Here is a revamped patch.
Samuel
=== modified file 'Makefile.util.def'
--- Makefile.util.def 2012-04-01 19:35:18 +
+++ Makefile.util.def 2012-04-22 18:26:01 +
@@ -170,7 +170,7 @@
ldadd = libgrubgcry.a;
ldadd = libgrubkern.a;
ldadd = grub-core/gnulib/libgnu.a;
- ldadd = '$(LIB
Vladimir 'φ-coder/phcoder' Serbinenko, le Mon 23 Apr 2012 13:17:15 +0200, a
écrit :
> On 23.04.2012 13:06, Samuel Thibault wrote:
> > Our need is to iterate over all disks that GRUB sees, to match that with
> > where the file resides.
> It looks like you have missed some of 2.00 developement.
I d
On Tue, Apr 24, 2012 at 12:54:03AM +0800, Bean wrote:
> Actually there is no problem. Windows 7 stores boot files in one
> partition and system files in another, so you won't find Windows
> directory in the boot partition. I constantly forget about it.
One cannot assume that the Windows 7 bootmg
Hi,
2012/4/24 Bean :
> BTW, I think i understand how this flag works. If the file is
> installed by Windows and not writable, it has a TrustedInstaller
> security descriptor and namespace is set to 0, for example
> C:\Windows\notepad.exe. If the file is modified, there is no
> TrustedInstaller de
Hi,
2012/4/24 Vladimir 'φ-coder/phcoder' Serbinenko :
> On 23.04.2012 18:03, Bean wrote:
>> Hi,
>>
>> 2012/4/23 Vladimir 'φ-coder/phcoder' Serbinenko :
>>> Have you tried with ntfs-3g?
>>> Also if the problem is with only one directory, it's better if it stays
>>> the way it is. The only quirk ass
On 23.04.2012 18:03, Bean wrote:
> Hi,
>
> 2012/4/23 Vladimir 'φ-coder/phcoder' Serbinenko :
>> Have you tried with ntfs-3g?
>> Also if the problem is with only one directory, it's better if it stays
>> the way it is. The only quirk associated with false assumption about
>> case-sensitivity is the
Hi,
2012/4/24 Bean :
> Hi,
>
> 2012/4/23 Vladimir 'φ-coder/phcoder' Serbinenko :
>> Have you tried with ntfs-3g?
>> Also if the problem is with only one directory, it's better if it stays
>> the way it is. The only quirk associated with false assumption about
>> case-sensitivity is the need to ref
Hi,
2012/4/23 Vladimir 'φ-coder/phcoder' Serbinenko :
> Have you tried with ntfs-3g?
> Also if the problem is with only one directory, it's better if it stays
> the way it is. The only quirk associated with false assumption about
> case-sensitivity is the need to reference the name in correct case
On 23.04.2012 17:26, Bean wrote:
> Hi,
>
> 2012/4/23 Vladimir 'φ-coder/phcoder' Serbinenko :
>> On 23.04.2012 16:41, Bean wrote:
>>> Hi all,
>>>
>>> This patch fix three issues:
>>>
>>> 1, In ntfsdoc, it says namespace = 0 means POSIX and it's case
>>> sensitive. However, i found out this is not su
On 23.04.2012 17:09, Vladimir 'φ-coder/phcoder' Serbinenko wrote:
> On 23.04.2012 17:02, Bean wrote:
>> Hi,
>>
>> I believe it only support raid level 0/1/4/5/6/10. It's easy to add
>> linear support, but i guess grub2 is currently in frozen state so you
>> might need to wait a bit.
> It depends on
Hi,
2012/4/23 Vladimir 'φ-coder/phcoder' Serbinenko :
> On 23.04.2012 16:41, Bean wrote:
>> Hi all,
>>
>> This patch fix three issues:
>>
>> 1, In ntfsdoc, it says namespace = 0 means POSIX and it's case
>> sensitive. However, i found out this is not sure in Windows 7 as some
>> system directory u
Op 23/04/2012 om 23:02:45 +0800, schreef Bean:
> I believe it only support raid level 0/1/4/5/6/10. It's easy to add
> linear support, but i guess grub2 is currently in frozen state so you
> might need to wait a bit.
Thanks, I'll wait.
>
> On Mon, Apr 23, 2012 at 10:25 PM, Leo Baltus
> wrote:
On 23.04.2012 17:02, Bean wrote:
> Hi,
>
> I believe it only support raid level 0/1/4/5/6/10. It's easy to add
> linear support, but i guess grub2 is currently in frozen state so you
> might need to wait a bit.
It depends on the kind of the problem. If it's just small probe issue
(like I suppose gi
On 23.04.2012 16:41, Bean wrote:
> Hi all,
>
> This patch fix three issues:
>
> 1, In ntfsdoc, it says namespace = 0 means POSIX and it's case
> sensitive. However, i found out this is not sure in Windows 7 as some
> system directory use this value as well. Now i ignore this flag and
> treat all fi
Hi,
I believe it only support raid level 0/1/4/5/6/10. It's easy to add
linear support, but i guess grub2 is currently in frozen state so you
might need to wait a bit.
