From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753644Ab1AYNug (ORCPT ); Tue, 25 Jan 2011 08:50:36 -0500 Received: from mga11.intel.com ([192.55.52.93]:57064 "EHLO mga11.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752615Ab1AYNuf (ORCPT ); Tue, 25 Jan 2011 08:50:35 -0500 Message-Id: X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.60,374,1291622400"; d="scan'208";a="880754421" Date: Tue, 25 Jan 2011 13:50:31 +0000 To: Knut Petersen Subject: Re: [BUG/REGRESSION] DRM / i915 / 2.6.37 and 2.6.38-rc*: DVI output gets disabled/reenabled under load Cc: jesse.barnes@intel.com, linux-kernel@vger.kernel.org References: <4D3DC997.5040605@t-online.de> <0d30dc$krk6f6@orsmga001.jf.intel.com> <4D3EB90C.9020901@t-online.de> <0d30dc$krt8da@orsmga001.jf.intel.com> <4D3EC393.3010403@t-online.de> From: Chris Wilson In-Reply-To: <4D3EC393.3010403@t-online.de> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 25 Jan 2011 13:35:31 +0100, Knut Petersen wrote: > Attached find the startup logs. Both debuglevel 0xf. 2.6.36: > [ 2.699943] [drm:intel_sdvo_debug_write], SDVOC: W: 02 (SDVO_CMD_GET_DEVICE_CAPS) > [ 2.736984] [drm:intel_sdvo_debug_response], SDVOC: R: 02 C2 01 01 01 3D 3E 00 (Success) 2.6.38: > [ 2.561275] [drm:intel_sdvo_debug_write], SDVOC: W: 02 (SDVO_CMD_GET_DEVICE_CAPS) > [ 2.565952] [drm:intel_sdvo_read_response], SDVOC: R: (Success) 02 C2 01 01 01 3D 3E 00 > [ 2.579730] [drm:intel_sdvo_get_capabilities], SDVO capabilities: > [ 2.579735] vendor_id: 2 > [ 2.579736] device_id: 194 > [ 2.579737] device_rev_id: 1 > [ 2.579738] sdvo_version_major: 1 > [ 2.579739] sdvo_version_minor: 1 > [ 2.579740] sdvo_inputs_mask: 1 > [ 2.579741] smooth_scaling: 1 > [ 2.579742] sharp_scaling: 1 > [ 2.579743] up_scaling: 1 > [ 2.579744] down_scaling: 1 > [ 2.579745] stall_support: 0 > [ 2.579746] output_flags: 62 Both report output_flags of 0x3e, which includes VGA and a few complementary TV out modes. The VGA-2 seems to be ignored by 2.6.36, but does seem legitimate. Hmm. -Chris -- Chris Wilson, Intel Open Source Technology Centre