Re: Does grub-mkrescue support UEFI hybrid USB/DVD images?
On Fri, Jan 23, 2015 at 1:15 AM, Vladimir 'φ-coder/phcoder' Serbinenko
wrote:
> On 22.01.2015 22:25, Kris Moore wrote:
>>
>> Hey all,
>>
>> We are trying to create hybrid USB/DVD images, and they work just fine
>> on BIOS. However it doesn't seem to work when creating UEFI.
>>
> Try this:
> diff --git a/grub-core/disk/efi/efidisk.c b/grub-core/disk/efi/efidisk.c
> index 3b12c34..983afdb 100644
> --- a/grub-core/disk/efi/efidisk.c
> +++ b/grub-core/disk/efi/efidisk.c
> @@ -862,9 +862,7 @@ grub_efidisk_get_device_name (grub_efi_handle_t *handle)
> if (! ctx.partition_name)
> {
> grub_disk_close (parent);
> - if (is_cdrom)
> - return grub_strdup (device_name);
> - return 0;
> + return grub_strdup (device_name);
How is it possible? So we have EFI device path that refers to
partition (or at least to something that does not start on device
first block) but grub does not see this partition? In case of CD-ROM
this happens because CD-ROM media subtype actually refers to ESP and
so points inside CD-ROM. But how it can happen for hard disk?
Kris, could you add output of device path (grub_efi_print_device_path)
at the beginning?
> }
>
> dev_name = grub_xasprintf ("%s,%s", parent->name,
>
>
>
> ___
> Grub-devel mailing list
> Grub-devel@gnu.org
> https://lists.gnu.org/mailman/listinfo/grub-devel
>
___
Grub-devel mailing list
Grub-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/grub-devel
Re: Does grub-mkrescue support UEFI hybrid USB/DVD images?
On 23.01.2015 10:40, Andrei Borzenkov wrote:
> On Fri, Jan 23, 2015 at 1:15 AM, Vladimir 'φ-coder/phcoder' Serbinenko
> wrote:
>> On 22.01.2015 22:25, Kris Moore wrote:
>>>
>>> Hey all,
>>>
>>> We are trying to create hybrid USB/DVD images, and they work just fine
>>> on BIOS. However it doesn't seem to work when creating UEFI.
>>>
>> Try this:
>> diff --git a/grub-core/disk/efi/efidisk.c b/grub-core/disk/efi/efidisk.c
>> index 3b12c34..983afdb 100644
>> --- a/grub-core/disk/efi/efidisk.c
>> +++ b/grub-core/disk/efi/efidisk.c
>> @@ -862,9 +862,7 @@ grub_efidisk_get_device_name (grub_efi_handle_t *handle)
>> if (! ctx.partition_name)
>> {
>> grub_disk_close (parent);
>> - if (is_cdrom)
>> - return grub_strdup (device_name);
>> - return 0;
>> + return grub_strdup (device_name);
>
> How is it possible? So we have EFI device path that refers to
> partition (or at least to something that does not start on device
> first block) but grub does not see this partition? In case of CD-ROM
> this happens because CD-ROM media subtype actually refers to ESP and
> so points inside CD-ROM. But how it can happen for hard disk?
>
I think his EFI uses eltorito even from the stick. In that case we end
up in exactly the same scenario as CD-ROM.
> Kris, could you add output of device path (grub_efi_print_device_path)
> at the beginning?
>
>
>> }
>>
>> dev_name = grub_xasprintf ("%s,%s", parent->name,
>>
>>
>>
>> ___
>> Grub-devel mailing list
>> Grub-devel@gnu.org
>> https://lists.gnu.org/mailman/listinfo/grub-devel
>>
>
> ___
> Grub-devel mailing list
> Grub-devel@gnu.org
> https://lists.gnu.org/mailman/listinfo/grub-devel
>
signature.asc
Description: OpenPGP digital signature
___
Grub-devel mailing list
Grub-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/grub-devel
Re: Does grub-mkrescue support UEFI hybrid USB/DVD images?
On Fri, Jan 23, 2015 at 12:45 PM, Vladimir 'φ-coder/phcoder'
Serbinenko wrote:
> On 23.01.2015 10:40, Andrei Borzenkov wrote:
>> On Fri, Jan 23, 2015 at 1:15 AM, Vladimir 'φ-coder/phcoder' Serbinenko
>> wrote:
>>> On 22.01.2015 22:25, Kris Moore wrote:
Hey all,
We are trying to create hybrid USB/DVD images, and they work just fine
on BIOS. However it doesn't seem to work when creating UEFI.
