Skip to contentSkip to conversation

enable Intel Alderlake RST VMD configuration

PATCH In review Oldřich Jedlička 3 participants 12 messages Aug 5, 2022 – Aug 25, 2022 MD RAID

lore
Share & tools
lore
#0 lore raw

cover enable Intel Alderlake RST VMD configuration

Hi,

I was testing RAID 0 on my Dell Precision 3570 with Intel i7-1255U Alderlake processor. The mdadm tool did not work initially (the RAID was not recognised), but I was able to get the RAID working when I simply set IMSM_NO_PLATFORM=1. Then I tracked the issue down to UEFI variables and discovered that the platform newly uses the 'RstVmdV' name. This exactly same variable was used in recent patch series called "mdadm-CI for-jes/20220728", so I updated the code to recognise the new variable name also for VMD controller case. I then verified the change by first running it in a debugger and then by putting it to initrd on Fedora 36 (kernel 5.17). The startup was fine, the RAID 0 was correctly setup.

I based my work on top of "mdadm-CI for-jes/20220728" patch series, it is required in order to apply the patch.

Regards,
Oldrich.

Oldřich Jedlička (1):
mdadm: enable Intel Alderlake RST VMD configuration

 platform-intel.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
Signature · 1 line
2.37.1
#1 lore raw

PATCH 1/1 mdadm: enable Intel Alderlake RST VMD configuration

Alderlake changed UEFI variable name to 'RstVmdV' also and for VMD devices, so check the updated name for VMD devices like it is done in the SATA case.

---
 platform-intel.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
platform-intel.c +12 -7
index a4d55a3..2f8e6af 100644
@@ -512,8 +512,8 @@ static const struct imsm_orom *find_imsm_hba_orom(struct sys_dev *hba)
 #define AHCI_PROP "RstSataV"
 #define AHCI_SSATA_PROP "RstsSatV"
 #define AHCI_TSATA_PROP "RsttSatV"
-#define AHCI_RST_PROP "RstVmdV"
-#define VMD_PROP "RstUefiV"
+#define RST_VMD_PROP "RstVmdV"
+#define RST_UEFI_PROP "RstUefiV"
 
 #define VENDOR_GUID \
 	EFI_GUID(0x193dfefa, 0xa445, 0x4302, 0x99, 0xd8, 0xef, 0x3a, 0xad, 0x1a, 0x04, 0xc6)
@@ -607,7 +607,8 @@ const struct imsm_orom *find_imsm_efi(struct sys_dev *hba)
 	struct orom_entry *ret;
 	static const char * const sata_efivars[] = {AHCI_PROP, AHCI_SSATA_PROP,
 						    AHCI_TSATA_PROP,
-						    AHCI_RST_PROP};
+						    RST_VMD_PROP};
+	static const char * const vmd_efivars[] = {RST_UEFI_PROP, RST_VMD_PROP};
 	unsigned long i;
 
 	if (check_env("IMSM_TEST_AHCI_EFI") || check_env("IMSM_TEST_SCU_EFI"))
@@ -640,10 +641,14 @@ const struct imsm_orom *find_imsm_efi(struct sys_dev *hba)
 
 		break;
 	case SYS_DEV_VMD:
-		if (!read_efi_variable(&orom, sizeof(orom), VMD_PROP,
-				       VENDOR_GUID))
-			break;
-		return NULL;
+		for (i = 0; i < ARRAY_SIZE(vmd_efivars); i++) {
+			if (!read_efi_variable(&orom, sizeof(orom),
+						vmd_efivars[i], VENDOR_GUID))
+				break;
+		}
+		if (i == ARRAY_SIZE(vmd_efivars))
+			return NULL;
+		break;
 	default:
 		return NULL;
 	}
Signature · 1 line
2.37.1
#2 lore raw

Re: [PATCH 1/1] mdadm: enable Intel Alderlake RST VMD configuration

Quoted 46 lines · On Fri, 5 Aug 2022 12:05:45 +0200 Oldřich Jedlička <[email protected]>…

Alderlake changed UEFI variable name to 'RstVmdV' also and for VMD devices, so check the updated name for VMD devices like it is done in the SATA case.

---
 platform-intel.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
platform-intel.c +2 -2
index a4d55a3..2f8e6af 100644
@@ -512,8 +512,8 @@ static const struct imsm_orom
 *find_imsm_hba_orom(struct sys_dev *hba) #define AHCI_PROP "RstSataV"
 #define AHCI_SSATA_PROP "RstsSatV"
 #define AHCI_TSATA_PROP "RsttSatV"
-#define AHCI_RST_PROP "RstVmdV"
-#define VMD_PROP "RstUefiV"
+#define RST_VMD_PROP "RstVmdV"
+#define RST_UEFI_PROP "RstUefiV"
 
 #define VENDOR_GUID \
 	EFI_GUID(0x193dfefa, 0xa445, 0x4302, 0x99, 0xd8, 0xef, 0x3a,

0xad, 0x1a, 0x04, 0xc6) @@ -607,7 +607,8 @@ const struct imsm_orom *find_imsm_efi(struct sys_dev *hba) struct orom_entry *ret;
static const char * const sata_efivars[] = {AHCI_PROP,
AHCI_SSATA_PROP, AHCI_TSATA_PROP,

platform-intel.c +2 -1
-						    AHCI_RST_PROP};
+						    RST_VMD_PROP};
+	static const char * const vmd_efivars[] = {RST_UEFI_PROP,

RST_VMD_PROP}; unsigned long i;

if (check_env("IMSM_TEST_AHCI_EFI") ||
check_env("IMSM_TEST_SCU_EFI")) @@ -640,10 +641,14 @@ const struct imsm_orom *find_imsm_efi(struct sys_dev *hba)

platform-intel.c +3 -4
 		break;
 	case SYS_DEV_VMD:
-		if (!read_efi_variable(&orom, sizeof(orom), VMD_PROP,
-				       VENDOR_GUID))
-			break;
-		return NULL;
+		for (i = 0; i < ARRAY_SIZE(vmd_efivars); i++) {
+			if (!read_efi_variable(&orom, sizeof(orom),
+						vmd_efivars[i],

VENDOR_GUID))

platform-intel.c const struct imsm_orom *find_imsm_efi(struct sys_dev *hba) · line 654

Quoted 8 lines
platform-intel.c +5 -0
+				break;
+		}
+		if (i == ARRAY_SIZE(vmd_efivars))
+			return NULL;
+		break;
 	default:
 		return NULL;
 	}

Hi,

please have a look at the following mail:
https://marc.info/?l=linux-raid&m=165969352101643&w=2

Regards,
Kinga Tanska

#3 lore raw

Re: [PATCH 1/1] mdadm: enable Intel Alderlake RST VMD configuration

pá 5. 8. 2022 v 13:56 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 61 lines
Quoted 46 lines · On Fri, 5 Aug 2022 12:05:45 +0200 Oldřich Jedlička <[email protected]>…

Alderlake changed UEFI variable name to 'RstVmdV' also and for VMD devices, so check the updated name for VMD devices like it is done in the SATA case.

---
 platform-intel.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
platform-intel.c +2 -2
index a4d55a3..2f8e6af 100644
@@ -512,8 +512,8 @@ static const struct imsm_orom
 *find_imsm_hba_orom(struct sys_dev *hba) #define AHCI_PROP "RstSataV"
 #define AHCI_SSATA_PROP "RstsSatV"
 #define AHCI_TSATA_PROP "RsttSatV"
-#define AHCI_RST_PROP "RstVmdV"
-#define VMD_PROP "RstUefiV"
+#define RST_VMD_PROP "RstVmdV"
+#define RST_UEFI_PROP "RstUefiV"
 
 #define VENDOR_GUID \
      EFI_GUID(0x193dfefa, 0xa445, 0x4302, 0x99, 0xd8, 0xef, 0x3a,

0xad, 0x1a, 0x04, 0xc6) @@ -607,7 +607,8 @@ const struct imsm_orom

*find_imsm_efi(struct sys_dev *hba) struct orom_entry *ret;
      static const char * const sata_efivars[] = {AHCI_PROP,

AHCI_SSATA_PROP, AHCI_TSATA_PROP,

platform-intel.c +2 -1
-                                                 AHCI_RST_PROP};
+                                                 RST_VMD_PROP};
+     static const char * const vmd_efivars[] = {RST_UEFI_PROP,

