openwrt/staging/blogic.git
7 years agodrm/amdgpu:cleanup indent/format for gfx_v9_0.c
Monk Liu [Mon, 1 May 2017 09:00:13 +0000 (17:00 +0800)]
drm/amdgpu:cleanup indent/format for gfx_v9_0.c

Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: clean doorbell after sending init table to mmsch
Frank Min [Tue, 2 May 2017 11:49:32 +0000 (19:49 +0800)]
drm/amdgpu: clean doorbell after sending init table to mmsch

According to HW design, need to clean doorbell after setup MMSCH
table.

Signed-off-by: Frank Min <Frank.Min@amd.com>
Signed-off-by: Xiangliang Yu <Xiangliang.Yu@amd.com>
Reviewed-by: Monk Liu <Monk.Liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/virt: change AI ack-irq message to debug level
Xiangliang Yu [Thu, 4 May 2017 06:02:31 +0000 (14:02 +0800)]
drm/amdgpu/virt: change AI ack-irq message to debug level

Change message to debug level as VI does.

Signed-off-by: Xiangliang Yu <Xiangliang.Yu@amd.com>
Reviewed-by: Monk Liu <Monk.Liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/psp: Do not load asd for SRIOV
Xiangliang Yu [Thu, 4 May 2017 03:05:13 +0000 (11:05 +0800)]
drm/amdgpu/psp: Do not load asd for SRIOV

If psp version doesn't match asd version, asd loading will be
failed. Add workaround to bypass it for sriov.

Signed-off-by: Daniel Wang <Daniel.Wang2@amd.com>
Signed-off-by: Xiangliang Yu <Xiangliang.Yu@amd.com>
Reviewed-by: Monk Liu <Monk.Liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: Bypass GMC/UVD/VCE hw_fini in SR-IOV
Trigger Huang [Wed, 26 Apr 2017 06:29:47 +0000 (02:29 -0400)]
drm/amdgpu: Bypass GMC/UVD/VCE hw_fini in SR-IOV

On vega10, some hw finish operations should not be applied in SR-IOV
case. This works as workaround to fix multi-VFs reboot/shutdown
issues.

Signed-off-by: Trigger Huang <trigger.huang@amd.com>
Reviewed-by: Xiangliang Yu <Xiangliang.Yu@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu:re-write sriov_reinit_early/late (v2)
Monk Liu [Wed, 26 Apr 2017 04:00:49 +0000 (12:00 +0800)]
drm/amdgpu:re-write sriov_reinit_early/late (v2)

1,this way we make those routines compatible with the sequence
  requirment for both Tonga and Vega10
2,ignore PSP hw init when doing TDR, because for SR-IOV device
the ucode won't get lost after VF FLR, so no need to invoke PSP
doing the ucode reloading again.

v2: squash in ARRAY_SIZE fix

Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Xiangliang Yu <Xiangliang.Yu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu:need som change on vega10 mailbox
Monk Liu [Fri, 21 Apr 2017 11:35:11 +0000 (19:35 +0800)]
drm/amdgpu:need som change on vega10 mailbox

if sriov gpu reset is invoked by job timeout, it is run
in a global work-queue which is very slow and better not call
msleep ortherwise it takes long time to get back CPU.

so make below changes:

1: Change msleep 1 to mdelay 5
2: Ignore the ack fail from pf after time out,
   because VF FLR will clear ack, sometime VF FLR is done
   prior to the beginning of poll_ack so we can ignore this ack

TODO:
Put job_timedout (and the following gpu reset) in a driver thread,
instead of the global work_struct.

Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Xiangliang Yu <Xiangliang.Yu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu:fix cannot receive rcv/ack irq bug
Monk Liu [Sun, 23 Apr 2017 03:25:59 +0000 (11:25 +0800)]
drm/amdgpu:fix cannot receive rcv/ack irq bug

Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Xiangliang Yu <Xiangliang.Yu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu:kiq reg access need timeout(v2)
Monk Liu [Fri, 5 May 2017 21:30:50 +0000 (17:30 -0400)]
drm/amdgpu:kiq reg access need timeout(v2)

this is to prevent fence forever waiting if FLR occured
during register accessing.

v2:
use define instead of hardcode for the timeout msec

Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx9: wait for completion in KIQ init
Alex Deucher [Mon, 17 Apr 2017 20:21:52 +0000 (16:21 -0400)]
drm/amdgpu/gfx9: wait for completion in KIQ init

We need to make sure the various init sequences submitted
to KIQ complete before testing the rings.

Reviewed-by: monk liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx9: use new KIQ packet defines
Alex Deucher [Mon, 17 Apr 2017 20:14:09 +0000 (16:14 -0400)]
drm/amdgpu/gfx9: use new KIQ packet defines

Rather than magic numbers.

Reviewed-by: monk liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: add KIQ packet defines to soc15d.h
Alex Deucher [Mon, 17 Apr 2017 19:54:55 +0000 (15:54 -0400)]
drm/amdgpu: add KIQ packet defines to soc15d.h

Will be used in subsequent commits rather rather than
magic numbers.

Reviewed-by: monk liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx9: clear the compute ring on reset
Alex Deucher [Mon, 17 Apr 2017 20:32:12 +0000 (16:32 -0400)]
drm/amdgpu/gfx9: clear the compute ring on reset

To be consistent with gfx8.

Reviewed-by: monk liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx9: create mqd backups
Alex Deucher [Mon, 17 Apr 2017 19:29:14 +0000 (15:29 -0400)]
drm/amdgpu/gfx9: create mqd backups

And properly synchronize them with the master during
queue init.

Reviewed-by: monk liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: Move kiq ring lock out of virt structure
Shaoyun Liu [Fri, 28 Apr 2017 21:18:26 +0000 (17:18 -0400)]
drm/amdgpu: Move kiq ring lock out of virt structure

The usage of kiq should not depend on the virtualization.

Signed-off-by: Shaoyun Liu <Shaoyun.Liu@amd.com>
Reviewed-by:Andres Rodriquez <andresx7@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: bump module verion for reserved vmid
Chunming Zhou [Mon, 24 Apr 2017 03:47:05 +0000 (11:47 +0800)]
drm/amdgpu: bump module verion for reserved vmid

Interface to reserve a vmid for a specific process to
add in shader debugging that requries a fixed vmid.

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: implement grab reserved vmid V4
Chunming Zhou [Fri, 21 Apr 2017 03:13:56 +0000 (11:13 +0800)]
drm/amdgpu: implement grab reserved vmid V4

Implement the vmid reservation.

v2: move sync waiting only when flush needs
v3: fix racy
v4: peek fence instead of get fence, and fix potential context starved.

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: add limitation for dedicated vm number v4
Chunming Zhou [Fri, 21 Apr 2017 07:51:04 +0000 (15:51 +0800)]
drm/amdgpu: add limitation for dedicated vm number v4

Limit reserved vmids to 1 to avoid taking too many
out of commission and starving the system.

v2: move #define to amdgpu_vm.h
v3: move reserved vmid counter to id_manager,
and increase counter before allocating vmid
v4: rename to reserved_vmid_num

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: reserve/unreserve vmid by vm ioctl v4
Chunming Zhou [Thu, 20 Apr 2017 08:18:48 +0000 (16:18 +0800)]
drm/amdgpu: reserve/unreserve vmid by vm ioctl v4

add reserve/unreserve vmid funtions. Used to reserve
vmids for certain shader debugging functionality that
required a fixed vmid for the life of the debug.

v3:
only reserve vmid from gfxhub
v4:
fix racy condition

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: add reserved vmid field in vm struct v2
Chunming Zhou [Thu, 20 Apr 2017 08:17:34 +0000 (16:17 +0800)]
drm/amdgpu: add reserved vmid field in vm struct v2

v2: rename dedicated_vmid to reserved_vmid

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: add vm ioctl
Chunming Zhou [Mon, 24 Apr 2017 03:09:04 +0000 (11:09 +0800)]
drm/amdgpu: add vm ioctl

It will be used for reserving vmid for shader debugging
that requires a fixed vmid.

v2: fix warning (Alex)

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: Enable chained IB MCBP support
Trigger Huang [Thu, 27 Apr 2017 07:09:31 +0000 (03:09 -0400)]
drm/amdgpu: Enable chained IB MCBP support

Support for MCBP/Virtualization in combination with chained IBs is
formal released on firmware feature version #46. So enable it
according to firmware feature version, otherwise, world switch will
hang.

Signed-off-by: Trigger Huang <trigger.huang@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Reviewed-by: Xiangliang Yu <Xiangliang.Yu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu:fix get wrong gfx always on cu masks.
Rex Zhu [Wed, 26 Apr 2017 08:32:22 +0000 (16:32 +0800)]
drm/amdgpu:fix get wrong gfx always on cu masks.

