<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="/rss.xsl.xml"?>
<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/">
<channel>
    <title>Changes in stack_layout.ll</title>
    <description></description>
    <language>en</language>
    <copyright>Copyright 2015</copyright>
    <generator>Java</generator><item>
        <title>0df277a1 - [test] Remove some legacy PM tests in llvm/test/Instrumentation/AddressSanitizer</title>
        <link>http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll#0df277a1</link>
        <description>[test] Remove some legacy PM tests in llvm/test/Instrumentation/AddressSanitizer

            List of files:
            /llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll</description>
        <pubDate>Sun, 05 Sep 2021 19:02:31 +0000</pubDate>
        <dc:creator>Arthur Eubanks &lt;aeubanks@google.com&gt;</dc:creator>
    </item>
<item>
        <title>7c2f2762 - [NewPM][ASan] Make ASan tests work under NPM</title>
        <link>http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll#7c2f2762</link>
        <description>[NewPM][ASan] Make ASan tests work under NPMUnder NPM, the asan-globals-md analysis is required but cannot be runwithin the asan function pass due to module analyses not being able torun from a function pass. So this pins all tests using &quot;-asan&quot; to thelegacy PM and adds a corresponding RUN line with-passes=&apos;require&lt;asan-globals-md&gt;,function(asan)&apos;.Now all tests in Instrumentation/AddressSanitizer pass when-enable-new-pm is by default on.Tests were automatically converted using the following python script andfailures were manually fixed up.import sysfor i in sys.argv:    with open(i, &apos;r&apos;) as f:        s = f.read()    with open(i, &apos;w&apos;) as f:        for l in s.splitlines():            if &quot;RUN:&quot; in l and &apos; -asan -asan-module &apos; in l and &apos;\\&apos; not in l:                f.write(l.replace(&apos; -asan -asan-module &apos;, &apos; -asan -asan-module -enable-new-pm=0 &apos;))                f.write(&apos;\n&apos;)                f.write(l.replace(&apos; -asan -asan-module &apos;, &quot; -passes=&apos;require&lt;asan-globals-md&gt;,function(asan),module(asan-module)&apos; &quot;))                f.write(&apos;\n&apos;)            elif &quot;RUN:&quot; in l and &apos; -asan &apos; in l and &apos;\\&apos; not in l:                f.write(l.replace(&apos; -asan &apos;, &apos; -asan -enable-new-pm=0 &apos;))                f.write(&apos;\n&apos;)                f.write(l.replace(&apos; -asan &apos;, &quot; -passes=&apos;require&lt;asan-globals-md&gt;,function(asan)&apos; &quot;))                f.write(&apos;\n&apos;)            else:                f.write(l)                f.write(&apos;\n&apos;)See https://bugs.llvm.org/show_bug.cgi?id=46611.Reviewed By: vitalybukaDifferential Revision: https://reviews.llvm.org/D83921

            List of files:
            /llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll</description>
        <pubDate>Sat, 18 Jul 2020 00:49:46 +0000</pubDate>
        <dc:creator>Arthur Eubanks &lt;aeubanks@google.com&gt;</dc:creator>
    </item>
<item>
        <title>3c6c14d1 - ASAN: Provide reliable debug info for local variables at -O0.</title>
        <link>http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll#3c6c14d1</link>
        <description>ASAN: Provide reliable debug info for local variables at -O0.The function stack poisioner conditionally stores local variableseither in an alloca or in malloc&apos;ated memory, which has theunfortunate side-effect, that the actual address of the variable isonly materialized when the variable is accessed, which means thatthose variables are mostly invisible to the debugger even whencompiling without optimizations.This patch stores the address of the local stack base into an alloca,which can be referred to by the debug info and is available throughoutthe function. This adds one extra pointer-sized alloca to each stackframe (but mem2reg can optimize it away again when optimizations areenabled, yielding roughly the same debug info quality as before inoptimized code).rdar://problem/30433661Differential Revision: https://reviews.llvm.org/D41034llvm-svn: 320415

            List of files:
            /llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll</description>
        <pubDate>Mon, 11 Dec 2017 20:43:21 +0000</pubDate>
        <dc:creator>Adrian Prantl &lt;aprantl@apple.com&gt;</dc:creator>
    </item>
<item>
        <title>f10061ec - Add address space mangling to lifetime intrinsics</title>
        <link>http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll#f10061ec</link>
        <description>Add address space mangling to lifetime intrinsicsIn preparation for allowing allocas to have non-0 addrspace.llvm-svn: 299876

            List of files:
            /llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll</description>
        <pubDate>Mon, 10 Apr 2017 20:18:21 +0000</pubDate>
        <dc:creator>Matt Arsenault &lt;Matthew.Arsenault@amd.com&gt;</dc:creator>
    </item>
<item>
        <title>d88e5201 - [asan] Append line number to variable name if line is available and in the same file as the function.</title>
        <link>http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll#d88e5201</link>
        <description>[asan] Append line number to variable name if line is available and in the same file as the function.PR30498Reviewers: eugenisDifferential Revision: https://reviews.llvm.org/D25715llvm-svn: 284546

