[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: workaround for MS Windows' insanity (ignores partition types / hidde
From: |
Marcel Partap |
Subject: |
Re: workaround for MS Windows' insanity (ignores partition types / hidden flag since 2018) |
Date: |
Mon, 18 Nov 2019 08:44:06 +0100 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.2.2 |
>> On Wed, 13 Nov 2019, 17:12 Marcel Partap, <address@hidden> wrote:
>> […] I propose to add a flag that overrides this code:
>> > static inline int
>> > grub_msdos_partition_is_empty (int type)
>> > {
>> > return (type == GRUB_PC_PARTITION_TYPE_NONE);
>> > }
>> to allow booting from partitions of type 0 empty/none.
>>
>From my limited impact analysis (rg -pzLuuC50 partition_is_empty), the only
>change in behaviour would be grub not failing to boot from such a partition.
Other opinions on this?
> Type 0 means that entry is empty and changing this is to expose bunch of
> garbage partition
Where would these come from? I never encountered a partition of type 0 in the
wild.. No software seems to put partitions into the part table, then flag them
as empty.. which is why this should be a no-impact fix, especially when it is a
grub config flag.
> and this is behavior followed by other consumers like Linux and BSD.
Yes of course the current behaviour is default because it is sane. But in a
world where insane software still plays a dominating rule, sometimes conscious
non-sane workarounds can be an apt kludge.
> However we can assist you in finding a better solution. Did you already
> consider alternative partition maps? I'd try msdos+gpt or msdos+BSD or
> msdos+sunpc.
Yes I tried with hybrid DOS/GPT partition scheme, no dice.
> Another workaround is to put some fake FAT structures that will end up as
> showing as a FS. You can even put some explanation files there
That might be a way, but I would have no idea how to do that and it is somewhat
more complex to apply..