Skip to contentSkip to conversation

Extend Intel service layer, FPGA manager and region

PATCH Applied [email protected] 5 participants 30 messages Feb 9, 2021 – May 2, 2021 1 message not in the archive

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

v5 cover Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
firmware: stratix10-svc: extend SVC driver to get the firmware version
fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
fpga: of-fpga-region: add authenticate-fpga-config property
dt-bindings: fpga: add authenticate-fpga-config property
fpga: stratix10-soc: extend driver for bitstream authentication

 .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
 drivers/firmware/stratix10-svc.c                   | 12 ++++-
 drivers/fpga/of-fpga-region.c                      | 24 ++++++---
 drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
 include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
 .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
 include/linux/fpga/fpga-mgr.h                      |  3 ++
 7 files changed, 125 insertions(+), 18 deletions(-)
Signature · 1 line
2.7.4
#1 lore raw

PATCH v5 1/7 firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0

From: Richard Gong <[email protected]>

Clean up COMMAND_RECONFIG_FLAG_PARTIAL flag by resetting it to 0, which aligns with the firmware settings.

Fixes 36847f9e3e56 ("firmware: stratix10-svc: correct reconfig flag and timeout values")

---
v5: new add, add the missing standalone patch

---
 include/linux/firmware/intel/stratix10-svc-client.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/firmware/intel/stratix10-svc-client.h +1 -1
index a93d859..f843c6a 100644
@@ -56,7 +56,7 @@
  * COMMAND_RECONFIG_FLAG_PARTIAL:
  * Set to FPGA configuration type (full or partial).
  */
-#define COMMAND_RECONFIG_FLAG_PARTIAL	1
+#define COMMAND_RECONFIG_FLAG_PARTIAL	0
 
 /**
  * Timeout settings for service clients:
Signature · 1 line
2.7.4
reviewer
#2 lore raw

Re: [PATCHv5 1/7] firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0

Quoted 18 lines · On 2/9/21 2:20 PM, [email protected] wrote:

From: Richard Gong <[email protected]>

Clean up COMMAND_RECONFIG_FLAG_PARTIAL flag by resetting it to 0, which aligns with the firmware settings.

Fixes 36847f9e3e56 ("firmware: stratix10-svc: correct reconfig flag and timeout values")

---
v5: new add, add the missing standalone patch

---
 include/linux/firmware/intel/stratix10-svc-client.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/firmware/intel/stratix10-svc-client.h +0 -0
index a93d859..f843c6a 100644

include/linux/firmware/intel/stratix10-svc-client.h line 59

Quoted 5 lines
include/linux/firmware/intel/stratix10-svc-client.h +1 -1
@@ -56,7 +56,7 @@
  * COMMAND_RECONFIG_FLAG_PARTIAL:
  * Set to FPGA configuration type (full or partial).
  */
-#define COMMAND_RECONFIG_FLAG_PARTIAL	1
+#define COMMAND_RECONFIG_FLAG_PARTIAL	0

Is this the stand alone fix split from v3's patch 1 ?

https://lore.kernel.org/linux-fpga/YBFW50LPP%[email protected]/

Tom

Quoted 3 lines
include/linux/firmware/intel/stratix10-svc-client.h +0 -0
 /**
  * Timeout settings for service clients:
#3 lore raw

Re: [PATCHv5 1/7] firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0

Hi Tom,

Quoted 28 lines · Tom Rix
Quoted 18 lines · On 2/9/21 2:20 PM, [email protected] wrote:

From: Richard Gong <[email protected]>

Clean up COMMAND_RECONFIG_FLAG_PARTIAL flag by resetting it to 0, which aligns with the firmware settings.

Fixes 36847f9e3e56 ("firmware: stratix10-svc: correct reconfig flag and timeout values")

---
v5: new add, add the missing standalone patch

---
  include/linux/firmware/intel/stratix10-svc-client.h | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/firmware/intel/stratix10-svc-client.h +0 -0
index a93d859..f843c6a 100644

include/linux/firmware/intel/stratix10-svc-client.h line 59

Quoted 5 lines
include/linux/firmware/intel/stratix10-svc-client.h +1 -1
@@ -56,7 +56,7 @@
   * COMMAND_RECONFIG_FLAG_PARTIAL:
   * Set to FPGA configuration type (full or partial).
   */
-#define COMMAND_RECONFIG_FLAG_PARTIAL	1
+#define COMMAND_RECONFIG_FLAG_PARTIAL	0

Is this the stand alone fix split from v3's patch 1 ?

Quoted 9 lines

https://lore.kernel.org/linux-fpga/YBFW50LPP%[email protected]/
Yes, it is a stand-alone patch.

Tom

Quoted 3 lines
include/linux/firmware/intel/stratix10-svc-client.h +0 -0
  /**
   * Timeout settings for service clients:

Regards,
Richard

reviewer Reviewed-by
#4 lore raw

Re: [PATCHv5 1/7] firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0

Quoted 33 lines · Richard Gong

Hi Tom,

Quoted 30 lines · Tom Rix
Quoted 18 lines · On 2/9/21 2:20 PM, [email protected] wrote:

From: Richard Gong <[email protected]>

Clean up COMMAND_RECONFIG_FLAG_PARTIAL flag by resetting it to 0, which aligns with the firmware settings.

Fixes 36847f9e3e56 ("firmware: stratix10-svc: correct reconfig flag and timeout values")

---
v5: new add, add the missing standalone patch
---
  include/linux/firmware/intel/stratix10-svc-client.h | 2 +-

  1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/firmware/intel/stratix10-svc-client.h +0 -0
index a93d859..f843c6a 100644

include/linux/firmware/intel/stratix10-svc-client.h line 59

Quoted 5 lines
include/linux/firmware/intel/stratix10-svc-client.h +1 -1
@@ -56,7 +56,7 @@
    * COMMAND_RECONFIG_FLAG_PARTIAL:
    * Set to FPGA configuration type (full or partial).
    */
-#define COMMAND_RECONFIG_FLAG_PARTIAL    1
+#define COMMAND_RECONFIG_FLAG_PARTIAL    0

Is this the stand alone fix split from v3's patch 1 ?

https://lore.kernel.org/linux-fpga/YBFW50LPP%[email protected]/
Yes, it is a stand-alone patch.

Thanks.

Quoted 9 lines
Quoted 6 lines

Tom

Quoted 2 lines
include/linux/firmware/intel/stratix10-svc-client.h +0 -0
     /**
    * Timeout settings for service clients:

Regards,
Richard

maintainer Applied
#5 lore raw

Re: [PATCHv5 1/7] firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0

Quoted 33 lines · Tom Rix
Quoted 31 lines · Richard Gong

Hi Tom,

Quoted 28 lines · Tom Rix
Quoted 18 lines · On 2/9/21 2:20 PM, [email protected] wrote:

From: Richard Gong <[email protected]>

Clean up COMMAND_RECONFIG_FLAG_PARTIAL flag by resetting it to 0, which aligns with the firmware settings.

Fixes 36847f9e3e56 ("firmware: stratix10-svc: correct reconfig flag and timeout values")

---
v5: new add, add the missing standalone patch
---
  include/linux/firmware/intel/stratix10-svc-client.h | 2 +-

  1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/firmware/intel/stratix10-svc-client.h +0 -0
index a93d859..f843c6a 100644

include/linux/firmware/intel/stratix10-svc-client.h line 59

Quoted 5 lines
include/linux/firmware/intel/stratix10-svc-client.h +1 -1
@@ -56,7 +56,7 @@
    * COMMAND_RECONFIG_FLAG_PARTIAL:
    * Set to FPGA configuration type (full or partial).
    */
-#define COMMAND_RECONFIG_FLAG_PARTIAL    1
+#define COMMAND_RECONFIG_FLAG_PARTIAL    0

Is this the stand alone fix split from v3's patch 1 ?

Quoted 17 lines
Quoted 2 lines
Quoted 2 lines

https://lore.kernel.org/linux-fpga/YBFW50LPP%[email protected]/
Yes, it is a stand-alone patch.

Thanks.

Quoted 9 lines
Quoted 6 lines

Tom

Quoted 2 lines
include/linux/firmware/intel/stratix10-svc-client.h +0 -0
     /**
    * Timeout settings for service clients:

Regards,
Richard

Applied to fixes.

Thanks,
Moritz

#6 lore raw

PATCH v5 2/7 firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag

In reply to [email protected]

From: Richard Gong <[email protected]>

Add COMMAND_AUTHENTICATE_BITSTREAM command flag for new added bitstream authentication feature. Authenticating a bitstream is to make sure a signed bitstream has the valid signatures.

Except for the actual configuration of the device, the bitstream authentication works the same way as FPGA configuration does. If the authentication passes, the signed bitstream will be programmed into QSPI flash memory and will be expected to boot without issues.

