On 1/18/19 11:07 AM, Greg KH wrote:
On Thu, Jan 10, 2019 at 12:51:33PM -0500, David Long wrote:
From: "David A. Long" <dave.long@xxxxxxxxxx>
V4.14 backport of spectre patches from Russell M. King's spectre branch.
Based on some recent private email threads I had about this series, here
is what I would like to see for the next time this is posted:
- Patch series such that someone moving from one tree to a newer one
does not experience regressions (i.e. a 4.19 version of this series,
and 4.20 where needed.)
The v4.19 and v4.9 versions are ready. Not sure I understand why there
would be a need for a v4.20 version even if it is one patch shy.
- a statement saying how this was tested. As part of that statement,
there better be something like "we tested using our reproducer and it
shows that there is no longer an issue." Yes, I know Spectre
reproducers are hard to come by, but they are out there and I do not
want to take a patch series that is not at least tested for the thing
that it is supposed to be solving.
I'll see what I can figure out for spectre testing but at best this
stuff is pretty non-deterministic.
-dl
