<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body>
<p>Hi, Teresa,</p>
<p>This looks really useful.</p>
<p>It seems like this general infrastructure could be useful for
diagnosing places where we have a lot of false sharing too. This
could be between cores, sockets, or devices. Looking forward,
especially with HMM and support under Linux for transparent
unified memory between CPUs and accelerators, I anticipate we'll
end up looking for places where some data accesses are causing
undesired page migrations. Between cores and sockets, the
interesting granularity is probably a cache line (and it looks
like this is what you have anyway). This seems similar to me, the
only difference being that we need to increment the count, not on
every access, but only when the relevant entity id is different
from the last one that accessed the relevant memory region. I'm
not saying that we need to support this, but I'm curious what you
think about this potential use case.<br>
</p>
<p> -Hal<br>
</p>
<div class="moz-cite-prefix">On 6/24/20 6:57 PM, Teresa Johnson via
llvm-dev wrote:<br>
</div>
<blockquote type="cite" cite="mid:CAAe5K+WShf2TgMwkMbWoQ-yHFFMdn4BfjV3zmo2d_E=fjSvgVQ@mail.gmail.com">
<div dir="ltr"><span id="gmail-docs-internal-guid-9e402852-7fff-d302-02c2-83b38a26ec2b">Hi
all,</span>
<div><span><br>
</span></div>
<div><span>I've included an RFC for a heap profiler design I've
been working on in conjunction with David Li. Please send
any questions or feedback. For sanitizer folks, one area of
feedback is on refactoring some of the *ASAN shadow setup
code (see the Shadow Memory section).</span></div>
<div><span><br>
</span></div>
<div><span>Thanks,</span></div>
<div><span>Teresa</span></div>
<div><span> <br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:3pt"><span style="font-size:26pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">RFC: Sanitizer-based Heap Profiler</span></p>
<h1 dir="ltr" style="line-height:1.38;margin-top:20pt;margin-bottom:6pt"><span style="font-size:20pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Summary</span></h1>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">This document provides an overview of an LLVM Sanitizer-based heap profiler design.</span></p>
<h1 dir="ltr" style="line-height:1.38;margin-top:20pt;margin-bottom:6pt"><span style="font-size:20pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Motivation</span></h1>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">The objective of heap memory profiling is to collect critical runtime information associated with heap memory references and information on heap memory allocations. The profile information will be used first for tooling, and subsequently to guide the compiler optimizer and allocation runtime to layout heap objects with improved spatial locality. As a result, DTLB and cache utilization will be improved, and program IPC (performance) will be increased due to reduced TLB and cache misses. More details on the heap profile guided optimizations will be shared in the future.</span></p>
<h1 dir="ltr" style="line-height:1.38;margin-top:20pt;margin-bottom:6pt"><span style="font-size:20pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Overview</span></h1>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">The profiler is based on compiler inserted instrumentation of load and store accesses, and utilizes runtime support to monitor heap allocations and profile data. The target consumer of the heap memory profile information is initially tooling and ultimately automatic data layout optimizations performed by the compiler and/or allocation runtime (with the support of new allocation runtime APIs).</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Each memory address is mapped to </span><a href="https://en.wikipedia.org/wiki/Shadow_memory" style="text-decoration-line:none" moz-do-not-send="true"><span style="font-size:11pt;font-family:Arial;background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;text-decoration-line:underline;vertical-align:baseline;white-space:pre-wrap">Shadow Memory</span></a><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">, similar to the approach used by the </span><a href="https://github.com/google/sanitizers/wiki/AddressSanitizer" style="text-decoration-line:none" moz-do-not-send="true"><span style="font-size:11pt;font-family:Arial;background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;text-decoration-line:underline;vertical-align:baseline;white-space:pre-wrap">Address Sanitizer</span></a><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> (ASAN). Unlike ASAN, which maps each 8 bytes of memory to 1 byte of shadow, the heap profiler maps 64 bytes of memory to 8 bytes of shadow. The shadow location implements the profile counter (incremented on accesses to the corresponding memory). This granularity was chosen to help avoid counter overflow, but it may be possible to consider mapping 32-bytes to 4 bytes. To avoid aliasing of shadow memory for different allocations, we must choose a minimum alignment carefully. As discussed further below, we can attain a 32-byte minimum alignment, instead of a 64-byte alignment, by storing necessary heap information for each allocation in a 32-byte header block.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">The compiler instruments each load and store to increment the associated shadow memory counter, in order to determine hotness.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">The heap profiler runtime is responsible for tracking allocations and deallocations, including the stack at each allocation, and information such as the allocation size and other statistics. I have implemented a prototype built using a stripped down and modified version of ASAN, however this will be a separate library utilizing sanitizer_common components.</span></p>
