bloat-o-meter: include read-only data section in report
authorJean Delvare <khali@linux-fr.org>
Tue, 22 Mar 2011 23:35:13 +0000 (16:35 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Wed, 23 Mar 2011 00:44:17 +0000 (17:44 -0700)
I'm not sure why the read-only data section is excluded from the report,
it seems as relevant as the other data sections (b and d).

I've stripped the symbols starting with __mod_ as they can have their
names dynamically generated and thus comparison between binaries is not
possible.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Andi Kleen <andi@firstfloor.org>
Acked-by: Nathan Lynch <ntl@pobox.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
scripts/bloat-o-meter

index 6501a50e17f0e157dab2dab3d64d8807fc3669da..6129020c41a94570ff49f5e42b5963a23105ba8e 100755 (executable)
@@ -17,7 +17,9 @@ def getsizes(file):
     sym = {}
     for l in os.popen("nm --size-sort " + file).readlines():
         size, type, name = l[:-1].split()
-        if type in "tTdDbB":
+        if type in "tTdDbBrR":
+            # strip generated symbols
+            if name[:6] == "__mod_": continue
             # function names begin with '.' on 64-bit powerpc
             if "." in name[1:]: name = "static." + name.split(".")[0]
             sym[name] = sym.get(name, 0) + int(size, 16)