Skip to contentSkip to conversation

aic79xx hostraid support

PATCH In review Arjan van de Ven 3 participants 7 messages Aug 21, 2004 – Aug 23, 2004

Subsystems (2)
SCSI AIC7XXX / AIC79XX SCSI driver
lore
Share & tools
lore

Arjan van de Ven

author
#0 lore raw

PATCH 1/1 aic79xx hostraid support

Hi,

Below is a patch ported to 2.6.8 by John A. Hull ([email protected]) to add support for the aic79xx hostraid family support. Please consider.

Greetings,
Arjan van de Ven

drivers/scsi/aic7xxx/aic79xx.h +2 -1
diff -purN linux-2.6.8/drivers/scsi/aic7xxx/aic79xx.h linux/drivers/scsi/aic7xxx/aic79xx.h
@@ -375,7 +375,8 @@ typedef enum {
 	AHD_RESET_POLL_ACTIVE = 0x200000,
 	AHD_UPDATE_PEND_CMDS  = 0x400000,
 	AHD_RUNNING_QOUTFIFO  = 0x800000,
-	AHD_HAD_FIRST_SEL     = 0x1000000
+	AHD_HAD_FIRST_SEL     = 0x1000000,
+ 	AHD_HOSTRAID_BOARD    = 0x2000000
 } ahd_flag;
 
 /************************* Hardware  SCB Definition ***************************/
drivers/scsi/aic7xxx/aic79xx_osm_pci.c +4 -0
diff -purN linux-2.6.8/drivers/scsi/aic7xxx/aic79xx_osm_pci.c linux/drivers/scsi/aic7xxx/aic79xx_osm_pci.c
@@ -64,6 +64,10 @@ static struct pci_device_id ahd_linux_pc
 		0x9005, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
 		PCI_CLASS_STORAGE_SCSI << 8, 0xFFFF00, 0
 	},
+	{
+		0x9005, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
+		PCI_CLASS_STORAGE_RAID << 8, 0xFFFF00, 0
+	},
 	{ 0 }
 };
drivers/scsi/aic7xxx/aic79xx_pci.c +24 -20

105 lines not on this page · raw message

author
#1 lore raw
Quoted 4 lines · Arjan van de Ven

Hi,

Below is a patch ported to 2.6.8 by John A. Hull ([email protected]) to add support for the aic79xx hostraid family support. Please consider.

since AHD_HOSTRAID_BOARD wasn't actually used... a more minimal patch:

drivers/scsi/aic7xxx/aic79xx_osm_pci.c +4 -0
diff -purN linux-2.6.8/drivers/scsi/aic7xxx/aic79xx_osm_pci.c linux/drivers/scsi/aic7xxx/aic79xx_osm_pci.c
@@ -64,6 +64,10 @@ static struct pci_device_id ahd_linux_pc
 		0x9005, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
 		PCI_CLASS_STORAGE_SCSI << 8, 0xFFFF00, 0
 	},
+	{
+		0x9005, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
+		PCI_CLASS_STORAGE_RAID << 8, 0xFFFF00, 0
+	},
 	{ 0 }
 };
drivers/scsi/aic7xxx/aic79xx_pci.c +15 -20
diff -purN linux-2.6.8/drivers/scsi/aic7xxx/aic79xx_pci.c linux/drivers/scsi/aic7xxx/aic79xx_pci.c
@@ -65,10 +65,10 @@ ahd_compose_id(u_int device, u_int vendo
 }
 
 #define ID_ALL_MASK			0xFFFFFFFFFFFFFFFFull
-#define ID_ALL_IROC_MASK		0xFFFFFF7FFFFFFFFFull
+#define ID_ALL_IROC_MASK		0xFF7FFFFFFFFFFFFFull
 #define ID_DEV_VENDOR_MASK		0xFFFFFFFF00000000ull
 #define ID_9005_GENERIC_MASK		0xFFF0FFFF00000000ull
-#define ID_9005_GENERIC_IROC_MASK	0xFFF0FF7F00000000ull
+#define ID_9005_GENERIC_IROC_MASK	0xFF70FFFF00000000ull
 