Bug: SWDEV-117987: Always on CU mask broken for gfx7+

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: fix s3 ring test failed on Vi caused by KIQ enabled.
Rex Zhu [Mon, 24 Apr 2017 14:00:09 +0000 (22:00 +0800)]
drm/amdgpu: fix s3 ring test failed on Vi caused by KIQ enabled.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/virt: change the place of virt_init_setting
Xiangliang Yu [Fri, 21 Apr 2017 06:06:09 +0000 (14:06 +0800)]
drm/amdgpu/virt: change the place of virt_init_setting

Change place of virt_init_setting function so that can cover the
cg and pg flags configuration.

Signed-off-by: Xiangliang Yu <Xiangliang.Yu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/virt: bypass cg and pg setting for SRIOV
Xiangliang Yu [Fri, 21 Apr 2017 06:01:29 +0000 (14:01 +0800)]
drm/amdgpu/virt: bypass cg and pg setting for SRIOV

GPU hypervisor cover all settings of CG and PG, so guest doesn't
need to do anything. Bypass it.

Signed-off-by: Frank Min <Frank.Min@amd.com>
Signed-off-by: Xiangliang Yu <Xiangliang.Yu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: drop support for per ASIC read registers
Christian König [Wed, 12 Apr 2017 10:53:18 +0000 (12:53 +0200)]
drm/amdgpu: drop support for per ASIC read registers

Only per family registers are still used.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: drop support for untouched registers
Christian König [Wed, 12 Apr 2017 10:49:54 +0000 (12:49 +0200)]
drm/amdgpu: drop support for untouched registers

I couldn't figure out what this was original good for, but we
don't use it any more.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: delete redundant kiq irq funcs type check in gfx8.
Rex Zhu [Tue, 18 Apr 2017 04:59:41 +0000 (12:59 +0800)]
drm/amdgpu: delete redundant kiq irq funcs type check in gfx8.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: fix typo in dmesg in gfx_v8_0_kiq_kcq_disable.
Rex Zhu [Tue, 18 Apr 2017 05:00:43 +0000 (13:00 +0800)]
drm/amdgpu: fix typo in dmesg in gfx_v8_0_kiq_kcq_disable.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: add HDMI audio support for si dce6
Xiaojie Yuan [Tue, 21 Feb 2017 03:41:40 +0000 (11:41 +0800)]
drm/amdgpu: add HDMI audio support for si dce6

Signed-off-by: Xiaojie Yuan <Xiaojie.Yuan@amd.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: add DP audio support for si dce6 (v3)
Xiaojie Yuan [Sun, 19 Feb 2017 11:56:17 +0000 (19:56 +0800)]
drm/amdgpu: add DP audio support for si dce6 (v3)

v2: refine dce_v6_0_audio_endpt_wreg() and unify inconsistent method names
v3: fix num_pins for tahiti, pitcairn, verde and oland

Signed-off-by: Xiaojie Yuan <Xiaojie.Yuan@amd.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Junwei Zhang <Jerry.Zhang@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx8: move CP_PQ_STATUS after doorbell range setting (v2)
Alex Deucher [Thu, 13 Apr 2017 18:35:02 +0000 (14:35 -0400)]
drm/amdgpu/gfx8: move CP_PQ_STATUS after doorbell range setting (v2)

I'm not sure if the order matters, but it seems like it makes
more sense to set this after the range is programmed.

v2: rebase (Alex)

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: set cpg doorbell for fiji and polaris.
Rex Zhu [Tue, 11 Apr 2017 07:44:29 +0000 (15:44 +0800)]
drm/amdgpu: set cpg doorbell for fiji and polaris.

add set_doorbell functions for mec and cpg.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx8: unify the HQD deactivation code
Alex Deucher [Tue, 11 Apr 2017 21:37:01 +0000 (17:37 -0400)]
drm/amdgpu/gfx8: unify the HQD deactivation code

This could be used in Andres' priority scheduling patch
as well.

Reviewed-by: Andres Rodriguez <andresx7@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx8: enable cp/rlc ints after we disable clockgating
Alex Deucher [Tue, 14 Mar 2017 19:00:23 +0000 (15:00 -0400)]
drm/amdgpu/gfx8: enable cp/rlc ints after we disable clockgating

Even if we disable clockgating, we still need to make sure the
cp/rlc interrupts are enabled for powergating which might still
be enabled.

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx7: enable cp/rlc ints after we disable clockgating
Alex Deucher [Tue, 14 Mar 2017 19:04:59 +0000 (15:04 -0400)]
drm/amdgpu/gfx7: enable cp/rlc ints after we disable clockgating

Even if we disable clockgating, we still need to make sure the
cp/rlc interrupts are enabled for powergating which might still
be enabled.

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx8: move MEC doorbell range setting
Alex Deucher [Mon, 10 Apr 2017 17:00:14 +0000 (13:00 -0400)]
drm/amdgpu/gfx8: move MEC doorbell range setting

It's global, not queue specific, so move it out of the
kiq register init function.

Tested-and-Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx8: fix resume of KIQ and KCQs
Alex Deucher [Fri, 31 Mar 2017 16:00:14 +0000 (12:00 -0400)]
drm/amdgpu/gfx8: fix resume of KIQ and KCQs

No need to reset the wptr and clear the rings.  The UNMAP_QUEUES
packet writes the current MQD state back the MQD on suspend,
so there is no need to reset it as well.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx8: properly disable the KCQs in hw_fini
Alex Deucher [Thu, 30 Mar 2017 17:06:32 +0000 (13:06 -0400)]
drm/amdgpu/gfx8: properly disable the KCQs in hw_fini

Use the UNMAP_QUEUES packet to have the KIQ properly
disable them.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx8: use new KIQ packet defines
Alex Deucher [Thu, 30 Mar 2017 16:34:06 +0000 (12:34 -0400)]
drm/amdgpu/gfx8: use new KIQ packet defines

Rather than open coding them.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx8: move SET_RESOURCES into the same command stream
Alex Deucher [Wed, 29 Mar 2017 19:31:22 +0000 (15:31 -0400)]
drm/amdgpu/gfx8: move SET_RESOURCES into the same command stream

As the KCQ setup.  This way we only have to wait once for the
entire MEC.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx8: wait once for all KCQs to be created
Alex Deucher [Wed, 29 Mar 2017 17:42:32 +0000 (13:42 -0400)]
drm/amdgpu/gfx8: wait once for all KCQs to be created

Rather than waiting for each queue.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: split gfx_v8_0_kiq_init_queue into two
Alex Deucher [Wed, 29 Mar 2017 17:23:44 +0000 (13:23 -0400)]
drm/amdgpu: split gfx_v8_0_kiq_init_queue into two

One for KIQ and one for the KCQ. This simplifies the logic and
allows for future optimizations.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx8: wait for completion in KIQ init
Alex Deucher [Wed, 29 Mar 2017 17:15:22 +0000 (13:15 -0400)]
drm/amdgpu/gfx8: wait for completion in KIQ init

We need to make sure the various init sequences submitted
to KIQ complete before testing the rings.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agoRevert "drm/amd/amdgpu: Disable GFX_PG on Carrizo until compute issues solved"
Alex Deucher [Fri, 17 Mar 2017 19:27:06 +0000 (15:27 -0400)]
Revert "drm/amd/amdgpu: Disable GFX_PG on Carrizo until compute issues solved"

Re-enable GFX PG.  It's working properly with MEC now that KIQ is
enabled.

Reviewed-by: Samuel Li <samuel.li@amd.com>
This reverts commit e9ef19aa1bdeac380662a112f1d03a7c3477527f.

7 years agodrm/amdgpu: Switch baremetal to use KIQ for compute ring management. (v3)
David Panariti [Tue, 28 Mar 2017 16:57:31 +0000 (12:57 -0400)]
drm/amdgpu: Switch baremetal to use KIQ for compute ring management. (v3)

KIQ is the Kernel Interface Queue for managing the MEC.  Rather than setting
up rings via direct MMIO of ring registers, the rings are configured via
special packets sent to the KIQ.  The allows the MEC to better manage shared
resources and certain power events.

v2: squash in s3/s4 fix from Rex
v3: further fixes from Rex

Signed-off-by: David Panariti <David.Panariti@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Tom St Denis <tom.stdenis@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx8: set doorbell range for polaris as well
Alex Deucher [Mon, 10 Apr 2017 16:48:02 +0000 (12:48 -0400)]
drm/amdgpu/gfx8: set doorbell range for polaris as well

Add missing chips to the doorbell range setup.  These
were missed in the KIQ code.  Fixes power and performance
regressions with KIQ.  Spotted by Rex.

Tested-and-Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx8: add additional MQD initialization
Alex Deucher [Fri, 7 Apr 2017 21:43:05 +0000 (17:43 -0400)]
drm/amdgpu/gfx8: add additional MQD initialization

Need to properly set the MTYPE and ROQ space setting.
This should fix performance regressions with KIQ enabled.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amd/powerplay: fix pcie dpm table for vega10
Eric Huang [Thu, 11 May 2017 15:55:48 +0000 (11:55 -0400)]
drm/amd/powerplay: fix pcie dpm table for vega10

This resolves pcie low speed problem.