>>> Try this:
>>> diff --git a/grub-core/disk/efi/efidisk.c b/grub-core/disk/efi/efidisk.c
>>> index 3b12c34..983afdb 100644
>>> --- a/grub-core/disk/efi/efidisk.c
>>> +++ b/grub-core/disk/efi/efidisk.c
>>> @@ -862,9 +862,7 @@ grub_efidisk_get_device_name (grub_efi_handle_t *handle)
>>> if (! ctx.partition_name)
>>> {
>>> grub_disk_close (parent);
>>> - if (is_cdrom)
>>> - return grub_strdup (device_name);
>>> - return 0;
>>> + return grub_strdup (device_name);
>>
>> How is it possible? So we have EFI device path that refers to
>> partition (or at least to something that does not start on device
>> first block) but grub does not see this partition? In case of CD-ROM
>> this happens because CD-ROM media subtype actually refers to ESP and
>> so points inside CD-ROM. But how it can happen for hard disk?
>>
> I think his EFI uses eltorito even from the stick. In that case we end
> up in exactly the same scenario as CD-ROM.
That's fine but then I expect CD-ROM subtype. I'd love to see full device path.
>> Kris, could you add output of device path (grub_efi_print_device_path)
>> at the beginning?
>>
>>
>>> }
>>>
>>> dev_name = grub_xasprintf ("%s,%s", parent->name,
>>>
>>>
>>>
>>> ___
>>> Grub-devel mailing list
>>> Grub-devel@gnu.org
>>> https://lists.gnu.org/mailman/listinfo/grub-devel
>>>
>>
>> ___
>> Grub-devel mailing list
>> Grub-devel@gnu.org
>> https://lists.gnu.org/mailman/listinfo/grub-devel
>>
>
>
>
> ___
> Grub-devel mailing list
> Grub-devel@gnu.org
> https://lists.gnu.org/mailman/listinfo/grub-devel
>
___
Grub-devel mailing list
Grub-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/grub-devel
Re: Does grub-mkrescue support UEFI hybrid USB/DVD images?
Hi, Vladimir wrote: > I think his EFI uses eltorito even from the stick. In that case we end > up in exactly the same scenario as CD-ROM. This contradicts all my beliefs about El Torito (and several wikis where i published them). I would like to see this theory verified: The first experiment would be to remove MBR and GPT. If the theory is right, then it should still boot. Next one should remove the El Torito catalog entry for the EFI boot image and check whether it does not boot any more. Finally it would be interesting to see whether it boots if the EFI boot image is only represented by MBR+GPT but not by El Torito. - To Kris Moore: If Vladimir's patch does fix the problem, then i would be willing to patch such an improved ISO to the experimental forms mentioned above. Just give me a download URL. I will then determine the necessary dd commands for you to execute. (Small ISO preferred.) Have a nice day :) Thomas ___ Grub-devel mailing list Grub-devel@gnu.org https://lists.gnu.org/mailman/listinfo/grub-devel
Re: Does grub-mkrescue support UEFI hybrid USB/DVD images?
On 01/23/2015 05:14, Thomas Schmitt wrote: > Hi, > > Vladimir wrote: >> I think his EFI uses eltorito even from the stick. In that case we end >> up in exactly the same scenario as CD-ROM. > This contradicts all my beliefs about El Torito > (and several wikis where i published them). > > I would like to see this theory verified: > > The first experiment would be to remove MBR and GPT. > If the theory is right, then it should still boot. > Next one should remove the El Torito catalog entry > for the EFI boot image and check whether it does not > boot any more. > Finally it would be interesting to see whether it > boots if the EFI boot image is only represented by > MBR+GPT but not by El Torito. > > - > To Kris Moore: > > If Vladimir's patch does fix the problem, then i would > be willing to patch such an improved ISO to the > experimental forms mentioned above. Just give me a > download URL. I will then determine the necessary dd > commands for you to execute. (Small ISO preferred.) > > > Have a nice day :) > > Thomas > > > ___ > Grub-devel mailing list > Grub-devel@gnu.org > https://lists.gnu.org/mailman/listinfo/grub-devel Vladimir's patch did fix the issue here, thanks! Thomas, Here's a very small ISO, no kernel or anything, around 40MB. www.pcbsd.org/~kris/test-efigrub.iso -- Kris Moore PC-BSD Software iXsystems ___ Grub-devel mailing list Grub-devel@gnu.org https://lists.gnu.org/mailman/listinfo/grub-devel
Re: Does grub-mkrescue support UEFI hybrid USB/DVD images?