 #define ID_AIC7901			0x800F9005FFFF9005ull
 #define ID_AHA_29320A			0x8000900500609005ull

@@ -92,6 +92,8 @@ ahd_compose_id(u_int device, u_int vendo
 #define ID_AIC7902_PCI_REV_B0		0x10
 #define SUBID_HP			0x0E11
 
+#define DEVID_9005_HOSTRAID(id) ((id) & 0x80)
+
 #define DEVID_9005_TYPE(id) ((id) & 0xF)
 #define		DEVID_9005_TYPE_HBA		0x0	/* Standard Card */
 #define		DEVID_9005_TYPE_HBA_2EXT	0x1	/* 2 External Ports */

@@ -134,18 +136,18 @@ struct ahd_pci_identity ahd_pci_ident_ta
 		"Adaptec 29320ALP Ultra320 SCSI adapter",
 		ahd_aic7901_setup
 	},
-	/* aic7901A based controllers */
+	/* aic7902 based controllers */	
 	{
 		ID_AHA_29320,
 		ID_ALL_MASK,
 		"Adaptec 29320 Ultra320 SCSI adapter",
-		ahd_aic7901A_setup
+		ahd_aic7902_setup
 	},
 	{
 		ID_AHA_29320B,
 		ID_ALL_MASK,
 		"Adaptec 29320B Ultra320 SCSI adapter",
-		ahd_aic7901A_setup
+		ahd_aic7902_setup
 	},
 	{
 		ID_AHA_29320LP,

@@ -153,7 +155,6 @@ struct ahd_pci_identity ahd_pci_ident_ta
 		"Adaptec 29320LP Ultra320 SCSI adapter",
 		ahd_aic7901A_setup
 	},
-	/* aic7902 based controllers */	
 	{
 		ID_AHA_39320,
 		ID_ALL_MASK,

@@ -196,22 +197,10 @@ struct ahd_pci_identity ahd_pci_ident_ta
 		"Adaptec (HP OEM) 39320D Ultra320 SCSI adapter",
 		ahd_aic7902_setup
 	},
-	{
-		ID_AHA_29320,
-		ID_ALL_MASK,
-		"Adaptec 29320 Ultra320 SCSI adapter",
-		ahd_aic7902_setup
-	},
-	{
-		ID_AHA_29320B,
-		ID_ALL_MASK,
-		"Adaptec 29320B Ultra320 SCSI adapter",
-		ahd_aic7902_setup
-	},
 	/* Generic chip probes for devices we don't know 'exactly' */
 	{
-		ID_AIC7901 & ID_DEV_VENDOR_MASK,
-		ID_DEV_VENDOR_MASK,
+		ID_AIC7901 & ID_9005_GENERIC_MASK,
+		ID_9005_GENERIC_MASK,
 		"Adaptec AIC7901 Ultra320 SCSI adapter",
 		ahd_aic7901_setup
 	},

@@ -294,6 +283,12 @@ ahd_find_pci_device(ahd_dev_softc_t pci)
 				 subdevice,
 				 subvendor);
 
+	/*
+	 * Controllers, mask out the IROC/HostRAID bit
+	 */
+	
+	full_id &= ID_ALL_IROC_MASK;
+
 	for (i = 0; i < ahd_num_pci_devs; i++) {
 		entry = &ahd_pci_ident_table[i];
 		if (entry->full_id == (full_id & entry->id_mask)) {
#2 lore raw

Those patches are from the more recent versions of the AIC drivers, from Justin's website, verbatim but somehow reworked to compile for older versions of the drivers.

Please don't claim that you wrote it or as your own work.

I've been sent _several_ such patches privately for review. This seems to be a version of one.

Also, those patches do no allow one to disable probing for HostRAID drivers, which some users of Linux prefer.