Signed-off-by: Eric Huang <JinHuiEric.Huang@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amd/powerplay: update vega10 smu interface version to E.
Rex Zhu [Wed, 10 May 2017 08:33:03 +0000 (16:33 +0800)]
drm/amd/powerplay: update vega10 smu interface version to E.

need update smu firmware to version 0x1c20.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewws-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amd/powerplay: delete dead code in vega10_thermal.c
Rex Zhu [Thu, 11 May 2017 06:05:27 +0000 (14:05 +0800)]
drm/amd/powerplay: delete dead code in vega10_thermal.c

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amd/powerplay: Add Vega10 Powertune Table v3 support.
Rex Zhu [Wed, 10 May 2017 13:51:24 +0000 (21:51 +0800)]
drm/amd/powerplay: Add Vega10 Powertune Table v3 support.

Handle the latest powerplay table format; includes Boost
State support.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewws-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amd/powerplay: convert from number of lanes to lane bits on vega10
Rex Zhu [Wed, 10 May 2017 08:18:34 +0000 (16:18 +0800)]
drm/amd/powerplay: convert from number of lanes to lane bits on vega10

We need a mask.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewws-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amd/powerplay: fix bug in processing CKS_Enable bit.
Rex Zhu [Wed, 10 May 2017 07:51:09 +0000 (15:51 +0800)]
drm/amd/powerplay: fix bug in processing CKS_Enable bit.

Typo in the mask.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amd/powerplay: add avfs fuse overdriver func.
Rex Zhu [Thu, 11 May 2017 20:12:21 +0000 (16:12 -0400)]
drm/amd/powerplay: add avfs fuse overdriver func.

Add a function to look up the AVFS fuse values for vega10
These are used to populate the avfs fuse table in the smu.

Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amd/powerplay: add power profile support for Vega10 (v2)
Eric Huang [Mon, 8 May 2017 19:11:41 +0000 (15:11 -0400)]
drm/amd/powerplay: add power profile support for Vega10 (v2)

This implements the workload specific interface of optimized
compute power profile for Vega10.

v2: squash in fix (Tom)

Signed-off-by: Eric Huang <JinHuiEric.Huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/gfx9: drop duplicate gfx info init (v3)
Alex Deucher [Tue, 2 May 2017 21:28:45 +0000 (17:28 -0400)]
drm/amdgpu/gfx9: drop duplicate gfx info init (v3)

Taken care of by gpu info firmware now.

v2: rebase
v3: rework based on latest firmware

Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Tested-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: parse the gpu_info firmware (v4)
Alex Deucher [Thu, 27 Apr 2017 20:58:01 +0000 (16:58 -0400)]
drm/amdgpu: parse the gpu_info firmware (v4)

And populate the gfx structures from it.

v2: update the structures updated by the table
v3: rework based on new table structure
v4: simplify things

Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Tested-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: add gpu_info firmware (v3)
Alex Deucher [Thu, 27 Apr 2017 03:40:37 +0000 (23:40 -0400)]
drm/amdgpu: add gpu_info firmware (v3)

Add a new gpu info firmware to store gpu specific configuration
data.  This allows us to store hw constants in a unified place.

v2: adjust structure and elements
v3: further restructure

Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>
Tested-by: Junwei Zhang <Jerry.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amd/powerplay: fix a signedness bugs
Dan Carpenter [Tue, 23 May 2017 05:13:45 +0000 (13:13 +0800)]
drm/amd/powerplay: fix a signedness bugs

Smatch complains about a signedness bug here:

        vega10_hwmgr.c:4202 vega10_force_clock_level()
        warn: always true condition '(i >= 0) => (0-u32max >= 0)'

Fixes: 7b52db39a4c2 ("drm/amd/powerplay: fix bug sclk/mclk
                     level can't be set on vega10.")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Eric Huang <JinHuiEric.Huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: fix NULL pointer panic of emit_gds_switch
Chunming Zhou [Thu, 11 May 2017 10:22:17 +0000 (18:22 +0800)]
drm/amdgpu: fix NULL pointer panic of emit_gds_switch

[  338.384770] BUG: unable to handle kernel NULL pointer dereference at           (null)
[  338.384817] IP: [<          (null)>]           (null)
[  338.385505] RIP: 0010:[<0000000000000000>]  [<          (null)>]           (null)
[  338.385950] Call Trace:
[  338.385993]  [<ffffffffa05d2313>] ? amdgpu_vm_flush+0x283/0x400 [amdgpu]
[  338.386025]  [<ffffffff811818d3>] ? printk+0x4d/0x4f
[  338.386074]  [<ffffffffa05d4906>] amdgpu_ib_schedule+0x4a6/0x4d0 [amdgpu]
[  338.386140]  [<ffffffffa0673e54>] amdgpu_job_run+0x64/0x180 [amdgpu]
[  338.386203]  [<ffffffffa0672e09>] amd_sched_main+0x2e9/0x4a0 [amdgpu]
[  338.386232]  [<ffffffff810bfce0>] ? prepare_to_wait_event+0x110/0x110
[  338.386295]  [<ffffffffa0672b20>] ? amd_sched_select_entity+0xe0/0xe0 [amdgpu]
[  338.386327]  [<ffffffff8109b423>] kthread+0xd3/0xf0
[  338.386349]  [<ffffffff8109b350>] ? kthread_park+0x60/0x60
[  338.386376]  [<ffffffff817e1ee5>] ret_from_fork+0x25/0x30
[  338.386401] Code:  Bad RIP value.
[  338.386420] RIP  [<          (null)>]           (null)
[  338.386443]  RSP <ffffc90001bd7d40>
[  338.386458] CR2: 0000000000000000
[  338.398508] ---[ end trace 4c66fcdc74b9a0a2 ]---

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/radeon: Unbreak HPD handling for r600+
Lyude [Thu, 11 May 2017 23:31:12 +0000 (19:31 -0400)]
drm/radeon: Unbreak HPD handling for r600+

We end up reading the interrupt register for HPD5, and then writing it
to HPD6 which on systems without anything using HPD5 results in
permanently disabling hotplug on one of the display outputs after the
first time we acknowledge a hotplug interrupt from the GPU.

This code is really bad. But for now, let's just fix this. I will
hopefully have a large patch series to refactor all of this soon.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Lyude <lyude@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amd/powerplay/smu7: disable mclk switching for high refresh rates
Alex Deucher [Thu, 11 May 2017 17:57:41 +0000 (13:57 -0400)]
drm/amd/powerplay/smu7: disable mclk switching for high refresh rates

Even if the vblank period would allow it, it still seems to
be problematic on some cards.

bug: https://bugs.freedesktop.org/show_bug.cgi?id=96868

Cc: stable@vger.kernel.org
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amd/powerplay/smu7: add vblank check for mclk switching (v2)
Alex Deucher [Thu, 11 May 2017 17:46:12 +0000 (13:46 -0400)]
drm/amd/powerplay/smu7: add vblank check for mclk switching (v2)

Check to make sure the vblank period is long enough to support
mclk switching.

v2: drop needless initial assignment (Nils)

bug: https://bugs.freedesktop.org/show_bug.cgi?id=96868

Cc: stable@vger.kernel.org
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/radeon/ci: disable mclk switching for high refresh rates (v2)
Alex Deucher [Thu, 11 May 2017 17:14:14 +0000 (13:14 -0400)]
drm/radeon/ci: disable mclk switching for high refresh rates (v2)

Even if the vblank period would allow it, it still seems to
be problematic on some cards.

v2: fix logic inversion (Nils)

bug: https://bugs.freedesktop.org/show_bug.cgi?id=96868

Cc: stable@vger.kernel.org
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu/ci: disable mclk switching for high refresh rates (v2)
Alex Deucher [Thu, 11 May 2017 17:10:02 +0000 (13:10 -0400)]
drm/amdgpu/ci: disable mclk switching for high refresh rates (v2)

Even if the vblank period would allow it, it still seems to
be problematic on some cards.

v2: fix logic inversion (Nils)

bug: https://bugs.freedesktop.org/show_bug.cgi?id=96868

Cc: stable@vger.kernel.org
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
7 years agodrm/amdgpu: fix fundamental suspend/resume issue
Christian König [Wed, 10 May 2017 18:06:58 +0000 (20:06 +0200)]
drm/amdgpu: fix fundamental suspend/resume issue

Reinitializing the VM manager during suspend/resume is a very very bad
idea since all the VMs are still active and kicking.

