* [PATCH v7 1/4] x86/sev: add SVSM vTPM probe/send_command functions
2025-04-10 13:51 [PATCH v7 0/4] Enlightened vTPM support for SVSM on SEV-SNP Stefano Garzarella
@ 2025-04-10 13:51 ` Stefano Garzarella
2025-04-10 13:51 ` [PATCH v7 2/4] svsm: add header with SVSM_VTPM_CMD helpers Stefano Garzarella
` (2 subsequent siblings)
3 siblings, 0 replies; 7+ messages in thread
From: Stefano Garzarella @ 2025-04-10 13:51 UTC (permalink / raw)
To: Borislav Petkov, Jarkko Sakkinen
Cc: Joerg Roedel, H. Peter Anvin, Dionna Glaze, Ingo Molnar,
linux-kernel, Dov Murik, linux-integrity, Dave Hansen,
Claudio Carvalho, Tom Lendacky, Thomas Gleixner, Peter Huewe,
Jason Gunthorpe, x86, James Bottomley, linux-coco,
Stefano Garzarella
From: Stefano Garzarella <sgarzare@redhat.com>
Add two new functions to probe and send commands to the SVSM vTPM.
They leverage the two calls defined by the AMD SVSM specification [1]
for the vTPM protocol: SVSM_VTPM_QUERY and SVSM_VTPM_CMD.
Expose snp_svsm_vtpm_send_command() to be used by a tpm driver.
[1] "Secure VM Service Module for SEV-SNP Guests"
Publication # 58019 Revision: 1.00
Co-developed-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Co-developed-by: Claudio Carvalho <cclaudio@linux.ibm.com>
Signed-off-by: Claudio Carvalho <cclaudio@linux.ibm.com>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
v7:
- avoided exporting snp_svsm_vtpm_probe() because we will only use it
internally
v5:
- added stubs when !CONFIG_AMD_MEM_ENCRYPT [Dionna]
- added Jarkko's R-b
v4:
- added Tom's R-b
- added functions documentation [Jarkko]
- simplified TPM_SEND_COMMAND check [Tom/Jarkko]
v3:
- removed link to the spec because those URLs are unstable [Borislav]
- squashed "x86/sev: add SVSM call macros for the vTPM protocol" patch
in this one [Borislav]
- slimmed down snp_svsm_vtpm_probe() [Borislav]
- removed features check and any print related [Tom]
---
arch/x86/include/asm/sev.h | 7 +++++
arch/x86/coco/sev/core.c | 58 ++++++++++++++++++++++++++++++++++++++
2 files changed, 65 insertions(+)
diff --git a/arch/x86/include/asm/sev.h b/arch/x86/include/asm/sev.h
index ba7999f66abe..d9ba035d8609 100644
--- a/arch/x86/include/asm/sev.h
+++ b/arch/x86/include/asm/sev.h
@@ -384,6 +384,10 @@ struct svsm_call {
#define SVSM_ATTEST_SERVICES 0
#define SVSM_ATTEST_SINGLE_SERVICE 1
+#define SVSM_VTPM_CALL(x) ((2ULL << 32) | (x))
+#define SVSM_VTPM_QUERY 0
+#define SVSM_VTPM_CMD 1
+
#ifdef CONFIG_AMD_MEM_ENCRYPT
extern u8 snp_vmpl;
@@ -481,6 +485,8 @@ void snp_msg_free(struct snp_msg_desc *mdesc);
int snp_send_guest_request(struct snp_msg_desc *mdesc, struct snp_guest_req *req,
struct snp_guest_request_ioctl *rio);
+int snp_svsm_vtpm_send_command(u8 *buffer);
+
void __init snp_secure_tsc_prepare(void);
void __init snp_secure_tsc_init(void);
@@ -524,6 +530,7 @@ static inline struct snp_msg_desc *snp_msg_alloc(void) { return NULL; }
static inline void snp_msg_free(struct snp_msg_desc *mdesc) { }
static inline int snp_send_guest_request(struct snp_msg_desc *mdesc, struct snp_guest_req *req,
struct snp_guest_request_ioctl *rio) { return -ENODEV; }
+static inline int snp_svsm_vtpm_send_command(u8 *buffer) { return -ENODEV; }
static inline void __init snp_secure_tsc_prepare(void) { }
static inline void __init snp_secure_tsc_init(void) { }
diff --git a/arch/x86/coco/sev/core.c b/arch/x86/coco/sev/core.c
index b0c1a7a57497..3bc5b47e7304 100644
--- a/arch/x86/coco/sev/core.c
+++ b/arch/x86/coco/sev/core.c
@@ -2625,6 +2625,64 @@ static int snp_issue_guest_request(struct snp_guest_req *req, struct snp_req_dat
return ret;
}
+/**
+ * snp_svsm_vtpm_probe() - Probe if SVSM provides a vTPM device
+ *
+ * This function checks that there is SVSM and that it supports at least
+ * TPM_SEND_COMMAND which is the only request we use so far.
+ *
+ * Return: true if the platform provides a vTPM SVSM device, false otherwise.
