This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Tremendous increase in compile times for 3.4.1 with -mcpu=G5



On Thursday, August 12, 2004, at 09:12 AM, Michael Matz wrote:


No, sbitmaps are not used in invalid_mode_change_p which is culprit
function. The problem is, that with -mcpu=G4 the subregs_of_mode bitmap
(_not_ sbitmap) is sparsely filled up to bit 981728 in 197 list elements.


For G5 it is filled up to bit 1621359 scattered over 9011 elements. I.e.
not as sparse. The function in question iterates over a subset of bits in
that bitmap, starting with a predefined one. It is searching for the
element containing this start bit which is so much slower with G5.


I suspect the same reason as David mentioned why this bitmap is filled
with more bits.  There simply are more references to subregs.

I've been trying to digest this. Is this a problem that gcc has with 64-bit ppc architecture, i.e., one that will also show up on aix and linux64?


Brad


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]