qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
To: Eduardo Habkost <ehabkost@redhat.com>,
	"Michael S. Tsirkin" <mst@redhat.com>
Cc: Ed Swierk <eswierk@skyportsystems.com>,
	qemu-devel@nongnu.org, Richard Smith <smithbone@gmail.com>
Subject: Re: [Qemu-devel] [PATCH for-2.5] piix: Document coreboot-specific RAM size config register
Date: Tue, 25 Aug 2015 12:06:57 +0300	[thread overview]
Message-ID: <55DC3031.7090204@gmail.com> (raw)
In-Reply-To: <1438974931-21128-1-git-send-email-ehabkost@redhat.com>

On 08/07/2015 10:15 PM, Eduardo Habkost wrote:
> The existing i440fx initialization code sets a PCI config register that
> isn't documented anywhere in the Intel 440FX datasheet. Register 0x57 is
> DRAMC (DRAM Control) and has nothing to do with the RAM size.
>
> This was implemented in commit ec5f92ce6ac8ec09056be77e03c941be188648fa
> because old coreboot code tried to read registers 0x5a-0x5f,0x56,0x57 to
> get the RAM size from QEMU, but I couldn't find out why coreboot did
> that. I assume it was a mistake, and the original code was supposed to
> be reading the DRB[0-7] registers (offsets 0x60-0x67).
>
> Document that coreboot-specific register offset in a macro and a
> comment, for future reference.
>
> Cc: Ed Swierk <eswierk@skyportsystems.com>
> Cc: Richard Smith <smithbone@gmail.com>
> Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
> ---
> References to coreboot commits:
> * Original commit adding code reading register offsets
>    0x5a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f, 0x56, 0x57 to Intel 440bx code in
>    coreboot:
>    cb8eab482ff09ec256456312ef2d6e7710123551
> * Commit adding the same register offsets to QEMU-specific
>    coreboot code:
>    9cf642bad3fdd2205ffdd83a3222a39855b1ceff
> * coreboot commit replacing the weird register offsets with
>    code that actually reads the DRB7 register, in the I440BX code:
>    1a9c892d58c746aef0cb530481c214e63a6a6871
> * coreboot commit replacing the weird register offets with
>    code reading the CMOS in QEMU-specific code:
>    7339f36961917814ed12d5a4e6f1fe19418b8aca
> ---
>   hw/pci-host/piix.c | 7 ++++++-
>   1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/hw/pci-host/piix.c b/hw/pci-host/piix.c
> index ad55f99..1cb25f3 100644
> --- a/hw/pci-host/piix.c
> +++ b/hw/pci-host/piix.c
> @@ -117,6 +117,11 @@ struct PCII440FXState {
>   #define I440FX_PAM_SIZE 7
>   #define I440FX_SMRAM    0x72
>
> +/* Older coreboot versions (4.0 and older) read a config register that doesn't
> + * exist in real hardware, to get the RAM size from QEMU.
> + */
> +#define I440FX_COREBOOT_RAM_SIZE 0x57
> +
>   static void piix3_set_irq(void *opaque, int pirq, int level);
>   static PCIINTxRoute piix3_route_intx_pin_to_irq(void *opaque, int pci_intx);
>   static void piix3_write_config_xen(PCIDevice *dev,
> @@ -394,7 +399,7 @@ PCIBus *i440fx_init(PCII440FXState **pi440fx_state,
>       if (ram_size > 255) {
>           ram_size = 255;
>       }
> -    d->config[0x57] = ram_size;
> +    d->config[I440FX_COREBOOT_RAM_SIZE] = ram_size;
Thanks! Another magic number has now an actual meaning.

Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>


Thanks,
Marcel

>
>       i440fx_update_memory_mappings(f);
>
>

      parent reply	other threads:[~2015-08-25  9:07 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-08-07 19:15 [Qemu-devel] [PATCH for-2.5] piix: Document coreboot-specific RAM size config register Eduardo Habkost
2015-08-10  1:48 ` Ed Swierk
2015-08-13 15:30   ` Richard Smith
2015-08-17 18:58     ` Eduardo Habkost
2015-08-25  9:52       ` Thomas Lamprecht
2015-08-25  9:06 ` Marcel Apfelbaum [this message]

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=55DC3031.7090204@gmail.com \
    --to=marcel.apfelbaum@gmail.com \
    --cc=ehabkost@redhat.com \
    --cc=eswierk@skyportsystems.com \
    --cc=marcel@redhat.com \
    --cc=mst@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=smithbone@gmail.com \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).