+ */
+static bool snp_svsm_vtpm_probe(void)
+{
+ struct svsm_call call = {};
+
+ /* The vTPM device is available only if a SVSM is present */
+ if (!snp_vmpl)
+ return false;
+
+ call.caa = svsm_get_caa();
+ call.rax = SVSM_VTPM_CALL(SVSM_VTPM_QUERY);
+
+ if (svsm_perform_call_protocol(&call))
+ return false;
+
+ /* Check platform commands contains TPM_SEND_COMMAND - platform command 8 */
+ return call.rcx_out & BIT_ULL(8);
+}
+
+/**
+ * snp_svsm_vtpm_send_command() - execute a vTPM operation on SVSM
+ * @buffer: A buffer used to both send the command and receive the response.
+ *
+ * This function executes a SVSM_VTPM_CMD call as defined by
+ * "Secure VM Service Module for SEV-SNP Guests" Publication # 58019 Revision: 1.00
+ *
+ * All command request/response buffers have a common structure as specified by
+ * the following table:
+ * Byte Size In/Out Description
+ * Offset (Bytes)
+ * 0x000 4 In Platform command
+ * Out Platform command response size
+ *
+ * Each command can build upon this common request/response structure to create
+ * a structure specific to the command.
+ * See include/linux/tpm_svsm.h for more details.
+ *
+ * Return: 0 on success, -errno on failure
+ */
+int snp_svsm_vtpm_send_command(u8 *buffer)
+{
+ struct svsm_call call = {};
+
+ call.caa = svsm_get_caa();
+ call.rax = SVSM_VTPM_CALL(SVSM_VTPM_CMD);
+ call.rcx = __pa(buffer);
+
+ return svsm_perform_call_protocol(&call);
+}
+EXPORT_SYMBOL_GPL(snp_svsm_vtpm_send_command);
+
static struct platform_device sev_guest_device = {
.name = "sev-guest",
.id = -1,
--
2.49.0
^ permalink raw reply [flat|nested] 7+ messages in thread* [PATCH v7 2/4] svsm: add header with SVSM_VTPM_CMD helpers
2025-04-10 13:51 [PATCH v7 0/4] Enlightened vTPM support for SVSM on SEV-SNP Stefano Garzarella
2025-04-10 13:51 ` [PATCH v7 1/4] x86/sev: add SVSM vTPM probe/send_command functions Stefano Garzarella
@ 2025-04-10 13:51 ` Stefano Garzarella
2025-04-10 13:51 ` [PATCH v7 3/4] tpm: add SNP SVSM vTPM driver Stefano Garzarella
2025-04-10 13:51 ` [PATCH v7 4/4] x86/sev: register tpm-svsm platform device Stefano Garzarella
3 siblings, 0 replies; 7+ messages in thread
From: Stefano Garzarella @ 2025-04-10 13:51 UTC (permalink / raw)
To: Borislav Petkov, Jarkko Sakkinen
Cc: Joerg Roedel, H. Peter Anvin, Dionna Glaze, Ingo Molnar,
linux-kernel, Dov Murik, linux-integrity, Dave Hansen,
Claudio Carvalho, Tom Lendacky, Thomas Gleixner, Peter Huewe,
Jason Gunthorpe, x86, James Bottomley, linux-coco,
Stefano Garzarella
From: Stefano Garzarella <sgarzare@redhat.com>
Helpers for the SVSM_VTPM_CMD calls used by the vTPM protocol defined by
the AMD SVSM spec [1].
The vTPM protocol follows the Official TPM 2.0 Reference Implementation
(originally by Microsoft, now part of the TCG) simulator protocol.
[1] "Secure VM Service Module for SEV-SNP Guests"
Publication # 58019 Revision: 1.00
Co-developed-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Co-developed-by: Claudio Carvalho <cclaudio@linux.ibm.com>
Signed-off-by: Claudio Carvalho <cclaudio@linux.ibm.com>
Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
v5:
- added Jarkko's R-b
v4:
- used svsm_vtpm_ prefix consistently [Jarkko]
- removed __packed where not needed [Jarkko]
- expanded headers to avoid obfuscation [Jarkko]
- used `buf` instead of `inbuf`/`outbuf` [Jarkko]
- added more documentation quoting the specification
- removed TPM_* macros since we only use TPM_SEND_COMMAND in one place
and don't want dependencies on external headers, but put the value
directly as specified in the AMD SVSM specification
- header renamed in tpm_svsm.h so it will fall under TPM DEVICE DRIVER
section [Borislav, Jarkko]
v3:
- renamed header and prefix to make clear it's related to the SVSM vTPM
protocol
- renamed fill/parse functions [Tom]
- removed link to the spec because those URLs are unstable [Borislav]
---
include/linux/tpm_svsm.h | 149 +++++++++++++++++++++++++++++++++++++++
1 file changed, 149 insertions(+)
create mode 100644 include/linux/tpm_svsm.h
diff --git a/include/linux/tpm_svsm.h b/include/linux/tpm_svsm.h
new file mode 100644
index 000000000000..38e341f9761a
--- /dev/null
+++ b/include/linux/tpm_svsm.h
@@ -0,0 +1,149 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2023 James.Bottomley@HansenPartnership.com
+ * Copyright (C) 2025 Red Hat, Inc. All Rights Reserved.
