readonly version

[Simon]

JB

Out of curiosity how much faster could Cecil be if there was a readonly version?

That is if u didnt need to worry about saving or modifying IL and was only returning it for analysis purposes

[Jb Evain]

Hi,

It's hardly possible to give a meaningful figure.
But I don't think it would account for much.

> --
> --
> mono-cecil

[Greg Young]

I researched this a while ago for MM (when looking at development of
lowmemuse). It did not seem to be a significant amount of overhead
caused by code there for writing capabilities. Maybe eek out a few %
but overall the performance did not seem to be there.

> --
> --
> mono-cecil



--
Le doute n'est pas une condition agréable, mais la certitude est absurde.

[Simon]

I was thinking that since most things could be assumed to be immutable the reading of the assembly could make use of multiple cores.

Would this not make a significant difference?

[Jb Evain]

On Jul 18, 2012, at 10:31 AM, Simon wrote:
> I was thinking that since most things could be assumed to be immutable the reading of the assembly could make use of multiple cores.
>
> Would this not make a significant difference?

In that case, no, I don't think it would make a significant difference. You'd spend a lot of time synchronizing given how the metadata is interwoven.