<h2 dir="ltr" style="line-height:1.38;margin-top:18pt;margin-bottom:6pt"><span style="font-size:16pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Compiler</span></h2>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">A simple HeapProfiler instrumentation pass instruments interesting memory accesses (loads, stores, atomics), with a simple load, increment, store of the associated shadow memory location (computed via a mask and shift to do the mapping of 64 bytes to 8 byte shadow, and add of the shadow offset). The handling is very similar to and based off of the ASAN instrumentation pass, with slightly different instrumentation code.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Various techniques can be used to reduce the overhead, by aggressively coalescing counter updates (e.g. given the 32-byte alignment, accesses known to be in the same 32-byte block, or across possible aliases since we don’t care about the dereferenced values). </span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Additionally, the Clang driver needs to set up to link with the runtime library, much as it does with the sanitizers.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">A </span><span style="font-size:11pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">-fheapprof</span><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> option is added to enable the instrumentation pass and runtime library linking. Similar to </span><span style="font-size:11pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">-fprofile-generate</span><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">, </span><span style="font-size:11pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">-fheapprof</span><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> will accept an argument specifying the directory in which to write the profile.</span></p>
<h2 dir="ltr" style="line-height:1.38;margin-top:18pt;margin-bottom:6pt"><span style="font-size:16pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Runtime</span></h2>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">The heap profiler runtime is responsible for tracking and reporting information about heap allocations and accesses, aggregated by allocation calling context. For example, the hotness, lifetime, and cpu affinity. </span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">A new heapprof library will be created within compiler-rt. It will leverage support within sanitizer_common, which already contains facilities like stack context tracking, needed by the heap profiler.</span></p>
<h3 dir="ltr" style="line-height:1.38;margin-top:16pt;margin-bottom:4pt"><span style="font-size:14pt;font-family:Arial;color:rgb(67,67,67);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Shadow Memory</span></h3>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">There are some basic facilities in sanitizer_common for mmap’ing the shadow memory, but most of the existing setup lives in the ASAN and HWASAN libraries. In the case of ASAN, there is support for both statically assigned shadow offsets (the default on most platforms), and for dynamically assigned shadow memory (implemented for Windows and currently also used for Android and iOS). According to kcc, recent experiments show that the performance with a dynamic shadow is close to that with a static mapping. In fact, that is the only approach currently used by HWASAN. Given the simplicity, the heap profiler will be implemented with a dynamic shadow as well.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">There are a number of functions in ASAN and HWASAN related to setup of the shadow that are duplicated but very nearly identical, at least for linux (which seems to be the only OS flavor currently supported for HWASAN). E.g. ReserveShadowMemoryRange, ProtectGap, and FindDynamicShadowStart (in ASAN there is another nearly identical copy in PremapShadow, used by Android, whereas in HW ASAN the premap handling is already commoned with the non-premap handling). Rather than make yet another copy of these mechanisms, I propose refactoring them into sanitizer_common versions. Like HWASAN, the initial version of the heap profiler will be supported for linux only, but other OSes can be added as needed similar to ASAN.</span></p>
<h3 dir="ltr" style="line-height:1.38;margin-top:16pt;margin-bottom:4pt"><span style="font-size:14pt;font-family:Arial;color:rgb(67,67,67);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">StackTrace and StackDepot</span></h3>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">The sanitizer already contains support for obtaining and representing a stack trace in a StackTrace object, and storing it in the StackDepot which “efficiently stores huge amounts of stack traces”. This is in the sanitizer_common subdirectory and the support is shared by ASAN and ThreadSanitizer. The StackDepot is essentially an unbounded hash table, where each StackTrace is assigned a unique id. ASAN stores this id in the alloc_context_id field in each ChunkHeader (in the redzone preceding each allocation). Additionally, there is support for symbolizing and printing StackTrace objects.</span></p>
<h3 dir="ltr" style="line-height:1.38;margin-top:16pt;margin-bottom:4pt"><span style="font-size:14pt;font-family:Arial;color:rgb(67,67,67);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">ChunkHeader</span></h3>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">The heap profiler needs to track several pieces of information for each allocation. Given the mapping of 64-bytes to 8-bytes shadow, we can achieve a minimum of 32-byte alignment by holding this information in a 32-byte header block preceding each allocation.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">In ASAN, each allocation is preceded by a 16-byte ChunkHeader. It contains information about the current allocation state, user requested size, allocation and free thread ids, the allocation context id (representing the call stack at allocation, assigned by the StackDepot as described above), and misc other bookkeeping. For heap profiling, this will be converted to a 32-byte header block.