Re: [PATCH] Allow specifying the boot device by its label instead of UUID

[darkpenguin]

On 28/09/23 09:16, Oskari Pirhonen wrote:
> I know you're just learning, so here's some more tips:
> 
> Updated patches are generally sent as a "vN" where N is the revision of
> the series. `git format-patch` can do this nicely with the `-v` arg.
> This would make the subject line look like "[PATCH vN] Allow specifying
> the boot device by its label instead of UUID".
> 
> Updated patches are also typically sent as a new thread instead of as a
> reply to the old one. Makes for easier review.

Yes! Thanks! It's not easy to figure out the proper procedure and
conventions because they are not in the manuals. :)


> On Wed, Sep 27, 2023 at 16:25:23 +0000, darkpenguin wrote:
>> Here it is, as a single patch, and in reply to the corresponding thread this 
>> time!
>>
> 
> The way git handles patches, applying this would make the above sentence
> the commit message. Which is probably not what you were after.
> 
>> ---
>>  util/grub-mkconfig.in     |  1 +
>>  util/grub-mkconfig_lib.in | 14 +++++++++++---
>>  util/grub.d/10_linux.in   |  3 +++
>>  3 files changed, 15 insertions(+), 3 deletions(-)
>>
>> diff --git a/util/grub-mkconfig.in b/util/grub-mkconfig.in
>> index 32c480dae..fb5000d3f 100644
>> --- a/util/grub-mkconfig.in
>> +++ b/util/grub-mkconfig.in
>> @@ -244,6 +244,7 @@ export GRUB_DEFAULT \
>>    GRUB_DISABLE_UUID \
>>    GRUB_DISABLE_LINUX_UUID \
>>    GRUB_DISABLE_LINUX_PARTUUID \
>> +  GRUB_ENABLE_LINUX_LABEL \
> 
> I still think this should be GRUB_DISABLE_LINUX_LABEL to match the other
> vars.

I don't mind, so let's do it that way then; I've found a mention of
GRUB_ENABLE_LINUX_LABEL on the Internet, and I thought that maybe it was
used before - so let's keep compatibility. But I think the user who
posted it was mistaken, so either way it fine.

Now that we know that GRUB has nothing to do with fstab, we can just use
GRUB_DISABLE_LINUX_LABEL.