PATCH: Momentum Computer Ocelot: Making the latest CVS tree build

PATCH: Momentum Computer Ocelot: Making the latest CVS tree build

[Matthew Dharm]

[-- Attachment #1: Type: text/plain, Size: 683 bytes --]

Attached is a patch against the latest CVS tree (as of about 3 hours
ago).  It makes the Momentum Computer Ocelot board build again --
apparently, some of the PCI code has been changed.

It seems likely to me that this is not the best way to do this
patch... if it's unacceptable to the powers that be (Ralf?), could
someone point out to me the new convention for board-specific PCI
initialization?

Matt Dharm

--
Matthew D. Dharm                            Senior Software Designer
Momentum Computer Inc.                      1815 Aston Ave.  Suite 107
(760) 431-8663 X-115                        Carlsbad, CA 92008-7310
Momentum Works For You                      www.momenco.com

[-- Attachment #2: mydiff --]
[-- Type: application/octet-stream, Size: 1383 bytes --]

--- arch/mips/gt64120/common/irq.c	2001/03/09 20:33:46	1.2
+++ arch/mips/gt64120/common/irq.c	2001/05/14 23:25:01
@@ -61,6 +61,7 @@
 #define DBG(x...)
 #endif
 
+void (*irq_setup)(void);
 
 /*
  * Generic no controller code
--- arch/mips/gt64120/momenco_ocelot/setup.c	2001/02/05 01:33:01	1.1
+++ arch/mips/gt64120/momenco_ocelot/setup.c	2001/05/14 23:25:01
@@ -77,7 +77,7 @@
 {
 	unsigned int i, j;
 
-	irq_setup = momenco_ocelot_irq_setup;
+	irq_setup = momenco_ocelot_irq_setup; 
 	board_time_init = gt64120_time_init;
 
 	mips_io_port_base = KSEG1;
--- fs/ext2/inode.c	2001/04/05 04:58:29	1.34
+++ fs/ext2/inode.c	2001/05/14 23:25:11
@@ -568,7 +568,7 @@
 
 changed:
 	while (partial > chain) {
-		bforget(partial->bh);
+		brelse(partial->bh);
 		partial--;
 	}
 	goto reread;
@@ -799,8 +799,8 @@
 				/* Writer: ->i_blocks */
 				inode->i_blocks -= blocks * count;
 				/* Writer: end */
-				ext2_free_blocks (inode, block_to_free, count);
 				mark_inode_dirty(inode);
+				ext2_free_blocks (inode, block_to_free, count);
 			free_this:
 				block_to_free = nr;
 				count = 1;
@@ -811,8 +811,8 @@
 		/* Writer: ->i_blocks */
 		inode->i_blocks -= blocks * count;
 		/* Writer: end */
-		ext2_free_blocks (inode, block_to_free, count);
 		mark_inode_dirty(inode);
+		ext2_free_blocks (inode, block_to_free, count);
 	}
 }
 

Re: PATCH: Momentum Computer Ocelot: Making the latest CVS tree build

[Ralf Baechle]

On Mon, May 14, 2001 at 04:35:43PM -0700, Matthew Dharm wrote:

> Attached is a patch against the latest CVS tree (as of about 3 hours
> ago).  It makes the Momentum Computer Ocelot board build again --
> apparently, some of the PCI code has been changed.

Except your patch doesn't even touch the PCI but interrupt code.

> It seems likely to me that this is not the best way to do this
> patch... if it's unacceptable to the powers that be (Ralf?), could
> someone point out to me the new convention for board-specific PCI
> initialization?

Applied, but without the two extra junk segment in the patch.

  Ralf