search for: isvectorti

What is the proper solution to disable auto-vectorization for unaligned data? I have an out of tree target and I added this: bool OpusTargetLowering::allowsUnalignedMemoryAccesses(EVT VT, bool *Fast) const { if (VT.isVector()) return false; .... } After that, I could see that vectorization is still done on unaligned data except that llvm will copy the data back and forth from the source

Hi, I am executing a C program using JIT. When the program exits, the control comes back to my program which initiates the JIT. Now, I want to read the value(not the LLVM nomenclature) of a global variable (not the LLVM nomenclature). I am able to get the global variable, using the following command: GlobalVariable *my_global = Mod->getNamedGlobal("MY_GLOBAL"); I want to know the

On Jul 19, 2013, at 3:14 PM, Francois Pichet <pichet2000 at gmail.com> wrote: > > What is the proper solution to disable auto-vectorization for unaligned data? > > I have an out of tree target and I added this: > > bool OpusTargetLowering::allowsUnalignedMemoryAccesses(EVT VT, bool *Fast) const { > if (VT.isVector()) > return false; > .... > } >

On Fri, Jul 19, 2013 at 1:14 PM, Francois Pichet <pichet2000 at gmail.com> wrote: > > What is the proper solution to disable auto-vectorization for unaligned > data? Why are you trying to do this? If auto-vectorization is making a given loop slower on your target, that means the cost metrics are off, and we should fix them. If code size is an issue, you should tell the optimizer

Ok any quick workaround to limit vectorization to 16-byte aligned 128-bit data then? All the memory copying done by ExpandUnalignedStore/ExpandUnalignedLoad is just too expensive. On Sat, Jul 20, 2013 at 12:52 PM, Arnold Schwaighofer < aschwaighofer at apple.com> wrote: > > On Jul 19, 2013, at 3:14 PM, Francois Pichet <pichet2000 at gmail.com> wrote: > > > > >

On Mon, Mar 29, 2010 at 01:15:30PM +0100, Renato Golin wrote: > Hi All, > > Which implies no one was expecting a UnionType there... > > Also, if I generate the object code directly, llc fails too... > > Is there any plan to implement the union type? The work-around is quite ugly... Sorry to Renato for getting two copeis of this, I cocked up the reply first time. Anyway,

Hi All, I've noticed the union type in the language manual [1] but it seems it's not used too much. According to the manual, the code: union { int a; double b; } a; Could be compiled to: %union.anon = type union { i32, double } @a = common global %union.anon zeroinitializer, align 8 ; <%union.anon*> [#uses=0] But when I try to assemble it, I get: $ llvm-as union.ll

On 11/17/2011 12:38 AM, Hal Finkel wrote: > Tobias, et al., > > Attached is the my autovectorization pass. Very nice. Will you be at the developer summit? Maybe we could discuss the integration there? Here a first review of the source code. > diff --git a/docs/Passes.html b/docs/Passes.html > index 5c42f3f..076effa 100644 > --- a/docs/Passes.html > +++ b/docs/Passes.html

No, I am afraid not without computing alignment based on the scalar code. In order to limit vectorization to 16-byte aligned data we need to know that data is 16-byte aligned. The way we vectorize we won’t know that until after we have vectorized. As you have observed we will pass “4” to getMemoryOpCost in the loop vectorizer (as that is the only thing that can be inferred from a consecutive

Hello, is there a good way to determine that the instruction is a vector instruction? Or the only way like it's size is multiple of the operands size? -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20170829/bbe6213b/attachment.html>

Hello, Say the original type is Integer i16*,&nbsp; I want to create a v16i16* type to replace it. static Type *getVectorPtr(Type *Ty) { &nbsp; &nbsp; PointerType *PointerTy = dyn_cast<PointerType&gt;(Ty); &nbsp; &nbsp; assert(PointerTy &amp;&amp; "PointerType expected"); &nbsp; &nbsp; unsigned addSpace = PointerTy-&gt;getAddressSpace();

Tobias, I've attached an updated patch. It contains a few bug fixes and many (refactoring and coding-convention) changes inspired by your comments. I'm currently trying to fix the bug responsible for causing a compile failure when compiling test-suite/MultiSource/Applications/obsequi/toggle_move.c; after the pass begins to fuse instructions in a basic block in this file, the aliasing

Tobias, et al., Attached is the my autovectorization pass. I've fixed a bug that appears when using -bb-vectorize-aligned-only, fixed some 80-col violations, etc., and at least on x86_64, all test cases pass except for a few; and all of these failures look like instruction-selection bugs. For example: MultiSource/Applications/ClamAV - fails to compile shared_sha256.c with an error: error in

Tobias, I've attached the latest version of my autovectorization patch. I was able to add support for using the ScalarEvolution analysis for load/store pairing (thanks for your help!). This led to a modest performance increase and a modest compile-time increase. This version also has a cutoff as you suggested (although the default value is set high (4000 instructions between pairs) because

On 11/23/2011 05:52 PM, Hal Finkel wrote: > On Mon, 2011-11-21 at 21:22 -0600, Hal Finkel wrote: >> > On Mon, 2011-11-21 at 11:55 -0600, Hal Finkel wrote: >>> > > Tobias, >>> > > >>> > > I've attached an updated patch. It contains a few bug fixes and many >>> > > (refactoring and coding-convention) changes inspired

Tobias, I've attached an updated copy of the patch. I believe that I accounted for all of your suggestions except for: 1. You said that I could make AA a member of the class and initialize it for each basic block. I suppose that I'd need to make it a pointer, but more generally, what is the thread-safely model that I should have in mind for the analysis passes (will multiple threads

On Mon, 2011-11-21 at 21:22 -0600, Hal Finkel wrote: > On Mon, 2011-11-21 at 11:55 -0600, Hal Finkel wrote: > > Tobias, > > > > I've attached an updated patch. It contains a few bug fixes and many > > (refactoring and coding-convention) changes inspired by your comments. > > > > I'm currently trying to fix the bug responsible for causing a compile

On Fri, 2011-12-02 at 17:07 +0100, Tobias Grosser wrote: > On 11/23/2011 05:52 PM, Hal Finkel wrote: > > On Mon, 2011-11-21 at 21:22 -0600, Hal Finkel wrote: > >> > On Mon, 2011-11-21 at 11:55 -0600, Hal Finkel wrote: > >>> > > Tobias, > >>> > > > >>> > > I've attached an updated patch. It contains a few bug fixes

On Mon, 2011-11-21 at 11:55 -0600, Hal Finkel wrote: > Tobias, > > I've attached an updated patch. It contains a few bug fixes and many > (refactoring and coding-convention) changes inspired by your comments. > > I'm currently trying to fix the bug responsible for causing a compile > failure when compiling >