+ *
+ * Helpers for the SVSM_VTPM_CMD calls used by the vTPM protocol defined by the
+ * AMD SVSM spec [1].
+ *
+ * The vTPM protocol follows the Official TPM 2.0 Reference Implementation
+ * (originally by Microsoft, now part of the TCG) simulator protocol.
+ *
+ * [1] "Secure VM Service Module for SEV-SNP Guests"
+ * Publication # 58019 Revision: 1.00
+ */
+#ifndef _TPM_SVSM_H_
+#define _TPM_SVSM_H_
+
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+
+#define SVSM_VTPM_MAX_BUFFER 4096 /* max req/resp buffer size */
+
+/**
+ * struct svsm_vtpm_request - Generic request for single word command
+ * @cmd: The command to send
+ *
+ * Defined by AMD SVSM spec [1] in section "8.2 SVSM_VTPM_CMD Call" -
+ * Table 15: vTPM Common Request/Response Structure
+ * Byte Size In/Out Description
+ * Offset (Bytes)
+ * 0x000 4 In Platform command
+ * Out Platform command response size
+ */
+struct svsm_vtpm_request {
+ u32 cmd;
+};
+
+/**
+ * struct svsm_vtpm_response - Generic response
+ * @size: The response size (zero if nothing follows)
+ *
+ * Defined by AMD SVSM spec [1] in section "8.2 SVSM_VTPM_CMD Call" -
+ * Table 15: vTPM Common Request/Response Structure
+ * Byte Size In/Out Description
+ * Offset (Bytes)
+ * 0x000 4 In Platform command
+ * Out Platform command response size
+ *
+ * Note: most TCG Simulator commands simply return zero here with no indication
+ * of success or failure.
+ */
+struct svsm_vtpm_response {
+ u32 size;
+};
+
+/**
+ * struct svsm_vtpm_cmd_request - Structure for a TPM_SEND_COMMAND request
+ * @cmd: The command to send (must be TPM_SEND_COMMAND)
+ * @locality: The locality
+ * @buf_size: The size of the input buffer following
+ * @buf: A buffer of size buf_size
+ *
+ * Defined by AMD SVSM spec [1] in section "8.2 SVSM_VTPM_CMD Call" -
+ * Table 16: TPM_SEND_COMMAND Request Structure
+ * Byte Size Meaning
+ * Offset (Bytes)
+ * 0x000 4 Platform command (8)
+ * 0x004 1 Locality (must-be-0)
+ * 0x005 4 TPM Command size (in bytes)
+ * 0x009 Variable TPM Command
+ *
+ * Note: the TCG Simulator expects @buf_size to be equal to the size of the
+ * specific TPM command, otherwise an TPM_RC_COMMAND_SIZE error is returned.
+ */
+struct svsm_vtpm_cmd_request {
+ u32 cmd;
+ u8 locality;
+ u32 buf_size;
+ u8 buf[];
+} __packed;
+
+/**
+ * struct svsm_vtpm_cmd_response - Structure for a TPM_SEND_COMMAND response
+ * @buf_size: The size of the output buffer following
+ * @buf: A buffer of size buf_size
+ *
+ * Defined by AMD SVSM spec [1] in section "8.2 SVSM_VTPM_CMD Call" -
+ * Table 17: TPM_SEND_COMMAND Response Structure
+ * Byte Size Meaning
+ * Offset (Bytes)
+ * 0x000 4 Response size (in bytes)
+ * 0x004 Variable Response
+ */
+struct svsm_vtpm_cmd_response {
+ u32 buf_size;
+ u8 buf[];
+};
+
+/**
+ * svsm_vtpm_cmd_request_fill() - Fill a TPM_SEND_COMMAND request to be sent to SVSM
+ * @req: The struct svsm_vtpm_cmd_request to fill
+ * @locality: The locality
+ * @buf: The buffer from where to copy the payload of the command
+ * @len: The size of the buffer
+ *
+ * Return: 0 on success, negative error code on failure.
+ */
+static inline int
+svsm_vtpm_cmd_request_fill(struct svsm_vtpm_cmd_request *req, u8 locality,
+ const u8 *buf, size_t len)
+{
+ if (len > SVSM_VTPM_MAX_BUFFER - sizeof(*req))
+ return -EINVAL;
+
+ req->cmd = 8; /* TPM_SEND_COMMAND */
+ req->locality = locality;
+ req->buf_size = len;
+
+ memcpy(req->buf, buf, len);
+
+ return 0;
+}
+
+/**
+ * svsm_vtpm_cmd_response_parse() - Parse a TPM_SEND_COMMAND response received from SVSM
+ * @resp: The struct svsm_vtpm_cmd_response to parse
+ * @buf: The buffer where to copy the response
+ * @len: The size of the buffer
+ *
+ * Return: buffer size filled with the response on success, negative error
+ * code on failure.
