Linux 2.5.64-ac1

Linux 2.5.64-ac1

[Alan Cox]

Linux 2.5.64-ac1
	Merge Linus 2.5.63
	Merge Linus 2.5.64
	Revert broken watchdog changes
o	Revert wrong -ac change to keyboard.c		(me)
o	Fix cpufreq compile				(Bob Miller)

Linux 2.5.62-ac1
	Merge Linus 2.5.62
o	UNEXPECTED_IO_APIC can be static		(Pavel Machek)
o	Update IPMI driver to version 18		(Corey Minyard)
o	Tons of spelling fixes				(Steven Cole)
o	FBdev updates					(James Simmons)
o	PC-9800 update					(Osamu Tomita)
o	Remove dead scripts				(Brian Gerst)
o	v850 updates					(Miles Bader)
o	Update 3c523 to new MCA api (untested)		(James Bottomley)
o	Toshiba keyboard workaround			(Pavel Machek)
o	Fix mremap file name in comments		(Paul Larson)
o	Firestream typo fixes				(Maciej Soltysiak)
o	Backport trident reset fix from 2.4		(Muli Ben-Yehuda)
o	Morse code panics are back!			(Tomas Szepe)
o	Fix aicasm build				(Bob Tracy)
o	Fixes for 700/710 drivers			(Rolf Eike Beer)
o	Spelling fixes					(Rolf Eike Beer)
o	Optimise CRC32					(Joakim Tjernlund)
o	Next batch of v850 updates			(Miles Bader)
o	Takayoshi Kochi has moved email			(Takayoshi Kochi)
o	SunRPC race fix					(Trond Myklebust)
o	Refix addr/port naming confusion in IDE iops	(me)
o	Forward port VIA APIC handling quirks		(me)
o	Forward port ALi magick quirk flag handler	(me)
	| Needs bt848 etc to acquire the fix too
o	Forward port IDE bases fix			(me)
o	Forward port pci irq search for legacy IDE	(me)

Linux 2.5.61-ac1
	Merge Linus 2.5.61
o	Fix aic7xxx makefile				(Sam Ravnborg)
o	Fix ieee1394 build on Alpha			(Ben Collins)
o	Fix isdn_net build with X.25			(Adriank Bunk)
o	Typo fix					(Steven Bosscher)
o	A pile of other typo fixes			(Steven Cole)
o	C99 initializers				(Art Haas)
o	dasd typo fix					(Maciej Soltysiak)
o	Remove an unused variable in sunrpc		(Robert Love)
o	Remove duplicate different BSD partition names	(Andries Brouwer)
o	PPC plural fix					(Steven Cole)
o	EISA driver class patches			(Marc Zyngier)
o	VIA Rhine updates				(Roger Luethi)
o	Further ppa scsi fix				(John Kim)
o	Kill unused __beep				(Hugh Dickins)
o	Merge visws support 				(Andrey Panin)
	| Some collisions with pc9800 but should be ok
o	Limits for upward growing stacks		(Matthew Wilcox)
o	ucLinux updates					(Greg Ungerer)
o	68328 frame buffer updates			(Greg Ungerer)
o	Merge ucLinux H8300 support			(Yoshinori Sato)
o	Fix aironet compile				(Ookhoi)
o	Fix DMA mask on OSS trident driver		(Ivan Kokshaysky)
o	Kill some old 2.4 glue code in DRM		(John Kim)
o	Fix compile of old "hd.c" driver		(Paul Gortmaker)
o	Add experimental BOCHS virtualisation		(Kevin Lawton)
o	Clean up intermezzo driver			(Adrian Bunk)
o	Clean up rio use of compatmac			(Adrian Bunk)
o	Remove 2.0 ifdefs from ipchains code		(Adrian Bunk)
o	Remove old junk from efs 			(Adrian Bunk)
o	Remove old 2.0/2.2 junk from media/video	(Adrian Bunk)
o	Remove unused variable in ali-ircc		(Adrian Bunk)
o	Remove 2.0 ifdefs from network drivers		(Adrian Bunk)
o	Clean up uglies in inia100			(Adrian Bunk)
o	Clean up uglies in i91u scsi 			(Adrian Bunk)
o	Clean up wan drivers 2.0/2.2 code		(Adrian Bunk)
o	Restore ontrack remap support			(Jim Houston)
	| I'd really like to see this get turned into device mapper..
o	Forward port emu10k1 driver to 2.5		(Rui Souza)
o	Fix boot on EPOX 4BEA-R and friends		(Alexandar Achenbach)
o	Switch alpha cia code to static inline		(Matt Reppert)
o	Fix pcmcia scsi compile breakages		(Mike Anderson)
o	EHCI workarounds				(David Brownell)

