On Sat, Aug 31, 2013 at 06:19:14PM -0400, Lidza Louina wrote: > This patch removes the error statements that follow > kzalloc. These are useless because kzalloc has its > own error statement and the driver's error handling > would never actually happen. > Gar! Wait! What? I'm sorry, I was totally unclear on that. These things totally can happen (once every thousand years) and we need to handle them. Don't bother with an error message but do clean up and return the error code. This was my bad. I appologize for setting you on the wrong path. Some allocation failures are easier to trigger than others. We handle them all the same way, with a cleanup and return an error. If we start handling some one way and some another way then it means that we have to sort through which errors are important and which are only theoretical. Sorting is more of a pain than just fixing everything no matter how minor. regards, dan carpenter _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