Hi, i inspect the image by xorriso-1.3.8 -indev test-efigrub.iso \ -report_el_torito plain \ -report_system_area plain and get Boot record : El Torito , MBR protective-msdos-label cyl-align-off GPT APM ... Volume id: 'PCBSD_INSTALL' El Torito catalog : 1763 1 El Torito cat path : /boot.catalog El Torito images : N Pltf B Emul Ld_seg Hdpt Ldsiz LBA El Torito boot img : 1 UEFI y none 0x 0x00 5760 140 El Torito img path : 1 /efi.img System area options: 0x0201 System area summary: MBR protective-msdos-label cyl-align-off GPT APM ISO image size/512 : 81552 Partition offset : 0 MBR heads per cyl : 64 MBR secs per head : 32 MBR partition table: N Status TypeStart Blocks MBR partition : 1 0x00 0xee181551 GPT: N Info GPT disk GUID : 70e9427b66c1b744b42113c17b268b07 GPT entry array: 20 176 separated GPT lba range : 64 81506 81551 GPT partition name : 1 4700610070003000 GPT partname local : 1 Gap0 GPT partition GUID : 1 70e9427b66c1b744b42213c17b268b07 GPT type GUID : 1 a2a0d0ebe5b9334487c068b6b72699c7 GPT partition flags: 1 0x1001 GPT start and size : 1 64 496 GPT partition name : 2 450046004900200062006f006f007400200070006100720074006900740069006f006e00 GPT partname local : 2 EFI boot partition GPT partition GUID : 2 70e9427b66c1b744b42313c17b268b07 GPT type GUID : 2 28732ac11ff8d211ba4b00a0c93ec93b GPT partition flags: 2 0x1001 GPT start and size : 2 560 5760 GPT partition path : 2 /efi.img GPT partition name : 3 4700610070003100 GPT partname local : 3 Gap1 GPT partition GUID : 3 70e9427b66c1b744b42013c17b268b07 GPT type GUID : 3 a2a0d0ebe5b9334487c068b6b72699c7 GPT partition flags: 3 0x1001 GPT start and size : 3 6320 75184 APM: N Info APM block size : 2048 APM gap fillers: 2 APM partition name : 1 Gap0 APM partition type : 1 ISO9660_data APM start and size : 1 16 1564 APM partition name : 2 HFSPLUS_Hybrid APM partition type : 2 Apple_HFS APM start and size : 2 1580 18646 APM partition name : 3 Gap1 APM partition type : 3 ISO9660_data APM start and size : 3 20226 162 So we have no entry points for BIOS but only for EFI. Those, nevertheless, are plenty. :)) Experiment 1: Remove MBR, GPT, APM, and backup GPT. (by zeroizing the whole ISO System Area and the last 512-block) cp test-efigrub.iso test.iso dd if=/dev/zero conv=notrunc bs=512 count=64 of=test.iso dd if=/dev/zero conv=notrunc bs=512 seek=81551 count=1 of=test.iso Inspection shows: El Torito catalog : 1763 1 El Torito cat path : /boot.catalog El Torito images : N Pltf B Emul Ld_seg Hdpt Ldsiz LBA El Torito boot img : 1 UEFI y none 0x 0x00 5760 140 El Torito img path : 1 /efi.img xorriso : NOTE : No System Area was loaded So if this boots, the firmware either really used El Torito, or it recognizes /efi.img as boot image by some naming rule. Experiment 2: Remove El Torito too. (by moving the Volume Descriptor Set Terminator to the position of the Boot Record) # using test.iso as emerged from experiment 1 dd if=test-efigrub.iso bs=2048 count=1 skip=18 | \ dd conv=notrunc bs=2048 count=1 seek=17 of=test.iso Inspection shows: Volume id: 'PCBSD_INSTALL' xorriso : NOTE : No El Torito information was loaded xorriso : NOTE : No System Area was loaded If this still boots, then EFI finds /efi.img by miracle. Experiment 3: Remove El Torito but keep MBR, GPT, APM. (by moving the Volume Descriptor Set Terminator to the position of the Boot Record) cp test-efigrub.iso test.iso dd if=test-efigrub.iso bs=2048 count=1 skip=18 | \ dd conv=notrunc bs=2048 count=1 seek=17 of=test.iso Inspection: Boot record : (system area only) , MBR protective-msdos-label cyl-align-off GPT APM ... Volume id: 'PCBSD_INSTALL' xorriso : NOTE : No El Torito information was loaded System area options: 0x0201 System area summary: MBR protective-msdos-label cyl-align-off GPT APM ISO image size/512 : 81552 Partition offset : 0 MBR heads per cyl : 64 MBR secs per head : 32 MBR partition table: N Status TypeStart Blocks MBR partition : 1 0x00 0xee181551 GPT: N Info GPT disk GUID : 70e9427b66c1b744b42113c17b268b07 GPT entry array: 20 176 separated GPT lba range : 64 81506 81551 GPT
Re: Does grub-mkrescue support UEFI hybrid USB/DVD images?