---
v5: no change
v4: remove change at COMMAND_RECONFIG_FLAG_PARTIAL flag & add that to a

    separate commit 27ad5309c247b6bde8a098e17e9bd9b1576b7f71.
v3: no change
v2: new added
---
 include/linux/firmware/intel/stratix10-svc-client.h | 5 +++++
 1 file changed, 5 insertions(+)
include/linux/firmware/intel/stratix10-svc-client.h +5 -0
index f843c6a..fa9581d 100644
@@ -55,8 +55,13 @@
  *
  * COMMAND_RECONFIG_FLAG_PARTIAL:
  * Set to FPGA configuration type (full or partial).
+ *
+ * COMMAND_AUTHENTICATE_BITSTREAM:
+ * Set for bitstream authentication, which makes sure a signed bitstream
+ * has valid signatures before committing it to device.
  */
 #define COMMAND_RECONFIG_FLAG_PARTIAL	0
+#define COMMAND_AUTHENTICATE_BITSTREAM	1
 
 /**
  * Timeout settings for service clients:
Signature · 1 line
2.7.4
#7 lore raw

PATCH v5 3/7 firmware: stratix10-svc: extend SVC driver to get the firmware version

In reply to [email protected]

From: Richard Gong <[email protected]>

Extend Intel service layer driver to get the firmware version running at FPGA device. Therefore FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version.

---
v5: no change
v4: no change
v3: new added, changes for getting firmware version

---
 drivers/firmware/stratix10-svc.c                    | 12 ++++++++++--
 include/linux/firmware/intel/stratix10-smc.h        | 21 +++++++++++++++++++--
 include/linux/firmware/intel/stratix10-svc-client.h |  4 ++++
 3 files changed, 33 insertions(+), 4 deletions(-)
drivers/firmware/stratix10-svc.c +10 -2
index 3aa489d..1443bbd 100644
@@ -306,6 +306,7 @@ static void svc_thread_recv_status_ok(struct stratix10_svc_data *p_data,
 		break;
 	case COMMAND_RSU_RETRY:
 	case COMMAND_RSU_MAX_RETRY:
+	case COMMAND_FIRMWARE_VERSION:
 		cb_data->status = BIT(SVC_STATUS_OK);
 		cb_data->kaddr1 = &res.a1;
 		break;
@@ -422,6 +423,11 @@ static int svc_normal_to_secure_thread(void *data)
 			a1 = 0;
 			a2 = 0;
 			break;
+		case COMMAND_FIRMWARE_VERSION:
+			a0 = INTEL_SIP_SMC_FIRMWARE_VERSION;
+			a1 = 0;
+			a2 = 0;
+			break;
 		default:
 			pr_warn("it shouldn't happen\n");
 			break;
@@ -487,11 +493,13 @@ static int svc_normal_to_secure_thread(void *data)
 
 			/*
 			 * be compatible with older version firmware which
-			 * doesn't support RSU notify or retry
+			 * doesn't support RSU notify, retry or bitstream
+			 * authentication.
 			 */
 			if ((pdata->command == COMMAND_RSU_RETRY) ||
 			    (pdata->command == COMMAND_RSU_MAX_RETRY) ||
-				(pdata->command == COMMAND_RSU_NOTIFY)) {
+			    (pdata->command == COMMAND_RSU_NOTIFY) ||
+			    (pdata->command == COMMAND_FIRMWARE_VERSION)) {
 				cbdata->status =
 					BIT(SVC_STATUS_NO_SUPPORT);
 				cbdata->kaddr1 = NULL;
include/linux/firmware/intel/stratix10-smc.h +19 -2
index c3e5ab0..505fcca 100644
@@ -321,8 +321,6 @@ INTEL_SIP_SMC_FAST_CALL_VAL(INTEL_SIP_SMC_FUNCID_FPGA_CONFIG_COMPLETED_WRITE)
 #define INTEL_SIP_SMC_ECC_DBE \
 	INTEL_SIP_SMC_FAST_CALL_VAL(INTEL_SIP_SMC_FUNCID_ECC_DBE)
 
-#endif
-
 /**
  * Request INTEL_SIP_SMC_RSU_NOTIFY
  *
@@ -404,3 +402,22 @@ INTEL_SIP_SMC_FAST_CALL_VAL(INTEL_SIP_SMC_FUNCID_FPGA_CONFIG_COMPLETED_WRITE)
 #define INTEL_SIP_SMC_FUNCID_RSU_MAX_RETRY 18
 #define INTEL_SIP_SMC_RSU_MAX_RETRY \
 	INTEL_SIP_SMC_FAST_CALL_VAL(INTEL_SIP_SMC_FUNCID_RSU_MAX_RETRY)
+
+/**
+ * Request INTEL_SIP_SMC_FIRMWARE_VERSION
+ *
+ * Sync call used to query the version of running firmware
+ *
+ * Call register usage:
+ * a0 INTEL_SIP_SMC_FIRMWARE_VERSION
+ * a1-a7 not used
+ *
+ * Return status:
+ * a0 INTEL_SIP_SMC_STATUS_OK or INTEL_SIP_SMC_STATUS_ERROR
+ * a1 running firmware version
+ */
+#define INTEL_SIP_SMC_FUNCID_FIRMWARE_VERSION 31
+#define INTEL_SIP_SMC_FIRMWARE_VERSION \
+	INTEL_SIP_SMC_FAST_CALL_VAL(INTEL_SIP_SMC_FUNCID_FIRMWARE_VERSION)
+
+#endif
include/linux/firmware/intel/stratix10-svc-client.h +4 -0
index fa9581d..193a2cf 100644
@@ -109,6 +109,9 @@ struct stratix10_svc_chan;
  *
  * @COMMAND_RSU_DCMF_VERSION: query firmware for the DCMF version, return status
  * is SVC_STATUS_OK or SVC_STATUS_ERROR
+ *
+ * @COMMAND_FIRMWARE_VERSION: query running firmware version, return status
+ * is SVC_STATUS_OK or SVC_STATUS_ERROR
  */
 enum stratix10_svc_command_code {
 	COMMAND_NOOP = 0,
@@ -122,6 +125,7 @@ enum stratix10_svc_command_code {
 	COMMAND_RSU_RETRY,
 	COMMAND_RSU_MAX_RETRY,
 	COMMAND_RSU_DCMF_VERSION,
+	COMMAND_FIRMWARE_VERSION,
 };
 
 /**
Signature · 1 line
2.7.4
maintainer Acked-by on v5 3/7
#8 lore raw

Re: [PATCHv5 3/7] firmware: stratix10-svc: extend SVC driver to get the firmware version

Quoted 8 lines · On Tue, Feb 09, 2021 at 04:20:29PM -0600, [email protected] wrote:

From: Richard Gong <[email protected]>

Extend Intel service layer driver to get the firmware version running at FPGA device. Therefore FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version.

Quoted 111 lines

---
v5: no change
v4: no change
v3: new added, changes for getting firmware version

---
 drivers/firmware/stratix10-svc.c                    | 12 ++++++++++--
 include/linux/firmware/intel/stratix10-smc.h        | 21 +++++++++++++++++++--
 include/linux/firmware/intel/stratix10-svc-client.h |  4 ++++
 3 files changed, 33 insertions(+), 4 deletions(-)
drivers/firmware/stratix10-svc.c +10 -2
index 3aa489d..1443bbd 100644
@@ -306,6 +306,7 @@ static void svc_thread_recv_status_ok(struct stratix10_svc_data *p_data,
 		break;
 	case COMMAND_RSU_RETRY:
 	case COMMAND_RSU_MAX_RETRY:
+	case COMMAND_FIRMWARE_VERSION:
 		cb_data->status = BIT(SVC_STATUS_OK);
 		cb_data->kaddr1 = &res.a1;
 		break;
@@ -422,6 +423,11 @@ static int svc_normal_to_secure_thread(void *data)
 			a1 = 0;
 			a2 = 0;
 			break;
+		case COMMAND_FIRMWARE_VERSION:
+			a0 = INTEL_SIP_SMC_FIRMWARE_VERSION;
+			a1 = 0;
+			a2 = 0;
+			break;
 		default:
 			pr_warn("it shouldn't happen\n");
 			break;
@@ -487,11 +493,13 @@ static int svc_normal_to_secure_thread(void *data)
 
