Patch for 256 disks in 2.4

Patch for 256 disks in 2.4

[Pete Zaitcev]

For those who do not follow, John Cagle allocated 8 more SCSI
disk majors. Here's a patch to make use of them. I am not sure
if we want a 2.5 version; it's going to be all devicefs anyhow...

It really is strange how many places know major assignments.
I hope I found all of them which matter. Also, I think nobody
uses min_major/max_major, do you guys know if that's right?
More comments?

-- Pete

diff -ur -X dontdiff linux-2.4.19-rc1/drivers/char/sysrq.c linux-2.4.19-rc1-p3/drivers/char/sysrq.c
--- linux-2.4.19-rc1/drivers/char/sysrq.c	Mon Jul  1 12:58:54 2002
+++ linux-2.4.19-rc1-p3/drivers/char/sysrq.c	Sat Jul 20 16:55:25 2002
@@ -107,6 +107,8 @@
 	unsigned int major;
 	major = MAJOR(dev);
 
+	if (SCSI_DISK_MAJOR(major))
+		return 1;
 	switch (major) {
 	case IDE0_MAJOR:
 	case IDE1_MAJOR:
@@ -118,14 +120,6 @@
 	case IDE7_MAJOR:
 	case IDE8_MAJOR:
 	case IDE9_MAJOR:
-	case SCSI_DISK0_MAJOR:
-	case SCSI_DISK1_MAJOR:
-	case SCSI_DISK2_MAJOR:
-	case SCSI_DISK3_MAJOR:
-	case SCSI_DISK4_MAJOR:
-	case SCSI_DISK5_MAJOR:
-	case SCSI_DISK6_MAJOR:
-	case SCSI_DISK7_MAJOR:
 	case XT_DISK_MAJOR:
 		return 1;
 	default:
diff -ur -X dontdiff linux-2.4.19-rc1/drivers/scsi/hosts.h linux-2.4.19-rc1-p3/drivers/scsi/hosts.h
--- linux-2.4.19-rc1/drivers/scsi/hosts.h	Thu Jul 18 14:40:14 2002
+++ linux-2.4.19-rc1-p3/drivers/scsi/hosts.h	Sat Jul 20 16:55:25 2002
@@ -507,8 +507,8 @@
     struct module * module;	  /* Used for loadable modules */
     unsigned char scsi_type;
     unsigned int major;
-    unsigned int min_major;      /* Minimum major in range. */ 
-    unsigned int max_major;      /* Maximum major in range. */
+    unsigned int _min_major_dontuse;
+    unsigned int _max_major_dontuse;
     unsigned int nr_dev;	  /* Number currently attached */
     unsigned int dev_noticed;	  /* Number of devices detected. */
     unsigned int dev_max;	  /* Current size of arrays */
diff -ur -X dontdiff linux-2.4.19-rc1/drivers/scsi/scsi_lib.c linux-2.4.19-rc1-p3/drivers/scsi/scsi_lib.c
--- linux-2.4.19-rc1/drivers/scsi/scsi_lib.c	Mon Jul  1 12:59:15 2002
+++ linux-2.4.19-rc1-p3/drivers/scsi/scsi_lib.c	Sat Jul 20 16:55:25 2002
@@ -776,8 +776,9 @@
  *
  * Lock status: No locks assumed to be held, but as it happens the
  *              io_request_lock is held when this is called.
+ *              We traverse scis_devicelist, so protect that.
  *
- * Returns:     Nothing
+ * Returns:     Pointer to a Scsi_Device_Template.
  *
  * Notes:       The requests in the request queue may have originated
  *              from any block device driver.  We need to find out which
@@ -791,6 +792,9 @@
 
 	ASSERT_LOCK(&io_request_lock, 1);
 
