一键释放iOS 64位App潜力

背景
远在iOS 11时期(2017年),苹果就发公告要求所有需要上架的应用都必须支持64位 。32位应用不再支持上架与运行 。
升级64位应用有什么好处呢?(以下内容纯摘抄 , 客官可以直接跳过)
寄存器更多,减少内存读写,加快执行速度
这里我们要注意的是:虚拟内存确实比纯32位多了,但是App到底能用多少,是否跟宣传一样接近16EB?下面将会展开聊聊,我们先来看一个Crash 。
一个长期存在的幽灵
我们先来看下面的一个内存导致的崩溃,JSC在使用尝试进行内存分配时,提示OOM导致了 。
Last Exception :0JavaScriptCore0x000000018b777570 _pas_panic_on_out_of_memory_error1JavaScriptCore0x000000018b72e918 _bmalloc_try_iso_allocate_impl_impl_slow2JavaScriptCore0x000000018b73d3d8 _bmalloc_heap_config_specialized_local_allocator_try_allocate_small_segregated_slow +59523JavaScriptCore0x000000018b7276f8 _bmalloc_allocate_impl_casual_case +8004JavaScriptCore0x000000018c60d494 JSC::PropertyTable::create(JSC::VM&, unsigned int) +2445JavaScriptCore0x000000018c66ba74 JSC::Structure::materializePropertyTable(JSC::VM&, bool) +3246JavaScriptCore0x000000018c66dfac JSC::Structure::changePrototypeTransition(JSC::VM&, JSC::Structure*, JSC::JSValue, JSC::DeferredStructureTransitionWatchpointFire&) +6127JavaScriptCore0x000000018c559930 JSC::JSObject::setPrototypeDirect(JSC::VM&, JSC::JSValue) +1928JavaScriptCore0x000000018c559e40 JSC::JSObject::setPrototypeWithCycleCheck(JSC::VM&, JSC::JSGlobalObject*, JSC::JSValue, bool) +3169JavaScriptCore0x000000018c4f580c JSC::globalFuncProtoSetter(JSC::JSGlobalObject*, JSC::CallFrame*) +19210 JavaScriptCore0x000000018ba1f7a8 _vmEntryToNative +28011 JavaScriptCore0x000000018c1b0cd0 JSC::Interpreter::executeCall(JSC::JSGlobalObject*, JSC::JSObject*, JSC::CallData const&, JSC::JSValue, JSC::ArgList const&) +61612 JavaScriptCore0x000000018c474ecc JSC::GetterSetter::callSetter(JSC::JSGlobalObject*, JSC::JSValue, JSC::JSValue, bool) +21213 JavaScriptCore0x000000018c5b6264 JSC::JSGenericTypedArrayView::put(JSC::JSCell*, JSC::JSGlobalObject*, JSC::PropertyName, JSC::JSValue, JSC::PutPropertySlot&) +61214 JavaScriptCore0x000000018c2c2ecc _llint_slow_path_put_by_id +3244// 忽略多余重复堆栈37 JavaScriptCore0x000000018ba1f5fc _vmEntryToJavaScript +26438 JavaScriptCore0x000000018c1b0c7c JSC::Interpreter::executeCall(JSC::JSGlobalObject*, JSC::JSObject*, JSC::CallData const&, JSC::JSValue, JSC::ArgList const&) +53239 JavaScriptCore0x000000018bac7ae4 _JSObjectCallAsFunction +56840 mttlite0x0000000102a54914 hippy::napi::JSCCtx::CallFunction(std::__1::shared_ptr const&, unsigned long, std::__1::shared_ptr const*) (js_native_api_value_jsc.cc:406)41 mttlite0x0000000102a664e0 _ZNSt3__110__function6__funcIZN11TimerModule5StartERKN5hippy4napi12CallbackInfoEbE3$_4NS_9allocatorIS8_EEFvvEEclEv (memory:3237)42 mttlite0x0000000102a63018 hippy::base::TaskRunner::Run() (memory:3237)43 mttlite0x0000000102a64974 ThreadEntry (thread.cc:0)44 libsystem_pthread.dylib0x00000001dc129348 __pthread_start +116------Exception Type: SIGTRAP Exception Codes: fault addr: 0x000000018b777570Crashed Thread: 48 hippy.js这个OOM问题 , 与iOS上常见的OOM不一样 。按照常规的理解 , 当App内存不足的时候,正常会触发系统的机制杀死App 。在系统日志中会留下相关日志,理论上不会在Bugly等异常上报中发现 。但这一类崩溃却一直在产生上报软件运行出错误代码1,并且低内存的崩溃堆栈表现形式有很多种 。