 			/*
 			 * be compatible with older version firmware which
-			 * doesn't support RSU notify or retry
+			 * doesn't support RSU notify, retry or bitstream
+			 * authentication.
 			 */
 			if ((pdata->command == COMMAND_RSU_RETRY) ||
 			    (pdata->command == COMMAND_RSU_MAX_RETRY) ||
-				(pdata->command == COMMAND_RSU_NOTIFY)) {
+			    (pdata->command == COMMAND_RSU_NOTIFY) ||
+			    (pdata->command == COMMAND_FIRMWARE_VERSION)) {
 				cbdata->status =
 					BIT(SVC_STATUS_NO_SUPPORT);
 				cbdata->kaddr1 = NULL;
include/linux/firmware/intel/stratix10-smc.h +19 -2
index c3e5ab0..505fcca 100644
@@ -321,8 +321,6 @@ INTEL_SIP_SMC_FAST_CALL_VAL(INTEL_SIP_SMC_FUNCID_FPGA_CONFIG_COMPLETED_WRITE)
 #define INTEL_SIP_SMC_ECC_DBE \
 	INTEL_SIP_SMC_FAST_CALL_VAL(INTEL_SIP_SMC_FUNCID_ECC_DBE)
 
-#endif
-
 /**
  * Request INTEL_SIP_SMC_RSU_NOTIFY
  *
@@ -404,3 +402,22 @@ INTEL_SIP_SMC_FAST_CALL_VAL(INTEL_SIP_SMC_FUNCID_FPGA_CONFIG_COMPLETED_WRITE)
 #define INTEL_SIP_SMC_FUNCID_RSU_MAX_RETRY 18
 #define INTEL_SIP_SMC_RSU_MAX_RETRY \
 	INTEL_SIP_SMC_FAST_CALL_VAL(INTEL_SIP_SMC_FUNCID_RSU_MAX_RETRY)
+
+/**
+ * Request INTEL_SIP_SMC_FIRMWARE_VERSION
+ *
+ * Sync call used to query the version of running firmware
+ *
+ * Call register usage:
+ * a0 INTEL_SIP_SMC_FIRMWARE_VERSION
+ * a1-a7 not used
+ *
+ * Return status:
+ * a0 INTEL_SIP_SMC_STATUS_OK or INTEL_SIP_SMC_STATUS_ERROR
+ * a1 running firmware version
+ */
+#define INTEL_SIP_SMC_FUNCID_FIRMWARE_VERSION 31
+#define INTEL_SIP_SMC_FIRMWARE_VERSION \
+	INTEL_SIP_SMC_FAST_CALL_VAL(INTEL_SIP_SMC_FUNCID_FIRMWARE_VERSION)
+
+#endif
include/linux/firmware/intel/stratix10-svc-client.h +4 -0
index fa9581d..193a2cf 100644
@@ -109,6 +109,9 @@ struct stratix10_svc_chan;
  *
  * @COMMAND_RSU_DCMF_VERSION: query firmware for the DCMF version, return status
  * is SVC_STATUS_OK or SVC_STATUS_ERROR
+ *
+ * @COMMAND_FIRMWARE_VERSION: query running firmware version, return status
+ * is SVC_STATUS_OK or SVC_STATUS_ERROR
  */
 enum stratix10_svc_command_code {
 	COMMAND_NOOP = 0,
@@ -122,6 +125,7 @@ enum stratix10_svc_command_code {
 	COMMAND_RSU_RETRY,
 	COMMAND_RSU_MAX_RETRY,
 	COMMAND_RSU_DCMF_VERSION,
+	COMMAND_FIRMWARE_VERSION,
 };
 
 /**
Signature · 1 line
2.7.4

Thanks,
Moritz

#9 lore raw

PATCH v5 4/7 fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag

In reply to [email protected]

From: Richard Gong <[email protected]>

Add FPGA_MGR_BITSTREAM_AUTHENTICATE flag for FPGA bitstream authentication, which makes sure a signed bitstream has valid signatures.

Except for the actual configuration of the device, the authentication works the same way as FPGA configuration does. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

---
v5: no change
v4: s/FPGA_MGR_BITSTREM_AUTHENTICATION/FPGA_MGR_BITSTREAM_AUTHENTICATE
v3: no change
v2: align all FPGA_MGR_* flags
update the commit messages

---
 include/linux/fpga/fpga-mgr.h | 3 +++
 1 file changed, 3 insertions(+)
include/linux/fpga/fpga-mgr.h +3 -0
index 2bc3030..a81b3a7 100644
@@ -67,12 +67,15 @@ enum fpga_mgr_states {
  * %FPGA_MGR_BITSTREAM_LSB_FIRST: SPI bitstream bit order is LSB first
  *
  * %FPGA_MGR_COMPRESSED_BITSTREAM: FPGA bitstream is compressed
+ *
+ * %FPGA_MGR_BITSTREAM_AUTHENTICATE: do FPGA bitstream authentication only
  */
 #define FPGA_MGR_PARTIAL_RECONFIG	BIT(0)
 #define FPGA_MGR_EXTERNAL_CONFIG	BIT(1)
 #define FPGA_MGR_ENCRYPTED_BITSTREAM	BIT(2)
 #define FPGA_MGR_BITSTREAM_LSB_FIRST	BIT(3)
 #define FPGA_MGR_COMPRESSED_BITSTREAM	BIT(4)
+#define FPGA_MGR_BITSTREAM_AUTHENTICATE	BIT(5)
 
 /**
  * struct fpga_image_info - information specific to a FPGA image
Signature · 1 line
2.7.4
#10 lore raw

PATCH v5 5/7 fpga: of-fpga-region: add authenticate-fpga-config property

In reply to [email protected]

From: Richard Gong <[email protected]>

Add authenticate-fpga-config property to support FPGA bitstream authentication, which makes sure a signed bitstream has valid signatures.

---
v5: no change
v4: add additional checks to make sure *only* authenticate
v3: no change
v2: changed in alphabetical order

---
 drivers/fpga/of-fpga-region.c | 24 +++++++++++++++++-------
 1 file changed, 17 insertions(+), 7 deletions(-)
drivers/fpga/of-fpga-region.c +17 -7
index e405309..5074479 100644
@@ -218,15 +218,25 @@ static struct fpga_image_info *of_fpga_region_parse_ov(
 
 	info->overlay = overlay;
 
-	/* Read FPGA region properties from the overlay */
-	if (of_property_read_bool(overlay, "partial-fpga-config"))
-		info->flags |= FPGA_MGR_PARTIAL_RECONFIG;
+	/*
+	 * Read FPGA region properties from the overlay.
+	 *
+	 * First check the integrity of the bitstream. If the
+	 * authentication is passed, the user can perform other
+	 * operations.
+	 */
+	if (of_property_read_bool(overlay, "authenticate-fpga-config")) {
+		info->flags |= FPGA_MGR_BITSTREAM_AUTHENTICATE;
+	} else {
+		if (of_property_read_bool(overlay, "partial-fpga-config"))
+			info->flags |= FPGA_MGR_PARTIAL_RECONFIG;
 
-	if (of_property_read_bool(overlay, "external-fpga-config"))
-		info->flags |= FPGA_MGR_EXTERNAL_CONFIG;
+		if (of_property_read_bool(overlay, "external-fpga-config"))
+			info->flags |= FPGA_MGR_EXTERNAL_CONFIG;
 
-	if (of_property_read_bool(overlay, "encrypted-fpga-config"))
-		info->flags |= FPGA_MGR_ENCRYPTED_BITSTREAM;
+		if (of_property_read_bool(overlay, "encrypted-fpga-config"))
+			info->flags |= FPGA_MGR_ENCRYPTED_BITSTREAM;
+	}
 
 	if (!of_property_read_string(overlay, "firmware-name",
 				     &firmware_name)) {
Signature · 1 line
2.7.4
#11 lore raw

PATCH v5 6/7 dt-bindings: fpga: add authenticate-fpga-config property

In reply to [email protected]

From: Richard Gong <[email protected]>

Add authenticate-fpga-config property for FPGA bitstream authentication, which makes sure a signed bitstream has valid signatures.

---
v5: rewrite the description to highlight two things with
authenticate-fpga-config flag
v4: explain authenticate-fpga-config flag further
v3: no change
v2: put authenticate-fpga-config above partial-fpga-config
update commit messages

---
 Documentation/devicetree/bindings/fpga/fpga-region.txt | 10 ++++++++++
 1 file changed, 10 insertions(+)
Documentation/devicetree/bindings/fpga/fpga-region.txt +10 -0
index e811cf8..dca0e37 100644
@@ -182,6 +182,16 @@ Optional properties:
 	This property is optional if the FPGA Manager handles the bridges.
         If the fpga-region is  the child of a fpga-bridge, the list should not
         contain the parent bridge.
+- authenticate-fpga-config : boolean, set if do bitstream authentication only.
+	If 'authenticate-fpga-config' is added then adding a new node or another
+	operation is not allowed.
+	Flag authenticate-fpga-config is used to check the integrity of the
+	bitstream.
+	Except for the actual configuration of the device, the authentication
+	works in the same way as FPGA configuration. If the authentication passes,
+	other operations such as full or partial reconfiguration can be performed.
+	When the bitstream into QSPI flash memory at device is programmed, it is
+	expected that there will be no issue when starting the device.
 - partial-fpga-config : boolean, set if partial reconfiguration is to be done,
 	otherwise full reconfiguration is done.
 - external-fpga-config : boolean, set if the FPGA has already been configured
Signature · 1 line
2.7.4
#12 lore raw

PATCH v5 7/7 fpga: stratix10-soc: extend driver for bitstream authentication

In reply to [email protected]

From: Richard Gong <[email protected]>

Extend FPGA manager driver to support FPGA bitstream authentication on Intel SocFPGA platforms.

