linux-kernel - Re: [GIT PULL] iomap: new code for 5.4

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAHk-=wgXBV57mz46ZB5XivjiSBGkM0cjuvnU2OWyfRF=+41NPQ@mail.gmail.com>
Date:   Wed, 18 Sep 2019 19:07:56 -0700
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     "Darrick J. Wong" <djwong@...nel.org>
Cc:     linux-fsdevel <linux-fsdevel@...r.kernel.org>,
        linux-xfs <linux-xfs@...r.kernel.org>,
        Dave Chinner <david@...morbit.com>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Eric Sandeen <sandeen@...deen.net>,
        Christoph Hellwig <hch@....de>,
        Andreas Gruenbacher <agruenba@...hat.com>,
        Bob Peterson <rpeterso@...hat.com>,
        cluster-devel <cluster-devel@...hat.com>
Subject: Re: [GIT PULL] iomap: new code for 5.4

On Wed, Sep 18, 2019 at 6:31 PM Linus Torvalds
<torvalds@...ux-foundation.org> wrote:
>
> Why would anybody use that odd "list_pop()" thing in a loop, when what
> it really seems to just want is that bog-standard
> "list_for_each_entry_safe()"

Side note: I do agree that the list_for_each_entry_safe() thing isn't
exactly beautiful, particularly since you need that extra variable for
the temporary "next" pointer.

It's one of the C++ features I'd really like to use in the kernel -
the whole "declare new variable in a for (;;) statement" thing.

In fact, it made it into C - it's there in C99 -  but we still use
"-std=gnu89" because of other problems with the c99 updates.

Anyway, I *would* be interested in cleaning up
list_for_each_entry_safe() if somebody has the energy and figures out
what we could do to get the c99 behavior without the breakage from
other sources.

For some background: the reason we use "gnu89" is because we use the
GNU extension with type cast initializers quite a bit, ie things like

    #define __RAW_SPIN_LOCK_UNLOCKED(lockname)      \
        (raw_spinlock_t) __RAW_SPIN_LOCK_INITIALIZER(lockname)

and that broke in c99 and gnu99, which considers those compound
literals and you can no longer use them as initializers.

See

    https://lore.kernel.org/lkml/20141019231031.GB9319@node.dhcp.inet.fi/

for some of the historical discussion about this. It really _is_ sad,
because variable declarations inside for-loops are very useful, and
would have the potential to make some of our "for_each_xyz()" macros a
lot prettier (and easier to use too).

So our list_for_each_entry_safe() thing isn't perfect, but that's no
reason to try to then make up completely new things.

             Linus