public inbox for stable@vger.kernel.org
 help / color / mirror / Atom feed
From: Sasha Levin <sasha.levin@oracle.com>
To: stable@vger.kernel.org, stable-commits@vger.kernel.org
Cc: "Ville Syrjälä" <ville.syrjala@linux.intel.com>,
	"Akash Goel" <akash.goel@intel.com>,
	"Chris Wilson" <chris@chris-wilson.co.uk>,
	"Jani Nikula" <jani.nikula@intel.com>,
	"Sasha Levin" <sasha.levin@oracle.com>
Subject: [added to the 4.1 stable tree] drm/i915: Make RPS EI/thresholds multiple of 25 on SNB-BDW
Date: Thu, 19 May 2016 00:19:20 -0400	[thread overview]
Message-ID: <1463631606-32540-21-git-send-email-sasha.levin@oracle.com> (raw)
In-Reply-To: <1463631606-32540-1-git-send-email-sasha.levin@oracle.com>

From: Ville Syrjälä <ville.syrjala@linux.intel.com>

This patch has been added to the 4.1 stable tree. If you have any
objections, please let us know.

===============

[ Upstream commit 4ea3959018d09edfa36a9e7b5ccdbd4ec4b99e49 ]

Somehow my SNB GT1 (Dell XPS 8300) gets very unhappy around
GPU hangs if the RPS EI/thresholds aren't suitably aligned.
It seems like scheduling/timer interupts stop working somehow
and things get stuck eg. in usleep_range().

I bisected the problem down to
commit 8a5864377b12 ("drm/i915/skl: Restructured the gen6_set_rps_thresholds function")
I observed that before all the values were at least multiples of 25,
but afterwards they are not. And rounding things up to the next multiple
of 25 does seem to help, so lets' do that. I also tried roundup(..., 5)
but that wasn't sufficient. Also I have no idea if we might need this sort of
thing on gen9+ as well.

These are the original EI/thresholds:
 LOW_POWER
  GEN6_RP_UP_EI          12500
  GEN6_RP_UP_THRESHOLD   11800
  GEN6_RP_DOWN_EI        25000
  GEN6_RP_DOWN_THRESHOLD 21250
 BETWEEN
  GEN6_RP_UP_EI          10250
  GEN6_RP_UP_THRESHOLD    9225
  GEN6_RP_DOWN_EI        25000
  GEN6_RP_DOWN_THRESHOLD 18750
 HIGH_POWER
  GEN6_RP_UP_EI           8000
  GEN6_RP_UP_THRESHOLD    6800
  GEN6_RP_DOWN_EI        25000
  GEN6_RP_DOWN_THRESHOLD 15000

These are after 8a5864377b12:
 LOW_POWER
  GEN6_RP_UP_EI          12500
  GEN6_RP_UP_THRESHOLD   11875
  GEN6_RP_DOWN_EI        25000
  GEN6_RP_DOWN_THRESHOLD 21250
 BETWEEN
  GEN6_RP_UP_EI          10156
  GEN6_RP_UP_THRESHOLD    9140
  GEN6_RP_DOWN_EI        25000
  GEN6_RP_DOWN_THRESHOLD 18750
 HIGH_POWER
  GEN6_RP_UP_EI           7812
  GEN6_RP_UP_THRESHOLD    6640
  GEN6_RP_DOWN_EI        25000
  GEN6_RP_DOWN_THRESHOLD 15000

And these are what we have after this patch:
 LOW_POWER
  GEN6_RP_UP_EI          12500
  GEN6_RP_UP_THRESHOLD   11875
  GEN6_RP_DOWN_EI        25000
  GEN6_RP_DOWN_THRESHOLD 21250
 BETWEEN
  GEN6_RP_UP_EI          10175
  GEN6_RP_UP_THRESHOLD    9150
  GEN6_RP_DOWN_EI        25000
  GEN6_RP_DOWN_THRESHOLD 18750
 HIGH_POWER
  GEN6_RP_UP_EI           7825
  GEN6_RP_UP_THRESHOLD    6650
  GEN6_RP_DOWN_EI        25000
  GEN6_RP_DOWN_THRESHOLD 15000

