On Wed, Apr 12, 2000 at 10:22:59PM -0500, Jim Studt wrote: > Brendan Simon. wrote > > I am a big PowerPC fan but I would like to know the reasoning behind the > > Linux/PPC people choosing unsigned char as the default char type. Why go > > against the convention ? I hope there is a really good reason that will > > help > > progress the C programming environment to better things. > > I'm told for ARM it is a performance issue. I've handled dozens of > these on the ARM port. In almost all of them it is a getc/getchar > that is being coerced down to a char before being checked agains EOF. > > getc and getchar are (int) functions, EOF is an int. You can't cast > down to a char until you know you have a character and not an EOF. > > Code with downcasts before checking EOF will fail one way on PPC and ARM > and the opposite way anytime it encounters a (char)255 in its input > stream. > > Fortunately some wise soul extended the ARM compiler to print a warning > when a `char' is compared to a negative constant. We collect these during > the builds and list them over at... > http://www.federated.com/~jim/arm/charhell.html > ... there are only about 150 packages left with this problem. > There are still packages like gnome-pim that can't be detected by the > compiler, but anytime a program hangs or fails to read a file this is > the first thing to look for on the ARM port. > > I'll write a little script to blow all the compiler detected problems > into the BTS. We'll see how many we can get cleaned up.
*applause* I have my build daemon filter these into a separate mailbox; I've got about 90 of them listed right now. I won't post the list since it's a subset of yours. Thanks for the excellent description of the problem! Dan /--------------------------------\ /--------------------------------\ | Daniel Jacobowitz |__| SCS Class of 2002 | | Debian GNU/Linux Developer __ Carnegie Mellon University | | [EMAIL PROTECTED] | | [EMAIL PROTECTED] | \--------------------------------/ \--------------------------------/
