On 11 July 2011 12:30, Dave Martin <dave.mar...@linaro.org> wrote: > On Mon, Jul 11, 2011 at 10:42:27AM +0100, Richard Sandiford wrote: >> Dave Martin <dave.mar...@linaro.org> writes: >> > IFUNC doesn't solve the problem because either it gets resolved >> > lazily (violating the above principle (*)), or we have to force _all_ >> > symbols to resolve at startup, with may have a significant impact on >> > startup time for large programs. >> >> IFUNCs are never resolved lazily; that's one way in which they differ >> from PLTs. >> >> (BTW, in response to a comment upthread, ARM does support IFUNCs now.) > > Do you know when support was merged in eglibc? > > For example, Linaro binutils 2.21.0.20110327-2ubuntu3 appears to support > IFUNC, but the natty version of eglibc (2.13-0ubuntu13) does not handle > it correctly.
That's why I didn't try and use it; I thought it was best to wait for it to percolate through. Richard's patch went into glibc-ports on April 26th and looks like it got imported into eglibc-ports on 5th May (svn rev 13608); there was also a fix from Joseph Myers in that area on 21st June. Dave _______________________________________________ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