Luben

Quoted 128 lines · Arjan van de Ven
Quoted 3 lines · Arjan van de Ven

Hi,

Below is a patch ported to 2.6.8 by John A. Hull ([email protected])

to add

Quoted 1 line

support for the aic79xx hostraid family support. Please consider.

since AHD_HOSTRAID_BOARD wasn't actually used... a more minimal patch:

diff -purN linux-2.6.8/drivers/scsi/aic7xxx/aic79xx_osm_pci.c linux/drivers/scsi/aic7xxx/aic79xx_osm_pci.c --- linux-2.6.8/drivers/scsi/aic7xxx/aic79xx_osm_pci.c 2004-08-14
07:36:33.000000000 +0200
+++ linux/drivers/scsi/aic7xxx/aic79xx_osm_pci.c 2004-08-21
15:27:18.902556250 +0200

drivers/scsi/aic7xxx/aic79xx_osm_pci.c +4 -0
@@ -64,6 +64,10 @@ static struct pci_device_id ahd_linux_pc
                0x9005, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
                PCI_CLASS_STORAGE_SCSI << 8, 0xFFFF00, 0
        },
+       {
+               0x9005, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
+               PCI_CLASS_STORAGE_RAID << 8, 0xFFFF00, 0
+       },
        { 0 }
 };

diff -purN linux-2.6.8/drivers/scsi/aic7xxx/aic79xx_pci.c linux/drivers/scsi/aic7xxx/aic79xx_pci.c
--- linux-2.6.8/drivers/scsi/aic7xxx/aic79xx_pci.c 2004-08-14
07:36:57.000000000 +0200
+++ linux/drivers/scsi/aic7xxx/aic79xx_pci.c 2004-08-21
15:27:18.903556130 +0200

drivers/scsi/aic7xxx/aic79xx_pci.c +4 -2
@@ -65,10 +65,10 @@ ahd_compose_id(u_int device, u_int vendo
 }
 
 #define ID_ALL_MASK                    0xFFFFFFFFFFFFFFFFull
-#define ID_ALL_IROC_MASK               0xFFFFFF7FFFFFFFFFull
+#define ID_ALL_IROC_MASK               0xFF7FFFFFFFFFFFFFull
 #define ID_DEV_VENDOR_MASK             0xFFFFFFFF00000000ull
 #define ID_9005_GENERIC_MASK           0xFFF0FFFF00000000ull
-#define ID_9005_GENERIC_IROC_MASK      0xFFF0FF7F00000000ull
+#define ID_9005_GENERIC_IROC_MASK      0xFF70FFFF00000000ull
 
 #define ID_AIC7901                     0x800F9005FFFF9005ull
 #define ID_AHA_29320A                  0x8000900500609005ull

@@ -92,6 +92,8 @@ ahd_compose_id(u_int device, u_int vendo
 #define ID_AIC7902_PCI_REV_B0          0x10
 #define SUBID_HP                       0x0E11
 
+#define DEVID_9005_HOSTRAID(id) ((id) & 0x80)
+
 #define DEVID_9005_TYPE(id) ((id) & 0xF)
 #define                DEVID_9005_TYPE_HBA             0x0     /*

Standard Card */
#define DEVID_9005_TYPE_HBA_2EXT 0x1 /* 2
External Ports */

drivers/scsi/aic7xxx/aic79xx_pci.c +11 -19
@@ -134,18 +136,18 @@ struct ahd_pci_identity ahd_pci_ident_ta
                "Adaptec 29320ALP Ultra320 SCSI adapter",
                ahd_aic7901_setup
        },
-       /* aic7901A based controllers */
+       /* aic7902 based controllers */
        {
                ID_AHA_29320,
                ID_ALL_MASK,
                "Adaptec 29320 Ultra320 SCSI adapter",
-               ahd_aic7901A_setup
+               ahd_aic7902_setup
        },
        {
                ID_AHA_29320B,
                ID_ALL_MASK,
                "Adaptec 29320B Ultra320 SCSI adapter",
-               ahd_aic7901A_setup
+               ahd_aic7902_setup
        },
        {
                ID_AHA_29320LP,

@@ -153,7 +155,6 @@ struct ahd_pci_identity ahd_pci_ident_ta
                "Adaptec 29320LP Ultra320 SCSI adapter",
                ahd_aic7901A_setup
        },
-       /* aic7902 based controllers */
        {
                ID_AHA_39320,
                ID_ALL_MASK,

