search for: heroics

I am a professor at UC Santa Cruz, but I also do consulting a Huawei. Chris Lattner told me that I should post this in the llvm-dev. HiSilicon (Santa Clara office) is looking for some developer capable of implementing the "heroic optimizations" (http://llvm.org/devmtg/2015-10/slides/Gerolf-PerformanceImprovementsAndHeadroom.pdf) in LLVM. Focus on SPEC2006 but also looking at the

Hi Tobias- The loop fusion you mention is the one in libquantum/cpu2006 ? Or something else in cpu2017 ? -Thx Dibyendu -----Original Message----- From: llvm-dev [mailto:llvm-dev-bounces at lists.llvm.org] On Behalf Of Tobias Grosser via llvm-dev Sent: Wednesday, August 16, 2017 10:10 AM To: renau at uncore.io; llvm-dev at lists.llvm.org Subject: Re: [llvm-dev] Heroic LLVM optimizations Hi

I'll be interested in seeing the improvements. As a reference, this is what I get in an Intel 6700K when I compare gcc 5.4 (Ofast flto) vs published Intel results. 23x in libquantum, and over 40% in many benchmarks. I think that it is mostly from AoS vs SoA and loop transformations. 5.4

...> https://lkml.org/lkml/2018/8/2/322 > > > > In particular the OOM notifier is worse than the shrinker because: > > The issue is that DEFLATE_ON_OOM is under-specified. > > > > > 1. It is last-resort, which means the system has already gone through > > heroics to prevent OOM. Those heroic reclaim efforts are expensive > > and impact application performance. > > That's *exactly* what "deflate on OOM" suggests. > It seems there are some use cases where "deflate on OOM" is desired and others where "deflate on...

...> https://lkml.org/lkml/2018/8/2/322 > > > > In particular the OOM notifier is worse than the shrinker because: > > The issue is that DEFLATE_ON_OOM is under-specified. > > > > > 1. It is last-resort, which means the system has already gone through > > heroics to prevent OOM. Those heroic reclaim efforts are expensive > > and impact application performance. > > That's *exactly* what "deflate on OOM" suggests. > It seems there are some use cases where "deflate on OOM" is desired and others where "deflate on...

There were apparently good reasons for moving away from OOM notifier callback: https://lkml.org/lkml/2018/7/12/314 https://lkml.org/lkml/2018/8/2/322 In particular the OOM notifier is worse than the shrinker because: 1. It is last-resort, which means the system has already gone through heroics to prevent OOM. Those heroic reclaim efforts are expensive and impact application performance. 2. It lacks understanding of NUMA or other OOM constraints. 3. It has a higher potential for bugs due to the subtlety of the callback context. Given the above, I think the shrinker API certai...

There were apparently good reasons for moving away from OOM notifier callback: https://lkml.org/lkml/2018/7/12/314 https://lkml.org/lkml/2018/8/2/322 In particular the OOM notifier is worse than the shrinker because: 1. It is last-resort, which means the system has already gone through heroics to prevent OOM. Those heroic reclaim efforts are expensive and impact application performance. 2. It lacks understanding of NUMA or other OOM constraints. 3. It has a higher potential for bugs due to the subtlety of the callback context. Given the above, I think the shrinker API certai...

In the capi.conf file, there is a bridge option that allow to "native bridging (CAPI line interconnect) if available", and I found this in the capi-user mailing list : -------------------------------- I suggest you put bridge=yes into each interface. Then, when Asterisk bridges two channels, it looks for the possibility to do a native bridge (call the bridge code of the channel

...> In particular the OOM notifier is worse than the shrinker because: > > > > The issue is that DEFLATE_ON_OOM is under-specified. > > > > > > > > 1. It is last-resort, which means the system has already gone > through > > > heroics to prevent OOM. Those heroic reclaim efforts are > expensive > > > and impact application performance. > > > > That's *exactly* what "deflate on OOM" suggests. > > > > > > It seems there are some use cases where "deflate on OO...