+ */
+static inline int
+svsm_vtpm_cmd_response_parse(const struct svsm_vtpm_cmd_response *resp, u8 *buf,
+ size_t len)
+{
+ if (len < resp->buf_size)
+ return -E2BIG;
+
+ if (resp->buf_size > SVSM_VTPM_MAX_BUFFER - sizeof(*resp))
+ return -EINVAL; // Invalid response from the platform TPM
+
+ memcpy(buf, resp->buf, resp->buf_size);
+
+ return resp->buf_size;
+}
+
+#endif /* _TPM_SVSM_H_ */
--
2.49.0
^ permalink raw reply [flat|nested] 7+ messages in thread* [PATCH v7 3/4] tpm: add SNP SVSM vTPM driver
2025-04-10 13:51 [PATCH v7 0/4] Enlightened vTPM support for SVSM on SEV-SNP Stefano Garzarella
2025-04-10 13:51 ` [PATCH v7 1/4] x86/sev: add SVSM vTPM probe/send_command functions Stefano Garzarella
2025-04-10 13:51 ` [PATCH v7 2/4] svsm: add header with SVSM_VTPM_CMD helpers Stefano Garzarella
@ 2025-04-10 13:51 ` Stefano Garzarella
2025-04-11 10:15 ` [tip: x86/sev] tpm: Add " tip-bot2 for Stefano Garzarella
2025-04-10 13:51 ` [PATCH v7 4/4] x86/sev: register tpm-svsm platform device Stefano Garzarella
3 siblings, 1 reply; 7+ messages in thread
From: Stefano Garzarella @ 2025-04-10 13:51 UTC (permalink / raw)
To: Borislav Petkov, Jarkko Sakkinen
Cc: Joerg Roedel, H. Peter Anvin, Dionna Glaze, Ingo Molnar,
linux-kernel, Dov Murik, linux-integrity, Dave Hansen,
Claudio Carvalho, Tom Lendacky, Thomas Gleixner, Peter Huewe,
Jason Gunthorpe, x86, James Bottomley, linux-coco,
Stefano Garzarella
From: Stefano Garzarella <sgarzare@redhat.com>
Add driver for the vTPM defined by the AMD SVSM spec [1].
The specification defines a protocol that a SEV-SNP guest OS can use to
discover and talk to a vTPM emulated by the Secure VM Service Module (SVSM)
in the guest context, but at a more privileged level (VMPL0).
The new tpm-svsm platform driver uses API exposed by x86/sev to send
commands and receive responses.
The device cannot be hot-plugged/unplugged as it is emulated by the
platform, so we can use module_platform_driver_probe(). The device will
be registered by the platform only when it's available, so the probe
function just needs to setup the tpm_chip.
This device does not support interrupts and sends responses to commands
synchronously.
In order to have .recv() called just after .send() in tpm_try_transmit(),
the .status() callback is not implemented as recently supported by
commit 980a573621ea ("tpm: Make chip->{status,cancel,req_canceled} opt").
[1] "Secure VM Service Module for SEV-SNP Guests"
Publication # 58019 Revision: 1.00
Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
v7:
- added Jarkko's R-b
- avoided to call snp_svsm_vtpm_probe() since the device will be
registered by SEV core only if the device is detected [Borislav]
- updated commit description
v6:
- removed the `locality` field (set to 0) and the FIXME comment [Jarkko]
v5:
- removed cancel/status/req_* ops after rebase on master that cotains
commit 980a573621ea ("tpm: Make chip->{status,cancel,req_canceled} opt")
v4:
- moved "asm" includes after the "linux" includes [Tom]
- allocated buffer separately [Tom/Jarkko/Jason]
v3:
- removed send_recv() ops and followed the ftpm driver implementing .status,
.req_complete_mask, .req_complete_val, etc. [Jarkko]
- removed link to the spec because those URLs are unstable [Borislav]
---
drivers/char/tpm/tpm_svsm.c | 125 ++++++++++++++++++++++++++++++++++++
drivers/char/tpm/Kconfig | 10 +++
drivers/char/tpm/Makefile | 1 +
3 files changed, 136 insertions(+)
create mode 100644 drivers/char/tpm/tpm_svsm.c
diff --git a/drivers/char/tpm/tpm_svsm.c b/drivers/char/tpm/tpm_svsm.c
new file mode 100644
index 000000000000..4280edf427d6
--- /dev/null
+++ b/drivers/char/tpm/tpm_svsm.c
@@ -0,0 +1,125 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2025 Red Hat, Inc. All Rights Reserved.
+ *
+ * Driver for the vTPM defined by the AMD SVSM spec [1].
+ *
+ * The specification defines a protocol that a SEV-SNP guest OS can use to
+ * discover and talk to a vTPM emulated by the Secure VM Service Module (SVSM)
+ * in the guest context, but at a more privileged level (usually VMPL0).
