Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.debian.kernel > #72195 > unrolled thread

"tweewide: Fix most Shebang lines" in stable updates

Started byBen Hutchings <ben@decadent.org.uk>
First post2021-06-15 01:00 +0200
Last post2021-06-15 18:00 +0200
Articles 2 — 2 participants

Back to article view | Back to linux.debian.kernel


Contents

  "tweewide: Fix most Shebang lines" in stable updates Ben Hutchings <ben@decadent.org.uk> - 2021-06-15 01:00 +0200
    Re: "tweewide: Fix most Shebang lines" in stable updates Salvatore Bonaccorso <carnil@debian.org> - 2021-06-15 18:00 +0200

#72195 — "tweewide: Fix most Shebang lines" in stable updates

FromBen Hutchings <ben@decadent.org.uk>
Date2021-06-15 01:00 +0200
Subject"tweewide: Fix most Shebang lines" in stable updates
Message-ID<Cq3IL-6rB-9@gated-at.bofh.it>

[Multipart message — attachments visible in raw view] — view raw

Commit c25ce589dca10d64dde139ae093abc258a32869c "tweewide: Fix most
Shebang lines" was backported to stable branches including 4.9 and
4.19.  This changes scripts interpreted by bash, perl, and python to
use /usr/bin/env instead of an absolute path to the interpreter.

I think that this carries some risk of regression for users building
custom kernels or out-of-tree module.  So perhaps we should revert it
for stretch and buster?

Ben.

-- 
Ben Hutchings
Teamwork is essential - it allows you to blame someone else.

[toc] | [next] | [standalone]


#72198

FromSalvatore Bonaccorso <carnil@debian.org>
Date2021-06-15 18:00 +0200
Message-ID<CqjDP-7M9-1@gated-at.bofh.it>
In reply to#72195
Hi Ben,

On Tue, Jun 15, 2021 at 12:52:31AM +0200, Ben Hutchings wrote:
> Commit c25ce589dca10d64dde139ae093abc258a32869c "tweewide: Fix most
> Shebang lines" was backported to stable branches including 4.9 and
> 4.19.  This changes scripts interpreted by bash, perl, and python to
> use /usr/bin/env instead of an absolute path to the interpreter.
> 
> I think that this carries some risk of regression for users building
> custom kernels or out-of-tree module.  So perhaps we should revert it
> for stretch and buster?

When this was asked for backporting and then queued for stable series
I was indeed worried this might cause issues. Other have raised as
well concern, but it was then appied.

https://lore.kernel.org/stable/YJvC2W9QTpc9JBp1@kroah.com/
and then
https://lore.kernel.org/stable/20210520203625.GA6187@amd/

How about the following: keep it as close to upstream, in case we get
reports, try again to convince upstream to revert the changes as
unsuitable for a stable series. Otherwise we would have to diverge.

But personally I would prefer to remain as close as possible to
upstream.

Regards,
Salvatore

[toc] | [prev] | [standalone]


Back to top | Article view | linux.debian.kernel


csiph-web