В Fri, 23 Jan 2015 10:45:08 +0100
Vladimir 'φ-coder/phcoder' Serbinenko пишет:
> On 23.01.2015 10:40, Andrei Borzenkov wrote:
> > On Fri, Jan 23, 2015 at 1:15 AM, Vladimir 'φ-coder/phcoder' Serbinenko
> > wrote:
> >> On 22.01.2015 22:25, Kris Moore wrote:
> >>>
> >>> Hey all,
> >>>
> >>> We are trying to create hybrid USB/DVD images, and they work just fine
> >>> on BIOS. However it doesn't seem to work when creating UEFI.
> >>>
> >> Try this:
> >> diff --git a/grub-core/disk/efi/efidisk.c b/grub-core/disk/efi/efidisk.c
> >> index 3b12c34..983afdb 100644
> >> --- a/grub-core/disk/efi/efidisk.c
> >> +++ b/grub-core/disk/efi/efidisk.c
> >> @@ -862,9 +862,7 @@ grub_efidisk_get_device_name (grub_efi_handle_t
> >> *handle)
> >> if (! ctx.partition_name)
> >> {
> >> grub_disk_close (parent);
> >> - if (is_cdrom)
> >> - return grub_strdup (device_name);
> >> - return 0;
> >> + return grub_strdup (device_name);
> >
> > How is it possible? So we have EFI device path that refers to
> > partition (or at least to something that does not start on device
> > first block) but grub does not see this partition? In case of CD-ROM
> > this happens because CD-ROM media subtype actually refers to ESP and
> > so points inside CD-ROM. But how it can happen for hard disk?
> >
> I think his EFI uses eltorito even from the stick. In that case we end
> up in exactly the same scenario as CD-ROM.
Not eltorito, but normal ESP.
Media paths
Handle 0x2f241518
/ACPI(a0341d0,0)/PCI(0,4)/HD(1,40,8c,8678dcb08dcc6944,2,2)/EndEntire
disk
block
device path
Handle 0x2f241298
/ACPI(a0341d0,0)/PCI(0,4)/HD(2,cc,1680,8678dcb08dcc6944,2,2)/EndEntire
simple FS
disk
c12a7328-f81f-11d2-ba4b-00a0c93ec93b
block
device path
Handle 0x2f241018
/ACPI(a0341d0,0)/PCI(0,4)/HD(3,174c,3548,8678dcb08dcc6944,2,2)/EndEntire
disk
block
device path
All of them match existing partitions. EFI boots from ESP as it should.
Boot path grub gets
/ACPI(a0341d0,0)/PCI(0,4)/HD(2,cc,1680,8678dcb08dcc6944,2,2)/EndEntire
which matches ESP. The reason it is not found at startup is missing part_gpt:
grub rescue> ls
(hd0) (hd0,apple4) (hd0,apple3) (hd0,apple2) (hd0,apple1) (fd0) (fd1) (cd0)
After adjusting prefix:
grub> insmod part_gpt
grub> ls
(proc) (hd0) (hd0,apple4) (hd0,apple3) (hd0,apple2) (hd0,apple1) (hd0,gpt3)
(hd0,gpt2) (hd0,gpt1) (fd0) (fd1) (cd0)
But returning (hd0,gpt2) would also be wrong here - ESP on CD image is
empty, we need to somehow jump from ESP to full ISO image.