+ *
+ * [1] "Secure VM Service Module for SEV-SNP Guests"
+ * Publication # 58019 Revision: 1.00
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/tpm_svsm.h>
+
+#include <asm/sev.h>
+
+#include "tpm.h"
+
+struct tpm_svsm_priv {
+ void *buffer;
+};
+
+static int tpm_svsm_send(struct tpm_chip *chip, u8 *buf, size_t len)
+{
+ struct tpm_svsm_priv *priv = dev_get_drvdata(&chip->dev);
+ int ret;
+
+ ret = svsm_vtpm_cmd_request_fill(priv->buffer, 0, buf, len);
+ if (ret)
+ return ret;
+
+ /*
+ * The SVSM call uses the same buffer for the command and for the
+ * response, so after this call, the buffer will contain the response
+ * that can be used by .recv() op.
+ */
+ return snp_svsm_vtpm_send_command(priv->buffer);
+}
+
+static int tpm_svsm_recv(struct tpm_chip *chip, u8 *buf, size_t len)
+{
+ struct tpm_svsm_priv *priv = dev_get_drvdata(&chip->dev);
+
+ /*
+ * The internal buffer contains the response after we send the command
+ * to SVSM.
+ */
+ return svsm_vtpm_cmd_response_parse(priv->buffer, buf, len);
+}
+
+static struct tpm_class_ops tpm_chip_ops = {
+ .flags = TPM_OPS_AUTO_STARTUP,
+ .recv = tpm_svsm_recv,
+ .send = tpm_svsm_send,
+};
+
+static int __init tpm_svsm_probe(struct platform_device *pdev)
+{
+ struct device *dev = &pdev->dev;
+ struct tpm_svsm_priv *priv;
+ struct tpm_chip *chip;
+ int err;
+
+ priv = devm_kmalloc(dev, sizeof(*priv), GFP_KERNEL);
+ if (!priv)
+ return -ENOMEM;
+
+ /*
+ * The maximum buffer supported is one page (see SVSM_VTPM_MAX_BUFFER
+ * in tpm_svsm.h).
+ */
+ priv->buffer = (void *)devm_get_free_pages(dev, GFP_KERNEL, 0);
+ if (!priv->buffer)
+ return -ENOMEM;
+
+ chip = tpmm_chip_alloc(dev, &tpm_chip_ops);
+ if (IS_ERR(chip))
+ return PTR_ERR(chip);
+
+ dev_set_drvdata(&chip->dev, priv);
+
+ err = tpm2_probe(chip);
+ if (err)
+ return err;
+
+ err = tpm_chip_register(chip);
+ if (err)
+ return err;
+
+ dev_info(dev, "SNP SVSM vTPM %s device\n",
+ (chip->flags & TPM_CHIP_FLAG_TPM2) ? "2.0" : "1.2");
+
+ return 0;
+}
+
+static void __exit tpm_svsm_remove(struct platform_device *pdev)
+{
+ struct tpm_chip *chip = platform_get_drvdata(pdev);
+
+ tpm_chip_unregister(chip);
+}
+
+/*
+ * tpm_svsm_remove() lives in .exit.text. For drivers registered via
+ * module_platform_driver_probe() this is ok because they cannot get unbound
+ * at runtime. So mark the driver struct with __refdata to prevent modpost
+ * triggering a section mismatch warning.
+ */
+static struct platform_driver tpm_svsm_driver __refdata = {
+ .remove = __exit_p(tpm_svsm_remove),
+ .driver = {
+ .name = "tpm-svsm",
+ },
+};
+
+module_platform_driver_probe(tpm_svsm_driver, tpm_svsm_probe);
+
+MODULE_DESCRIPTION("SNP SVSM vTPM Driver");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:tpm-svsm");
diff --git a/drivers/char/tpm/Kconfig b/drivers/char/tpm/Kconfig
index fe4f3a609934..dddd702b2454 100644
--- a/drivers/char/tpm/Kconfig
+++ b/drivers/char/tpm/Kconfig
@@ -234,5 +234,15 @@ config TCG_FTPM_TEE
help
This driver proxies for firmware TPM running in TEE.
+config TCG_SVSM
+ tristate "SNP SVSM vTPM interface"
+ depends on AMD_MEM_ENCRYPT
+ help
+ This is a driver for the AMD SVSM vTPM protocol that a SEV-SNP guest
+ OS can use to discover and talk to a vTPM emulated by the Secure VM
+ Service Module (SVSM) in the guest context, but at a more privileged
+ level (usually VMPL0). To compile this driver as a module, choose M
+ here; the module will be called tpm_svsm.
