]> www.infradead.org Git - users/jedix/linux-maple.git/log
users/jedix/linux-maple.git
3 years agovdpa: Mark vdpa_config_ops.get_vq_notification as optional
Eugenio Pérez [Thu, 4 Nov 2021 19:52:48 +0000 (20:52 +0100)]
vdpa: Mark vdpa_config_ops.get_vq_notification as optional

Since vhost_vdpa_mmap checks for its existence before calling it.

Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Link: https://lore.kernel.org/r/20211104195248.2088904-1-eperezma@redhat.com
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
3 years agovdpa: Avoid duplicate call to vp_vdpa get_status
Eugenio Pérez [Thu, 4 Nov 2021 19:58:33 +0000 (20:58 +0100)]
vdpa: Avoid duplicate call to vp_vdpa get_status

It has no sense to call get_status twice, since we already have a
variable for that.

Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Link: https://lore.kernel.org/r/20211104195833.2089796-1-eperezma@redhat.com
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
3 years agoeni_vdpa: Simplify 'eni_vdpa_probe()'
Christophe JAILLET [Sun, 7 Nov 2021 16:14:56 +0000 (17:14 +0100)]
eni_vdpa: Simplify 'eni_vdpa_probe()'

When 'pcim_enable_device()' is used, some resources become automagically
managed.
There is no need to call 'pci_free_irq_vectors()' when the driver is
removed. The same will already be done by 'pcim_release()'.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Link: https://lore.kernel.org/r/02045bdcbbb25f79bae4827f66029cfcddc90381.1636301587.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
3 years agonet/mlx5_vdpa: Offer VIRTIO_NET_F_MTU when setting MTU
Eli Cohen [Wed, 24 Nov 2021 17:09:49 +0000 (19:09 +0200)]
net/mlx5_vdpa: Offer VIRTIO_NET_F_MTU when setting MTU

Make sure to offer VIRTIO_NET_F_MTU since we configure the MTU based on
what was queried from the device.

This allows the virtio driver to allocate large enough buffers based on
the reported MTU.

Signed-off-by: Eli Cohen <elic@nvidia.com>
Link: https://lore.kernel.org/r/20211124170949.51725-1-elic@nvidia.com
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Si-Wei Liu <si-wei.liu@oracle.com>
3 years agovirtio-mem: prepare fake page onlining code for granularity smaller than MAX_ORDER - 1
David Hildenbrand [Fri, 26 Nov 2021 13:42:09 +0000 (14:42 +0100)]
virtio-mem: prepare fake page onlining code for granularity smaller than MAX_ORDER - 1

Let's prepare our fake page onlining code for subblock size smaller than
MAX_ORDER - 1: we might get called for ranges not covering properly
aligned MAX_ORDER - 1 pages. We have to detect the order to use
dynamically.

Signed-off-by: David Hildenbrand <david@redhat.com>
Link: https://lore.kernel.org/r/20211126134209.17332-3-david@redhat.com
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: Eric Ren <renzhengeek@gmail.com>
3 years agovirtio-mem: prepare page onlining code for granularity smaller than MAX_ORDER - 1
David Hildenbrand [Fri, 26 Nov 2021 13:42:08 +0000 (14:42 +0100)]
virtio-mem: prepare page onlining code for granularity smaller than MAX_ORDER - 1

Let's prepare our page onlining code for subblock size smaller than
MAX_ORDER - 1: we'll get called for a MAX_ORDER - 1 page but might have
some subblocks in the range plugged and some unplugged. In that case,
fallback to subblock granularity to properly only expose the plugged
parts to the buddy.

Signed-off-by: David Hildenbrand <david@redhat.com>
Link: https://lore.kernel.org/r/20211126134209.17332-2-david@redhat.com
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Reviewed-by: Eric Ren <renzhengeek@gmail.com>
3 years agovdpa: add driver_override support
Stefano Garzarella [Fri, 26 Nov 2021 16:47:53 +0000 (17:47 +0100)]
vdpa: add driver_override support

`driver_override` allows to control which of the vDPA bus drivers
binds to a vDPA device.

If `driver_override` is not set, the previous behaviour is followed:
devices use the first vDPA bus driver loaded (unless auto binding
is disabled).

Tested on Fedora 34 with driverctl(8):
  $ modprobe virtio-vdpa
  $ modprobe vhost-vdpa
  $ modprobe vdpa-sim-net

  $ vdpa dev add mgmtdev vdpasim_net name dev1

  # dev1 is attached to the first vDPA bus driver loaded
  $ driverctl -b vdpa list-devices
    dev1 virtio_vdpa

  $ driverctl -b vdpa set-override dev1 vhost_vdpa

  $ driverctl -b vdpa list-devices
    dev1 vhost_vdpa [*]

  Note: driverctl(8) integrates with udev so the binding is
  preserved.

Suggested-by: Jason Wang <jasowang@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Link: https://lore.kernel.org/r/20211126164753.181829-3-sgarzare@redhat.com
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agodocs: document sysfs ABI for vDPA bus
Stefano Garzarella [Fri, 26 Nov 2021 16:47:52 +0000 (17:47 +0100)]
docs: document sysfs ABI for vDPA bus

Add missing documentation of sysfs ABI for vDPA bus in
the new Documentation/ABI/testing/sysfs-bus-vdpa file.

Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Link: https://lore.kernel.org/r/20211126164753.181829-2-sgarzare@redhat.com
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
3 years agoifcvf/vDPA: fix misuse virtio-net device config size for blk dev
Zhu Lingshan [Wed, 1 Dec 2021 08:12:55 +0000 (16:12 +0800)]
ifcvf/vDPA: fix misuse virtio-net device config size for blk dev

This commit fixes a misuse of virtio-net device config size issue
for virtio-block devices.

A new member config_size in struct ifcvf_hw is introduced and would
be initialized through vdpa_dev_add() to record correct device
config size.

To be more generic, rename ifcvf_hw.net_config to ifcvf_hw.dev_config,
the helpers ifcvf_read/write_net_config() to ifcvf_read/write_dev_config()

Signed-off-by: Zhu Lingshan <lingshan.zhu@intel.com>
Reported-and-suggested-by: Stefano Garzarella <sgarzare@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Fixes: 6ad31d162a4e ("vDPA/ifcvf: enable Intel C5000X-PL virtio-block for vDPA")
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20211201081255.60187-1-lingshan.zhu@intel.com
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agovduse: moving kvfree into caller
Guanjun [Mon, 6 Dec 2021 08:48:18 +0000 (16:48 +0800)]
vduse: moving kvfree into caller

This free action should be moved into caller 'vduse_ioctl' in
concert with the allocation.

No functional change.

Signed-off-by: Guanjun <guanjun@linux.alibaba.com>
Link: https://lore.kernel.org/r/1638780498-55571-1-git-send-email-guanjun@linux.alibaba.com
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agohwrng: virtio - unregister device before reset
Michael S. Tsirkin [Thu, 25 Nov 2021 18:00:17 +0000 (13:00 -0500)]
hwrng: virtio - unregister device before reset

unregister after reset is clearly wrong - device
can be used while it's reset. There's an attempt to
protect against that using hwrng_removed but it
seems racy since access can be in progress
when the flag is set.

Just unregister, then reset seems simpler and cleaner.
NB: we might be able to drop hwrng_removed in a follow-up patch.

Signed-off-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agovirtio: wrap config->reset calls
Michael S. Tsirkin [Wed, 13 Oct 2021 10:55:44 +0000 (06:55 -0400)]
virtio: wrap config->reset calls

This will enable cleanups down the road.
The idea is to disable cbs, then add "flush_queued_cbs" callback
as a parameter, this way drivers can flush any work
queued after callbacks have been disabled.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://lore.kernel.org/r/20211013105226.20225-1-mst@redhat.com
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agodrm/amd/display: Revert W/A for hard hangs on DCN20/DCN21
Mario Limonciello [Fri, 7 Jan 2022 16:44:17 +0000 (10:44 -0600)]
drm/amd/display: Revert W/A for hard hangs on DCN20/DCN21

The WA from commit 2a50edbf10c8 ("drm/amd/display: Apply w/a for hard hang
on HPD") and commit 1bd3bc745e7f ("drm/amd/display: Extend w/a for hard
hang on HPD to dcn20") causes a regression in s0ix where the system will
fail to resume properly on many laptops.  Pull the workarounds out to
avoid that s0ix regression in the common case.  This HPD hang happens with
an external device in special circumstances and a new W/A will need to be
developed for this in the future.

Cc: stable@vger.kernel.org
Cc: Qingqing Zhuo <qingqing.zhuo@amd.com>
Reported-by: Scott Bruce <smbruce@gmail.com>
Reported-by: Chris Hixon <linux-kernel-bugs@hixontech.com>
Reported-by: spasswolf@web.de
Link: https://bugzilla.kernel.org/show_bug.cgi?id=215436
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1821
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1852
Fixes: 2a50edbf10c8 ("drm/amd/display: Apply w/a for hard hang on HPD")
Fixes: 1bd3bc745e7f ("drm/amd/display: Extend w/a for hard hang on HPD to dcn20")
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
3 years agodrm/amdgpu: drop flags check for CHIP_IP_DISCOVERY
Alex Deucher [Fri, 14 Jan 2022 14:59:29 +0000 (09:59 -0500)]
drm/amdgpu: drop flags check for CHIP_IP_DISCOVERY

Support for IP based discovery is in place now so this
check is no longer required.

