![]() | |||||||||||||
Messages in this thread |
On Sun, 31 May 1998, Nicholas J. Leon wrote: > * it was my understand that perl was a no-no for mainstream inclusion. Cuz > if it isn't, I reduce my two pronged approach (1 .sh, 1.c) into a single > perl script. Well, I've seen it used for the NCR scsi drivers for eons now, so that's why I didn't consider it taboo to user it. > * the footprints of these two are greatly different. I show about a 1000 > byte increase in runtime size in my version. Much less than the other. Well, booting a kernel with and a kernel without both give the same memory stats on bootup: Memory: .... (748k kernel code, 392k reserved, 956k data, 32k init) Which I thought was very strange, but there is a 364 byte difference between the two kernel's vmlinux file, and a 2230 byte difference in the bzImage files. It could be that alignment or somesuch is hiding the difference at runtime. [snip] > (The above was not said with a bitter tone to it, though mild > disappointment at the loss of control might accurate enough :) It was never my intention to "take over" this feature. Since I had already done 90% of this when I posted originally, I just wanted to show a different implementation. Probably the biggest difference (the System.map 'feature' aside, since that can only be done as a module) is the ability to free up the memory and remove the proc entry. In this case it may be overkill, because the additional code required to release the memory is offset somewhat by the smallness of the data that it is releasing. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Andrew Anderson http://amelia.db.erau.edu/~andrew/ if(!(family_tree=fork())){redneck=TRUE;} - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu | ||||||||||||
| Last update: 2005-03-22 12:42 [from the cache] ©2003-2008 | |||||||||||||