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Note that we could instead use the metadata section, similar to other sanitizers, which is stored in a separate location. However, as described above, storing the header block with each allocation enables 32-byte alignment without aliasing shadow counters for the same 64 bytes of memory.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">In the prototype heap profiler implementation, the header contains the following fields:</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">// Should be 32 bytes</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">struct ChunkHeader {</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // 1-st 4 bytes</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Carry over from ASAN (available, allocated, quarantined). Will be</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // reduced to 1 bit (available or allocated). </span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 chunk_state : 8;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Carry over from ASAN. Used to determine the start of user allocation.</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 from_memalign : 1;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // 23 bits available</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // 2-nd 4 bytes</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Carry over from ASAN (comment copied verbatim).</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // This field is used for small sizes. For large sizes it is equal to</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // SizeClassMap::kMaxSize and the actual size is stored in the</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // SecondaryAllocator's metadata.</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 user_requested_size : 29;</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // 3-rd 4 bytes</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 cpu_id; // Allocation cpu id</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // 4-th 4 bytes</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Allocation timestamp in ms from a baseline timestamp computed at</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // the start of profiling (to keep this within 32 bits).</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 timestamp_ms;</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // 5-th and 6-th 4 bytes</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Carry over from ASAN. Used to identify allocation stack trace.</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u64 alloc_context_id;</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // 7-th and 8-th 4 bytes</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // UNIMPLEMENTED in prototype - needs instrumentation and IR support.</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u64 data_type_id; // hash of type name</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">};</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">As noted, the chunk state can be reduced to a single bit (no need for quarantined memory in the heap profiler). The header contains a placeholder for the data type hash, which is not yet implemented as it needs instrumentation and IR support.</span></p>
<h3 dir="ltr" style="line-height:1.38;margin-top:16pt;margin-bottom:4pt"><span style="font-size:14pt;font-family:Arial;color:rgb(67,67,67);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Heap Info Block (HIB)</span></h3>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">On a deallocation, information from the corresponding shadow block(s) and header are recorded in a Heap Info Block (HIB) object. The access count is computed from the shadow memory locations for the allocation, as well as the percentage of accessed 64-byte blocks (i.e. the percentage of non-zero 8-byte shadow locations for the whole allocation). Other information such as the deallocation timestamp (for lifetime computation) and deallocation cpu id (to determine migrations) are recorded along with the information in the chunk header recorded on allocation.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">The prototyped HIB object tracks the following:</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">struct HeapInfoBlock {</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Total allocations at this stack context</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 alloc_count;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Access count computed from all allocated 64-byte blocks (track total</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // across all allocations, and the min and max).</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u64 total_access_count, min_access_count, max_access_count;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Allocated size (track total across all allocations, and the min and max).</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u64 total_size;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 min_size, max_size;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Lifetime (track total across all allocations, and the min and max).</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u64 total_lifetime;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 min_lifetime, max_lifetime;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Percent utilization of allocated 64-byte blocks (track total</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // across all allocations, and the min and max). The utilization is</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // defined as the percentage of 8-byte shadow counters corresponding to</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // the full allocation that are non-zero.</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u64 total_percent_utilized;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 min_percent_utilized, max_percent_utilized;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Allocation and deallocation timestamps from the most recent merge into</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // the table with this stack context.