[Qemu-devel] [PATCH] boot a linux kernel from non-ide device

[Qemu-devel] [PATCH] boot a linux kernel from non-ide device

[Glauber de Oliveira Costa]

[-- Attachment #1: Type: text/plain, Size: 461 bytes --]

Since it's now possible to use the -drive option, the test for something
in the index 0 of the IDE bus is too restrictive.

A better idea, IMHO, is to check if the user specified any bootable device,
and only if not, fallback to the default, compatible behaviour of checking
hda regardless of the presence of a boot=on arg.

-- 
Glauber de Oliveira Costa.
"Free as in Freedom"
http://glommer.net

"The less confident you are, the more serious you have to act."

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: qemu.patch --]
[-- Type: text/x-patch; name=qemu.patch, Size: 2048 bytes --]

commit 65b8fcf3bc5b754b720a04fa7efed6fdcd472a6a
Author: Glauber de Oliveira Costa <gcosta@redhat.com>
Date:   Fri Feb 8 05:02:47 2008 -0200

    [PATCH] boot a linux kernel from non-ide device
    
    Since it's now possible to use the -drive option, the test for something
    in the index 0 of the IDE bus is too restrictive.
    
    A better idea, IMHO, is to check if the user specified any bootable device,
    and only if not, fallback to the default, compatible behaviour of checking
    hda regardless of the presence of a boot=on arg.

diff --git a/qemu/hw/pc.c b/qemu/hw/pc.c
index 5ce28ab..a9b0f71 100644
--- a/qemu/hw/pc.c
+++ b/qemu/hw/pc.c
@@ -398,11 +398,14 @@ static void generate_bootsect(uint32_t gpr[8], uint16_t segs[6], uint16_t ip)
 {
     uint8_t bootsect[512], *p;
     int i;
-    int hda;
-
-    hda = drive_get_index(IF_IDE, 0, 0);
-    if (hda == -1) {
-	fprintf(stderr, "A disk image must be given for 'hda' when booting "
+    int hda = -1, boot_device;
+
+    if (extboot_drive != -1)
+        boot_device = extboot_drive;
+    else if ((hda = drive_get_index(IF_IDE, 0, 0)) != -1) 
+	boot_device = hda;
+    else {
+	fprintf(stderr, "A bootable disk image must be given when booting "
 		"a Linux kernel\n");
 	exit(1);
     }
@@ -410,7 +413,7 @@ static void generate_bootsect(uint32_t gpr[8], uint16_t segs[6], uint16_t ip)
     memset(bootsect, 0, sizeof(bootsect));
 
     /* Copy the MSDOS partition table if possible */
-    bdrv_read(drives_table[hda].bdrv, 0, bootsect, 1);
+    bdrv_read(drives_table[boot_device].bdrv, 0, bootsect, 1);
 
     /* Make sure we have a partition signature */
     bootsect[510] = 0x55;
@@ -447,7 +450,7 @@ static void generate_bootsect(uint32_t gpr[8], uint16_t segs[6], uint16_t ip)
     *p++ = segs[1];		/* CS */
     *p++ = segs[1] >> 8;
 
-    bdrv_set_boot_sector(drives_table[hda].bdrv, bootsect, sizeof(bootsect));
+    bdrv_set_boot_sector(drives_table[boot_device].bdrv, bootsect, sizeof(bootsect));
 }
 
 static int load_kernel(const char *filename, uint8_t *addr,

Re: [Qemu-devel] [PATCH] boot a linux kernel from non-ide device

[Anthony Liguori]

Glauber de Oliveira Costa wrote:
> Since it's now possible to use the -drive option, the test for something
> in the index 0 of the IDE bus is too restrictive.
>
> A better idea, IMHO, is to check if the user specified any bootable device,
> and only if not, fallback to the default, compatible behaviour of checking
> hda regardless of the presence of a boot=on arg.
>   

extboot isn't in upstream QEMU yet so this patch won't apply.

Regards,

Anthony LIguori

Fwd: [PATCH] boot a linux kernel from non-ide device

[Glauber de Oliveira Costa]

[-- Attachment #1: Type: text/plain, Size: 959 bytes --]

Reposting to kvm-devel, since aliguori notices that I'm relying on
non-upstream features of qemu

---------- Forwarded message ----------
From: Glauber de Oliveira Costa <glommer-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date: Feb 8, 2008 5:05 AM
Subject: [PATCH] boot a linux kernel from non-ide device
To: qemu-devel-qX2TKyscuCcdnm+yROfE0A@public.gmane.org


Since it's now possible to use the -drive option, the test for something
in the index 0 of the IDE bus is too restrictive.

A better idea, IMHO, is to check if the user specified any bootable device,
and only if not, fallback to the default, compatible behaviour of checking
hda regardless of the presence of a boot=on arg.