This can lead to random VM faults after resume when new processes
become the same client ID assigned.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
7 years agoMerge tag 'drm-misc-next-2017-05-16' of git://anongit.freedesktop.org/git/drm-misc...
Dave Airlie [Thu, 18 May 2017 02:57:06 +0000 (12:57 +1000)]
Merge tag 'drm-misc-next-2017-05-16' of git://anongit.freedesktop.org/git/drm-misc into drm-next

UAPI Changes:
- Return -ENODEV instead of -ENXIO when creating cma fb w/o valid gem (Daniel)
- Add aspect ratio and custom scaling propertis to connector state (Maarten)

Cross-subsystem Changes:
- None

Core Changes:
- Add Laurent as bridge reviewer and Andrzej as bridge maintainer (Archit)
- Maintain new STM driver through -misc (Yannick)
- Misc doc improvements (as is tradition) (Daniel)
- Add driver-private objects to atomic state (Dhinakaran)
- Deprecate preclose hook in modern drivers (use postclose) (Daniel)
- Add hwmode to vblank struct. This fixes mode access in irq context and reduced
  a bunch of boilerplate (Daniel)

Driver Changes:
- vc4: Add out-fence support to vc4 V3D rendering (Eric)
- stm: Add stm32f429 display hw and am-480272h3tmqw-t01h panel support (Yannick)
- vc4: Remove 256MB cma limit from vc4 (Eric)
- dw-hdmi: Disable audio when inactive, instead of always enabled (Romain)
- zte: Add support for VGA to the ZTE driver (Shawn)
- i915: Track DP MST bandwidth and check it in atomic_check (Dhinakaran)
- vgem: Enable gem dmabuf import iface to facilitate ion testing (Laura)
- vc4: Add support for Cygnus (new dt compat string + couple bug fixes) (Eric)
- pl111: Add driver for pl111 CLCD display controller (Eric/Tom)
- vgem: Subclass drm_device instead of standalone platform device (Chris)

Cc: Archit Taneja <architt@codeaurora.org>
Cc: Eric Anholt <eric@anholt.net>
Cc: Yannick Fertre <yannick.fertre@st.com>
Cc: Romain Perier <romain.perier@collabora.com>
Cc: Navare, Manasi D <manasi.d.navare@intel.com>
Cc: Shawn Guo <shawn.guo@linaro.org>
Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Cc: Laura Abbott <labbott@redhat.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Tom Cooksey <tom.cooksey@arm.com>
Cc: Daniel Vetter <daniel.vetter@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
* tag 'drm-misc-next-2017-05-16' of git://anongit.freedesktop.org/git/drm-misc: (72 commits)
  drm: add missing declaration to drm_blend.h
  drm/dp: Wait up all outstanding tx waiters
  drm/dp: Read the tx msg state once after checking for an event
  drm/prime: Forward declare struct device
  drm/vblank: Lock down vblank->hwmode more
  drm/vblank: drop the mode argument from drm_calc_vbltimestamp_from_scanoutpos
  drm/vblank: Add FIXME comments about moving the vblank ts hooks
  drm/vblank: Switch to bool in_vblank_irq in get_vblank_timestamp
  drm/vblank: Switch drm_driver->get_vblank_timestamp to return a bool
  drm/vgem: Convert to a struct drm_device subclass
  gpu: drm: gma500: remove dead code
  drm/sti: Adjust two checks for null pointers in sti_hqvdp_probe()
  drm/sti: Fix typos in a comment line
  drm/sti: Fix a typo in a comment line
  drm/sti: Replace 17 seq_puts() calls by seq_putc()
  drm/sti: Reduce function calls for sequence output at five places
  drm/sti: use seq_puts to display a string
  drm: Nerf the preclose callback for modern drivers
  drm/exynos: Merge pre/postclose hooks
  drm/tegra: switch to postclose
  ...

7 years agodrm: add missing declaration to drm_blend.h
Tomi Valkeinen [Tue, 16 May 2017 12:34:45 +0000 (15:34 +0300)]
drm: add missing declaration to drm_blend.h

drm_blend.h is missing declaration for 'struct drm_plane'. Add it.

Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1494938085-21805-1-git-send-email-tomi.valkeinen@ti.com
7 years agodrm/dp: Wait up all outstanding tx waiters
Chris Wilson [Sat, 13 May 2017 10:52:01 +0000 (11:52 +0100)]
drm/dp: Wait up all outstanding tx waiters

As we can have multiple tx in the queue, with individual waiters, make
sure that all are woken when any state changes (so that we are sure the
right owner of the txmsg is woken).

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20170513105201.17658-2-chris@chris-wilson.co.uk
7 years agodrm/dp: Read the tx msg state once after checking for an event
Chris Wilson [Sat, 13 May 2017 10:52:00 +0000 (11:52 +0100)]
drm/dp: Read the tx msg state once after checking for an event

Both as an exercise to document that we are reading the state outside of
the appropriate mutex and to ensure that we only read the value once
before the multiple comparisons, use READ_ONCE.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20170513105201.17658-1-chris@chris-wilson.co.uk
7 years agoLinux 4.12-rc1
Linus Torvalds [Sat, 13 May 2017 20:19:49 +0000 (13:19 -0700)]
Linux 4.12-rc1

7 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
Linus Torvalds [Sat, 13 May 2017 17:25:05 +0000 (10:25 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/dtor/input

Pull some more input subsystem updates from Dmitry Torokhov:
 "An updated xpad driver with a few more recognized device IDs, and a
  new psxpad-spi driver, allowing connecting Playstation 1 and 2 joypads
  via SPI bus"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
  Input: cros_ec_keyb - remove extraneous 'const'
  Input: add support for PlayStation 1/2 joypads connected via SPI
  Input: xpad - add USB IDs for Mad Catz Brawlstick and Razer Sabertooth
  Input: xpad - sync supported devices with xboxdrv
  Input: xpad - sort supported devices by USB ID

7 years agoMerge tag 'upstream-4.12-rc1' of git://git.infradead.org/linux-ubifs
Linus Torvalds [Sat, 13 May 2017 17:23:12 +0000 (10:23 -0700)]
Merge tag 'upstream-4.12-rc1' of git://git.infradead.org/linux-ubifs

Pull UBI/UBIFS updates from Richard Weinberger:

 - new config option CONFIG_UBIFS_FS_SECURITY

 - minor improvements

 - random fixes

* tag 'upstream-4.12-rc1' of git://git.infradead.org/linux-ubifs:
  ubi: Add debugfs file for tracking PEB state
  ubifs: Fix a typo in comment of ioctl2ubifs & ubifs2ioctl
  ubifs: Remove unnecessary assignment
  ubifs: Fix cut and paste error on sb type comparisons
  ubi: fastmap: Fix slab corruption
  ubifs: Add CONFIG_UBIFS_FS_SECURITY to disable/enable security labels
  ubi: Make mtd parameter readable
  ubi: Fix section mismatch

7 years agoMerge branch 'for-linus-4.12-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 13 May 2017 17:20:02 +0000 (10:20 -0700)]
Merge branch 'for-linus-4.12-rc1' of git://git./linux/kernel/git/rw/uml

Pull UML fixes from Richard Weinberger:
 "No new stuff, just fixes"

* 'for-linus-4.12-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml:
  um: Add missing NR_CPUS include
  um: Fix to call read_initrd after init_bootmem
  um: Include kbuild.h instead of duplicating its macros
  um: Fix PTRACE_POKEUSER on x86_64
  um: Set number of CPUs
  um: Fix _print_addr()

7 years agoMerge branch 'akpm' (patches from Andrew)
Linus Torvalds [Sat, 13 May 2017 16:49:35 +0000 (09:49 -0700)]
Merge branch 'akpm' (patches from Andrew)

Merge misc fixes from Andrew Morton:
 "15 fixes"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
  mm, docs: update memory.stat description with workingset* entries
  mm: vmscan: scan until it finds eligible pages
  mm, thp: copying user pages must schedule on collapse
  dax: fix PMD data corruption when fault races with write
  dax: fix data corruption when fault races with write
  ext4: return to starting transaction in ext4_dax_huge_fault()
  mm: fix data corruption due to stale mmap reads
  dax: prevent invalidation of mapped DAX entries
  Tigran has moved
  mm, vmalloc: fix vmalloc users tracking properly
  mm/khugepaged: add missed tracepoint for collapse_huge_page_swapin
  gcov: support GCC 7.1
  mm, vmstat: Remove spurious WARN() during zoneinfo print
  time: delete current_fs_time()
  hwpoison, memcg: forcibly uncharge LRU pages

7 years agomm, docs: update memory.stat description with workingset* entries
Roman Gushchin [Fri, 12 May 2017 22:47:09 +0000 (15:47 -0700)]
mm, docs: update memory.stat description with workingset* entries

Commit 4b4cea91691d ("mm: vmscan: fix IO/refault regression in cache
workingset transition") introduced three new entries in memory stat
file:

 - workingset_refault
 - workingset_activate
 - workingset_nodereclaim

This commit adds a corresponding description to the cgroup v2 docs.