            List of files:
            /llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll</description>
        <pubDate>Tue, 18 Oct 2016 23:29:41 +0000</pubDate>
        <dc:creator>Vitaly Buka &lt;vitalybuka@google.com&gt;</dc:creator>
    </item>
<item>
        <title>8ed1d819 - [asan] Skip promotable allocas to improve performance at -O0</title>
        <link>http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll#8ed1d819</link>
        <description>[asan] Skip promotable allocas to improve performance at -O0Currently, the ASan executables built with -O0 are unnecessarily slow.The main reason is that ASan instrumentation pass inserts redundantchecks around promotable allocas. These allocas do not get instrumentedunder -O1 because they get converted to virtual registered by mem2reg.With this patch, ASan instrumentation pass will only instrument nonpromotable allocas, giving us a speedup of 39% on a collection ofbenchmarks with -O0. (There is no measurable speedup at -O1.)llvm-svn: 230724

            List of files:
            /llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll</description>
        <pubDate>Fri, 27 Feb 2015 03:12:36 +0000</pubDate>
        <dc:creator>Anna Zaks &lt;ganna@apple.com&gt;</dc:creator>
    </item>
<item>
        <title>4b7f413e - [ASan] Change fake stack and local variables handling.</title>
        <link>http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll#4b7f413e</link>
        <description>[ASan] Change fake stack and local variables handling.This commit changes the way we get fake stack from ASan runtime(to find use-after-return errors) and the way we represent localvariables:  - __asan_stack_malloc function now returns pointer to newly allocated    fake stack frame, or NULL if frame cannot be allocated. It doesn&apos;t    take pointer to real stack as an input argument, it is calculated    inside the runtime.  - __asan_stack_free function doesn&apos;t take pointer to real stack as    an input argument. Now this function is never called if fake stack    frame wasn&apos;t allocated.  - __asan_init version is bumped to reflect changes in the ABI.  - new flag &quot;-asan-stack-dynamic-alloca&quot; allows to store all the    function local variables in a dynamic alloca, instead of the static    one. It reduces the stack space usage in use-after-return mode    (dynamic alloca will not be called if the local variables are stored    in a fake stack), and improves the debug info quality for local    variables (they will not be described relatively to %rbp/%rsp, which    are assumed to be clobbered by function calls). This flag is turned    off by default for now, but I plan to turn it on after more    testing.llvm-svn: 224062

            List of files:
            /llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll</description>
        <pubDate>Thu, 11 Dec 2014 21:53:03 +0000</pubDate>
        <dc:creator>Alexey Samsonov &lt;vonosmas@gmail.com&gt;</dc:creator>
    </item>
<item>
        <title>7aafd31d - [ASan] Add -asan-module to the ASan .ll tests.</title>
        <link>http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll#7aafd31d</link>
        <description>[ASan] Add -asan-module to the ASan .ll tests.After the -asan pass had been split into -asan (function-level) and -asan-module (module-level) some of thetests have silently stopped working, because they didn&apos;t instrument the globals anymore.We&apos;ve decided to have every test using both passes, irrespective of the presence of globals in it.llvm-svn: 204335

            List of files:
            /llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll</description>
        <pubDate>Thu, 20 Mar 2014 11:16:34 +0000</pubDate>
        <dc:creator>Alexander Potapenko &lt;glider@google.com&gt;</dc:creator>
    </item>
<item>
        <title>cb66fe37 - [ASan] Fix the tests broken by r198018 to check for private linkage of ASan-generated globals.</title>
        <link>http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll#cb66fe37</link>
        <description>[ASan] Fix the tests broken by r198018 to check for private linkage of ASan-generated globals.llvm-svn: 198020

            List of files:
            /llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll</description>
        <pubDate>Wed, 25 Dec 2013 17:06:04 +0000</pubDate>
        <dc:creator>Alexander Potapenko &lt;glider@google.com&gt;</dc:creator>
    </item>
<item>
        <title>4fb7801b - [asan] rewrite asan&apos;s stack frame layout</title>
        <link>http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll#4fb7801b</link>
        <description>[asan] rewrite asan&apos;s stack frame layoutSummary:Rewrite asan&apos;s stack frame layout.First, most of the stack layout logic is moved into a separte fileto make it more testable and (potentially) useful for other projects.Second, make the frames more compact by using adaptive redzones(smaller for small objects, larger for large objects).Third, try to minimized gaps due to large alignments (this is hypothetical sincetoday we don&apos;t see many stack vars aligned by more than 32).The frames indeed become more compact, but I&apos;ll still need to run more benchmarksbefore committing, but I am sking for review now to get early feedback.This change will be accompanied by a trivial change in compiler-rt teststo match the new frame sizes.Reviewers: samsonov, dvyukovReviewed By: samsonovCC: llvm-commitsDifferential Revision: http://llvm-reviews.chandlerc.com/D2324llvm-svn: 196568

            List of files:
            /llvm-project-15.0.7/llvm/test/Instrumentation/AddressSanitizer/stack_layout.ll</description>
        <pubDate>Fri, 06 Dec 2013 09:00:17 +0000</pubDate>
        <dc:creator>Kostya Serebryany &lt;kcc@google.com&gt;</dc:creator>
    </item>
</channel>
</rss>
