[lkml]   [2020]   [May]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [git pull] coredump infoleak fix
On Thu, May 28, 2020 at 12:19:32PM -0700, Linus Torvalds wrote:
> On Thu, May 28, 2020 at 12:17 PM Al Viro <> wrote:
> >
> > Might make sense to change the summary of that pull request to something
> > like
> > make sure we don't forget to report the xstate components that happen
> > to be in init state - both for coredump and for PTRACE_GETREGSET
> Note that this has nothing to do with x86 per se.
> It's more about ->getregs() being a horrid interface, and being easy
> to get wrong in general. The fact that xstate is complex is just one
> such trigger.

The only one I've ran into so far, fortunately... Almost all instances
write sequentially; the only exceptions are this one (buggy) and ia64
horrors with unwind.

I certainly agree that ->get() is an atrocity...

 \ /
  Last update: 2020-05-28 21:29    [W:0.041 / U:8.224 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site