Re: Why is the code in .sys drivers slightly different in memory than on disk!?
From: heinz baer (heinz_baer_at_my-deja.com)
Date: 02/02/04
- Next message: Cinthia Lee: "Re: How does a tablet support pressure-sensitivity?"
- Previous message: BobF: "Re: Driver code sample for software devices?"
- Maybe in reply to: Wayne: "Re: Why is the code in .sys drivers slightly different in memory than on disk!?"
- Next in thread: Maxim S. Shatskih: "Re: Why is the code in .sys drivers slightly different in memory than on disk!?"
- Reply: Maxim S. Shatskih: "Re: Why is the code in .sys drivers slightly different in memory than on disk!?"
- Messages sorted by: [ date ] [ thread ]
Date: 1 Feb 2004 18:08:12 -0800
I had not heard of UPX and so downloaded it and read about it,
discovering there are a number of misconceptions here:
> you lose CPU time on decompression.
Bottom line is since the decompression is faster than the data
transfer rate of the hard drive, compressed drivers would load FASTER.
If every driver used UPX the system would start faster while reducing
the disk foot print.
> ...which will multiply the memory load on the machine several times.
No. Direct quote from their home page: "There is no memory overhead
for compressed executables because of in-place decompression."
> IMHO compressing EXEs in modern OSes is a degree in brain damage.
The hard drive is increasingly becoming the major bottleneck in the
system so I could appreciate the advantage of everything loading
quicker.
> use NTFS compression
Doesn't help the speed of downloading drivers from the net and that's
where the best gains are to be had because most people are on dialup.
If NTFS was the answer, no one would be using JPG, GIF, or PNG for
images; it would all be BMP files for instance. The internet makes
file size today as important as it was in the days software was put on
floppies. If Microsoft adds a -COMPRESS linker switch, I don't see why
anyone would want to turn it off.
- Next message: Cinthia Lee: "Re: How does a tablet support pressure-sensitivity?"
- Previous message: BobF: "Re: Driver code sample for software devices?"
- Maybe in reply to: Wayne: "Re: Why is the code in .sys drivers slightly different in memory than on disk!?"
- Next in thread: Maxim S. Shatskih: "Re: Why is the code in .sys drivers slightly different in memory than on disk!?"
- Reply: Maxim S. Shatskih: "Re: Why is the code in .sys drivers slightly different in memory than on disk!?"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