Linux 2.5.60-ac1 (not published)
	Includes Linus BK snapshot
	Merge relevant pieces from old -ac		(me)
	| Dropped visws and stuff thats been redone
	| also dropped out IRQ stacks (port is tricky!)
o	Fix build of cciss driver			(me)
o	Fix build of 3036 tv tuner			(me)
o	Remove i2o_lan					(me)
o	Fix i2o_scsi					(Randy Dunlap)
o	Fix iph5526 scsi changes (not fixed DMA)	(me)
o	Make starfire compile				(me)
o	Make mca-legacy warn if used			(me)
o	Make sim710 build with EISA			(me)
o	Make ultrastor compile				(me)
o	Make aha152x/aha154x build			(Randy Dunlap)
o	Fix aha154x/mca bits				(me)
o	Fix fd_mcs build				(me)
o	Fix NCR53c406a.c				(me)
o	Fix sym53c416.c					(me)
o	Fix ibmmca compile				(me)
o	Fix ppa compile					(me)
o	Fix NCR539x compile				(John Kim)
o	Fix mca_53c9x compile				(me)

[patch] 2.5.64-ac1: fix apic.c compile

[Adrian Bunk]

The -ac patch adds a second definition of enable_NMI_through_LVT0 to 
apic.c resulting in the following compile error:

<--  snip  -->

...
  gcc -Wp,-MD,arch/i386/kernel/.apic.o.d -D__KERNEL__ -Iinclude -Wall 
-Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common 
-pipe -mpreferred-stack-boundary=2 -march=k6 
-Iinclude/asm-i386/mach-default -nostdinc -iwithprefix include    
-DKBUILD_BASENAME=apic -DKBUILD_MODNAME=apic -c -o 
arch/i386/kernel/apic.o arch/i386/kernel/apic.c
arch/i386/kernel/apic.c:72: redefinition of `enable_NMI_through_LVT0'
arch/i386/kernel/apic.c:60: `enable_NMI_through_LVT0' previously defined here
{standard input}: Assembler messages:
{standard input}:126: Error: symbol `enable_NMI_through_LVT0' is already defined
make[1]: *** [arch/i386/kernel/apic.o] Error 1

<--  snip  -->


Please remove the following part of the -ac patch:


--- linux-2.5.64/arch/i386/kernel/apic.c	2003-03-06 17:04:22.000000000 +0000
+++ linux-2.5.64-ac1/arch/i386/kernel/apic.c	2003-03-06 17:08:59.000000000 +0000
@@ -66,6 +68,18 @@
 	apic_write_around(APIC_LVT0, v);
 }
 
+void enable_NMI_through_LVT0 (void * dummy)
+{
+	unsigned int v, ver;
+
+	ver = apic_read(APIC_LVR);
+	ver = GET_APIC_VERSION(ver);
+	v = APIC_DM_NMI;			/* unmask and set to NMI */
+	if (!APIC_INTEGRATED(ver))		/* 82489DX */
+		v |= APIC_LVT_LEVEL_TRIGGER;
+	apic_write_around(APIC_LVT0, v);
+}
+
 int get_maxlvt(void)
 {
 	unsigned int v, ver, maxlvt;


cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: Linux 2.5.64-ac1

[Steven Cole]

On Thu, 2003-03-06 at 12:15, Alan Cox wrote:
> 
> Linux 2.5.64-ac1

I had problems booting 2.5.62-ac1
http://marc.theaimsgroup.com/?l=linux-kernel&m=104578421426861&w=2
and Alan responded

On Thu, 2003-02-20 at 16:57, Alan Cox wrote:
> > ide_xlate_1024+0xf5
> > read_dev_sector+0x69
> > handle_ide_mess+0x179
>
> Ok I broke it with the change to the partiton stuff I put back. If you drop
> that partition tweak out it ought to boot.

