similar to: [LLVMdev] Missing InstCombine optimization.

> 1. hi*8 -> hi << 3 > > 2. ((idx >> 3) << 3) -> idx & -8 > > 3. hi*8+lo -> hi*8 | lo > > 4. (idx & -8) | (idx & 7) -> idx & (-8 | 7) -> idx > > > > After 155362 pattern #2 is deferred to DAGCombine stage, so InstCombine is > unable to apply pattern #4: > > 4*. ((idx >>

Hi all. I'm looking for an advice on how to deal with inefficient code generation for Intel Nehalem/Westmere architecture on 64-bit platform for the attached test.cpp (LLVM IR is in test.cpp.ll). The inner loop has 11 iterations and eventually unrolled. Test.lea.s is the assembly code of the outer loop. It simply has 11 loads, 11 FP add, 11 FP mull, 1 FP store and lea+mov for index

This sounds like llvm.org/pr13320. On 17 September 2013 18:20, Bader, Aleksey A <aleksey.a.bader at intel.com> wrote: > Hi all. > > > > I’m looking for an advice on how to deal with inefficient code generation > for Intel Nehalem/Westmere architecture on 64-bit platform for the attached > test.cpp (LLVM IR is in test.cpp.ll). > > The inner loop has 11 iterations

The two address pass is only concerned about register pressure. It sounds like it should be taught about profitability. In cases where profitability can only be determined with something machinetracemetric then it probably should live it to more sophisticated pass like regalloc. In this case, we probably need a profitability target hook which knows about lea. We should also consider disabling

On Fri, May 15, 2015 at 11:50 AM, David Chisnall < David.Chisnall at cl.cam.ac.uk> wrote: > On 15 May 2015, at 17:53, Chris Bieneman <beanz at apple.com> wrote: > > > > +1 to lib/Target/SPIRV/(Reader|Writer) > > > > I really like this idea. I’ve talked with some people on both the LLVM > and Khronos sides and I really think adding SPIR-V support to LLVM

+ llvm-dev From: Sumner, Brian Sent: Tuesday, May 10, 2016 3:11 PM To: Anastasia Stulova <Anastasia.Stulova at arm.com>; Liu, Yaxun (Sam) <Yaxun.Liu at amd.com>; cfe-dev (cfe-dev at lists.llvm.org) <cfe-dev at lists.llvm.org>; Pan, Xiuli <xiuli.pan at intel.com>; Bader, Alexey (alexey.bader at intel.com) <alexey.bader at intel.com> Cc: Stellard, Thomas

I accidentally realized today that any domU''s using the paravirt disk driver potentially suffer from poor performance when they get handed in a physical volume and partitioning is done inside the guest. The physical volume passed in has to be one that has the compat 512 logical sector size but hints its real sector size (eg. 4096) as physical sector size. In dom0 handling is correct and

Ok, downloading the BackupPC RPM directly did not work to myriad dependencies and sub-dependencies. How do I add the Test repository to my yum config, please? Aleksey On 7/31/08, Aleksey Tsalolikhin <atsaloli.tech at gmail.com> wrote: > Hi. I just installed CentOS for the first time, to make a BackupPC server. > > However, when I try to install BackupPC using yum, I get the

I am currently looking at a bug report[1] which is happening when a Xen PVM guest with multiple VCPUs is running a high IO database load (a test script is available in the bug report). In experimenting it seems that this happens (or is getting more likely) when the number of VCPUs is 8 or higher (though I have not tried 6, only 2 and 4), having autogroup enabled seems to make it more likely, too

When I try to start R from my desktop, an "information"-window pops up: "Fatal error: Invalid HOMEDRIVE". I already deleted and reinstalled the program, but the problem lasts. How can I solve this problem? I would be very glad, if you can give me a hint, what??s going wrong. Best regards Sabine Bader -- Sabine Bader Hamburger Str. 83 44135 Dortmund 0231/7950683

I don't see any problem mangling it to be honest even though there seems to be only one prototype anyways. We could add restrict in as well. Cheers, Anastasia ________________________________ From: Hongbin Zheng <etherzhhb at gmail.com> Sent: 17 September 2016 05:32:54 To: Liu, Yaxun (Sam) Cc: cfe-dev at lists.llvm.org; llvm-dev; Bader, Alexey (alexey.bader at intel.com); Anastasia

Full_Name: Martin Karl-Friedrich Bader Version: 2.8.0 OS: 10.5.3 Submission from: (NULL) (131.152.1.1) Hi there, when using las=1 on the y-axis the labels keep their centered justification, e.g. if you plot a graph using the axis command: axis(side=2, at=c(0,50, 100, 125), lab=c("0","50","100", " "), las=1), then the 0 and the 50 are centered in relation

Hi all, Is it planned to add support of rotations to IR? Thanks, Aleksey Zasenko -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20151119/c897b935/attachment.html>

On 23.01.2015 15:23, Ian Campbell wrote: > On Fri, 2015-01-23 at 14:56 +0100, Stefan Bader wrote: >> On 22.01.2015 17:02, Ian Campbell wrote: >>> On Thu, 2015-01-22 at 13:02 +0100, Stefan Bader wrote: >>>> That one modifies xen-init-list to look at domains managed by the xl namespace. >>>> This additionally works around a bug in xl which will report a

Hi to all of vorbis developers! I think it is time to do core optimizations to vorbis-coder/decoder in asm. I`d like to help. What you think about that?

Hi to all of vorbis developers! I think it is time to do core optimizations to vorbis-coder/decoder in asm. I`d like to help. What you think about that?

Hi R experts, I am looking for a simple error handling approach, whereby I could stop function execution with a customized error message. For example: for (i in 1:10) { if (i == 5) # I'd like to be able to stop right here with an error message I have complete control over } The problem with stop() is that I cannot control completely what gets printed to the terminal, even with

Hello, Dear SSH developers. I'm currently studying Business Information Technologies at the University of Applied Sciences in Oulu, Finland. I'm about to start my own online computer security related magazine. I would like to make an article about the helpful ssh protocol for security professionals. I have made sequence diagrams on how the SSH 2.0 protocol works, however i'm not sure

On 22.01.2015 17:02, Ian Campbell wrote: > On Thu, 2015-01-22 at 13:02 +0100, Stefan Bader wrote: >> That one modifies xen-init-list to look at domains managed by the xl namespace. >> This additionally works around a bug in xl which will report a syntax error when >> "xl list -l" is called while domains in a different namespace exist. Which is >> the case when

Dear R users, I have a question on using weighted.mean() while aggregating a data frame. I have a data frame with columns Sub, Length and Slope: > x[1:5,] Sub Length Slope 1 2 351.547 0.0025284969 2 2 343.738 0.0025859390 3 1 696.659 0.0015948968 4 2 5442.338 0.0026132544 5 1 209.483 0.0005304225 and I would like to calculate the weighted.mean of Slope, using Length