RST_VMD_PROP}; unsigned long i;

      if (check_env("IMSM_TEST_AHCI_EFI") ||
check_env("IMSM_TEST_SCU_EFI")) @@ -640,10 +641,14 @@ const struct
imsm_orom *find_imsm_efi(struct sys_dev *hba)
              break;
      case SYS_DEV_VMD:
platform-intel.c +3 -4
-             if (!read_efi_variable(&orom, sizeof(orom), VMD_PROP,
-                                    VENDOR_GUID))
-                     break;
-             return NULL;
+             for (i = 0; i < ARRAY_SIZE(vmd_efivars); i++) {
+                     if (!read_efi_variable(&orom, sizeof(orom),
+                                             vmd_efivars[i],

VENDOR_GUID))

platform-intel.c const struct imsm_orom *find_imsm_efi(struct sys_dev *hba) · line 654

Quoted 8 lines
platform-intel.c +5 -0
+                             break;
+             }
+             if (i == ARRAY_SIZE(vmd_efivars))
+                     return NULL;
+             break;
      default:
              return NULL;
      }

Hi,

Quoted 2 lines

please have a look at the following mail:
https://marc.info/?l=linux-raid&m=165969352101643&w=2

Sorry for double-posting, I received rejection emails regarding HTML content. Gmail switched to HTML.

Hi, the described issue applies specifically in the SYS_DEV_SATA (SATA configuration) case, so it should not apply to SYS_DEV_VMD (VMD configuration) one.

For me, the platform output looks reasonable (I have RAID 0 active):

#> sudo mdadm --detail-platform
Platform : Intel(R) Rapid Storage Technology
Version : 19.0.7.5579
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 32
Max Volumes : 2 per array, 4 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:00/0000:00:0e.0 (VMD) NVMe under VMD : /dev/nvme0n1 (S6P1NS0T318266R) NVMe under VMD : /dev/nvme1n1 (S6P1NS0T318223V)

Without the patch the platform isn't even recognized. Common to both changes is the usage of the new UEFI variable 'RstVmdV', not the changes to the controller.

Regards,
Oldrich.

Quoted 3 lines

Regards,
Kinga Tanska

#4 lore raw

Re: [PATCH 1/1] mdadm: enable Intel Alderlake RST VMD configuration

Hello Oldřich,

Quoted 9 lines · On Fri, 5 Aug 2022 14:50:36 +0200 Oldřich Jedlička <[email protected]> wrote:

pá 5. 8. 2022 v 13:56 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 7 lines
Quoted 3 lines · On Fri, 5 Aug 2022 12:05:45 +0200 Oldřich Jedlička <[email protected]>…

Alderlake changed UEFI variable name to 'RstVmdV' also and for VMD devices, so check the updated name for VMD devices like it is done in the SATA case.

Alderlake is RST so it doesn't changed, it is different. It is a support for RST family product.

Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
---
 platform-intel.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
Quoted 12 lines
Quoted 4 lines
Quoted 4 lines
platform-intel.c +0 -0
index a4d55a3..2f8e6af 100644
Quoted 8 lines

platform-intel.c static const struct imsm_orom · line 516

Quoted 8 lines
platform-intel.c +2 -2
@@ -512,8 +512,8 @@ static const struct imsm_orom
 *find_imsm_hba_orom(struct sys_dev *hba) #define AHCI_PROP "RstSataV"
 #define AHCI_SSATA_PROP "RstsSatV"
 #define AHCI_TSATA_PROP "RsttSatV"
-#define AHCI_RST_PROP "RstVmdV"
-#define VMD_PROP "RstUefiV"
+#define RST_VMD_PROP "RstVmdV"
+#define RST_UEFI_PROP "RstUefiV"

There are two products RSTe/ VROC and RST. Here you are adding support for RST platform. Please name it accordingly (yeah, I know that naming is confusing). I propose:

#define RST_VMD_PROP "RstVmdV"
#define VROC_VMD_PROP "RstUefiV"

Quoted 60 lines
Quoted 34 lines
Quoted 25 lines
platform-intel.c +0 -0
 #define VENDOR_GUID \
      EFI_GUID(0x193dfefa, 0xa445, 0x4302, 0x99, 0xd8, 0xef, 0x3a,

0xad, 0x1a, 0x04, 0xc6) @@ -607,7 +607,8 @@ const struct imsm_orom

*find_imsm_efi(struct sys_dev *hba) struct orom_entry *ret;
      static const char * const sata_efivars[] = {AHCI_PROP,

AHCI_SSATA_PROP, AHCI_TSATA_PROP,

platform-intel.c +2 -1
-                                                 AHCI_RST_PROP};
+                                                 RST_VMD_PROP};
+     static const char * const vmd_efivars[] = {RST_UEFI_PROP,

RST_VMD_PROP}; unsigned long i;

      if (check_env("IMSM_TEST_AHCI_EFI") ||
check_env("IMSM_TEST_SCU_EFI")) @@ -640,10 +641,14 @@ const struct
imsm_orom *find_imsm_efi(struct sys_dev *hba)
              break;
      case SYS_DEV_VMD:
platform-intel.c +3 -4
-             if (!read_efi_variable(&orom, sizeof(orom), VMD_PROP,
-                                    VENDOR_GUID))
-                     break;
-             return NULL;
+             for (i = 0; i < ARRAY_SIZE(vmd_efivars); i++) {
+                     if (!read_efi_variable(&orom, sizeof(orom),
+                                             vmd_efivars[i],

VENDOR_GUID))

platform-intel.c const struct imsm_orom *find_imsm_efi(struct sys_dev *hba) · line 654

Quoted 8 lines
platform-intel.c +5 -0
+                             break;
+             }
+             if (i == ARRAY_SIZE(vmd_efivars))
+                     return NULL;
+             break;
      default:
              return NULL;
      }
Quoted 4 lines

Hi,

please have a look at the following mail:
https://marc.info/?l=linux-raid&m=165969352101643&w=2

Hi, the described issue applies specifically in the SYS_DEV_SATA (SATA configuration) case, so it should not apply to SYS_DEV_VMD (VMD configuration) one.

For me, the platform output looks reasonable (I have RAID 0 active):

#> sudo mdadm --detail-platform
Platform : Intel(R) Rapid Storage Technology
Version : 19.0.7.5579
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 32
Max Volumes : 2 per array, 4 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:00/0000:00:0e.0 (VMD) NVMe under VMD : /dev/nvme0n1 (S6P1NS0T318266R) NVMe under VMD : /dev/nvme1n1 (S6P1NS0T318223V)

Quoted 4 lines

Without the patch the platform isn't even recognized. Common to both changes is the usage of the new UEFI variable 'RstVmdV', not the changes to the controller.

This version is different than one sent by Coly. We will test that to see if it doesn't cause regression in our VROC product.
I noted some nits, we will test this version anyway.

Thanks,
Mariusz

#5 lore raw

Re: [PATCH 1/1] mdadm: enable Intel Alderlake RST VMD configuration

Hi Mariusz,

čt 11. 8. 2022 v 12:18 odesílatel Mariusz Tkaczyk
<[email protected]> napsal:

Quoted 104 lines

Hello Oldřich,

Quoted 9 lines · On Fri, 5 Aug 2022 14:50:36 +0200 Oldřich Jedlička <[email protected]> wrote:

pá 5. 8. 2022 v 13:56 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 7 lines
Quoted 3 lines · On Fri, 5 Aug 2022 12:05:45 +0200 Oldřich Jedlička <[email protected]>…

Alderlake changed UEFI variable name to 'RstVmdV' also and for VMD devices, so check the updated name for VMD devices like it is done in the SATA case.

Alderlake is RST so it doesn't changed, it is different. It is a support for RST family product.

Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
---
 platform-intel.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
Quoted 12 lines
Quoted 4 lines
Quoted 4 lines
platform-intel.c +0 -0
index a4d55a3..2f8e6af 100644
Quoted 8 lines

platform-intel.c static const struct imsm_orom · line 516

Quoted 8 lines
platform-intel.c +2 -2
@@ -512,8 +512,8 @@ static const struct imsm_orom
 *find_imsm_hba_orom(struct sys_dev *hba) #define AHCI_PROP "RstSataV"
 #define AHCI_SSATA_PROP "RstsSatV"
 #define AHCI_TSATA_PROP "RsttSatV"
-#define AHCI_RST_PROP "RstVmdV"
-#define VMD_PROP "RstUefiV"
+#define RST_VMD_PROP "RstVmdV"
+#define RST_UEFI_PROP "RstUefiV"

There are two products RSTe/ VROC and RST. Here you are adding support for RST platform. Please name it accordingly (yeah, I know that naming is confusing). I propose:

#define RST_VMD_PROP "RstVmdV"
#define VROC_VMD_PROP "RstUefiV"

Quoted 60 lines
Quoted 34 lines
Quoted 25 lines
platform-intel.c +0 -0
 #define VENDOR_GUID \
      EFI_GUID(0x193dfefa, 0xa445, 0x4302, 0x99, 0xd8, 0xef, 0x3a,

0xad, 0x1a, 0x04, 0xc6) @@ -607,7 +607,8 @@ const struct imsm_orom

*find_imsm_efi(struct sys_dev *hba) struct orom_entry *ret;
      static const char * const sata_efivars[] = {AHCI_PROP,

AHCI_SSATA_PROP, AHCI_TSATA_PROP,

platform-intel.c +2 -1
-                                                 AHCI_RST_PROP};
+                                                 RST_VMD_PROP};
+     static const char * const vmd_efivars[] = {RST_UEFI_PROP,

RST_VMD_PROP}; unsigned long i;

      if (check_env("IMSM_TEST_AHCI_EFI") ||
check_env("IMSM_TEST_SCU_EFI")) @@ -640,10 +641,14 @@ const struct
imsm_orom *find_imsm_efi(struct sys_dev *hba)
              break;
      case SYS_DEV_VMD:
platform-intel.c +3 -4
-             if (!read_efi_variable(&orom, sizeof(orom), VMD_PROP,
-                                    VENDOR_GUID))
-                     break;
-             return NULL;
+             for (i = 0; i < ARRAY_SIZE(vmd_efivars); i++) {
+                     if (!read_efi_variable(&orom, sizeof(orom),
+                                             vmd_efivars[i],

VENDOR_GUID))

platform-intel.c const struct imsm_orom *find_imsm_efi(struct sys_dev *hba) · line 654

Quoted 8 lines
platform-intel.c +5 -0
+                             break;
+             }
+             if (i == ARRAY_SIZE(vmd_efivars))
+                     return NULL;
+             break;
      default:
              return NULL;
      }
Quoted 4 lines

Hi,

please have a look at the following mail:
https://marc.info/?l=linux-raid&m=165969352101643&w=2

Hi, the described issue applies specifically in the SYS_DEV_SATA (SATA configuration) case, so it should not apply to SYS_DEV_VMD (VMD configuration) one.

For me, the platform output looks reasonable (I have RAID 0 active):

#> sudo mdadm --detail-platform
Platform : Intel(R) Rapid Storage Technology
Version : 19.0.7.5579
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 32
Max Volumes : 2 per array, 4 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:00/0000:00:0e.0 (VMD) NVMe under VMD : /dev/nvme0n1 (S6P1NS0T318266R) NVMe under VMD : /dev/nvme1n1 (S6P1NS0T318223V)

Quoted 8 lines
Quoted 4 lines

Without the patch the platform isn't even recognized. Common to both changes is the usage of the new UEFI variable 'RstVmdV', not the changes to the controller.

This version is different than one sent by Coly. We will test that to see if it doesn't cause regression in our VROC product.
I noted some nits, we will test this version anyway.

Should I update the description (to just mention adding support for Alderlake RST on VMD platform), change the defines in the patch, and send V2, or you will send (possibly) updated version after your testing?

The patch was tested on my Dell Precision 3570 notebook with Intel i7-1255U CPU and two 2TB NVMe disks.

Cheers,
Oldřich.

Quoted 3 lines

Thanks,
Mariusz

#6 lore raw

Re: [PATCH 1/1] mdadm: enable Intel Alderlake RST VMD configuration

Quoted 117 lines · On Thu, 11 Aug 2022 21:38:59 +0200 Oldřich Jedlička <[email protected]>…

Hi Mariusz,

čt 11. 8. 2022 v 12:18 odesílatel Mariusz Tkaczyk
<[email protected]> napsal:

Quoted 104 lines

Hello Oldřich,

Quoted 9 lines · On Fri, 5 Aug 2022 14:50:36 +0200 Oldřich Jedlička <[email protected]> wrote:

pá 5. 8. 2022 v 13:56 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 7 lines
Quoted 3 lines · On Fri, 5 Aug 2022 12:05:45 +0200 Oldřich Jedlička <[email protected]>…

Alderlake changed UEFI variable name to 'RstVmdV' also and for VMD devices, so check the updated name for VMD devices like it is done in the SATA case.

Alderlake is RST so it doesn't changed, it is different. It is a support for RST family product.

Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
---
 platform-intel.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
Quoted 12 lines
Quoted 4 lines
Quoted 4 lines
platform-intel.c +0 -0
index a4d55a3..2f8e6af 100644
Quoted 8 lines

platform-intel.c static const struct imsm_orom · line 516

Quoted 8 lines
platform-intel.c +2 -2
@@ -512,8 +512,8 @@ static const struct imsm_orom
 *find_imsm_hba_orom(struct sys_dev *hba) #define AHCI_PROP "RstSataV"
 #define AHCI_SSATA_PROP "RstsSatV"
 #define AHCI_TSATA_PROP "RsttSatV"
-#define AHCI_RST_PROP "RstVmdV"
-#define VMD_PROP "RstUefiV"
+#define RST_VMD_PROP "RstVmdV"
+#define RST_UEFI_PROP "RstUefiV"  

There are two products RSTe/ VROC and RST. Here you are adding support for RST platform. Please name it accordingly (yeah, I know that naming is confusing). I propose:

#define RST_VMD_PROP "RstVmdV"
#define VROC_VMD_PROP "RstUefiV"

Quoted 60 lines
Quoted 34 lines
Quoted 25 lines
platform-intel.c +0 -0
 #define VENDOR_GUID \
      EFI_GUID(0x193dfefa, 0xa445, 0x4302, 0x99, 0xd8, 0xef, 0x3a,

0xad, 0x1a, 0x04, 0xc6) @@ -607,7 +607,8 @@ const struct imsm_orom

*find_imsm_efi(struct sys_dev *hba) struct orom_entry *ret;
      static const char * const sata_efivars[] = {AHCI_PROP,

AHCI_SSATA_PROP, AHCI_TSATA_PROP,

platform-intel.c +2 -1
-                                                 AHCI_RST_PROP};
+                                                 RST_VMD_PROP};
+     static const char * const vmd_efivars[] = {RST_UEFI_PROP,

RST_VMD_PROP}; unsigned long i;

      if (check_env("IMSM_TEST_AHCI_EFI") ||
check_env("IMSM_TEST_SCU_EFI")) @@ -640,10 +641,14 @@ const struct
imsm_orom *find_imsm_efi(struct sys_dev *hba)
              break;
      case SYS_DEV_VMD:
platform-intel.c +3 -4
-             if (!read_efi_variable(&orom, sizeof(orom), VMD_PROP,
-                                    VENDOR_GUID))
-                     break;
-             return NULL;
+             for (i = 0; i < ARRAY_SIZE(vmd_efivars); i++) {
+                     if (!read_efi_variable(&orom, sizeof(orom),
+                                             vmd_efivars[i],

VENDOR_GUID))

platform-intel.c const struct imsm_orom *find_imsm_efi(struct sys_dev *hba) · line 654

Quoted 8 lines
platform-intel.c +5 -0
+                             break;
+             }
+             if (i == ARRAY_SIZE(vmd_efivars))
+                     return NULL;
+             break;
      default:
              return NULL;
      }  
Quoted 4 lines

Hi,

please have a look at the following mail:
https://marc.info/?l=linux-raid&m=165969352101643&w=2

Hi, the described issue applies specifically in the SYS_DEV_SATA (SATA configuration) case, so it should not apply to SYS_DEV_VMD (VMD configuration) one.