+
source "drivers/char/tpm/st33zp24/Kconfig"
endif # TCG_TPM
diff --git a/drivers/char/tpm/Makefile b/drivers/char/tpm/Makefile
index 2b004df8c04b..9de1b3ea34a9 100644
--- a/drivers/char/tpm/Makefile
+++ b/drivers/char/tpm/Makefile
@@ -45,3 +45,4 @@ obj-$(CONFIG_TCG_CRB) += tpm_crb.o
obj-$(CONFIG_TCG_ARM_CRB_FFA) += tpm_crb_ffa.o
obj-$(CONFIG_TCG_VTPM_PROXY) += tpm_vtpm_proxy.o
obj-$(CONFIG_TCG_FTPM_TEE) += tpm_ftpm_tee.o
+obj-$(CONFIG_TCG_SVSM) += tpm_svsm.o
--
2.49.0
^ permalink raw reply [flat|nested] 7+ messages in thread* [tip: x86/sev] tpm: Add SNP SVSM vTPM driver
2025-04-10 13:51 ` [PATCH v7 3/4] tpm: add SNP SVSM vTPM driver Stefano Garzarella
@ 2025-04-11 10:15 ` tip-bot2 for Stefano Garzarella
0 siblings, 0 replies; 7+ messages in thread
From: tip-bot2 for Stefano Garzarella @ 2025-04-11 10:15 UTC (permalink / raw)
To: linux-tip-commits
Cc: Stefano Garzarella, Borislav Petkov (AMD),
Jarkko Sakkinen, x86, linux-kernel
The following commit has been merged into the x86/sev branch of tip:
Commit-ID: 93b7c6b3ce918c3d24de82dcff7a87b8bd631b2e
Gitweb: https://git.kernel.org/tip/93b7c6b3ce918c3d24de82dcff7a87b8bd631b2e
Author: Stefano Garzarella <sgarzare@redhat.com>
AuthorDate: Thu, 10 Apr 2025 15:51:15 +02:00
Committer: Borislav Petkov (AMD) <bp@alien8.de>
CommitterDate: Thu, 10 Apr 2025 16:24:29 +02:00
tpm: Add SNP SVSM vTPM driver
Add driver for the vTPM defined by the AMD SVSM spec [1].
The specification defines a protocol that a SEV-SNP guest OS can use to
discover and talk to a vTPM emulated by the Secure VM Service Module (SVSM) in
the guest context, but at a more privileged level (VMPL0).
The new tpm-svsm platform driver uses API exposed by the x86/sev core
implementation interface to a SVSM to send commands and receive responses.
The device cannot be hot-plugged/unplugged as it is emulated by the platform,
so module_platform_driver_probe() can be used. The device will be registered
by the platform only when it's available, so the probe function just needs to
setup the tpm_chip.
This device does not support interrupts and sends responses to commands
synchronously.
In order to have .recv() called just after .send() in tpm_try_transmit(), the
.status() callback is not implemented as recently supported by commit
980a573621ea ("tpm: Make chip->{status,cancel,req_canceled} opt").
[1] "Secure VM Service Module for SEV-SNP Guests"
Publication # 58019 Revision: 1.00
[ bp: Massage commit message. ]
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org>
Link: https://lore.kernel.org/r/20250410135118.133240-4-sgarzare@redhat.com
---
drivers/char/tpm/Kconfig | 10 +++-
drivers/char/tpm/Makefile | 1 +-
drivers/char/tpm/tpm_svsm.c | 125 +++++++++++++++++++++++++++++++++++-
3 files changed, 136 insertions(+)
create mode 100644 drivers/char/tpm/tpm_svsm.c
diff --git a/drivers/char/tpm/Kconfig b/drivers/char/tpm/Kconfig
index fe4f3a6..dddd702 100644
--- a/drivers/char/tpm/Kconfig
+++ b/drivers/char/tpm/Kconfig
@@ -234,5 +234,15 @@ config TCG_FTPM_TEE
help
This driver proxies for firmware TPM running in TEE.
+config TCG_SVSM
+ tristate "SNP SVSM vTPM interface"
+ depends on AMD_MEM_ENCRYPT
+ help
+ This is a driver for the AMD SVSM vTPM protocol that a SEV-SNP guest
+ OS can use to discover and talk to a vTPM emulated by the Secure VM
+ Service Module (SVSM) in the guest context, but at a more privileged
+ level (usually VMPL0). To compile this driver as a module, choose M
+ here; the module will be called tpm_svsm.
+
source "drivers/char/tpm/st33zp24/Kconfig"
endif # TCG_TPM
diff --git a/drivers/char/tpm/Makefile b/drivers/char/tpm/Makefile
index 2b004df..9de1b3e 100644
--- a/drivers/char/tpm/Makefile
+++ b/drivers/char/tpm/Makefile
@@ -45,3 +45,4 @@ obj-$(CONFIG_TCG_CRB) += tpm_crb.o
obj-$(CONFIG_TCG_ARM_CRB_FFA) += tpm_crb_ffa.o
obj-$(CONFIG_TCG_VTPM_PROXY) += tpm_vtpm_proxy.o
obj-$(CONFIG_TCG_FTPM_TEE) += tpm_ftpm_tee.o
+obj-$(CONFIG_TCG_SVSM) += tpm_svsm.o
diff --git a/drivers/char/tpm/tpm_svsm.c b/drivers/char/tpm/tpm_svsm.c
new file mode 100644
index 0000000..4280edf
--- /dev/null
+++ b/drivers/char/tpm/tpm_svsm.c
@@ -0,0 +1,125 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2025 Red Hat, Inc. All Rights Reserved.