Cc: stable@vger.kernel.org
Cc: Akash Goel <akash.goel@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Testcase: igt/kms_pipe_crc_basic/hang-read-crc-pipe-B
Fixes: 8a5864377b12 ("drm/i915/skl: Restructured the gen6_set_rps_thresholds function")
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1461159836-9108-1-git-send-email-ville.syrjala@linux.intel.com
Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Patrik Jakobsson <patrik.jakobsson@linux.intel.com>
(cherry picked from commit 8a292d016d1cc4938ff14b4df25328230b08a408)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
---
 drivers/gpu/drm/i915/i915_reg.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index a30db4b..d20d818 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2452,7 +2452,14 @@ enum skl_disp_power_wells {
 #define GEN6_RP_STATE_LIMITS	(MCHBAR_MIRROR_BASE_SNB + 0x5994)
 #define GEN6_RP_STATE_CAP	(MCHBAR_MIRROR_BASE_SNB + 0x5998)
 
-#define INTERVAL_1_28_US(us)	(((us) * 100) >> 7)
+/*
+ * Make these a multiple of magic 25 to avoid SNB (eg. Dell XPS
+ * 8300) freezing up around GPU hangs. Looks as if even
+ * scheduling/timer interrupts start misbehaving if the RPS
+ * EI/thresholds are "bad", leading to a very sluggish or even
+ * frozen machine.
+ */
+#define INTERVAL_1_28_US(us)	roundup(((us) * 100) >> 7, 25)
 #define INTERVAL_1_33_US(us)	(((us) * 3)   >> 2)
 #define GT_INTERVAL_FROM_US(dev_priv, us) (IS_GEN9(dev_priv) ? \
 				INTERVAL_1_33_US(us) : \
-- 
2.5.0


  parent reply	other threads:[~2016-05-19  4:21 UTC|newest]

