Date: Sun, 3 Feb 2019 11:01:54 -0800 From: bob prohaska <fbsd@www.zefox.net> To: Mark Millard <marklmi@yahoo.com> Cc: freebsd-arm@freebsd.org Subject: Re: What is clang error code 254? Message-ID: <20190203190154.GA22224@www.zefox.net> In-Reply-To: <18EBA0BD-B092-42FE-A165-4A26584BC9FF@yahoo.com> References: <18EBA0BD-B092-42FE-A165-4A26584BC9FF@yahoo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Jan 24, 2019 at 11:08:20PM -0800, Mark Millard wrote: > bob prohaska fbsd at www.zefox.net wrote on > Thu Jan 24 18:22:40 UTC 2019 : > > > The pattern of clang crashes during buildworld on RPI3 continues, > > most consistently reporting Error code 254. > > > > I've looked on the web for clang error codes and find nothing. > > Even the llvm.org website isn't helpful. There are references > > to error 254, but nothing seems recent, and no error list is > > mentioned. > > I found it traces to: > > // Return a special value to indicate that the process received an unhandled > // signal during execution as opposed to failing to execute. > WaitResult.ReturnCode = -2; > > in /usr/src/contrib/llvm/lib/Support/Unix/Program.inc . (I did the > trace between that and main by inspection but I'll not list the details > here.) > > -2 treated as an unsigned-byte is 254 for the rpi3 (and most common > FreeBSD machines [all?]). > I gather from all this that error 254 isn't highly informative then.... I want to thank you for pointing out use of WITH_META_MODE and NO_SILENT. They made working through failing buildworlds much easier, but working up the revision chain didn't help the problem. It looks as if backing the revision down to r339376 does help eliminate the segfaults, though the test won't be done for another day or two. Thanks for reading, and all your guidance. bob prohaska
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20190203190154.GA22224>