From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Albert Herranz <albert_herranz@yahoo.es>
Cc: linuxppc-dev@lists.ozlabs.org
Subject: Re: [RFC PATCH v2 1/4] wii: bootwrapper: add fixup to calc useable mem2
Date: Sat, 12 Dec 2009 09:03:46 +1100 [thread overview]
Message-ID: <1260569026.16132.367.camel@pasglop> (raw)
In-Reply-To: <1260297833-17625-2-git-send-email-albert_herranz@yahoo.es>
On Tue, 2009-12-08 at 19:43 +0100, Albert Herranz wrote:
> The top portion of MEM2 (the second 64MB memory block) in the Nintendo
> Wii video game console is used by the firmware running on the Starlet
> processor.
>
> Add code to calculate the portion of MEM2 safely useable by the
> Broadway processor. When running under the 'mini' firmware this is
> easily determined from an in-memory header. Otherwise, a safe default
> is used.
>
> Signed-off-by: Albert Herranz <albert_herranz@yahoo.es>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
> ---
> arch/powerpc/boot/wii.c | 118 +++++++++++++++++++++++++++++++++++++++++++++-
> 1 files changed, 115 insertions(+), 3 deletions(-)
>
> diff --git a/arch/powerpc/boot/wii.c b/arch/powerpc/boot/wii.c
> index d0e2625..2ebaec0 100644
> --- a/arch/powerpc/boot/wii.c
> +++ b/arch/powerpc/boot/wii.c
> @@ -22,10 +22,120 @@
>
> BSS_STACK(8192);
>
> -#define HW_REG(x) ((void *)(x))
> +#define HW_REG(x) ((void *)(x))
>
> -#define EXI_CTRL HW_REG(0x0d800070)
> -#define EXI_CTRL_ENABLE (1<<0)
> +#define EXI_CTRL HW_REG(0x0d800070)
> +#define EXI_CTRL_ENABLE (1<<0)
> +
> +#define MEM2_TOP (0x10000000 + 64*1024*1024)
> +#define FIRMWARE_DEFAULT_SIZE (12*1024*1024)
> +
> +
> +struct mipc_infohdr {
> + char magic[3];
> + u8 version;
> + u32 mem2_boundary;
> + u32 ipc_in;
> + size_t ipc_in_size;
> + u32 ipc_out;
> + size_t ipc_out_size;
> +};
> +
> +static int mipc_check_address(u32 pa)
> +{
> + /* only MEM2 addresses */
> + if (pa < 0x10000000 || pa > 0x14000000)
> + return -EINVAL;
> + return 0;
> +}
> +
> +static struct mipc_infohdr *mipc_get_infohdr(void)
> +{
> + struct mipc_infohdr **hdrp, *hdr;
> +
> + /* 'mini' header pointer is the last word of MEM2 memory */
> + hdrp = (struct mipc_infohdr **)0x13fffffc;
> + if (mipc_check_address((u32)hdrp)) {
> + printf("mini: invalid hdrp %08X\n", (u32)hdrp);
> + hdr = NULL;
> + goto out;
> + }
> +
> + hdr = *hdrp;
> + if (mipc_check_address((u32)hdr)) {
> + printf("mini: invalid hdr %08X\n", (u32)hdr);
> + hdr = NULL;
> + goto out;
> + }
> + if (memcmp(hdr->magic, "IPC", 3)) {
> + printf("mini: invalid magic\n");
> + hdr = NULL;
> + goto out;
> + }
> +
> +out:
> + return hdr;
> +}
> +
> +static int mipc_get_mem2_boundary(u32 *mem2_boundary)
> +{
> + struct mipc_infohdr *hdr;
> + int error;
> +
> + hdr = mipc_get_infohdr();
> + if (!hdr) {
> + error = -1;
> + goto out;
> + }
> +
> + if (mipc_check_address(hdr->mem2_boundary)) {
> + printf("mini: invalid mem2_boundary %08X\n",
> + hdr->mem2_boundary);
> + error = -EINVAL;
> + goto out;
> + }
> + *mem2_boundary = hdr->mem2_boundary;
> + error = 0;
> +out:
> + return error;
> +
> +}
> +
> +static void platform_fixups(void)
> +{
> + void *mem;
> + u32 reg[4];
> + u32 mem2_boundary;
> + int len;
> + int error;
> +
> + mem = finddevice("/memory");
> + if (!mem)
> + fatal("Can't find memory node\n");
> +
> + /* two ranges of (address, size) words */
> + len = getprop(mem, "reg", reg, sizeof(reg));
> + if (len != sizeof(reg)) {
> + /* nothing to do */
> + goto out;
> + }
> +
> + /* retrieve MEM2 boundary from 'mini' */
> + error = mipc_get_mem2_boundary(&mem2_boundary);
> + if (error) {
> + /* if that fails use a sane value */
> + mem2_boundary = MEM2_TOP - FIRMWARE_DEFAULT_SIZE;
> + }
> +
> + if (mem2_boundary > reg[2] && mem2_boundary < reg[2] + reg[3]) {
> + reg[3] = mem2_boundary - reg[2];
> + printf("top of MEM2 @ %08X\n", reg[2] + reg[3]);
> + setprop(mem, "reg", reg, sizeof(reg));
> + }
> +
> +out:
> + return;
> +}
>
> void platform_init(unsigned long r3, unsigned long r4, unsigned long r5)
> {
> @@ -42,5 +152,7 @@ void platform_init(unsigned long r3, unsigned long r4, unsigned long r5)
>
> if (ug_probe())
> console_ops.write = ug_console_write;
> +
> + platform_ops.fixups = platform_fixups;
> }
>
next prev parent reply other threads:[~2009-12-11 22:03 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-12-08 18:43 [RFC PATCH v2 0/4] powerpc: wii: mem2 as ram support Albert Herranz
2009-12-08 18:43 ` [RFC PATCH v2 1/4] wii: bootwrapper: add fixup to calc useable mem2 Albert Herranz
2009-12-11 22:03 ` Benjamin Herrenschmidt [this message]
2009-12-08 18:43 ` [RFC PATCH v2 2/4] wii: use both mem1 and mem2 as ram Albert Herranz
2009-12-11 22:05 ` Benjamin Herrenschmidt
2009-12-08 18:43 ` [RFC PATCH v2 3/4] powerpc: allow ioremap within reserved memory regions Albert Herranz
2009-12-11 22:13 ` Benjamin Herrenschmidt
2009-12-12 0:33 ` Albert Herranz
2009-12-12 4:24 ` Benjamin Herrenschmidt
2009-12-08 18:43 ` [RFC PATCH v2 4/4] powerpc: wii: allow ioremap within the memory hole Albert Herranz
2009-12-11 22:13 ` Benjamin Herrenschmidt
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1260569026.16132.367.camel@pasglop \
--to=benh@kernel.crashing.org \
--cc=albert_herranz@yahoo.es \
--cc=linuxppc-dev@lists.ozlabs.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.