So I guess your patch is the least evil. I will add some comments to
grub_efidisk_get_device_name() and grub-mkrescue to explain what
happens here after you push it.
pgp6eamYbMEqd.pgp
Description: OpenPGP digital signature
___
Grub-devel mailing list
Grub-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/grub-devel
Re: Does grub-mkrescue support UEFI hybrid USB/DVD images?
> But returning (hd0,gpt2) would also be wrong here - ESP on CD image is
> empty, we need to somehow jump from ESP to full ISO image.
>
Returning hd0,gpt2 is fine. By convention ()/file refers to parent disk
of whatever root partition is (as opposed to /file).
So just including part_gpt in core should be enough. But perhaps we
should put my patch as well to increase reliability. Consider scenario
of installing on disk with gpt and apt and using apt partitions. GRUB
will properly detect that it needs part_apple but part_gpt is not needed
to access partition.
Most of the install tools specify partition explicitly, so having
partition is non-critical. Having disk on the hand is critical.
@Kris: Please try (after removing the other patch)
diff --git a/util/grub-mkrescue.c b/util/grub-mkrescue.c
index e719839..e02340f 100644
--- a/util/grub-mkrescue.c
+++ b/util/grub-mkrescue.c
@@ -653,6 +653,9 @@ main (int argc, char *argv[])
char *efiimgfat;
grub_install_mkdir_p (efidir_efi_boot);
+ grub_install_push_module ("part_gpt");
+ grub_install_push_module ("part_msdos");
+
imgname = grub_util_path_concat (2, efidir_efi_boot, "bootia64.efi");
make_image_fwdisk_abs (GRUB_INSTALL_PLATFORM_IA64_EFI,
"ia64-efi", imgname);
free (imgname);
@@ -713,6 +716,8 @@ main (int argc, char *argv[])
free (efiimgfat);
free (efidir_efi);
free (efidir);
+ grub_install_pop_module ();
+ grub_install_pop_module ();
}
grub_install_push_module ("part_apple");
signature.asc
Description: OpenPGP digital signature
___
Grub-devel mailing list
Grub-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/grub-devel
Re: Does grub-mkrescue support UEFI hybrid USB/DVD images?
Hi, Andrei Borzenkov wrote: > Not eltorito, but normal ESP. That's somewhat appeasing. > returning (hd0,gpt2) would also be wrong here - ESP on CD image is > empty, we need to somehow jump from ESP to full ISO image. One could allow nested partitions and put whole-ISO partitions into the various tables. Debian "amd64" ISOs do it that way. They boot on BIOS via ISOLINUX and on EFI via GRUB2. MBR partition table: N Status TypeStart Blocks MBR partition : 1 0x80 0x000 454656 MBR partition : 2 0x00 0xef 3440 896 MBR partition path : 2 /boot/grub/efi.img ... GPT partname local : 1 ISOHybrid ... GPT start and size : 1 0 454600 ... GPT partname local : 2 ISOHybrid1 ... GPT start and size : 2 3440 896 GPT partition path : 2 /boot/grub/efi.img MBR partition 1 and GPT partition 1 cover the whole ISO. Partitions 2 mark the ESP. But Vladimir is clearly opposed to nested or overlapping partitions. And all partition editors seem to join his opinion. Have a nice day :) Thomas ___ Grub-devel mailing list Grub-devel@gnu.org https://lists.gnu.org/mailman/listinfo/grub-devel
Re: Does grub-mkrescue support UEFI hybrid USB/DVD images?
В Fri, 23 Jan 2015 19:29:07 +0100
Vladimir 'φ-coder/phcoder' Serbinenko пишет:
>
> > But returning (hd0,gpt2) would also be wrong here - ESP on CD image is
> > empty, we need to somehow jump from ESP to full ISO image.
> >
> Returning hd0,gpt2 is fine. By convention ()/file refers to parent disk
> of whatever root partition is (as opposed to /file).
Not sure I follow. If you include part_gpt, grub will set prefix to
(hd0,gpt2)/boot/grub. But /boot/grub does NOT exist there - it exists
on (hd0), so correct prefix is (hd0)/boot/grub.
> So just including part_gpt in core should be enough. But perhaps we
> should put my patch as well to increase reliability. Consider scenario
> of installing on disk with gpt and apt and using apt partitions. GRUB
> will properly detect that it needs part_apple but part_gpt is not needed
> to access partition.
