This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: ia64: Don't use static branch prediction
- To: bernds at redhat dot co dot uk (Bernd Schmidt)
- Subject: Re: ia64: Don't use static branch prediction
- From: cgd at sibyte dot com (Chris G. Demetriou)
- Date: 28 Oct 2000 09:54:22 -0700
- Cc: gcc-patches at gcc dot gnu dot org
- References: <Pine.LNX.4.21.0010281734080.2467-100000@host117.cygnus> <mailpost.972751061.10582@postal.sibyte.com>
bernds@redhat.co.uk (Bernd Schmidt) writes:
> /* Guess top and bottom 10% statically predicted. */
does this comment deserve updating? (I'd guess so.)
> - else if (pred_val < REG_BR_PROB_BASE * 9 / 10)
> + else if (pred_val < REG_BR_PROB_BASE / 100 * 98)
in practice, the ordering here is OK because of the (large-ish) value
of REG_BR_PROB_BASE... but why switch it around?
cgd