First I'd like that add that it's not technically us trying to run this on vintage PIII :-).
So I was able to resolve the issue by passing the /arch:SSE command line parameter. Thanks! On Wednesday, October 5, 2016 at 9:33:06 AM UTC-5, Adam Cozzette wrote: > > That's good to know that you're using protocol buffers on some vintage > hardware! :) I wonder if it's not the code per se that's being miscompiled > for the architecture but perhaps just the serialized descriptor data that's > being linked into the binary. It might be interesting to try saving the > descriptor data to disk from one of the affected machines and then > examining it after the fact to see if it's truncated or somehow otherwise > corrupted. > > On Wed, Oct 5, 2016 at 7:06 AM, Ingmar Koecher <[email protected] > <javascript:>> wrote: > >> Sorry, I forgot to answer your question. Yes, it happens every time. But >> only on Pentium 3 (!) machines. >> >> On Tuesday, October 4, 2016 at 12:37:24 PM UTC-5, Adam Cozzette wrote: >>> >>> It sounds like the static descriptor data is somehow getting corrupted. >>> On the affected systems does it happen reproducibly every time the >>> application starts up? >>> >>> On Mon, Oct 3, 2016 at 10:54 AM, Ingmar Koecher <[email protected]> >>> wrote: >>> >>>> I can now also confirm that this problem only occurs on Windows Server >>>> 2003 x86 hosts. >>>> >>>> >>>> On Monday, October 3, 2016 at 8:48:48 AM UTC-5, Ingmar Koecher wrote: >>>>> >>>>> We've been using protocol buffers for a couple of months (version 2, >>>>> exclusively on Windows) and while it works on 99.9% of systems, on a few >>>>> systems an application won't start with the following errors: >>>>> >>>>> [libprotobuf ERROR google\protobuf\descriptor_database.cc:315] >>>>> Invalid file descriptor data passed to >>>>> EncodedDescriptorDatabase::Add(). >>>>> >>>>> [libprotobuf FATAL >>>>> google\protocol_buffers\src\google\protobuf\descriptor.cc:1018] >>>>> CHECK failed: >>>>> generated_database_->Add(encoded_file_descriptor, size): >>>>> >>>>> Unfortunately I don't have access to those systems, so I can't debug >>>>> it. I've tried to send special instrumented versions to capture the >>>>> output, >>>>> but I haven't been able to figure out what the issue is. >>>>> >>>>> I have a suspicion that it only affects Windows 2003 Servers, but I'm >>>>> not sure. The protocol buffers are statically linked from Visual Studio >>>>> 2010. >>>>> >>>>> Do these error messages ring a bell with anyone? >>>>> >>>>> >>>>> Thanks! >>>>> >>>> -- >>>> You received this message because you are subscribed to the Google >>>> Groups "Protocol Buffers" group. >>>> To unsubscribe from this group and stop receiving emails from it, send >>>> an email to [email protected]. >>>> To post to this group, send email to [email protected]. >>>> Visit this group at https://groups.google.com/group/protobuf. >>>> For more options, visit https://groups.google.com/d/optout. >>>> >>> >>> -- >> You received this message because you are subscribed to the Google Groups >> "Protocol Buffers" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected] <javascript:>. >> To post to this group, send email to [email protected] >> <javascript:>. >> Visit this group at https://groups.google.com/group/protobuf. >> For more options, visit https://groups.google.com/d/optout. >> > > -- You received this message because you are subscribed to the Google Groups "Protocol Buffers" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/protobuf. For more options, visit https://groups.google.com/d/optout.