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
3 years agodrm/amdgpu: Fix rejecting Tahiti GPUs
Lukas Fink [Fri, 14 Jan 2022 06:51:41 +0000 (07:51 +0100)]
drm/amdgpu: Fix rejecting Tahiti GPUs

eb4fd29afd4a ("drm/amdgpu: bind to any 0x1002 PCI diplay class device") added
generic bindings to amdgpu so that that it binds to all display class devices
with VID 0x1002 and then rejects those in amdgpu_pci_probe.

Unfortunately it reuses a driver_data value of 0 to detect those new bindings,
which is already used to denote CHIP_TAHITI ASICs.

The driver_data value given to those new bindings was changed in
dd0761fd24ea1 ("drm/amdgpu: set CHIP_IP_DISCOVERY as the asic type by default")
to CHIP_IP_DISCOVERY (=36), but it seems that the check in amdgpu_pci_probe
was forgotten to be changed. Therefore, it still rejects Tahiti GPUs.

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1860
Fixes: eb4fd29afd4a ("drm/amdgpu: bind to any 0x1002 PCI diplay class device")
Cc: stable@vger.kernel.org
Signed-off-by: Lukas Fink <lukas.fink1@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
3 years agodrm/amdgpu: don't do resets on APUs which don't support it
Alex Deucher [Thu, 13 Jan 2022 03:38:51 +0000 (22:38 -0500)]
drm/amdgpu: don't do resets on APUs which don't support it

It can cause a hang.  This is normally not enabled for GPU
hangs on these asics, but was recently enabled for handling
aborted suspends.  This causes hangs on some platforms
on suspend.

Fixes: daf8de0874ab5b ("drm/amdgpu: always reset the asic in suspend (v2)")
Cc: stable@vger.kernel.org
Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1858
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
3 years agodrm/amdgpu: invert the logic in amdgpu_device_should_recover_gpu()
Alex Deucher [Tue, 11 Jan 2022 22:41:44 +0000 (17:41 -0500)]
drm/amdgpu: invert the logic in amdgpu_device_should_recover_gpu()

Rather than opting into GPU recovery support, default to on, and
opt out if it's not working on a particular GPU.  This avoids the
need to add new asics to this list since this is a core feature.

Reviewed-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
3 years agodrm/amdgpu: Enable recovery on yellow carp
CHANDAN VURDIGERE NATARAJ [Tue, 11 Jan 2022 13:32:26 +0000 (19:02 +0530)]
drm/amdgpu: Enable recovery on yellow carp

Add yellow carp to devices which support recovery

Signed-off-by: CHANDAN VURDIGERE NATARAJ <chandan.vurdigerenataraj@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
3 years agoMAINTAINERS: Add Helge as fbdev maintainer
Helge Deller [Fri, 14 Jan 2022 18:11:21 +0000 (19:11 +0100)]
MAINTAINERS: Add Helge as fbdev maintainer

The fbdev layer is orphaned, but seems to need some care.
So I'd like to step up as new maintainer.

Signed-off-by: Helge Deller <deller@gmx.de>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
3 years agox86/fpu: Fix inline prefix warnings
Yang Zhong [Thu, 13 Jan 2022 18:08:25 +0000 (13:08 -0500)]
x86/fpu: Fix inline prefix warnings

Fix sparse warnings in xstate and remove inline prefix.

Fixes: 980fe2fddcff ("x86/fpu: Extend fpu_xstate_prctl() with guest permissions")
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Reported-by: kernel test robot <lkp@intel.com>
Message-Id: <20220113180825.322333-1-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agoselftest: kvm: Add amx selftest
Yang Zhong [Thu, 23 Dec 2021 14:53:22 +0000 (09:53 -0500)]
selftest: kvm: Add amx selftest

This selftest covers two aspects of AMX.  The first is triggering #NM
exception and checking the MSR XFD_ERR value.  The second case is
loading tile config and tile data into guest registers and trapping to
the host side for a complete save/load of the guest state.  TMM0
is also checked against memory data after save/restore.

Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20211223145322.2914028-4-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agoselftest: kvm: Move struct kvm_x86_state to header
Yang Zhong [Thu, 23 Dec 2021 14:53:21 +0000 (09:53 -0500)]
selftest: kvm: Move struct kvm_x86_state to header

Those changes can avoid dereferencing pointer compile issue
when amx_test.c reference state->xsave.

Move struct kvm_x86_state definition to processor.h.

Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20211223145322.2914028-3-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agoselftest: kvm: Reorder vcpu_load_state steps for AMX
Paolo Bonzini [Thu, 23 Dec 2021 14:53:20 +0000 (09:53 -0500)]
selftest: kvm: Reorder vcpu_load_state steps for AMX

For AMX support it is recommended to load XCR0 after XFD, so
that KVM does not see XFD=0, XCR=1 for a save state that will
eventually be disabled (which would lead to premature allocation
of the space required for that save state).

It is also required to load XSAVE data after XCR0 and XFD, so
that KVM can trigger allocation of the extra space required to
store AMX state.

Adjust vcpu_load_state to obey these new requirements.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20211223145322.2914028-2-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agokvm: x86: Disable interception for IA32_XFD on demand
Kevin Tian [Wed, 5 Jan 2022 12:35:32 +0000 (04:35 -0800)]
kvm: x86: Disable interception for IA32_XFD on demand

Always intercepting IA32_XFD causes non-negligible overhead when this
register is updated frequently in the guest.

Disable r/w emulation after intercepting the first WRMSR(IA32_XFD)
with a non-zero value.

Disable WRMSR emulation implies that IA32_XFD becomes out-of-sync
with the software states in fpstate and the per-cpu xfd cache. This
leads to two additional changes accordingly:

  - Call fpu_sync_guest_vmexit_xfd_state() after vm-exit to bring
    software states back in-sync with the MSR, before handle_exit_irqoff()
    is called.

  - Always trap #NM once write interception is disabled for IA32_XFD.
    The #NM exception is rare if the guest doesn't use dynamic
    features. Otherwise, there is at most one exception per guest
    task given a dynamic feature.

p.s. We have confirmed that SDM is being revised to say that
when setting IA32_XFD[18] the AMX register state is not guaranteed
to be preserved. This clarification avoids adding mess for a creative
guest which sets IA32_XFD[18]=1 before saving active AMX state to
its own storage.

Signed-off-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-22-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agox86/fpu: Provide fpu_sync_guest_vmexit_xfd_state()
Thomas Gleixner [Wed, 5 Jan 2022 12:35:31 +0000 (04:35 -0800)]
x86/fpu: Provide fpu_sync_guest_vmexit_xfd_state()

KVM can disable the write emulation for the XFD MSR when the vCPU's fpstate
is already correctly sized to reduce the overhead.

When write emulation is disabled the XFD MSR state after a VMEXIT is
unknown and therefore not in sync with the software states in fpstate and
the per CPU XFD cache.

Provide fpu_sync_guest_vmexit_xfd_state() which has to be invoked after a
VMEXIT before enabling interrupts when write emulation is disabled for the
XFD MSR.

It could be invoked unconditionally even when write emulation is enabled
for the price of a pointless MSR read.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-21-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agokvm: selftests: Add support for KVM_CAP_XSAVE2
Wei Wang [Wed, 5 Jan 2022 12:35:30 +0000 (04:35 -0800)]
kvm: selftests: Add support for KVM_CAP_XSAVE2

When KVM_CAP_XSAVE2 is supported, userspace is expected to allocate
buffer for KVM_GET_XSAVE2 and KVM_SET_XSAVE using the size returned
by KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2).

Signed-off-by: Wei Wang <wei.w.wang@intel.com>
Signed-off-by: Guang Zeng <guang.zeng@intel.com>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-20-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agokvm: x86: Add support for getting/setting expanded xstate buffer
Guang Zeng [Wed, 5 Jan 2022 12:35:29 +0000 (04:35 -0800)]
kvm: x86: Add support for getting/setting expanded xstate buffer

With KVM_CAP_XSAVE, userspace uses a hardcoded 4KB buffer to get/set
xstate data from/to KVM. This doesn't work when dynamic xfeatures
(e.g. AMX) are exposed to the guest as they require a larger buffer
size.

Introduce a new capability (KVM_CAP_XSAVE2). Userspace VMM gets the
required xstate buffer size via KVM_CHECK_EXTENSION(KVM_CAP_XSAVE2).
KVM_SET_XSAVE is extended to work with both legacy and new capabilities
by doing properly-sized memdup_user() based on the guest fpu container.
KVM_GET_XSAVE is kept for backward-compatible reason. Instead,
KVM_GET_XSAVE2 is introduced under KVM_CAP_XSAVE2 as the preferred
interface for getting xstate buffer (4KB or larger size) from KVM
Link: https://lkml.org/lkml/2021/12/15/510)
Also, update the api doc with the new KVM_GET_XSAVE2 ioctl.