...> In particular the OOM notifier is worse than the shrinker because: > > > > The issue is that DEFLATE_ON_OOM is under-specified. > > > > > > > > 1. It is last-resort, which means the system has already gone > through > > > heroics to prevent OOM. Those heroic reclaim efforts are > expensive > > > and impact application performance. > > > > That's *exactly* what "deflate on OOM" suggests. > > > > > > It seems there are some use cases where "deflate on OO...

On May 3, 2011, at 4:08 PM, David A. Greene wrote: >> >> It's just that an REX prefix is required on some instructions when >> %xmm8 is used. Is it worth it to undo LICM just for that? In this >> case, probably. In general, no. > > Ah, so you're saying the regression is due to the inner loop icache > footprint increasing. Ok, that makes total sense to

> Function pointers are where things get fun. To do these, we need to > determine at run time whether we need to call the ISR or the mainline > version of a function This sounds convenient but it may well be overkill. On a PIC-class platform we can probably consider it to be a design flaw if the programmer doesn't know whether a function pointer will be dereferenced from interrupt

On Jul 29, 2019, at 1:30 PM, Michael Kruse <llvmdev at meinersbur.de> wrote: >> Have you been following what is happening on the MLIR side of the world? It directly supports multi-dimensional arrays with both value and buffer semantics (e.g. load, store, and DMA operations). It is specifically focused on solving these sorts of problems, and is the current proposed direction for the

On Nov 18, 2015, at 12:16 PM, Chris Adams <linux at cmadams.net> wrote: > > Once upon a time, Warren Young <wyml at etr-usa.com> said: >> - They?re serious server-grade machines, not borderline flimsy boxes competing largely on price. Built in and supported from Silicon Valley, not China. :) > > iXsystems sells rebadged SuperMicro stuff, nothing special (not made

...https://lkml.org/lkml/2018/7/12/314 > https://lkml.org/lkml/2018/8/2/322 > > In particular the OOM notifier is worse than the shrinker because: The issue is that DEFLATE_ON_OOM is under-specified. > > 1. It is last-resort, which means the system has already gone through > heroics to prevent OOM. Those heroic reclaim efforts are expensive > and impact application performance. That's *exactly* what "deflate on OOM" suggests. Assume you are using virtio-balloon for some weird way of memory hotunplug (which is what some people do) and you want to minimize...

.../2018/8/2/322 > > > > In particular the OOM notifier is worse than the shrinker because: > > The issue is that DEFLATE_ON_OOM is under-specified. > > > > >? 1. It is last-resort, which means the system has already gone through > >? ? ?heroics to prevent OOM. Those heroic reclaim efforts are expensive > >? ? ?and impact application performance. > > That's *exactly* what "deflate on OOM" suggests. > > > It seems there are some use cases where "deflate on OOM" is desired and > othe...

Hello first, sorry for my bad english but i'm french I'm using totaly Ubuntu Dapper for 4 months, and Wine 0.9.9 I would like to submit you a very usefull software for windows and who doesn't work with Wine : it's Http File Server ( or HFS for close friends :D ) The HFS developer knows that HFS doesn't work with Wine, but doesn't know why, so we need your help :D A

Hello, I need to detect both fax and answer machine, and it should be valuable that the detection will be run by the Diva card itself. So : - I read Diva Documentation, and I found that the Diva could send some specific DTMF, if I had "[..] enabled [this functionnality] by the application for a designated controller through a manufacturer request command 9 [...]", but I didn't

Hi all, I have set up Asterisk and I am able to register with my SIP provider and receive calls. When I try to register with Asterisk from outside I can place calls but tthe other person can't hear me. Have opened port 5060 UDP as well as port 10000 to 20000 UDP. Any ideas? Thanks, Christian

Looks like the Wiki got spammed. Dan