@@ -196,22 +197,10 @@ struct ahd_pci_identity ahd_pci_ident_ta
                "Adaptec (HP OEM) 39320D Ultra320 SCSI adapter",
                ahd_aic7902_setup
        },
-       {
-               ID_AHA_29320,
-               ID_ALL_MASK,
-               "Adaptec 29320 Ultra320 SCSI adapter",
-               ahd_aic7902_setup
-       },
-       {
-               ID_AHA_29320B,
-               ID_ALL_MASK,
-               "Adaptec 29320B Ultra320 SCSI adapter",
-               ahd_aic7902_setup
-       },
        /* Generic chip probes for devices we don't know 'exactly' */
        {
-               ID_AIC7901 & ID_DEV_VENDOR_MASK,
-               ID_DEV_VENDOR_MASK,
+               ID_AIC7901 & ID_9005_GENERIC_MASK,
+               ID_9005_GENERIC_MASK,
                "Adaptec AIC7901 Ultra320 SCSI adapter",
                ahd_aic7901_setup
        },

@@ -294,6 +283,12 @@ ahd_find_pci_device(ahd_dev_softc_t pci)
                                 subdevice,
                                 subvendor);
 
+       /*
+        * Controllers, mask out the IROC/HostRAID bit
+        */
+      
+       full_id &= ID_ALL_IROC_MASK;
+
        for (i = 0; i < ahd_num_pci_devs; i++) {
                entry = &ahd_pci_ident_table[i];
                if (entry->full_id == (full_id & entry->id_mask)) {
-

To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

author
#3 lore raw
Quoted 5 lines · Luben Tuikov

Those patches are from the more recent versions of the AIC drivers, from Justin's website, verbatim but somehow reworked to compile for older versions of the drivers.

Please don't claim that you wrote it or as your own work.

I do not and never have claimed it as my own work.

Do you agree that this patch is a useful addition for now, until you start sending similar small improvements to linux-scsi for the drivers?

(I for sure think it would be a bad thing if aic7xxx entirely halted pending that)

#4 lore raw
Quoted 6 lines · Arjan van de Ven
Quoted 4 lines · Luben Tuikov

Those patches are from the more recent versions of the AIC drivers, from Justin's website, verbatim but somehow reworked to compile for older versions of the drivers.

Quoted 7 lines
Quoted 1 line

Please don't claim that you wrote it or as your own work.

I do not and never have claimed it as my own work.

Do you agree that this patch is a useful addition for now, until you start sending similar small improvements to linux-scsi for the drivers?

Yes, sure, if it works for people _and_ they realize that it is:
a) temporary,
b) taken from the later version of the driver on Justin's website, and
c) it will be scrapped completely since the later versions which I'll be sending *contain and absolve* this or any other similar patches.

Quoted 2 lines

(I for sure think it would be a bad thing if aic7xxx entirely halted pending that)

Yes, I agree.

Luben

author
#5 lore raw
Quoted 12 lines · Luben Tuikov
Quoted 11 lines · Arjan van de Ven
Quoted 5 lines · Luben Tuikov

Those patches are from the more recent versions of the AIC drivers, from Justin's website, verbatim but somehow reworked to compile for older versions of the drivers.

Please don't claim that you wrote it or as your own work.

I do not and never have claimed it as my own work.

Quoted 8 lines
Quoted 2 lines

Do you agree that this patch is a useful addition for now, until you start sending similar small improvements to linux-scsi for the drivers?

Yes, sure, if it works for people _and_ they realize that it is:
a) temporary,
b) taken from the later version of the driver on Justin's website, and
c) it will be scrapped completely since the later versions which I'll be sending *contain and absolve* this or any other similar patches.

sure since I assume your later drop will also just support the hardware I dont see a user regression in this... I mean, when you replace the code the hardware keeps working, end user remains happy. Nobody is married to this patch; I just care about getting this new hardware working without having to go to a whopping gigantic patch immediately.

So it seems we are basically in agreement ;)

Thanks.

#6 lore raw
Quoted 1 line · Arjan van de Ven

So it seems we are basically in agreement ;)

OK, I'll put it in, then.

James