Oliver Neukum wrote: >>(Using SMP_CACHE_BYTES I guess, though it's not clear to me that >>synchronizing with other CPUs would have the same requirement as >>synchronizing to various devices.) Might be good to stick such >>data at the end of structures, since when they're kmalloced in >>cacheline-aligned chunks, aligning the start of such buffers is >>sufficient to prevent any cacheline sharing. > > > Does this mean that gcc will _not_ pad the end of such a member > of a structure to ensure that the next one will not share a cache line, > if an ordinary alignment directive is issued ?
When I saw what "info gcc" said about __aligned__ variable attributes, it talked about aligning the _start_ of the variable/member/etc. Not the end ... that'd be controlled by the next variable/member. See my other post ... start of next structure is controlled by a structure level alignment attribute (forcing extra padding at end), while start of next _member_ is controlled by member level attributes. - Dave _______________________________________________________________ Don't miss the 2002 Sprint PCS Application Developer's Conference August 25-28 in Las Vegas - http://devcon.sprintpcs.com/adp/index.cfm?source=osdntextlink _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
