<div dir="ltr"><br><br><div class="gmail_quote"><div dir="ltr">On Thu, Jun 4, 2015 at 4:09 PM Alexei Starovoitov <<a href="mailto:alexei.starovoitov@gmail.com">alexei.starovoitov@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">On Thu, Jun 4, 2015 at 3:52 PM, Eric Christopher <<a href="mailto:echristo@gmail.com" target="_blank">echristo@gmail.com</a>> wrote:<br>
> You were correct in your earlier assumption, this shouldn't have been added<br>
> by default yet.<br>
><br>
> I'm going to go ahead and revert this patch in a bit unless someone tells me<br>
> not to bother, and you can propose your patch to do, essentially, the same<br>
> thing with the rationale from this message if you'd like.<br>
<br>
of course. consider it proposed.<br>
llvmdev was cc-ed already.<br>
<br></blockquote><div><br></div><div>Probably best to have a new email with a subject of "Make BPF non-experimental" or some such to make it more obvious to people reading the list.</div><div><br></div><div>Thanks and sorry for the extra work here :)</div><div><br></div><div>-eric</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
> FWIW I'm not against it, you've been an active maintainer and the BPF port<br>
> hasn't been terrible to update (from looking at the patches) or keep updated<br>
> (in the few occasions I've done it).<br>
<br>
Great.<br>
<br>
> -eric<br>
><br>
> On Thu, Jun 4, 2015 at 2:38 PM Alexei Starovoitov<br>
> <<a href="mailto:alexei.starovoitov@gmail.com" target="_blank">alexei.starovoitov@gmail.com</a>> wrote:<br>
>><br>
>> On Thu, Jun 4, 2015 at 5:51 AM, Daniel Sanders<br>
>> <<a href="mailto:daniel.sanders@imgtec.com" target="_blank">daniel.sanders@imgtec.com</a>> wrote:<br>
>> > Author: dsanders<br>
>> > Date: Thu Jun 4 07:51:20 2015<br>
>> > New Revision: 239035<br>
>> ><br>
>> > Include BPF target in CMake builds.<br>
>> ><br>
>> > Modified:<br>
>> > llvm/trunk/CMakeLists.txt<br>
>> ><br>
>> > --- llvm/trunk/CMakeLists.txt (original)<br>
>> > +++ llvm/trunk/CMakeLists.txt Thu Jun 4 07:51:20 2015<br>
>> > @@ -169,6 +169,7 @@ set(LLVM_INCLUDE_DIR ${CMAKE_CURRENT_BIN<br>
>> > set(LLVM_ALL_TARGETS<br>
>> > AArch64<br>
>> > ARM<br>
>> > + BPF<br>
>> > CppBackend<br>
>><br>
>> wow :)<br>
>> this is what I wanted to propose for the last few months.<br>
>> I had a chat with Chandler about graduating BPF backend<br>
>> from experimental, but I want to make sure everyone is ok<br>
>> before proceeding.<br>
>> Current status of it:<br>
>> - running on x64, arm64, s390 architectures<br>
>> - projects that use it: perf, tc, ovs<br>
>> - there are few front-ends in the works:<br>
>> one is translating language X to C and then using clang/llvm,<br>
>> another is generating llvm IR directly,<br>
>> yet another is using clang rewriter to augment C language<br>
>> for tracing/networking needs.<br>
>> one project tried to hack clang overall and was abandoned.<br>
>> - I've seen people embedding binary llc with bpf support<br>
>> in their github projects.<br>
>> - broken llvm build with bpf backend was reported many<br>
>> times with few folks even provided simple patches to unbreak it.<br>
>> - we've started to prepare buildbot specific to bpf, but hit<br>
>> configuration issues with zorg. Hopefully it will be functional soon.<br>
>> - as far as I can see all across the board api refactoring that<br>
>> was done by a bunch of people over the last 5 month<br>
>> didn't cause any problems for them or for the backend<br>
>><br>
>> bpf backend has been in-tree since January and has enough<br>
>> users, so I think it's time to graduate it from experimental.<br>
>> Honestly I was surprised to see this patch... I was hoping<br>
>> it will be my honors to do it ;)<br>
</blockquote></div></div>