Link: http://lkml.kernel.org/r/1494530293-31236-1-git-send-email-guro@fb.com
Signed-off-by: Roman Gushchin <guro@fb.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Li Zefan <lizefan@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agomm: vmscan: scan until it finds eligible pages
Minchan Kim [Fri, 12 May 2017 22:47:06 +0000 (15:47 -0700)]
mm: vmscan: scan until it finds eligible pages

Although there are a ton of free swap and anonymous LRU page in elgible
zones, OOM happened.

  balloon invoked oom-killer: gfp_mask=0x17080c0(GFP_KERNEL_ACCOUNT|__GFP_ZERO|__GFP_NOTRACK), nodemask=(null),  order=0, oom_score_adj=0
  CPU: 7 PID: 1138 Comm: balloon Not tainted 4.11.0-rc6-mm1-zram-00289-ge228d67e9677-dirty #17
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Ubuntu-1.8.2-1ubuntu1 04/01/2014
  Call Trace:
   oom_kill_process+0x21d/0x3f0
   out_of_memory+0xd8/0x390
   __alloc_pages_slowpath+0xbc1/0xc50
   __alloc_pages_nodemask+0x1a5/0x1c0
   pte_alloc_one+0x20/0x50
   __pte_alloc+0x1e/0x110
   __handle_mm_fault+0x919/0x960
   handle_mm_fault+0x77/0x120
   __do_page_fault+0x27a/0x550
   trace_do_page_fault+0x43/0x150
   do_async_page_fault+0x2c/0x90
   async_page_fault+0x28/0x30
  Mem-Info:
  active_anon:424716 inactive_anon:65314 isolated_anon:0
   active_file:52 inactive_file:46 isolated_file:0
   unevictable:0 dirty:27 writeback:0 unstable:0
   slab_reclaimable:3967 slab_unreclaimable:4125
   mapped:133 shmem:43 pagetables:1674 bounce:0
   free:4637 free_pcp:225 free_cma:0
  Node 0 active_anon:1698864kB inactive_anon:261256kB active_file:208kB inactive_file:184kB unevictable:0kB isolated(anon):0kB isolated(file):0kB mapped:532kB dirty:108kB writeback:0kB shmem:172kB writeback_tmp:0kB unstable:0kB all_unreclaimable? no
  DMA free:7316kB min:32kB low:44kB high:56kB active_anon:8064kB inactive_anon:0kB active_file:0kB inactive_file:0kB unevictable:0kB writepending:0kB present:15992kB managed:15908kB mlocked:0kB slab_reclaimable:464kB slab_unreclaimable:40kB kernel_stack:0kB pagetables:24kB bounce:0kB free_pcp:0kB local_pcp:0kB free_cma:0kB
  lowmem_reserve[]: 0 992 992 1952
  DMA32 free:9088kB min:2048kB low:3064kB high:4080kB active_anon:952176kB inactive_anon:0kB active_file:36kB inactive_file:0kB unevictable:0kB writepending:88kB present:1032192kB managed:1019388kB mlocked:0kB slab_reclaimable:13532kB slab_unreclaimable:16460kB kernel_stack:3552kB pagetables:6672kB bounce:0kB free_pcp:56kB local_pcp:24kB free_cma:0kB
  lowmem_reserve[]: 0 0 0 959
  Movable free:3644kB min:1980kB low:2960kB high:3940kB active_anon:738560kB inactive_anon:261340kB active_file:188kB inactive_file:640kB unevictable:0kB writepending:20kB present:1048444kB managed:1010816kB mlocked:0kB slab_reclaimable:0kB slab_unreclaimable:0kB kernel_stack:0kB pagetables:0kB bounce:0kB free_pcp:832kB local_pcp:60kB free_cma:0kB
  lowmem_reserve[]: 0 0 0 0
  DMA: 1*4kB (E) 0*8kB 18*16kB (E) 10*32kB (E) 10*64kB (E) 9*128kB (ME) 8*256kB (E) 2*512kB (E) 2*1024kB (E) 0*2048kB 0*4096kB = 7524kB
  DMA32: 417*4kB (UMEH) 181*8kB (UMEH) 68*16kB (UMEH) 48*32kB (UMEH) 14*64kB (MH) 3*128kB (M) 1*256kB (H) 1*512kB (M) 2*1024kB (M) 0*2048kB 0*4096kB = 9836kB
  Movable: 1*4kB (M) 1*8kB (M) 1*16kB (M) 1*32kB (M) 0*64kB 1*128kB (M) 2*256kB (M) 4*512kB (M) 1*1024kB (M) 0*2048kB 0*4096kB = 3772kB
  378 total pagecache pages
  17 pages in swap cache
  Swap cache stats: add 17325, delete 17302, find 0/27
  Free swap  = 978940kB
  Total swap = 1048572kB
  524157 pages RAM
  0 pages HighMem/MovableOnly
  12629 pages reserved
  0 pages cma reserved
  0 pages hwpoisoned
  [ pid ]   uid  tgid total_vm      rss nr_ptes nr_pmds swapents oom_score_adj name
  [  433]     0   433     4904        5      14       3       82             0 upstart-udev-br
  [  438]     0   438    12371        5      27       3      191         -1000 systemd-udevd

With investigation, skipping page of isolate_lru_pages makes reclaim
void because it returns zero nr_taken easily so LRU shrinking is
effectively nothing and just increases priority aggressively.  Finally,
OOM happens.

The problem is that get_scan_count determines nr_to_scan with eligible
zones so although priority drops to zero, it couldn't reclaim any pages
if the LRU contains mostly ineligible pages.

get_scan_count:

        size = lruvec_lru_size(lruvec, lru, sc->reclaim_idx);
size = size >> sc->priority;

Assumes sc->priority is 0 and LRU list is as follows.

N-N-N-N-H-H-H-H-H-H-H-H-H-H-H-H-H-H-H-H

(Ie, small eligible pages are in the head of LRU but others are
 almost ineligible pages)

In that case, size becomes 4 so VM want to scan 4 pages but 4 pages from
tail of the LRU are not eligible pages.  If get_scan_count counts
skipped pages, it doesn't reclaim any pages remained after scanning 4
pages so it ends up OOM happening.

This patch makes isolate_lru_pages try to scan pages until it encounters
eligible zones's pages.

[akpm@linux-foundation.org: clean up mind-bending `for' statement.  Tweak comment text]
Fixes: 3db65812d688 ("Revert "mm, vmscan: account for skipped pages as a partial scan"")
Link: http://lkml.kernel.org/r/1494457232-27401-1-git-send-email-minchan@kernel.org
Signed-off-by: Minchan Kim <minchan@kernel.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agomm, thp: copying user pages must schedule on collapse
David Rientjes [Fri, 12 May 2017 22:47:03 +0000 (15:47 -0700)]
mm, thp: copying user pages must schedule on collapse

We have encountered need_resched warnings in __collapse_huge_page_copy()
while doing {clear,copy}_user_highpage() over HPAGE_PMD_NR source pages.

mm->mmap_sem is held for write, but the iteration is well bounded.

Reschedule as needed.

Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1705101426380.109808@chino.kir.corp.google.com
Signed-off-by: David Rientjes <rientjes@google.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agodax: fix PMD data corruption when fault races with write
Ross Zwisler [Fri, 12 May 2017 22:47:00 +0000 (15:47 -0700)]
dax: fix PMD data corruption when fault races with write

This is based on a patch from Jan Kara that fixed the equivalent race in
the DAX PTE fault path.

Currently DAX PMD read fault can race with write(2) in the following
way:

CPU1 - write(2)                 CPU2 - read fault
                                dax_iomap_pmd_fault()
                                  ->iomap_begin() - sees hole

dax_iomap_rw()
  iomap_apply()
    ->iomap_begin - allocates blocks
    dax_iomap_actor()
      invalidate_inode_pages2_range()
        - there's nothing to invalidate

                                  grab_mapping_entry()
  - we add huge zero page to the radix tree
    and map it to page tables

The result is that hole page is mapped into page tables (and thus zeros
are seen in mmap) while file has data written in that place.

Fix the problem by locking exception entry before mapping blocks for the
fault.  That way we are sure invalidate_inode_pages2_range() call for
racing write will either block on entry lock waiting for the fault to
finish (and unmap stale page tables after that) or read fault will see
already allocated blocks by write(2).

Fixes: 9f141d6ef6258 ("dax: Call ->iomap_begin without entry lock during dax fault")
Link: http://lkml.kernel.org/r/20170510172700.18991-1-ross.zwisler@linux.intel.com
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agodax: fix data corruption when fault races with write
Jan Kara [Fri, 12 May 2017 22:46:57 +0000 (15:46 -0700)]
dax: fix data corruption when fault races with write

Currently DAX read fault can race with write(2) in the following way:

CPU1 - write(2) CPU2 - read fault
dax_iomap_pte_fault()
  ->iomap_begin() - sees hole
dax_iomap_rw()
  iomap_apply()
    ->iomap_begin - allocates blocks
    dax_iomap_actor()
      invalidate_inode_pages2_range()
        - there's nothing to invalidate
  grab_mapping_entry()
  - we add zero page in the radix tree
    and map it to page tables

The result is that hole page is mapped into page tables (and thus zeros
are seen in mmap) while file has data written in that place.

Fix the problem by locking exception entry before mapping blocks for the
fault.  That way we are sure invalidate_inode_pages2_range() call for
racing write will either block on entry lock waiting for the fault to
finish (and unmap stale page tables after that) or read fault will see
already allocated blocks by write(2).

