From mboxrd@z Thu Jan  1 00:00:00 1970
From: Ferruh Yigit <ferruh.yigit@intel.com>
Subject: Re: [PATCH] Revert "eal: set name to threads"
Date: Thu, 19 Nov 2015 16:59:35 +0000
Message-ID: <20151119165935.GA4888@sivlogin002.ir.intel.com>
References: <103431910.iACTIb73WP@xps13>
 <1447932137-16886-1-git-send-email-ferruh.yigit@intel.com>
 <564DB761.5090203@redhat.com> <8530847.T7t25DiR7m@xps13>
 <20151119133924.GA28008@sivlogin002.ir.intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
To: Thomas Monjalon <thomas.monjalon@6wind.com>, dev@dpdk.org,
 Panu Matilainen <pmatilai@redhat.com>
Return-path: <dev-bounces@dpdk.org>
Received: from mga01.intel.com (mga01.intel.com [192.55.52.88])
 by dpdk.org (Postfix) with ESMTP id DBBC08D91
 for <dev@dpdk.org>; Thu, 19 Nov 2015 17:59:39 +0100 (CET)
Content-Disposition: inline
In-Reply-To: <20151119133924.GA28008@sivlogin002.ir.intel.com>
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
Sender: "dev" <dev-bounces@dpdk.org>

On Thu, Nov 19, 2015 at 01:39:24PM +0000, Ferruh Yigit wrote:
> On Thu, Nov 19, 2015 at 01:32:44PM +0100, Thomas Monjalon wrote:
> > 2015-11-19 13:49, Panu Matilainen:
> > > On 11/19/2015 01:22 PM, Ferruh Yigit wrote:
> > > > This reverts commit 67b6d3039e9edbc4624c878c6930be5e126e8b58.
> > > >
> > > > Reverted patch uses pthread_setname_np() function, this function added
> > > > into glibc in version 2.12 and cause a compile error in older glibc
> > > > versions:
> > > > error: implicit declaration of function "pthread_setname_np"
> > > >
> > > > Main purpose of reverted patch is to name threads, without
> > > > pthread_setname_np() function, patch does not mean much, so reverting
> > > > patch for sake of compatibility with older glibc versions.
> > > 
> > > Debuggability is important too. Rather than revert, why not wrap it in 
> > > rte_thread_setname() or such and just make it a no-op with glibc 
> > > versions where pthread_setname_np() is not available?
> > 
> > +1
> 
> Which means adding compile time glibc version check which I was trying to avoid,
> I believe we should not add glibc version dependencies unless feature is really required,
> but sure I can implement that way, will send v2.
> 
I tried defining weak symbol within DPDK, and library already has strong version, this was nice try, thanks to Sergio,
but this also does not work because how linker works, weak symbol in binary overrides the strong in shared library..