[syslinux] [PATCH 0/9] linux/syslinux: support ext2/3/4 device

[Ady]

> On 01/01/2015 09:05 PM, Robert Yang wrote:
> > Hello,
> > 
> > Happy New Year!
> > 
> > These patches make syslinux/linux support ext2/3/4, and it doesn't
> > require the root privilege, I'd like to add a separate e2fs/syslinux, if
> > that is more appropriate, it should be easy to do that.
> > 
> > I put these patches on github so that you can easily get them in case
> > you'd like to test them. (The repo's name is sys_tmp, which avoids
> > confusing others, I will remove the repo when these patches are
> > reviewed).
> > 
> 
> In my opinion, linux/syslinux really should just be part of extlinux,
> since it is exactly equivalent to mounting the filesystem, running
> extlinux, and unmounting the filesystem.  That automatically adds
> support for all filesystems.
> 
> mtools/syslinux is the one that doesn't require root privileges, and
> where libext2fs belongs.  Ideally the FAT support ought to be
> internalized into a library, too.
> 
> The one thing with mtools/syslinux is that some people like to compile
> it for non-Linux platforms (how successfully, I don't know) and so
> ext*fs support should probably be conditionalized.
> 
> 	-hpa
> 
> _______________________________________________
> Syslinux mailing list
> Submissions to Syslinux at zytor.com
> Unsubscribe or set options at:
> http://www.zytor.com/mailman/listinfo/syslinux
> 

The use-case of linux/syslinux is that it doesn't have/need 
dependencies (other than root privileges). This is supposed to be 
useful, for instance for smaller OS installers or alike. If it were to 
be merged with the extlinux command, wouldn't this use-case change (or 
get lost)?

Also, please let's not forget that most users don't rebuild Syslinux 
with specific flags, so whatever ends up being the standard, that's 
what users are "forced" to use.

As I said before (but probably most members decided tl;dr), I'd rather 
see a separate installer rather than reducing the use-case alternatives 
(not that I have a say in the matter). If the use-cases can be maintain 
while also reducing the amount of alternative installers available for 
common users (and not having to rebuild official upstream Syslinux), 
that would be ideal.

Regards,
Ady.