Fixes: 9f141d6ef6258a3a37a045842d9ba7e68f368956
Link: http://lkml.kernel.org/r/20170510085419.27601-5-jack@suse.cz
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agoext4: return to starting transaction in ext4_dax_huge_fault()
Jan Kara [Fri, 12 May 2017 22:46:54 +0000 (15:46 -0700)]
ext4: return to starting transaction in ext4_dax_huge_fault()

DAX will return to locking exceptional entry before mapping blocks for a
page fault to fix possible races with concurrent writes.  To avoid lock
inversion between exceptional entry lock and transaction start, start
the transaction already in ext4_dax_huge_fault().

Fixes: 9f141d6ef6258a3a37a045842d9ba7e68f368956
Link: http://lkml.kernel.org/r/20170510085419.27601-4-jack@suse.cz
Signed-off-by: Jan Kara <jack@suse.cz>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agomm: fix data corruption due to stale mmap reads
Jan Kara [Fri, 12 May 2017 22:46:50 +0000 (15:46 -0700)]
mm: fix data corruption due to stale mmap reads

Currently, we didn't invalidate page tables during invalidate_inode_pages2()
for DAX.  That could result in e.g. 2MiB zero page being mapped into
page tables while there were already underlying blocks allocated and
thus data seen through mmap were different from data seen by read(2).
The following sequence reproduces the problem:

 - open an mmap over a 2MiB hole

 - read from a 2MiB hole, faulting in a 2MiB zero page

 - write to the hole with write(3p). The write succeeds but we
   incorrectly leave the 2MiB zero page mapping intact.

 - via the mmap, read the data that was just written. Since the zero
   page mapping is still intact we read back zeroes instead of the new
   data.

Fix the problem by unconditionally calling invalidate_inode_pages2_range()
in dax_iomap_actor() for new block allocations and by properly
invalidating page tables in invalidate_inode_pages2_range() for DAX
mappings.

Fixes: c6dcf52c23d2d3fb5235cec42d7dd3f786b87d55
Link: http://lkml.kernel.org/r/20170510085419.27601-3-jack@suse.cz
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agodax: prevent invalidation of mapped DAX entries
Ross Zwisler [Fri, 12 May 2017 22:46:47 +0000 (15:46 -0700)]
dax: prevent invalidation of mapped DAX entries

Patch series "mm,dax: Fix data corruption due to mmap inconsistency",
v4.

This series fixes data corruption that can happen for DAX mounts when
page faults race with write(2) and as a result page tables get out of
sync with block mappings in the filesystem and thus data seen through
mmap is different from data seen through read(2).

The series passes testing with t_mmap_stale test program from Ross and
also other mmap related tests on DAX filesystem.

This patch (of 4):

dax_invalidate_mapping_entry() currently removes DAX exceptional entries
only if they are clean and unlocked.  This is done via:

  invalidate_mapping_pages()
    invalidate_exceptional_entry()
      dax_invalidate_mapping_entry()

However, for page cache pages removed in invalidate_mapping_pages()
there is an additional criteria which is that the page must not be
mapped.  This is noted in the comments above invalidate_mapping_pages()
and is checked in invalidate_inode_page().

For DAX entries this means that we can can end up in a situation where a
DAX exceptional entry, either a huge zero page or a regular DAX entry,
could end up mapped but without an associated radix tree entry.  This is
inconsistent with the rest of the DAX code and with what happens in the
page cache case.

We aren't able to unmap the DAX exceptional entry because according to
its comments invalidate_mapping_pages() isn't allowed to block, and
unmap_mapping_range() takes a write lock on the mapping->i_mmap_rwsem.

Since we essentially never have unmapped DAX entries to evict from the
radix tree, just remove dax_invalidate_mapping_entry().

Fixes: c6dcf52c23d2 ("mm: Invalidate DAX radix tree entries only if appropriate")
Link: http://lkml.kernel.org/r/20170510085419.27601-2-jack@suse.cz
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Reported-by: Jan Kara <jack@suse.cz>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: <stable@vger.kernel.org> [4.10+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agoTigran has moved
Andrew Morton [Fri, 12 May 2017 22:46:44 +0000 (15:46 -0700)]
Tigran has moved

Cc: Tigran Aivazian <aivazian.tigran@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agomm, vmalloc: fix vmalloc users tracking properly
Michal Hocko [Fri, 12 May 2017 22:46:41 +0000 (15:46 -0700)]
mm, vmalloc: fix vmalloc users tracking properly

Commit 1f5307b1e094 ("mm, vmalloc: properly track vmalloc users") has
pulled asm/pgtable.h include dependency to linux/vmalloc.h and that
turned out to be a bad idea for some architectures.  E.g.  m68k fails
with

   In file included from arch/m68k/include/asm/pgtable_mm.h:145:0,
                    from arch/m68k/include/asm/pgtable.h:4,
                    from include/linux/vmalloc.h:9,
                    from arch/m68k/kernel/module.c:9:
   arch/m68k/include/asm/mcf_pgtable.h: In function 'nocache_page':
>> arch/m68k/include/asm/mcf_pgtable.h:339:43: error: 'init_mm' undeclared (first use in this function)
    #define pgd_offset_k(address) pgd_offset(&init_mm, address)

as spotted by kernel build bot. nios2 fails for other reason

  In file included from include/asm-generic/io.h:767:0,
                   from arch/nios2/include/asm/io.h:61,
                   from include/linux/io.h:25,
                   from arch/nios2/include/asm/pgtable.h:18,
                   from include/linux/mm.h:70,
                   from include/linux/pid_namespace.h:6,
                   from include/linux/ptrace.h:9,
                   from arch/nios2/include/uapi/asm/elf.h:23,
                   from arch/nios2/include/asm/elf.h:22,
                   from include/linux/elf.h:4,
                   from include/linux/module.h:15,
                   from init/main.c:16:
  include/linux/vmalloc.h: In function '__vmalloc_node_flags':
  include/linux/vmalloc.h:99:40: error: 'PAGE_KERNEL' undeclared (first use in this function); did you mean 'GFP_KERNEL'?

which is due to the newly added #include <asm/pgtable.h>, which on nios2
includes <linux/io.h> and thus <asm/io.h> and <asm-generic/io.h> which
again includes <linux/vmalloc.h>.

Tweaking that around just turns out a bigger headache than necessary.
This patch reverts 1f5307b1e094 and reimplements the original fix in a
different way.  __vmalloc_node_flags can stay static inline which will
cover vmalloc* functions.  We only have one external user
(kvmalloc_node) and we can export __vmalloc_node_flags_caller and
provide the caller directly.  This is much simpler and it doesn't really
need any games with header files.

[akpm@linux-foundation.org: coding-style fixes]
[mhocko@kernel.org: revert old comment]
Link: http://lkml.kernel.org/r/20170509211054.GB16325@dhcp22.suse.cz
Fixes: 1f5307b1e094 ("mm, vmalloc: properly track vmalloc users")
Link: http://lkml.kernel.org/r/20170509153702.GR6481@dhcp22.suse.cz
Signed-off-by: Michal Hocko <mhocko@suse.com>
Cc: Tobias Klauser <tklauser@distanz.ch>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agomm/khugepaged: add missed tracepoint for collapse_huge_page_swapin
SeongJae Park [Fri, 12 May 2017 22:46:38 +0000 (15:46 -0700)]
mm/khugepaged: add missed tracepoint for collapse_huge_page_swapin

One return case of `__collapse_huge_page_swapin()` does not invoke
tracepoint while every other return case does.  This commit adds a
tracepoint invocation for the case.

Link: http://lkml.kernel.org/r/20170507101813.30187-1-sj38.park@gmail.com
Signed-off-by: SeongJae Park <sj38.park@gmail.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agogcov: support GCC 7.1
Martin Liska [Fri, 12 May 2017 22:46:35 +0000 (15:46 -0700)]
gcov: support GCC 7.1

Starting from GCC 7.1, __gcov_exit is a new symbol expected to be
implemented in a profiling runtime.

[akpm@linux-foundation.org: coding-style fixes]
[mliska@suse.cz: v2]
Link: http://lkml.kernel.org/r/e63a3c59-0149-c97e-4084-20ca8f146b26@suse.cz
Link: http://lkml.kernel.org/r/8c4084fa-3885-29fe-5fc4-0d4ca199c785@suse.cz
Signed-off-by: Martin Liska <mliska@suse.cz>
Acked-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agomm, vmstat: Remove spurious WARN() during zoneinfo print
Reza Arbab [Fri, 12 May 2017 22:46:32 +0000 (15:46 -0700)]
mm, vmstat: Remove spurious WARN() during zoneinfo print

After commit e2ecc8a79ed4 ("mm, vmstat: print non-populated zones in
zoneinfo"), /proc/zoneinfo will show unpopulated zones.

A memoryless node, having no populated zones at all, was previously
ignored, but will now trigger the WARN() in is_zone_first_populated().

Remove this warning, as its only purpose was to warn of a situation that
has since been enabled.