For me, the platform output looks reasonable (I have RAID 0 active):

#> sudo mdadm --detail-platform
Platform : Intel(R) Rapid Storage Technology
Version : 19.0.7.5579
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 32
Max Volumes : 2 per array, 4 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:00/0000:00:0e.0 (VMD) NVMe under VMD : /dev/nvme0n1 (S6P1NS0T318266R) NVMe under VMD : /dev/nvme1n1 (S6P1NS0T318223V)

Quoted 8 lines
Quoted 4 lines

Without the patch the platform isn't even recognized. Common to both changes is the usage of the new UEFI variable 'RstVmdV', not the changes to the controller.

This version is different than one sent by Coly. We will test that to see if it doesn't cause regression in our VROC product.
I noted some nits, we will test this version anyway.

Quoted 8 lines

Should I update the description (to just mention adding support for Alderlake RST on VMD platform), change the defines in the patch, and send V2, or you will send (possibly) updated version after your testing?

The patch was tested on my Dell Precision 3570 notebook with Intel i7-1255U CPU and two 2TB NVMe disks.

Please do. This is your patch, so I don't want to take a your glory :)

Thanks,
Mariusz

#7 lore raw

Re: [PATCH 1/1] mdadm: enable Intel Alderlake RST VMD configuration

In reply to Oldřich Jedlička
Quoted 90 lines · On Fri, 5 Aug 2022 14:50:36 +0200 Oldřich Jedlička <[email protected]> wrote:

pá 5. 8. 2022 v 13:56 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 62 lines
Quoted 47 lines · On Fri, 5 Aug 2022 12:05:45 +0200 Oldřich Jedlička <[email protected]>…

Alderlake changed UEFI variable name to 'RstVmdV' also and for VMD devices, so check the updated name for VMD devices like it is done in the SATA case.

---
 platform-intel.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
platform-intel.c +0 -0
index a4d55a3..2f8e6af 100644
@@ -512,8 +512,8 @@ static const struct imsm_orom
 *find_imsm_hba_orom(struct sys_dev *hba) #define AHCI_PROP

"RstSataV" #define AHCI_SSATA_PROP "RstsSatV"
#define AHCI_TSATA_PROP "RsttSatV"

platform-intel.c +2 -2
-#define AHCI_RST_PROP "RstVmdV"
-#define VMD_PROP "RstUefiV"
+#define RST_VMD_PROP "RstVmdV"
+#define RST_UEFI_PROP "RstUefiV"
 
 #define VENDOR_GUID \
      EFI_GUID(0x193dfefa, 0xa445, 0x4302, 0x99, 0xd8, 0xef, 0x3a,

0xad, 0x1a, 0x04, 0xc6) @@ -607,7 +607,8 @@ const struct imsm_orom

*find_imsm_efi(struct sys_dev *hba) struct orom_entry *ret;
      static const char * const sata_efivars[] = {AHCI_PROP,

AHCI_SSATA_PROP, AHCI_TSATA_PROP,

platform-intel.c +2 -1
-                                                 AHCI_RST_PROP};
+                                                 RST_VMD_PROP};
+     static const char * const vmd_efivars[] = {RST_UEFI_PROP,

RST_VMD_PROP}; unsigned long i;

      if (check_env("IMSM_TEST_AHCI_EFI") ||
check_env("IMSM_TEST_SCU_EFI")) @@ -640,10 +641,14 @@ const struct
imsm_orom *find_imsm_efi(struct sys_dev *hba)
              break;
      case SYS_DEV_VMD:

- if (!read_efi_variable(&orom, sizeof(orom),
VMD_PROP,

platform-intel.c +3 -3
-                                    VENDOR_GUID))
-                     break;
-             return NULL;
+             for (i = 0; i < ARRAY_SIZE(vmd_efivars); i++) {
+                     if (!read_efi_variable(&orom, sizeof(orom),
+                                             vmd_efivars[i],

VENDOR_GUID))

platform-intel.c const struct imsm_orom *find_imsm_efi(struct sys_dev *hba) · line 654

Quoted 8 lines
platform-intel.c +5 -0
+                             break;
+             }
+             if (i == ARRAY_SIZE(vmd_efivars))
+                     return NULL;
+             break;
      default:
              return NULL;
      }

Hi,

Quoted 2 lines

please have a look at the following mail:
https://marc.info/?l=linux-raid&m=165969352101643&w=2

Sorry for double-posting, I received rejection emails regarding HTML content. Gmail switched to HTML.

Hi, the described issue applies specifically in the SYS_DEV_SATA (SATA configuration) case, so it should not apply to SYS_DEV_VMD (VMD configuration) one.

For me, the platform output looks reasonable (I have RAID 0 active):

#> sudo mdadm --detail-platform
Platform : Intel(R) Rapid Storage Technology
Version : 19.0.7.5579
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 32
Max Volumes : 2 per array, 4 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:00/0000:00:0e.0 (VMD) NVMe under VMD : /dev/nvme0n1 (S6P1NS0T318266R) NVMe under VMD : /dev/nvme1n1 (S6P1NS0T318223V)

Quoted 10 lines

Without the patch the platform isn't even recognized. Common to both changes is the usage of the new UEFI variable 'RstVmdV', not the changes to the controller.

Regards,
Oldrich.

Quoted 3 lines

Regards,
Kinga Tanska

Hello,

I've done test to check if your patch doesn't change way of recognizing controllers. I've got two SATA controllers - first one with AHCI mode and the other one with RAID MODE enabled. Command "mdadm --detail-platform" should display info only about controller with RAID MODE:

mdadm: imsm capabilities not found for controller:
/sys/devices/pci0000:00/0000:00:17.0 (type SATA) Platform : Intel(R) Rapid Storage Technology enterprise Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:11.5 (SATA)
Port4 : - non-disk device (TEAC DV-W28S-B) - Port0 : - no device attached -
Port1 : - no device attached -
Port2 : - no device attached -
Port3 : - no device attached -
Port5 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 24
Max Volumes : 2 per array, 24 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:17/0000:17:05.5 (VMD) I/O Controller : /sys/devices/pci0000:d7/0000:d7:05.5 (VMD) NVMe under VMD : /dev/nvme3n1 (PHLJ915000201P0FGN) NVMe under VMD : /dev/nvme2n1 (PHLJ915003201P0FGN) I/O Controller : /sys/devices/pci0000:85/0000:85:05.5 (VMD) I/O Controller : /sys/devices/pci0000:ae/0000:ae:05.5 (VMD) I/O Controller : /sys/devices/pci0000:5d/0000:5d:05.5 (VMD) NVMe under VMD : /dev/nvme0n1 (PHFT536600GT400GGN) NVMe under VMD : /dev/nvme1n1 (CVFT523100122P0KGN)

But with your patch this command returns info about all controllers which is not correct:

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:17.0 (SATA)
Port2 : /dev/sdd (WDEBMLJ2)
Port3 : /dev/sde (BTPR2300034W120LGN) Port0 : /dev/sdb (CVTS5396007S240JGN) Port1 : /dev/sdc (ZDE0XM9Z)
Port4 : - no device attached -
Port5 : - no device attached -
Port6 : - no device attached -
Port7 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:11.5 (SATA)
Port4 : - non-disk device (TEAC DV-W28S-B) - Port0 : - no device attached -
Port1 : - no device attached -
Port2 : - no device attached -
Port3 : - no device attached -
Port5 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 24
Max Volumes : 2 per array, 24 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:17/0000:17:05.5 (VMD) I/O Controller : /sys/devices/pci0000:d7/0000:d7:05.5 (VMD) NVMe under VMD : /dev/nvme3n1 (PHLJ915000201P0FGN) NVMe under VMD : /dev/nvme2n1 (PHLJ915003201P0FGN) I/O Controller : /sys/devices/pci0000:85/0000:85:05.5 (VMD) I/O Controller : /sys/devices/pci0000:ae/0000:ae:05.5 (VMD) I/O Controller : /sys/devices/pci0000:5d/0000:5d:05.5 (VMD) NVMe under VMD : /dev/nvme0n1 (PHFT536600GT400GGN) NVMe under VMD : /dev/nvme1n1 (CVFT523100122P0KGN)

Please analyze it.
Regards,
Kinga Tanska

#8 lore raw

Re: [PATCH 1/1] mdadm: enable Intel Alderlake RST VMD configuration

čt 18. 8. 2022 v 16:12 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 205 lines
Quoted 90 lines · On Fri, 5 Aug 2022 14:50:36 +0200 Oldřich Jedlička <[email protected]> wrote:

pá 5. 8. 2022 v 13:56 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 62 lines
Quoted 47 lines · On Fri, 5 Aug 2022 12:05:45 +0200 Oldřich Jedlička <[email protected]>…

Alderlake changed UEFI variable name to 'RstVmdV' also and for VMD devices, so check the updated name for VMD devices like it is done in the SATA case.

