similar to: [LLVMdev] r135144

Hi Jack, The vector-select patch is not ready yet. R135144 is only a small increment. I am close to finishing the type-legalization work, after which implementing the vector-select x86 support would be easy. I will write to LLVM-dev when I am close to finishing this. Thanks for checking, Nadav -----Original Message----- From: llvmdev-bounces at cs.uiuc.edu [mailto:llvmdev-bounces at

Current dragonegg svn has all of the -fplugin-arg-dragonegg-enable-gcc-optzns bugs for usage with -ffast-math -O3 addressed except for those related to PR2314. Using the -fno-tree-vectorize option, we can evaluate the current state of -fplugin-arg-dragonegg-enable-gcc-optzns with the Polyhedron 2005 benchmarks compared to stock dragonegg and stock gcc 4.5.4. The runtime benchmarks below show that

Hi Jack, thanks for these numbers. Can you also please measure compile times? I'm thinking of enabling gcc optimizations by default, but I don't want to increase compile times, which means choosing a value for the -fplugin-arg-dragonegg-llvm-ir-optimize option that is low enough to get good compile times, yet high enough to get fast code. It would be great if you could play around with

Duncan, Below are the tabulated compile times and executable sizes. A) gcc 4.5.4svn using -msse3 -ffast-math -O3 -fno-tree-vectorize B) gcc 4.5.4svn/dragonegg using -msse3 -ffast-math -O3 -fno-tree-vectorize -fplugin-arg-dragonegg-enable-gcc-optzns C) gcc 4.5.4svn/dragonegg using -msse3 -ffast-math -O3 -fno-tree-vectorize Compile time (seconds) Benchmark A) stock B) gcc 4.5.4/ C)

Hi Jack, thanks for doing this. > Below are the tabulated compile times and executable sizes. > > A) gcc 4.5.4svn using -msse3 -ffast-math -O3 -fno-tree-vectorize > B) gcc 4.5.4svn/dragonegg using -msse3 -ffast-math -O3 -fno-tree-vectorize -fplugin-arg-dragonegg-enable-gcc-optzns > C) gcc 4.5.4svn/dragonegg using -msse3 -ffast-math -O3 -fno-tree-vectorize These numbers really

I agree that a single vector index is sufficient for many cases. Matt Pharr (from the ISPC compiler), showed me an interesting case where there is a single pointer into an array. In this case we need to have two indices, where the first index is zero. Once the basic patch is in, we can start looking at adding support for arrays and multiple indices. Nadav -----Original Message----- From: David

On Thu, Jun 09, 2011 at 03:44:40PM +0200, Duncan Sands wrote: > Hi Jack, thanks for doing this. > >> Below are the tabulated compile times and executable sizes. >> >> A) gcc 4.5.4svn using -msse3 -ffast-math -O3 -fno-tree-vectorize >> B) gcc 4.5.4svn/dragonegg using -msse3 -ffast-math -O3 -fno-tree-vectorize -fplugin-arg-dragonegg-enable-gcc-optzns >> C)

On Thu, Feb 7, 2013 at 3:23 PM, Nadav Rotem <nrotem at apple.com> wrote: > > On Feb 7, 2013, at 3:17 PM, ryan <stdstack at gmail.com> wrote: > > > Thanks a lot for the reply, Nadav! > > I see that using the cost model in target-independent analysis is not > good. But I really like the idea of having instruction > > cost estimation available. Also the

I've updated the release notes for all API changes I've made since 2.9 was branched: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20110711/123990.html > Preferably all in the same place, in order, with (approx.) revision > numbers where it happened. It'd make our jobs much easier to merge our > local base with trunk. I haven't gone into quite that much

https://bugzilla.mindrot.org/show_bug.cgi?id=2314 Bug ID: 2314 Summary: ls -l a_symlink does not list the link permissions Product: Portable OpenSSH Version: -current Hardware: Other OS: All Status: NEW Severity: normal Priority: P5 Component: sftp Assignee: unassigned-bugs