--
Glauber de Oliveira Costa.
"Free as in Freedom"
http://glommer.net

"The less confident you are, the more serious you have to act."



-- 
Glauber de Oliveira Costa.
"Free as in Freedom"
http://glommer.net

"The less confident you are, the more serious you have to act."

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: qemu.patch --]
[-- Type: text/x-patch; name=qemu.patch, Size: 2048 bytes --]

commit 65b8fcf3bc5b754b720a04fa7efed6fdcd472a6a
Author: Glauber de Oliveira Costa <gcosta@redhat.com>
Date:   Fri Feb 8 05:02:47 2008 -0200

    [PATCH] boot a linux kernel from non-ide device
    
    Since it's now possible to use the -drive option, the test for something
    in the index 0 of the IDE bus is too restrictive.
    
    A better idea, IMHO, is to check if the user specified any bootable device,
    and only if not, fallback to the default, compatible behaviour of checking
    hda regardless of the presence of a boot=on arg.

diff --git a/qemu/hw/pc.c b/qemu/hw/pc.c
index 5ce28ab..a9b0f71 100644
--- a/qemu/hw/pc.c
+++ b/qemu/hw/pc.c
@@ -398,11 +398,14 @@ static void generate_bootsect(uint32_t gpr[8], uint16_t segs[6], uint16_t ip)
 {
     uint8_t bootsect[512], *p;
     int i;
-    int hda;
-
-    hda = drive_get_index(IF_IDE, 0, 0);
-    if (hda == -1) {
-	fprintf(stderr, "A disk image must be given for 'hda' when booting "
+    int hda = -1, boot_device;
+
+    if (extboot_drive != -1)
+        boot_device = extboot_drive;
+    else if ((hda = drive_get_index(IF_IDE, 0, 0)) != -1) 
+	boot_device = hda;
+    else {
+	fprintf(stderr, "A bootable disk image must be given when booting "
 		"a Linux kernel\n");
 	exit(1);
     }
@@ -410,7 +413,7 @@ static void generate_bootsect(uint32_t gpr[8], uint16_t segs[6], uint16_t ip)
     memset(bootsect, 0, sizeof(bootsect));
 
     /* Copy the MSDOS partition table if possible */
-    bdrv_read(drives_table[hda].bdrv, 0, bootsect, 1);
+    bdrv_read(drives_table[boot_device].bdrv, 0, bootsect, 1);
 
     /* Make sure we have a partition signature */
     bootsect[510] = 0x55;
@@ -447,7 +450,7 @@ static void generate_bootsect(uint32_t gpr[8], uint16_t segs[6], uint16_t ip)
     *p++ = segs[1];		/* CS */
     *p++ = segs[1] >> 8;
 
-    bdrv_set_boot_sector(drives_table[hda].bdrv, bootsect, sizeof(bootsect));
+    bdrv_set_boot_sector(drives_table[boot_device].bdrv, bootsect, sizeof(bootsect));
 }
 
 static int load_kernel(const char *filename, uint8_t *addr,

[-- Attachment #3: Type: text/plain, Size: 228 bytes --]

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

[-- Attachment #4: Type: text/plain, Size: 186 bytes --]

_______________________________________________
kvm-devel mailing list
kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/kvm-devel

Re: Fwd: [PATCH] boot a linux kernel from non-ide device

[Anthony Liguori]

This patch seems reasonable to me.

But FWIW, with extboot, it's possible to implement the -kernel option in 
a saner way.  extboot already has code to take over int19 and load a 
kernel from memory on boot.  It was based on the old -kernel support in 
QEMU (prior to hpa's rewrite) so it's not enabled at the moment.  It 
should be pretty easy to update it though.

This approach would allow -kernel to be used without any disk (which 
also solves your problem, but in a different way).  We can also 
eliminate all the boot sector hijacking silliness.

Regards,

Anthony Liguori

Glauber de Oliveira Costa wrote:
> Reposting to kvm-devel, since aliguori notices that I'm relying on
> non-upstream features of qemu
>
> ---------- Forwarded message ----------
> From: Glauber de Oliveira Costa <glommer-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
> Date: Feb 8, 2008 5:05 AM
> Subject: [PATCH] boot a linux kernel from non-ide device
> To: qemu-devel-qX2TKyscuCcdnm+yROfE0A@public.gmane.org
>
>
> Since it's now possible to use the -drive option, the test for something
> in the index 0 of the IDE bus is too restrictive.
>
> A better idea, IMHO, is to check if the user specified any bootable device,
> and only if not, fallback to the default, compatible behaviour of checking
> hda regardless of the presence of a boot=on arg.
>
> --
> Glauber de Oliveira Costa.
> "Free as in Freedom"
> http://glommer.net
>
> "The less confident you are, the more serious you have to act."
>
>
>
>   


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/