This is the mail archive of the gcc-patches@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] |
On 14/10/2004, at 6:31 PM, Andrew Pinski wrote:On Oct 14, 2004, at 9:28 PM, Geoffrey Keating wrote:This is a small optimisation improvement on Darwin; it's better to separate jump tables from instructions because it makes better use of separate instruction and data caches.
Are you sure that this helps at all, the last time I heard Dale tried this and it did not help at all.
Right, it seemed to make no difference either way (although I think the measurement was pre-G5). What's the data backing up your claim?
The other reason for doing it is to make cleaner disassembly possible in gdb and shark.
...but this seems reasonable, so I have no immediate objection. If there's a regression on Spec I'll change my mind though.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |