Thread (15 messages) flat view 15 messages, 2 authors, 2014-09-26

Re: eBPF verifier thoughts (Re: [PATCH v15 net-next 00/11] eBPF syscall, verifier, testsuite)

From: Alexei Starovoitov <hidden>
Date: 2014-09-26 22:03:35
Also in: lkml, netdev

On Fri, Sep 26, 2014 at 2:47 PM, Andy Lutomirski [off-list ref] wrote:
Can't you just disallow the 1-byte write to the stack?
of course not.
That would be extremely limiting to users.
Can you actually see yourself living with stack that only
allows 8-byte writes/reads?
The stack usage will increase a lot, since all char/short
stack variables will become 8-byte...
I'll try it in Python.  I bet I can get it to be shorter than the current code.
Awesome challenge! :)
I'll buy you a drink of your choice if you can achieve that.
Also I'll send you our C programs that we use for
testing to make sure, your python verifier can analyze them.
If it passes, I'll be glad to rip mine out. Seriously.
Deal?
Yes, but does it work reliably?
I'm not saying current verifier is perfect. It can get better.
So far it was enough to let people code freely on top of it.
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help