Signed-off-by: Guang Zeng <guang.zeng@intel.com>
Signed-off-by: Wei Wang <wei.w.wang@intel.com>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-19-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agox86/fpu: Add uabi_size to guest_fpu
Thomas Gleixner [Wed, 5 Jan 2022 12:35:28 +0000 (04:35 -0800)]
x86/fpu: Add uabi_size to guest_fpu

Userspace needs to inquire KVM about the buffer size to work
with the new KVM_SET_XSAVE and KVM_GET_XSAVE2. Add the size info
to guest_fpu for KVM to access.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Wei Wang <wei.w.wang@intel.com>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-18-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agokvm: x86: Add CPUID support for Intel AMX
Jing Liu [Wed, 5 Jan 2022 12:35:27 +0000 (04:35 -0800)]
kvm: x86: Add CPUID support for Intel AMX

Extend CPUID emulation to support XFD, AMX_TILE, AMX_INT8 and
AMX_BF16. Adding those bits into kvm_cpu_caps finally activates all
previous logics in this series.

Hide XFD on 32bit host kernels. Otherwise it leads to a weird situation
where KVM tells userspace to migrate MSR_IA32_XFD and then rejects
attempts to read/write the MSR.

Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-17-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agokvm: x86: Add XCR0 support for Intel AMX
Jing Liu [Wed, 5 Jan 2022 12:35:26 +0000 (04:35 -0800)]
kvm: x86: Add XCR0 support for Intel AMX

Two XCR0 bits are defined for AMX to support XSAVE mechanism. Bit 17
is for tilecfg and bit 18 is for tiledata.

The value of XCR0[17:18] is always either 00b or 11b. Also, SDM
recommends that only 64-bit operating systems enable Intel AMX by
setting XCR0[18:17]. 32-bit host kernel never sets the tile bits in
vcpu->arch.guest_supported_xcr0.

Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-16-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agokvm: x86: Disable RDMSR interception of IA32_XFD_ERR
Jing Liu [Wed, 5 Jan 2022 12:35:25 +0000 (04:35 -0800)]
kvm: x86: Disable RDMSR interception of IA32_XFD_ERR

This saves one unnecessary VM-exit in guest #NM handler, given that the
MSR is already restored with the guest value before the guest is resumed.

Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-15-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agokvm: x86: Emulate IA32_XFD_ERR for guest
Jing Liu [Wed, 5 Jan 2022 12:35:24 +0000 (04:35 -0800)]
kvm: x86: Emulate IA32_XFD_ERR for guest

Emulate read/write to IA32_XFD_ERR MSR.

Only the saved value in the guest_fpu container is touched in the
emulation handler. Actual MSR update is handled right before entering
the guest (with preemption disabled)

Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Zeng Guang <guang.zeng@intel.com>
Signed-off-by: Wei Wang <wei.w.wang@intel.com>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-14-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agokvm: x86: Intercept #NM for saving IA32_XFD_ERR
Jing Liu [Wed, 5 Jan 2022 12:35:23 +0000 (04:35 -0800)]
kvm: x86: Intercept #NM for saving IA32_XFD_ERR

Guest IA32_XFD_ERR is generally modified in two places:

  - Set by CPU when #NM is triggered;
  - Cleared by guest in its #NM handler;

Intercept #NM for the first case when a nonzero value is written
to IA32_XFD. Nonzero indicates that the guest is willing to do
dynamic fpstate expansion for certain xfeatures, thus KVM needs to
manage and virtualize guest XFD_ERR properly. The vcpu exception
bitmap is updated in XFD write emulation according to guest_fpu::xfd.

Save the current XFD_ERR value to the guest_fpu container in the #NM
VM-exit handler. This must be done with interrupt disabled, otherwise
the unsaved MSR value may be clobbered by host activity.

The saving operation is conducted conditionally only when guest_fpu:xfd
includes a non-zero value. Doing so also avoids misread on a platform
which doesn't support XFD but #NM is triggered due to L1 interception.

Queueing #NM to the guest is postponed to handle_exception_nmi(). This
goes through the nested_vmx check so a virtual vmexit is queued instead
when #NM is triggered in L2 but L1 wants to intercept it.

Restore the host value (always ZERO outside of the host #NM
handler) before enabling interrupt.

Restore the guest value from the guest_fpu container right before
entering the guest (with interrupt disabled).

Suggested-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-13-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agox86/fpu: Prepare xfd_err in struct fpu_guest
Jing Liu [Wed, 5 Jan 2022 12:35:22 +0000 (04:35 -0800)]
x86/fpu: Prepare xfd_err in struct fpu_guest

When XFD causes an instruction to generate #NM, IA32_XFD_ERR
contains information about which disabled state components are
being accessed. The #NM handler is expected to check this
information and then enable the state components by clearing
IA32_XFD for the faulting task (if having permission).

If the XFD_ERR value generated in guest is consumed/clobbered
by the host before the guest itself doing so, it may lead to
non-XFD-related #NM treated as XFD #NM in host (due to non-zero
value in XFD_ERR), or XFD-related #NM treated as non-XFD #NM in
guest (XFD_ERR cleared by the host #NM handler).

Introduce a new field in fpu_guest to save the guest xfd_err value.
KVM is expected to save guest xfd_err before interrupt is enabled
and restore it right before entering the guest (with interrupt
disabled).

Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-12-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agokvm: x86: Add emulation for IA32_XFD
Jing Liu [Wed, 5 Jan 2022 12:35:21 +0000 (04:35 -0800)]
kvm: x86: Add emulation for IA32_XFD

Intel's eXtended Feature Disable (XFD) feature allows the software
to dynamically adjust fpstate buffer size for XSAVE features which
have large state.

Because guest fpstate has been expanded for all possible dynamic
xstates at KVM_SET_CPUID2, emulation of the IA32_XFD MSR is
straightforward. For write just call fpu_update_guest_xfd() to
update the guest fpu container once all the sanity checks are passed.
For read simply return the cached value in the container.

Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Zeng Guang <guang.zeng@intel.com>
Signed-off-by: Wei Wang <wei.w.wang@intel.com>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-11-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agox86/fpu: Provide fpu_update_guest_xfd() for IA32_XFD emulation
Kevin Tian [Wed, 5 Jan 2022 12:35:20 +0000 (04:35 -0800)]
x86/fpu: Provide fpu_update_guest_xfd() for IA32_XFD emulation

Guest XFD can be updated either in the emulation path or in the
restore path.

Provide a wrapper to update guest_fpu::fpstate::xfd. If the guest
fpstate is currently in-use, also update the per-cpu xfd cache and
the actual MSR.

Signed-off-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-10-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agokvm: x86: Enable dynamic xfeatures at KVM_SET_CPUID2
Jing Liu [Wed, 5 Jan 2022 12:35:19 +0000 (04:35 -0800)]
kvm: x86: Enable dynamic xfeatures at KVM_SET_CPUID2

KVM can request fpstate expansion in two approaches:

  1) When intercepting guest updates to XCR0 and XFD MSR;

  2) Before vcpu runs (e.g. at KVM_SET_CPUID2);

The first option doesn't waste memory for legacy guest if it doesn't
support XFD. However doing so introduces more complexity and also
imposes an order requirement in the restoring path, i.e. XCR0/XFD
must be restored before XSTATE.

Given that the agreement is to do the static approach. This is
considered a better tradeoff though it does waste 8K memory for
legacy guest if its CPUID includes dynamically-enabled xfeatures.

Successful fpstate expansion requires userspace VMM to acquire
guest xstate permissions before calling KVM_SET_CPUID2.

Also take the chance to adjust the indent in kvm_set_cpuid().

Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-9-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agox86/fpu: Provide fpu_enable_guest_xfd_features() for KVM
Sean Christopherson [Wed, 5 Jan 2022 12:35:18 +0000 (04:35 -0800)]
x86/fpu: Provide fpu_enable_guest_xfd_features() for KVM

Provide a wrapper for expanding the guest fpstate buffer according
to requested xfeatures. KVM wants to call this wrapper to manage
any dynamic xstate used by the guest.

Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20220105123532.12586-8-yang.zhong@intel.com>
[Remove unnecessary 32-bit check. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agox86/fpu: Add guest support to xfd_enable_feature()
Thomas Gleixner [Wed, 5 Jan 2022 12:35:17 +0000 (04:35 -0800)]
x86/fpu: Add guest support to xfd_enable_feature()

Guest support for dynamically enabled FPU features requires a few
modifications to the enablement function which is currently invoked from
the #NM handler:

  1) Use guest permissions and sizes for the update

  2) Update fpu_guest state accordingly

  3) Take into account that the enabling can be triggered either from a
     running guest via XSETBV and MSR_IA32_XFD write emulation or from
     a guest restore. In the latter case the guests fpstate is not the
     current tasks active fpstate.

Split the function and implement the guest mechanics throughout the
callchain.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-7-yang.zhong@intel.com>
[Add 32-bit stub for __xfd_enable_feature. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agox86/fpu: Make XFD initialization in __fpstate_reset() a function argument
Jing Liu [Wed, 5 Jan 2022 12:35:16 +0000 (04:35 -0800)]
x86/fpu: Make XFD initialization in __fpstate_reset() a function argument

vCPU threads are different from native tasks regarding to the initial XFD
value. While all native tasks follow a fixed value (init_fpstate::xfd)
established by the FPU core at boot, vCPU threads need to obey the reset
value (i.e. ZERO) defined by the specification, to meet the expectation of
the guest.

