From: John Levon <movement@us...> - 2004-05-07 14:59:42
Update of /cvsroot/oprofile/oprofile
In directory sc8-pr-cvs1.sourceforge.net:/tmp/cvs-serv19556
push some things back past 0.8 release
RCS file: /cvsroot/oprofile/oprofile/TODO,v
retrieving revision 1.381
retrieving revision 1.382
diff -u -p -d -r1.381 -r1.382
--- TODO 4 Apr 2004 17:22:19 -0000 1.381
+++ TODO 7 May 2004 14:59:31 -0000 1.382
@@ -3,20 +3,14 @@ This is an (incomplete) list of some of
If you're interested in hacking on any of these, please contact the list first
for some pointers and/or read HACKING and doc/CodingStyle.
- o stress test opstack: compile a Big Application w/o frame pointer and look
- how driver and opstack react.
- o TRACE_END is no longer sent by the driver, keep it a bit for compatibility
+ o anton's bug on the list about dcache aliasing
o callgraph_container.cpp: the inner loop in add() is very fragile!
- o child sample count for recursive function is not a bug but a feature, it
- needs documentation.
o need --callgraph in oprof_start
+ o stress test opstack: compile a Big Application w/o frame pointer and look
+ how driver and opstack react.
o add event aliases for common things like icache misses, we must start to
think about metrics including simple like event alias mapped to two or more
@@ -32,7 +26,10 @@ for some pointers and/or read HACKING an
Before 1.0 little stuff
+ o TRACE_END is no longer sent by the driver, keep it a bit for compatibility
o callgraph patch: better way to skip ignored backtrace ?
+ o child sample count for recursive function is not a bug but a feature, it
+ needs documentation.
o opcontrol --reset should avoid to reload the module if it's unloaded
o zwane problem with wrong text offset showed an interesting problem: if
op_bfd.cpp get any symbol below text offset for vmlinux or a module then
@@ -65,7 +62,6 @@ Before 1.0 little stuff
o odb_insert() can fail on ftruncate or mremap() in db_manage.c but we don't
try to recover gracefully.
o output column shortname headers for opreport -l
- o separate debug info stuff
o is relative_to_absolute_path guaranteeing a trailing '/' documented ?
o create_path API is weird
o move oprofiled.log to OP_SAMPLE_DIR/current ?
@@ -78,8 +74,6 @@ Before 1.0 little stuff
o the way we show kernel modules in 2.5 is not very obvious - "/oprofile"
o sample-file: / binary: don't work in any useful way - can we fix this
by peeking at binary: value and faking the split_sample_file somehow ?
- o increase encapsulation but, at least for now, not at the cost of
- getter()/setter(), this can go to Later section too.
@@ -107,6 +101,8 @@ General checks to make
+ o increase encapsulation but, at least for now, not at the cost of
+ getter()/setter(), this can go to Later section too.
o we should notice an opcontrol config change (--separate etc.) and
auto-restart the daemon if necessary (Run)
o we can add lots more unit tests yet