Displaying 20 results from an estimated 111 matches for "unnatural".
Did you mean:
natural
2009 Feb 12
2
[LLVMdev] Unnatural loops with O0
Hi Chris,
Is there a compelling reason why llvm-gcc does not always produce natural
loops. Is it a code size issue, or are there performance implications as
well? I am seeing a simple 'while' loop compiled to an unnatural loop,
without any gotos, breaks, or continues. What is the reason for this?
Marc
On Wed, Feb 11, 2009 at 6:51 PM, Chris Lattner <clattner at apple.com> wrote:
>
> On Feb 11, 2009, at 1:13 PM, Marc de Kruijf wrote:
>
> > I am reviving this thread because I am seeing the same...
2009 Feb 12
0
[LLVMdev] Unnatural loops with O0
Hi Marc,
> Is there a compelling reason why llvm-gcc does not always produce natural
> loops. Is it a code size issue, or are there performance implications as
> well? I am seeing a simple 'while' loop compiled to an unnatural loop,
> without any gotos, breaks, or continues. What is the reason for this?
is it already an unnatural loop when it comes out of the gcc parts of
llvm-gcc (you can check this by compiling with: -O0 -emit-llvm)? Or
is it llvm itself that creates the unnatural loops?
Ciao,
Duncan.
2009 Feb 12
0
[LLVMdev] Unnatural loops with O0
On Feb 11, 2009, at 1:13 PM, Marc de Kruijf wrote:
> I am reviving this thread because I am seeing the same thing
> (unnatural loops produced by llvm-gcc), but it is not limited to -O0
> -- I am seeing it for -O2 and -O3 as well.
> Some of my research work is relying on LoopInfo to provide loop
> information for all loops, but it is missing these loops. Is there
> any work in the pipeline that aims to fi...
2009 Feb 11
2
[LLVMdev] Unnatural loops with O0
I am reviving this thread because I am seeing the same thing (unnatural
loops produced by llvm-gcc), but it is not limited to -O0 -- I am seeing it
for -O2 and -O3 as well.
Some of my research work is relying on LoopInfo to provide loop information
for all loops, but it is missing these loops. Is there any work in the
pipeline that aims to fix this?
Many thanks,
Marc...
2008 Jun 21
0
[LLVMdev] Unnatural loops with O0
On Jun 11, 2008, at 6:27 AM, Florian Brandner wrote:
> On Thursday 08 May 2008 18:33:48 Adrian Prantl wrote:
>> we noticed that llvmgcc4.2-2.2 sometimes generates non-natural loops
>> when compiling to bytecode without any optimizations. Apparently what
>> happens is that the loop header is duplicated, which results in two
>> entry points for the loop.
>
> this is
2008 Jun 11
1
[LLVMdev] Unnatural loops with O0
On Thursday 08 May 2008 18:33:48 Adrian Prantl wrote:
> we noticed that llvmgcc4.2-2.2 sometimes generates non-natural loops
> when compiling to bytecode without any optimizations. Apparently what
> happens is that the loop header is duplicated, which results in two
> entry points for the loop.
this is actually a problem with the tailduplication pass of llvm. it does not
consider
2008 May 08
3
[LLVMdev] Unnatural loops with O0
Hello everybody,
we noticed that llvmgcc4.2-2.2 sometimes generates non-natural loops
when compiling to bytecode without any optimizations. Apparently what
happens is that the loop header is duplicated, which results in two
entry points for the loop. Since this could obstruct subsequent loop
optimizations, it might be interesting to further investigate this behavior.
To show the problem, I have
2007 Jun 03
5
can I use template to assemble a html page in an controller and redirect to another controller?
Hi,all!
I want to assemble a html page and put it into flash and redirect to
another controller ,for example:
flash[:aaa]="<html><body><div>hi</div></body></html>"
but what I need is to assemble some more complicate html page,so I
want to read the render result of a template before to send it to
browser.
Any thought?
Thanks in advance :-)
2012 Sep 13
6
[LLVMdev] Proposal: New IR instruction for casting between address spaces
...also be fine to force ptrtoint and inttoptr, although it does take one instruction more.
>>
>
> The problem with using ptrtoint and inttoptr is that one has to pick an intermediate integer type that is safe to convert to and from. Since the pointer size is target dependent, it seems unnatural to use ptrtoint and inttoptr for that.
If we were to add first class support for this, we'd have to add three new instructions: trunc_ptr, sext_ptr, and zext_ptr, all of which would be used for pointer to pointer conversions.
Why is it so bad to use ptrtoint/inttoptr with some large-enough...
2006 Jul 06
3
[PATCH] Fix argument check of xm reboot command (3)
Hi,
I tested unlikely operations about the xm reboot/shutdown command.
As a result, I found the following problems.
Problem 1: Can reboot/shutdown Domain-0 by mistake.
