[PATCH 5/6] bootwrapper: Add support for the DINK firmware

[PATCH 5/6] bootwrapper: Add support for the DINK firmware

[Mark A. Greer]

This patch adds the firmware operations that support the DINK firmware
from Freescale.

Signed-off-by: Mark A. Greer <mgreer@mvista.com>
--

 dink.c |   56 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 56 insertions(+)
--

diff --git a/arch/powerpc/boot/dink.c b/arch/powerpc/boot/dink.c
new file mode 100644
index 0000000..aa446e2
--- /dev/null
+++ b/arch/powerpc/boot/dink.c
@@ -0,0 +1,56 @@
+/*
+ * Sandpoint specific fixups.
+ *
+ * Author: Mark A. Greer <mgreer@mvista.com>
+ *
+ * 2006 (c) MontaVista, Software, Inc.  This file is licensed under
+ * the terms of the GNU General Public License version 2.  This program
+ * is licensed "as is" without any warranty of any kind, whether express
+ * or implied.
+ */
+
+#include <stdarg.h>
+#include <stddef.h>
+#include "types.h"
+#include "string.h"
+#include "stdio.h"
+#include "page.h"
+#include "ops.h"
+
+#define	MB	(1024*1024)
+
+extern char _end[];
+
+static u32	cur_base;
+static u32	end_of_ram = 32 * MB;
+
+static void *
+dink_malloc(u32 size)
+{
+	void *area = NULL;
+	static u8 first_time = 1;
+
+	if (first_time) {
+		cur_base = _ALIGN_UP((unsigned long)_end, MB);
+		first_time = 0;
+	}
+
+	if ((cur_base + size) < end_of_ram) {
+		area = (void *)cur_base;
+		cur_base += _ALIGN_UP(size, MB);
+	}
+
+	return area;
+}
+
+static struct fw_ops dink_fw_ops;
+
+struct fw_ops *
+dink_init(void)
+{
+	dink_fw_ops.malloc = dink_malloc;
+	dink_fw_ops.free = NULL;
+	dink_fw_ops.exit = NULL;
+
+	return &dink_fw_ops;
+}

Re: [PATCH 5/6] bootwrapper: Add support for the DINK firmware

[Mark A. Greer]

On Wed, Jul 19, 2006 at 04:14:21PM -0700, Mark A. Greer wrote:
> This patch adds the firmware operations that support the DINK firmware
> from Freescale.
> 
> Signed-off-by: Mark A. Greer <mgreer@mvista.com>
> --
> 
>  dink.c |   56 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>  1 files changed, 56 insertions(+)
> --
> 
> diff --git a/arch/powerpc/boot/dink.c b/arch/powerpc/boot/dink.c
> new file mode 100644
> index 0000000..aa446e2
> --- /dev/null
> +++ b/arch/powerpc/boot/dink.c
> @@ -0,0 +1,56 @@
> +/*
> + * Sandpoint specific fixups.

Obviously an incorrect comment.  Will fix.

Mark
--

Re: [PATCH 5/6] bootwrapper: Add support for the DINK firmware

[Tom Rini]

On Wed, Jul 19, 2006 at 04:14:21PM -0700, Mark A. Greer wrote:

> This patch adds the firmware operations that support the DINK firmware
> from Freescale.

This isn't really DINK support (which would mean actually talking
back/with DINK, parsing the infos that it does pass along, and it does,
iirc, but I don't recall if it's at all ever correct / useful :)) but
just a, um, I suck at naming things (see arch/ppc/boot/simple) but
non-nonsense, no help 'firmware' ops.  Please think up a name that
doesn't suck (as a name I would come up with would).  Thanks!

-- 
Tom Rini

Re: [PATCH 5/6] bootwrapper: Add support for the DINK firmware

[Mark A. Greer]

On Wed, Aug 02, 2006 at 09:19:34AM -0700, Tom Rini wrote:
> On Wed, Jul 19, 2006 at 04:14:21PM -0700, Mark A. Greer wrote:
> 
> > This patch adds the firmware operations that support the DINK firmware
> > from Freescale.
> 
> This isn't really DINK support (which would mean actually talking
> back/with DINK,

You seem to be thinking that this is the final, end-all solution to
everything.  It isn't.  Its a step in the direction that we are trying
go (IIUC).  dink.c is about the easiest one there will be.  of.c has a
slightly more complicated fw_ops.  We still need to figure out what uboot
needs and what PIBS needs and what...

As things are added, it'll evolve.

> parsing the infos that it does pass along, and it does,
> iirc,

Not that I know of.

> but I don't recall if it's at all ever correct / useful :)) but
> just a, um, I suck at naming things (see arch/ppc/boot/simple) but
> non-nonsense, no help 'firmware' ops.  Please think up a name that
> doesn't suck (as a name I would come up with would).  Thanks!

