Skip to contentSkip to conversation

TPM: STMicroelectronics u-boot driver I2C

PATCH New Mathias leblanc 3 participants 3 messages Apr 22, 2013 – Apr 23, 2013 Kbuild

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 TPM: STMicroelectronics u-boot driver I2C

From: Mathias Leblanc <[email protected]>

* STMicroelectronics version 1.2.0, Copyright (C) 2013
* This is free software, and you are welcome to redistribute it
* under certain conditions.

This is the driver for TPM chip from ST Microelectronics.

If you have a TPM security chip from STMicroelectronics working with an I2C, read the README file and add the correct defines regarding the tpm in the configuration file of your board.
This file is located in include/configs/your_board.h

The driver will be accessible from within uboot terminal.

---
 Makefile                       |    2 +-
 README                         |   14 +-
 common/cmd_tpm.c               |  170 +++++++++++-
 drivers/tpm/Makefile           |    1 +
 drivers/tpm/compatibility.h    |   52 ++++
 drivers/tpm/tis_i2c.c          |  248 +++++++++++++++++
 drivers/tpm/tpm.c              |  477 ++++++++++++++++++++++++++++++++
 drivers/tpm/tpm.h              |  170 +++++++++++
 drivers/tpm/tpm_i2c_st.c       |  599 ++++++++++++++++++++++++++++++++++++++++
 include/configs/coreboot.h     |    4 +-
 include/configs/omap3_beagle.h |    8 +
 include/tpm.h                  |   17 ++
 12 files changed, 1757 insertions(+), 5 deletions(-)
 create mode 100644 drivers/tpm/compatibility.h
 create mode 100644 drivers/tpm/tis_i2c.c
 create mode 100644 drivers/tpm/tpm.c
 create mode 100644 drivers/tpm/tpm.h
 create mode 100644 drivers/tpm/tpm_i2c_st.c
Makefile +1 -1
index 12763ce..ef954a4 100644
@@ -314,7 +314,7 @@ endif
 LIBS-y += drivers/rtc/librtc.o
 LIBS-y += drivers/serial/libserial.o
 LIBS-y += drivers/sound/libsound.o
-LIBS-$(CONFIG_GENERIC_LPC_TPM) += drivers/tpm/libtpm.o
+LIBS-$(CONFIG_TPM) += drivers/tpm/libtpm.o
 LIBS-y += drivers/twserial/libtws.o
 LIBS-y += drivers/usb/eth/libusb_eth.o
 LIBS-y += drivers/usb/gadget/libusb_gadget.o
README +13 -1
index 7f2506a..3b1b633 100644
@@ -1201,7 +1201,7 @@ The following options need to be configured:
 			If this option is set, the driver enables cache flush.
 
 - TPM Support:
-		CONFIG_GENERIC_LPC_TPM
+		CONFIG_TPM
 		Support for generic parallel port TPM devices. Only one device
 		per system is supported at this time.
 
@@ -1210,6 +1210,18 @@ The following options need to be configured:
 			to. Contemporary x86 systems usually map it at
 			0xfed40000.
 
+		CONFIG_ST_TPM_I2C
+		Define to compile the ST TPM I2C DRIVER.
+
+		CONFIG_TPM_I2C_BUS
+		Define the bus number of the board.
+
+		CONFIG_TPM_I2C_ADDR
+		Define the address of the TPM.
+
+		CONFIG_CMD_TPM
+		Define to use some TPM u-boot commands.
+
 - USB Support:
 		At the moment only the UHCI host controller is
 		supported (PIP405, MIP405, MPC5200); define
common/cmd_tpm.c +169 -1

178 lines not on this page · raw message

drivers/tpm/Makefile +1 -0
index be11c8b..802b54e 100644
@@ -24,6 +24,7 @@ include $(TOPDIR)/config.mk
 LIB := $(obj)libtpm.o
 
 COBJS-$(CONFIG_GENERIC_LPC_TPM) = generic_lpc_tpm.o
+COBJS-$(CONFIG_ST_TPM_I2C) = tis_i2c.o tpm.o tpm_i2c_st.o
 
