On Wed, 2023-05-17 at 16:06 +0200, Greg Kroah-Hartman wrote:
On Wed, May 17, 2023 at 04:02:35PM +0200, Greg Kroah-Hartman wrote:
On Wed, May 17, 2023 at 02:56:21PM +0200, Ben Hutchings wrote:
On Fri, 2023-03-10 at 14:33 +0100, Greg Kroah-Hartman wrote:
From: Sean Christopherson seanjc@google.com
[ Upstream commit e4a9af799e5539b0feb99571f0aaed5a3c81dc5a ]
For commands with small input/output buffers, use the local stack to "allocate" the structures used to communicate with the PSP. Now that __sev_do_cmd_locked() gracefully handles vmalloc'd buffers, there's no reason to avoid using the stack, e.g. CONFIG_VMAP_STACK=y will just work.
[...]
Julien Cristau reported a regression in ccp - the WARN_ON_ONCE(!virt_addr_valid(data)) is now being triggered. I believe this was introduced by the above commit, which depends on:
commit 8347b99473a313be6549a5b940bc3c56a71be81c Author: Sean Christopherson seanjc@google.com Date: Tue Apr 6 15:49:48 2021 -0700 crypto: ccp: Play nice with vmalloc'd memory for SEV command structs
Ben.
Thanks for letting me know, now queued up.
Nope, now dropped, it breaks the build :(
I've now looked further and found that we need both:
d5760dee127b crypto: ccp: Reject SEV commands with mismatching command buffer 8347b99473a3 crypto: ccp: Play nice with vmalloc'd memory for SEV command structs
(Not yet tested; I'll ask Julien if he can do that.)
Ben.