> Most of the install tools specify partition explicitly, so having
> partition is non-critical. Having disk on the hand is critical.
>
> @Kris: Please try (after removing the other patch)
> diff --git a/util/grub-mkrescue.c b/util/grub-mkrescue.c
> index e719839..e02340f 100644
> --- a/util/grub-mkrescue.c
> +++ b/util/grub-mkrescue.c
> @@ -653,6 +653,9 @@ main (int argc, char *argv[])
>char *efiimgfat;
>grub_install_mkdir_p (efidir_efi_boot);
>
> + grub_install_push_module ("part_gpt");
> + grub_install_push_module ("part_msdos");
> +
>imgname = grub_util_path_concat (2, efidir_efi_boot, "bootia64.efi");
>make_image_fwdisk_abs (GRUB_INSTALL_PLATFORM_IA64_EFI,
> "ia64-efi", imgname);
>free (imgname);
> @@ -713,6 +716,8 @@ main (int argc, char *argv[])
>free (efiimgfat);
>free (efidir_efi);
>free (efidir);
> + grub_install_pop_module ();
> + grub_install_pop_module ();
> }
>
>grub_install_push_module ("part_apple");
>
>
pgpHmbWCO22Qe.pgp
Description: OpenPGP digital signature
___
Grub-devel mailing list
Grub-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/grub-devel
Re: gnulib update policy
В Sun, 7 Dec 2014 20:34:57 +0300 Andrei Borzenkov пишет: > I have two pending patches for gnulib; each one is included upstream. > What is the policy in this case - update individual files, add patches > to be applied at autogen step or update from upstream (latter I'd > rather avoid at this stage)? ping? ___ Grub-devel mailing list Grub-devel@gnu.org https://lists.gnu.org/mailman/listinfo/grub-devel
Re: Does grub-mkrescue support UEFI hybrid USB/DVD images?
В Fri, 23 Jan 2015 21:38:27 +0300
Andrei Borzenkov пишет:
> В Fri, 23 Jan 2015 19:29:07 +0100
> Vladimir 'φ-coder/phcoder' Serbinenko пишет:
>
> >
> > > But returning (hd0,gpt2) would also be wrong here - ESP on CD image is
> > > empty, we need to somehow jump from ESP to full ISO image.
> > >
> > Returning hd0,gpt2 is fine. By convention ()/file refers to parent disk
> > of whatever root partition is (as opposed to /file).
>
> Not sure I follow. If you include part_gpt, grub will set prefix to
> (hd0,gpt2)/boot/grub. But /boot/grub does NOT exist there - it exists
> on (hd0), so correct prefix is (hd0)/boot/grub.
>
Ah, OK, I see. Sorry.
> > So just including part_gpt in core should be enough. But perhaps we
> > should put my patch as well to increase reliability. Consider scenario
> > of installing on disk with gpt and apt and using apt partitions. GRUB
> > will properly detect that it needs part_apple but part_gpt is not needed
> > to access partition.
> > Most of the install tools specify partition explicitly, so having
> > partition is non-critical. Having disk on the hand is critical.
> >
> > @Kris: Please try (after removing the other patch)
Yes, it works.
> > diff --git a/util/grub-mkrescue.c b/util/grub-mkrescue.c
> > index e719839..e02340f 100644
> > --- a/util/grub-mkrescue.c
> > +++ b/util/grub-mkrescue.c
> > @@ -653,6 +653,9 @@ main (int argc, char *argv[])
> >char *efiimgfat;
> >grub_install_mkdir_p (efidir_efi_boot);
> >
> > + grub_install_push_module ("part_gpt");
> > + grub_install_push_module ("part_msdos");
> > +
> >imgname = grub_util_path_concat (2, efidir_efi_boot, "bootia64.efi");
> >make_image_fwdisk_abs (GRUB_INSTALL_PLATFORM_IA64_EFI,
> > "ia64-efi", imgname);
> >free (imgname);
> > @@ -713,6 +716,8 @@ main (int argc, char *argv[])
> >free (efiimgfat);
> >free (efidir_efi);
> >free (efidir);
> > + grub_install_pop_module ();
> > + grub_install_pop_module ();
> > }
> >
> >grub_install_push_module ("part_apple");
> >
> >
>
pgpGAexama8cY.pgp
Description: OpenPGP digital signature
___
Grub-devel mailing list
Grub-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/grub-devel
Re: Does grub-mkrescue support UEFI hybrid USB/DVD images?
