|
From: | Fabrice Bellard |
Subject: | Re: [Qemu-devel] Re: [RFC][PATCH] x86: CS limit checks |
Date: | Thu, 17 Jul 2008 21:29:24 +0200 |
User-agent: | Thunderbird 2.0.0.5 (X11/20070727) |
[...]
And then there is the open question how much performance can be gained with compile-time optimization for those guests who do use segmentation.
The goal is to optimize the cases where segmentation is not used ! It is just an extension of the existing "HF_ADDSEG" optimization.
The worst case is very roughly about 50% slowdown right now (/w vs. w/o -seg-checks). As answered privately, some -no-seg-checks switch could remain a useful optimization.
Not sure. I believe most recent OSes (i.e. those where speed really matters) have segment registers loaded so that no runtime checks are necessary.
Fabrice.
[Prev in Thread] | Current Thread | [Next in Thread] |