Older blog entries for madhatter (starting at number 26)

Linux 2.4.x ptrace exploit

what a doozy, and I didn't want to have to run a 2.5 kernel... so i tried some things. Solar Designer happens to have a experimental security patch for 2.4.20 (www.openwall.com), but unfortunately it didn't stop the exploit from working. Next, St. Jude (sf.net/projects/stjude). I've had success with using St. Jude before, and it seems to stop this exploit from working. I also tested some other recent security problems (xfree86 XLOCALEDIR env overflow, and cupsd remote overflow) and it was able to stop BOTH of them from working. Good job Tim Lawless.
tk:
I've wondered that for a while too.... at least with a numerical-type name we would all have a unique identifier. Numbers as names does strike me as a chilling thought for some reason.
GCC quirk

i was fooling around with SIMD instructions on my notebook today and found something with gcc inline assembly:

asm volatile("xor %%eax,%%eax\n");

it looks like a pretty reasonable statement, but unfortunately it will not compile. gas will tell you that %%eax is an invalid register. it was throwing me for a loop for a while until mpr pointed out to me that gcc did not strip the first parentheses from the register name for some reason. gas doesn't want "%%eax", it wants "%eax", so I was sitting there with my thumb up my butt for about 5 minutes until I started messing with my inline statement, then doing gcc -S to see what the assembler output looked like. I couldn't get gcc to strip the first parentheses until I changed it to this:

asm volatile("xor %%eax,%%eax\n" : );

no, that's not supposed to be a smiley face. If you at least put the output operand colon there gcc will strip the first parentheses, otherwise it will assemble it without touching it and gas will go kaput. does anyone know why this happens?

hey, if you are into modeling and movement algorithms, this is alot of fun: www.sodaplay.com

19 Nov 2002 (updated 10 Dec 2002 at 18:29 UTC) »

gone

28 Oct 2002 (updated 10 Dec 2002 at 18:29 UTC) »

gone

24 Oct 2002 (updated 10 Dec 2002 at 18:30 UTC) »

gone

24 Oct 2002 (updated 6 Mar 2007 at 05:21 UTC) »
obsolete
23 Oct 2002 (updated 6 Mar 2007 at 05:20 UTC) »
obsolete

17 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!