overhead:
420 overhead = float(items[1][:-1])
423 cur_symbol = Symbol(symbol_name, comm, overhead, children_overhead)
428 overhead = float(items[0][:-1])
431 cur_symbol = Symbol(symbol_name, comm, overhead, 0)
[all …]
|
| /system/memory/libmemunreachable/ |
| D | README.md | 6 …overhead native memory leak detector. It uses an imprecise mark-and-sweep garbage collector pass …
8 …overhead) mode, the returned data on leaks is limited to the address, approximate (upper bound) si…
|
| /system/extras/libatrace_rust/ |
| D | README.md | 116 cheap - on the order of 5-10 ns. There is a 10-20% overhead from the wrapper, mostly caused by
121 The subscriber uses the bindings and adds its own overhead that depends on usage:
|
| /system/core/libsparse/ |
| D | sparse.cpp | 278 int overhead = sizeof(sparse_header_t) + 2 * sizeof(chunk_header_t) + sizeof(uint32_t); in move_chunks_up_to_len() local
279 len -= overhead; in move_chunks_up_to_len()
|
| /system/logging/logd/ |
| D | README.compression.md | 30 * High CPU overhead
|
| /system/extras/libatrace_rust/benchmark/ |
| D | README.md | 22 We use it to check overhead over the base `libatrace_rust`.
|
| /system/chre/platform/shared/idl/ |
| D | host_messages.fbs | 258 /// overhead in serializing and deserializing. The format is as follows:
363 /// overhead in serializing and deserializing. The format is as follows:
|
| /system/extras/ioblame/ |
| D | README | 91 (eg block level prefetching, filesystem metadata overhead etc).
|
| /system/libcppbor/ |
| D | README.md | 136 actually uses the callback-based method, plus has whatever overhead
|
| /system/chre/doc/ |
| D | framework_overview.md | 193 means avoiding language features that can impose runtime overhead, due to the
|
| D | compatibility.md | 109 Since these NSL compatibility behaviors carry some amount of overhead (even if
|
| /system/chre/chpp/ |
| D | RELEASE_NOTES.md | 224 The TX buffer will be slightly larger to accommodate the transport layer encoding overhead.
|
| D | README.md | 11 …ices as necessary. CHPP allows these services to coexist on a single link with very small overhead.
|
| /system/extras/simpleperf/doc/ |
| D | README.md | 129 4. Takes more CPU time than stack frame based call graphs. So it has higher overhead, and can't
|
| D | executable_commands_reference.md | 596 events. As the perf event is cpu-cycles, the overhead is the percentage of CPU cycles used in each
|
| /system/update_engine/ |
| D | README.md | 17 overhead) without interrupting the user. The only downside for users is a
|