 COBJS	:= $(COBJS-y)
 SRCS	:= $(COBJS:.o=.c)
drivers/tpm/compatibility.h new file +52 -0

52 lines not on this page · raw message

drivers/tpm/tis_i2c.c new file +248 -0

248 lines not on this page · raw message

drivers/tpm/tpm.c new file +477 -0

477 lines not on this page · raw message

drivers/tpm/tpm.h new file +169 -0

169 lines not on this page · raw message

drivers/tpm/tpm_i2c_st.c new file +599 -0

599 lines not on this page · raw message

include/configs/coreboot.h +2 -2
index 87daf62..fda6512 100644
@@ -76,8 +76,8 @@
 					 CONFIG_SYS_SCSI_MAX_LUN)
 #endif
 
-/* Generic TPM interfaced through LPC bus */
-#define CONFIG_GENERIC_LPC_TPM
+/* Generic TPM interfaced through LPC / I2C bus */
+#define CONFIG_TPM
 #define CONFIG_TPM_TIS_BASE_ADDRESS        0xfed40000
 
 /*-----------------------------------------------------------------------
include/configs/omap3_beagle.h +8 -0
index 48ce4c0..ef381f8 100644
@@ -111,6 +111,14 @@
 #define STATUS_LED_BOOT			STATUS_LED_BIT
 #define STATUS_LED_GREEN		STATUS_LED_BIT1
 
+/* TPM */
+#define CONFIG_CMD_TPM
+#define CONFIG_TPM
+#define CONFIG_ST_TPM_DEBUG
+#define CONFIG_ST_TPM_I2C
+#define CONFIG_TPM_I2C_BUS	1
+#define CONFIG_TPM_I2C_ADDR	0x13
+
 /* Enable Multi Bus support for I2C */
 #define CONFIG_I2C_MULTI_BUS		1
 
include/tpm.h +18 -0
index 6b21e9c..77a1514 100644
@@ -68,4 +68,21 @@ int tis_close(void);
 int tis_sendrecv(const uint8_t *sendbuf, size_t send_size, uint8_t *recvbuf,
 			size_t *recv_len);
 
+/*
+ * tis_sendrecv_hash()
+ *
+ * Send the requested data to the TPM for hash in LOC 4
+ * and then try to get its response
+ *
+ * @sendbuf - buffer of the data to hash
+ * @send_size size of the data to send
+ * @recvbuf - memory to save the response to
+ * @recv_len - pointer to the size of the response buffer
+ *
+ * Returns 0 on success (and places the number of response bytes at recv_len)
+ * or -1 on failure.
+ */
+int tis_sendrecv_hash(const uint8_t *sendbuf, size_t send_size,
+	 uint8_t *recvbuf,
+			size_t *recv_len);
+

#endif /* _INCLUDE_TPM_H_ */

Signature · 1 line
1.7.1
#1 lore raw

PATCH 1/1 TPM: STMicroelectronics u-boot driver I2C

Quoted 5 lines · Mathias leblanc

From: Mathias Leblanc <[email protected]>

* STMicroelectronics version 1.2.0, Copyright (C) 2013
* This is free software, and you are welcome to redistribute it
* under certain conditions.

Just leave these lines out of the commit message.

Quoted 8 lines

This is the driver for TPM chip from ST Microelectronics.

If you have a TPM security chip from STMicroelectronics working with an I2C, read the README file and add the correct defines regarding the tpm in the configuration file of your board.
This file is located in include/configs/your_board.h

The driver will be accessible from within uboot terminal.

Please see
http://www.denx.de/wiki/view/U-Boot/Patches#Sending_updated_patch_versions
as this is the 3rd or 4th posting of these changes.

