From owner-freebsd-emulation@FreeBSD.ORG Tue Sep 9 16:05:03 2008 Return-Path: Delivered-To: freebsd-emulation@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id AF4D210656B8 for ; Tue, 9 Sep 2008 16:05:03 +0000 (UTC) (envelope-from freebsd-emulation@m.gmane.org) Received: from ciao.gmane.org (main.gmane.org [80.91.229.2]) by mx1.freebsd.org (Postfix) with ESMTP id 31C548FC1D for ; Tue, 9 Sep 2008 16:05:02 +0000 (UTC) (envelope-from freebsd-emulation@m.gmane.org) Received: from list by ciao.gmane.org with local (Exim 4.43) id 1Kd4qY-0000k2-I9 for freebsd-emulation@freebsd.org; Tue, 09 Sep 2008 15:09:38 +0000 Received: from 81.210.229.80 ([81.210.229.80]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Tue, 09 Sep 2008 15:09:38 +0000 Received: from saper by 81.210.229.80 with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Tue, 09 Sep 2008 15:09:38 +0000 X-Injected-Via-Gmane: http://gmane.org/ To: freebsd-emulation@freebsd.org From: Marcin Cieslak Date: Tue, 09 Sep 2008 17:09:27 +0200 Lines: 64 Message-ID: <48C691A7.5060302@system.pl> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-ripemd160; protocol="application/pgp-signature"; boundary="------------enig4C256BDCE0C1E5115289672A" X-Complaints-To: usenet@ger.gmane.org X-Gmane-NNTP-Posting-Host: 81.210.229.80 User-Agent: Mozilla/5.0 (X11; U; FreeBSD amd64; en-US; rv:1.8.1.16) Gecko/20080721 MultiZilla/1.8.3.4e SeaMonkey/1.1.11 X-Enigmail-Version: 0.95.6 Sender: news Cc: rdivacky@FreeBSD.org, rink@FreeBSD.org, Robert Watson Subject: Re: kern/102956: [linux] [patch] Add partial support for SO_PEERCRED in Linux emulation X-BeenThere: freebsd-emulation@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Development of Emulators of other operating systems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 09 Sep 2008 16:05:03 -0000 This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enig4C256BDCE0C1E5115289672A Content-Type: text/plain; charset=ISO-8859-2; format=flowed Content-Transfer-Encoding: quoted-printable Hello, > (1) The value of LINUX_SO_PEERCRED is incorrect for Alpha, it should be= 18 on > that platform. Well, in the meantime Alpha support is gone... > (2) I'm a bit worried about pid not being set, but this may (may) be OK= =2E On > Linux, generally speaking you are guaranteed that either you get (0, -1= , > -1) or (pid, uid, gid), but not a blend of both. As we support the pid > for SCM_CREDS, we might also consider adding a LOCAL_PEERPID for use by= > the linux emulator to query the remote pid (we'd need to add that where= > the peercred is currently cached though). Will remote PID always be available? >=20 > (3) LOG_WARNING should perhaps be LOG_DEBUG or something more consisten= t with > the res of the linuxulator. I agree. This should be LOG_DEBUG. > FYI, I'm not sure I like that we just pass all other socket options=20 through to > getsockopt() without transformation or an error, it seems failure-prone= =2E We > may end up returning invalid data, etc, but that's not caused by this p= atch, > but a generally poor failure mode in the linuxulator. I agree. Probably we should explicitly list all supported socket option. = I think most of the translation layer is coded for "fixing known=20 differences" vs. "explicit support for X, Y, Z returning EA, EB or EC". --Marcin --------------enig4C256BDCE0C1E5115289672A Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- iQCVAwUBSMaRqz2W2v2wY27ZAQMSjQQAk1cWPPpb+a/lzUqVTDK5UPF62rjUVABD 5najGil8hJTcwvErzY12FLISIE+pql6rGcKKfVuP3RKR19fBmGMxeRviu8pWdF+2 wrBvcOelWavZH2qEgqd8F25F+3x2SK+QnB4SHme+OYdlOUZyGUtIZaUUWH0sbhki 7UvSZLisI+8= =9JK1 -----END PGP SIGNATURE----- --------------enig4C256BDCE0C1E5115289672A--