linux-kernel - [RFC PATCH v2 0/6] SEAMCALL Wrappers

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20241203010317.827803-1-rick.p.edgecombe@intel.com>
Date: Mon,  2 Dec 2024 17:03:10 -0800
From: Rick Edgecombe <rick.p.edgecombe@...el.com>
To: kvm@...r.kernel.org,
	pbonzini@...hat.com,
	seanjc@...gle.com,
	dave.hansen@...el.com
Cc: isaku.yamahata@...il.com,
	kai.huang@...el.com,
	linux-kernel@...r.kernel.org,
	tony.lindgren@...ux.intel.com,
	xiaoyao.li@...el.com,
	yan.y.zhao@...el.com,
	rick.p.edgecombe@...el.com,
	x86@...nel.org,
	adrian.hunter@...el.com
Subject: [RFC PATCH v2 0/6] SEAMCALL Wrappers

Hi,

This is a followup to the "SEAMCALL Wrappers" RFC[0] that spun out of 
Dave’s comments on the SEAMCALL wrappers in the “TDX vCPU/VM creation” 
series [1]. To try to summarize Dave’s comments, he noted that the 
SEAMCALL wrappers were very thin, not even using proper types for things 
where types exist.

The last discussion was to use struct pages instead of u64. It is pretty 
much what Dave suggested with a minor tweak to instead include the tdcx 
page count in the TD struct instead of the vCPU one.

This is because it will not vary between vCPUs. Doing it that way 
basically preserves the existing data duplication, but these counts are 
basically "global metadata". The global metadata patches export them as a 
size, but KVM wants to use them as a page count. So we should not be 
including these counts in each TD scoped structure as is currently done. To
address the duplication we need to change the "global metadata patches"
to export the count instead of size.

Otherwise, in the spirit of looking to find better types for the other raw 
u64's, I played around again with the out params of 
tdh_phymem_page_reclaim(). In the end I opted for better names and a 
comment rather than anything fancier.

Here is the branch with the VM/vCPU caller adjustments as the last commit:
https://github.com/intel/tdx/tree/seamcall-rfc-v2

Thanks,

Rick

[0]
https://lore.kernel.org/kvm/20241115202028.1585487-1-rick.p.edgecombe@intel.com/
[1]
https://lore.kernel.org/kvm/20241030190039.77971-1-rick.p.edgecombe@intel.com/

Rick Edgecombe (6):
  x86/virt/tdx: Add SEAMCALL wrappers for TDX KeyID management
  x86/virt/tdx: Add SEAMCALL wrappers for TDX TD creation
  x86/virt/tdx: Add SEAMCALL wrappers for TDX vCPU creation
  x86/virt/tdx: Add SEAMCALL wrappers for TDX page cache management
  x86/virt/tdx: Add SEAMCALL wrappers for TDX VM/vCPU field access
  x86/virt/tdx: Add SEAMCALL wrappers for TDX flush operations

 arch/x86/include/asm/tdx.h  |  38 ++++++
 arch/x86/virt/vmx/tdx/tdx.c | 240 ++++++++++++++++++++++++++++++++++++
 arch/x86/virt/vmx/tdx/tdx.h |  38 ++++--
 3 files changed, 309 insertions(+), 7 deletions(-)

-- 
2.47.1