+	if (SCSI_DISK_MAJOR(major))
+		major = SCSI_DISK0_MAJOR;
+
 	for (spnt = scsi_devicelist; spnt; spnt = spnt->next) {
 		/*
 		 * Search for a block device driver that supports this
@@ -799,22 +803,6 @@
 		if (spnt->blk && spnt->major == major) {
 			return spnt;
 		}
-		/*
-		 * I am still not entirely satisfied with this solution,
-		 * but it is good enough for now.  Disks have a number of
-		 * major numbers associated with them, the primary
-		 * 8, which we test above, and a secondary range of 7
-		 * different consecutive major numbers.   If this ever
-		 * becomes insufficient, then we could add another function
-		 * to the structure, and generalize this completely.
-		 */
-		if( spnt->min_major != 0 
-		    && spnt->max_major != 0
-		    && major >= spnt->min_major
-		    && major <= spnt->max_major )
-		{
-			return spnt;
-		}
 	}
 	return NULL;
 }
diff -ur -X dontdiff linux-2.4.19-rc1/drivers/scsi/sd.c linux-2.4.19-rc1-p3/drivers/scsi/sd.c
--- linux-2.4.19-rc1/drivers/scsi/sd.c	Mon Jul  1 12:59:15 2002
+++ linux-2.4.19-rc1-p3/drivers/scsi/sd.c	Sat Jul 20 16:55:25 2002
@@ -65,10 +65,10 @@
  *  static const char RCSid[] = "$Header:";
  */
 
-/* system major --> sd_gendisks index */
-#define SD_MAJOR_IDX(i)		(MAJOR(i) & SD_MAJOR_MASK)
+/* device number --> sd_gendisks index */
+#define SD_MAJOR_IDX(i)		( ((MAJOR(i) & 0x80) >> 4) + (MAJOR(i) & 7) )
 /* sd_gendisks index --> system major */
-#define SD_MAJOR(i) (!(i) ? SCSI_DISK0_MAJOR : SCSI_DISK1_MAJOR-1+(i))
+#define SD_MAJOR(i)		(sd_major[i])
 
 #define SD_PARTITION(dev)	((SD_MAJOR_IDX(dev) << 8) | (MINOR(dev) & 255))
 
@@ -96,6 +96,25 @@
 static int *sd_hardsizes;	/* Hardware sector size */
 static int *sd_max_sectors;
 
+static const unsigned int sd_major[N_SD_MAJORS] = {
+	SCSI_DISK0_MAJOR,	/* 0x08 */
+	SCSI_DISK1_MAJOR,	/* 0x41 */
+	SCSI_DISK2_MAJOR,	/* 0x42 */
+	SCSI_DISK3_MAJOR,	/* 0x43 */
+	SCSI_DISK4_MAJOR,	/* 0x44 */
+	SCSI_DISK5_MAJOR,	/* 0x45 */
+	SCSI_DISK6_MAJOR,	/* 0x46 */
+	SCSI_DISK7_MAJOR,	/* 0x47 */
+	SCSI_DISK10_MAJOR,	/* 0x80 */
+	SCSI_DISK11_MAJOR,	/* 0x81 */
+	SCSI_DISK12_MAJOR,	/* 0x82 */
+	SCSI_DISK13_MAJOR,	/* 0x83 */
+	SCSI_DISK14_MAJOR,	/* 0x84 */
+	SCSI_DISK15_MAJOR,	/* 0x85 */
+	SCSI_DISK16_MAJOR,	/* 0x86 */
+	SCSI_DISK17_MAJOR,	/* 0x87 */
+};
+
 static int check_scsidisk_media_change(kdev_t);
 static int fop_revalidate_scsidisk(kdev_t);
 
@@ -114,11 +133,6 @@
 	tag:"sd",
 	scsi_type:TYPE_DISK,
 	major:SCSI_DISK0_MAJOR,
-        /*
-         * Secondary range of majors that this driver handles.
-         */
-	min_major:SCSI_DISK1_MAJOR,
-	max_major:SCSI_DISK7_MAJOR,
 	blk:1,
 	detect:sd_detect,
 	init:sd_init,
diff -ur -X dontdiff linux-2.4.19-rc1/drivers/scsi/sd.h linux-2.4.19-rc1-p3/drivers/scsi/sd.h
--- linux-2.4.19-rc1/drivers/scsi/sd.h	Mon Jul  1 12:59:15 2002
+++ linux-2.4.19-rc1-p3/drivers/scsi/sd.h	Sat Jul 20 16:55:25 2002
@@ -40,9 +40,7 @@
  */
 extern kdev_t sd_find_target(void *host, int tgt);
 
