mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Kevin O'Connor" <kevin@koconnor.net>
To: Yinghai Lu <yinghai@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>, "Rafael J. Wysocki" <rjw@sisk.pl>,
	Stefan Reinauer <stepan@coresystems.de>,
	linux-kernel@vger.kernel.org, coreboot@coreboot.org,
	Jaswinder Singh Rajput <jaswinder@infradead.org>
Subject: Re: MPTable can not be high-memory on Linux
Date: Sat, 28 Feb 2009 18:41:28 -0500	[thread overview]
Message-ID: <20090228234127.GA29964@morn.localdomain> (raw)
In-Reply-To: <49A23EE0.8000102@kernel.org>

On Sun, Feb 22, 2009 at 10:14:56PM -0800, Yinghai Lu wrote:
> please check
> 
> [PATCH] x86: check physptr with max_low_pfn on 32bit

Thanks for looking at this issue.

Unfortunately, the kernel still does not work with the applied patch.
Neither 32bit nor 64bit kernels boot.  (Note, I applied the patch to
Linus' git, and I had to replace mpf->physptr with mpf->mpf_physptr.)

The error messages from before and after patching appear to be the
same.


In order to aid in testing, I modified SeaBIOS to force the mptable
into high memory even in qemu mode.  A modified bios.bin file can be
found at:

http://linuxtogo.org/~kevin/SeaBIOS/bios.bin-high-mptable

To use it with qemu, the steps are roughly:

mkdir testbios
cp /usr/share/qemu/*.bin testbios/
cp bios.bin-high-mptable testbios/bios.bin
qemu -L testbios/ -hda mytestlinux.img -serial file:mylog


The modified bios.bin can also be built manually by running:

git clone git://linuxtogo.org/home/kevin/seabios.git
cd seabios
patch -p1 < patch-from-below
make
cp out/bios.bin ../bios.bin-high-mptable

-Kevin


diff --git a/src/config.h b/src/config.h
index 56e5302..1ecc86f 100644
--- a/src/config.h
+++ b/src/config.h
@@ -21,7 +21,7 @@
 #define CONFIG_DEBUG_LEVEL 1
 
 // Send debugging information to serial port
-#define CONFIG_DEBUG_SERIAL 0
+#define CONFIG_DEBUG_SERIAL 1
 
 // Support for int13 floppy drive access
 #define CONFIG_FLOPPY_SUPPORT 1
@@ -77,7 +77,7 @@
 // Support finding a UUID (for smbios) via "magic" outl sequence.
 #define CONFIG_UUID_BACKDOOR 1
 // Support generation of ACPI tables (for emulators)
-#define CONFIG_ACPI 1
+#define CONFIG_ACPI 0
 // Support bios callbacks specific to via vgabios.
 #define CONFIG_VGAHOOKS 0
 // Support S3 resume handler.
diff --git a/src/mptable.c b/src/mptable.c
index 9e030fe..79cc7ba 100644
--- a/src/mptable.c
+++ b/src/mptable.c
@@ -19,7 +19,7 @@ mptable_init(void)
     dprintf(3, "init MPTable\n");
 
     int smp_cpus = smp_probe();
-    if (smp_cpus <= 1)
+    if (0 && smp_cpus <= 1)
         // Building an mptable on uniprocessor machines confuses some OSes.
         return;
 
@@ -39,7 +39,9 @@ mptable_init(void)
     /* floating pointer structure */
     struct mptable_floating_s *floating = (void*)start;
     memset(floating, 0, sizeof(*floating));
-    struct mptable_config_s *config = (void*)&floating[1];
+//    struct mptable_config_s *config = (void*)&floating[1];
+    struct mptable_config_s *config = (void*)(RamSize - 64*1024);
+    add_e820((u32)config, 64*1024, E820_RESERVED);
     floating->signature = MPTABLE_SIGNATURE;
     floating->physaddr = (u32)config;
     floating->length = 1;

  parent reply	other threads:[~2009-02-28 23:41 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <499DB40C.1060205@coresystems.de>
     [not found] ` <499DB480.1000909@coresystems.de>
     [not found]   ` <499DB6D8.5040706@coresystems.de>
     [not found]     ` <499DBDA0.2010606@coresystems.de>
     [not found]       ` <499DCE79.8020809@coresystems.de>
2009-02-22 15:25         ` Kevin O'Connor
2009-02-22 17:06           ` Rafael J. Wysocki
2009-02-22 21:33             ` Yinghai Lu
2009-02-22 22:32               ` Kevin O'Connor
2009-02-22 22:58                 ` Ingo Molnar
2009-02-23  6:14                   ` Yinghai Lu
2009-02-23  6:42                     ` Ingo Molnar
2009-02-28 23:41                     ` Kevin O'Connor [this message]
2009-03-01  3:10                       ` Yinghai Lu
2009-03-01 18:04                         ` Kevin O'Connor
2009-03-02  3:23                           ` [PATCH] x86: ioremap mptable -v2 Yinghai Lu
2009-03-02 10:18                             ` Ingo Molnar
2009-03-02 10:19                               ` Ingo Molnar
2009-03-02 20:07                               ` Yinghai Lu
2009-03-02 20:29                                 ` Ingo Molnar
2009-03-02 20:46                                   ` Yinghai Lu
2009-03-02 20:57                                     ` Ingo Molnar
2009-02-22 17:53           ` MPTable can not be high-memory on Linux Andi Kleen

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=20090228234127.GA29964@morn.localdomain \
    --to=kevin@koconnor.net \
    --cc=coreboot@coreboot.org \
    --cc=jaswinder@infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=rjw@sisk.pl \
    --cc=stepan@coresystems.de \
    --cc=yinghai@kernel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®