similar to: [LLVMdev] Target specific info available to Clang (and others)

On 10 March 2015 at 16:20, Mehdi Amini <mehdi.amini at apple.com> wrote: > I’d like to avoid as much as possible adding compilation time to the process unless it is necessary (I’m not saying it does not worth it here). As I said in the review, it may end up being faster, because of the amount of crap we'll remove from all tools and LLVM. > It is not clear to me why do you need

Hi all, I am looking at the scheduling model of the ThunderX2. I am trying to figure out the cost of the LDADDALX instruction. The following program’s output is: name LDADDALX; class 872 microops 65535 I would have assumed that the microops are less than 20. The ThunderX2 has a detailed cost model for LSE. Could somebody tell me what I am doing wrong? Cheers, Tom #define GET_REGINFO_ENUM

> On Sep 22, 2017, at 10:34 AM, Thorsten Schütt via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > Hi all, > > I am looking at the scheduling model of the ThunderX2. I am trying to figure out the cost of the LDADDALX instruction. The following program’s output is: > name LDADDALX; class 872 > microops 65535 > I would have assumed that the microops are less than

On Fri, Sep 29, 2017 at 7:51 PM, Andrew Trick via llvm-dev < llvm-dev at lists.llvm.org> wrote: > > > > On Sep 22, 2017, at 10:34 AM, Thorsten Schütt via llvm-dev < > llvm-dev at lists.llvm.org> wrote: > > > > > > > #define GET_REGINFO_ENUM > > #include "AArch64GenRegisterInfo.inc" > > > > #define GET_INSTRINFO_ENUM >

hi, i am trying to find out all the Triple (CPU information) of all the ARM subtargets supported by LLVM 3.3. to do this, i traced all the calls to class ARMSubtarget(), because this class initialized the Triple. i supposed that all the ARM machines need to initialize their own Triple info with ARMSubtarget(). however, to my frustrated, this doesnt seem true, and i still cannot find them

On 24 Oct 2014, at 18:17, Renato Golin <renato.golin at linaro.org> wrote: > I'm beginning to think that the "nice" feature 1 is not worth the two > big problems 2 and 3. If we tie Clang builds with back-end builds and > force it not to have support for other arches (because the info isn't > available if you don't build its back-end), than all that info can

Thanks Bradley. I see that the features set in /ARM.td/ get written to the generated file /<build>/llvm/lib/Target/ARM/ARMGenSubtargetInfo.inc./ Here the ProcA9 features appear in /ARMFeatureKV/ table: /{ "a9", "Cortex-A9 ARM processors", { ARM::ProcA9 }, { *ARM::FeatureFP16* } }, /With your change, the features for ProcA9 in the above entry are empty.//This

Add several .gitignore rules to various directories to ensure a clean worktree after a default build. Signed-off-by: Ramkumar Ramachandra <artagnon at gmail.com> --- Just cloned and built LLVM. This annoyed me. Here's a trivial patch. .gitignore | 10 ++++++++++ bindings/ocaml/llvm/.gitignore | 1 + docs/.gitignore

From: Andrew Trick [mailto:atrick at apple.com] Sent: 24 January 2014 21:52 To: Daniel Sanders Cc: LLVM Developers Mailing List (llvmdev at cs.uiuc.edu) Subject: Re: New machine model questions On Jan 24, 2014, at 2:21 AM, Daniel Sanders <Daniel.Sanders at imgtec.com<mailto:Daniel.Sanders at imgtec.com>> wrote: Hi Andrew, I seem to be making good progress on the P5600 scheduler

On Wed Oct 22 2014 at 12:33:53 PM Chris Lattner <clattner at apple.com> wrote: > On Oct 20, 2014, at 10:41 PM, Eric Christopher <echristo at gmail.com> wrote: > >>> So the storage for DataLayout right now is on a per-subtarget basis. > >>> I.e. if you don't construct one in the module the backend will make > >>> one up based on information in

System : CentOS 4.2 2.6.9-22.0.1.ELsmp System fully up-to-date. 3GB RAM 3ware 9000S card, with Raid5 array. I think that's about all relevant info ... Had file on disk (not array), attempted to mv file to array. Went fine till 2.4GB was copied, then it slowed down to a meg every few minutes. Free memory was ~50MB (Typically is 1.5-2GB), and cache was 2.5GB. Stopped the move, however cache

Hi list I''m writing some migrations to move a legacy db on a rails app. I have a bunch of tables which need some custom modifications and the columns updated_at to be added etc... Now since the rails related columns will be the same for many (but not all) tables I was thinking... can I write a Migration that sounds like that class AddRailsFields < ActiveRecord::Migration def

I have this code in my controller and i cannot seem to figure out how to get it to paginate correctly.. Is there a trick to do something like this? There are a few others at www.rubyonrailsforum.com that are having the same issue. Any incite would be great, thanks in advance! (over 117 views and no answer in rubyonrailsforum???) def display_location if params[:search] &&

Rails assumes tables that have a column called type are using a single inheritance pattern. For example, if you had a model called Article and then had a model called News and another called Feature that both inherit from Article. You can use a single table to represent both child types data. The type column tells rails which model should be used to create the object. sean -----Original

I have ext2 compiled in the kernel (so i can mount the initrd image), all other filesystems are modules. The initrd image contains aic7xxx, jbd, and ext3 modules. /etc/mtab reports the filesystem mounted ext3 /proc/mounts reports it mounted ext2 /proc/filesystems lists ext3 If I umount other partitions that are ext3 the module becomes 'unused'. The filesystem was created

Chris Mason wrote: > > > Hi guys, > > Here's a new patch against 2.4.8-pre8, updated to Al's new super > handling. The differences between the original are small, > but they are big enough that I want extra testing from the > LVM guys (and ext3/XFS). ext3 will probably lock up on unmount with 2.4.8-pre8. The fix is to replace fsync_dev with fsync_no_super in

> ext3 will probably lock up on unmount with 2.4.8-pre8. The fix is > to replace fsync_dev with fsync_no_super in fs/ext3/super.c and > fs/jbd/recovery.c. I'll be generating a new patchset this weekend. Actually if you dont fix recovery.c it will hang the machine when you mount an fs that needs recovering. With the old patches at least umount of ext3 seems fine as is. Alan

Well ... After all let me introduce. I''m a student from Informatic Sciences. Now I try to develop a website using Scriptacolus Library. I have some doubts about some things I need in my website. 1) How I can save user customization page? For example, I offer the posibility to move some elements around the page, organize some elements for example in a menu or submenu, positioning some DIV

On Sat, Nov 7, 2009 at 5:22 PM, Cristian Rojas <intipu@googlemail.com> wrote: > try this on domU, had similar problem and worked for me: > > ethtool -K eth0 tx off > > Cris > > -------------------------------------------------- Hi Chris, Thanx, I have tried that one too, from a google search, with no luck. BUT, it seems that one of the upstream routers might have had

https://bugzilla.mindrot.org/show_bug.cgi?id=2078 Bug ID: 2078 Summary: Documentation claims ~.ssh/config must not be accessible by others when actually it must not be readable Classification: Unclassified Product: Portable OpenSSH Version: 5.9p1 Hardware: Other OS: Linux