---
 platform-intel.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
platform-intel.c +0 -0
index a4d55a3..2f8e6af 100644
@@ -512,8 +512,8 @@ static const struct imsm_orom
 *find_imsm_hba_orom(struct sys_dev *hba) #define AHCI_PROP

"RstSataV" #define AHCI_SSATA_PROP "RstsSatV"
#define AHCI_TSATA_PROP "RsttSatV"

platform-intel.c +2 -2
-#define AHCI_RST_PROP "RstVmdV"
-#define VMD_PROP "RstUefiV"
+#define RST_VMD_PROP "RstVmdV"
+#define RST_UEFI_PROP "RstUefiV"
 
 #define VENDOR_GUID \
      EFI_GUID(0x193dfefa, 0xa445, 0x4302, 0x99, 0xd8, 0xef, 0x3a,

0xad, 0x1a, 0x04, 0xc6) @@ -607,7 +607,8 @@ const struct imsm_orom

*find_imsm_efi(struct sys_dev *hba) struct orom_entry *ret;
      static const char * const sata_efivars[] = {AHCI_PROP,

AHCI_SSATA_PROP, AHCI_TSATA_PROP,

platform-intel.c +2 -1
-                                                 AHCI_RST_PROP};
+                                                 RST_VMD_PROP};
+     static const char * const vmd_efivars[] = {RST_UEFI_PROP,

RST_VMD_PROP}; unsigned long i;

      if (check_env("IMSM_TEST_AHCI_EFI") ||
check_env("IMSM_TEST_SCU_EFI")) @@ -640,10 +641,14 @@ const struct
imsm_orom *find_imsm_efi(struct sys_dev *hba)
              break;
      case SYS_DEV_VMD:

- if (!read_efi_variable(&orom, sizeof(orom),
VMD_PROP,

platform-intel.c +3 -3
-                                    VENDOR_GUID))
-                     break;
-             return NULL;
+             for (i = 0; i < ARRAY_SIZE(vmd_efivars); i++) {
+                     if (!read_efi_variable(&orom, sizeof(orom),
+                                             vmd_efivars[i],

VENDOR_GUID))

platform-intel.c const struct imsm_orom *find_imsm_efi(struct sys_dev *hba) · line 654

Quoted 8 lines
platform-intel.c +5 -0
+                             break;
+             }
+             if (i == ARRAY_SIZE(vmd_efivars))
+                     return NULL;
+             break;
      default:
              return NULL;
      }

Hi,

Quoted 2 lines

please have a look at the following mail:
https://marc.info/?l=linux-raid&m=165969352101643&w=2

Sorry for double-posting, I received rejection emails regarding HTML content. Gmail switched to HTML.

Hi, the described issue applies specifically in the SYS_DEV_SATA (SATA configuration) case, so it should not apply to SYS_DEV_VMD (VMD configuration) one.

For me, the platform output looks reasonable (I have RAID 0 active):

#> sudo mdadm --detail-platform
Platform : Intel(R) Rapid Storage Technology
Version : 19.0.7.5579
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 32
Max Volumes : 2 per array, 4 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:00/0000:00:0e.0 (VMD) NVMe under VMD : /dev/nvme0n1 (S6P1NS0T318266R) NVMe under VMD : /dev/nvme1n1 (S6P1NS0T318223V)

Quoted 10 lines

Without the patch the platform isn't even recognized. Common to both changes is the usage of the new UEFI variable 'RstVmdV', not the changes to the controller.

Regards,
Oldrich.

Quoted 3 lines

Regards,
Kinga Tanska

Hello,

I've done test to check if your patch doesn't change way of recognizing controllers. I've got two SATA controllers - first one with AHCI mode and the other one with RAID MODE enabled. Command "mdadm --detail-platform" should display info only about controller with RAID MODE:

mdadm: imsm capabilities not found for controller:
/sys/devices/pci0000:00/0000:00:17.0 (type SATA) Platform : Intel(R) Rapid Storage Technology enterprise Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:11.5 (SATA)
Port4 : - non-disk device (TEAC DV-W28S-B) - Port0 : - no device attached -
Port1 : - no device attached -
Port2 : - no device attached -
Port3 : - no device attached -
Port5 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 24
Max Volumes : 2 per array, 24 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:17/0000:17:05.5 (VMD) I/O Controller : /sys/devices/pci0000:d7/0000:d7:05.5 (VMD) NVMe under VMD : /dev/nvme3n1 (PHLJ915000201P0FGN) NVMe under VMD : /dev/nvme2n1 (PHLJ915003201P0FGN) I/O Controller : /sys/devices/pci0000:85/0000:85:05.5 (VMD) I/O Controller : /sys/devices/pci0000:ae/0000:ae:05.5 (VMD) I/O Controller : /sys/devices/pci0000:5d/0000:5d:05.5 (VMD) NVMe under VMD : /dev/nvme0n1 (PHFT536600GT400GGN) NVMe under VMD : /dev/nvme1n1 (CVFT523100122P0KGN)

But with your patch this command returns info about all controllers which is not correct:

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:17.0 (SATA)
Port2 : /dev/sdd (WDEBMLJ2)
Port3 : /dev/sde (BTPR2300034W120LGN) Port0 : /dev/sdb (CVTS5396007S240JGN) Port1 : /dev/sdc (ZDE0XM9Z)
Port4 : - no device attached -
Port5 : - no device attached -
Port6 : - no device attached -
Port7 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:11.5 (SATA)
Port4 : - non-disk device (TEAC DV-W28S-B) - Port0 : - no device attached -
Port1 : - no device attached -
Port2 : - no device attached -
Port3 : - no device attached -
Port5 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 24
Max Volumes : 2 per array, 24 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:17/0000:17:05.5 (VMD) I/O Controller : /sys/devices/pci0000:d7/0000:d7:05.5 (VMD) NVMe under VMD : /dev/nvme3n1 (PHLJ915000201P0FGN) NVMe under VMD : /dev/nvme2n1 (PHLJ915003201P0FGN) I/O Controller : /sys/devices/pci0000:85/0000:85:05.5 (VMD) I/O Controller : /sys/devices/pci0000:ae/0000:ae:05.5 (VMD) I/O Controller : /sys/devices/pci0000:5d/0000:5d:05.5 (VMD) NVMe under VMD : /dev/nvme0n1 (PHFT536600GT400GGN) NVMe under VMD : /dev/nvme1n1 (CVFT523100122P0KGN)

Quoted 1 line

Please analyze it.

Hi Kinga Tanska,

My patch touches only the VMD controllers. I see that there are no changes to VMD display in your output. My patch is only about VMD.

I guess that your issue is identical to what you already wrote to RAID mailing list 5th of August. Correct? I cannot investigate this for you, I am neither the author of the change, nor I have RST enterprise RAID controller, sorry. Please contact the patch submitter/author.

Regards,
Oldrich.

Quoted 2 lines

Regards,
Kinga Tanska

#9 lore raw

Re: [PATCH 1/1] mdadm: enable Intel Alderlake RST VMD configuration

čt 18. 8. 2022 v 16:53 odesílatel Oldřich Jedlička <[email protected]> napsal:

Quoted 212 lines

čt 18. 8. 2022 v 16:12 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 205 lines
Quoted 90 lines · On Fri, 5 Aug 2022 14:50:36 +0200 Oldřich Jedlička <[email protected]> wrote:

pá 5. 8. 2022 v 13:56 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 62 lines
Quoted 47 lines · On Fri, 5 Aug 2022 12:05:45 +0200 Oldřich Jedlička <[email protected]>…

Alderlake changed UEFI variable name to 'RstVmdV' also and for VMD devices, so check the updated name for VMD devices like it is done in the SATA case.