Let the caller supply an argument and adjust the host and guest related
invocations accordingly.

Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220105123532.12586-6-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
3 years agomodule: fix signature check failures when using in-kernel decompression
Dmitry Torokhov [Fri, 14 Jan 2022 00:51:52 +0000 (16:51 -0800)]
module: fix signature check failures when using in-kernel decompression

The new flag MODULE_INIT_COMPRESSED_FILE unintentionally trips check in
module_sig_check(). The check was supposed to catch case when version
info or magic was removed from a signed module, making signature
invalid, but it was coded too broadly and was catching this new flag as
well.

Change the check to only test the 2 particular flags affecting signature
validity.

Fixes: b1ae6dc41eaa ("module: add in-kernel support for decompressing")
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
3 years agoALSA: hda/realtek: fix speakers and micmute on HP 855 G8
Alexander Sergeyev [Fri, 14 Jan 2022 16:50:50 +0000 (19:50 +0300)]
ALSA: hda/realtek: fix speakers and micmute on HP 855 G8

There are several PCI ids associated with HP EliteBook 855 G8 Notebook
PC. Commit 0e68c4b11f1e6 ("ALSA: hda/realtek: fix mute/micmute LEDs for
HP 855 G8") covers 0x103c:0x8896, while this commit covers 0x103c:0x8895
which needs some additional work on top of the quirk from 0e68c4b11f1e6.

Note that the device can boot up with working speakers and micmute LED
without this patch, but the success rate would be quite low (order of
16 working boots across 709 boots) at least for the built-in drivers
scenario. This also means that there are some timing issues during early
boot and this patch is a workaround.

With this patch applied speakers and headphones are consistenly working,
as well as mute/micmute LEDs and the internal microphone.

Signed-off-by: Alexander Sergeyev <sergeev917@gmail.com>
Link: https://lore.kernel.org/r/20220114165050.ouw2nknuspclynro@localhost.localdomain
Signed-off-by: Takashi Iwai <tiwai@suse.de>
3 years agoovl: don't fail copy up if no fileattr support on upper
Miklos Szeredi [Fri, 14 Jan 2022 15:57:56 +0000 (16:57 +0100)]
ovl: don't fail copy up if no fileattr support on upper

Christoph Fritz is reporting that failure to copy up fileattr when upper
doesn't support fileattr or xattr results in a regression.

Return success in these failure cases; this reverts overlayfs to the old
behavior.

Add a pr_warn_once() in these cases to still let the user know about the
copy up failures.

Reported-by: Christoph Fritz <chf.fritz@googlemail.com>
Fixes: 72db82115d2b ("ovl: copy up sync/noatime fileattr flags")
Cc: <stable@vger.kernel.org> # v5.15
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
3 years agoSUNRPC: Don't dereference xprt->snd_task if it's a cookie
Chuck Lever [Thu, 13 Jan 2022 17:20:36 +0000 (12:20 -0500)]
SUNRPC: Don't dereference xprt->snd_task if it's a cookie

Fixes: e26d9972720e ("SUNRPC: Clean up scheduling of autoclose")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
3 years agoxprtrdma: Remove definitions of RPCDBG_FACILITY
Chuck Lever [Thu, 13 Jan 2022 17:20:30 +0000 (12:20 -0500)]
xprtrdma: Remove definitions of RPCDBG_FACILITY

Deprecated. dprintk is no longer used in xprtrdma.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
3 years agoxprtrdma: Remove final dprintk call sites from xprtrdma
Chuck Lever [Thu, 13 Jan 2022 17:20:23 +0000 (12:20 -0500)]
xprtrdma: Remove final dprintk call sites from xprtrdma

Deprecated. This information is available via tracepoints.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
3 years agoMerge branch 'i2c/for-mergewindow' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 14 Jan 2022 15:19:38 +0000 (16:19 +0100)]
Merge branch 'i2c/for-mergewindow' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux

Pull i2c updates from Wolfram Sang:
 "Mostly driver updates and refactorization.

  The removal of the XLR driver and the i801 refactoring stand out a
  little. In the core, we enabled async suspend/resume for I2C
  controllers and their clients. No issues were reported during the test
  phase in -next. We will see how this goes for mainline"

* 'i2c/for-mergewindow' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux: (54 commits)
  i2c: sh_mobile: remove unneeded semicolon
  i2c: riic: Use platform_get_irq() to get the interrupt
  i2c: sh_mobile: Use platform_get_irq_optional() to get the interrupt
  i2c: bcm2835: Use platform_get_irq() to get the interrupt
  i2c: aspeed: Remove unused includes
  dt-bindings: i2c: aspeed: Drop stray '#interrupt-cells'
  i2c: sh_mobile: update to new DMAENGINE API when terminating
  i2c: rcar: update to new DMAENGINE API when terminating
  i2c: exynos5: Fix getting the optional clock
  i2c: designware-pci: Convert to use dev_err_probe()
  i2c: designware-pci: use __maybe_unused for PM functions
  i2c: designware-pci: Group MODULE_*() macros
  i2c: designware-pci: Add a note about struct dw_scl_sda_cfg usage
  i2c: designware-pci: Fix to change data types of hcnt and lcnt parameters
  i2c: designware: Do not complete i2c read without RX_FULL interrupt
  eeprom: at24: Add support for 24c1025 EEPROM
  dt-bindings: at24: add at24c1025
  i2c: tegra: use i2c_timings for bus clock freq
  dt-bindings: at24: Rework special case compatible handling
  i2c: i801: Don't clear status flags twice in interrupt mode
  ...

3 years agoMerge tag 'char-misc-5.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregk...
Linus Torvalds [Fri, 14 Jan 2022 15:02:28 +0000 (16:02 +0100)]
Merge tag 'char-misc-5.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc

Pull char/misc and other driver updates from Greg KH:
 "Here is the large set of char, misc, and other "small" driver
  subsystem changes for 5.17-rc1.

  Lots of different things are in here for char/misc drivers such as:

   - habanalabs driver updates

   - mei driver updates

   - lkdtm driver updates

   - vmw_vmci driver updates

   - android binder driver updates

   - other small char/misc driver updates

  Also smaller driver subsystems have also been updated, including:

   - fpga subsystem updates

   - iio subsystem updates

   - soundwire subsystem updates

   - extcon subsystem updates

   - gnss subsystem updates

   - phy subsystem updates

   - coresight subsystem updates

   - firmware subsystem updates

   - comedi subsystem updates

   - mhi subsystem updates

   - speakup subsystem updates

   - rapidio subsystem updates

   - spmi subsystem updates

   - virtual driver updates

   - counter subsystem updates

  Too many individual changes to summarize, the shortlog contains the
  full details.

  All of these have been in linux-next for a while with no reported
  issues"

* tag 'char-misc-5.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (406 commits)
  counter: 104-quad-8: Fix use-after-free by quad8_irq_handler
  dt-bindings: mux: Document mux-states property
  dt-bindings: ti-serdes-mux: Add defines for J721S2 SoC
  counter: remove old and now unused registration API
  counter: ti-eqep: Convert to new counter registration
  counter: stm32-lptimer-cnt: Convert to new counter registration
  counter: stm32-timer-cnt: Convert to new counter registration
  counter: microchip-tcb-capture: Convert to new counter registration
  counter: ftm-quaddec: Convert to new counter registration
  counter: intel-qep: Convert to new counter registration
  counter: interrupt-cnt: Convert to new counter registration
  counter: 104-quad-8: Convert to new counter registration
  counter: Update documentation for new counter registration functions
  counter: Provide alternative counter registration functions
  counter: stm32-timer-cnt: Convert to counter_priv() wrapper
  counter: stm32-lptimer-cnt: Convert to counter_priv() wrapper
  counter: ti-eqep: Convert to counter_priv() wrapper
  counter: ftm-quaddec: Convert to counter_priv() wrapper
  counter: intel-qep: Convert to counter_priv() wrapper
  counter: microchip-tcb-capture: Convert to counter_priv() wrapper
  ...

3 years agoovl: fix NULL pointer dereference in copy up warning
Christoph Fritz [Wed, 12 Jan 2022 18:33:21 +0000 (19:33 +0100)]
ovl: fix NULL pointer dereference in copy up warning

This patch is fixing a NULL pointer dereference to get a recently
introduced warning message working.

Fixes: 5b0a414d06c3 ("ovl: fix filattr copy-up failure")
Signed-off-by: Christoph Fritz <chf.fritz@googlemail.com>
Cc: <stable@vger.kernel.org> # v5.15
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
3 years agoperf cputopo: Fix CPU topology reading on s/390
Thomas Richter [Wed, 24 Nov 2021 09:03:43 +0000 (10:03 +0100)]
perf cputopo: Fix CPU topology reading on s/390

Commit fdf1e29b6118c18f ("perf expr: Add metric literals for topology.")
fails on s390:

 # ./perf test -Fv 7
   ...
 # FAILED tests/expr.c:173 #num_dies >= #num_packages
   ---- end ----
   Simple expression parser: FAILED!
 #

Investigating this issue leads to these functions:
 build_cpu_topology()
   +--> has_die_topology(void)
        {
           struct utsname uts;

           if (uname(&uts) < 0)
                  return false;
           if (strncmp(uts.machine, "x86_64", 6))
                  return false;
           ....
        }

which always returns false on s390. The caller build_cpu_topology()
checks has_die_topology() return value. On false the
the struct cpu_topology::die_cpu_list is not contructed and has zero
entries. This leads to the failing comparison: #num_dies >= #num_packages.
s390 of course has a positive number of packages.

