similar to: Disable combining of loads and stores in instcombine

I’m writing a pass for some custom hardware — we’d like to split arrays across hardware elements; this doesn’t work if consecutive writes to characters get combined to a word. On Apr 16, 2019, 8:17 PM -0700, Tom Stellard <tstellar at redhat.com>, wrote: > On 04/16/2019 11:38 AM, Neil Ryan via llvm-dev wrote: > > LLVM's optimizer combines stores to consecutive characters into a

> On Apr 17, 2019, at 5:02 AM, Arsenault, Matthew via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > This won’t happen with volatile load/store This is mostly true today, but AFAICT the LLVM memory model doesn’t actually offer this guarantee. It merely says that LLVM treats volatile like C / C++ treats volatile… which isn’t much of a guarantee because C / C++ volatile doesn’t

IIRC it’s not strictly possible to determine what array a load/store is based on. I don’t believe the decomposition is always possible, as information is lost when accesses are combined. Neil On Apr 17, 2019, 12:31 PM -0700, Arsenault, Matthew <Matthew.Arsenault at amd.com>, wrote: > This is really a codegen problem. You can decompose the load/store however you like in the backend.

So, volatile’s been a fine solution — the issue is volatile pointers would perform the load every time; ideally memory accesses would be able to be cached. This is why I’ve been leaning towards disabling the part of instcombine that combines memory accesses instead of using volatile — there should be better performance. On Apr 17, 2019, 9:54 AM -0700, Jameson Nash <vtjnash at gmail.com>,

> Why do you want this? The goal is to share arrays between multiple tiles on a manycore architecture by splitting arrays between tiles. With a DRF memory model, it makes sense to elide multiple loads to the same memory location between barriers.; IIRC the semantics for volatile don’t allow this eliding. -------------- next part -------------- An HTML attachment was scrubbed... URL:

> But why is it desirable to avoid combining adjacent stores? If you’ve got DRF code then the combination can’t be observed. It’s more that the consecutive stores would be going to different tiles. If multiple stores are combined in IR, I don’t think they’d be able to decoupled in IR, unless there’s a way to always determine which global object an arbitrary GEP is pointing to. --------------

Ok I think we have some common ground - CSE should choose the aliased pointer over the non-aliased one because we don't want the no-aliasing information to creep outwards from the inlined callsite. I'll put together a patch in the coming days and add y'all as reviewers so you get visibility. Cheers, -Neil. On Wed, Jan 22, 2020 at 4:47 PM Jeroen Dobbelaere < Jeroen.Dobbelaere at

At a high level, EarlyCSE should be intersecting the metadata of instructions that it combines. If it doesn't, and also doesn't drop the metadata, that seems like a bug, regardless of anything else. On 1/22/20 9:30 AM, Jeroen Dobbelaere wrote: Hi Neil, Hall, - as far as 'C' is concerned, this is input code is valid, as the pointers are not used to modify objects. - as far as

So I've been narrowing down a very fun issue in our Burst compiler stack with respect to noalias support, and I've managed to basically boil this down to the following failure (see https://godbolt.org/z/-mdjPV): int called(int* __restrict__ a, int* b, int* c) { return *a + *b + *c; } int foo(int * x, int * y) { return *x + *y + called(x, x, y); } int bar(int * x, int * y) { return

I have a couple of data sets that I want to combine into one data frame. One set contains a number of records on individual observations and includes a geographic descriptor called dacode. The dacode is not unique in that table. The other table contains a number of socio-economic variables for each of the geographic areas identified in the other table. This second table also includes a variable

Hi Tobi, it should just work, but depends on the OS version. ./configure ?help tells you all the configure options, including: --with-maxminddb-includedir path to maxminddb include directory [default=auto] --with-maxminddb-libdir path to maxminddb library directory [default=auto] Neil > On 14 May 2019, at 17:44, Tobi via dovecot <dovecot at dovecot.org>

Hi Tobi, This looks like you haven?t included the libmaxmind libraries before running configure. GeoIP support is only compiled in if it finds the right libs. This would be libmaxminddb-dev on Ubuntu for example. Neil >> Hi list >> >> hope it's okay to ask weakforced questions here as well, but I could not >> find a dedicated mailinglist for wforce. >>

Hello: Is it possible to use Samba to send documents to a Microsoft Shared Fax Printer? How would I give it the phone number and other recipient details that normally comes in the popup window when I print to a shared fax printer from Windows? Thanks, Neil -- Neil Aggarwal, (214)986-3533, www.JAMMConsulting.com FREE! Eliminate junk email and reclaim your inbox. Visit

Hello everyone: I tried following the instructions on this page to set up a VNC server: http://wiki.centos.org/HowTos/VNC-Server I can telnet to port 5901 from the machine itself: # telnet 127.0.0.1 5901 Trying 127.0.0.1... Connected to 127.0.0.1. Escape character is '^]'. RFB 003.008 But, when I try to connect to it from the outside world, I get a connection timeout. My

Hello: I have a brother HL-1440 laser printer set-up and working fine on my RedHat 7.3 Linux box. The printer queue is named brotherLaser. I added a samba share for the printer with these options in my smb.conf file: [brotherLaser] path = /var/spool/lpd/brotherLaser writable = yes comment = Brother Laser Printer printable = yes public = yes I then went

Aki Tuomi <aki.tuomi at dovecot.fi> writes: >> On 24 February 2018 at 15:47 Neil Jerram <neil at ossau.homelinux.net> wrote: [...] >> > Feb 24 10:24:24 arudy dovecot[1712]: imap(neil): Warning: >> > Transaction log file >> > /home/neil/dovecot-mail/.imap/INBOX/dovecot.index.log was locked >> > for 98 seconds (Mailbox was synchronized) [...]

I have an app that has the following in the routes file: namespace "admin" do # ADMINISTRATIVE ROUTES ONLY root :to => ''home#index'' resources :comments do member do get :approve get :reject end end resources :users do member do get :block get :unblock end end end When browing to

Hello: Does anyone know a free or low cost online backup system for CentOS? Thanks, Neil -- Neil Aggarwal, (281)846-8957, http://www.JAMMConsulting.com CentOS 5.4 KVM VPS $55/mo, no setup fee, no contract, dedicated 64bit CPU 1GB dedicated RAM, 40GB RAID storage, 500GB/mo premium BW, Zero downtime

Hello: I have a Supermicro SuperServer 5015P-TR. It has a Pentium D and is running CentOS x86_64. Can this machine use 8GB RAM? I think it should because it is running a 64 bit OS, but I am concerned because it is fundamentally 32 bit hardware. Anyone have any experience in this area? Thanks, Neil -- Neil Aggarwal, (832)245-7314, www.JAMMConsulting.com Eliminate junk email and reclaim your

Aki Tuomi <aki.tuomi at dovecot.fi> writes: > On 24 February 2018 at 12:45 Neil Jerram < neil at ossau.homelinux.net> wrote: > > Please could you help me to understand and fix why my dovecot IMAP > performance is so bad? I've read through a lot of the > performance-related material on the website, but I don't think that any > of it could account for