---
v5: no change
v4: s/FPGA_MGR_BITSTREM_AUTHENTICATION/FPGA_MGR_BITSTREAM_AUTHENTICATE
v3: add handle to retriev the firmware version to keep driver
back compatible
v2: use flag defined in stratix10-svc driver

---
 drivers/fpga/stratix10-soc.c | 62 +++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 56 insertions(+), 6 deletions(-)
drivers/fpga/stratix10-soc.c +56 -6
index 657a70c..9ab7afd 100644
@@ -24,6 +24,10 @@
 #define S10_BUFFER_TIMEOUT (msecs_to_jiffies(SVC_RECONFIG_BUFFER_TIMEOUT_MS))
 #define S10_RECONFIG_TIMEOUT (msecs_to_jiffies(SVC_RECONFIG_REQUEST_TIMEOUT_MS))
 
+#define INVALID_FIRMWARE_VERSION	0xFFFF
+typedef void (*s10_callback)(struct stratix10_svc_client *client,
+			     struct stratix10_svc_cb_data *data);
+
 /*
  * struct s10_svc_buf
  * buf:  virtual address of buf provided by service layer
@@ -40,11 +44,13 @@ struct s10_priv {
 	struct completion status_return_completion;
 	struct s10_svc_buf svc_bufs[NUM_SVC_BUFS];
 	unsigned long status;
+	unsigned int fw_version;
 };
 
 static int s10_svc_send_msg(struct s10_priv *priv,
 			    enum stratix10_svc_command_code command,
-			    void *payload, u32 payload_length)
+			    void *payload, u32 payload_length,
+			    s10_callback callback)
 {
 	struct stratix10_svc_chan *chan = priv->chan;
 	struct device *dev = priv->client.dev;
@@ -57,6 +63,7 @@ static int s10_svc_send_msg(struct s10_priv *priv,
 	msg.command = command;
 	msg.payload = payload;
 	msg.payload_length = payload_length;
+	priv->client.receive_cb = callback;
 
 	ret = stratix10_svc_send(chan, &msg);
 	dev_dbg(dev, "stratix10_svc_send returned status %d\n", ret);
@@ -134,6 +141,29 @@ static void s10_unlock_bufs(struct s10_priv *priv, void *kaddr)
 }
 
 /*
+ * s10_fw_version_callback - callback for the version of running firmware
+ * @client: service layer client struct
+ * @data: message from service layer
+ */
+static void s10_fw_version_callback(struct stratix10_svc_client *client,
+				    struct stratix10_svc_cb_data *data)
+{
+	struct s10_priv *priv = client->priv;
+	unsigned int *version = (unsigned int *)data->kaddr1;
+
+	if (data->status == BIT(SVC_STATUS_OK))
+		priv->fw_version = *version;
+	else if (data->status == BIT(SVC_STATUS_NO_SUPPORT))
+		dev_warn(client->dev,
+			 "FW doesn't support bitstream authentication\n");
+	else
+		dev_err(client->dev, "Failed to get FW version %lu\n",
+			BIT(data->status));
+
+	complete(&priv->status_return_completion);
+}
+
+/*
  * s10_receive_callback - callback for service layer to use to provide client
  * (this driver) messages received through the mailbox.
  * client: service layer client struct
@@ -186,13 +216,22 @@ static int s10_ops_write_init(struct fpga_manager *mgr,
 	if (info->flags & FPGA_MGR_PARTIAL_RECONFIG) {
 		dev_dbg(dev, "Requesting partial reconfiguration.\n");
 		ctype.flags |= BIT(COMMAND_RECONFIG_FLAG_PARTIAL);
+	} else if (info->flags & FPGA_MGR_BITSTREAM_AUTHENTICATE) {
+		if (priv->fw_version == INVALID_FIRMWARE_VERSION) {
+			dev_err(dev, "FW doesn't support\n");
+			return -EINVAL;
+		}
+
+		dev_dbg(dev, "Requesting bitstream authentication.\n");
+		ctype.flags |= BIT(COMMAND_AUTHENTICATE_BITSTREAM);
 	} else {
 		dev_dbg(dev, "Requesting full reconfiguration.\n");
 	}
 
 	reinit_completion(&priv->status_return_completion);
 	ret = s10_svc_send_msg(priv, COMMAND_RECONFIG,
-			       &ctype, sizeof(ctype));
+			       &ctype, sizeof(ctype),
+			       s10_receive_callback);
 	if (ret < 0)
 		goto init_done;
 
@@ -259,7 +298,7 @@ static int s10_send_buf(struct fpga_manager *mgr, const char *buf, size_t count)
 	svc_buf = priv->svc_bufs[i].buf;
 	memcpy(svc_buf, buf, xfer_sz);
 	ret = s10_svc_send_msg(priv, COMMAND_RECONFIG_DATA_SUBMIT,
-			       svc_buf, xfer_sz);
+			       svc_buf, xfer_sz, s10_receive_callback);
 	if (ret < 0) {
 		dev_err(dev,
 			"Error while sending data to service layer (%d)", ret);
@@ -303,7 +342,7 @@ static int s10_ops_write(struct fpga_manager *mgr, const char *buf,
 
 			ret = s10_svc_send_msg(
 				priv, COMMAND_RECONFIG_DATA_CLAIM,
-				NULL, 0);
+				NULL, 0, s10_receive_callback);
 			if (ret < 0)
 				break;
 		}
@@ -357,7 +396,8 @@ static int s10_ops_write_complete(struct fpga_manager *mgr,
 	do {
 		reinit_completion(&priv->status_return_completion);
 
-		ret = s10_svc_send_msg(priv, COMMAND_RECONFIG_STATUS, NULL, 0);
+		ret = s10_svc_send_msg(priv, COMMAND_RECONFIG_STATUS,
+				       NULL, 0, s10_receive_callback);
 		if (ret < 0)
 			break;
 
@@ -411,8 +451,9 @@ static int s10_probe(struct platform_device *pdev)
 	if (!priv)
 		return -ENOMEM;
 
+	priv->fw_version = INVALID_FIRMWARE_VERSION;
 	priv->client.dev = dev;
-	priv->client.receive_cb = s10_receive_callback;
+	priv->client.receive_cb = NULL;
 	priv->client.priv = priv;
 
 	priv->chan = stratix10_svc_request_channel_byname(&priv->client,
@@ -440,6 +481,15 @@ static int s10_probe(struct platform_device *pdev)
 		goto probe_err;
 	}
 
+	/* get the running firmware version */
+	ret = s10_svc_send_msg(priv, COMMAND_FIRMWARE_VERSION,
+			       NULL, 0, s10_fw_version_callback);
+	if (ret) {
+		dev_err(dev, "couldn't get firmware version\n");
+		fpga_mgr_free(mgr);
+		goto probe_err;
+	}
+
 	platform_set_drvdata(pdev, mgr);
 	return ret;
 
Signature · 1 line
2.7.4
#13 lore raw
In reply to [email protected]

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
firmware: stratix10-svc: extend SVC driver to get the firmware version
fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
fpga: of-fpga-region: add authenticate-fpga-config property
dt-bindings: fpga: add authenticate-fpga-config property
fpga: stratix10-soc: extend driver for bitstream authentication

 .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
 drivers/firmware/stratix10-svc.c                   | 12 ++++-
 drivers/fpga/of-fpga-region.c                      | 24 ++++++---
 drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
 include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
 .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
 include/linux/fpga/fpga-mgr.h                      |  3 ++
 7 files changed, 125 insertions(+), 18 deletions(-)

--
2.7.4

reviewer
#14 lore raw

The first patch is a fix that is targeted for stable.

Tom

Quoted 49 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
firmware: stratix10-svc: extend SVC driver to get the firmware version
fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
fpga: of-fpga-region: add authenticate-fpga-config property
dt-bindings: fpga: add authenticate-fpga-config property
fpga: stratix10-soc: extend driver for bitstream authentication

 .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
 drivers/firmware/stratix10-svc.c                   | 12 ++++-
 drivers/fpga/of-fpga-region.c                      | 24 ++++++---
 drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
 include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
 .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
 include/linux/fpga/fpga-mgr.h                      |  3 ++
 7 files changed, 125 insertions(+), 18 deletions(-)

--
2.7.4

#15 lore raw

The first patch of the version 5 patch set is a fix for the mainline, I submitted a separate patch for a fix at the stable.