</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 alloc_timestamp, dealloc_timestamp;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Allocation and deallocation cpu ids from the most recent merge into</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // the table with this stack context.</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 alloc_cpu_id, dealloc_cpu_id;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Count of allocations at this stack context that had a different</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // allocation and deallocation cpu id.</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 num_migrated_cpu;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Number of times the lifetime of the entry being merged had its lifetime</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // overlap with the previous entry merged with this stack context (by</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // comparing the new alloc/dealloc timestamp with the one last recorded in</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // the entry in the table.</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 num_lifetime_overlaps;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Number of times the alloc/dealloc cpu of the entry being merged was the</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // same as that of the previous entry merged with this stack context</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 num_same_alloc_cpu;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u32 num_same_dealloc_cpu;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // Hash of type name (UNIMPLEMENTED). This needs instrumentation support and</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> // possibly IR changes.</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap"> u64 data_type_id;</span></p>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:10pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">}</span></p>
<h3 dir="ltr" style="line-height:1.38;margin-top:16pt;margin-bottom:4pt"><span style="font-size:14pt;font-family:Arial;color:rgb(67,67,67);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">HIB Table</span></h3>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">The Heap Info Block Table, which is a multi-way associative cache, holds HIB objects from deallocated objects. It is indexed by the stack allocation context id from the chunk header, and currently utilizes a simple mod with a prime number close to a power of two as the hash (because of the way the stack context ids are assigned, a mod of a power of two performs very poorly). Thus far, only 4-way associativity has been evaluated. </span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">HIB entries are added or merged into the HIB Table on each deallocation. If an entry with a matching stack alloc context id is found in the Table, the newly deallocated information is merged into the existing entry. Each HIB Table entry currently tracks the min, max and total value of the various fields for use in computing and reporting the min, max and average when the Table is ultimately dumped.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">If no entry with a matching stack alloc context id is found, a new entry is created. If this causes an eviction, the evicted entry is dumped immediately (by default to stderr, otherwise to a specified report file). Later post processing can merge dumped entries with the same stack alloc context id.</span></p>
<h3 dir="ltr" style="line-height:1.38;margin-top:16pt;margin-bottom:4pt"><span style="font-size:14pt;font-family:Arial;color:rgb(67,67,67);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Initialization</span></h3>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">For ASAN, an __asan_init function initializes the memory allocation tracking support, and the ASAN instrumentation pass in LLVM creates a global constructor to invoke it. The heap profiler prototype adds a new __heapprof_init function, which performs heap profile specific initialization, and the heap profile instrumentation pass calls this new init function instead by a generated global constructor. It currently additionally invokes __asan_init since we are leveraging a modified ASAN runtime. Eventually, this should be changed to initialize refactored common support.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Note that __asan init is also placed in the .preinit_array when it is available, so it is invoked even earlier than global constructors. Currently, it is not possible to do this for __heapprof_init, as it calls timespec_get in order to get a baseline timestamp (as described in the ChunkHeader comments the timestamps (ms) are actually offsets from the baseline timestamp, in order to fit into 32 bits), and system calls cannot be made that early (dl_init is not complete). Since the constructor priority is 1, it should be executed early enough that there are very few allocations before it runs, and likely the best solution is to simply ignore any allocations before initialization.</span></p>
<h3 dir="ltr" style="line-height:1.38;margin-top:16pt;margin-bottom:4pt"><span style="font-size:14pt;font-family:Arial;color:rgb(67,67,67);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Dumping</span></h3>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">For the prototype, the profile is dumped as text with a compact raw format to limit its size. Ultimately it should be dumped in a more compact binary format (i.e. into a different section of the raw instrumentation based profile, with llvm-profdata performing post-processing) which is TBD.</span></p>
