Thread (28 messages) 28 messages, 10 authors, 2020-08-14

Re: [Linux-kernel-mentees] [PATCH net] rds: Prevent kernel-infoleak in rds_notify_queue_get()

From: Jason Gunthorpe <jgg@ziepe.ca>
Date: 2020-08-14 17:07:15
Also in: linux-kernel-mentees, linux-rdma, lkml

On Sat, Aug 08, 2020 at 03:57:33PM -0700, Jack Leadford wrote:
Hello!

Thanks to Jason for getting this conversation back on track.

Yes: in general, {} or a partial initializer /will/ zero padding bits.

However, there is a bug in some versions of GCC where {} will /not/ zero
padding bits; actually, Jason's test program in this mail
https://lore.kernel.org/lkml/20200731143604.GF24045@ziepe.ca/ (local)
has the right ingredients to trigger the bug, but the GCC
versions used are outside of the bug window. :)
It seems fine, at least Godbolt doesn't show a bug with that code.

Can you share the test that does fail?

This seems like the sort of security sensitive bug that should be
addressed in gcc, not worked around in the kernel code :\

Jason
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help