El 22/01/15 a las 22:25, Kris Moore escribió: Hey all, We are trying to create hybrid USB/DVD images, and they work just fine on BIOS. However it doesn't seem to work when creating UEFI. We are using the following: # grub-mkrescue -d "/usr/local/lib/grub/x86_64-efi" -o -- -volid "OURLABEL" This image does boot UEFI via DVD, but when copied to a USB stick with dd it fails to boot with the following: Hi, I'm going to describe what I use for making a USB/CDROM EFI/x86 hybrid image for building SG2D (An improved Grub2 Disk). But my question (after reading other replies to this question) is why Grub2 should be patched for what he wants ? Is it a non standard EFI boot maybe? Thank you. Debian way == Install these two packages: grub-efi-amd64-bin grub-pc-bin Run: grub-mkrescue -o ./my_grub.iso Upstream way (I actually want to ask here if this recipe could be improved, but that would be probably need another email thread) Make sure no Debian grub package is found in chroot apt-get remove '.grub.' Grub2 (efi+pc)) Build and installation # Make sure we are testing the right revision mkdir grub_upstream cd grub_upstream git clone git://git.savannah.gnu.org/grub.git . git checkout COMMIT-TO-TEST # Remove any non git related file - BEGIN git reset --hard git clean -f -d git clean -f -x -d # Remove any non git related file - END # Build efi based grub and install it - BEGIN ./autogen.sh && ./configure --enable-grub-mkfont --with-platform=efi && make # Dont forget to delete any rest of previous installation - BEGIN sudo rm -rf /usr/local/etc/grub.d/ /usr/local/bin/grub-* /usr/local/share/grub/ # Dont forget to delete any rest of previous installation - END sudo make install # Build efi based grub and install it - END make clean # Prepare for next build ./configure --enable-grub-mkfont --with-platform=pc && make # Prepare for a i386 build sudo make install # Install i386 build TL;DR: Make sure no Grub Debian package is installed, install grub twice, one configured for x86, the other one configured for efi. Then grub-mkrescue detects both modules and adds to the final image which by default is ISO/USB hybrid. By the way. I was told how to build the image without doing make install (thus not having to remove any Debian Grub package) so I might re-adapt this recipe in the future. grub-mkrescue -o ./my_grub.iso adrian15 -- Support free software. Donate to Super Grub Disk. Apoya el software libre. Dona a Super Grub Disk. http://www.supergrubdisk.org/donate/ ___ Grub-devel mailing list Grub-devel@gnu.org https://lists.gnu.org/mailman/listinfo/grub-devel
Re: pvgrub2 and old grub menu.lst
В Thu, 3 Apr 2014 14:50:02 +0800
Michael Chang пишет:
> On Wed, Apr 02, 2014 at 08:13:14PM +0400, Andrey Borzenkov wrote:
> > В Wed, 2 Apr 2014 14:29:50 +0400
> > Andrey Borzenkov пишет:
> >
> > > On Wed, Apr 2, 2014 at 12:43 PM, Michael Chang wrote:
> > > > On Tue, Apr 01, 2014 at 08:06:44PM +0400, Andrey Borzenkov wrote:
> > > >> В Tue, 1 Apr 2014 15:46:22 +0800
> > > >> Michael Chang пишет:
> > > >>
> > > >> > While trying to boot xen pv guest with old grub config (menu.lst),
> > > >> > the pvgrub2 loads it's menu successfully, but when you attemp to boot
> > > >> > the kernel it failed with the following message.
> > > >> >
> > > >> > error: couldn't load file /boot/vmlinuz-3.0.76-0.11-xen.
> > > >> > error: you need to load the kernel first.
> > > >> >
> > > >> > Press any key to continue..."
> > > >> >
> > > >> > The kernel is there and boots fine with pygrub.
> > > >> >
> > > >> > After some investigation, it boils down to the menu.lst has
> > > >> > explicitly
> > > >> > set it's root device to (hd0,1) and therefore the root device is not
> > > >> > recognized by pvgrub2.
