0
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
1 Date: Wed, 20 Jun 2001 12:32:22 -0500
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
2 From: Vikram Adve <vadve@cs.uiuc.edu>
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
3 To: Chris Lattner <lattner@cs.uiuc.edu>
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
4 Subject: .NET vs. our VM
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
5
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
6 One significant difference between .NET CLR and our VM is that the CLR
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
7 includes full information about classes and inheritance. In fact, I just
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
8 sat through the paper on adding templates to .NET CLR, and the speaker
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
9 indicated that the goal seems to be to do simple static compilation (very
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
10 little lowering or optimization). Also, the templates implementation in CLR
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
11 "relies on dynamic class loading and JIT compilation".
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
12
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
13 This is an important difference because I think there are some significant
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
14 advantages to have a much lower level VM layer, and do significant static
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
15 analysis and optimization.
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
16
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
17 I also talked to the lead guy for KAI's C++ compiler (Arch Robison) and he
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
18 said that SGI and other commercial compilers have included options to export
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
19 their *IR* next to the object code (i.e., .il files) and use them for
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
20 link-time code generation. In fact, he said that the .o file was nearly
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
21 empty and was entirely generated from the .il at link-time. But he agreed
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
22 that this limited the link-time interprocedural optimization to modules
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
23 compiled by the same compiler, whereas our approach allows us to link and
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
24 optimize modules from multiple different compilers. (Also, of course, they
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
25 don't do anything for runtime optimization).
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
26
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
27 All issues to bring up in Related Work.
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
28
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
29 --Vikram
|
Kaito Tokumori <e105711@ie.u-ryukyu.ac.jp>
parents:
diff
changeset
|
30
|