diff options
author | Stefan Richter <stefanr@s5r6.in-berlin.de> | 2010-06-20 22:52:55 +0200 |
---|---|---|
committer | Stefan Richter <stefanr@s5r6.in-berlin.de> | 2010-06-20 23:11:56 +0200 |
commit | 604f45167824e18ad5766e51ecf1d4d65f15118d (patch) | |
tree | 40deeca88e1788e98d54d82f429a45ce276c4135 /include/linux/firewire-cdev.h | |
parent | 0244f57302f7e8bebd2f1ab58767eac2e9f678a6 (diff) | |
download | blackbird-obmc-linux-604f45167824e18ad5766e51ecf1d4d65f15118d.tar.gz blackbird-obmc-linux-604f45167824e18ad5766e51ecf1d4d65f15118d.zip |
firewire: cdev: freeze FW_CDEV_VERSION due to libraw1394 bug
libraw1394 v2.0.0...v2.0.5 takes FW_CDEV_VERSION from an externally
installed header file and uses it to declare its own implementation
level in FW_CDEV_IOC_GET_INFO. This is wrong; it should set the real
version for which it was actually written.
If we add features to the kernel ABI that require the kernel to check
a client's implementation level, we can not trust the client version if
it was set from FW_CDEV_VERSION.
Hence freeze FW_CDEV_VERSION at the current value (no damage has been
done yet), clearly document FW_CDEV_VERSION as a dummy version and what
clients are expected to do with fw_cdev_get_info.version, and use a new
defined constant (which is not placed into the exported header file) as
kernel implementation level.
Note, in order to check in client program source code which features are
present in an externally installed linux/firewire-cdev.h, use
preprocessor directives like
#ifdef FW_CDEV_IOC_ALLOCATE_ISO_RESOURCE
or
#ifdef FW_CDEV_EVENT_ISO_RESOURCE_ALLOCATED
instead of a check of FW_CDEV_VERSION.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Diffstat (limited to 'include/linux/firewire-cdev.h')
-rw-r--r-- | include/linux/firewire-cdev.h | 22 |
1 files changed, 16 insertions, 6 deletions
diff --git a/include/linux/firewire-cdev.h b/include/linux/firewire-cdev.h index 6ffb24a1f2f2..0d0cc07358af 100644 --- a/include/linux/firewire-cdev.h +++ b/include/linux/firewire-cdev.h @@ -219,7 +219,7 @@ union fw_cdev_event { struct fw_cdev_event_response response; struct fw_cdev_event_request request; struct fw_cdev_event_iso_interrupt iso_interrupt; - struct fw_cdev_event_iso_resource iso_resource; + struct fw_cdev_event_iso_resource iso_resource; /* added in 2.6.30 */ }; /* available since kernel version 2.6.22 */ @@ -252,22 +252,32 @@ union fw_cdev_event { #define FW_CDEV_IOC_GET_CYCLE_TIMER2 _IOWR('#', 0x14, struct fw_cdev_get_cycle_timer2) /* - * FW_CDEV_VERSION History + * ABI version history * 1 (2.6.22) - initial version + * (2.6.24) - added %FW_CDEV_IOC_GET_CYCLE_TIMER * 2 (2.6.30) - changed &fw_cdev_event_iso_interrupt.header if * &fw_cdev_create_iso_context.header_size is 8 or more + * - added %FW_CDEV_IOC_*_ISO_RESOURCE*, + * %FW_CDEV_IOC_GET_SPEED, %FW_CDEV_IOC_SEND_BROADCAST_REQUEST, + * %FW_CDEV_IOC_SEND_STREAM_PACKET * (2.6.32) - added time stamp to xmit &fw_cdev_event_iso_interrupt * (2.6.33) - IR has always packet-per-buffer semantics now, not one of * dual-buffer or packet-per-buffer depending on hardware * 3 (2.6.34) - made &fw_cdev_get_cycle_timer reliable + * - added %FW_CDEV_IOC_GET_CYCLE_TIMER2 */ -#define FW_CDEV_VERSION 3 +#define FW_CDEV_VERSION 3 /* Meaningless; don't use this macro. */ /** * struct fw_cdev_get_info - General purpose information ioctl - * @version: The version field is just a running serial number. - * We never break backwards compatibility, but may add more - * structs and ioctls in later revisions. + * @version: The version field is just a running serial number. Both an + * input parameter (ABI version implemented by the client) and + * output parameter (ABI version implemented by the kernel). + * A client must not fill in an %FW_CDEV_VERSION defined from an + * included kernel header file but the actual version for which + * the client was implemented. This is necessary for forward + * compatibility. We never break backwards compatibility, but + * may add more structs, events, and ioctls in later revisions. * @rom_length: If @rom is non-zero, at most rom_length bytes of configuration * ROM will be copied into that user space address. In either * case, @rom_length is updated with the actual length of the |