On Tue, Jun 24, 2008 at 4:19 PM, Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote: >> /* >> + * write_string() is passed a nul-terminated kernelspace >> + * buffer of maximum length determined by max_write_len. >> + * Returns 0 or -ve error code. >> + */ >> + int (*write_string)(struct cgroup *cgrp, struct cftype *cft, >> + const char *buffer); > > Everything seems to use size_t (or ssize_t?) except for the ->write_string > return value. Can any of this be improved? What other things are you including as "everything"? write_string() returns 0 on success or a -ve error code on failure - it doesn't have the concept of writing some fraction of the passed bytes. The functions that deal in size_t/ssize_t (along with userspace buffers, files and position pointers) are the glue that interfaces with the filesystem layer. My aim (which is furthered by this patch series) is to keep as much of that as possible in the cgroup layer itself, and to reduce filesystem glue in the cgroup subsystems. The raw file interface is still exposed by cgroups for those subsystems that really need it, but it should be the exception rather than the rule. Paul _______________________________________________ Containers mailing list Containers@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/containers