-#define N_SD_MAJORS	8
-
-#define SD_MAJOR_MASK	(N_SD_MAJORS - 1)
+#define N_SD_MAJORS	16
 
 #endif
 
diff -ur -X dontdiff linux-2.4.19-rc1/fs/partitions/check.c linux-2.4.19-rc1-p3/fs/partitions/check.c
--- linux-2.4.19-rc1/fs/partitions/check.c	Mon Jul  1 12:59:29 2002
+++ linux-2.4.19-rc1-p3/fs/partitions/check.c	Sat Jul 20 16:55:25 2002
@@ -97,6 +97,10 @@
  * a pointer to that same buffer (for convenience).
  */
 
+/* The major calculation part duplicates SD_MAJOR_INDEX verbatim. */
+#define SCSI_DEVICE_NR(M,m)	\
+    (( ( (((M) & 0x80) >> 4) + ((M) & 7) ) << (8 - 4)) + ((m) >> 4))
+
 char *disk_name (struct gendisk *hd, int minor, char *buf)
 {
 	const char *maj = hd->major_name;
@@ -147,8 +151,8 @@
 			sprintf(buf, "%s%d", maj, unit);
 			return buf;
 	}
-	if (hd->major >= SCSI_DISK1_MAJOR && hd->major <= SCSI_DISK7_MAJOR) {
-		unit = unit + (hd->major - SCSI_DISK1_MAJOR + 1) * 16;
+	if (SCSI_DISK_MAJOR(hd->major)) {
+		unit = SCSI_DEVICE_NR(hd->major, minor);
 		if (unit+'a' > 'z') {
 			unit -= 26;
 			sprintf(buf, "sd%c%c", 'a' + unit / 26, 'a' + unit % 26);
diff -ur -X dontdiff linux-2.4.19-rc1/include/linux/blk.h linux-2.4.19-rc1-p3/include/linux/blk.h
--- linux-2.4.19-rc1/include/linux/blk.h	Mon Jul  1 12:59:39 2002
+++ linux-2.4.19-rc1-p3/include/linux/blk.h	Sat Jul 20 16:55:26 2002
@@ -143,7 +143,10 @@
 
 #define DEVICE_NAME "scsidisk"
 #define TIMEOUT_VALUE (2*HZ)
-#define DEVICE_NR(device) (((MAJOR(device) & SD_MAJOR_MASK) << (8 - 4)) + (MINOR(device) >> 4))
+/* The major calculation part duplicates SD_MAJOR_INDEX verbatim. */
+#define DEVICE_NR(device)	\
+    (( ( ((MAJOR(device) & 0x80) >> 4) + (MAJOR(device) & 7) ) << (8 - 4)) + \
+     (MINOR(device) >> 4))
 
 /* Kludge to use the same number for both char and block major numbers */
 #elif  (MAJOR_NR == MD_MAJOR) && defined(MD_DRIVER)
diff -ur -X dontdiff linux-2.4.19-rc1/include/linux/major.h linux-2.4.19-rc1-p3/include/linux/major.h
--- linux-2.4.19-rc1/include/linux/major.h	Mon Jul  1 12:59:40 2002
+++ linux-2.4.19-rc1-p3/include/linux/major.h	Sat Jul 20 16:55:26 2002
@@ -119,6 +119,15 @@
 
 #define ATARAID_MAJOR		114
 
+#define SCSI_DISK10_MAJOR	128
+#define SCSI_DISK11_MAJOR	129
+#define SCSI_DISK12_MAJOR	130
+#define SCSI_DISK13_MAJOR	131
+#define SCSI_DISK14_MAJOR	132
+#define SCSI_DISK15_MAJOR	133
+#define SCSI_DISK16_MAJOR	134
+#define SCSI_DISK17_MAJOR	135
+
 #define DASD_MAJOR      94	/* Official assignations from Peter */
 
 #define MDISK_MAJOR     95	/* Official assignations from Peter */
@@ -170,7 +179,8 @@
  */
 
 #define SCSI_DISK_MAJOR(M) ((M) == SCSI_DISK0_MAJOR || \
-  ((M) >= SCSI_DISK1_MAJOR && (M) <= SCSI_DISK7_MAJOR))
+  ((M) >= SCSI_DISK1_MAJOR && (M) <= SCSI_DISK7_MAJOR) || \
+  ((M) >= SCSI_DISK10_MAJOR && (M) <= SCSI_DISK17_MAJOR))
   
 #define SCSI_BLK_MAJOR(M) \
   (SCSI_DISK_MAJOR(M)	\

Re: Patch for 256 disks in 2.4

[Kurt Garloff]

[-- Attachment #1: Type: text/plain, Size: 1734 bytes --]

Hi Pete,

On Sat, Jul 20, 2002 at 07:57:29PM -0400, Pete Zaitcev wrote:
> For those who do not follow, John Cagle allocated 8 more SCSI
> disk majors.