Same problem with 2.5.64-ac1.

I backed out the same partitions stuff as before, and 2.5.64-ac1 boots
fine.  This is the resulting diff.

P.S.  This isn't a spelling fix this time.

Steven

--- linux-2.5.64-ac1-borken/fs/partitions/msdos.c	Thu Mar  6 13:05:16 2003
+++ linux-2.5.64-ac1/fs/partitions/msdos.c	Tue Mar  4 20:29:24 2003
@@ -385,87 +385,6 @@
 	{SOLARIS_X86_PARTITION, parse_solaris_x86},
 	{0, NULL},
 };
-/*
- * Look for various forms of IDE disk geometry translation
- */
-static int handle_ide_mess(struct block_device *bdev)
-{
-#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_IDE_MODULE)
-	Sector sect;
-	unsigned char *data;
-	kdev_t dev = to_kdev_t(bdev->bd_dev);
-	unsigned int sig;
-	int heads = 0;
-	struct partition *p;
-	int i;
-#ifdef CONFIG_BLK_DEV_IDE_MODULE
-	if (!ide_xlate_1024)
-		return 1;
-#endif
-	/*
-	 * The i386 partition handling programs very often
-	 * make partitions end on cylinder boundaries.
-	 * There is no need to do so, and Linux fdisk doesn't always
-	 * do this, and Windows NT on Alpha doesn't do this either,
-	 * but still, this helps to guess #heads.
-	 */
-	data = read_dev_sector(bdev, 0, &sect);
-	if (!data)
-		return -1;
-	if (!msdos_magic_present(data + 510)) {
-		put_dev_sector(sect);
-		return 0;
-	}
-	sig = le16_to_cpu(*(unsigned short *)(data + 2));
-	p = (struct partition *) (data + 0x1be);
-	for (i = 0; i < 4; i++) {
-		struct partition *q = &p[i];
-		if (NR_SECTS(q)) {
-			if ((q->sector & 63) == 1 &&
-			    (q->end_sector & 63) == 63)
-				heads = q->end_head + 1;
-			break;
-		}
-	}
-	if (SYS_IND(p) == EZD_PARTITION) {
-		/*
-		 * Accesses to sector 0 must go to sector 1 instead.
-		 */
-		if (ide_xlate_1024(bdev, -1, heads, " [EZD]"))
-			goto reread;
-	} else if (SYS_IND(p) == DM6_PARTITION) {
-
-		/*
-		 * Everything on the disk is offset by 63 sectors,
-		 * including a "new" MBR with its own partition table.
-		 */
-		if (ide_xlate_1024(bdev, 1, heads, " [DM6:DDO]"))
-			goto reread;
-	} else if (sig <= 0x1ae &&
-		   data[sig] == 0xAA && data[sig+1] == 0x55 &&
-		   (data[sig+2] & 1)) {
-		/* DM6 signature in MBR, courtesy of OnTrack */
-		(void) ide_xlate_1024 (bdev, 0, heads, " [DM6:MBR]");
-	} else if (SYS_IND(p) == DM6_AUX1PARTITION ||
-		   SYS_IND(p) == DM6_AUX3PARTITION) {
-		/*
-		 * DM6 on other than the first (boot) drive
-		 */
-		(void) ide_xlate_1024(bdev, 0, heads, " [DM6:AUX]");
-	} else {
-		(void) ide_xlate_1024(bdev, 2, heads, " [PTBL]");
-	}
-	put_dev_sector(sect);
-	return 1;
-
-reread:
-	put_dev_sector(sect);
-	/* Flush the cache */
-	invalidate_bdev(bdev, 1);
-	truncate_inode_pages(bdev->bd_inode->i_mapping, 0);
-#endif /* defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_IDE_MODULE) */
-	return 1;
-}
  
 int msdos_partition(struct parsed_partitions *state, struct block_device *bdev)
 {
@@ -474,11 +393,7 @@
 	unsigned char *data;
 	struct partition *p;
 	int slot;
-	int err;
 