以上的JSC崩溃问题已经存在很长一段时间了(至少2年),而且崩溃堆栈都集中在JSC执行JS代码的过程中,长期缺乏JS相关的监控与Debug工具导致该问题一直无法解决 。
虽然堆栈上有明确的原因说明是OOM,但我们观察到有不少用户实际上物理内存空间还是足够的:
两年前,冲浪的时候偶然看来了来自微视同学的Case总结:《OOM与内存》
当时跟hippy SDK的同事也讨论过是否存在类似的内存不足情况 。但由于大家对JSC黑盒都不熟悉,而且崩溃的JS堆栈也不确切 。当时的建议是:少在后台加载JSC 。最终也并没有解决该问题 。
两年后,当浏览器集成,类似的JS崩溃直接翻倍(21H2 0.08% -> 22H1 0.16%) 。没办法,还是要看类似JSC和Dart VM的内存分配机制是怎样的,再挖掘一下是否存在解(缓)决(解)方案 。
JSC、的虚拟内存分配
翻阅相关虚拟机的内存管理相关代码 , 可以找到底层的内存分配基本实现都是基于mmap处理的 。
// WebKit bmalloc VMAllocateinline void* tryVMAllocate(size_t vmSize, VMTag usage = VMTag::Malloc){vmValidate(vmSize);void* result = mmap(0, vmSize, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANON | BMALLOC_NORESERVE, static_cast(usage), 0);if (result == MAP_FAILED)return nullptr;return result;}// Dart VM的虚拟内存VirtualMemory* VirtualMemory::Allocate(intptr_t size,bool is_executable,const char* name) {ASSERT(Utils::IsAligned(size, PageSize()));const int prot = PROT_READ | PROT_WRITE | (is_executable ? PROT_EXEC : 0);int map_flags = MAP_PRIVATE | MAP_ANONYMOUS;#if (defined(DART_HOST_OS_MACOS) && !defined(DART_HOST_OS_IOS))if (is_executable && IsAtLeastOS10_14()) {map_flags |= MAP_JIT;}#endif// defined(DART_HOST_OS_MACOS)// Some 64-bit microarchitectures store only the low 32-bits of targets as// part of indirect branch prediction, predicting that the target's upper bits// will be same as the call instruction's address. This leads to misprediction// for indirect calls crossing a 4GB boundary. We ask mmap to place our// generated code near the VM binary to avoid this.void* hint = is_executable ? reinterpret_cast(&Allocate) : nullptr;void* address = mmap(hint, size, prot, map_flags, -1, 0);if (address == MAP_FAILED) {return nullptr;}return new VirtualMemory(address, size);}VirtualMemory::~VirtualMemory() {if (address_ != nullptr) {if (munmap(address_, size_) != 0) {int error = errno;const int kBufferSize = 1024;char error_buf[kBufferSize];FATAL("munmap error: %d (%s)", error,Utils::StrError(error, error_buf, kBufferSize));}}}当包含时 , 并且fd传入-1时 , mmap将直接使用虚拟内存进行分配,不需要依赖文件描述符 。
mmap在xnu上的实现
