similar to: [RFC] Lanai backend

Hi All, I'd like to propose to upgrade BPF backend to first class citizen. Current status of it: - used with linux kernel on x64, arm64, s390 architectures - projects that use it: perf, tc, ovs - there are few front-ends in the works: . one is translating custom language to C and then using clang/llvm . another is generating llvm IR directly . yet another is using clang rewriter to

On Thu, Jun 4, 2015 at 5:51 AM, Daniel Sanders <daniel.sanders at imgtec.com> wrote: > Author: dsanders > Date: Thu Jun 4 07:51:20 2015 > New Revision: 239035 > > Include BPF target in CMake builds. > > Modified: > llvm/trunk/CMakeLists.txt > > --- llvm/trunk/CMakeLists.txt (original) > +++ llvm/trunk/CMakeLists.txt Thu Jun 4 07:51:20 2015 > @@

Hi, Alexei On 2015/7/30 1:13, Alexei Starovoitov wrote: > On 7/29/15 2:38 AM, He Kuang wrote: >> Hi, Alexei >> >> On 2015/7/28 10:18, Alexei Starovoitov wrote: >>> On 7/25/15 3:04 AM, He Kuang wrote: >>>> I noticed that for 64-bit elf format, the reloc sections have >>>> 'Addend' in the entry, but there's no 'Addend' info

Hi everyone, I am working with Alexei Starovoitov to contribute an LLVM buildbot for the experimental BPF backend. I am following the steps at [1] to setup a buildbot and I was mostly successful: I was able to setup a slave and a temporary master to check its base config. Now I'm not sure about the next step: patching the "slaves.py" and "builders.py" files in zorg. * Is

On Wed, Jun 15, 2016 at 2:37 PM, Richard Henderson via iovisor-dev <iovisor-dev at lists.iovisor.org> wrote: > This same value for EM_BPF is being propagated to glibc, > elfutils, and binutils. great! Can you share the link to glibc and the other patches? > diff --git a/include/llvm/Support/ELF.h b/include/llvm/Support/ELF.h > index 352fd8a..fb8ff71 100644 > ---

Thank you. Alexei just committed the change to SVN. Regards, Marco Leogrande Sent by a carbon-based life form; hence, it may contain repetitions, inaccuracies, logical fallacies and repetitions. On Wed, Jun 17, 2015 at 2:52 PM, Galina Kistanova <gkistanova at gmail.com> wrote: > LGTM. > Please commit. > > Thanks > > Galina > > On Tue, Jun 16, 2015 at 8:32 PM,

Galina, thanks again for your help. I have cooked a new patch that should make a bit more sense. I have added the new builder under _get_experimental_scheduled_builders(), as the comment above that function suggested me it would be a good idea. :) Does it look a reasonable start? Regards, Marco Leogrande Sent by a carbon-based life form; hence, it may contain repetitions, inaccuracies,

Hello Marco, Welcome aboard! slaves.py keeps the build slaves definitions, builders.py keeps the builders definitions. You have to have both. The steps of adding a new slave is here: http://llvm.org/docs/HowToAddABuilder.html. Please make sure you done the step # 10 before bringing your slave up, otherwise it wouldn't be authorized by the master and will be blacklisted after multiple

Dear there, I was trying to use clang to cross compile executable directly on my mac to bpf ISA: clang -target bpf-linux-gnu aaa.c But I guess maybe the linker does not support BPF? (with -c clang can generate unlinked objs) Unexpected arch UNREACHABLE executed at llvm/tools/clang/lib/Driver/Tools.cpp:8113 Should I use other linkers to link clang generated BPF objs? Thanks, Xiaochu

Dear there, I wasn't able to generate obj file for bpf using llc: llc -march bpf -filetype=obj foo.bc Anyone knows what is the missing part in their backend? I want to make it work for my project. Thanks, Xiaochu -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20150802/e0b55467/attachment.html>

You've raised an important point here Pete, and while I disagree pretty strongly with it (regardless of whether Lanai makes sense or not), I'm glad that you've surfaced it where we can clearly look at the issue. The idea of "it really should have users outside of just the people who have access to the HW" I think is deeply problematic for the project as a whole. Where does

On 19 July 2016 at 17:04, Martin J. O'Riordan via llvm-dev <llvm-dev at lists.llvm.org> wrote: > Presumably if my out-of-tree backend was to be pushed to LLVM, it too would be considered experimental. Yes. Though, not all out-of-tree back-ends end up upstream for different reasons. A few basic rules to get accepted are if: * the target exists and can be easily purchased / emulated

On Wed, Aug 05, 2015 at 04:28:13PM +0800, Wangnan (F) wrote: > > It doesn't work for me at first since in my llvm there's only > llvm.bpf.load.*. > > I think llvm.bpf.store.* belone to some patches you haven't posted yet? nope. only loads have special instructions ld_abs/ld_ind which are represented by these intrinsics. stores, so far, are done via single

On Tue, Feb 9, 2016 at 10:30 PM Chris Lattner <clattner at apple.com> wrote: > > > On Feb 9, 2016, at 10:24 PM, Chandler Carruth via llvm-dev < > llvm-dev at lists.llvm.org> wrote: > > > > You've raised an important point here Pete, and while I disagree pretty > strongly with it (regardless of whether Lanai makes sense or not), I'm glad > that

On 8/4/15 11:51 PM, Wangnan (F) wrote: > void bpf_store_half(void *skb, int off, int val) > asm("llvm.bpf.store.half"); > int func() > { > bpf_store_half(0, 0, 0); > return 0; > } > > Compiled with: > > $ clang -g -target bpf -O2 -S -c test.c > > And get this: > > .text > .globl func > .align

Send again since llvmdev is moved to llvm-dev at lists.llvm.org On 2015/8/5 9:58, Wangnan (F) wrote: > > > On 2015/8/4 17:01, Wangnan (F) wrote: >> For people who in llvmdev: >> >> This mail is belong to a thread in linux kernel mailing list, the >> first message >> can be retrived from: >> >> http://lkml.kernel.org/r/55B1535E.8090406 at

Hi Alexei, I haven't tried to run code yet. The clang frontend can generate machine code for bpf but llc does not do the job. When i use : llc -march bpf -filetype=obj arith.bc It has this error message: Assertion failed: (isa<X>(Val) && "cast<Ty>() argument of incompatible type!"), function cast, file

On 2015/8/4 3:44, Alexei Starovoitov wrote: [SNIP] >> I'll post 2 LLVM patches by replying this mail. Please have a look and >> help me >> send them to LLVM if you think my code is correct. > > [SNIP] > patch 2: > do we really need to hack clang? > Can you just define a function that aliases to intrinsic, > like we do for ld_abs/ld_ind ? > void

On 7/19/2016 6:12 AM, Renato Golin via llvm-dev wrote: > > I don't see why not. LGTM. Same here. -Krzysztof -- Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation

> On Jul 19, 2016, at 12:57 PM, Pete Cooper via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > Hi Renato >> On Jul 19, 2016, at 9:42 AM, Renato Golin via llvm-dev <llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org>> wrote: >> >> A few basic rules to get accepted are if: >> * the target exists and can be easily purchased / emulated