On Fri, Oct 21, 2011 at 2:11 PM, John David Anglin <[email protected]> wrote: > On 10/21/2011 1:42 PM, John David Anglin wrote: >> >> Yes. The issue here is expand_call doesn't appear to be aware that >> loading the >> address of a __thread variable may require a call. Thus, it just emits an >> insn to >> load r26. This load is dead because r26 is reloaded for the call to >> __tls_get_addr. > > It appears this may be caused by treating TLS symbols as legitimate > constants.
Which they are not in any of the *_DYNAMIC modes, because the dynamic linker is involved in their placement. Cheers, Carlos. -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected] Archive: http://lists.debian.org/cadzpyizjwcefnlzyhd74wdyfvrgnpd6j+2crhjr7zkl0xtw...@mail.gmail.com