Regards,
Richard

Quoted 55 lines · Tom Rix

The first patch is a fix that is targeted for stable.

Tom

Quoted 49 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
firmware: stratix10-svc: extend SVC driver to get the firmware version
fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
fpga: of-fpga-region: add authenticate-fpga-config property
dt-bindings: fpga: add authenticate-fpga-config property
fpga: stratix10-soc: extend driver for bitstream authentication

  .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
  drivers/firmware/stratix10-svc.c                   | 12 ++++-
  drivers/fpga/of-fpga-region.c                      | 24 ++++++---
  drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
  include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
  .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
  include/linux/fpga/fpga-mgr.h                      |  3 ++
  7 files changed, 125 insertions(+), 18 deletions(-)

--
2.7.4

#16 lore raw
In reply to Gong, Richard

Hi Moritz,

Thanks for approving the 1st patch of my version 5 patchest, which submitted on 02/09/21.

Can you help review the remaining 6 patches from the same version 5 patchset? I need your ACKs to move forward, or please let me know if additional work is need.

Many thanks for your time again!

Regards,
Richard

Quoted 49 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
firmware: stratix10-svc: extend SVC driver to get the firmware version
fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
fpga: of-fpga-region: add authenticate-fpga-config property
dt-bindings: fpga: add authenticate-fpga-config property
fpga: stratix10-soc: extend driver for bitstream authentication

  .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
  drivers/firmware/stratix10-svc.c                   | 12 ++++-
  drivers/fpga/of-fpga-region.c                      | 24 ++++++---
  drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
  include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
  .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
  include/linux/fpga/fpga-mgr.h                      |  3 ++
  7 files changed, 125 insertions(+), 18 deletions(-)

--
2.7.4

reviewer
#17 lore raw
Quoted 4 lines · Richard Gong

Hi Moritz,

Thanks for approving the 1st patch of my version 5 patchest, which submitted on 02/09/21.

This change

e23bd83368af ("firmware: stratix10-svc: fix kernel-doc markups")

Makes a lot of formatting changes in the same files as this patchset, including the first patch.

It would be good to try applying this patchset to char-misc-next and resubmit if there are conflicts.

Quoted 2 lines

Can you help review the remaining 6 patches from the same version 5 patchset? I need your ACKs to move forward, or please let me know if additional work is need.

These changes look good to me.

I was looking at the patchset again seeing if the firmware/ parts could be split out.

Even though stratix10 is a fpga, from the MAINTAINERS file it is not clear to me if linux-fpga owns them and they come in on Moritz's branch.  I think this change is needed to the MAINTAINERS file to make that clearer.

MAINTAINERS +0 -0
index aa84121c5611..1f68e9ff76de 100644
@@ -9193,7 +9193,8 @@ F:    tools/power/x86/intel-speed-select/


 INTEL STRATIX10 FIRMWARE DRIVERS
 M:    Richard Gong <[email protected]>
-L:    [email protected]
+R:    Tom Rix <[email protected]>
+L:    [email protected]
 S:    Maintained
 F:    Documentation/ABI/testing/sysfs-devices-platform-stratix10-rsu F:    Documentation/devicetree/bindings/firmware/intel,stratix10-svc.txt

I also added myself as a reviewer because I want to help out.

Tom

Quoted 58 lines

Many thanks for your time again!

Regards,
Richard

Quoted 49 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
   firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
   firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
   firmware: stratix10-svc: extend SVC driver to get the firmware version
   fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
   fpga: of-fpga-region: add authenticate-fpga-config property
   dt-bindings: fpga: add authenticate-fpga-config property
   fpga: stratix10-soc: extend driver for bitstream authentication

  .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++ drivers/firmware/stratix10-svc.c                   | 12 ++++- drivers/fpga/of-fpga-region.c                      | 24 ++++++--- drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++--- include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++- .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++- include/linux/fpga/fpga-mgr.h                      |  3 ++

  7 files changed, 125 insertions(+), 18 deletions(-)
Signature · 1 line
2.7.4

A message that is not in the archive (sent off-list or not yet received): MWHPR11MB0015516D86D02A0FE5423D6387669@MWHPR11MB0015.namprd11.prod.outlook.com

#19 lore raw

Hi Tom,

Quoted 11 lines · Richard Gong
Quoted 4 lines

Hi Moritz,

Thanks for approving the 1st patch of my version 5 patchest, which submitted on 02/09/21.

This change

e23bd83368af ("firmware: stratix10-svc: fix kernel-doc markups")

This patch e23bd83368af is not from my version 5 patch set.

Quoted 11 lines

Makes a lot of formatting changes in the same files as this patchset, including the first patch.

It would be good to try applying this patchset to char-misc-next and resubmit if there are conflicts.

Quoted 2 lines

Can you help review the remaining 6 patches from the same version 5 patchset? I need your ACKs to move forward, or please let me know if additional work is need.

These changes look good to me.

I was looking at the patchset again seeing if the firmware/ parts could be split out.

No, we can't split out the firmware parts.

Quoted 18 lines

Even though stratix10 is a fpga, from the MAINTAINERS file it is not clear to me if linux-fpga owns them and they come in on Moritz's branch.  I think this change is needed to the MAINTAINERS file to make that clearer.

MAINTAINERS +2 -1
index aa84121c5611..1f68e9ff76de 100644
@@ -9193,7 +9193,8 @@ F:    tools/power/x86/intel-speed-select/
 
  INTEL STRATIX10 FIRMWARE DRIVERS
  M:    Richard Gong <[email protected]>
-L:    [email protected]
+R:    Tom Rix <[email protected]>
+L:    [email protected]
  S:    Maintained
  F:    Documentation/ABI/testing/sysfs-devices-platform-stratix10-rsu
  F:    Documentation/devicetree/bindings/firmware/intel,stratix10-svc.txt
Quoted 4 lines

I also added myself as a reviewer because I want to help out.

Tom

Regards,
Richard

Quoted 60 lines
Quoted 59 lines

Many thanks for your time again!

Regards,
Richard

Quoted 49 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
   firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
   firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
   firmware: stratix10-svc: extend SVC driver to get the firmware version
   fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
   fpga: of-fpga-region: add authenticate-fpga-config property
   dt-bindings: fpga: add authenticate-fpga-config property
   fpga: stratix10-soc: extend driver for bitstream authentication

   .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
   drivers/firmware/stratix10-svc.c                   | 12 ++++-
   drivers/fpga/of-fpga-region.c                      | 24 ++++++---
   drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
   include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
   .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
   include/linux/fpga/fpga-mgr.h                      |  3 ++
   7 files changed, 125 insertions(+), 18 deletions(-)
Signature · 1 line
2.7.4
reviewer
#20 lore raw
Quoted 9 lines · Richard Gong

Hi Tom,

Quoted 6 lines · Richard Gong
Quoted 3 lines

Hi Moritz,

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

Thanks for approving the 1st patch of my version 5 patchest, which submitted on 02/09/21.

This change

e23bd83368af ("firmware: stratix10-svc: fix kernel-doc markups")

This patch e23bd83368af is not from my version 5 patch set.

Correct.

But since it is already in char-misc-next, your version 5 patchset will conflict with it.

I could not apply this patchset to my unoffical fpga-testing.

I am suggesting you do a test application of your patchset against char-misc-next.

And if you find there are issues, rebase your patchset.

Quoted 6 lines
Quoted 6 lines

Makes a lot of formatting changes in the same files as this patchset, including the first patch.

It would be good to try applying this patchset to char-misc-next and resubmit if there are conflicts.

Quoted 1 line
Quoted 7 lines
Quoted 5 lines
Quoted 1 line

Can you help review the remaining 6 patches from the same version 5 patchset? I need your ACKs to move forward, or please let me know if additional work is need.

These changes look good to me.

I was looking at the patchset again seeing if the firmware/ parts could be split out.

No, we can't split out the firmware parts.

ok

Tom

Quoted 86 lines
Quoted 17 lines

Even though stratix10 is a fpga, from the MAINTAINERS file it is not clear to me if linux-fpga owns them and they come in on Moritz's branch.  I think this change is needed to the MAINTAINERS file to make that clearer.

MAINTAINERS +0 -0
index aa84121c5611..1f68e9ff76de 100644
@@ -9193,7 +9193,8 @@ F:    tools/power/x86/intel-speed-select/

    INTEL STRATIX10 FIRMWARE DRIVERS
  M:    Richard Gong <[email protected]>

MAINTAINERS +2 -1
-L:    [email protected]
+R:    Tom Rix <[email protected]>
+L:    [email protected]

  S:    Maintained
  F:    Documentation/ABI/testing/sysfs-devices-platform-stratix10-rsu F:    Documentation/devicetree/bindings/firmware/intel,stratix10-svc.txt

Quoted 4 lines

I also added myself as a reviewer because I want to help out.