Guess I suck too b/c I can't think of anything better.  As things
evolve, fw_ops may go away all together.  The problem is, I'm not
exactly sure where we're going to end up so this is the best that
I can do with the info that I have.

Mark

Re: [PATCH 5/6] bootwrapper: Add support for the DINK firmware

[Tom Rini]

On Wed, Aug 02, 2006 at 11:14:22AM -0700, Mark A. Greer wrote:
> On Wed, Aug 02, 2006 at 09:19:34AM -0700, Tom Rini wrote:
> > On Wed, Jul 19, 2006 at 04:14:21PM -0700, Mark A. Greer wrote:
> > 
> > > This patch adds the firmware operations that support the DINK firmware
> > > from Freescale.
> > 
> > This isn't really DINK support (which would mean actually talking
> > back/with DINK,
> 
> You seem to be thinking that this is the final, end-all solution to
> everything.  It isn't.  Its a step in the direction that we are trying
> go (IIUC).  dink.c is about the easiest one there will be.  of.c has a
> slightly more complicated fw_ops.  We still need to figure out what uboot
> needs and what PIBS needs and what...

Right.  I'm just saying that unless you're planning on making this talk
with / use DINK (I swear, the DINK manual talks about this a bit, but
I'm too lazy to look just this second), this isn't the DINK one, this is
the (will be needed and used) bare metal one.

-- 
Tom Rini

Re: [PATCH 5/6] bootwrapper: Add support for the DINK firmware

[Mark A. Greer]

On Wed, Aug 02, 2006 at 11:15:14AM -0700, Tom Rini wrote:
> On Wed, Aug 02, 2006 at 11:14:22AM -0700, Mark A. Greer wrote:
> > On Wed, Aug 02, 2006 at 09:19:34AM -0700, Tom Rini wrote:
> > > On Wed, Jul 19, 2006 at 04:14:21PM -0700, Mark A. Greer wrote:
> > > 
> > > > This patch adds the firmware operations that support the DINK firmware
> > > > from Freescale.
> > > 
> > > This isn't really DINK support (which would mean actually talking
> > > back/with DINK,
> > 
> > You seem to be thinking that this is the final, end-all solution to
> > everything.  It isn't.  Its a step in the direction that we are trying
> > go (IIUC).  dink.c is about the easiest one there will be.  of.c has a
> > slightly more complicated fw_ops.  We still need to figure out what uboot
> > needs and what PIBS needs and what...
> 
> Right.  I'm just saying that unless you're planning on making this talk
> with / use DINK (I swear, the DINK manual talks about this a bit, but
> I'm too lazy to look just this second), this isn't the DINK one, this is
> the (will be needed and used) bare metal one.

Ah, okay. Sure.

Re: [PATCH 5/6] bootwrapper: Add simple memory allocator

[Mark A. Greer]

This patch adds a simple memory allocator to the bootwrapper.

Signed-off-by: Mark A. Greer <mgreer@mvista.com>
--

 simple_alloc.c |   48 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 48 insertions(+)
--

diff --git a/arch/powerpc/boot/simple_alloc.c b/arch/powerpc/boot/simple_alloc.c
new file mode 100644
index 0000000..ccf61b2
--- /dev/null
+++ b/arch/powerpc/boot/simple_alloc.c
@@ -0,0 +1,48 @@
+/*
+ * Simple memory allocator.
+ *
+ * Author: Mark A. Greer <mgreer@mvista.com>
+ *
+ * 2006 (c) MontaVista, Software, Inc.  This file is licensed under
+ * the terms of the GNU General Public License version 2.  This program
+ * is licensed "as is" without any warranty of any kind, whether express
+ * or implied.
+ */
+
+/* Some day, put in a real memory allocator */
+
+#include <stddef.h>
+#include "types.h"
+#include "page.h"
+#include "ops.h"
+
+#define	MB	(1024*1024)
+
+extern char _end[];
+
+static u32	cur_base;
+static u32	end_of_ram = 32 * MB;
+
+static void *simple_malloc(u32 size)
+{
+	void *area = NULL;
+	static u8 first_time = 1;
+
+	if (first_time) {
+		cur_base = _ALIGN_UP((unsigned long)_end, MB);
+		first_time = 0;
+	}
+
+	if ((cur_base + size) < end_of_ram) {
+		area = (void *)cur_base;
+		cur_base += _ALIGN_UP(size, MB);
+	}
+
+	return area;
+}
+
+void simple_alloc_init(void)
+{
+	platform_ops.malloc = simple_malloc;
+	platform_ops.free = NULL;
+}