Re: mails from markus@trippesldorf.de don't make it to libc-alpha@sourceware.org

[octoploid <octoploid@yandex.com>]

01.02.2017, 17:44, "octoploid" <octoploid@yandex.com>:
> Hi,
>
> I cannot post messages to libc-alpha@sourceware.org with my
> markus@trippesldorf.de account anymore.
> (Posting to gcc-patches@gcc.gnu.org however works fine).
>
> All mails are just silently rejected after 24h.
>
> Could you please remove my email address from the black list?

Here is an example:

Hi. This is the qmail-send program at mail.ud10.udmedia.de.
I'm afraid I wasn't able to deliver your message to the following addresses.
This is a permanent error; I've given up. Sorry it didn't work out.

<libc-alpha@sourceware.org>:
recipient did not talk proper QMTP (#4.3.0)
I'm not going to try again; this message has been in the queue too long.

--- Below this line is a copy of the message.

Return-Path: <markus@trippelsdorf.de>
Received: (qmail 30027 invoked from network); 31 Jan 2017 17:10:06 +0100
Received: from ip5b405f78.dynamic.kabel-deutschland.de (HELO x4) (ud10?360p3@91.64.95.120)
  by mail.ud10.udmedia.de with ESMTPSA (ECDHE-RSA-AES256-SHA encrypted, authenticated); 31 Jan 2017 17:10:06 +0100
Date: Tue, 31 Jan 2017 17:10:05 +0100
From: Markus Trippelsdorf <markus@trippelsdorf.de>
To: Carlos O'Donell <carlos@redhat.com>
Cc: DJ Delorie <dj@redhat.com>, libc-alpha@sourceware.org
Subject: Re: [patch] malloc per-thread cache ready for review
Message-ID: <20170131161005.GD288@x4>
References: <xn8tptn33k.fsf@greed.delorie.com>
 <014d5212-5009-07e4-62c2-c88586c095a5@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <014d5212-5009-07e4-62c2-c88586c095a5@redhat.com>

On 2017.01.31 at 10:14 -0500, Carlos O'Donell wrote:
> On 01/29/2017 01:16 PM, DJ Delorie wrote:
> >
> > Well, thank you for testing it at least :-)
> >
> > I'll work on debugging that.  Hopefully it's something stupid I "fixed"
> > recently, and not a design flaw somewhere...
>
> Whatever email Markus sent you is not on the list?
>
> Markus, would you mind sharing on list?

I will try. The mail was rejected without any given reason.

Here it is:

On 2017.01.26 at 14:29 -0500, DJ Delorie wrote:
>
> Tested on x86-64 and x86-32, including replacing the system glibc.

I gave your patch a quick spin on a x86_64-pc-linux-gnu machine.
Unfortunately it turned out to be quite unusable.
For example gcc would just hang when trying to compile simple programs
(after e.g. cc1plus finishes, it just stays idle, instead of calling the
assembler and linker next).

And sometimes gcc would ICE like:

cc1: arena.c:713: detach_arena: Assertion `replaced_arena->attached_threads > 0' failed.
In file included from malloc.c:1869:0:
arena.c: In function ‘arena_get2.part.5’:
arena.c:937:1: internal compiler error: Aborted
 }
 ^
0x7f4dcc5f23af ???
        /home/markus/glibc/signal/../sysdeps/unix/sysv/linux/x86_64/sigaction.c:0
0x7f4dcc5f232e __GI_raise
        ../sysdeps/unix/sysv/linux/raise.c:51
0x7f4dcc5f3f3f __GI_abort
        /home/markus/glibc/stdlib/abort.c:89
0x7f4dcc6403d9 __malloc_assert
        /home/markus/glibc/malloc/malloc.c:296
0x7f4dcc5dbd00 detach_arena
        /home/markus/glibc/malloc/arena.c:713
0x7f4dcc640fa3 detach_arena
        /home/markus/glibc/malloc/arena.c:833
0x7f4dcc640fa3 _int_new_arena
        /home/markus/glibc/malloc/arena.c:773
0x7f4dcc640fa3 arena_get2
        /home/markus/glibc/malloc/arena.c:962
0x7f4dcc648eca arena_get2
        /home/markus/glibc/malloc/arena.c:930
0x7f4dcc648eca __GI___libc_malloc
        /home/markus/glibc/malloc/malloc.c:2988
0x7f4dcc5dc46f __libc_start_main
        ../csu/libc-start.c:295
0x122e509 ???
        ../sysdeps/x86_64/start.S:120


--
Markus