[snip]

common/cmd_tpm.c static int do_tpm_many(cmd_tbl_t *cmdtp, int flag, · line 149

Quoted 10 lines
common/cmd_tpm.c +2 -0
index 0970a6f..2b7bf35 100644
@@ -145,10 +145,177 @@ static int do_tpm_many(cmd_tbl_t *cmdtp, int flag,
 	return rv;
 }
 
+static int do_tpm_hash(cmd_tbl_t *cmdtp, int flag, int argc,
+char * const argv[])

checkpatch warning there, please run tools/checkpatch.pl and fix all errors.

Quoted 9 lines
common/cmd_tpm.c +9 -0
+	u8 startup[] = {
+		0x00, 0xc1,
+		0x00, 0x00, 0x00, 0x0c,
+		0x00, 0x00, 0x00, 0x99,
+		0x00, 0x01
+	};
+
+	u8 selftestfull[] = {
+		0x00, 0xc1,

common/cmd_tpm.c static int do_tpm_many(cmd_tbl_t *cmdtp, int flag, · line 178

Quoted 10 lines
common/cmd_tpm.c +10 -0
+		0x00, 0x00, 0x00, 0x0a,
+		0x00, 0x00, 0x00, 0x50
+	};
+
+	u8 readpcr17[] = {
+		0x00, 0xc1,
+		0x00, 0x00, 0x00, 0x0e,
+		0x00, 0x00, 0x00, 0x15,
+		0x00, 0x00, 0x00, 0x11
+	};

Comment what these are and/or where they come from.

[snip]

common/cmd_tpm.c static int do_tpm_many(cmd_tbl_t *cmdtp, int flag, · line 170

Quoted 2 lines
common/cmd_tpm.c +2 -0
+	if (!
+	tis_sendrecv(readpcr17, sizeof(readpcr17), &response, &read_size)) {

Funny place to break the line, split it on the tis_sendrecv params. And fix anything else like this too.

[snip]

Quoted 9 lines
common/cmd_tpm.c +9 -0
+	u8 startup[] = {
+		0x00, 0xc1,
+		0x00, 0x00, 0x00, 0x0c,
+		0x00, 0x00, 0x00, 0x99,
+		0x00, 0x01
+	};
+	u8 selftestfull[] = {
+		0x00, 0xc1,
+		0x00, 0x00, 0x00, 0x0a,

common/cmd_tpm.c static int do_tpm_many(cmd_tbl_t *cmdtp, int flag, · line 284

Quoted 10 lines
common/cmd_tpm.c +10 -0
+		0x00, 0x00, 0x00, 0x50
+	};
+	u8 getpermflags[] = {
+		0x00, 0xc1,
+		0x00, 0x00, 0x00, 0x16,
+		0x00, 0x00, 0x00, 0x65,
+		0x00, 0x00, 0x00, 0x04,
+		0x00, 0x00, 0x00, 0x04,
+		0x00, 0x00, 0x01, 0x08
+	};

startup and selftestfull are the same as before? Shouldn't duplicate them then.

common/cmd_tpm.c static int do_tpm_many(cmd_tbl_t *cmdtp, int flag, · line 289

Quoted 5 lines
common/cmd_tpm.c +5 -0
+
+
+	CHECK(tis_init());
+
+

Too much whitespace around the line.

[snip]

common/cmd_tpm.c static int do_tpm_many(cmd_tbl_t *cmdtp, int flag, · line 196

Quoted 2 lines
common/cmd_tpm.c +2 -0
+/* extended error numbers from linux (see errno.h) */
+#define	ECANCELED	125	/* Operation Canceled */

'#define<space>' please.

common/cmd_tpm.c static int do_tpm_many(cmd_tbl_t *cmdtp, int flag, · line 197

Quoted 1 line
common/cmd_tpm.c +1 -0
+#define msleep(t) udelay((t)*1000)

We already have a few msleep compatibility defines. Can you please do a separate prepatch that adds msleep to <common.h> after the extern for udelay?

common/cmd_tpm.c static int do_tpm_many(cmd_tbl_t *cmdtp, int flag, · line 200

Quoted 3 lines
common/cmd_tpm.c +3 -0
+
+/* Timer frequency. Corresponds to msec timer resolution*/
+#define HZ             1000

