search for: pakin

I was suprised to find that some bitcode I'm generating isn't getting optimized. Here, I'm doing the equivalent of "myarray[5]++" (on an "extern int *myarray"), repeated three times: @myarray = external global i32* define void @update_array() #0 { %1 = load i32** @myarray, align 8 %2 = getelementptr inbounds i32* %1, i64 5 %3 = load

Hey Scott, On Wed, Apr 24, 2013 at 1:40 PM, Scott Pakin <pakin at lanl.gov> wrote: ... > > Is there some semantic reason that the increments aren't allowed to be > combined, or is this a missed optimization opportunity in LLVM? > > I believe that the wildcard is the extern keyword. Since the external symbol isn't resolved...

...timized if myarray is const, because the optimizer knows that const memory is not modified by stores. It also gets optimized if myarray is an actual array, because then the address of the array is constant, rather than being a value loaded from memory. Dan On Wed, Apr 24, 2013 at 10:40 AM, Scott Pakin <pakin at lanl.gov> wrote: > I was suprised to find that some bitcode I'm generating isn't getting > optimized. Here, I'm doing the equivalent of "myarray[5]++" (on an > "extern int *myarray"), repeated three times: > > @myarray = external...

Andrzej, On 7/14/19 8:02 AM, Andrzej Warzynski wrote: > This is still very early stages, but you can be my guinea pig: > > https://github.com/banach-space/llvm-tutor > > It's a tutorial that I've been preparing recently and am hoping to > present somewhere at some point :-) I believe that it already covers 1), > 2) and 3). I haven't had the time to work on 4).

> From: llvmdev-bounces at cs.uiuc.edu [mailto:llvmdev-bounces at cs.uiuc.edu] > On Behalf Of Scott Pakin > Subject: [LLVMdev] Another missed optimization opportunity? > I'm doing the equivalent of "myarray[5]++" (on an > "extern int *myarray"), repeated three times: > I had expected the three increments by 1 to > be collapsed into a single increment by 3: &gt...

> From: llvmdev-bounces at cs.uiuc.edu [mailto:llvmdev-bounces at cs.uiuc.edu] > On Behalf Of Scott Pakin > Subject: Re: [LLVMdev] Another missed optimization opportunity? > > Is this a potential aliasing effect? Since myarray is defined as a > > pointer, not an array, it's theoretically possible that the address > > therein refers to the same memory location as the pointer...

Hi Scott, On 11/05/13 00:09, Scott Pakin wrote: > Is it possible for a pass to get access to the command line options > passed to it? That is, if I use the CommandLine library to define > > cl::opt<int> Foo("foo", ...); > cl::opt<int> Bar("bar", ...); > cl::opt<bool> B...

On 04/24/2013 01:29 PM, Caldarale, Charles R wrote: > Is this a potential aliasing effect? Since myarray is defined as a pointer, not an array, it's theoretically possible that the address therein refers to the same memory location as the pointer itself. I was thinking along those lines, but I haven't been able to come up with a specific instance of what could possibly be aliased.

Is it possible for a pass to get access to the command line options passed to it? That is, if I use the CommandLine library to define cl::opt<int> Foo("foo", ...); cl::opt<int> Bar("bar", ...); cl::opt<bool> Baz("baz", ...); and the user runs "opt -load mypass.so -foo=123 -std-compile-opts -baz", can I somehow get a string

Hi Scott, On 24/04/13 19:40, Scott Pakin wrote: > I was suprised to find that some bitcode I'm generating isn't getting > optimized. Here, I'm doing the equivalent of "myarray[5]++" (on an > "extern int *myarray"), repeated three times: does your bitcode contain data layout information? Ciao, Du...

I'm looking for a "Hello, world" pass example that ideally has all of the following properties: 1) Complete. That is, it should have functional CMakeLists.txt, etc. files. 2) Out-of-source. That is, it should build against a binary LLVM installation. 3) Modern. That is, I don't want to use any deprecated APIs or CMake macros. 4) Clang-compatible.

Dear All, I need some witty computer related proverbs or one liners that some of have used in the past as footers for your emails for example. I particularly remember a PERL based transliteration joke/footer that when run in PERL produces a witty output.... This is to fix a problem internally created by an over zealous salesman at my company pinning up A4 sheets with bullSh*t sales proverbs

Florian, On 7/10/19 2:17 PM, Florian Hahn wrote: > http://www.cs.cornell.edu/~asampson/blog/llvm.html describes how to > create an out-of-source pass (see 'Let’s Write a Pass’). I think it > is for LLVM 3.8 though. Thanks for pointing that out. I had actually recently stumbled across that page (and the GitHub repository it references, https://github.com/sampsyo/llvm-pass-skeleton)