-	err = handle_ide_mess(bdev);
-	if (err <= 0)
-		return err;
 	data = read_dev_sector(bdev, 0, &sect);
 	if (!data)
 		return -1;
@@ -524,21 +439,6 @@
 			state->parts[slot].flags = 1;
 	}
 
-	/*
-	 *  Check for old-style Disk Manager partition table
-	 */
-	if (msdos_magic_present(data + 0xfc)) {
-		p = (struct partition *) (0x1be + data);
-		for (slot = 4 ; slot < 16 ; slot++, state->next++) {
-			p--;
-			if (state->next == state->limit)
-				break;
-			if (!(START_SECT(p) && NR_SECTS(p)))
-				continue;
-			put_partition(state, state->next,
-						START_SECT(p), NR_SECTS(p));
-		}
-	}
 	printk("\n");
 
 	/* second pass - output for each on a separate line */
@@ -556,7 +456,7 @@
 		if (!subtypes[n].parse)
 			continue;
 		subtypes[n].parse(state, bdev, START_SECT(p)*sector_size,
-						NR_SECTS(p)*sector_size, n);
+						NR_SECTS(p)*sector_size, slot);
 	}
 	put_dev_sector(sect);
 	return 1;

Re: Linux 2.5.64-ac1

[Alan Cox]

On Thu, 2003-03-06 at 21:24, Steven Cole wrote:
> I backed out the same partitions stuff as before, and 2.5.64-ac1 boots
> fine.  This is the resulting diff.

Backing it out isnt an option in the end, it has to get fixed 8(

Re: Linux 2.5.64-ac1

[Andries Brouwer]

On Thu, Mar 06, 2003 at 10:51:16PM +0000, Alan Cox wrote:
> On Thu, 2003-03-06 at 21:24, Steven Cole wrote:
> > I backed out the same partitions stuff as before, and 2.5.64-ac1 boots
> > fine.  This is the resulting diff.
> 
> Backing it out isnt an option in the end, it has to get fixed 8(

Usually I try to follow partition and geometry stuff, but this
is a discussion I missed.  What is wrong?

Andries

Re: Linux 2.5.64-ac1

[Steven Cole]

On Thu, 2003-03-06 at 15:58, Andries Brouwer wrote:
> On Thu, Mar 06, 2003 at 10:51:16PM +0000, Alan Cox wrote:
> > On Thu, 2003-03-06 at 21:24, Steven Cole wrote:
> > > I backed out the same partitions stuff as before, and 2.5.64-ac1 boots
> > > fine.  This is the resulting diff.
> > 
> > Backing it out isnt an option in the end, it has to get fixed 8(
> 
> Usually I try to follow partition and geometry stuff, but this
> is a discussion I missed.  What is wrong?
> 
> Andries

Here are the symptoms if that will help.  
Copied by hand when it died on boot.

Call Trace:

ide_xlate_1024+0xf5
read_dev_sector+0x69
handle_ide_mess+0x179
msdos_partition+0x3c
call_console_drivers+0xeb
printk+0x17d
check_partition+0xac
register_disk+0xd0
blk_register_region+0x24
add_disk+0x35
exact_match+0x0
exact_lock+0x0
sd_attach+0x291
scsi_register_device+0x94
init+0x62
init+0x0
kernel_thread_helper+0x5

Steven

Re: Linux 2.5.64-ac1

[Alan Cox]

On Thu, 2003-03-06 at 22:58, Andries Brouwer wrote:
> On Thu, Mar 06, 2003 at 10:51:16PM +0000, Alan Cox wrote:
> > On Thu, 2003-03-06 at 21:24, Steven Cole wrote:
> > > I backed out the same partitions stuff as before, and 2.5.64-ac1 boots
> > > fine.  This is the resulting diff.
> > 
> > Backing it out isnt an option in the end, it has to get fixed 8(
> 
> Usually I try to follow partition and geometry stuff, but this
> is a discussion I missed.  What is wrong?

I've got a bug somewhere in the ide_xlate stuff I restored because some
people do need it. The bug is in my code not in the partition stuff
though.