stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg KH <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: torvalds@linux-foundation.org, akpm@linux-foundation.org,
	alan@lxorguk.ukuu.org.uk, Rob Castle <futuredub@gmail.com>,
	Eric Appleman <erappleman@gmail.com>,
	Keith Packard <keithp@keithp.com>,
	Kenneth Graunke <kenneth@whitecape.org>,
	Daniel Vetter <daniel.vetter@ffwll.ch>,
	Dave Airlie <airlied@redhat.com>
Subject: [ 23/75] drm/i915: Set the Stencil Cache eviction policy to non-LRA mode.
Date: Fri, 04 May 2012 13:42:47 -0700	[thread overview]
Message-ID: <20120504204226.355894568@linuxfoundation.org> (raw)
In-Reply-To: <20120504204258.GA12552@kroah.com>

3.3-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Kenneth Graunke <kenneth@whitecape.org>

commit 3a69ddd6f872180b6f61fda87152b37202118fbc upstream.

Clearing bit 5 of CACHE_MODE_0 is necessary to prevent GPU hangs in
OpenGL programs such as Google MapsGL, Google Earth, and gzdoom when
using separate stencil buffers.  Without it, the GPU tries to use the
LRA eviction policy, which isn't supported.  This was supposed to be off
by default, but seems to be on for many machines.

This cannot be done in gen6_init_clock_gating with most of the other
workaround bits; the render ring needs to exist.  Otherwise, the
register write gets dropped on the floor (one printk will show it
changed, but a second printk immediately following shows the value
reverts to the old one).

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=47535
Cc: Rob Castle <futuredub@gmail.com>
Cc: Eric Appleman <erappleman@gmail.com>
Cc: aaron667@gmx.net
Cc: Keith Packard <keithp@keithp.com>
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 drivers/gpu/drm/i915/i915_reg.h         |    1 +
 drivers/gpu/drm/i915/intel_ringbuffer.c |    8 ++++++++
 2 files changed, 9 insertions(+)

--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -523,6 +523,7 @@
 #define   CM0_MASK_SHIFT          16
 #define   CM0_IZ_OPT_DISABLE      (1<<6)
 #define   CM0_ZR_OPT_DISABLE      (1<<5)
+#define	  CM0_STC_EVICT_DISABLE_LRA_SNB	(1<<5)
 #define   CM0_DEPTH_EVICT_DISABLE (1<<4)
 #define   CM0_COLOR_EVICT_DISABLE (1<<3)
 #define   CM0_DEPTH_WRITE_DISABLE (1<<1)
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -417,6 +417,14 @@ static int init_render_ring(struct intel
 	if (INTEL_INFO(dev)->gen >= 6) {
 		I915_WRITE(INSTPM,
 			   INSTPM_FORCE_ORDERING << 16 | INSTPM_FORCE_ORDERING);
+
+		/* From the Sandybridge PRM, volume 1 part 3, page 24:
+		 * "If this bit is set, STCunit will have LRA as replacement
+		 *  policy. [...] This bit must be reset.  LRA replacement
+		 *  policy is not supported."
+		 */
+		I915_WRITE(CACHE_MODE_0,
+			   CM0_STC_EVICT_DISABLE_LRA_SNB << CM0_MASK_SHIFT);
 	}
 
 	return ret;



  parent reply	other threads:[~2012-05-04 20:42 UTC|newest]