# xm list
Name ID Mem(MiB) VCPUs State Time(s)
Domain-0 0 1024 2 r----- 31.0
# xm reboot 0
Error: Can''t specify Domain-0
# xm reboot 00
#
2012 Jul 25
6
[LLVMdev] X86 FMA4
We're migrating to LLVM 3.1 and trying to use the upstream FMA patterns.
Why is VFMADDSD4 defined with vector types? Is this simply because the
gcc intrinsic uses vector types? It's quite unnatural if you have a
compiler that generates FMAs as opposed to requiring user intrinsics.
-Dave
2004 Sep 30
2
How to save graphics in portable way in batch mode?
...the postscript() in prior of
plot() call and then 15 times the dev.off() as epilogue
to plot() call.
5. Hm, one could create function similar to Sweave()
(see R-1.9.1\src\library\utils\R\Sweave.R) and
benefit from automatic name generation and saving.
But it is quite unnatural, isn't it?
so, it would be nice to know what is the "right way".
P.S. BTW, will be there any specific changes in R-2.x?
Thank you a priori for your comments,
Valery.
2007 Jun 02
0
To glenpool who claremore
the unnatural war was still raging between the two brothers, Cleopatra, drawing toward the close of his life, being in fact over eighty years of were brought to bear upon her in the soft and voluptuous clime where the such a condition as to make it extremely dangerous and difficult of
Mediterranean, at a point t...
2007 Jun 02
0
To glenpool who claremore
the unnatural war was still raging between the two brothers, Cleopatra, drawing toward the close of his life, being in fact over eighty years of were brought to bear upon her in the soft and voluptuous clime where the such a condition as to make it extremely dangerous and difficult of
Mediterranean, at a point t...
2014 Sep 10
2
[LLVMdev] awkward object file abstractions
...oint directly to the symbol
// entry in the memory mapped object file. SymbolPimpl cannot contain any
// virtual functions because then it could not point into the memory mapped
// file.
Is this still a goal and if so, why? Building symbol objects on top
of the file's memory image causes some unnatural acts, e.g. no virtual
functions, for what otherwise could be an unsurprising class
hierarchy.
Thanks,
-steve
2007 Jun 30
1
[LLVMdev] Repository Layout
Hello, Reid.
> Fortunately, with SVN, making such a change isn't hard and its tracked
> in the commits. Anyone object or have other ideas in this realm?
I'm also confused. At one hand, proposed layout (with BTT in the top)
seems to be "unnatural", when we will create new tags/branches.
In theory, each module can have its own version, thus if the want to
"tag" it, we have either copy the whole repository to new tag, wasting
space and checkout times, or "tag" just a subset of the repository,
which seems to be prett...
2007 Jul 27
4
Q: extracting data from lm
...at R is unlike anything I've
ever seen, and it's far from trivial to get my head around some of it --
for example, how I could have known a priori that the above is the way to
get the nu4 coefficient is beyond me). Anyway, having figured out how to
get the estimate of the coefficient, I not-unnaturally wanted also to find
a way to access the std. error of the estimate (the value 0.01025 in the
summary). But I am completely mystified as to how to do it :-(
Any help gratefully (VERY gratefully) received, and I apologise if this is
a really, really stupid question and that the answer lies somewhe...
2012 Jul 26
0
[LLVMdev] X86 FMA4
...t; Sent: Wednesday, July 25, 2012 3:26 PM
> Subject: [LLVMdev] X86 FMA4
>
> We're migrating to LLVM 3.1 and trying to use the upstream FMA patterns.
>
> Why is VFMADDSD4 defined with vector types? Is this simply because the
> gcc intrinsic uses vector types? It's quite unnatural if you have a
> compiler that generates FMAs as opposed to requiring user intrinsics.
>
> -Dave
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu
> http://lis...
2012 Sep 13
0
[LLVMdev] Proposal: New IR instruction for casting between address spaces
...e to force ptrtoint and inttoptr, although it does take one instruction more.
>>>
>>
>> The problem with using ptrtoint and inttoptr is that one has to pick an intermediate integer type that is safe to convert to and from. Since the pointer size is target dependent, it seems unnatural to use ptrtoint and inttoptr for that.
>
> If we were to add first class support for this, we'd have to add three new instructions: trunc_ptr, sext_ptr, and zext_ptr, all of which would be used for pointer to pointer conversions.
>
> Why is it so bad to use ptrtoint/inttoptr wit...
2012 Sep 13
0
[LLVMdev] Proposal: New IR instruction for casting between address spaces
...oint and inttoptr, although it does take one instruction
> more.
> >>
> >
> > The problem with using ptrtoint and inttoptr is that one has to pick
> an intermediate integer type that is safe to convert to and from.
> Since the pointer size is target dependent, it seems unnatural to use
> ptrtoint and inttoptr for that.
>
> If we were to add first class support for this, we'd have to add three
> new instructions: trunc_ptr, sext_ptr, and zext_ptr, all of which would
> be used for pointer to pointer conversions.
[Villmow, Micah] Would we really need to ad...