Hi, So, I owe you all an apology. I was totally unaware that the llvm.org server, running at UIUC, was still in operation and hosting an active LNT instance. I was under the assumption the LNT server ran elsewhere. As a result, the ability for the LNT bots to submit run information to the "old" LNT server is now gone. I am very sorry for this oversight. In speaking with Chris Matthews this afternoon he suggested we ask all of you if you would care to redirect your LNT bots to the new LNT instance currently running at https://lnt.llvm.org. In the meantime, I am going to work to get a dump of the "old" LNT database off of the UIUC machine and see if I can make it accessible somewhere. Chris mentioned, if the historical data is of huge importance to anyone, we could most likely get it surfaced publicly again, though this will take some time to do. Again, I am very sorry for the oversight here. I was quite focused on the SVN move and I just totally missed this. Respectfully, Mike Edwards -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20170627/8f777075/attachment.html>
On 06/27/2017 04:52 PM, Mike Edwards via llvm-dev wrote:> Hi, > So, I owe you all an apology. I was totally unaware that the llvm.org > <http://llvm.org> server, running at UIUC, was still in operation and > hosting an active LNT instance. I was under the assumption the LNT > server ran elsewhere. As a result, the ability for the LNT bots to > submit run information to the "old" LNT server is now gone. I am very > sorry for this oversight. > > In speaking with Chris Matthews this afternoon he suggested we ask all > of you if you would care to redirect your LNT bots to the new LNT > instance currently running at https://lnt.llvm.org. In the meantime, I > am going to work to get a dump of the "old" LNT database off of the > UIUC machine and see if I can make it accessible somewhere. Chris > mentioned, if the historical data is of huge importance to anyone, we > could most likely get it surfaced publicly again, though this will > take some time to do.I think it is preferable for all of our public lnt data to go to the same instance. The historical data seems useful, although the other server was configured with an old baseline that made the data hard to interpret. That having been said, the server at lnt.llvm.org also shows some pretty bad regressions against the baseline it uses (e.g. http://lnt.llvm.org/db_default/v4/nts/graph?plot.0=1335.1604618.3&highlight_run=56388 - perf of uudecode -- although it is not clear to me what would have caused that around that revision, and if it was some hardware/config change, we need to be careful about resetting the baselines to compensate). -Hal> > Again, I am very sorry for the oversight here. I was quite focused on > the SVN move and I just totally missed this. > > Respectfully, > Mike Edwards > > > _______________________________________________ > LLVM Developers mailing list > llvm-dev at lists.llvm.org > http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev-- Hal Finkel Lead, Compiler Technology and Programming Languages Leadership Computing Facility Argonne National Laboratory -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20170627/6a7e7038/attachment.html>
On Tue, Jun 27, 2017, at 11:52 PM, Mike Edwards via llvm-dev wrote:> Hi, > So, I owe you all an apology. I was totally unaware that the llvm.org > server, running at UIUC, was still in operation and hosting an active LNT > instance. I was under the assumption the LNT server ran elsewhere. As a > result, the ability for the LNT bots to submit run information to the > "old" > LNT server is now gone. I am very sorry for this oversight. > > In speaking with Chris Matthews this afternoon he suggested we ask all of > you if you would care to redirect your LNT bots to the new LNT instance > currently running at https://lnt.llvm.org. In the meantime, I am going > to > work to get a dump of the "old" LNT database off of the UIUC machine and > see if I can make it accessible somewhere. Chris mentioned, if the > historical data is of huge importance to anyone, we could most likely get > it surfaced publicly again, though this will take some time to do. > > Again, I am very sorry for the oversight here. I was quite focused on > the > SVN move and I just totally missed this.I just switched my perf servers (and Renatos). @Galina, can you restart the buildbots tonight? Best, Tobias
> On 28 Jun 2017, at 09:22, Tobias Grosser <tobias.grosser at inf.ethz.ch> wrote: > > On Tue, Jun 27, 2017, at 11:52 PM, Mike Edwards via llvm-dev wrote: >> Hi, >> So, I owe you all an apology. I was totally unaware that the llvm.org >> server, running at UIUC, was still in operation and hosting an active LNT >> instance. I was under the assumption the LNT server ran elsewhere. As a >> result, the ability for the LNT bots to submit run information to the >> "old" >> LNT server is now gone. I am very sorry for this oversight. >> >> In speaking with Chris Matthews this afternoon he suggested we ask all of >> you if you would care to redirect your LNT bots to the new LNT instance >> currently running at https://lnt.llvm.org. In the meantime, I am going >> to >> work to get a dump of the "old" LNT database off of the UIUC machine and >> see if I can make it accessible somewhere. Chris mentioned, if the >> historical data is of huge importance to anyone, we could most likely get >> it surfaced publicly again, though this will take some time to do. >> >> Again, I am very sorry for the oversight here. I was quite focused on >> the >> SVN move and I just totally missed this. > > I just switched my perf servers (and Renatos). > > @Galina, can you restart the buildbots tonight? > > Best, > TobiasAnd I also just switched my perf bot, which is driven by a separate buildbot master. It should hopefully submit its first results in a couple of hours to the new address. I think this means all the existing public perf bots have been switched over. Thanks! Kristof
On 28 June 2017 at 08:22, Tobias Grosser <tobias.grosser at inf.ethz.ch> wrote:> I just switched my perf servers (and Renatos).Thanks Tobi! :)
Hi Galina, FYI I took this opportunity to also add a new builder for GlobalISel on AArch64. It has been behaving very well in our internal buildmaster for more than a week. Thanks, Diana On 28 June 2017 at 09:22, Tobias Grosser via llvm-dev <llvm-dev at lists.llvm.org> wrote:> On Tue, Jun 27, 2017, at 11:52 PM, Mike Edwards via llvm-dev wrote: >> Hi, >> So, I owe you all an apology. I was totally unaware that the llvm.org >> server, running at UIUC, was still in operation and hosting an active LNT >> instance. I was under the assumption the LNT server ran elsewhere. As a >> result, the ability for the LNT bots to submit run information to the >> "old" >> LNT server is now gone. I am very sorry for this oversight. >> >> In speaking with Chris Matthews this afternoon he suggested we ask all of >> you if you would care to redirect your LNT bots to the new LNT instance >> currently running at https://lnt.llvm.org. In the meantime, I am going >> to >> work to get a dump of the "old" LNT database off of the UIUC machine and >> see if I can make it accessible somewhere. Chris mentioned, if the >> historical data is of huge importance to anyone, we could most likely get >> it surfaced publicly again, though this will take some time to do. >> >> Again, I am very sorry for the oversight here. I was quite focused on >> the >> SVN move and I just totally missed this. > > I just switched my perf servers (and Renatos). > > @Galina, can you restart the buildbots tonight? > > Best, > Tobias > _______________________________________________ > LLVM Developers mailing list > llvm-dev at lists.llvm.org > http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev