In this case however grepping for "alignmentto" won't help. (cc @risc_v there is a blank missing in bfd/elf32-riscv.c and bfd/elf64-riscv.c)pic.twitter.com/QaIFSN956O
Show this thread
Typos is error messages are a feature. They make it easier to find relevant information via google and/or grep for the right instance of the message in source code.pic.twitter.com/eCCeW2gZjy
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This is why I strongly prefer error messages to be associated with error *codes*, like E4077 or some such. This allows you to search for root causes and solutions much, much easier. This practice fell out of favor about when Unix took off, and we've suffered since.
-
I somewhat agree, though I'd like to see some more advanced yet more user friendly error code somehow. Exactly what, I have no real idea...not enough experience here
(BTW I see we both use more than 140 chars! 
) And DOCUMENTATION!!! - Show replies
New conversation -
-
-
I don't think we need to go so extreme; but, IBM mainframes often use 6- to 8-character codes for each error, warning, and info message, typically in TPPPXXXX format, where T is E, W, or I for error, warning, or info, PPP is a unique program ID, etc.
-
So there is historical precedent for the basic idea at least. :)
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.