Fix this by adding s390 architecture to support CPU die list.

Output after:
 # ./perf test -Fv 7
  7: Simple expression parser                                        :
  --- start ---
  division by zero
  syntax error
  ---- end ----
  Simple expression parser: Ok
 #

Fixes: fdf1e29b6118c18f ("perf expr: Add metric literals for topology.")
Reviewed-by: Ian Rogers <irogers@google.com>
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Cc: Heiko Carstens <hca@linux.ibm.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Sumanth Korikkar <sumanthk@linux.ibm.com>
Cc: Sven Schnelle <svens@linux.ibm.com>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Link: https://lore.kernel.org/r/20211124090343.9436-1-tmricht@linux.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
3 years agoperf metricgroup: Fix use after free in metric__new()
José Expósito [Wed, 8 Dec 2021 17:11:13 +0000 (18:11 +0100)]
perf metricgroup: Fix use after free in metric__new()

We shouldn't free() something that will be used in the next line, fix
it.

Fixes: b85a4d61d3022608 ("perf metric: Allow modifiers on metrics")
Addresses-Coverity-ID: 1494000
Signed-off-by: José Expósito <jose.exposito89@gmail.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20211208171113.22089-1-jose.exposito89@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
3 years agoMerge tag 'amd-drm-next-5.17-2022-01-12' of https://gitlab.freedesktop.org/agd5f...
Daniel Vetter [Fri, 14 Jan 2022 14:42:27 +0000 (15:42 +0100)]
Merge tag 'amd-drm-next-5.17-2022-01-12' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-5.17-2022-01-12:

amdgpu:
- SR-IOV fixes
- Suspend/resume fixes
- Display fixes
- DMCUB fixes
- DP alt mode fixes
- RAS fixes
- UBSAN fix
- Navy Flounder VCN fix
- ttm resource manager cleanup
- default_groups change for kobj_type
- vkms fix
- Aldebaran fixes

amdkfd:
- SDMA ECC interrupt fix
- License clarification
- Pointer check fix
- DQM fixes for hawaii
- default_groups change for kobj_type
- Typo fixes

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220113030537.5758-1-alexander.deucher@amd.com
3 years agoMerge tag 'drm-misc-next-fixes-2022-01-14' of git://anongit.freedesktop.org/drm/drm...
Daniel Vetter [Fri, 14 Jan 2022 14:41:47 +0000 (15:41 +0100)]
Merge tag 'drm-misc-next-fixes-2022-01-14' of git://anongit.freedesktop.org/drm/drm-misc into drm-next

 * atomic helpers: Fix error messages
 * mipi-dbi: Fix buffer mapping

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/YeGHu7qU92pjuQOn@linux-uq9g
3 years agolibperf tests: Update a use of the new cpumap API
Ian Rogers [Fri, 14 Jan 2022 06:51:05 +0000 (22:51 -0800)]
libperf tests: Update a use of the new cpumap API

Fixes a build breakage.

Fixes: 6d18804b963b78dc ("perf cpumap: Give CPUs their own type")
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: colin ian king <colin.king@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Shunsuke Nakamura <nakamura.shun@fujitsu.com>
Link: http://lore.kernel.org/lkml/20220114065105.1806542-1-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
3 years agoperf arm: Fix off-by-one directory path
Ian Rogers [Fri, 14 Jan 2022 06:48:22 +0000 (22:48 -0800)]
perf arm: Fix off-by-one directory path

Relative path include works in the regular build due to -I paths but may
fail in other situations.

Fixes: 83869019c74cc2d0 ("perf arch: Support register names from all archs")
Reviewed-by: German Gomez <german.gomez@arm.com>
Signed-off-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexandre Truong <alexandre.truong@arm.com>
Cc: Athira Jajeev <atrajeev@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@arm.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220114064822.1806019-1-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
3 years agotools arch x86: Sync the msr-index.h copy with the kernel sources
Arnaldo Carvalho de Melo [Fri, 7 Aug 2020 11:45:47 +0000 (08:45 -0300)]
tools arch x86: Sync the msr-index.h copy with the kernel sources

To pick up the changes in:

  89aa94b4a218339b ("x86/msr: Add AMD CPPC MSR definitions")

Addressing these tools/perf build warnings:

    diff -u tools/arch/x86/include/asm/msr-index.h arch/x86/include/asm/msr-index.h
    Warning: Kernel ABI header at 'tools/arch/x86/include/asm/msr-index.h' differs from latest version at 'arch/x86/include/asm/msr-index.h'

That makes the beautification scripts to pick some new entries:

  $ tools/perf/trace/beauty/tracepoints/x86_msr.sh > before
  $ cp arch/x86/include/asm/msr-index.h tools/arch/x86/include/asm/msr-index.h
  $ tools/perf/trace/beauty/tracepoints/x86_msr.sh > after
  $ diff -u before after
  --- before 2022-01-13 10:59:51.743416890 -0300
  +++ after 2022-01-13 11:00:00.776644178 -0300
  @@ -303,6 +303,11 @@
    [0xc0010299 - x86_AMD_V_KVM_MSRs_offset] = "AMD_RAPL_POWER_UNIT",
    [0xc001029a - x86_AMD_V_KVM_MSRs_offset] = "AMD_CORE_ENERGY_STATUS",
    [0xc001029b - x86_AMD_V_KVM_MSRs_offset] = "AMD_PKG_ENERGY_STATUS",
  +       [0xc00102b0 - x86_AMD_V_KVM_MSRs_offset] = "AMD_CPPC_CAP1",
  +       [0xc00102b1 - x86_AMD_V_KVM_MSRs_offset] = "AMD_CPPC_ENABLE",
  +       [0xc00102b2 - x86_AMD_V_KVM_MSRs_offset] = "AMD_CPPC_CAP2",
  +       [0xc00102b3 - x86_AMD_V_KVM_MSRs_offset] = "AMD_CPPC_REQ",
  +       [0xc00102b4 - x86_AMD_V_KVM_MSRs_offset] = "AMD_CPPC_STATUS",
    [0xc00102f0 - x86_AMD_V_KVM_MSRs_offset] = "AMD_PPIN_CTL",
    [0xc00102f1 - x86_AMD_V_KVM_MSRs_offset] = "AMD_PPIN",
   };
  $

And this gets rebuilt:

  CC       /tmp/build/perf/trace/beauty/tracepoints/x86_msr.o
  INSTALL  trace_plugins
  LD       /tmp/build/perf/trace/beauty/tracepoints/perf-in.o
  LD       /tmp/build/perf/trace/beauty/perf-in.o
  LD       /tmp/build/perf/perf-in.o
  LINK     /tmp/build/perf/perf

Now one can trace systemwide asking to see backtraces to where those
MSRs are being read/written with:

  # perf trace -e msr:*_msr/max-stack=32/ --filter="msr>=AMD_CPPC_CAP1 && msr<=AMD_CPPC_STATUS"
  ^C#

If we use -v (verbose mode) we can see what it does behind the scenes:

  # perf trace -v -e msr:*_msr/max-stack=32/ --filter="msr>=AMD_CPPC_CAP1 && msr<=AMD_CPPC_STATUS"
  <SNIP>
  New filter for msr:read_msr: (msr>=0xc00102b0 && msr<=0xc00102b4) && (common_pid != 2612102 && common_pid != 3841)
  New filter for msr:write_msr: (msr>=0xc00102b0 && msr<=0xc00102b4) && (common_pid != 2612102 && common_pid != 3841)
  <SNIP>
  ^C#

Example with a frequent msr:

  # perf trace -v -e msr:*_msr/max-stack=32/ --filter="msr==IA32_SPEC_CTRL" --max-events 2
  Using CPUID AuthenticAMD-25-21-0
  0x48
  New filter for msr:read_msr: (msr==0x48) && (common_pid != 2612129 && common_pid != 3841)
  0x48
  New filter for msr:write_msr: (msr==0x48) && (common_pid != 2612129 && common_pid != 3841)
  mmap size 528384B
  Looking at the vmlinux_path (8 entries long)
  symsrc__init: build id mismatch for vmlinux.
  Using /proc/kcore for kernel data
  Using /proc/kallsyms for symbols
       0.000 Timer/2525383 msr:write_msr(msr: IA32_SPEC_CTRL, val: 6)
                                         do_trace_write_msr ([kernel.kallsyms])
                                         do_trace_write_msr ([kernel.kallsyms])
                                         __switch_to_xtra ([kernel.kallsyms])
                                         __switch_to ([kernel.kallsyms])
                                         __schedule ([kernel.kallsyms])
                                         schedule ([kernel.kallsyms])
                                         futex_wait_queue_me ([kernel.kallsyms])
                                         futex_wait ([kernel.kallsyms])
                                         do_futex ([kernel.kallsyms])
                                         __x64_sys_futex ([kernel.kallsyms])
                                         do_syscall_64 ([kernel.kallsyms])
                                         entry_SYSCALL_64_after_hwframe ([kernel.kallsyms])
                                         __futex_abstimed_wait_common64 (/usr/lib64/libpthread-2.33.so)
       0.030 :0/0 msr:write_msr(msr: IA32_SPEC_CTRL, val: 2)
                                         do_trace_write_msr ([kernel.kallsyms])
                                         do_trace_write_msr ([kernel.kallsyms])
                                         __switch_to_xtra ([kernel.kallsyms])
                                         __switch_to ([kernel.kallsyms])
                                         __schedule ([kernel.kallsyms])
                                         schedule_idle ([kernel.kallsyms])
                                         do_idle ([kernel.kallsyms])
                                         cpu_startup_entry ([kernel.kallsyms])
                                         secondary_startup_64_no_verify ([kernel.kallsyms])
  #

Acked-by: Huang Rui <ray.huang@amd.com>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/all/YeA2PAvHV+uHRhLj@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
3 years agoMerge tag 'for-linus' of git://github.com/openrisc/linux
Linus Torvalds [Fri, 14 Jan 2022 14:32:09 +0000 (15:32 +0100)]
Merge tag 'for-linus' of git://github.com/openrisc/linux

Pull OpenRISC updates from Stafford Horne:
 "A few fixups and enhancements for OpenRISC:

   - Fix to add proper wrapper for clone3 to save callee saved regs

   - Cleanups for clone, fork and switch

   - Add support for common clk so OpenRISC and use more drivers"

* tag 'for-linus' of git://github.com/openrisc/linux:
  openrisc: init: Add support for common clk
  openrisc: Add clone3 ABI wrapper
  openrisc: Use delay slot for clone and fork wrappers
  openrisc: Cleanup switch code and comments

3 years agoMerge tag 'powerpc-5.17-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc...
Linus Torvalds [Fri, 14 Jan 2022 14:17:26 +0000 (15:17 +0100)]
Merge tag 'powerpc-5.17-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux

Pull powerpc updates from Michael Ellerman:

 - Optimise radix KVM guest entry/exit by 2x on Power9/Power10.

 - Allow firmware to tell us whether to disable the entry and uaccess
   flushes on Power10 or later CPUs.

 - Add BPF_PROBE_MEM support for 32 and 64-bit BPF jits.

 - Several fixes and improvements to our hard lockup watchdog.

 - Activate HAVE_DYNAMIC_FTRACE_WITH_REGS on 32-bit.

 - Allow building the 64-bit Book3S kernel without hash MMU support, ie.
   Radix only.

 - Add KUAP (SMAP) support for 40x, 44x, 8xx, Book3E (64-bit).

 - Add new encodings for perf_mem_data_src.mem_hops field, and use them
   on Power10.

 - A series of small performance improvements to 64-bit interrupt entry.

 - Several commits fixing issues when building with the clang integrated
   assembler.

 - Many other small features and fixes.

Thanks to Alan Modra, Alexey Kardashevskiy, Ammar Faizi, Anders Roxell,
Arnd Bergmann, Athira Rajeev, Cédric Le Goater, Christophe JAILLET,
Christophe Leroy, Christoph Hellwig, Daniel Axtens, David Yang, Erhard
Furtner, Fabiano Rosas, Greg Kroah-Hartman, Guo Ren, Hari Bathini, Jason
Wang, Joel Stanley, Julia Lawall, Kajol Jain, Kees Cook, Laurent Dufour,
Madhavan Srinivasan, Mark Brown, Minghao Chi, Nageswara R Sastry, Naresh
Kamboju, Nathan Chancellor, Nathan Lynch, Nicholas Piggin, Nick Child,
Oliver O'Halloran, Peiwei Hu, Randy Dunlap, Ravi Bangoria, Rob Herring,
Russell Currey, Sachin Sant, Sean Christopherson, Segher Boessenkool,
Thadeu Lima de Souza Cascardo, Tyrel Datwyler, Xiang wangx, and Yang
Guang.

* tag 'powerpc-5.17-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux: (240 commits)
  powerpc/xmon: Dump XIVE information for online-only processors.
  powerpc/opal: use default_groups in kobj_type
  powerpc/cacheinfo: use default_groups in kobj_type
  powerpc/sched: Remove unused TASK_SIZE_OF
  powerpc/xive: Add missing null check after calling kmalloc
  powerpc/floppy: Remove usage of the deprecated "pci-dma-compat.h" API
  selftests/powerpc: Add a test of sigreturning to an unaligned address
  powerpc/64s: Use EMIT_WARN_ENTRY for SRR debug warnings
  powerpc/64s: Mask NIP before checking against SRR0
  powerpc/perf: Fix spelling of "its"
  powerpc/32: Fix boot failure with GCC latent entropy plugin
  powerpc/code-patching: Replace patch_instruction() by ppc_inst_write() in selftests
  powerpc/code-patching: Move code patching selftests in its own file
  powerpc/code-patching: Move instr_is_branch_{i/b}form() in code-patching.h
  powerpc/code-patching: Move patch_exception() outside code-patching.c
  powerpc/code-patching: Use test_trampoline for prefixed patch test
  powerpc/code-patching: Fix patch_branch() return on out-of-range failure
  powerpc/code-patching: Reorganise do_patch_instruction() to ease error handling
  powerpc/code-patching: Fix unmap_patch_area() error handling
  powerpc/code-patching: Fix error handling in do_patch_instruction()
  ...

3 years agoMerge tag 'drm-misc-fixes-2022-01-14' of git://anongit.freedesktop.org/drm/drm-misc...
Daniel Vetter [Fri, 14 Jan 2022 14:15:56 +0000 (15:15 +0100)]
Merge tag 'drm-misc-fixes-2022-01-14' of git://anongit.freedesktop.org/drm/drm-misc into drm-next

Two DT bindings fixes for meson, a device refcounting fix for sun4i, a
probe fix for vga16fb, a locking fix for the CMA dma-buf heap and a
compilation fix for ttm.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
[danvet: I made sure I have exactly the same conflict resolution as
Linus in 8d0749b4f83b ("Merge tag 'drm-next-2022-01-07' of
git://anongit.freedesktop.org/drm/drm") to avoid further conflict fun.
From: Maxime Ripard <maxime@cerno.tech>
Link: https://patchwork.freedesktop.org/patch/msgid/20220114125454.zs46ny52lrxk3ljz@houat
3 years agoKVM: arm64: pkvm: Use the mm_ops indirection for cache maintenance
Marc Zyngier [Fri, 14 Jan 2022 08:57:58 +0000 (08:57 +0000)]
KVM: arm64: pkvm: Use the mm_ops indirection for cache maintenance

CMOs issued from EL2 cannot directly use the kernel helpers,
as EL2 doesn't have a mapping of the guest pages. Oops.

Instead, use the mm_ops indirection to use helpers that will
perform a mapping at EL2 and allow the CMO to be effective.

Fixes: 25aa28691bb9 ("KVM: arm64: Move guest CMOs to the fault handlers")
Reviewed-by: Quentin Perret <qperret@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20220114125038.1336965-1-maz@kernel.org
3 years agoMerge tag 'mips_5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux
Linus Torvalds [Fri, 14 Jan 2022 14:08:36 +0000 (15:08 +0100)]
Merge tag 'mips_5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux

Pull MIPS updates from Thomas Bogendoerfer:

 - add support for more BCM47XX based devices

 - add MIPS support for brcmstb PCIe controller

 - add Loongson 2K1000 reset driver

 - remove board support for rbtx4938/rbtx4939

 - remove support for TX4939 SoCs

 - fixes and cleanups

* tag 'mips_5.17' of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux: (59 commits)
  MIPS: ath79: drop _machine_restart again
  PCI: brcmstb: Augment driver for MIPs SOCs
  MIPS: bmips: Remove obsolete DMA mapping support
  MIPS: bmips: Add support PCIe controller device nodes
  dt-bindings: PCI: Add compatible string for Brcmstb 74[23]5 MIPs SOCs
  MIPS: compressed: Fix build with ZSTD compression
  MIPS: BCM47XX: Add support for Netgear WN2500RP v1 & v2
  MIPS: BCM47XX: Add support for Netgear R6300 v1
  MIPS: BCM47XX: Add LEDs and buttons for Asus RTN-10U
  MIPS: BCM47XX: Add board entry for Linksys WRT320N v1
  MIPS: BCM47XX: Define Linksys WRT310N V2 buttons
  MIPS: Remove duplicated include in local.h
  MIPS: retire "asm/llsc.h"
  MIPS: rework local_t operation on MIPS64
  MIPS: fix local_{add,sub}_return on MIPS64
  mips/pci: remove redundant ret variable
  MIPS: Loongson64: Add missing of_node_put() in ls2k_reset_init()
  MIPS: new Kconfig option ZBOOT_LOAD_ADDRESS
  MIPS: enable both vmlinux.gz.itb and vmlinuz for generic
  MIPS: signal: Return immediately if call fails
  ...