On Mon, Apr 23, 2012 at 10:25 PM, Leo Baltus
wrote:
> Op 23/04/2012 om 16:00:26 +0200, schreef Mads Kiilerich:
>> On 04/23/2012
Hi all,
This patch fix three issues:
1, In ntfsdoc, it says namespace = 0 means POSIX and it's case
sensitive. However, i found out this is not sure in Windows 7 as some
system directory use this value as well. Now i ignore this flag and
treat all files in ntfs as case insensitive.
2, Previous v
Op 23/04/2012 om 16:00:26 +0200, schreef Mads Kiilerich:
> On 04/23/2012 03:19 PM, Leo Baltus wrote:
> >Hi,
> >
> >I just tested with grub2-1.99-13 (fedora16) on an md-device with raid-
> >level 'linear'.
> >
> >On grub2-install is says:
>
> Exactly which full command is it that fails?
Im am work
On 04/23/2012 03:19 PM, Leo Baltus wrote:
Hi,
I just tested with grub2-1.99-13 (fedora16) on an md-device with raid-
level 'linear'.
On grub2-install is says:
Exactly which full command is it that fails?
What do you have in /boot/grub2/device.map?
/Mads
error: unsupported RAID level: -1.
Hi,
I just tested with grub2-1.99-13 (fedora16) on an md-device with raid-
level 'linear'.
On grub2-install is says:
error: unsupported RAID level: -1.
/sbin/grub2-probe: error: no such disk.
Auto-detection of a filesystem of /dev/md0 failed.
Is this really unsupported?
--
Leo Baltus, intern
Okay,
I modified the code. But still getting some error messages on the
beginning of starting grub (and later on: "error: file not found, you
need to load the kernel first")
#set keybits = "32";
grub_outb(32, 0x20A);
grub_outw(-16832, 0x20C);
bits9 = grub_inb(0x209);
echo "bits9= "
echo ${bit
On 23.04.2012 14:54, Joachim Mammele wrote:
> Hi,
> accrding to Vladimirs hints I adapted my grub-configuration.
>
> What I changed in /boot/grub/grub.cfg is the following: (I'm aware
> that editing grub.cfg isn't the best option as it gets rewritten on
> executing the config-scripts)
> (the whole
Hi,
accrding to Vladimirs hints I adapted my grub-configuration.
What I changed in /boot/grub/grub.cfg is the following: (I'm aware that
editing grub.cfg isn't the best option as it gets rewritten on executing
the config-scripts)
(the whole file can be found here http://pastebin.com/K57RHjwn)
On 23.04.2012 13:06, Samuel Thibault wrote:
> Vladimir 'φ-coder/phcoder' Serbinenko, le Mon 23 Apr 2012 12:45:09 +0200, a
> écrit :
>> On 23.04.2012 11:40, Samuel Thibault wrote:
>>> Vladimir 'φ-coder/phcoder' Serbinenko, le Mon 23 Apr 2012 10:52:09 +0200, a
>>> écrit :
map was never intende
Vladimir 'φ-coder/phcoder' Serbinenko, le Mon 23 Apr 2012 12:45:09 +0200, a
écrit :
> On 23.04.2012 11:40, Samuel Thibault wrote:
> > Vladimir 'φ-coder/phcoder' Serbinenko, le Mon 23 Apr 2012 10:52:09 +0200, a
> > écrit :
> >> map was never intended to be globally visible. At most perhaps
> >> "s
On 23.04.2012 11:40, Samuel Thibault wrote:
> Vladimir 'φ-coder/phcoder' Serbinenko, le Mon 23 Apr 2012 10:52:09 +0200, a
> écrit :
>> map was never intended to be globally visible. At most perhaps
>> "semi-local" due to difficulties of split between getroot.c and hostdisk.c
>> Moreover the behavi
Vladimir 'φ-coder/phcoder' Serbinenko, le Mon 23 Apr 2012 10:52:09 +0200, a
écrit :
> map was never intended to be globally visible. At most perhaps
> "semi-local" due to difficulties of split between getroot.c and hostdisk.c
> Moreover the behaviour of find_root_device must be independent of what
On 22.04.2012 20:24, Samuel Thibault wrote:
> + sprintf (devpath + strlen (devpath), "s%d", winner_part + 1);
> +}
Also:
- It's probably wrong to assume that Hurd partition numbering is the
same as GRUB's, especially if several partmaps are present.
- Don't use x + strlen (x), keep pointer
On 22.04.2012 20:24, Samuel Thibault wrote:
> Hello,
>
> GNU/Hurd is moving to userland partition support, i.e. /dev/hd0s* slices
> are actually served by userland translators which use libparted. The
> grub probe code needs to be fixed for that, here is a proposed patch.
>
> The principle is that
On 23.04.2012 09:29, Michael Chang wrote:
> I'm cherry picking your patch into our package and will let you know
> once I found any problem.
>
> Btw, When I'm working on this, I'm not sure the *is_part = 1 should
> also be set somewhere after devmapper_out: ? Please correct me if I'm
> wrong.
No it
I'm cherry picking your patch into our package and will let you know
once I found any problem.
Btw, When I'm working on this, I'm not sure the *is_part = 1 should
also be set somewhere after devmapper_out: ? Please correct me if I'm
wrong.
devmapper_out:
if (! mapper_name && node)
{
30 matches
Mail list logo