Have those officially been assigned to SCSI disks?
So disks 128 -- 255 have majors 128 thr. 135?

> Here's a patch to make use of them. I am not sure
> if we want a 2.5 version; it's going to be all devicefs anyhow...

I've written a patch for sd that makes the allocation of majors
dynamic. The driver just takes 8 at sd_init and further majors are 
allocated when disks are attached. Which saves a lot of memory for
all the gendisk and hd_struct stuff in case you do not have a lot of 
SCSI disks connected. The patch does support up to 160 SD majors, 
though currently, it won't succeed getting more than 132 majors.

> It really is strange how many places know major assignments.

Indeed. I missed the sysrq stuff in my patch, BTW. 
Do you have any idea why we can't just sync all mounted filesystems
in do_emergency_sync()? Or at least all that are backed by a real 
device? The test for IDE and SCSI majors seems bogus to me. What
about DASD? LVM? EVMS? MD? Loop? NBD? DRBD? What's the rationale 
of restricting the sync to only IDE and SCSI? Deadlock avoidance?

> I hope I found all of them which matter. Also, I think nobody
> uses min_major/max_major, do you guys know if that's right?

It's right. Kill them! I killed them as well in my patch.

> More comments?

I'm gonna post my patches tomorrow ...

Regards,
-- 
Kurt Garloff  <garloff@suse.de>                          Eindhoven, NL
GPG key: See mail header, key servers         Linux kernel development
SuSE Linux AG, Nuernberg, DE                            SCSI, Security

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: Patch for 256 disks in 2.4

[Pete Zaitcev]

> Date: Mon, 22 Jul 2002 19:08:40 +0200
> From: Kurt Garloff <garloff@suse.de>

> > For those who do not follow, John Cagle allocated 8 more SCSI
> > disk majors.
> 
> Have those officially been assigned to SCSI disks?
> So disks 128 -- 255 have majors 128 thr. 135?

I do not understand what your problem is. Do you refuse to recognise
John as the LANANA chair or something?

My patch is done in accordance with this:
http://www.lanana.org/docs/device-list/devices.txt

> > Here's a patch to make use of them. I am not sure
> > if we want a 2.5 version; it's going to be all devicefs anyhow...
> 
> I've written a patch for sd that makes the allocation of majors
> dynamic. The driver just takes 8 at sd_init and further majors are 
> allocated when disks are attached. Which saves a lot of memory for
> all the gendisk and hd_struct stuff in case you do not have a lot of 
> SCSI disks connected. The patch does support up to 160 SD majors, 
> though currently, it won't succeed getting more than 132 majors.

That's wonderful, but we cannot ship that. There is no userland
support to create device nodes in dynamic fashion and to ensure
that they do not conflict. This is why Arjan filed for and received
additional majors. Dynamic solutions need some time to float about
the community, I think.

BTW, DASD does the same thing already. I never saw any memo or document
explaining how to use this capability properly. Perhaps SuSE people
support it. Kurt, can you tell anything about it?

> Do you have any idea why we can't just sync all mounted filesystems
> in do_emergency_sync()?
>  DASD? LVM? EVMS? MD? Loop? NBD? DRBD? What's the rationale 
> of restricting the sync to only IDE and SCSI? Deadlock avoidance?

I suspect it is a deadlock prevention thing, too. I cannot say if
it ever worked satisfactory... :)

> I'm gonna post my patches tomorrow ...

Thanks, that's interesting. Like I said, they are not likely to
get to the distro soon, but I'd love to look at them.

