This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [RFC] Expanding the hwcap
- From: Roland McGrath <roland at hack dot frob dot com>
- To: rsa at us dot ibm dot com
- Cc: libc-alpha at sourceware dot org, rth at twiddle dot net, Benjamin Herrenschmidt <benh at kernel dot crashing dot org>, Steven Munroe <sjmunroe at us dot ibm dot com>
- Date: Wed, 17 Oct 2012 12:53:08 -0700 (PDT)
- Subject: Re: [RFC] Expanding the hwcap
- References: <1350425312.25040.6342.camel@localhost.localdomain>
AT_HWCAP2 seems adequate for the time being, as others have said.
But I don't think there's anything wrong with the extensible vDSO approach.
It doesn't need to induce any runtime overhead (beyond the startup time
cost of finding the magic in the vDSO). struct rtld_global_ro can just
contain a fixed-sized array large enough for all the bits that libc code
actually tests. The startup code will copy into that from the vDSO values,
rather than the code testing the bits following a pointer into the vDSO.
Thanks,
Roland