+ *
+ * Driver for the vTPM defined by the AMD SVSM spec [1].
+ *
+ * The specification defines a protocol that a SEV-SNP guest OS can use to
+ * discover and talk to a vTPM emulated by the Secure VM Service Module (SVSM)
+ * in the guest context, but at a more privileged level (usually VMPL0).
+ *
+ * [1] "Secure VM Service Module for SEV-SNP Guests"
+ * Publication # 58019 Revision: 1.00
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/tpm_svsm.h>
+
+#include <asm/sev.h>
+
+#include "tpm.h"
+
+struct tpm_svsm_priv {
+ void *buffer;
+};
+
+static int tpm_svsm_send(struct tpm_chip *chip, u8 *buf, size_t len)
+{
+ struct tpm_svsm_priv *priv = dev_get_drvdata(&chip->dev);
+ int ret;
+
+ ret = svsm_vtpm_cmd_request_fill(priv->buffer, 0, buf, len);
+ if (ret)
+ return ret;
+
+ /*
+ * The SVSM call uses the same buffer for the command and for the
+ * response, so after this call, the buffer will contain the response
+ * that can be used by .recv() op.
+ */
+ return snp_svsm_vtpm_send_command(priv->buffer);
+}
+
+static int tpm_svsm_recv(struct tpm_chip *chip, u8 *buf, size_t len)
+{
+ struct tpm_svsm_priv *priv = dev_get_drvdata(&chip->dev);
+
+ /*
+ * The internal buffer contains the response after we send the command
+ * to SVSM.
+ */
+ return svsm_vtpm_cmd_response_parse(priv->buffer, buf, len);
+}
+
+static struct tpm_class_ops tpm_chip_ops = {
+ .flags = TPM_OPS_AUTO_STARTUP,
+ .recv = tpm_svsm_recv,
+ .send = tpm_svsm_send,
+};
+
+static int __init tpm_svsm_probe(struct platform_device *pdev)
+{
+ struct device *dev = &pdev->dev;
+ struct tpm_svsm_priv *priv;
+ struct tpm_chip *chip;
+ int err;
+
+ priv = devm_kmalloc(dev, sizeof(*priv), GFP_KERNEL);
+ if (!priv)
+ return -ENOMEM;
+
+ /*
+ * The maximum buffer supported is one page (see SVSM_VTPM_MAX_BUFFER
+ * in tpm_svsm.h).
+ */
+ priv->buffer = (void *)devm_get_free_pages(dev, GFP_KERNEL, 0);
+ if (!priv->buffer)
+ return -ENOMEM;
+
+ chip = tpmm_chip_alloc(dev, &tpm_chip_ops);
+ if (IS_ERR(chip))
+ return PTR_ERR(chip);
+
+ dev_set_drvdata(&chip->dev, priv);
+
+ err = tpm2_probe(chip);
+ if (err)
+ return err;
+
+ err = tpm_chip_register(chip);
+ if (err)
+ return err;
+
+ dev_info(dev, "SNP SVSM vTPM %s device\n",
+ (chip->flags & TPM_CHIP_FLAG_TPM2) ? "2.0" : "1.2");
+
+ return 0;
+}
+
+static void __exit tpm_svsm_remove(struct platform_device *pdev)
+{
+ struct tpm_chip *chip = platform_get_drvdata(pdev);
+
+ tpm_chip_unregister(chip);
+}
+
+/*
+ * tpm_svsm_remove() lives in .exit.text. For drivers registered via
+ * module_platform_driver_probe() this is ok because they cannot get unbound
+ * at runtime. So mark the driver struct with __refdata to prevent modpost
+ * triggering a section mismatch warning.
+ */
+static struct platform_driver tpm_svsm_driver __refdata = {
+ .remove = __exit_p(tpm_svsm_remove),
+ .driver = {
+ .name = "tpm-svsm",
+ },
+};
+
+module_platform_driver_probe(tpm_svsm_driver, tpm_svsm_probe);
+
+MODULE_DESCRIPTION("SNP SVSM vTPM Driver");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:tpm-svsm");
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH v7 4/4] x86/sev: register tpm-svsm platform device
2025-04-10 13:51 [PATCH v7 0/4] Enlightened vTPM support for SVSM on SEV-SNP Stefano Garzarella
` (2 preceding siblings ...)
2025-04-10 13:51 ` [PATCH v7 3/4] tpm: add SNP SVSM vTPM driver Stefano Garzarella
@ 2025-04-10 13:51 ` Stefano Garzarella
2025-04-11 10:15 ` [tip: x86/sev] x86/sev: Register " tip-bot2 for Stefano Garzarella
3 siblings, 1 reply; 7+ messages in thread
From: Stefano Garzarella @ 2025-04-10 13:51 UTC (permalink / raw)
To: Borislav Petkov, Jarkko Sakkinen
Cc: Joerg Roedel, H. Peter Anvin, Dionna Glaze, Ingo Molnar,
linux-kernel, Dov Murik, linux-integrity, Dave Hansen,
Claudio Carvalho, Tom Lendacky, Thomas Gleixner, Peter Huewe,
Jason Gunthorpe, x86, James Bottomley, linux-coco,
Stefano Garzarella
From: Stefano Garzarella <sgarzare@redhat.com>
SNP platform can provide a vTPM device emulated by SVSM.