Tom

Regards,
Richard

Quoted 60 lines
Quoted 59 lines

Many thanks for your time again!

Regards,
Richard

Quoted 49 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
    firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
    firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
    firmware: stratix10-svc: extend SVC driver to get the firmware version
    fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
    fpga: of-fpga-region: add authenticate-fpga-config property
    dt-bindings: fpga: add authenticate-fpga-config property
    fpga: stratix10-soc: extend driver for bitstream authentication

   .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++ drivers/firmware/stratix10-svc.c                   | 12 ++++- drivers/fpga/of-fpga-region.c                      | 24 ++++++--- drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++--- include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++- .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++- include/linux/fpga/fpga-mgr.h                      |  3 ++

   7 files changed, 125 insertions(+), 18 deletions(-)

--
2.7.4

#21 lore raw

Hi Tom,

Quoted 20 lines · Tom Rix
Quoted 8 lines · Richard Gong

Hi Tom >>

Quoted 6 lines · Richard Gong
Quoted 3 lines

Hi Moritz,

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

Thanks for approving the 1st patch of my version 5 patchest, which submitted on 02/09/21.

This change

e23bd83368af ("firmware: stratix10-svc: fix kernel-doc markups")

This patch e23bd83368af is not from my version 5 patch set.

Correct.

Quoted 8 lines

But since it is already in char-misc-next, your version 5 patchset will conflict with it.

I could not apply this patchset to my unoffical fpga-testing.

I am suggesting you do a test application of your patchset against char-misc-next.

And if you find there are issues, rebase your patchset.

I tried to apply my patchset to the top of char-misc-next, but I didn't see any conflicts.

c7582d1 fpga: stratix10-soc: extend driver for bitstream authentication 2c9ecd3 dt-bindings: fpga: add authenticate-fpga-config property 6244115 fpga: of-fpga-region: add authenticate-fpga-config property da274c9 fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag 9f93cad firmware: stratix10-svc: extend SVC driver to get the firmware version eda6b51 firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag 91aff09 firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0 83be46e Merge v5.12-rc3 into char-misc-next

Regards,
Richard

Quoted 105 lines
Quoted 6 lines
Quoted 6 lines

Makes a lot of formatting changes in the same files as this patchset, including the first patch.

It would be good to try applying this patchset to char-misc-next and resubmit if there are conflicts.

Quoted 1 line
Quoted 7 lines
Quoted 5 lines
Quoted 1 line

Can you help review the remaining 6 patches from the same version 5 patchset? I need your ACKs to move forward, or please let me know if additional work is need.

These changes look good to me.

I was looking at the patchset again seeing if the firmware/ parts could be split out.

No, we can't split out the firmware parts.

ok

Tom

Quoted 86 lines
Quoted 19 lines

Even though stratix10 is a fpga, from the MAINTAINERS file it is not clear to me if linux-fpga owns them and they come in on Moritz's branch.  I think this change is needed to the MAINTAINERS file to make that clearer.

MAINTAINERS +2 -1
index aa84121c5611..1f68e9ff76de 100644
@@ -9193,7 +9193,8 @@ F:    tools/power/x86/intel-speed-select/
     INTEL STRATIX10 FIRMWARE DRIVERS
   M:    Richard Gong <[email protected]>
-L:    [email protected]
+R:    Tom Rix <[email protected]>
+L:    [email protected]
   S:    Maintained
   F:    Documentation/ABI/testing/sysfs-devices-platform-stratix10-rsu
   F:    Documentation/devicetree/bindings/firmware/intel,stratix10-svc.txt
 

I also added myself as a reviewer because I want to help out.

Quoted 2 lines

Tom

Regards,
Richard

Quoted 60 lines
Quoted 59 lines

Many thanks for your time again!

Regards,
Richard

Quoted 49 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
    firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
    firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
    firmware: stratix10-svc: extend SVC driver to get the firmware version
    fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
    fpga: of-fpga-region: add authenticate-fpga-config property
    dt-bindings: fpga: add authenticate-fpga-config property
    fpga: stratix10-soc: extend driver for bitstream authentication

    .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
    drivers/firmware/stratix10-svc.c                   | 12 ++++-
    drivers/fpga/of-fpga-region.c                      | 24 ++++++---
    drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
    include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
    .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
    include/linux/fpga/fpga-mgr.h                      |  3 ++
    7 files changed, 125 insertions(+), 18 deletions(-)
Signature · 1 line
2.7.4
maintainer
#22 lore raw
In reply to Gong, Richard

Hi Richard, Russ,

Quoted 46 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
firmware: stratix10-svc: extend SVC driver to get the firmware version
fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
fpga: of-fpga-region: add authenticate-fpga-config property
dt-bindings: fpga: add authenticate-fpga-config property
fpga: stratix10-soc: extend driver for bitstream authentication

 .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
 drivers/firmware/stratix10-svc.c                   | 12 ++++-
 drivers/fpga/of-fpga-region.c                      | 24 ++++++---
 drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
 include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
 .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
 include/linux/fpga/fpga-mgr.h                      |  3 ++
 7 files changed, 125 insertions(+), 18 deletions(-)
Quoted 3 lines

--
2.7.4

Apologies for the epic delay in getting back to this, I took another look at this patchset and Russ' patchset.

TL;DR I'm not really a fan of using device-tree overlays for this (and again, apologies, I should've voiced this earlier ...).

Anyways, let's find a common API for this and Russ' work, they're trying to achieve the same / similar thing, they should use the same API.

I'd like to re-invetigate the possiblity to extend FPGA Manager with 'secure update' ops that work for both these use-cases (and I susspect hte XRT patchset will follow with a similar requirement, right after).

- Moritz

reviewer
#23 lore raw
Quoted 58 lines · Moritz Fischer

Hi Richard, Russ,

Quoted 46 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
firmware: stratix10-svc: extend SVC driver to get the firmware version
fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
fpga: of-fpga-region: add authenticate-fpga-config property
dt-bindings: fpga: add authenticate-fpga-config property
fpga: stratix10-soc: extend driver for bitstream authentication

 .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
 drivers/firmware/stratix10-svc.c                   | 12 ++++-
 drivers/fpga/of-fpga-region.c                      | 24 ++++++---
 drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
 include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
 .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
 include/linux/fpga/fpga-mgr.h                      |  3 ++
 7 files changed, 125 insertions(+), 18 deletions(-)
Quoted 3 lines

--
2.7.4

Apologies for the epic delay in getting back to this, I took another look at this patchset and Russ' patchset.

TL;DR I'm not really a fan of using device-tree overlays for this (and again, apologies, I should've voiced this earlier ...).

Quoted 6 lines

Anyways, let's find a common API for this and Russ' work, they're trying to achieve the same / similar thing, they should use the same API.

I'd like to re-invetigate the possiblity to extend FPGA Manager with 'secure update' ops that work for both these use-cases (and I susspect hte XRT patchset will follow with a similar requirement, right after).

The xrt patchset makes heavy use of device trees.

What is the general guidance for device tree usage ?

Tom

Quoted 3 lines

- Moritz

maintainer
#24 lore raw

Tom,

Quoted 63 lines · Tom Rix
Quoted 61 lines · Moritz Fischer

Hi Richard, Russ,

Quoted 46 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
firmware: stratix10-svc: extend SVC driver to get the firmware version
fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
fpga: of-fpga-region: add authenticate-fpga-config property
dt-bindings: fpga: add authenticate-fpga-config property
fpga: stratix10-soc: extend driver for bitstream authentication

 .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
 drivers/firmware/stratix10-svc.c                   | 12 ++++-
 drivers/fpga/of-fpga-region.c                      | 24 ++++++---
 drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
 include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
 .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
 include/linux/fpga/fpga-mgr.h                      |  3 ++
 7 files changed, 125 insertions(+), 18 deletions(-)
Quoted 3 lines

--
2.7.4

Apologies for the epic delay in getting back to this, I took another look at this patchset and Russ' patchset.

TL;DR I'm not really a fan of using device-tree overlays for this (and again, apologies, I should've voiced this earlier ...).

Anyways, let's find a common API for this and Russ' work, they're trying to achieve the same / similar thing, they should use the same API.

Quoted 7 lines
Quoted 3 lines

I'd like to re-invetigate the possiblity to extend FPGA Manager with 'secure update' ops that work for both these use-cases (and I susspect hte XRT patchset will follow with a similar requirement, right after).

The xrt patchset makes heavy use of device trees.