Hello, I started working on adding vector support for the SELECT and CMP instructions in the codegen (bugs: 3384, 1784, 2314).  Currently, the codegen scalarizes vector CMPs into multiple scalar CMPs.  It is easy to add similar scalarization support to the SELECT instruction.  However, using multiple scalar operations is slower than using vector operations. In LLVM, vector-compare operations

On Wed, Oct 31, 2012 at 05:17:28PM -0700, Eli Friedman wrote: > On Wed, Oct 31, 2012 at 5:01 PM, Jack Howarth <howarth at bromo.med.uc.edu> wrote: > > On Wed, Oct 31, 2012 at 04:43:42PM -0700, Eli Friedman wrote: > >> On Wed, Oct 31, 2012 at 4:33 PM, Jack Howarth <howarth at bromo.med.uc.edu> wrote: > >> > On Wed, Oct 31, 2012 at 03:53:50PM -0700, Eli

Hi. I am new to the listm so, first of all I'd like to say hello to everyone here in the list. I hope this list is an adequate place to post this; else, please, kindly direct me to the right list :) I write because I am having an issue with nouveau and mplayer. I have been only able to reproduce it with mplayer, but I think the issue lies in nouveau, either the kernel part or libdrm. The

On 07/14/2011 11:23 AM, Jay Foad wrote: > I've updated the release notes for all API changes I've made since 2.9 > was branched: > > http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20110711/123990.html Thanks. But what about the API changes related to the new type system (OpaqueType gone, etc.)? Is that documented somewhere? Shouldn't it be in the release

On Wed, Oct 31, 2012 at 04:43:42PM -0700, Eli Friedman wrote: > On Wed, Oct 31, 2012 at 4:33 PM, Jack Howarth <howarth at bromo.med.uc.edu> wrote: > > On Wed, Oct 31, 2012 at 03:53:50PM -0700, Eli Friedman wrote: > >> On Wed, Oct 31, 2012 at 8:43 AM, Jack Howarth <howarth at bromo.med.uc.edu> wrote: > >> > While testing llvm/polly svn on

Am I doing something wrong? I can't get this dial command to timeout.... Dial(Zap/g1/xxxxxxx,20) -- Gary White admin@netpathway.com Network Administrator Internet Pathway 105 D East Church Street Voice: 601-776-3355 P. O. Box 777 Fax: 601-776-2314 Quitman, MS 39355

Hi, It's my first try to setup RODC using Samba 4.8. We have latest Samba 4.7 environnement with 2 DC and some file servers. Joining the DC to the domain is OK using samba-tool domain join command. The domain controller appears in the DC list (MMC) However, users cannot be authenticated. Samba is running but these ports are closed : netbios-ssn 139/tcp # NETBIOS session service

Hy all of you, I've successfully installed a freepbx solution with 10 extensions : - 5 on Linksys SPA922 - 1 on Linksys SPA942 - 1 on Thomson ST022 Everything seems to work fine with all the hardphones excepts last week. The thomson has a strange behaviour. It can reach french mobile cell phones but when it reaches "fix" phones, the correspondant can't hear the caller. What

On Wed, Oct 31, 2012 at 5:26 PM, Jack Howarth <howarth at bromo.med.uc.edu> wrote: > On Wed, Oct 31, 2012 at 05:17:28PM -0700, Eli Friedman wrote: >> On Wed, Oct 31, 2012 at 5:01 PM, Jack Howarth <howarth at bromo.med.uc.edu> wrote: >> > On Wed, Oct 31, 2012 at 04:43:42PM -0700, Eli Friedman wrote: >> >> On Wed, Oct 31, 2012 at 4:33 PM, Jack Howarth

On Thu, Jun 09, 2011 at 03:44:40PM +0200, Duncan Sands wrote: > Hi Jack, thanks for doing this. > >> Below are the tabulated compile times and executable sizes. >> >> A) gcc 4.5.4svn using -msse3 -ffast-math -O3 -fno-tree-vectorize >> B) gcc 4.5.4svn/dragonegg using -msse3 -ffast-math -O3 -fno-tree-vectorize -fplugin-arg-dragonegg-enable-gcc-optzns >> C)