---
 platform-intel.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
platform-intel.c +0 -0
index a4d55a3..2f8e6af 100644
@@ -512,8 +512,8 @@ static const struct imsm_orom
 *find_imsm_hba_orom(struct sys_dev *hba) #define AHCI_PROP

"RstSataV" #define AHCI_SSATA_PROP "RstsSatV"
#define AHCI_TSATA_PROP "RsttSatV"

platform-intel.c +2 -2
-#define AHCI_RST_PROP "RstVmdV"
-#define VMD_PROP "RstUefiV"
+#define RST_VMD_PROP "RstVmdV"
+#define RST_UEFI_PROP "RstUefiV"
 
 #define VENDOR_GUID \
      EFI_GUID(0x193dfefa, 0xa445, 0x4302, 0x99, 0xd8, 0xef, 0x3a,

0xad, 0x1a, 0x04, 0xc6) @@ -607,7 +607,8 @@ const struct imsm_orom

*find_imsm_efi(struct sys_dev *hba) struct orom_entry *ret;
      static const char * const sata_efivars[] = {AHCI_PROP,

AHCI_SSATA_PROP, AHCI_TSATA_PROP,

platform-intel.c +2 -1
-                                                 AHCI_RST_PROP};
+                                                 RST_VMD_PROP};
+     static const char * const vmd_efivars[] = {RST_UEFI_PROP,

RST_VMD_PROP}; unsigned long i;

      if (check_env("IMSM_TEST_AHCI_EFI") ||
check_env("IMSM_TEST_SCU_EFI")) @@ -640,10 +641,14 @@ const struct
imsm_orom *find_imsm_efi(struct sys_dev *hba)
              break;
      case SYS_DEV_VMD:

- if (!read_efi_variable(&orom, sizeof(orom),
VMD_PROP,

platform-intel.c +3 -3
-                                    VENDOR_GUID))
-                     break;
-             return NULL;
+             for (i = 0; i < ARRAY_SIZE(vmd_efivars); i++) {
+                     if (!read_efi_variable(&orom, sizeof(orom),
+                                             vmd_efivars[i],

VENDOR_GUID))

platform-intel.c const struct imsm_orom *find_imsm_efi(struct sys_dev *hba) · line 654

Quoted 8 lines
platform-intel.c +5 -0
+                             break;
+             }
+             if (i == ARRAY_SIZE(vmd_efivars))
+                     return NULL;
+             break;
      default:
              return NULL;
      }

Hi,

Quoted 2 lines

please have a look at the following mail:
https://marc.info/?l=linux-raid&m=165969352101643&w=2

Sorry for double-posting, I received rejection emails regarding HTML content. Gmail switched to HTML.

Hi, the described issue applies specifically in the SYS_DEV_SATA (SATA configuration) case, so it should not apply to SYS_DEV_VMD (VMD configuration) one.

For me, the platform output looks reasonable (I have RAID 0 active):

#> sudo mdadm --detail-platform
Platform : Intel(R) Rapid Storage Technology
Version : 19.0.7.5579
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 32
Max Volumes : 2 per array, 4 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:00/0000:00:0e.0 (VMD) NVMe under VMD : /dev/nvme0n1 (S6P1NS0T318266R) NVMe under VMD : /dev/nvme1n1 (S6P1NS0T318223V)

Quoted 10 lines

Without the patch the platform isn't even recognized. Common to both changes is the usage of the new UEFI variable 'RstVmdV', not the changes to the controller.

Regards,
Oldrich.

Quoted 3 lines

Regards,
Kinga Tanska

Hello,

I've done test to check if your patch doesn't change way of recognizing controllers. I've got two SATA controllers - first one with AHCI mode and the other one with RAID MODE enabled. Command "mdadm --detail-platform" should display info only about controller with RAID MODE:

mdadm: imsm capabilities not found for controller:
/sys/devices/pci0000:00/0000:00:17.0 (type SATA) Platform : Intel(R) Rapid Storage Technology enterprise Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:11.5 (SATA)
Port4 : - non-disk device (TEAC DV-W28S-B) - Port0 : - no device attached -
Port1 : - no device attached -
Port2 : - no device attached -
Port3 : - no device attached -
Port5 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 24
Max Volumes : 2 per array, 24 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:17/0000:17:05.5 (VMD) I/O Controller : /sys/devices/pci0000:d7/0000:d7:05.5 (VMD) NVMe under VMD : /dev/nvme3n1 (PHLJ915000201P0FGN) NVMe under VMD : /dev/nvme2n1 (PHLJ915003201P0FGN) I/O Controller : /sys/devices/pci0000:85/0000:85:05.5 (VMD) I/O Controller : /sys/devices/pci0000:ae/0000:ae:05.5 (VMD) I/O Controller : /sys/devices/pci0000:5d/0000:5d:05.5 (VMD) NVMe under VMD : /dev/nvme0n1 (PHFT536600GT400GGN) NVMe under VMD : /dev/nvme1n1 (CVFT523100122P0KGN)

But with your patch this command returns info about all controllers which is not correct:

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:17.0 (SATA)
Port2 : /dev/sdd (WDEBMLJ2)
Port3 : /dev/sde (BTPR2300034W120LGN) Port0 : /dev/sdb (CVTS5396007S240JGN) Port1 : /dev/sdc (ZDE0XM9Z)
Port4 : - no device attached -
Port5 : - no device attached -
Port6 : - no device attached -
Port7 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:11.5 (SATA)
Port4 : - non-disk device (TEAC DV-W28S-B) - Port0 : - no device attached -
Port1 : - no device attached -
Port2 : - no device attached -
Port3 : - no device attached -
Port5 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 24
Max Volumes : 2 per array, 24 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:17/0000:17:05.5 (VMD) I/O Controller : /sys/devices/pci0000:d7/0000:d7:05.5 (VMD) NVMe under VMD : /dev/nvme3n1 (PHLJ915000201P0FGN) NVMe under VMD : /dev/nvme2n1 (PHLJ915003201P0FGN) I/O Controller : /sys/devices/pci0000:85/0000:85:05.5 (VMD) I/O Controller : /sys/devices/pci0000:ae/0000:ae:05.5 (VMD) I/O Controller : /sys/devices/pci0000:5d/0000:5d:05.5 (VMD) NVMe under VMD : /dev/nvme0n1 (PHFT536600GT400GGN) NVMe under VMD : /dev/nvme1n1 (CVFT523100122P0KGN)

Quoted 1 line

Please analyze it.

Hi Kinga Tanska,

Quoted 7 lines

My patch touches only the VMD controllers. I see that there are no changes to VMD display in your output. My patch is only about VMD.

I guess that your issue is identical to what you already wrote to RAID mailing list 5th of August. Correct? I cannot investigate this for you, I am neither the author of the change, nor I have RST enterprise RAID controller, sorry. Please contact the patch submitter/author.

I just checked the patch code, I cannot debug this. You are most probably affected by patch which you already identified:

[PATCH 19/23] mdadm: enable Intel Alderlake RSTe configuration.

It tries to exclude SATA in AHCI mode (PCI class 0x010601) from the changed EFI variable check. What is your lspci -nn output? It could help in verifying that my guess is correct.

Cheers,
Oldrich.

Quoted 6 lines

Regards,
Oldrich.

Quoted 2 lines

Regards,
Kinga Tanska

#10 lore raw

Re: [PATCH 1/1] mdadm: enable Intel Alderlake RST VMD configuration

Quoted 231 lines · On Thu, 18 Aug 2022 17:21:20 +0200 Oldřich Jedlička <[email protected]>…

čt 18. 8. 2022 v 16:53 odesílatel Oldřich Jedlička <[email protected]> napsal:

Quoted 217 lines

čt 18. 8. 2022 v 16:12 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 210 lines
Quoted 95 lines · On Fri, 5 Aug 2022 14:50:36 +0200 Oldřich Jedlička <[email protected]> wrote:

pá 5. 8. 2022 v 13:56 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 67 lines
Quoted 52 lines · On Fri, 5 Aug 2022 12:05:45 +0200 Oldřich Jedlička <[email protected]>…

Alderlake changed UEFI variable name to 'RstVmdV' also and for VMD devices, so check the updated name for VMD devices like it is done in the SATA case.

