Re: [PATCH 17/19] ceph: ioctls

[Sage Weil <sage@newdream.net>]

On Thu, 23 Jul 2009, Andi Kleen wrote:
> What I find more suspicious is that it's the direct network
> data structure. Presumably that's in a defined endian?
> So that means the application would already need to 
> change to network endian order at the ioctl level?
> 
> That seems wrong if true. I think defining
> a ioctl directly based on a network protocol header is 
> likely a bad idea.

That's a good point.  I've defined an ioctl-specific struct, so the entire 
interface is defined in the header:

---
#include <linux/ioctl.h>
#include <linux/types.h>

#define CEPH_IOCTL_MAGIC 0x97

/* just use u64 to align sanely on all archs */
struct ceph_ioctl_layout {
	__u64 stripe_unit, stripe_count, object_size;
	__u64 data_pool;
};

#define CEPH_IOC_GET_LAYOUT _IOR(CEPH_IOCTL_MAGIC, 1,		\
				   struct ceph_ioctl_layout)
#define CEPH_IOC_SET_LAYOUT _IOW(CEPH_IOCTL_MAGIC, 2,		\
				   struct ceph_ioctl_layout)
---

The remaining issues are:
- this isn't complete.  I haven't made up my mind how to list, describe, 
  and choose the data pool to store a file in; for now it's ignored.
- where the header should go.  I don't see any other file system ioctls in 
  include/linux other than ext[23]; the rest are all in fs/*.

sage