Re: PR25394 patch: debuginfod groom vs. scan race condition

2020-01-20 Thread Mark Wielaard
Hi Frank, On Mon, 2020-01-20 at 12:10 -0500, Frank Ch. Eigler wrote: > > Nice work. And I think I almost understand it. I am just a little > > confused about thread_main_fts_source_paths also being an "idler": > > Yup, it's a bit subtle. > > > > while (! interrupted) > > > { > > > -

Re: PR25394 patch: debuginfod groom vs. scan race condition

2020-01-20 Thread Frank Ch. Eigler
Hi - > Nice work. And I think I almost understand it. I am just a little > confused about thread_main_fts_source_paths also being an "idler": Yup, it's a bit subtle. > > while (! interrupted) > > { > > - set_metric("thread_timer", "role","traverse", rescan_timer); > > - // set_

Re: PR25394 patch: debuginfod groom vs. scan race condition

2020-01-20 Thread Mark Wielaard
Hi Frank, On Sun, 2020-01-19 at 20:43 -0500, Frank Ch. Eigler wrote: > This was a doozie. As the commit text hints, I haven't found a way > of > testing this on an elfutils/tests snack-sized data set (except perhaps > by injecting long sleeps?), but the changes work on the larger > archives such

PR25394 patch: debuginfod groom vs. scan race condition

2020-01-19 Thread Frank Ch. Eigler
Hi - This was a doozie. As the commit text hints, I haven't found a way of testing this on an elfutils/tests snack-sized data set (except perhaps by injecting long sleeps?), but the changes work on the larger archives such as the debuginfod.systemtap.org set. commit 34e67018914cf9ebbef070659657