Re: Please revert 21564fd2a3deb48200b595332f9ed4c9f311f2a7

Alan Cox wrote:
2.6.22's behaviour is the bug. 2.6.21 you couldn't load random binary
crap into the kernel without logging a taint. 2.6.22 you can. This means
every single 2.6.22 bug report has to be assumed to be caused by binary
module crap as a starting point which slows down debug immensely.

Er, what? What do you mean by "load random binary crap"? Are you worried that a malicious kernel module might modify paravirt_ops? How is that different from a malicious module doing any of the infinite other things a malicious module can do? A module will only register a taint if its playing by the rules anyway, and a module playing by the rules won't touch paravirt_ops directly.

J

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Relevant Pages

  • Re: Fw: signed kernel modules?
    ... fact that the kernel does have to worry more about licensing issues ... admit that I didn't like not seeing the compiler output on first use, ... Damn crap. ... process has used cunning hacks for years. ...
    (Linux-Kernel)
  • Re: [GIT PATCH] UIO patches for 2.6.21
    ... By using the kernel's userspace interface, ... "derived work" of the kernel. ... I think we should have a taint flag for UIO type drivers. ... if we make it entirely clear that the flag is nonpolitical. ...
    (Linux-Kernel)
  • Re: [RFD] Freezing of kernel threads
    ... you're just adding crap to make up for other crap. ... *not* how we do kernel develpment. ... So stop talking about freezing when you're talkign abotu CPU hotplug. ... much more than the freezer, and which has never really caused any issues, ...
    (Linux-Kernel)
  • [PATCH 1/1] Taint kernel after WARN_ON(condition)
    ... warning is issued with WARN_ON. ... is helpful to record this within the kernel state as a taint. ... This patch applies to Linus's git tree. ...
    (Linux-Kernel)
  • Re: Fw: signed kernel modules?
    ... At least until an attacker uploads a malicious module or modifies one of the ... > what's in the module, the loader, or the kernel. ... > either of these is crap. ... send the line "unsubscribe linux-kernel" in ...
    (Linux-Kernel)