3 years agoMerge tag 'sound-5.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai...
Linus Torvalds [Fri, 14 Jan 2022 13:55:38 +0000 (14:55 +0100)]
Merge tag 'sound-5.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound

Pull sound updates from Takashi Iwai:
 "It's a relatively calm development cycle, but still lots of updates in
  the driver side like Intel SOF. Below are some highlights:

  ALSA / ASoC core:
   - A new kselftest for ALSA control API
   - PCM NO_REWINDS support
   - Potential race fixes around control removals
   - Unify x86 SG-buffer memory allocation code
   - Cleanups and race fixes for ASoC DPCM locking

  ASoC:
   - Refinements and cleanups around the delay() APIs
   - Wider use of dev_err_probe().
   - Continuing cleanups and improvements to the SOF code
   - Support for pin switches in simple-card derived cards
   - Support for AMD Renoir ACP, Asahi Kasei Microdevices AKM4375, Intel
     systems using NAU8825 and MAX98390, Mediatek MT8915, nVidia Tegra20
     S/PDIF, Qualcomm systems using ALC5682I-VS and Texas Instruments
     TLV320ADC3xxx

  HD-audio / USB-audio:
   - Fix deadlock at HD-audio codec unbinding
   - Fixes for Tegra194 HD-audio, new HDA support for CS35L41 codec
   - Quirks for Lenovo and HP machines, Gigabyte mobo, Bose device

  Misc:
   - Fix virmidi drain behavior

  Note that the merge of CS35L41 codec support is still half-baked, and
  at least one ACPI change is missing. Although this won't hinder the
  kernel build itself, we're going to catch up before RC1"

* tag 'sound-5.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (415 commits)
  ALSA: hda: intel-dsp-config: reorder the config table
  ALSA: hda: intel-dsp-config: add JasperLake support
  ALSA: hda: cs35l41: fix double free on error in probe()
  ALSA: hda: Fix dependencies of CS35L41 on SPI/I2C buses
  ALSA: hda: Fix dependency on ASoC cs35l41 codec
  ASoC: cs35l41: Add support for hibernate memory retention mode
  ASoC: cs35l41: Update handling of test key registers
  ALSA: intel_hdmi: Check for error num after setting mask
  ASoC: wcd9335: Keep a RX port value for each SLIM RX mux
  ASoC: amd: acp: acp-mach: Change default RT1019 amp dev id
  ALSA: virmidi: Remove duplicated code
  ALSA: seq: virmidi: Add a drain operation
  ASoC: topology: Fix typo
  ASoC: fsl_asrc: refine the check of available clock divider
  ASoC: Intel: bytcr_rt5640: Add support for external GPIO jack-detect
  ASoC: Intel: bytcr_rt5640: Support retrieving the codec IRQ from the AMCR0F28 ACPI dev
  ASoC: rt5640: Add support for boards with an external jack-detect GPIO
  ASoC: rt5640: Allow snd_soc_component_set_jack() to override the codec IRQ
  ASoC: rt5640: Change jack_work to a delayed_work
  ASoC: rt5640: Fix possible NULL pointer deref on resume
  ...

3 years agoio_uring: fix UAF due to missing POLLFREE handling
Pavel Begunkov [Fri, 14 Jan 2022 11:59:10 +0000 (11:59 +0000)]
io_uring: fix UAF due to missing POLLFREE handling

Fixes a problem described in 50252e4b5e989
("aio: fix use-after-free due to missing POLLFREE handling")
and copies the approach used there.

In short, we have to forcibly eject a poll entry when we meet POLLFREE.
We can't rely on io_poll_get_ownership() as can't wait for potentially
running tw handlers, so we use the fact that wqs are RCU freed. See
Eric's patch and comments for more details.

Reported-by: Eric Biggers <ebiggers@google.com>
Link: https://lore.kernel.org/r/20211209010455.42744-6-ebiggers@kernel.org
Reported-and-tested-by: syzbot+5426c7ed6868c705ca14@syzkaller.appspotmail.com
Fixes: 221c5eb233823 ("io_uring: add support for IORING_OP_POLL")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/4ed56b6f548f7ea337603a82315750449412748a.1642161259.git.asml.silence@gmail.com
[axboe: drop non-functional change from patch]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
3 years agodrm/mipi-dbi: Fix source-buffer address in mipi_dbi_buf_copy
Thomas Zimmermann [Tue, 11 Jan 2022 13:26:34 +0000 (14:26 +0100)]
drm/mipi-dbi: Fix source-buffer address in mipi_dbi_buf_copy

Set the source-buffer address after mapping the buffer into the
kernel's address space. Makes MIPI DBI helpers work again.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Fixes: c47160d8edcd ("drm/mipi-dbi: Remove dependency on GEM CMA helper library")
Reviewed-by: Noralf Trønnes <noralf@tronnes.org>
Reported-by: Noralf Trønnes <noralf@tronnes.org>
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Maxime Ripard <mripard@kernel.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20220111132634.18302-1-tzimmermann@suse.de
(cherry picked from commit 9d31993451f6bb4059a9b9eec4856b2225e36df0)
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
3 years agoMerge tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Linus Torvalds [Fri, 14 Jan 2022 13:37:34 +0000 (14:37 +0100)]
Merge tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi

Pull SCSI updates from James Bottomley:
 "This series consists of the usual driver updates (ufs, pm80xx, lpfc,
  mpi3mr, mpt3sas, hisi_sas, libsas) and minor updates and bug fixes.

  The most impactful change is likely the switch from GFP_DMA to
  GFP_KERNEL in a bunch of drivers, but even that shouldn't affect too
  many people"

* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (121 commits)
  scsi: mpi3mr: Bump driver version to 8.0.0.61.0
  scsi: mpi3mr: Fixes around reply request queues
  scsi: mpi3mr: Enhanced Task Management Support Reply handling
  scsi: mpi3mr: Use TM response codes from MPI3 headers
  scsi: mpi3mr: Add io_uring interface support in I/O-polled mode
  scsi: mpi3mr: Print cable mngnt and temp threshold events
  scsi: mpi3mr: Support Prepare for Reset event
  scsi: mpi3mr: Add Event acknowledgment logic
  scsi: mpi3mr: Gracefully handle online FW update operation
  scsi: mpi3mr: Detect async reset that occurred in firmware
  scsi: mpi3mr: Add IOC reinit function
  scsi: mpi3mr: Handle offline FW activation in graceful manner
  scsi: mpi3mr: Code refactor of IOC init - part2
  scsi: mpi3mr: Code refactor of IOC init - part1
  scsi: mpi3mr: Fault IOC when internal command gets timeout
  scsi: mpi3mr: Display IOC firmware package version
  scsi: mpi3mr: Handle unaligned PLL in unmap cmnds
  scsi: mpi3mr: Increase internal cmnds timeout to 60s
  scsi: mpi3mr: Do access status validation before adding devices
  scsi: mpi3mr: Add support for PCIe Managed Switch SES device
  ...

3 years agodrm: fix error found in some cases after the patch d1af5cd86997
Claudio Suarez [Thu, 2 Dec 2021 09:51:12 +0000 (10:51 +0100)]
drm: fix error found in some cases after the patch d1af5cd86997

The patch d1af5cd86997 ("drm: get rid of DRM_DEBUG_* log
calls in drm core, files drm_a*.c") fails when the drm_device
cannot be found in the parameter plane_state->crtc.
Fix it using plane_state->plane.

Reported-by: kernel test robot <oliver.sang@intel.com>
Fixes: 6e22dc358377 ("drm: get rid of DRM_DEBUG_* log calls in drm core, files drm_a*.c")
Signed-off-by: Claudio Suarez <cssk@net-c.es>
[danvet: fix Fixes: line]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/YaiXEARd8z2C463h@gineta.localdomain
(cherry picked from commit 521d459b1935628a7caa0753429f880dae2dbfc9)
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
3 years agoMerge tag 'drm-misc-next-fixes-2022-01-13' of git://anongit.freedesktop.org/drm/drm...
Daniel Vetter [Fri, 14 Jan 2022 12:39:50 +0000 (13:39 +0100)]
Merge tag 'drm-misc-next-fixes-2022-01-13' of git://anongit.freedesktop.org/drm/drm-misc into drm-next

 * Fix use of CRTC state's active vs enable in atomic helper

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/Yd/i/zj9vEHisSSB@linux-uq9g
3 years agoMerge branch 'for-5.17/kallsyms' into for-linus
Petr Mladek [Fri, 14 Jan 2022 12:36:32 +0000 (13:36 +0100)]
Merge branch 'for-5.17/kallsyms' into for-linus

3 years agoMerge tag 'drm-intel-next-fixes-2022-01-13' of git://anongit.freedesktop.org/drm...
Daniel Vetter [Fri, 14 Jan 2022 12:34:39 +0000 (13:34 +0100)]
Merge tag 'drm-intel-next-fixes-2022-01-13' of git://anongit.freedesktop.org/drm/drm-intel into drm-next

- Hold runtime PM wakelock during PXP unbind (Juston Li)
- Three fixes for the TTM backend fault handling (Matthew Auld)
- Make sure to unmap when purging in the TTM backend (Matthew Auld)

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/Yd/xzyCM87rfrwQT@tursulin-mobl2
3 years agodrm/ttm: fix compilation on ARCH=um
Johannes Berg [Mon, 20 Dec 2021 10:15:22 +0000 (11:15 +0100)]
drm/ttm: fix compilation on ARCH=um

Even if it's probably not really useful, it can get selected
by e.g. randconfig builds, and then failing to compile is an
annoyance. Unfortunately, it's hard to fix in Kconfig, since
DRM_TTM is selected by many things that don't really depend
on any specific architecture, and just depend on PCI (which
is indeed now available in ARCH=um via simulation/emulation).

