From: Dan S. <da...@sh...> - 2004-01-26 12:45:22
|
--- umid.c.orig 2004-01-26 23:10:38.000000000 +1030 +++ umid.c 2004-01-26 23:11:41.000000000 +1030 @@ -205,7 +205,7 @@ printf("Failed to malloc uml_dir - error = %d\n", errno); uml_dir = name; - return(0); + return(1); } sprintf(uml_dir, "%s/", name); } (and the sf list seems to have nearly completely stopped over the last 36 hours.) -- Dan Shearer da...@sh... |
From: Jeff D. <jd...@ad...> - 2004-01-28 01:22:57
|
da...@sh... said: > - return(0); > + return(1); I did the return(0) on purpose because, even though initcall procedures are declared as returning an int, do_initcalls ignores it. So, I added a comment there instead of changing the return value. > In addition, umid.c seems to be a bit inconsistent as to what rates an > exit(1) and what just does return(1). Stuff that's done early, before the kernel is actually running, rates an exit(1). Everything else gets a return(1). I don't always follow this rule, but I did a quick check of umid.c and it looks OK. Jeff Jeff |
From: Dan S. <da...@sh...> - 2004-01-26 13:01:32
|
In addition, umid.c seems to be a bit inconsistent as to what rates an exit(1) and what just does return(1). On Mon, Jan 26, 2004 at 11:17:08PM +1030, Dan Shearer wrote: > --- umid.c.orig 2004-01-26 23:10:38.000000000 +1030 > +++ umid.c 2004-01-26 23:11:41.000000000 +1030 > @@ -205,7 +205,7 @@ > printf("Failed to malloc uml_dir - error = %d\n", > errno); > uml_dir = name; > - return(0); > + return(1); > } > sprintf(uml_dir, "%s/", name); > } -- Dan Shearer da...@sh... |