What is the general guidance for device tree usage ?

I'm not generally against using device tree, it has its place. To describe hardware (and hardware *changes* with overlays) :)

What I don't like about this particular implementation w.r.t device-tree usage is that it uses DT overlays as a mechanism to program the flash -- in place of having an API to do so.

One could add device-nodes during the DT overlay application, while the FPGA doesn't actually get programmed with a new runtime image -- meaning live DT and actual hardware state diverged -- worst case it'd crash.

So when roughly at the same time (from the same company even) we have two patchsets that do similar things with radically different APIs I think we should pause, and reflect on whether we can come up with something that works for both :)

TL;DR the firmware parts to authenticate the bitstream look fine to me, the way we tie it into the FPGA region I'm not a fan of.

- Moritz

#25 lore raw

Moritz,

Quoted 68 lines · Moritz Fischer

Tom,

Quoted 65 lines · Tom Rix
Quoted 64 lines · Moritz Fischer

Hi Richard, Russ,

Quoted 46 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
firmware: stratix10-svc: extend SVC driver to get the firmware version
fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
fpga: of-fpga-region: add authenticate-fpga-config property
dt-bindings: fpga: add authenticate-fpga-config property
fpga: stratix10-soc: extend driver for bitstream authentication

 .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
 drivers/firmware/stratix10-svc.c                   | 12 ++++-
 drivers/fpga/of-fpga-region.c                      | 24 ++++++---
 drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
 include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
 .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
 include/linux/fpga/fpga-mgr.h                      |  3 ++
 7 files changed, 125 insertions(+), 18 deletions(-)
Quoted 3 lines

--
2.7.4

Apologies for the epic delay in getting back to this, I took another look at this patchset and Russ' patchset.

TL;DR I'm not really a fan of using device-tree overlays for this (and again, apologies, I should've voiced this earlier ...).

Anyways, let's find a common API for this and Russ' work, they're trying to achieve the same / similar thing, they should use the same API.

I'd like to re-invetigate the possiblity to extend FPGA Manager with 'secure update' ops that work for both these use-cases (and I susspect hte XRT patchset will follow with a similar requirement, right after).

Richard and I had an initial conversation today. I'll start looking at how secure operations can be integrated into the fpga manager.

More to come...

Thanks,
- Russ

Quoted 23 lines
Quoted 3 lines

The xrt patchset makes heavy use of device trees.

What is the general guidance for device tree usage ?

I'm not generally against using device tree, it has its place. To describe hardware (and hardware *changes* with overlays) :)

What I don't like about this particular implementation w.r.t device-tree usage is that it uses DT overlays as a mechanism to program the flash -- in place of having an API to do so.

One could add device-nodes during the DT overlay application, while the FPGA doesn't actually get programmed with a new runtime image -- meaning live DT and actual hardware state diverged -- worst case it'd crash.

So when roughly at the same time (from the same company even) we have two patchsets that do similar things with radically different APIs I think we should pause, and reflect on whether we can come up with something that works for both :)

TL;DR the firmware parts to authenticate the bitstream look fine to me, the way we tie it into the FPGA region I'm not a fan of.

- Moritz

maintainer
#26 lore raw

Hi Russ,

Quoted 68 lines · Russ Weight

Moritz,

Quoted 65 lines · Moritz Fischer

Tom,

Quoted 62 lines · Tom Rix
Quoted 61 lines · Moritz Fischer

Hi Richard, Russ,

Quoted 46 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
firmware: stratix10-svc: extend SVC driver to get the firmware version
fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
fpga: of-fpga-region: add authenticate-fpga-config property
dt-bindings: fpga: add authenticate-fpga-config property
fpga: stratix10-soc: extend driver for bitstream authentication

 .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
 drivers/firmware/stratix10-svc.c                   | 12 ++++-
 drivers/fpga/of-fpga-region.c                      | 24 ++++++---
 drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
 include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
 .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
 include/linux/fpga/fpga-mgr.h                      |  3 ++
 7 files changed, 125 insertions(+), 18 deletions(-)
Quoted 3 lines

--
2.7.4

Apologies for the epic delay in getting back to this, I took another look at this patchset and Russ' patchset.

TL;DR I'm not really a fan of using device-tree overlays for this (and again, apologies, I should've voiced this earlier ...).

Anyways, let's find a common API for this and Russ' work, they're trying to achieve the same / similar thing, they should use the same API.

Quoted 7 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines

I'd like to re-invetigate the possiblity to extend FPGA Manager with 'secure update' ops that work for both these use-cases (and I susspect hte XRT patchset will follow with a similar requirement, right after).

Richard and I had an initial conversation today. I'll start looking at how secure operations can be integrated into the fpga manager.

More to come...

Great, feel free to send RFCs ahead.

Cheers,
Moritz

#27 lore raw

Hi Moritz,

Quoted 73 lines · Moritz Fischer

Hi Russ,

Quoted 71 lines · Russ Weight

Moritz,

Quoted 68 lines · Moritz Fischer

Tom,

Quoted 65 lines · Tom Rix
Quoted 64 lines · Moritz Fischer

Hi Richard, Russ,

Quoted 46 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
firmware: stratix10-svc: extend SVC driver to get the firmware version
fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
fpga: of-fpga-region: add authenticate-fpga-config property
dt-bindings: fpga: add authenticate-fpga-config property
fpga: stratix10-soc: extend driver for bitstream authentication

 .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
 drivers/firmware/stratix10-svc.c                   | 12 ++++-
 drivers/fpga/of-fpga-region.c                      | 24 ++++++---
 drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
 include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
 .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
 include/linux/fpga/fpga-mgr.h                      |  3 ++
 7 files changed, 125 insertions(+), 18 deletions(-)
Quoted 3 lines

--
2.7.4

Apologies for the epic delay in getting back to this, I took another look at this patchset and Russ' patchset.

TL;DR I'm not really a fan of using device-tree overlays for this (and again, apologies, I should've voiced this earlier ...).

Anyways, let's find a common API for this and Russ' work, they're trying to achieve the same / similar thing, they should use the same API.

I'd like to re-invetigate the possiblity to extend FPGA Manager with 'secure update' ops that work for both these use-cases (and I susspect hte XRT patchset will follow with a similar requirement, right after).

Quoted 7 lines
Quoted 3 lines

Richard and I had an initial conversation today. I'll start looking at how secure operations can be integrated into the fpga manager.

More to come...

Great, feel free to send RFCs ahead.

Cheers,
Moritz

I have completed a comparison of the security manager and the FPGA manager to see how the secure update functions can be integrated into the FPGA manager. I'll send that out separately as an RFC document (it is about 150 lines).

FYI: In my conversations with Richard, we have learned that what we are trying to accomplish is not as similar as it seemed. Richard is effectively wanting to do a "dry-run" of an existing FPGA Manager function to verify authentication of an image. Based on the results, higher-level code may choose to write the image to flash.

The security manager is all about providing a common user API (sysfs interface) for tranferring an image to an FPGA card while managing a completion time (including authentication and FLASH) of up to 40 minutes.

- Russ

maintainer
#28 lore raw
Quoted 88 lines · Russ Weight

Hi Moritz,

Quoted 73 lines · Moritz Fischer

Hi Russ,

Quoted 71 lines · Russ Weight

Moritz,

Quoted 68 lines · Moritz Fischer

Tom,

Quoted 65 lines · Tom Rix
Quoted 64 lines · Moritz Fischer

Hi Richard, Russ,

Quoted 46 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
firmware: stratix10-svc: extend SVC driver to get the firmware version
fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
fpga: of-fpga-region: add authenticate-fpga-config property
dt-bindings: fpga: add authenticate-fpga-config property
fpga: stratix10-soc: extend driver for bitstream authentication

 .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
 drivers/firmware/stratix10-svc.c                   | 12 ++++-
 drivers/fpga/of-fpga-region.c                      | 24 ++++++---
 drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
 include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
 .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
 include/linux/fpga/fpga-mgr.h                      |  3 ++
 7 files changed, 125 insertions(+), 18 deletions(-)
Quoted 3 lines

--
2.7.4

Apologies for the epic delay in getting back to this, I took another look at this patchset and Russ' patchset.

TL;DR I'm not really a fan of using device-tree overlays for this (and again, apologies, I should've voiced this earlier ...).

Anyways, let's find a common API for this and Russ' work, they're trying to achieve the same / similar thing, they should use the same API.

I'd like to re-invetigate the possiblity to extend FPGA Manager with 'secure update' ops that work for both these use-cases (and I susspect hte XRT patchset will follow with a similar requirement, right after).

Quoted 7 lines
Quoted 3 lines

Richard and I had an initial conversation today. I'll start looking at how secure operations can be integrated into the fpga manager.

More to come...

Great, feel free to send RFCs ahead.