-- Pete

Re: Patch for 256 disks in 2.4

[Christoph Hellwig]

On Mon, Jul 22, 2002 at 04:48:56PM -0400, Pete Zaitcev wrote:
> > I've written a patch for sd that makes the allocation of majors
> > dynamic. The driver just takes 8 at sd_init and further majors are 
> > allocated when disks are attached. Which saves a lot of memory for
> > all the gendisk and hd_struct stuff in case you do not have a lot of 
> > SCSI disks connected. The patch does support up to 160 SD majors, 
> > though currently, it won't succeed getting more than 132 majors.
> 
> That's wonderful, but we cannot ship that. There is no userland
> support to create device nodes in dynamic fashion and to ensure
> that they do not conflict. This is why Arjan filed for and received
> additional majors. Dynamic solutions need some time to float about
> the community, I think.

I might be stupid, but it looks to me like we want both the new majors
and the kernel structs dynamically allocated when they get actually used..

Re: Patch for 256 disks in 2.4

[Kurt Garloff]

[-- Attachment #1: Type: text/plain, Size: 2687 bytes --]

Hi Pete,

On Mon, Jul 22, 2002 at 04:48:56PM -0400, Pete Zaitcev wrote:
> > From: Kurt Garloff <garloff@suse.de>
> 
> > > For those who do not follow, John Cagle allocated 8 more SCSI
> > > disk majors.
> > 
> > Have those officially been assigned to SCSI disks?
> > So disks 128 -- 255 have majors 128 thr. 135?
> 
> I do not understand what your problem is. Do you refuse to recognise
> John as the LANANA chair or something?

Strange. I was just asking. Why would you think I would be silly and
refuse to recognize somebody?

> My patch is done in accordance with this:
> http://www.lanana.org/docs/device-list/devices.txt

OK, I should have checked there before asking here, probably.

> > SCSI disks connected. The patch does support up to 160 SD majors, 
> > though currently, it won't succeed getting more than 132 majors.
> 
> That's wonderful, but we cannot ship that. There is no userland
> support to create device nodes in dynamic fashion and to ensure
> that they do not conflict.

There will be.

> This is why Arjan filed for and received
> additional majors. Dynamic solutions need some time to float about
> the community, I think.

I don't object to having some more static ones. Fewer users will need
userspace tools for handling the dev nodes then ;-)
And of course, I'll adapt my patch to grab the assigned ones before
the unknown ones ...

> BTW, DASD does the same thing already. I never saw any memo or document
> explaining how to use this capability properly. Perhaps SuSE people
> support it. Kurt, can you tell anything about it?

I don't know much about DASD. They allocate block majors dynamically
starting from 255 backwards as far as I know. So, dev nodes need to
be created dynamically, I guess.

> > Do you have any idea why we can't just sync all mounted filesystems
> > in do_emergency_sync()?
> >  DASD? LVM? EVMS? MD? Loop? NBD? DRBD? What's the rationale 
> > of restricting the sync to only IDE and SCSI? Deadlock avoidance?
> 
> I suspect it is a deadlock prevention thing, too. I cannot say if
> it ever worked satisfactory... :)

Well, Alt-SysRq-S does work; but it obviously misses to sync a number
of filesystems.

> > I'm gonna post my patches tomorrow ...
> 
> Thanks, that's interesting. Like I said, they are not likely to
> get to the distro soon, but I'd love to look at them.

Well, I would be astonished if you adopted before we do ;-)

Regards,
-- 
Kurt Garloff  <garloff@suse.de>                          Eindhoven, NL
GPG key: See mail header, key servers         Linux kernel development
SuSE Linux AG, Nuernberg, DE                            SCSI, Security

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: Patch for 256 disks in 2.4

[Kurt Garloff]

[-- Attachment #1: Type: text/plain, Size: 534 bytes --]

On Mon, Jul 22, 2002 at 09:57:00PM +0100, Christoph Hellwig wrote:
> I might be stupid, but it looks to me like we want both the new majors
> and the kernel structs dynamically allocated when they get actually used..

Sure. The sd dynamic major allocation uses the registered block majors first.

Regards,
-- 
Kurt Garloff  <garloff@suse.de>                          Eindhoven, NL
GPG key: See mail header, key servers         Linux kernel development
SuSE Linux AG, Nuernberg, DE                            SCSI, Security

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]