NetBSD Problem Report #43327
From khorben@defora.org Wed May 19 11:06:40 2010
Return-Path: <khorben@defora.org>
Received: from mail.netbsd.org (mail.netbsd.org [204.152.190.11])
by www.NetBSD.org (Postfix) with ESMTP id 019C463B873
for <gnats-bugs@gnats.NetBSD.org>; Wed, 19 May 2010 11:06:39 +0000 (UTC)
Message-Id: <20100519110635.57C74D0@kwarx.defora.lan>
Date: Wed, 19 May 2010 13:06:35 +0200 (CEST)
From: Pierre Pronchery <khorben@defora.org>
To: gnats-bugs@gnats.NetBSD.org
Subject: Allowing suspend with radeondrm on NetBSD 5
X-Send-Pr-Version: 3.95
>Number: 43327
>Category: kern
>Synopsis: It is currently not possible to suspend with radeondrm configured
>Confidential: no
>Severity: non-critical
>Priority: medium
>Responsible: mrg
>State: closed
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Wed May 19 11:10:00 +0000 2010
>Closed-Date: Sun Jun 20 06:51:05 +0000 2010
>Last-Modified: Sun Jun 20 15:00:03 +0000 2010
>Originator: Pierre Pronchery
>Release: NetBSD 5.1_RC1
>Organization:
>Environment:
System: NetBSD kwarx.defora.lan 5.1_RC1 NetBSD 5.1_RC1 (GENERIC) #2: Wed May 5 04:21:36 CEST 2010 khorben@kwarx.defora.lan:/usr/obj/sys/arch/amd64/compile/GENERIC amd64
Architecture: x86_64
Machine: amd64
>Description:
A fix was committed two months ago by mrg@ to "make suspend/resume actually work" with radeondrm configured. I have just backported and tested these changes, which I think would be nice to have in the netbsd-5 branch.
>How-To-Repeat:
Enable radeondrm in the kernel and try to suspend; it refuses:
| Devices without power management support: radeondrm0
With the following patch, suspend and resume works fine, even though I could observe the following errors with my Lenovo ThinkPad T60 on docking station:
| ACPI Error (utmutex-0422): Mutex [4] is not acquired, cannot release [20080321] ACPI Error (dswstate-0259): Failed to extend the result stack [20080321]
| ACPI Error (dswstate-0185): Result stack is empty! State=0xffff800004878000 [20080321]
| ACPI Error (utmutex-0422): ACPI Error (psparse-0627): Mutex [4] is not acquired, cannot releaseMethod parse/execution failed [20080321]
[\_GPE._L01] (Node 0xffff800004019be0), AE_NO_MEMORY ACPI Exception (evgpe-0711): AE_NO_MEMORY, while evaluating GPE method [_L01] [20080321]
| ACPI Error (utmutex-0422): Mutex [1] is not acquired, cannot release [20080321] ACPI Error (utmutex-0422): Mutex [0] is not acquired, cannot release [20080321]
| ACPI Error (exutils-0249): Could not release AML Interpreter mutex [20080321] ioapic0 reenabling
| ACPI Error (utmutex-0422): Mutex [0] is not acquired, cannot release [20080321] ACPI Error (exutils-0249): Could not release AML Interpreter mutex [20080321]
>Fix:
Index: sys/external/bsd/drm/dist/bsd-core/radeon_drv.c
===================================================================
RCS file: /cvsroot/src/sys/external/bsd/drm/dist/bsd-core/radeon_drv.c,v
retrieving revision 1.2.10.1
diff -p -u -r1.2.10.1 radeon_drv.c
--- sys/external/bsd/drm/dist/bsd-core/radeon_drv.c 20 Jun 2009 23:36:59 -0000 1.2.10.1
