From owner-freebsd-current@freebsd.org Sat Aug 13 07:26:30 2016 Return-Path: Delivered-To: freebsd-current@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 879DABB8258 for ; Sat, 13 Aug 2016 07:26:30 +0000 (UTC) (envelope-from mmacy@nextbsd.org) Received: from sender163-mail.zoho.com (sender163-mail.zoho.com [74.201.84.163]) (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 7AD381696 for ; Sat, 13 Aug 2016 07:26:30 +0000 (UTC) (envelope-from mmacy@nextbsd.org) Received: from mail.zoho.com by mx.zohomail.com with SMTP id 1471073181901813.6290259503966; Sat, 13 Aug 2016 00:26:21 -0700 (PDT) Date: Sat, 13 Aug 2016 00:26:21 -0700 From: Matthew Macy To: "Lundberg, Johannes" Cc: "FreeBSD Current" Message-ID: <15682cb6097.af30fe2f123452.8740060224087388607@nextbsd.org> In-Reply-To: References: Subject: Re: Wayland work status MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Priority: Medium User-Agent: Zoho Mail X-Mailer: Zoho Mail X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 13 Aug 2016 07:26:30 -0000 >=20 > =E2=80=8BWayland relies on kqueue which is not implemented in drm's 3.8 = or 4.6 > branches. I'm working on this now for drm-next-4.6 and it is almost > complete. > I will probably implement it also in the 3.8 =E2=80=8Bbranch to be able = to run > Wayland on both to compare and find bugs in linuxkpi more easily. Will > share patch for 3.8 branch when done. >=20 drm-next has working kevent support following shortly after the drm-v4.7 ta= g. Your kqfilter implementation would be ideal for 3.8 but is not suitable = for drm-next as it substantially modifies vendor code. I think your impleme= ntation does in fact work or is very close. Your apparent problem stemmed f= rom an index overwrite in your modified kmscube.c test case. A working kmscube.c can be found at: https://github.com/FreeBSDDesktop/kmscube With no arguments it will use select. If you pass -k it will use kevent. -M