Cheers,
Moritz

I have completed a comparison of the security manager and the FPGA manager to see how the secure update functions can be integrated into the FPGA manager. I'll send that out separately as an RFC document (it is about 150 lines).

Quoted 5 lines

FYI: In my conversations with Richard, we have learned that what we are trying to accomplish is not as similar as it seemed. Richard is effectively wanting to do a "dry-run" of an existing FPGA Manager function to verify authentication of an image. Based on the results, higher-level code may choose to write the image to flash.

Ok, as mentioned on the other thread. Let's replace the overlay code with a sysfs entry for dry-run or verify-image or something of that sort.

I don't see what the overlay mechanism adds in this case since we're not trying to load drivers.

- Moritz

#29 lore raw
In reply to Moritz Fischer

Hi Moritz,

Quoted 81 lines · Moritz Fischer

Tom,

Quoted 63 lines · Tom Rix
Quoted 61 lines · Moritz Fischer

Hi Richard, Russ,

Quoted 46 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
firmware: stratix10-svc: extend SVC driver to get the firmware version
fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
fpga: of-fpga-region: add authenticate-fpga-config property
dt-bindings: fpga: add authenticate-fpga-config property
fpga: stratix10-soc: extend driver for bitstream authentication

  .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++
  drivers/firmware/stratix10-svc.c                   | 12 ++++-
  drivers/fpga/of-fpga-region.c                      | 24 ++++++---
  drivers/fpga/stratix10-soc.c                       | 62 +++++++++++++++++++---
  include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++-
  .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++-
  include/linux/fpga/fpga-mgr.h                      |  3 ++
  7 files changed, 125 insertions(+), 18 deletions(-)
Quoted 3 lines

--
2.7.4

Apologies for the epic delay in getting back to this, I took another look at this patchset and Russ' patchset.

TL;DR I'm not really a fan of using device-tree overlays for this (and again, apologies, I should've voiced this earlier ...).

Anyways, let's find a common API for this and Russ' work, they're trying to achieve the same / similar thing, they should use the same API.

Quoted 7 lines
Quoted 3 lines

I'd like to re-invetigate the possiblity to extend FPGA Manager with 'secure update' ops that work for both these use-cases (and I susspect hte XRT patchset will follow with a similar requirement, right after).

The xrt patchset makes heavy use of device trees.

What is the general guidance for device tree usage ?

I'm not generally against using device tree, it has its place. To describe hardware (and hardware *changes* with overlays) :)

What I don't like about this particular implementation w.r.t device-tree usage is that it uses DT overlays as a mechanism to program the flash -- in place of having an API to do so.

Quoted 9 lines

One could add device-nodes during the DT overlay application, while the FPGA doesn't actually get programmed with a new runtime image -- meaning live DT and actual hardware state diverged -- worst case it'd crash.

So when roughly at the same time (from the same company even) we have two patchsets that do similar things with radically different APIs I think we should pause, and reflect on whether we can come up with something that works for both :)

I discussed with Russ and studies his patches, came to realize that the work we had to accomplish was not same or similar. What I want to achieve is to verify the identity of the bitstream, which is like doing a "dry-run" to FPGA configuration.

Performing FPGA configuration (full or partial) through the device tree overlay is a method widely used by our customers.

Russ's approach utilizes a different user API which is a set of sysfs files.

If we depart from device tree overlay, then the end-user must utilize 2 different mechanism or APIs (device tree overlay is used for full/partial configuration, and sysfs is used for bitstream authentication). Similarly low-level FPGA manager driver also needs to add additional codes. For the end-user the single and simple mechanism is always better choice, device tree overlay should be a better way to achieve that goal.

Regards,
Richard

Quoted 5 lines

TL;DR the firmware parts to authenticate the bitstream look fine to me, the way we tie it into the FPGA region I'm not a fan of.

- Moritz

#30 lore raw

Hi Moritz,

Not sure if you have chance to view Russ's comments and my comments. Pleas let us know what you think so I can act accordingly.

I had a few discussions with Russ, and we all realized that the goals we were trying to achieve were not as similar as they seemed.

Regards,
Richard

Quoted 112 lines · Richard Gong

Hi Moritz,

Quoted 82 lines · Moritz Fischer

Tom,

Quoted 64 lines · Tom Rix
Quoted 62 lines · Moritz Fischer

Hi Richard, Russ,

Quoted 45 lines · Richard

Hi Moritz,

Sorry for asking.

When you have chance, can you help review the version 5 patchset submitted on 02/09/21?

Regards,
Richard

-----Original Message-----
From: [email protected] <[email protected]>
Sent: Tuesday, February 9, 2021 4:20 PM
To: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Cc: Gong, Richard <[email protected]>
Subject: [PATCHv5 0/7] Extend Intel service layer, FPGA manager and region

From: Richard Gong <[email protected]>

This is 5th submission of Intel service layer and FPGA patches, which includes the missing standalone patch in the 4th submission.

This submission includes additional changes for Intel service layer driver to get the firmware version running at FPGA SoC device. Then FPGA manager driver, one of Intel service layer driver's client, can decide whether to handle the newly added bitstream authentication function based on the retrieved firmware version. So that we can maintain FPGA manager driver the back compatible.

Bitstream authentication makes sure a signed bitstream has valid signatures.

The customer sends the bitstream via FPGA framework and overlay, the firmware will authenticate the bitstream but not program the bitstream to device. If the authentication passes, the bitstream will be programmed into QSPI flash and will be expected to boot without issues.

Extend Intel service layer, FPGA manager and region drivers to support the bitstream authentication feature.

Richard Gong (7):
   firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
   firmware: stratix10-svc: add COMMAND_AUTHENTICATE_BITSTREAM flag
   firmware: stratix10-svc: extend SVC driver to get the firmware version
   fpga: fpga-mgr: add FPGA_MGR_BITSTREAM_AUTHENTICATE flag
   fpga: of-fpga-region: add authenticate-fpga-config property
   dt-bindings: fpga: add authenticate-fpga-config property
   fpga: stratix10-soc: extend driver for bitstream authentication

  .../devicetree/bindings/fpga/fpga-region.txt       | 10 ++++ drivers/firmware/stratix10-svc.c                   | 12 ++++- drivers/fpga/of-fpga-region.c                      | 24 ++++++--- drivers/fpga/stratix10-soc.c                       | 62
+++++++++++++++++++---
  include/linux/firmware/intel/stratix10-smc.h       | 21 +++++++- .../linux/firmware/intel/stratix10-svc-client.h    | 11 +++- include/linux/fpga/fpga-mgr.h                      |  3 ++

Quoted 5 lines
  7 files changed, 125 insertions(+), 18 deletions(-)
Signature · 1 line
2.7.4

Apologies for the epic delay in getting back to this, I took another look at this patchset and Russ' patchset.

TL;DR I'm not really a fan of using device-tree overlays for this (and again, apologies, I should've voiced this earlier ...).

Anyways, let's find a common API for this and Russ' work, they're trying to achieve the same / similar thing, they should use the same API.

Quoted 7 lines
Quoted 3 lines

I'd like to re-invetigate the possiblity to extend FPGA Manager with 'secure update' ops that work for both these use-cases (and I susspect hte XRT patchset will follow with a similar requirement, right after).

The xrt patchset makes heavy use of device trees.

What is the general guidance for device tree usage ?

I'm not generally against using device tree, it has its place. To describe hardware (and hardware *changes* with overlays) :)

What I don't like about this particular implementation w.r.t device-tree usage is that it uses DT overlays as a mechanism to program the flash -- in place of having an API to do so.

Quoted 9 lines

One could add device-nodes during the DT overlay application, while the FPGA doesn't actually get programmed with a new runtime image -- meaning live DT and actual hardware state diverged -- worst case it'd crash.

So when roughly at the same time (from the same company even) we have two patchsets that do similar things with radically different APIs I think we should pause, and reflect on whether we can come up with something that works for both :)

I discussed with Russ and studies his patches, came to realize that the work we had to accomplish was not same or similar. What I want to achieve is to verify the identity of the bitstream, which is like doing a "dry-run" to FPGA configuration.

Performing FPGA configuration (full or partial) through the device tree overlay is a method widely used by our customers.

Russ's approach utilizes a different user API which is a set of sysfs files.

If we depart from device tree overlay, then the end-user must utilize 2 different mechanism or APIs (device tree overlay is used for full/partial configuration, and sysfs is used for bitstream authentication). Similarly low-level FPGA manager driver also needs to add additional codes. For the end-user the single and simple mechanism is always better choice, device tree overlay should be a better way to achieve that goal.

Regards,
Richard

Quoted 5 lines

TL;DR the firmware parts to authenticate the bitstream look fine to me, the way we tie it into the FPGA region I'm not a fan of.

- Moritz