Peter da Silva wrote:
* I published this idea on Usenet 20 years ago, and there's prior art in Intel's "ISIS" operating system... which was incidentally one of the precursors of CP/M and MS-DOS, so it's pretty patent-resistant.
Now that pretty much slaughters my bright idea, doesn't it? Too bad I already spent all that money assuming I'm gonna become rich in a couple of days. What took you so long?
Well, there is still some hope for me - maybe people care about their statically linked binaries, and then there are all those versions of libc out there, and then there's the small amount of programs which are not written on top of libc.
Of course there are the other problems - no portable way to find out what syscall ptrace traced except for taking all that scary code from strace, no portable way to open an image file except for using all that scary code from bfd, and so on.
As a matter of fact - yes, I do find the argument-list-too-long *EXTREMELY HATEFUL*, because there are hairy scripts that fail because of it periodically and I have to manually fix stuff or just give up, depending on the circumstances. So I guess I will implement the entry-point-in-the-middle workaround. Thanks for the tip. I'll have to look into the statically-linked-binary-with-no-source-available thing (I could overwrite _start, but that's, um, suboptimal), but there probably aren't many of these.
Allow me to express my gratitude by not enumerating the memory limitations and other resource constraints of the various devices I had the dubious pleasure to program (1K FOR THE ARGUMENTS? THAT'S ONE SHITLOAD OF WASTED MEMORY!). I think the fact that my not-so-top-notch workstation has 2G of real memory says it all.