Thread overview: 88+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-05-04 20:42 [ 00/75] 3.3.5-stable review Greg KH
2012-05-04 20:42 ` [ 01/75] nfs: Enclose hostname in brackets when needed in nfs_do_root_mount Greg KH
2012-05-04 20:42 ` [ 02/75] NFSv4: Ensure that the LOCK code sets exception->inode Greg KH
2012-05-04 20:42 ` [ 03/75] NFSv4: Ensure that we check lock exclusive/shared type against open modes Greg KH
2012-05-04 20:42 ` [ 04/75] NFS: put open context on error in nfs_pagein_multi Greg KH
2012-05-04 20:42 ` [ 05/75] NFS: put open context on error in nfs_flush_multi Greg KH
2012-05-04 20:42 ` [ 06/75] x86, microcode: Fix sysfs warning during module unload on unsupported CPUs Greg KH
2012-05-04 20:42 ` [ 07/75] x86, microcode: Ensure that module is only loaded on supported AMD CPUs Greg KH
2012-05-04 20:42 ` [ 08/75] x86, apic: APIC code touches invalid MSR on P5 class machines Greg KH
2012-05-04 20:42 ` [ 09/75] x86/platform: Remove incorrect error message in x86_default_fixup_cpu_id() Greg KH
2012-05-04 20:42 ` [ 10/75] Revert "autofs: work around unhappy compat problem on x86-64" Greg KH
2012-05-04 20:42 ` [ 11/75] xen: correctly check for pending events when restoring irq flags Greg KH
2012-05-04 20:42 ` [ 12/75] xen/smp: Fix crash when booting with ACPI hotplug CPUs Greg KH
2012-05-04 20:42 ` [ 13/75] ASoC: dapm: Ensure power gets managed for line widgets Greg KH
2012-05-04 20:42 ` [ 14/75] ASoC: wm8994: Improve sequencing of AIF channel enables Greg KH
2012-05-04 20:42 ` [ 15/75] dmaengine: at_hdmac: remove clear-on-read in atc_dostart() Greg KH
2012-05-04 20:42 ` [ 16/75] sched: Fix OOPS when build_sched_domains() percpu allocation fails Greg KH
2012-05-04 20:42 ` [ 17/75] tracing: Fix stacktrace of latency tracers (irqsoff and friends) Greg KH
2012-05-04 20:42 ` [ 18/75] hwmon: fam15h_power: fix bogus values with current BIOSes Greg KH
2012-05-04 20:42 ` [ 19/75] hwmon: (fam15h_power) Fix pci_device_id array Greg KH
2012-05-04 20:42 ` [ 20/75] dell-laptop: Terminate quirks list properly Greg KH
2012-05-04 20:42 ` [ 21/75] drm/radeon/kms: need to set up ss on DP bridges as well Greg KH
2012-05-04 20:42 ` [ 22/75] drm/i915: handle input/output sdvo timings separately in mode_set Greg KH
2012-05-04 20:42 ` Greg KH [this message]
2012-05-04 20:42 ` [ 24/75] drm/i915: fix integer overflow in i915_gem_execbuffer2() Greg KH
2012-05-04 20:42 ` [ 25/75] drm/i915: fix integer overflow in i915_gem_do_execbuffer() Greg KH
2012-05-04 20:42 ` [ 26/75] i387: ptrace breaks the lazy-fpu-restore logic Greg KH
2012-05-04 20:42 ` [ 27/75] nl80211: ensure interface is up in various APIs Greg KH
2012-05-04 20:42 ` [ 28/75] ALSA: HDA: Add external mic quirk for Asus Zenbook UX31E Greg KH
2012-05-04 20:42 ` [ 29/75] USB: cdc-wdm: fix race leading leading to memory corruption Greg KH
2012-05-04 20:42 ` [ 30/75] USB: EHCI: fix crash during suspend on ASUS computers Greg KH
2012-05-04 20:42 ` [ 31/75] USB: gadget: storage gadgets send wrong error code for unknown commands Greg KH
2012-05-04 20:42 ` [ 32/75] usb: gadget: dummy: do not call pullup() on udc_stop() Greg KH
2012-05-04 20:42 ` [ 33/75] usb gadget: uvc: uvc_request_data::length field must be signed Greg KH
2012-05-04 20:42 ` [ 34/75] pipes: add a "packetized pipe" mode for writing Greg KH
2012-05-04 20:42 ` [ 35/75] autofs: make the autofsv5 packet file descriptor use a packetized pipe Greg KH
2012-05-04 20:43 ` [ 36/75] crypto: talitos - properly lock access to global talitos registers Greg KH
2012-05-04 20:43 ` [ 37/75] Input: synaptics - fix regression with "image sensor" trackpads Greg KH
2012-05-04 20:43 ` [ 38/75] USB: ehci-tegra: remove redundant gpio_set_value Greg KH
2012-05-04 20:43 ` [ 39/75] ARM: 7396/1: errata: only handle ARM erratum #326103 on affected cores Greg KH
2012-05-04 20:43 ` [ 40/75] ARM: 7403/1: tls: remove covert channel via TPIDRURW Greg KH