> > > >> >
> > > >> > title Xen -- SUSE Linux Enterprise Server 11 SP3 - 3.0.76-0.11
> > > >> > root (hd0,1)
> > > >> > kernel /boot/vmlinuz-3.0.76-0.11-xen root=/dev/xvda2
> > > >> > resume=/dev/xvda1
> > > >> > splash=silent showopts initrd /boot/initrd-3.0.76-0.11.xen
> > > >> >
> > > >> > I'd like to know whether old grub configuration is considered or
> > > >> > planned
> > > >> > to work with pvgrub2 ? And is it a viable suggestion to have (hdX,Y)
> > > >> > a
> > > >> > valid xen disk name because it's quite common in some existing
> > > >> > configs ?
> > > >> >
> > > >>
> > > >> I believe this technically should work; but this needs someone who
> > > >> knows how Xen disks is mapped to hd number in legacy grub.
> > > >>
> > > >> It would be something like (completely untested)
> > > >
> > > > Thank you for the patch. I applied it and tried to test it. The
> > > > virtdisks[i].handle never matched with num as it's id of the disk (on
> > > > my system it is 51712).
> > > >
> > >
> > > As I said, completely untested :)
> > >
> > > > Please pardon my ignorant, can num be the index to the virtdisks array
> > > > to get the mapped xen disk ?
> > > >
> > >
> > > pv-grub additionally sorts by handle:
> > > http://xenbits.xen.org/gitweb/?p=xen.git;a=blob;f=stubdom/grub/mini-os.c;h=9d4bcc76d5f3e9907889948ee0dcea32e19c185b;hb=HEAD#l73
> > >
> > > So to actually map pv-grub hdX we need to do the same.
> >
> > Could you test patch below (compile-tested)?
>
> This patch works very well for me. It can boot into system using legacy
> menu.lst.
>
> I also setup two disks for testing, and set root=hdX,Y to list the
> file in repective disk correct. Then swapping the disk order in xen
> config and again the result is correct.
>
> FWIW here my acknowledge for my positive test result.
>
> Tested-By: Michael Chang .
>
> Feel free to let me know anything I can help in this topic, and really
> appreciate for your work and time on this.
>
Finally pushed.
> Thanks,
> Michael
>
> >
> > From: Andrey Borzenkov
> > Subject: [PATCH] accept also hdX as alias to native Xen disk name
> >
> > To assign correct disk numbers, sort disks by increasing order of handle
> > value. This allows reusing legacy pv-grub menu.lst which is using hdX
> > names.
> >
> > Suggested-By: Michael Chang
> >
> > ---
> > grub-core/disk/xen/xendisk.c | 45
> > +++-
> > 1 file changed, 36 insertions(+), 9 deletions(-)
> >
> > diff --git a/grub-core/disk/xen/xendisk.c b/grub-core/disk/xen/xendisk.c
> > index 2b11c2a..b18a923 100644
> > --- a/grub-core/disk/xen/xendisk.c
> > +++ b/grub-core/disk/xen/xendisk.c
> > @@ -40,6 +40,7 @@ struct virtdisk
> >grub_xen_evtchn_t evtchn;
> >void *dma_page;
> >grub_xen_grant_t dma_grant;
> > + struct virtdisk *compat_next;
> > };
> >
> > #define xen_wmb() mb()
> > @@ -47,6 +48,7 @@ struct virtdisk
> >
> > static struct virtdisk *virtdisks;
> > static grub_size_t vdiskcnt;
> > +struct virtdisk *compat_head;
> >
> > static int
> > grub_virtdisk_iterate (grub_disk_dev_iterate_hook_t hook, void *hook_data,
> > @@ -66,20 +68,32 @@ grub_virtdisk_iterate (grub_disk_dev_iterate_hook_t
> > hook, void *hook_data,
> > static grub_err_t
> > grub_virtdisk_open (const char *name, grub_disk_t disk)
> > {
> > - grub_size_t i;
> > + int i;
> >grub_uint32_t secsize;
> >char fdir[200];
> >char *buf;
> > + int num = -1;
> > + struct virtdisk *vd;
> >
> > - for (i = 0; i < vdiskcnt; i++)
> > -if (grub_strcmp (name, virtdisks[i].fullname) == 0)
> > + /* For compatibility with pv-grub legacy menu.lst accept hdX as disk
> > name */
> > + if (name[0] == 'h' && name[1] == 'd' && name[2])
> > +{
> > + num = grub_strtoul (name + 2, 0, 10);
> > + if (grub_errno)
> > + {
> > + grub_errno = 0;
> > + num = -1;
> > +