Aside: The "per-node stats" are still printed under the first populated
zone, but that's not necessarily the first stanza any more.  I'm not
sure which criteria is more important with regard to not breaking
parsers, but it looks a little weird to the eye.

Fixes: e2ecc8a79ed4 ("mm, vmstat: print node-based stats in zoneinfo file")
Link: http://lkml.kernel.org/r/1493854905-10918-1-git-send-email-arbab@linux.vnet.ibm.com
Signed-off-by: Reza Arbab <arbab@linux.vnet.ibm.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Anshuman Khandual <khandual@linux.vnet.ibm.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agotime: delete current_fs_time()
Deepa Dinamani [Fri, 12 May 2017 22:46:29 +0000 (15:46 -0700)]
time: delete current_fs_time()

All uses of the current_fs_time() function have been replaced by other
time interfaces.

And, its use cases can be fulfilled by current_time() or ktime_get_*
variants.

Link: http://lkml.kernel.org/r/1491613030-11599-13-git-send-email-deepa.kernel@gmail.com
Signed-off-by: Deepa Dinamani <deepa.kernel@gmail.com>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agohwpoison, memcg: forcibly uncharge LRU pages
Michal Hocko [Fri, 12 May 2017 22:46:26 +0000 (15:46 -0700)]
hwpoison, memcg: forcibly uncharge LRU pages

Laurent Dufour has noticed that hwpoinsoned pages are kept charged.  In
his particular case he has hit a bad_page("page still charged to
cgroup") when onlining a hwpoison page.  While this looks like something
that shouldn't happen in the first place because onlining hwpages and
returning them to the page allocator makes only little sense it shows a
real problem.

hwpoison pages do not get freed usually so we do not uncharge them (at
least not since commit 0a31bc97c80c ("mm: memcontrol: rewrite uncharge
API")).  Each charge pins memcg (since e8ea14cc6ead ("mm: memcontrol:
take a css reference for each charged page")) as well and so the
mem_cgroup and the associated state will never go away.  Fix this leak
by forcibly uncharging a LRU hwpoisoned page in delete_from_lru_cache().
We also have to tweak uncharge_list because it cannot rely on zero ref
count for these pages.

[akpm@linux-foundation.org: coding-style fixes]
Fixes: 0a31bc97c80c ("mm: memcontrol: rewrite uncharge API")
Link: http://lkml.kernel.org/r/20170502185507.GB19165@dhcp22.suse.cz
Signed-off-by: Michal Hocko <mhocko@suse.com>
Reported-by: Laurent Dufour <ldufour@linux.vnet.ibm.com>
Tested-by: Laurent Dufour <ldufour@linux.vnet.ibm.com>
Reviewed-by: Balbir Singh <bsingharora@gmail.com>
Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
7 years agoMerge branch 'libnvdimm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdim...
Linus Torvalds [Fri, 12 May 2017 22:43:10 +0000 (15:43 -0700)]
Merge branch 'libnvdimm-fixes' of git://git./linux/kernel/git/nvdimm/nvdimm

Pull libnvdimm fixes from Dan Williams:
 "Incremental fixes and a small feature addition on top of the main
  libnvdimm 4.12 pull request:

   - Geert noticed that tinyconfig was bloated by BLOCK selecting DAX.
     The size regression is fixed by moving all dax helpers into the
     dax-core and only specifying "select DAX" for FS_DAX and
     dax-capable drivers. He also asked for clarification of the
     NR_DEV_DAX config option which, on closer look, does not need to be
     a config option at all. Mike also throws in a DEV_DAX_PMEM fixup
     for good measure.

   - Ben's attention to detail on -stable patch submissions caught a
     case where the recent fixes to arch_copy_from_iter_pmem() missed a
     condition where we strand dirty data in the cache. This is tagged
     for -stable and will also be included in the rework of the pmem api
     to a proposed {memcpy,copy_user}_flushcache() interface for 4.13.

   - Vishal adds a feature that missed the initial pull due to pending
     review feedback. It allows the kernel to clear media errors when
     initializing a BTT (atomic sector update driver) instance on a pmem
     namespace.

   - Ross noticed that the dax_device + dax_operations conversion broke
     __dax_zero_page_range(). The nvdimm unit tests fail to check this
     path, but xfstests immediately trips over it. No excuse for missing
     this before submitting the 4.12 pull request.

  These all pass the nvdimm unit tests and an xfstests spot check. The
  set has received a build success notification from the kbuild robot"

* 'libnvdimm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  filesystem-dax: fix broken __dax_zero_page_range() conversion
  libnvdimm, btt: ensure that initializing metadata clears poison
  libnvdimm: add an atomic vs process context flag to rw_bytes
  x86, pmem: Fix cache flushing for iovec write < 8 bytes
  device-dax: kill NR_DEV_DAX
  block, dax: move "select DAX" from BLOCK to FS_DAX
  device-dax: Tell kbuild DEV_DAX_PMEM depends on DEV_DAX

7 years agoMerge tag 'sound-fix-4.12-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai...
Linus Torvalds [Fri, 12 May 2017 19:10:38 +0000 (12:10 -0700)]
Merge tag 'sound-fix-4.12-rc1' of git://git./linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "This contains a one-liner change that has a significant impact:
  disabling the build of OSS. It's been unmaintained for long time, and
  we'd like to drop the stuff. Finally, as the first step, stop the
  build. Let's see whether it works without much complaints.

  Other than that, there are two small fixes for HD-audio"

* tag 'sound-fix-4.12-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  sound: Disable the build of OSS drivers
  ALSA: hda: Fix cpu lockup when stopping the cmd dmas
  ALSA: hda - Add mute led support for HP EliteBook 840 G3

7 years agoMerge tag 'for-v4.12-2' of git://git.kernel.org/pub/scm/linux/kernel/git/sre/linux...
Linus Torvalds [Fri, 12 May 2017 19:02:21 +0000 (12:02 -0700)]
Merge tag 'for-v4.12-2' of git://git./linux/kernel/git/sre/linux-power-supply

Pull more power-supply updates from Sebastian Reichel:
 "The power-supply subsystem has a few more changes for the v4.12 merge
  window:

   - New battery driver for AXP20X and AXP22X PMICs

   - Improve max17042_battery for usage on x86

   - Misc small cleanups & fixes"

* tag 'for-v4.12-2' of git://git.kernel.org/pub/scm/linux/kernel/git/sre/linux-power-supply: (34 commits)
  power: supply: cpcap-charger: Keep trickle charger bits disabled
  power: supply: cpcap-charger: Fix enable for 3.8V charge setting
  power: supply: cpcap-charger: Fix charge voltage configuration
  power: supply: cpcap-charger: Fix charger name
  power: supply: twl4030-charger: make twl4030_bci_property_is_writeable static
  power: supply: sbs-battery: Add alert callback
  mailmap: add Sebastian Reichel
  power: supply: avoid unused twl4030-madc.h
  power: supply: sbs-battery: Correct supply status with current draw
  power: supply: sbs-battery: Don't ignore the first external power change
  power: supply: pda_power: move from timer to delayed_work
  power: supply: max17042_battery: Add support for the SCOPE property
  power: supply: max17042_battery: Add support for the CHARGE_NOW property
  power: supply: max17042_battery: Add support for the CHARGE_FULL_DESIGN property
  power: supply: max17042_battery: mAh readings depend on r_sns value
  power: supply: max17042_battery: Add support for the VOLT_MIN property
  power: supply: max17042_battery: Add support for the TECHNOLOGY attribute
  power: supply: max17042_battery: Add external_power_changed callback
  power: supply: max17042_battery: Add support for the STATUS property
  power: supply: max17042_battery: Add default platform_data fallback data
  ...

7 years agoMerge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/rzhang/linux
Linus Torvalds [Fri, 12 May 2017 18:58:45 +0000 (11:58 -0700)]
Merge branch 'next' of git://git./linux/kernel/git/rzhang/linux

Pull thermal management updates from Zhang Rui:

 - Fix a problem where orderly_shutdown() is called for multiple times
   due to multiple critical overheating events raised in a short period
   by platform thermal driver. (Keerthy)

 - Introduce a backup thermal shutdown mechanism, which invokes
   kernel_power_off()/emergency_restart() directly, after
   orderly_shutdown() being issued for certain amount of time(specified
   via Kconfig). This is useful in certain conditions that userspace may
   be unable to power off the system in a clean manner and leaves the
   system in a critical state, like in the middle of driver probing
   phase. (Keerthy)

 - Introduce a new interface in thermal devfreq_cooling code so that the
   driver can provide more precise data regarding actual power to the
   thermal governor every time the power budget is calculated. (Lukasz
   Luba)

 - Introduce BCM 2835 soc thermal driver and northstar thermal driver,
   within a new sub-folder. (Rafał Miłecki)

 - Introduce DA9062/61 thermal driver. (Steve Twiss)

 - Remove non-DT booting on TI-SoC driver. Also add support to fetching
   coefficients from DT. (Keerthy)

 - Refactorf RCAR Gen3 thermal driver. (Niklas Söderlund)

