[Bug translator/2149] check return value for _stp_map_set_xx()

[joshua dot i dot stone at intel dot com]

------- Additional Comments From joshua dot i dot stone at intel dot com  2006-01-12 20:49 -------
I don't want to belittle the importance of checking return values, but filled
maps could be avoided by turning on map->wrap.  Perhaps we need pragmas or
something to let the user turn this on...

And BTW - the current replacement policy when wrapping is LRU.  But, when
aggregating pmaps (_stp_pmap_agg), it favors data from the later cpus in the
for_each_cpu.  I'm not sure of a "fair" way to correct this, except to let the
pmap->agg have more entries (MAXMAPENTRIES*NUM_CPUS).

-- 


http://sourceware.org/bugzilla/show_bug.cgi?id=2149

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.