The "tpm-svsm" device can be handled by the platform driver added
by the previous commit in drivers/char/tpm/tpm_svsm.c
Register the platform device only when SVSM is available and it
supports vTPM commands as checked by snp_svsm_vtpm_probe().
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
v7:
- added Jarkko's R-b
- call snp_svsm_vtpm_probe() before registering the device [Borislav]
- s/device/devices in pr_info [Tom]
- updated commit description
v6:
- added Tom's R-b
v4:
- explained better why we register it anyway in the commit message
---
arch/x86/coco/sev/core.c | 11 ++++++++++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/arch/x86/coco/sev/core.c b/arch/x86/coco/sev/core.c
index 3bc5b47e7304..54272c9777cf 100644
--- a/arch/x86/coco/sev/core.c
+++ b/arch/x86/coco/sev/core.c
@@ -2688,6 +2688,11 @@ static struct platform_device sev_guest_device = {
.id = -1,
};
+static struct platform_device tpm_svsm_device = {
+ .name = "tpm-svsm",
+ .id = -1,
+};
+
static int __init snp_init_platform_device(void)
{
if (!cc_platform_has(CC_ATTR_GUEST_SEV_SNP))
@@ -2696,7 +2701,11 @@ static int __init snp_init_platform_device(void)
if (platform_device_register(&sev_guest_device))
return -ENODEV;
- pr_info("SNP guest platform device initialized.\n");
+ if (snp_svsm_vtpm_probe() &&
+ platform_device_register(&tpm_svsm_device))
+ return -ENODEV;
+
+ pr_info("SNP guest platform devices initialized.\n");
return 0;
}
device_initcall(snp_init_platform_device);
--
2.49.0
^ permalink raw reply [flat|nested] 7+ messages in thread* [tip: x86/sev] x86/sev: Register tpm-svsm platform device
2025-04-10 13:51 ` [PATCH v7 4/4] x86/sev: register tpm-svsm platform device Stefano Garzarella
@ 2025-04-11 10:15 ` tip-bot2 for Stefano Garzarella
0 siblings, 0 replies; 7+ messages in thread
From: tip-bot2 for Stefano Garzarella @ 2025-04-11 10:15 UTC (permalink / raw)
To: linux-tip-commits
Cc: Stefano Garzarella, Borislav Petkov (AMD),
Tom Lendacky, Jarkko Sakkinen, x86, linux-kernel
The following commit has been merged into the x86/sev branch of tip:
Commit-ID: e396dd85172c6098e3b70b17e91424edc7bb2d8f
Gitweb: https://git.kernel.org/tip/e396dd85172c6098e3b70b17e91424edc7bb2d8f
Author: Stefano Garzarella <sgarzare@redhat.com>
AuthorDate: Thu, 10 Apr 2025 15:51:16 +02:00
Committer: Borislav Petkov (AMD) <bp@alien8.de>
CommitterDate: Thu, 10 Apr 2025 16:25:33 +02:00
x86/sev: Register tpm-svsm platform device
SNP platform can provide a vTPM device emulated by SVSM.
The "tpm-svsm" device can be handled by the platform driver registered by the
x86/sev core code.
Register the platform device only when SVSM is available and it supports vTPM
commands as checked by snp_svsm_vtpm_probe().
[ bp: Massage commit message. ]
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Reviewed-by: Jarkko Sakkinen <jarkko@kernel.org>
Link: https://lore.kernel.org/r/20250410135118.133240-5-sgarzare@redhat.com
---
arch/x86/coco/sev/core.c | 11 ++++++++++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/arch/x86/coco/sev/core.c b/arch/x86/coco/sev/core.c
index ecd09da..654a4cc 100644
--- a/arch/x86/coco/sev/core.c
+++ b/arch/x86/coco/sev/core.c
@@ -2688,6 +2688,11 @@ static struct platform_device sev_guest_device = {
.id = -1,
};
+static struct platform_device tpm_svsm_device = {
+ .name = "tpm-svsm",
+ .id = -1,
+};
+
static int __init snp_init_platform_device(void)
{
if (!cc_platform_has(CC_ATTR_GUEST_SEV_SNP))
@@ -2696,7 +2701,11 @@ static int __init snp_init_platform_device(void)
if (platform_device_register(&sev_guest_device))
return -ENODEV;
- pr_info("SNP guest platform device initialized.\n");
+ if (snp_svsm_vtpm_probe() &&
+ platform_device_register(&tpm_svsm_device))
+ return -ENODEV;
+
+ pr_info("SNP guest platform devices initialized.\n");
return 0;
}
device_initcall(snp_init_platform_device);
^ permalink raw reply [flat|nested] 7+ messages in thread