kupsch

@mplegendre, @vsoch there is now a branch kupsch/parse-callsites-preview-3 that is based on the tip of master. Rebase your branch to it, and the test problems should be fixed

That is correct. AMD recommended using the processor column from https://llvm.org/docs/AMDGPUUsage.html#processors On 7/11/22 3:39 PM, bbiiggppiigg wrote: > @bigtrak > > So every architecture might have variants > . >...

Tony said the official arch names to product names are on this web page: https://llvm.org/docs/AMDGPUUsage.html#processors Jim On 7/11/22 2:54 PM, Bolo -- Josef Burger wrote: > > > > >...

@wcohen, I have made changes to allow Dyninst to successfully build with `-DLIGHTWEIGHT_SYMTAB=YES` (use symlite) on the branch [kupsch/fix-symlite](/dyninst/dyninst/tree/kupsch/fix-symlite). This hasn't been tested as I does play well with the...

The problem is that libdyninstAPI is not built with LIGHTWEIGHT_SYMTAB, which the testsuite uses. Are you using libdyninstAPI? If not, we can look at fixing the testsuite. If you are...

@wchohen, We discussed this issue at our group meeting and symlite is currently not a supported library, and we do not have the resources at this time to bring it...

Couldn't log in, but I now have a copy of it. The files are from https://github.com/mxz297/dyninst-tools.git plus some large (~5GB) result files. We'll have to decide if we need those.

I can do it, but it looks like we're switching to a new doc system, so I'll wait until that that change is complete.

After a bit of investigation, I found the cause of the problem. 40 of the files in parseAPI are in a group called ROSE_SRC and this group adds `-w` to...

I was thinking the same thing on both points.