Josef Weidendorfer 840c6d8437 Before LL, instrument outstanding helper calls. ...

Callgrind, Cachegrind, and Lackey call
helpers for memory accesses in bunches, to reduce
register save/restore overhead (and merge load/store
within same instruction into a "modify" event).

The calls should not be done within a RMW section
enclosed by LL/SC instructions, as this reduces the
chance of SC to succeed, and can result in hangs.
For Callgrind, this definitly helped MIPS, and was
committed in r13136. Do the same for Cachegrind/Lackey.

git-svn-id: svn://svn.valgrind.org/valgrind/trunk@13143

2012-11-26 18:16:58 +00:00

..

docs

Use last-level cache for simulation. Even if there are more than

2012-10-07 21:03:27 +00:00

tests

s390x machines z196 and zec12 have 4 levels of caches. Adopt the test suite

2012-11-07 07:57:01 +00:00

cg_annotate.in

Added cg_diff.

2010-06-30 05:23:34 +00:00

cg_arch.h

Fix more Char/HChar mixups. Closing in...

2012-11-10 22:29:54 +00:00

cg_branchpred.c

Update copyright dates to include 2012.

2012-08-05 15:46:46 +00:00

cg_diff.in

Added a --mod-funcname option to cg_diff.

2010-12-17 00:45:19 +00:00

cg_main.c

Before LL, instrument outstanding helper calls.

2012-11-26 18:16:58 +00:00

cg_merge.c

Changes to allow compilation with -Wwrite-strings. That compiler option

2012-11-23 16:17:43 +00:00

cg_sim.c

Cachegrind: introduce special case for Ir

2012-10-30 00:28:29 +00:00

cg-arch.c

Fix more Char/HChar mixups. Closing in...

2012-11-10 22:29:54 +00:00

Makefile.am

This patch is the first installment of the cache info reorganisation.

2012-10-07 19:47:04 +00:00