On Wed, 2006-11-15 at 12:44 +0000, Phil Blundell wrote: > I'm also concerned that, even on the CATS machines, it might be that > the build is only succeeding by chance and any real-world workload > would cause the same bug to resurface. So, overall, with our current > state of knowledge I kind of feel we'd be better off not shipping mono > at all on arm.
Mono bootstraps during the build, and uses a real-world application (workload) during that, which is the C# compiler used to compile the complete Base Class Library of Mono (which is quite big [0]). The C# compiler (mcs) was written in C# and thus runs on the Mono runtime, like every other program also. The points where the build fails is during compiling the BCL. Means it's a runtime problem, also proved by other non-mono (not mono source package, but mono based software) failing build logs. Also the failing happens not always at the same build "position", but its always the first few files that get compiled. The build runs 8 to 10 hours on a arm box, a race condition that always happened in the first hour on netwinder, but not at all on cats, means it's not a runtime issue that seem to ever happen on cats. So I think this issue is a not reason to not ship mono for arm. [0] Total Physical Source Lines of Code (SLOC) = 1,173,190 -- Regards, Mirco 'meebey' Bauer PGP-Key: http://keyserver.noreply.org/pks/lookup?op=get&search=0xEEF946C8 -----BEGIN GEEK CODE BLOCK----- Version: 3.12 GIT d s-:+ a-- C++ UL++++$ P L++$>+++$ E- W+++$ N o? K- w++>! O---- M- V? PS PE+ Y- PGP++ t 5+ X++ R tv+ b+ DI? D+ G>++ e h! r->++ y? ------END GEEK CODE BLOCK------
signature.asc
Description: This is a digitally signed message part