---
 platform-intel.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
platform-intel.c +0 -0
index a4d55a3..2f8e6af 100644
@@ -512,8 +512,8 @@ static const struct imsm_orom
 *find_imsm_hba_orom(struct sys_dev *hba) #define AHCI_PROP

"RstSataV" #define AHCI_SSATA_PROP "RstsSatV"
#define AHCI_TSATA_PROP "RsttSatV"

platform-intel.c +2 -2
-#define AHCI_RST_PROP "RstVmdV"
-#define VMD_PROP "RstUefiV"
+#define RST_VMD_PROP "RstVmdV"
+#define RST_UEFI_PROP "RstUefiV"
 
 #define VENDOR_GUID \
      EFI_GUID(0x193dfefa, 0xa445, 0x4302, 0x99, 0xd8,

0xef, 0x3a, 0xad, 0x1a, 0x04, 0xc6) @@ -607,7 +607,8 @@ const struct imsm_orom *find_imsm_efi(struct sys_dev *hba) struct orom_entry *ret; static const char * const sata_efivars[] = {AHCI_PROP, AHCI_SSATA_PROP, AHCI_TSATA_PROP,
-
AHCI_RST_PROP};
+
RST_VMD_PROP};
+ static const char * const vmd_efivars[] =
{RST_UEFI_PROP, RST_VMD_PROP}; unsigned long i;

      if (check_env("IMSM_TEST_AHCI_EFI") ||
check_env("IMSM_TEST_SCU_EFI")) @@ -640,10 +641,14 @@ const
struct imsm_orom *find_imsm_efi(struct sys_dev *hba)
              break;
      case SYS_DEV_VMD:

- if (!read_efi_variable(&orom, sizeof(orom),
VMD_PROP,

platform-intel.c +1 -3
-                                    VENDOR_GUID))
-                     break;
-             return NULL;
+             for (i = 0; i < ARRAY_SIZE(vmd_efivars); i++)

{
+ if (!read_efi_variable(&orom,
sizeof(orom),
+
vmd_efivars[i], VENDOR_GUID))

platform-intel.c const struct imsm_orom *find_imsm_efi(struct sys_dev *hba) · line 654

Quoted 8 lines
platform-intel.c +5 -0
+                             break;
+             }
+             if (i == ARRAY_SIZE(vmd_efivars))
+                     return NULL;
+             break;
      default:
              return NULL;
      }  

Hi,

Quoted 2 lines

please have a look at the following mail:
https://marc.info/?l=linux-raid&m=165969352101643&w=2

Sorry for double-posting, I received rejection emails regarding HTML content. Gmail switched to HTML.

Hi, the described issue applies specifically in the SYS_DEV_SATA (SATA configuration) case, so it should not apply to SYS_DEV_VMD (VMD configuration) one.

For me, the platform output looks reasonable (I have RAID 0 active):
#> sudo mdadm --detail-platform
Platform : Intel(R) Rapid Storage Technology
Version : 19.0.7.5579
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 32
Max Volumes : 2 per array, 4 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:00/0000:00:0e.0 (VMD) NVMe under VMD : /dev/nvme0n1 (S6P1NS0T318266R) NVMe under VMD : /dev/nvme1n1 (S6P1NS0T318223V)

Quoted 10 lines

Without the patch the platform isn't even recognized. Common to both changes is the usage of the new UEFI variable 'RstVmdV', not the changes to the controller.

Regards,
Oldrich.

Quoted 3 lines

Regards,
Kinga Tanska

Hello,

I've done test to check if your patch doesn't change way of recognizing controllers. I've got two SATA controllers - first one with AHCI mode and the other one with RAID MODE enabled. Command "mdadm --detail-platform" should display info only about controller with RAID MODE:

mdadm: imsm capabilities not found for controller:
/sys/devices/pci0000:00/0000:00:17.0 (type SATA) Platform :
Intel(R) Rapid Storage Technology enterprise Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:11.5 (SATA)
Port4 : - non-disk device (TEAC DV-W28S-B) - Port0 : - no device attached -
Port1 : - no device attached -
Port2 : - no device attached -
Port3 : - no device attached -
Port5 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 24
Max Volumes : 2 per array, 24 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:17/0000:17:05.5 (VMD) I/O Controller : /sys/devices/pci0000:d7/0000:d7:05.5 (VMD) NVMe under VMD : /dev/nvme3n1 (PHLJ915000201P0FGN) NVMe under VMD : /dev/nvme2n1 (PHLJ915003201P0FGN) I/O Controller : /sys/devices/pci0000:85/0000:85:05.5 (VMD) I/O Controller : /sys/devices/pci0000:ae/0000:ae:05.5 (VMD) I/O Controller : /sys/devices/pci0000:5d/0000:5d:05.5 (VMD) NVMe under VMD : /dev/nvme0n1 (PHFT536600GT400GGN) NVMe under VMD : /dev/nvme1n1 (CVFT523100122P0KGN)

But with your patch this command returns info about all controllers which is not correct:

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:17.0 (SATA)
Port2 : /dev/sdd (WDEBMLJ2)
Port3 : /dev/sde (BTPR2300034W120LGN) Port0 : /dev/sdb (CVTS5396007S240JGN) Port1 : /dev/sdc (ZDE0XM9Z)
Port4 : - no device attached -
Port5 : - no device attached -
Port6 : - no device attached -
Port7 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:11.5 (SATA)
Port4 : - non-disk device (TEAC DV-W28S-B) - Port0 : - no device attached -
Port1 : - no device attached -
Port2 : - no device attached -
Port3 : - no device attached -
Port5 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 24
Max Volumes : 2 per array, 24 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:17/0000:17:05.5 (VMD) I/O Controller : /sys/devices/pci0000:d7/0000:d7:05.5 (VMD) NVMe under VMD : /dev/nvme3n1 (PHLJ915000201P0FGN) NVMe under VMD : /dev/nvme2n1 (PHLJ915003201P0FGN) I/O Controller : /sys/devices/pci0000:85/0000:85:05.5 (VMD) I/O Controller : /sys/devices/pci0000:ae/0000:ae:05.5 (VMD) I/O Controller : /sys/devices/pci0000:5d/0000:5d:05.5 (VMD) NVMe under VMD : /dev/nvme0n1 (PHFT536600GT400GGN) NVMe under VMD : /dev/nvme1n1 (CVFT523100122P0KGN)

Quoted 1 line

Please analyze it.

Hi Kinga Tanska,

Quoted 8 lines

My patch touches only the VMD controllers. I see that there are no changes to VMD display in your output. My patch is only about VMD.

I guess that your issue is identical to what you already wrote to RAID mailing list 5th of August. Correct? I cannot investigate this for you, I am neither the author of the change, nor I have RST enterprise RAID controller, sorry. Please contact the patch submitter/author.

I just checked the patch code, I cannot debug this. You are most probably affected by patch which you already identified:

Quoted 15 lines

[PATCH 19/23] mdadm: enable Intel Alderlake RSTe configuration.

It tries to exclude SATA in AHCI mode (PCI class 0x010601) from the changed EFI variable check. What is your lspci -nn output? It could help in verifying that my guess is correct.

Cheers,
Oldrich.

Quoted 6 lines

Regards,
Oldrich.

Quoted 2 lines

Regards,
Kinga Tanska

Hi Oldrich,

I've noticed that your patch is on top of mdadm-CI for-jes/20220728 branch. This branch contains previous patch for Alderlake RST VMD support ([PATCH 19/23] mdadm: enable Intel Alderlake RSTe configuration). This patch causes defect, which I've described. You were right saying that this patch affected me.
Please move your patch to the HEAD and then I've been able to test it again.

Thanks,
Kinga

#11 lore raw

Re: [PATCH 1/1] mdadm: enable Intel Alderlake RST VMD configuration

út 23. 8. 2022 v 16:03 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 253 lines
Quoted 231 lines · On Thu, 18 Aug 2022 17:21:20 +0200 Oldřich Jedlička <[email protected]>…

čt 18. 8. 2022 v 16:53 odesílatel Oldřich Jedlička <[email protected]> napsal:

Quoted 217 lines

čt 18. 8. 2022 v 16:12 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 210 lines
Quoted 95 lines · On Fri, 5 Aug 2022 14:50:36 +0200 Oldřich Jedlička <[email protected]> wrote:

pá 5. 8. 2022 v 13:56 odesílatel Kinga Tanska
<[email protected]> napsal:

Quoted 67 lines
Quoted 52 lines · On Fri, 5 Aug 2022 12:05:45 +0200 Oldřich Jedlička <[email protected]>…

Alderlake changed UEFI variable name to 'RstVmdV' also and for VMD devices, so check the updated name for VMD devices like it is done in the SATA case.

