Re: patch rfc: PR28661: debuginfod thread-pool

2021-12-10 Thread Mark Wielaard
Hi Frank, On Thu, 2021-12-09 at 19:10 -0500, Frank Ch. Eigler via Elfutils-devel wrote: > So I would recommend to simply add a testcase that just uses no- > > option, > > -C and -C 256 (or take any arbitrary number, 1 might be an interesting > > corner case) and see that if you have 4 (8, 16, ...

Re: patch rfc: PR28661: debuginfod thread-pool

2021-12-09 Thread Frank Ch. Eigler via Elfutils-devel
Hi, Mark - > So I would recommend to simply add a testcase that just uses no-option, > -C and -C 256 (or take any arbitrary number, 1 might be an interesting > corner case) and see that if you have 4 (8, 16, ...?) debuginfod-find > (or curl metric) processes doing some parallel queries works as >

Re: patch rfc: PR28661: debuginfod thread-pool

2021-12-09 Thread Frank Ch. Eigler via Elfutils-devel
Hi - > [...] > If you can use ulimit -u or ulimit -T in the run-test.sh script then > please use that, but that probably requires launching sub-shells and > you quickly end up in shell-hell. A problem I found with that is that ulimit -u appears to be systemwide in the sense that a new process/thr

Re: patch rfc: PR28661: debuginfod thread-pool

2021-12-09 Thread Mark Wielaard
Hi Frank, On Wed, 2021-12-08 at 22:55 -0500, Frank Ch. Eigler via Elfutils-devel wrote: > While I think this patch itself is fine, and works around the > libmicrohttpd bug that motivated it, I don't know how to test it > seriously in the testsuite. (We can certainly try few -C options for > parsi

patch rfc: PR28661: debuginfod thread-pool

2021-12-08 Thread Frank Ch. Eigler via Elfutils-devel
Hi - While I think this patch itself is fine, and works around the libmicrohttpd bug that motivated it, I don't know how to test it seriously in the testsuite. (We can certainly try few -C options for parsing & operability.) The error edge cases only appear to occur under very high load and task