From: Theodore Ts'o Subject: Re: [PATCH] e4defrag: fix build when posix_fadvise is missing Date: Wed, 1 Jan 2014 12:22:05 -0500 Message-ID: <20140101172205.GC17519@thunk.org> References: <33245e3808058c72b66931ac14aea8d5dc6d1ba5.1388572103.git.baruch@tkos.co.il> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-ext4@vger.kernel.org To: Baruch Siach Return-path: Received: from imap.thunk.org ([74.207.234.97]:44950 "EHLO imap.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754072AbaAARWK (ORCPT ); Wed, 1 Jan 2014 12:22:10 -0500 Content-Disposition: inline In-Reply-To: <33245e3808058c72b66931ac14aea8d5dc6d1ba5.1388572103.git.baruch@tkos.co.il> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Wed, Jan 01, 2014 at 12:28:23PM +0200, Baruch Siach wrote: > uClibc declares posix_fadvise() even when the architecture does not provide > one. The static posix_fadvise() signature is not compatible with POSIX. Rename > the internal implementation to fix this. If the architecture doesn't provide posix_fadvise(), does that imply that __NR_fadvise64_64 also doesn't exist? Or do you mean that for some reason, uClibc is not providing posix_fadvise on all architectures, even though the kernel supports it? That seems wierd. - Ted