---
 platform-intel.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
platform-intel.c +0 -0
index a4d55a3..2f8e6af 100644
@@ -512,8 +512,8 @@ static const struct imsm_orom
 *find_imsm_hba_orom(struct sys_dev *hba) #define AHCI_PROP

"RstSataV" #define AHCI_SSATA_PROP "RstsSatV"
#define AHCI_TSATA_PROP "RsttSatV"

platform-intel.c +2 -2
-#define AHCI_RST_PROP "RstVmdV"
-#define VMD_PROP "RstUefiV"
+#define RST_VMD_PROP "RstVmdV"
+#define RST_UEFI_PROP "RstUefiV"
 
 #define VENDOR_GUID \
      EFI_GUID(0x193dfefa, 0xa445, 0x4302, 0x99, 0xd8,

0xef, 0x3a, 0xad, 0x1a, 0x04, 0xc6) @@ -607,7 +607,8 @@ const struct imsm_orom *find_imsm_efi(struct sys_dev *hba) struct orom_entry *ret; static const char * const sata_efivars[] = {AHCI_PROP, AHCI_SSATA_PROP, AHCI_TSATA_PROP,
-
AHCI_RST_PROP};
+
RST_VMD_PROP};
+ static const char * const vmd_efivars[] =
{RST_UEFI_PROP, RST_VMD_PROP}; unsigned long i;

      if (check_env("IMSM_TEST_AHCI_EFI") ||
check_env("IMSM_TEST_SCU_EFI")) @@ -640,10 +641,14 @@ const
struct imsm_orom *find_imsm_efi(struct sys_dev *hba)
              break;
      case SYS_DEV_VMD:

- if (!read_efi_variable(&orom, sizeof(orom),
VMD_PROP,

platform-intel.c +1 -3
-                                    VENDOR_GUID))
-                     break;
-             return NULL;
+             for (i = 0; i < ARRAY_SIZE(vmd_efivars); i++)

{
+ if (!read_efi_variable(&orom,
sizeof(orom),
+
vmd_efivars[i], VENDOR_GUID))

platform-intel.c

Quoted 8 lines
platform-intel.c +5 -0
+                             break;
+             }
+             if (i == ARRAY_SIZE(vmd_efivars))
+                     return NULL;
+             break;
      default:
              return NULL;
      }

Hi,

Quoted 2 lines

please have a look at the following mail:
https://marc.info/?l=linux-raid&m=165969352101643&w=2

Sorry for double-posting, I received rejection emails regarding HTML content. Gmail switched to HTML.

Hi, the described issue applies specifically in the SYS_DEV_SATA (SATA configuration) case, so it should not apply to SYS_DEV_VMD (VMD configuration) one.

For me, the platform output looks reasonable (I have RAID 0 active):
#> sudo mdadm --detail-platform
Platform : Intel(R) Rapid Storage Technology
Version : 19.0.7.5579
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 32
Max Volumes : 2 per array, 4 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:00/0000:00:0e.0 (VMD) NVMe under VMD : /dev/nvme0n1 (S6P1NS0T318266R) NVMe under VMD : /dev/nvme1n1 (S6P1NS0T318223V)

Quoted 10 lines

Without the patch the platform isn't even recognized. Common to both changes is the usage of the new UEFI variable 'RstVmdV', not the changes to the controller.

Regards,
Oldrich.

Quoted 3 lines

Regards,
Kinga Tanska

Hello,

I've done test to check if your patch doesn't change way of recognizing controllers. I've got two SATA controllers - first one with AHCI mode and the other one with RAID MODE enabled. Command "mdadm --detail-platform" should display info only about controller with RAID MODE:

mdadm: imsm capabilities not found for controller:
/sys/devices/pci0000:00/0000:00:17.0 (type SATA) Platform :
Intel(R) Rapid Storage Technology enterprise Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:11.5 (SATA)
Port4 : - non-disk device (TEAC DV-W28S-B) - Port0 : - no device attached -
Port1 : - no device attached -
Port2 : - no device attached -
Port3 : - no device attached -
Port5 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 24
Max Volumes : 2 per array, 24 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:17/0000:17:05.5 (VMD) I/O Controller : /sys/devices/pci0000:d7/0000:d7:05.5 (VMD) NVMe under VMD : /dev/nvme3n1 (PHLJ915000201P0FGN) NVMe under VMD : /dev/nvme2n1 (PHLJ915003201P0FGN) I/O Controller : /sys/devices/pci0000:85/0000:85:05.5 (VMD) I/O Controller : /sys/devices/pci0000:ae/0000:ae:05.5 (VMD) I/O Controller : /sys/devices/pci0000:5d/0000:5d:05.5 (VMD) NVMe under VMD : /dev/nvme0n1 (PHFT536600GT400GGN) NVMe under VMD : /dev/nvme1n1 (CVFT523100122P0KGN)

But with your patch this command returns info about all controllers which is not correct:

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:17.0 (SATA)
Port2 : /dev/sdd (WDEBMLJ2)
Port3 : /dev/sde (BTPR2300034W120LGN) Port0 : /dev/sdb (CVTS5396007S240JGN) Port1 : /dev/sdc (ZDE0XM9Z)
Port4 : - no device attached -
Port5 : - no device attached -
Port6 : - no device attached -
Port7 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 8
Max Volumes : 2 per array, 8 per controller
I/O Controller : /sys/devices/pci0000:00/0000:00:11.5 (SATA)
Port4 : - non-disk device (TEAC DV-W28S-B) - Port0 : - no device attached -
Port1 : - no device attached -
Port2 : - no device attached -
Port3 : - no device attached -
Port5 : - no device attached -

Platform : Intel(R) Rapid Storage Technology enterprise
Version : 5.3.0.1052
RAID Levels : raid0 raid1 raid10 raid5
Chunk Sizes : 4k 8k 16k 32k 64k 128k
2TB volumes : supported
2TB disks : supported
Max Disks : 24
Max Volumes : 2 per array, 24 per controller
3rd party NVMe : supported
I/O Controller : /sys/devices/pci0000:17/0000:17:05.5 (VMD) I/O Controller : /sys/devices/pci0000:d7/0000:d7:05.5 (VMD) NVMe under VMD : /dev/nvme3n1 (PHLJ915000201P0FGN) NVMe under VMD : /dev/nvme2n1 (PHLJ915003201P0FGN) I/O Controller : /sys/devices/pci0000:85/0000:85:05.5 (VMD) I/O Controller : /sys/devices/pci0000:ae/0000:ae:05.5 (VMD) I/O Controller : /sys/devices/pci0000:5d/0000:5d:05.5 (VMD) NVMe under VMD : /dev/nvme0n1 (PHFT536600GT400GGN) NVMe under VMD : /dev/nvme1n1 (CVFT523100122P0KGN)

Quoted 1 line

Please analyze it.

Hi Kinga Tanska,

Quoted 8 lines

My patch touches only the VMD controllers. I see that there are no changes to VMD display in your output. My patch is only about VMD.

I guess that your issue is identical to what you already wrote to RAID mailing list 5th of August. Correct? I cannot investigate this for you, I am neither the author of the change, nor I have RST enterprise RAID controller, sorry. Please contact the patch submitter/author.

I just checked the patch code, I cannot debug this. You are most probably affected by patch which you already identified:

Quoted 15 lines

[PATCH 19/23] mdadm: enable Intel Alderlake RSTe configuration.

It tries to exclude SATA in AHCI mode (PCI class 0x010601) from the changed EFI variable check. What is your lspci -nn output? It could help in verifying that my guess is correct.

Cheers,
Oldrich.

Quoted 6 lines

Regards,
Oldrich.

Quoted 2 lines

Regards,
Kinga Tanska

Hi Oldrich,

Quoted 7 lines

I've noticed that your patch is on top of mdadm-CI for-jes/20220728 branch. This branch contains previous patch for Alderlake RST VMD support ([PATCH 19/23] mdadm: enable Intel Alderlake RSTe configuration). This patch causes defect, which I've described. You were right saying that this patch affected me.
Please move your patch to the HEAD and then I've been able to test it again.

Hi Kinga,

Thanks, I will do it during the weekend.

Cheers,
Oldrich.

Quoted 2 lines

Thanks,
Kinga