documenting PrintCompilation output

[Krystal Mok]

Hi Christian,

I quickly looked over the file and noticed that the links for CR 7022998 are
> wrong.

Oops, my bad. Copy-and-paste bug. Fixed it.
Thanks for pointing that out!

BTW, the compilation level in tiered mode in HS22-b02 is still printed as
"0". I'm not sure what's the cleaner way of getting the intended "-"; this
one certainly does the trick:

diff -r a594deb1d6dc src/share/vm/compiler/compileBroker.cpp
--- a/src/share/vm/compiler/compileBroker.cpp Mon Aug 22 11:00:39 2011 -0700
+++ b/src/share/vm/compiler/compileBroker.cpp Wed Aug 24 23:06:21 2011 +0800
@@ -322,8 +322,8 @@
   st->print("%c%c%c%c%c ", compile_type, sync_char, exception_char,
blocking_char, native_char);

   if (TieredCompilation) {
-    if (comp_level != -1)  st->print("%d ", comp_level);
-    else                   st->print("- ");
+    if (comp_level != CompLevel_none)  st->print("%d ", comp_level);
+    else                               st->print("- ");
   }
   st->print("     ");  // more indent


Or maybe it's better to create the nmethod object for the native wrapper
with a compilation level of -1, in nmethod::new_native_nmethod

Regards,
Kris Mok

On Wed, Aug 24, 2011 at 7:32 PM, Christian Thalinger <
christian.thalinger at oracle.com> wrote:

> I quickly looked over the file and noticed that the links for CR 7022998
> are wrong.
>
> -- Christian
>
> On Aug 24, 2011, at 7:40 AM, Krystal Mok wrote:
>
> > Hi all,
> >
> > I wrote a note describing the output of PrintCompilation from HotSpot VM:
> https://gist.github.com/1165804#file_notes.md
> > I thought there might be other people interested, so I'm posting the link
> here.
> >
> > Please correct me if I've made any mistakes in the note, especially the
> details. I'm not familiar with the HotSpot VMs before OpenJDK came out, so
> I'm not too sure if I got it right.
> >
> > P.S. Any chances of integrating notes like this one into the HotSpot
> Internals wiki (
> http://wikis.sun.com/display/HotSpotInternals/PrintCompilation)?
> >
> > Regards,
> > Kris Mok
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/hotspot-dev/attachments/20110824/731bf1b0/attachment.html