+++ sys/external/bsd/drm/dist/bsd-core/radeon_drv.c 19 May 2010 10:41:47 -0000
@@ -134,6 +134,38 @@ MODULE_DEPEND(radeon, drm, 1, 1, 1);
#elif defined(__NetBSD__)
+static bool
+radeondrm_suspend(device_t self PMF_FN_ARGS)
+{
+ struct drm_device *dev = device_private(self);
+ drm_radeon_cp_stop_t stop_args;
+ bool rv = true;
+
+ stop_args.flush = stop_args.idle = 0;
+ DRM_LOCK();
+ if (drm_find_file_by_proc(dev, curlwp->l_proc) &&
+ radeon_cp_stop(dev, &stop_args, dev->lock.file_priv) != 0)
+ rv = false;
+ DRM_UNLOCK();
+
+ return rv;
+}
+
+static bool
+radeondrm_resume(device_t self PMF_FN_ARGS)
+{
+ struct drm_device *dev = device_private(self);
+ bool rv = true;
+
+ DRM_LOCK();
+ if (drm_find_file_by_proc(dev, curlwp->l_proc) &&
+ radeon_cp_resume(dev, NULL, NULL) != 0)
+ rv = false;
+ DRM_UNLOCK();
+
+ return rv;
+}
+
static int
radeondrm_probe(device_t parent, cfdata_t match, void *aux)
{
@@ -152,11 +184,22 @@ radeondrm_attach(device_t parent, device
radeon_configure(dev);
+ if (!pmf_device_register(self, radeondrm_suspend, radeondrm_resume))
+ aprint_error_dev(self, "couldn't establish power handler\n");
+
drm_attach(self, pa, radeon_pciidlist);
}
+static int
+radeondrm_detach(device_t self, int flags)
+{
+ pmf_device_deregister(self);
+
+ return drm_detach(self, flags);
+}
+
CFATTACH_DECL_NEW(radeondrm, sizeof(struct drm_device),
- radeondrm_probe, radeondrm_attach, drm_detach, drm_activate);
+ radeondrm_probe, radeondrm_attach, radeondrm_detach, NULL);
#ifdef _MODULE
>Release-Note:
>Audit-Trail:
Responsible-Changed-From-To: kern-bug-people->mrg
Responsible-Changed-By: mrg@NetBSD.org
Responsible-Changed-When: Mon, 24 May 2010 02:18:05 +0000
Responsible-Changed-Why:
mine.
State-Changed-From-To: open->pending-pullups
State-Changed-By: mrg@NetBSD.org
State-Changed-When: Mon, 24 May 2010 02:18:05 +0000
State-Changed-Why:
i've forwarded this patch to the netbsd 5 releng team for consideration.
State-Changed-From-To: pending-pullups->closed
State-Changed-By: mrg@NetBSD.org
State-Changed-When: Sun, 20 Jun 2010 06:51:05 +0000
State-Changed-Why:
this was pulled up to netbsd-5.1. thanks.
From: Jeff Rizzo <riz@tastylime.net>
To: gnats-bugs@NetBSD.org
Cc:
Subject: Re: kern/43327 (It is currently not possible to suspend with radeondrm
configured)
Date: Sun, 20 Jun 2010 07:58:42 -0700
On 6/19/10 11:51 PM, mrg@NetBSD.org wrote:
> Synopsis: It is currently not possible to suspend with radeondrm configured
>
> State-Changed-From-To: pending-pullups->closed
> State-Changed-By: mrg@NetBSD.org
> State-Changed-When: Sun, 20 Jun 2010 06:51:05 +0000
> State-Changed-Why:
> this was pulled up to netbsd-5.1. thanks.
>
>
>
Actually, this pullup is stalled until after 5.1 is released, but it
should be pulled up right after.
>Unformatted:
(Contact us)
$NetBSD: query-full-pr,v 1.39 2013/11/01 18:47:49 spz Exp $
$NetBSD: gnats_config.sh,v 1.8 2006/05/07 09:23:38 tsutsui Exp $
Copyright © 1994-2007
The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.