2012-05-04 20:43 ` [ 41/75] ARM: 7406/1: hotplug: copy the affinity mask when forcefully migrating IRQs Greg KH
2012-05-04 20:43 ` [ 42/75] MIPS: ath79: fix AR933X WMAC reset code Greg KH
2012-05-04 20:43 ` [ 43/75] SCSI: libsas: fix sas_find_bcast_phy() in the presence of vacant phys Greg KH
2012-05-04 20:43 ` [ 44/75] SCSI: libsas: fix false positive device attached conditions Greg KH
2012-05-04 20:43 ` [ 45/75] efi: Add new variable attributes Greg KH
2012-05-04 20:43 ` [ 46/75] efi: Validate UEFI boot variables Greg KH
2012-05-04 20:43 ` [ 47/75] x86, efi: Fix pointer math issue in handle_ramdisks() Greg KH
2012-05-04 20:43 ` [ 48/75] tools/include: Add byteshift headers for endian access Greg KH
2012-05-04 20:43 ` [ 49/75] x86, mkpiggy: Dont open code put_unaligned_le32() Greg KH
2012-05-04 20:43 ` [ 50/75] x86, boot: Restrict CFLAGS for hostprogs Greg KH
2012-05-04 20:43 ` [ 51/75] x86, efi: Fix endian issues and unaligned accesses Greg KH
2012-05-04 20:43 ` [ 52/75] x86, boot: Correct CFLAGS for hostprogs Greg KH
2012-05-04 20:43 ` [ 53/75] x86, efi: Add dedicated EFI stub entry point Greg KH
2012-05-04 20:43 ` [ 54/75] powerpc/85xx: dont call of_platform_bus_probe() twice Greg KH
2012-05-04 20:43 ` [ 55/75] PM / Hibernate: fix the number of pages used for hibernate/thaw buffering Greg KH
2012-05-04 20:43 ` [ 56/75] sched: Fix nohz load accounting -- again! Greg KH
2012-05-04 22:03   ` Doug Smythies
     [not found]     ` <002101cd2e38$3adfbc80$b09f3580$@net>
2012-05-22 16:39       ` Doug Smythies
2012-05-04 20:43 ` [ 57/75] exit_signal: simplify the "we have changed execution domain" logic Greg KH
2012-05-07  1:57   ` Ben Hutchings
2012-05-08  0:31     ` Greg KH
2012-05-09  3:07       ` Ben Hutchings
2012-05-04 20:43 ` [ 58/75] exit_signal: fix the "parent has changed security " Greg KH
2012-05-04 20:43 ` [ 59/75] md/raid5: Fix a bug about judging if the operation is syncing or replacing Greg KH
2012-05-04 20:43 ` [ 60/75] efivars: Improve variable validation Greg KH
2012-05-04 20:43 ` [ 61/75] hwmon: (coretemp) Increase CPU core limit Greg KH
2012-05-04 20:43 ` [ 62/75] nouveau: initialise has_optimus variable Greg KH
2012-05-04 20:43 ` [ 63/75] hwmon: (coretemp) fix oops on cpu unplug Greg KH
2012-05-04 20:43 ` [ 64/75] libata: skip old error history when counting probe trials Greg KH
2012-05-04 20:43 ` [ 65/75] b43: only reload config after successful initialization Greg KH
2012-05-04 20:43 ` [ 66/75] i2c: pnx: Disable clk in suspend Greg KH
2012-05-04 20:43 ` [ 67/75] ipw2200: Fix race condition in the command completion acknowledge Greg KH
2012-05-07 14:37   ` Ben Hutchings
2012-05-07 23:45     ` Stanislav Yakovlev
2012-05-08  0:14       ` Ben Hutchings
2012-05-04 20:43 ` [ 68/75] mac80211: fix AP mode EAP tx for VLAN stations Greg KH
2012-05-04 20:43 ` [ 69/75] rtlwifi: Fix oops on unload Greg KH
2012-05-04 20:43 ` [ 70/75] wl1251: fix crash on remove due to premature kfree Greg KH
2012-05-04 20:43 ` [ 71/75] wl1251: fix crash on remove due to leftover work item Greg KH
2012-05-04 20:43 ` [ 72/75] iwlwifi: do not nulify ctx->vif on reset Greg KH
2012-05-04 20:43 ` [ 73/75] iwlwifi: use correct released ucode version Greg KH
2012-05-07 14:41   ` Ben Hutchings
2012-05-07 19:24     ` Venkataraman, Meenakshi
2012-05-09  3:03       ` Ben Hutchings
2012-05-04 20:43 ` [ 74/75] iwlwifi: fix hardware queue programming Greg KH
2012-05-04 20:43 ` [ 75/75] iwlwifi: use 6000G2B for 6030 device series Greg KH
2012-05-05 19:39 ` [ 00/75] 3.3.5-stable review Greg KH

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20120504204226.355894568@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=airlied@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=daniel.vetter@ffwll.ch \
    --cc=erappleman@gmail.com \
    --cc=futuredub@gmail.com \
    --cc=keithp@keithp.com \
    --cc=kenneth@whitecape.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=torvalds@linux-foundation.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).