<h4 dir="ltr" style="line-height:1.38;margin-top:14pt;margin-bottom:4pt"><span style="font-size:12pt;font-family:Arial;color:rgb(102,102,102);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">HIB Dumping</span></h4>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">As noted earlier, HIB Table entries are created as memory is deallocated. At the end of the run (or whenever dumping is requested, discussed later), HIB entries need to be created for allocations that are still live. Conveniently, the sanitizer allocator already contains a mechanism to walk through all chunks of memory it is tracking (</span><span style="font-size:11pt;font-family:"Courier New";color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">ForEachChunk</span><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">). The heap profiler simply looks for all chunks with a chunk state of allocated, and creates a HIB the same as would be done on deallocation, adding each to the table.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">A HIB Table mechanism for printing each entry is then invoked.</span></p>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">By default, the dumping occurs:</span></p>
<ul style="margin-top:0px;margin-bottom:0px">
<li dir="ltr" style="list-style-type:disc;font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre"><p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">on evictions</span></p></li>
<li dir="ltr" style="list-style-type:disc;font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre"><p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">full table at exit (when the static Allocator object is destructed)</span></p></li>
</ul>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">For running in a load testing scenario, we will want to add a mechanism to provoke finalization (merging currently live allocations) and dumping of the HIB Table before exit. This would be similar to the __llvm_profile_dump facility used for normal PGO counter dumping.</span></p>
<h4 dir="ltr" style="line-height:1.38;margin-top:14pt;margin-bottom:4pt"><span style="font-size:12pt;font-family:Arial;color:rgb(102,102,102);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Stack Trace Dumping</span></h4>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">There is existing support for dumping symbolized StackTrace objects. A wrapper to dump all StackTrace objects in the StackDepot will be added. This new interface is invoked just after the HIB Table is dumped (on exit or via dumping interface).</span></p>
<h4 dir="ltr" style="line-height:1.38;margin-top:14pt;margin-bottom:4pt"><span style="font-size:12pt;font-family:Arial;color:rgb(102,102,102);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Memory Map Dumping</span></h4>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">In cases where we may want to symbolize as a post processing step, we may need the memory map (from /proc/self/smaps). Specifically, this is needed to symbolize binaries using ASLR (Address Space Layout Randomization). There is already support for reading this file and dumping it to the specified report output file (DumpProcessMap()). This is invoked when the profile output file is initialized (HIB Table construction), so that the memory map is available at the top of the raw profile.</span></p>
<h1 dir="ltr" style="line-height:1.38;margin-top:20pt;margin-bottom:6pt"><span style="font-size:20pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-weight:400;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Current Status and Next Steps</span></h1>
<br>
<p dir="ltr" style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">As mentioned earlier, I have a working prototype based on a simplified stripped down version of ASAN. My current plan is to do the following:</span></p>
<ol style="margin-top:0px;margin-bottom:0px">
<li style="list-style-type:decimal;font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre"><p style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Refactor out some of the shadow setup code common between ASAN and HWASAN into sanitizer_common.</span></p></li>
<li style="list-style-type:decimal;font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre"><p style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Rework my prototype into a separate heapprof library in compiler-rt, using sanitizer_common support where possible, and send patches for review.</span></p></li>
<li style="list-style-type:decimal;font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre"><p style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Send patches for the heap profiler instrumentation pass and related clang options.</span></p></li>
<li style="list-style-type:decimal;font-size:11pt;font-family:Arial;color:rgb(0,0,0);background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre"><p style="line-height:1.38;margin-top:0pt;margin-bottom:0pt"><span style="font-size:11pt;background-color:transparent;font-variant-numeric:normal;font-variant-east-asian:normal;vertical-align:baseline;white-space:pre-wrap">Design/implement binary profile format</span></p></li>
</ol>
</span>
<div><br>
</div>
-- <br>
<div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature">
<div dir="ltr">
<div><span style="font-family:Times;font-size:medium">
<table cellspacing="0" cellpadding="0">
<tbody>
<tr style="color:rgb(85,85,85);font-family:sans-serif;font-size:small">
<td style="border-top:2px solid rgb(213,15,37)" nowrap="nowrap">Teresa Johnson |</td>
<td style="border-top:2px solid rgb(51,105,232)" nowrap="nowrap"> Software Engineer |</td>
<td style="border-top:2px solid rgb(0,153,57)" nowrap="nowrap"> <a href="mailto:tejohnson@google.com" target="_blank" moz-do-not-send="true">tejohnson@google.com</a> |</td>
<td style="border-top:2px solid rgb(238,178,17)" nowrap="nowrap"><br>
</td>
</tr>
</tbody>
</table>
</span></div>
</div>
</div>
</div>
</div>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<pre class="moz-quote-pre" wrap="">_______________________________________________
LLVM Developers mailing list
<a class="moz-txt-link-abbreviated" href="mailto:llvm-dev@lists.llvm.org">llvm-dev@lists.llvm.org</a>
<a class="moz-txt-link-freetext" href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a>
</pre>
</blockquote>
<pre class="moz-signature" cols="72">--
Hal Finkel
Lead, Compiler Technology and Programming Languages
Leadership Computing Facility
Argonne National Laboratory</pre>
</body>
</html>