Please use CONFIG_SYS_HZ

Quoted 1 line
common/cmd_tpm.c +1 -0
+++ b/drivers/tpm/tis_i2c.c

[snip]

Quoted 2 lines

+ * [backport from https://github.com/theopolis/u-boot-sboot/
+ * blob/master/drivers/tpm/tis_i2c.c]

You're giving the original author credit in the file already, yes? If not, please do, and drop these lines.

common/cmd_tpm.c

Quoted 5 lines
common/cmd_tpm.c +5 -0
+/* Define in board config */
+#ifndef CONFIG_TPM_I2C_BUS
+	#define CONFIG_TPM_I2C_BUS 0
+	#define CONFIG_TPM_I2C_ADDR 0
+#endif

No extra indentation.

[snip]

common/cmd_tpm.c

Quoted 3 lines
common/cmd_tpm.c +3 -0
+	if (i2c_probe(tpm.slave_addr) && i2c_probe(tpm.slave_addr)) {
+		debug(
+		"%s: fail to probe i2c addr 0x%x\n", __func__, tpm.slave_addr);

Split after __func__ instead.

drivers/tpm/tpm.c

Quoted 1 line
drivers/tpm/tpm.c +0 -0

[snip]

We have this as drivers/tpm/slb9635_i2c/tpm.c now, so I think we need to be renaming the existing one, same with tpm.h

[snip]

Quoted 1 line
drivers/tpm/tpm_i2c_st.c +0 -0

[snip]

Quoted 10 lines
drivers/tpm/tpm_i2c_st.c +10 -0
+/*
+ * write8_reg
+ * Send byte to the TIS register according to the ST33ZP24 I2C protocol.
+ * @param: tpm_register, the tpm tis register where the data should be written
+ * @param: tpm_data, the tpm_data to write inside the tpm_register
+ * @param: tpm_size, The length of the data
+ * @return: Returns zero in case of success else the negative error code.
+ */
+static int write8_reg(u8 addr, u8 tpm_register,
+		      u8 *tpm_data, u16 tpm_size)

drivers/tpm/tpm_i2c_st.c line 143

Quoted 10 lines
drivers/tpm/tpm_i2c_st.c +10 -0
+{
+	u8 data;
+	data = tpm_register;
+	memcpy(&(tpm_dev.buf[0]), &data, sizeof(data));
+	memcpy(&(tpm_dev.buf[0])+1, tpm_data, tpm_size);
+
+	return i2c_write(addr, 0, 0, &tpm_dev.buf[0],
+				tpm_size + 1);
+
+} /* write8_reg() */

Since this is kerneldoc/etc style commenting, it should be /** at the start, yes? And we don't do comments like that at the end of a function. And since it is kerneldoc style, can you do a template for them? See doc/DocBook/ Thanks!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20130422/ef503703/attachment.pgp>
#2 lore raw

PATCH 1/1 TPM: STMicroelectronics u-boot driver I2C

In reply to Mathias leblanc

Hi Mathias,

Quoted 6 lines · Mathias leblanc

From: Mathias Leblanc <[email protected]>

* STMicroelectronics version 1.2.0, Copyright (C) 2013
* This is free software, and you are welcome to redistribute it
* under certain conditions.

Quoted 8 lines

This is the driver for TPM chip from ST Microelectronics.

If you have a TPM security chip from STMicroelectronics working with an I2C, read the README file and add the correct defines regarding the tpm in the configuration file of your board.
This file is located in include/configs/your_board.h

The driver will be accessible from within uboot terminal.

Please can you take a look at current mainline TPM support?

Also there are some new patches on top of that:

http://patchwork.ozlabs.org/patch/236211/
http://patchwork.ozlabs.org/patch/236212/
http://patchwork.ozlabs.org/patch/236214/
http://patchwork.ozlabs.org/patch/236215/
http://patchwork.ozlabs.org/patch/236213/

Please take a look, and perhaps rebase on top of that?

Thanks for your efforts on this.

Regards,
Simon