Messages in this thread | | | Date | Sat, 10 Jul 2010 00:02:52 +0200 | From | Sam Ravnborg <> | Subject | Re: [PATCH][GIT PULL][for 2.6.35] tracing: Add alignment to syscall metadata declarations |
| |
On Fri, Jul 09, 2010 at 05:05:50PM -0400, Steven Rostedt wrote: > On Fri, 2010-07-09 at 22:53 +0200, Sam Ravnborg wrote: > > > But in this case we have no control of the value of "." (current address) > > when we have processed (_ftrace_events) so it may even be at a 2 byte boundary. > > The linker will add padding as needed to satisfy the alignmnet of > > __syscalls_metadata - but that padding will be inbetween "." and the first > > member in __syscalls_metadata. > > Fine, but this is a separate issue. I doubt the "ALIGN(8);" would have > helped us anyway. Remember what the issue we had: > > ffffffff8173c438 <__start_syscalls_metadata>: > ... > > ffffffff8173c440 <__syscall_meta__mmap>: > > > __start_syscalls_metadata was already aligned to 8, but for some strange > reason, gcc decided to align the first member to 16.
I found some more info in following bug: https://bugzillafiles.novell.org/attachment.cgi?id=344563
Jeff Mahoney says: GCC 4.5 introduced behavior that forces the alignment of structures to use the largest possible value. The default value is 32 bytes, so if some structures are defined with a 4-byte alignment and others aren't declared with an alignment constraint at all - it will align at 32-bytes.
So according to this we can rely on 32 byte alignment and this is trivial to add in vmlinux.lds.h.
Sam
| |