/* * mmap stub, with preemptory failures due to extra parameter checking * mandated for conformance. * * This is for UNIX03 only. */void *mmap(void *addr, size_t len, int prot, int flags, int fildes, off_t off){/** Preemptory failures:** ooff is not a multiple of the page size* oflags does not contain either MAP_PRIVATE or MAP_SHARED* olen is zero*/extern void cerror_nocancel(int);if ((off & PAGE_MASK) ||(((flags & MAP_PRIVATE) != MAP_PRIVATE) &&((flags & MAP_SHARED) != MAP_SHARED)) ||(len == 0)) {cerror_nocancel(EINVAL);return(MAP_FAILED);}void *ptr = __mmap(addr, len, prot, flags, fildes, off);if (__syscall_logger) {int stackLoggingFlags = stack_logging_type_vm_allocate;if (flags & MAP_ANON) {stackLoggingFlags |= (fildes & VM_FLAGS_ALIAS_MASK);} else {stackLoggingFlags |= stack_logging_type_mapped_file_or_shared_mem;}__syscall_logger(stackLoggingFlags, (uintptr_t)mach_task_self(), (uintptr_t)len, 0, (uintptr_t)ptr, 0);}return ptr;}上面的调用会传递到内核.c的实现函数mmap( p, *uap, *)
/* * XXX Internally, we use VM_PROT_* somewhat interchangeably, but the correct * XXX usage is PROT_* from an interface perspective.Thus the values of * XXX VM_PROT_* and PROT_* need to correspond. */intmmap(proc_t p, struct mmap_args *uap, user_addr_t *retval){/** 上面忽略了一部分代码*/result = vm_map_enter_mem_object(user_map,&user_addr, user_size,0, alloc_flags, vmk_flags,tag,IPC_PORT_NULL, 0, FALSE,prot, maxprot,(flags & MAP_SHARED) ?VM_INHERIT_SHARE :VM_INHERIT_DEFAULT);/* If a non-binding address was specified for this anonymous* mapping, retry the mapping with a zero base* in the event the mapping operation failed due to* lack of space between the address and the map's maximum.*/if ((result == KERN_NO_SPACE) && ((flags & MAP_FIXED) == 0) && user_addr && (num_retries++ == 0)) {user_addr = vm_map_page_size(user_map);goto map_anon_retry;}/** 下面忽略了一部分代码*/}其中又会调用.c内部的ect,而该方法最终会在中依据对象进行内存分配:
// 下面这个只截了个头 , 大概带一下,我也没调过代码~/* *Routine:vm_map_enter * *Description: *Allocate a range in the specified virtual address map. *The resulting range will refer to memory defined by *the given memory object and offset into that object. * *Arguments are as defined in the vm_map call. */kern_return_tvm_map_enter(vm_map_tmap,vm_map_offset_t*address,/* IN/OUT */vm_map_size_tsize,vm_map_offset_tmask,intflags,vm_map_kernel_flags_tvmk_flags,vm_tag_talias,vm_object_tobject,vm_object_offset_toffset,boolean_tneeds_copy,vm_prot_tcur_protection,vm_prot_tmax_protection,vm_inherit_tinheritance)其中在分配过程中会对→作判断,即最大的可分配空间 。
xnu上虚拟内存的分配范围
本来我只是观察到苹果在iOS15上增加了com.apple…–limit的能力声明 。本着死马当活马医的想法,尝试在新版本上添加该声明以缓解一部分问题 。
结果偶然看到部分开发者提问:该能力可配合com.apple…–使用 。看到后我一下子反应过来 , 顺手搜到了今年二月国外有大佬做了相关的探索:

一键释放iOS 64位App潜力

文章插图
Size : An of on iOS
文章阐述了iOS的内存管理机制和虚拟内存空间分配在不同的机型上存在上限 , 代码如下:
#define ARM64_MIN_MAX_ADDRESS (SHARED_REGION_BASE_ARM64 + SHARED_REGION_SIZE_ARM64 + 0x20000000) // end of shared region + 512MB for various purposesconst vm_map_offset_t min_max_offset = ARM64_MIN_MAX_ADDRESS; // end of shared region + 512MB for various purposesif (arm64_pmap_max_offset_default) {max_offset_ret = arm64_pmap_max_offset_default;} else if (max_mem > 0xC0000000) {max_offset_ret = min_max_offset + 0x138000000; // Max offset is 13.375GB for devices with > 3GB of memory} else if (max_mem > 0x40000000) {max_offset_ret = min_max_offset + 0x38000000;// Max offset is 9.375GB for devices with > 1GB and <= 3GB of memory} else {max_offset_ret = min_max_offset;}并且总结了一个上限值与机型表格:
RAM
Space
> 3 GiB
15.375 GiB
7.375 GiB
– XS – 13
– iPad Air (4th )
– iPad Pro (12.9-inch), (10.5-inch), (11-inch)
> 1 GiB
11.375 GiB
3.375 GiB
– 6s – X, SE, XR
– iPad (5th ) – iPad (8th )
– iPad Air 2, iPad Air (3rd )
– iPad mini 4, iPad mini (5th )
– iPad Pro (9.7-inch)
(64位系统下1008字节,32位系统下496)
内存扩展前失败阈值约 * 1009 = 6.63 GB
内存扩展后失败阈值约 * 1009 = 53.33 GB
当然,在xnu的单元测试代码中,也可找到jumbo mode相关的测试代码 , 与上面的测试结果完全一致,即最多可分配53GB的空间 。
#define GB (1ULL * 1024 * 1024 * 1024)/* * This test expects the entitlement to be the enabling factor for a process to * allocate at least this many GB of VA space. i.e. with the entitlement, n GB * must be allocatable; whereas without it, it must be less. * This value was determined experimentally to fit on applicable devices and to * be clearly distinguishable from the default VA limit. */#define ALLOC_TEST_GB 53T_DECL(TESTNAME,"Verify that a required entitlement is present in order to be granted an extra-large ""VA space on arm64",T_META_NAMESPACE("xnu.vm"),T_META_CHECK_LEAKS(false)){int i;void*res;if (!dt_64_bit_kernel()) {T_SKIP("This test is only applicable to arm64");}T_LOG("Attemping to allocate VA space in 1 GB chunks.");for (i = 0; i < (ALLOC_TEST_GB * 2); i++) {res = mmap(NULL, 1 * GB, PROT_NONE, MAP_PRIVATE | MAP_ANON, 0, 0);if (res == MAP_FAILED) {if (errno != ENOMEM) {T_WITH_ERRNO;T_LOG("mmap failed: stopped at %d of %d GB allocated", i, ALLOC_TEST_GB);}break;} else {T_LOG("%d: %pn", i, res);}}#if defined(ENTITLED)T_EXPECT_GE_INT(i, ALLOC_TEST_GB, "Allocate at least %d GB of VA space", ALLOC_TEST_GB);#elseT_EXPECT_LT_INT(i, ALLOC_TEST_GB, "Not permitted to allocate %d GB of VA space", ALLOC_TEST_GB);#endif}可见,当开启com.apple…–时,内核的可分配空间确实有明显提升 。
上线效果与结论
从QQ浏览器的上线效果来看,JS相关的内存分配Crash在14.0以上系统几乎全部消失 。上线第一天App崩溃率环比下降接近50%,效果显著 。
简单总结:
苹果很少在公开文档中说明64位App在虚拟内存使用上存在限制 。而且很多App也并没有像浏览器内一样,为业务灵活性而选择将hippy、等技术进行大规模的组合使用,所以可能很多App其实并不会遇到虚拟内存不足的情况 。上线效果也说明浏览器在混合开发的场景下,内存优化仍然存在很大的空间 。因为 仅能缓解虚拟内存不足的情况,并不意味着App的物理内存也得到增加,对FOOM的治理仍然需要持续 。鉴于司内有不少的著名组件都会使用mmap机制进行内存管理软件运行出错误代码1,建议在使用相关组件时 , 控制好mmap的大小 。如果有需要在 12 Pro、M1 iPad、M1上运行应用,并希望解放更多的物理内存,建议增加com.apple…–limit的能力声明,实测在 13 Pro下可以增加1GB的可用物理内存 。和类似框架在项目中使用较多的,建议需要考虑多个的复用,减少创建重复内容,司内外都有实践证明该措施十分有效 。对于一类的内存优化 , 可翻阅的相关代码 。vm在创建时允许外部传参控制vm行为,包括:old heap size、leak vm等 。合适的参数可比较有效控制内存占用 。
以上源码相关的内容仅个人阅读理解,如有错误请指出 。
【一键释放iOS 64位App潜力】本文到此结束,希望对大家有所帮助 。