 - Small fix on MTK and intel-soc-dts thermal driver. (Dawei Chien,
   Brian Bian)

* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/rzhang/linux: (25 commits)
  thermal: core: Add a back up thermal shutdown mechanism
  thermal: core: Allow orderly_poweroff to be called only once
  Thermal: Intel SoC DTS: Change interrupt request behavior
  trace: thermal: add another parameter 'power' to the tracing function
  thermal: devfreq_cooling: add new interface for direct power read
  thermal: devfreq_cooling: refactor code and add get_voltage function
  thermal: mt8173: minor mtk_thermal.c cleanups
  thermal: bcm2835: move to the broadcom subdirectory
  thermal: broadcom: ns: specify myself as MODULE_AUTHOR
  thermal: da9062/61: Thermal junction temperature monitoring driver
  Documentation: devicetree: thermal: da9062/61 TJUNC temperature binding
  thermal: broadcom: add Northstar thermal driver
  dt-bindings: thermal: add support for Broadcom's Northstar thermal
  thermal: bcm2835: add thermal driver for bcm2835 SoC
  dt-bindings: Add thermal zone to bcm2835-thermal example
  thermal: rcar_gen3_thermal: add suspend and resume support
  thermal: rcar_gen3_thermal: store device match data in private structure
  thermal: rcar_gen3_thermal: enable hardware interrupts for trip points
  thermal: rcar_gen3_thermal: record and check number of TSCs found
  thermal: rcar_gen3_thermal: check that TSC exists before memory allocation
  ...

7 years agoMerge tag 'drm-fixes-for-v4.12-rc1' of git://people.freedesktop.org/~airlied/linux
Linus Torvalds [Fri, 12 May 2017 18:48:26 +0000 (11:48 -0700)]
Merge tag 'drm-fixes-for-v4.12-rc1' of git://people.freedesktop.org/~airlied/linux

Pull drm fixes from Dave Airlie:
 "AMD, nouveau, one i915, and one EDID fix for v4.12-rc1

  Some fixes that it would be good to have in rc1. It contains the i915
  quiet fix that you reported.

  It also has an amdgpu fixes pull, with lots of ongoing work on Vega10
  which is new in this kernel and is preliminary support so may have a
  fair bit of movement.

  Otherwise a few non-Vega10 AMD fixes, one EDID fix and some nouveau
  regression fixers"

* tag 'drm-fixes-for-v4.12-rc1' of git://people.freedesktop.org/~airlied/linux: (144 commits)
  drm/i915: Make vblank evade warnings optional
  drm/nouveau/therm: remove ineffective workarounds for alarm bugs
  drm/nouveau/tmr: avoid processing completed alarms when adding a new one
  drm/nouveau/tmr: fix corruption of the pending list when rescheduling an alarm
  drm/nouveau/tmr: handle races with hw when updating the next alarm time
  drm/nouveau/tmr: ack interrupt before processing alarms
  drm/nouveau/core: fix static checker warning
  drm/nouveau/fb/ram/gf100-: remove 0x10f200 read
  drm/nouveau/kms/nv50: skip core channel cursor update on position-only changes
  drm/nouveau/kms/nv50: fix source-rect-only plane updates
  drm/nouveau/kms/nv50: remove pointless argument to window atomic_check_acquire()
  drm/amd/powerplay: refine pwm1_enable callback functions for CI.
  drm/amd/powerplay: refine pwm1_enable callback functions for vi.
  drm/amd/powerplay: refine pwm1_enable callback functions for Vega10.
  drm/amdgpu: refine amdgpu pwm1_enable sysfs interface.
  drm/amdgpu: add amd fan ctrl mode enums.
  drm/amd/powerplay: add more smu message on Vega10.
  drm/amdgpu: fix dependency issue
  drm/amd: fix init order of sched job
  drm/amdgpu: add some additional vega10 pci ids
  ...

7 years agoMerge branch 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/nab/target...
Linus Torvalds [Fri, 12 May 2017 18:44:13 +0000 (11:44 -0700)]
Merge branch 'for-next' of git://git./linux/kernel/git/nab/target-pending

Pull SCSI target updates from Nicholas Bellinger:
 "Things were a lot more calm than previously expected. It's primarily
  fixes in various areas, with most of the new functionality centering
  around TCMU backend driver work that Xiubo Li has been driving.

  Here's the summary on the feature side:

   - Make T10-PI verify configurable for emulated (FILEIO + RD) backends
    (Dmitry Monakhov)
   - Allow target-core/TCMU pass-through to use in-kernel SPC-PR logic
    (Bryant Ly + MNC)
   - Add TCMU support for growing ring buffer size (Xiubo Li + MNC)
   - Add TCMU support for global block data pool (Xiubo Li + MNC)

  and on the bug-fix side:

   - Fix COMPARE_AND_WRITE non GOOD status handling for READ phase
    failures (Gary Guo + nab)
   - Fix iscsi-target hang with explicitly changing per NodeACL
    CmdSN number depth with concurrent login driven session
    reinstatement.  (Gary Guo + nab)
   - Fix ibmvscsis fabric driver ABORT task handling (Bryant Ly)
   - Fix target-core/FILEIO zero length handling (Bart Van Assche)

  Also, there was an OOPs introduced with the WRITE_VERIFY changes that
  I ended up reverting at the last minute, because as not unusual Bart
  and I could not agree on the fix in time for -rc1. Since it's specific
  to a conformance test, it's been reverted for now.

  There is a separate patch in the queue to address the underlying
  control CDB write overflow regression in >= v4.3 separate from the
  WRITE_VERIFY revert here, that will be pushed post -rc1"

* 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending: (30 commits)
  Revert "target: Fix VERIFY and WRITE VERIFY command parsing"
  IB/srpt: Avoid that aborting a command triggers a kernel warning
  IB/srpt: Fix abort handling
  target/fileio: Fix zero-length READ and WRITE handling
  ibmvscsis: Do not send aborted task response
  tcmu: fix module removal due to stuck thread
  target: Don't force session reset if queue_depth does not change
  iscsi-target: Set session_fall_back_to_erl0 when forcing reinstatement
  target: Fix compare_and_write_callback handling for non GOOD status
  tcmu: Recalculate the tcmu_cmd size to save cmd area memories
  tcmu: Add global data block pool support
  tcmu: Add dynamic growing data area feature support
  target: fixup error message in target_tg_pt_gp_tg_pt_gp_id_store()
  target: fixup error message in target_tg_pt_gp_alua_access_type_store()
  target/user: PGR Support
  target: Add WRITE_VERIFY_16
  Documentation/target: add an example script to configure an iSCSI target
  target: Use kmalloc_array() in transport_kmap_data_sg()
  target: Use kmalloc_array() in compare_and_write_callback()
  target: Improve size determinations in two functions
  ...

7 years agoMerge branch 'work.sane_pwd' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Linus Torvalds [Fri, 12 May 2017 18:39:59 +0000 (11:39 -0700)]
Merge branch 'work.sane_pwd' of git://git./linux/kernel/git/viro/vfs

Pull misc vfs updates from Al Viro:
 "Making sure that something like a referral point won't end up as pwd
  or root.

  The main part is the last commit (fixing mntns_install()); that one
  fixes a hard-to-hit race. The fchdir() commit is making fchdir(2) a
  bit more robust - it should be impossible to get opened files (even
  O_PATH ones) for referral points in the first place, so the existing
  checks are OK, but checking the same thing as in chdir(2) is just as
  cheap.

  The path_init() commit removes a redundant check that shouldn't have
  been there in the first place"

* 'work.sane_pwd' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  make sure that mntns_install() doesn't end up with referral for root
  path_init(): don't bother with checking MAY_EXEC for LOOKUP_ROOT
  make sure that fchdir() won't accept referral points, etc.

7 years agoMerge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 12 May 2017 17:45:36 +0000 (10:45 -0700)]
Merge branch 'perf-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull perf updates/fixes from Ingo Molnar:
 "Mostly tooling updates, but also two kernel fixes: a call chain
  handling robustness fix and an x86 PMU driver event definition fix"

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf/callchain: Force USER_DS when invoking perf_callchain_user()
  tools build: Fixup sched_getcpu feature test
  perf tests kmod-path: Don't fail if compressed modules aren't supported
  perf annotate: Fix AArch64 comment char
  perf tools: Fix spelling mistakes
  perf/x86: Fix Broadwell-EP DRAM RAPL events
  perf config: Refactor a duplicated code for obtaining config file name
  perf symbols: Allow user probes on versioned symbols
  perf symbols: Accept symbols starting at address 0
  tools lib string: Adopt prefixcmp() from perf and subcmd
  perf units: Move parse_tag_value() to units.[ch]
  perf ui gtk: Move gtk .so name to the only place where it is used
  perf tools: Move HAS_BOOL define to where perl headers are used
  perf memswap: Split the byteswap memory range wrappers from util.[ch]
  perf tools: Move event prototypes from util.h to event.h
  perf buildid: Move prototypes from util.h to build-id.h