From: Chao Gao <chao.gao@intel.com>
To: linux-coco@lists.linux.dev, linux-kernel@vger.kernel.org,
x86@kernel.org, reinette.chatre@intel.com, ira.weiny@intel.com,
kai.huang@intel.com, dan.j.williams@intel.com,
yilun.xu@linux.intel.com, sagis@google.com,
vannapurve@google.com, paulmck@kernel.org, nik.borisov@suse.com
Cc: Chao Gao <chao.gao@intel.com>,
Farrah Chen <farrah.chen@intel.com>,
"Kirill A. Shutemov" <kas@kernel.org>,
Dave Hansen <dave.hansen@linux.intel.com>,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
"H. Peter Anvin" <hpa@zytor.com>
Subject: [PATCH v2 16/21] x86/virt/seamldr: Handle TDX Module update failures
Date: Tue, 30 Sep 2025 19:53:00 -0700 [thread overview]
Message-ID: <20251001025442.427697-17-chao.gao@intel.com> (raw)
In-Reply-To: <20251001025442.427697-1-chao.gao@intel.com>
Failures encountered after a successful module shutdown are unrecoverable,
e.g., there is no way to restore the old TDX Module.
All subsequent SEAMCALLs to the TDX Module will fail and so TDs have to be
killed.
Report failures through sysfs attributes and log a message to clarify that
SEAMCALL errors are expected in this situation.
To prevent TDX Module update failures, admins are encouraged to use the
user space tool [1] that will perform compatibility and integrity checks
that guarantee TDX Module update success (unless the system's update limit
is exceeded, but the kernel will prevent an update attempt in this case).
Signed-off-by: Chao Gao <chao.gao@intel.com>
Tested-by: Farrah Chen <farrah.chen@intel.com>
Link: https://github.com/intel/tdx-module-binaries/blob/main/version_select_and_load.py # [1]
---
arch/x86/virt/vmx/tdx/seamldr.c | 15 ++++++++++++++-
arch/x86/virt/vmx/tdx/tdx.c | 6 ++++++
arch/x86/virt/vmx/tdx/tdx.h | 1 +
drivers/virt/coco/tdx-host/tdx-host.c | 4 ++++
4 files changed, 25 insertions(+), 1 deletion(-)
diff --git a/arch/x86/virt/vmx/tdx/seamldr.c b/arch/x86/virt/vmx/tdx/seamldr.c
index b9daf11e1064..a5aff04a85b9 100644
--- a/arch/x86/virt/vmx/tdx/seamldr.c
+++ b/arch/x86/virt/vmx/tdx/seamldr.c
@@ -264,6 +264,14 @@ static void ack_state(void)
}
}
+static void print_update_failure_message(void)
+{
+ static atomic_t printed = ATOMIC_INIT(0);
+
+ if (atomic_inc_return(&printed) == 1)
+ pr_err("update failed, SEAMCALLs will report failure until TDs killed\n");
+}
+
/*
* See multi_cpu_stop() from where this multi-cpu state-machine was
* adopted, and the rationale for touch_nmi_watchdog()
@@ -293,8 +301,13 @@ static int do_seamldr_install_module(void *params)
break;
}
- if (ret)
+ if (ret) {
atomic_inc(&tdp_data.failed);
+ if (curstate > TDP_SHUTDOWN) {
+ tdx_module_set_error();
+ print_update_failure_message();
+ }
+ }
ack_state();
} else {
touch_nmi_watchdog();
diff --git a/arch/x86/virt/vmx/tdx/tdx.c b/arch/x86/virt/vmx/tdx/tdx.c
index 7019a149ec4b..26357be18fa9 100644
--- a/arch/x86/virt/vmx/tdx/tdx.c
+++ b/arch/x86/virt/vmx/tdx/tdx.c
@@ -1219,6 +1219,12 @@ int tdx_module_shutdown(void)
return 0;
}
+void tdx_module_set_error(void)
+{
+ /* Called from stop_machine(). no need to hold tdx_module_lock */
+ tdx_module_status = TDX_MODULE_ERROR;
+}
+
static bool is_pamt_page(unsigned long phys)
{
struct tdmr_info_list *tdmr_list = &tdx_tdmr_list;
diff --git a/arch/x86/virt/vmx/tdx/tdx.h b/arch/x86/virt/vmx/tdx/tdx.h
index 1c4da9540ae0..5b9a2d63808c 100644
--- a/arch/x86/virt/vmx/tdx/tdx.h
+++ b/arch/x86/virt/vmx/tdx/tdx.h
@@ -120,5 +120,6 @@ struct tdmr_info_list {
};
int tdx_module_shutdown(void);
+void tdx_module_set_error(void);
#endif
diff --git a/drivers/virt/coco/tdx-host/tdx-host.c b/drivers/virt/coco/tdx-host/tdx-host.c
index 418e90797689..47c5ba115993 100644
--- a/drivers/virt/coco/tdx-host/tdx-host.c
+++ b/drivers/virt/coco/tdx-host/tdx-host.c
@@ -37,6 +37,10 @@ static ssize_t version_show(struct device *dev, struct device_attribute *attr,
const struct tdx_sys_info *tdx_sysinfo = tdx_get_sysinfo();
const struct tdx_sys_info_version *ver;
+ /*
+ * Inform userspace that the TDX module isn't in a usable state,
+ * possibly due to a failed update.
+ */
if (!tdx_sysinfo)
return -ENXIO;
--
2.47.3
next prev parent reply other threads:[~2025-10-01 2:55 UTC|newest]
Thread overview: 75+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-01 2:52 [PATCH v2 00/21] Runtime TDX Module update support Chao Gao
2025-10-01 2:52 ` [PATCH v2 01/21] x86/virt/tdx: Print SEAMCALL leaf numbers in decimal Chao Gao
2025-10-01 2:52 ` [PATCH v2 02/21] x86/virt/tdx: Use %# prefix for hex values in SEAMCALL error messages Chao Gao
2025-10-01 2:52 ` [PATCH v2 03/21] x86/virt/tdx: Move low level SEAMCALL helpers out of <asm/tdx.h> Chao Gao
2025-10-01 2:52 ` [PATCH v2 04/21] x86/virt/tdx: Prepare to support P-SEAMLDR SEAMCALLs Chao Gao
2025-10-01 2:52 ` [PATCH v2 05/21] x86/virt/seamldr: Introduce a wrapper for " Chao Gao
2025-10-01 2:52 ` [PATCH v2 06/21] x86/virt/seamldr: Retrieve P-SEAMLDR information Chao Gao
2025-10-01 2:52 ` [PATCH v2 07/21] coco/tdx-host: Expose P-SEAMLDR information via sysfs Chao Gao
2025-10-30 21:54 ` Sagi Shahar
2025-10-30 23:05 ` dan.j.williams
2025-10-31 14:31 ` Sagi Shahar
2025-10-01 2:52 ` [PATCH v2 08/21] coco/tdx-host: Implement FW_UPLOAD sysfs ABI for TDX Module updates Chao Gao
2025-10-01 2:52 ` [PATCH v2 09/21] x86/virt/seamldr: Block TDX Module updates if any CPU is offline Chao Gao
2025-10-01 2:52 ` [PATCH v2 10/21] x86/virt/seamldr: Verify availability of slots for TDX Module updates Chao Gao
2025-10-01 2:52 ` [PATCH v2 11/21] x86/virt/seamldr: Allocate and populate a module update request Chao Gao
2025-10-01 2:52 ` [PATCH v2 12/21] x86/virt/seamldr: Introduce skeleton for TDX Module updates Chao Gao
2025-10-01 2:52 ` [PATCH v2 13/21] x86/virt/seamldr: Abort updates if errors occurred midway Chao Gao
2025-10-01 2:52 ` [PATCH v2 14/21] x86/virt/seamldr: Shut down the current TDX module Chao Gao
2025-10-01 2:52 ` [PATCH v2 15/21] x86/virt/tdx: Reset software states after TDX module shutdown Chao Gao
2025-10-01 2:53 ` Chao Gao [this message]
2025-10-28 2:53 ` [PATCH v2 16/21] x86/virt/seamldr: Handle TDX Module update failures Chao Gao
2025-10-01 2:53 ` [PATCH v2 17/21] x86/virt/seamldr: Install a new TDX Module Chao Gao
2025-10-01 2:53 ` [PATCH v2 18/21] x86/virt/seamldr: Do TDX per-CPU initialization after updates Chao Gao
2025-10-01 2:53 ` [PATCH v2 19/21] x86/virt/tdx: Establish contexts for the new TDX Module Chao Gao
2025-10-01 2:53 ` [PATCH v2 20/21] x86/virt/tdx: Update tdx_sysinfo and check features post-update Chao Gao
2025-10-01 2:53 ` [PATCH v2 21/21] x86/virt/tdx: Enable TDX Module runtime updates Chao Gao
2025-10-14 15:32 ` [PATCH v2 00/21] Runtime TDX Module update support Vishal Annapurve
2025-10-15 8:54 ` Reshetova, Elena
2025-10-15 14:19 ` Vishal Annapurve
2025-10-16 6:48 ` Reshetova, Elena
2025-10-15 15:02 ` Dave Hansen
2025-10-16 6:46 ` Reshetova, Elena
2025-10-16 17:47 ` Vishal Annapurve
2025-10-17 10:08 ` Reshetova, Elena
2025-10-18 0:01 ` Vishal Annapurve
2025-10-21 13:42 ` Reshetova, Elena
2025-10-22 7:14 ` Chao Gao
2025-10-22 15:42 ` Vishal Annapurve
2025-10-23 20:31 ` Vishal Annapurve
2025-10-23 21:10 ` Dave Hansen
2025-10-23 22:00 ` Vishal Annapurve
2025-10-24 7:43 ` Chao Gao
2025-10-24 18:02 ` Dave Hansen
2025-10-24 19:40 ` dan.j.williams
2025-10-24 20:00 ` Sean Christopherson
2025-10-24 20:14 ` Dave Hansen
2025-10-24 21:09 ` Vishal Annapurve
2025-10-24 20:13 ` Dave Hansen
2025-10-24 21:12 ` dan.j.williams
2025-10-24 21:19 ` Dave Hansen
2025-10-25 0:54 ` Vishal Annapurve
2025-10-25 1:42 ` dan.j.williams
2025-10-25 11:55 ` Vishal Annapurve
2025-10-25 12:01 ` Vishal Annapurve
2025-10-26 21:30 ` dan.j.williams
2025-10-26 22:01 ` Vishal Annapurve
2025-10-27 18:53 ` dan.j.williams
2025-10-28 0:42 ` Vishal Annapurve
2025-10-28 2:13 ` dan.j.williams
2025-10-28 17:00 ` Erdem Aktas
2025-10-29 0:56 ` Sean Christopherson
2025-10-29 2:17 ` dan.j.williams
2025-10-29 13:48 ` Sean Christopherson
2025-10-30 17:01 ` Vishal Annapurve
2025-10-31 2:53 ` Chao Gao
2025-11-19 22:44 ` Sagi Shahar
2025-11-20 2:47 ` Chao Gao
2025-10-28 23:48 ` Vishal Annapurve
2025-10-28 20:29 ` dan.j.williams
2025-10-28 20:32 ` dan.j.williams
2025-10-31 16:55 ` Sagi Shahar
2025-10-31 17:57 ` Vishal Annapurve
2025-11-01 2:18 ` Chao Gao
2025-11-01 2:05 ` Chao Gao
2025-11-12 14:09 ` Chao Gao
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=20251001025442.427697-17-chao.gao@intel.com \
--to=chao.gao@intel.com \
--cc=bp@alien8.de \
--cc=dan.j.williams@intel.com \
--cc=dave.hansen@linux.intel.com \
--cc=farrah.chen@intel.com \
--cc=hpa@zytor.com \
--cc=ira.weiny@intel.com \
--cc=kai.huang@intel.com \
--cc=kas@kernel.org \
--cc=linux-coco@lists.linux.dev \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=nik.borisov@suse.com \
--cc=paulmck@kernel.org \
--cc=reinette.chatre@intel.com \
--cc=sagis@google.com \
--cc=tglx@linutronix.de \
--cc=vannapurve@google.com \
--cc=x86@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).