Thread overview: 69+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-05-19  4:19 [added to the 4.1 stable tree] Revert "usb: hub: do not clear BOS field during reset device" Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] stable: remove artifact created on backport Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] iwlwifi: pcie: lower the debug level for RSA semaphore access Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ASoC: rt5640: Correct the digital interface data select Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] regulator: s2mps11: Fix invalid selector mask and voltages for buck9 Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] libahci: save port map for forced port map Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ata: ahci-platform: Add ports-implemented DT bindings Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] regmap: spmi: Fix regmap_spmi_ext_read in multi-byte case Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] iio: ak8975: Fix NULL pointer exception on early interrupt Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] efi: Fix out-of-bounds read in variable_matches() Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] USB: serial: cp210x: add ID for Link ECU Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] USB: serial: cp210x: add Straizona Focusers device ids Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] [media] v4l2-dv-timings.h: fix polarity for 4k formats Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] MD: make bio mergeable Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ALSA: hda - Add dock support for ThinkPad X260 Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] workqueue: fix ghost PENDING flag while doing MQ IO Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] drm/dp/mst: Get validated port ref in drm_dp_update_payload_part1() Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] drm/dp/mst: Restore primary hub guid on resume Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] cxl: Keep IRQ mappings on context teardown Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] drm/i915/ddi: Fix eDP VDD handling during booting and suspend/resume Sasha Levin
2016-05-19  4:19 ` Sasha Levin [this message]
2016-05-19  4:19 ` [added to the 4.1 stable tree] drm/radeon: fix vertical bars appear on monitor (v2) Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ARM: SoCFPGA: Fix secondary CPU startup in thumb2 kernel Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ARM: cpuidle: Pass on arm_cpuidle_suspend()'s return value Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] IB/security: Restrict use of the write() interface Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] mm/huge_memory: replace VM_NO_THP VM_BUG_ON with actual VMA check Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] mm: vmscan: reclaim highmem zone if buffer_heads is over limit Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] mm: soft-offline: don't free target page in successful page migration Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] mm: check __PG_HWPOISON separately from PAGE_FLAGS_CHECK_AT_* Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ALSA: usb-audio: Quirk for yet another Phoenix Audio devices (v2) Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] EDAC: i7core, sb_edac: Don't return NOTIFY_BAD from mce_decoder callback Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] atomic_open(): fix the handling of create_error Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] Drivers: hv: ring_buffer.c: fix comment style Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] Drivers: hv_vmbus: Fix signal to host condition Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] Drivers: hv: vmbus: Fix signaling logic in hv_need_to_signal_on_read() Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] powerpc: Fix bad inline asm constraint in create_zero_mask() Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] Minimal fix-up of bad hashing behavior of hash_64() Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] tracing: Don't display trigger file for events that can't be enabled Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] drm/radeon: make sure vertical front porch is at least 1 Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] MAINTAINERS: Remove asterisk from EFI directory names Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ACPICA: Dispatcher: Update thread ID for recursive method calls Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] crypto: hash - Fix page length clamping in hash walk Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] x86/sysfb_efi: Fix valid BAR address range check Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] fs/pnode.c: treat zero mnt_group_id-s as unequal Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] propogate_mnt: Handle the first propogated copy being a slave Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] drm/radeon: fix DP link training issue with second 4K monitor Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] mm, cma: prevent nr_isolated_* counters from going negative Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] x86/tsc: Read all ratio bits from MSR_PLATFORM_INFO Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] parisc: fix a bug when syscall number of tracee is __NR_Linux_syscalls Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] get_rock_ridge_filename(): handle malformed NM entries Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ALSA: hda - Apply fix for white noise on Asus N550JV, too Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ALSA: hda - Fix white noise on Asus UX501VW headset Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] Input: max8997-haptic - fix NULL pointer dereference Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] drm/i915: Bail out of pipe config compute loop on LPT Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ALSA: hda - Fix broken reconfig Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ALSA: hda - Asus N750JV external subwoofer fixup Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ALSA: hda - Fix white noise on Asus N750JV headphone Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ALSA: hda - Fix subwoofer pin on ASUS N751 and N551 Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] ALSA: usb-audio: Yet another Phoneix Audio device quirk Sasha Levin
2016-05-19  4:19 ` [added to the 4.1 stable tree] perf/core: Disable the event on a truncated AUX record Sasha Levin
2016-05-23  6:59   ` Alexander Shishkin
2016-05-30 21:50     ` Sasha Levin
2016-05-19  4:20 ` [added to the 4.1 stable tree] tools lib traceevent: Do not reassign parg after collapse_tree() Sasha Levin
2016-05-19  4:20 ` [added to the 4.1 stable tree] workqueue: fix rebind bound workers warning Sasha Levin
2016-05-19  4:20 ` [added to the 4.1 stable tree] drm/radeon: fix DP mode validation Sasha Levin
2016-05-19  4:20 ` [added to the 4.1 stable tree] ocfs2: fix SGID not inherited issue Sasha Levin
2016-05-19  4:20 ` [added to the 4.1 stable tree] ocfs2: revert using ocfs2_acl_chmod to avoid inode cluster lock hang Sasha Levin
2016-05-19  4:20 ` [added to the 4.1 stable tree] ocfs2: fix posix_acl_create deadlock Sasha Levin
2016-05-19  4:20 ` [added to the 4.1 stable tree] nf_conntrack: avoid kernel pointer value leak in slab name Sasha Levin

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=1463631606-32540-21-git-send-email-sasha.levin@oracle.com \
    --to=sasha.levin@oracle.com \
    --cc=akash.goel@intel.com \
    --cc=chris@chris-wilson.co.uk \
    --cc=jani.nikula@intel.com \
    --cc=stable-commits@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=ville.syrjala@linux.intel.com \
    /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