All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jonathan Cameron <jonathan.cameron@huawei.com>
To: Dan Williams <dan.j.williams@intel.com>
Cc: <linux-pci@vger.kernel.org>, <linux-coco@lists.linux.dev>,
	Xu Yilun <yilun.xu@linux.intel.com>
Subject: Re: [PATCH v2 7/8] PCI/TSM: Add pci_tsm_guest_req() for managing TDIs
Date: Thu, 13 Nov 2025 12:04:15 +0000	[thread overview]
Message-ID: <20251113120415.000026cf@huawei.com> (raw)
In-Reply-To: <20251113021446.436830-8-dan.j.williams@intel.com>

On Wed, 12 Nov 2025 18:14:45 -0800
Dan Williams <dan.j.williams@intel.com> wrote:

> A PCIe device function interface assigned to a TVM is a TEE Device
> Interface (TDI). A TDI instantiated by pci_tsm_bind() needs additional
> steps taken by the TVM to be accepted into the TVM's Trusted Compute
> Boundary (TCB) and transitioned to the RUN state.
> 
> pci_tsm_guest_req() is a channel for the guest to request TDISP collateral,
> like Device Interface Reports, and effect TDISP state changes, like
> LOCKED->RUN transititions. Similar to IDE establishment and pci_tsm_bind(),
> these are long running operations involving SPDM message passing via the
> DOE mailbox.
> 
> The path for a TVM to invoke pci_tsm_guest_req() is:
> * TSM triggers exit via guest-to-host-interface ABI (implementation specific)
> * VMM invokes handler (KVM handle_exit() -> userspace io)
> * handler issues request (userspace io handler -> ioctl() ->
>   pci_tsm_guest_req())
> * handler supplies response
> * VMM posts response, notifies/re-enters TVM
> 
> This path is purely a transport for messages from TVM to platform TSM. By
> design the host kernel does not and must not care about the content of
> these messages. I.e. the host kernel is not in the TCB of the TVM.
> 
> As this is an opaque passthrough interface, similar to fwctl, the kernel
> requires that implementations stay within the bounds defined by 'enum
> pci_tsm_req_scope'. Violation of those expectations likely has market and
> regulatory consequences. Out of scope requests are blocked by default.
> 
> Co-developed-by: Xu Yilun <yilun.xu@linux.intel.com>
> Signed-off-by: Xu Yilun <yilun.xu@linux.intel.com>
> Signed-off-by: Dan Williams <dan.j.williams@intel.com>

Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>



  reply	other threads:[~2025-11-13 12:04 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-13  2:14 [PATCH v2 0/8] PCI/TSM: Finalize "Link" TSM infrastructure Dan Williams
2025-11-13  2:14 ` [PATCH v2 1/8] drivers/virt: Drop VIRT_DRIVERS build dependency Dan Williams
2025-11-13 11:28   ` Jonathan Cameron
2025-12-02 23:44   ` Nathan Chancellor
2025-12-03  1:51     ` dan.j.williams
2025-11-13  2:14 ` [PATCH v2 2/8] PCI/TSM: Drop stub for pci_tsm_doe_transfer() Dan Williams
2025-11-13 11:29   ` Jonathan Cameron
2025-11-13  2:14 ` [PATCH v2 3/8] resource: Introduce resource_assigned() for discerning active resources Dan Williams
2025-11-13 11:36   ` Jonathan Cameron
2025-11-13  2:14 ` [PATCH v2 4/8] PCI/IDE: Add Address Association Register setup for downstream MMIO Dan Williams
2025-11-13 11:48   ` Jonathan Cameron
2025-11-13 18:54   ` kernel test robot
2025-11-14  1:02   ` [PATCH v3 " Dan Williams
2025-11-19  9:02   ` [PATCH v2 " kernel test robot
2025-11-13  2:14 ` [PATCH v2 5/8] PCI/IDE: Initialize an ID for all IDE streams Dan Williams
2025-11-13 11:52   ` Jonathan Cameron
2025-11-17 11:11   ` Xu Yilun
2025-11-13  2:14 ` [PATCH v2 6/8] PCI/TSM: Add pci_tsm_bind() helper for instantiating TDIs Dan Williams
2025-11-13 12:01   ` Jonathan Cameron
2025-11-13 20:41     ` dan.j.williams
2025-11-17 11:30   ` Xu Yilun
2025-11-13  2:14 ` [PATCH v2 7/8] PCI/TSM: Add pci_tsm_guest_req() for managing TDIs Dan Williams
2025-11-13 12:04   ` Jonathan Cameron [this message]
2025-11-17 11:57   ` Xu Yilun
2025-11-13  2:14 ` [PATCH v2 8/8] PCI/TSM: Add 'dsm' and 'bound' attributes for dependent functions Dan Williams
2025-11-17 14:58   ` Xu Yilun

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20251113120415.000026cf@huawei.com \
    --to=jonathan.cameron@huawei.com \
    --cc=dan.j.williams@intel.com \
    --cc=linux-coco@lists.linux.dev \
    --cc=linux-pci@vger.kernel.org \
    --cc=yilun.xu@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.