Fix this in the code instead by just ifdef'ing the relevant
two lines that depend on "real X86".

Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20211220111519.a4c8c6eff702.Ie4cf4e68698f6a9f546b83379bc52c266504424f@changeid
3 years agonet: stmmac: Fix "Unbalanced pm_runtime_enable!" warning
Kai-Heng Feng [Fri, 14 Jan 2022 07:14:30 +0000 (15:14 +0800)]
net: stmmac: Fix "Unbalanced pm_runtime_enable!" warning

If the device is PCI based like intel-eth-pci, pm_runtime_enable() is
already called by pci_pm_init().

So only pm_runtime_enable() when it's not already enabled.

Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agolib82596: Fix IRQ check in sni_82596_probe
Miaoqian Lin [Fri, 14 Jan 2022 06:57:24 +0000 (06:57 +0000)]
lib82596: Fix IRQ check in sni_82596_probe

platform_get_irq() returns negative error number instead 0 on failure.
And the doc of platform_get_irq() provides a usage example:

    int irq = platform_get_irq(pdev, 0);
    if (irq < 0)
        return irq;

Fix the check of return value to catch errors correctly.

Fixes: 115978859272 ("i825xx: Move the Intel 82586/82593/82596 based drivers")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: apple: bmac: Fix build since dev_addr constification
Michael Ellerman [Fri, 14 Jan 2022 03:13:16 +0000 (14:13 +1100)]
net: apple: bmac: Fix build since dev_addr constification

Since commit adeef3e32146 ("net: constify netdev->dev_addr") the bmac
driver no longer builds with the following errors (pmac32_defconfig):

  linux/drivers/net/ethernet/apple/bmac.c: In function ‘bmac_probe’:
  linux/drivers/net/ethernet/apple/bmac.c:1287:20: error: assignment of read-only location ‘*(dev->dev_addr + (sizetype)j)’
   1287 |   dev->dev_addr[j] = rev ? bitrev8(addr[j]): addr[j];
        |                    ^

Fix it by making the modifications to a local macaddr variable and then
passing that to eth_hw_addr_set().

We don't use the existing addr variable because the bitrev8() would
mutate it, but it is already used unreversed later in the function.

Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: apple: mace: Fix build since dev_addr constification
Michael Ellerman [Fri, 14 Jan 2022 03:12:52 +0000 (14:12 +1100)]
net: apple: mace: Fix build since dev_addr constification

Since commit adeef3e32146 ("net: constify netdev->dev_addr") the mace
driver no longer builds with various errors (pmac32_defconfig):

  linux/drivers/net/ethernet/apple/mace.c: In function ‘mace_probe’:
  linux/drivers/net/ethernet/apple/mace.c:170:20: error: assignment of read-only location ‘*(dev->dev_addr + (sizetype)j)’
    170 |   dev->dev_addr[j] = rev ? bitrev8(addr[j]): addr[j];
        |                    ^
  linux/drivers/net/ethernet/apple/mace.c: In function ‘mace_reset’:
  linux/drivers/net/ethernet/apple/mace.c:349:32: warning: passing argument 2 of ‘__mace_set_address’ discards ‘const’ qualifier from pointer target type
    349 |     __mace_set_address(dev, dev->dev_addr);
        |                             ~~~^~~~~~~~~~
  linux/drivers/net/ethernet/apple/mace.c:93:62: note: expected ‘void *’ but argument is of type ‘const unsigned char *’
     93 | static void __mace_set_address(struct net_device *dev, void *addr);
        |                                                        ~~~~~~^~~~
  linux/drivers/net/ethernet/apple/mace.c: In function ‘__mace_set_address’:
  linux/drivers/net/ethernet/apple/mace.c:388:36: error: assignment of read-only location ‘*(dev->dev_addr + (sizetype)i)’
    388 |  out_8(&mb->padr, dev->dev_addr[i] = p[i]);
        |                                    ^

Fix it by making the modifications to a local macaddr variable and then
passing that to eth_hw_addr_set(), as well as adding some missing const
qualifiers.

Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agokselftests/net: list all available tests in usage()
Li Zhijian [Fri, 14 Jan 2022 03:02:46 +0000 (11:02 +0800)]
kselftests/net: list all available tests in usage()

So that users can run/query them easily.

$ ./fcnal-test.sh -h
usage: fcnal-test.sh OPTS

-4          IPv4 tests only
-6          IPv6 tests only
-t <test>   Test name/set to run
-p          Pause on fail
-P          Pause after each test
-v          Be verbose

Tests:
ipv4_ping ipv4_tcp ipv4_udp ipv4_bind ipv4_runtime ipv4_netfilter ipv6_ping ipv6_tcp ipv6_udp ipv6_bind ipv6_runtime ipv6_netfilter use_cases

Suggested-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agonet: usb: Correct reset handling of smsc95xx
Markus Reichl [Thu, 13 Jan 2022 20:01:11 +0000 (21:01 +0100)]
net: usb: Correct reset handling of smsc95xx

On boards with LAN9514 and no preconfigured MAC address we don't get an
ip address from DHCP after commit a049a30fc27c ("net: usb: Correct PHY handling
of smsc95xx") anymore. Adding an explicit reset before starting the phy
fixes the issue.

[1]
https://lore.kernel.org/netdev/199eebbd6b97f52b9119c9fa4fd8504f8a34de18.camel@collabora.com/

From: Gabriel Hojda <ghojda@yo2urs.ro>
Fixes: a049a30fc27c ("net: usb: Correct PHY handling of smsc95xx")
Signed-off-by: Gabriel Hojda <ghojda@yo2urs.ro>
Signed-off-by: Markus Reichl <m.reichl@fivetechno.de>
Tested-by: Alexander Stein <alexander.stein@ew.tq-group.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agobcmgenet: add WOL IRQ check
Sergey Shtylyov [Thu, 13 Jan 2022 19:46:07 +0000 (22:46 +0300)]
bcmgenet: add WOL IRQ check

The driver neglects to check the result of platform_get_irq_optional()'s
call and blithely passes the negative error codes to devm_request_irq()
(which takes *unsigned* IRQ #), causing it to fail with -EINVAL.
Stop calling devm_request_irq() with the invalid IRQ #s.

Fixes: 8562056f267d ("net: bcmgenet: request Wake-on-LAN interrupt")
Signed-off-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoHID: vivaldi: Minor cleanups
Stephen Boyd [Fri, 7 Jan 2022 23:23:05 +0000 (15:23 -0800)]
HID: vivaldi: Minor cleanups

Perform some minor cleanups on this driver. Include header files for
struct definitions that are used, drop a forward declaration that isn't
useful, and mark a sysfs attribute static as it isn't used outside this
file.

Cc: Sean O'Brien <seobrien@chromium.org>
Cc: Ting Shen <phoenixshen@chromium.org>
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
3 years agotracing: Add ustring operation to filtering string pointers
Steven Rostedt [Fri, 14 Jan 2022 01:08:40 +0000 (20:08 -0500)]
tracing: Add ustring operation to filtering string pointers

Since referencing user space pointers is special, if the user wants to
filter on a field that is a pointer to user space, then they need to
specify it.

Add a ".ustring" attribute to the field name for filters to state that the
field is pointing to user space such that the kernel can take the
appropriate action to read that pointer.

Link: https://lore.kernel.org/all/yt9d8rvmt2jq.fsf@linux.ibm.com/
Fixes: 77360f9bbc7e ("tracing: Add test for user space strings when filtering on string pointers")
Tested-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
3 years agoHID: vivaldi: fix handling devices not using numbered reports
Dmitry Torokhov [Fri, 7 Jan 2022 20:09:36 +0000 (12:09 -0800)]
HID: vivaldi: fix handling devices not using numbered reports

Unfortunately details of USB HID transport bled into HID core and
handling of numbered/unnumbered reports is quite a mess, with
hid_report_len() calculating the length according to USB rules,
and hid_hw_raw_request() adding report ID to the buffer for both
numbered and unnumbered reports.

Untangling it all requres a lot of changes in HID, so for now let's
handle this in the driver.

[jkosina@suse.cz: microoptimize field->report->id to report->id]
Fixes: 14c9c014babe ("HID: add vivaldi HID driver")
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Tested-by: Stephen Boyd <swboyd@chromium.org> # CoachZ
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
3 years agoata: pata_ali: remove redundant return statement
Colin Ian King [Wed, 12 Jan 2022 23:47:41 +0000 (23:47 +0000)]
ata: pata_ali: remove redundant return statement

A return statement is unnecessarily complicated, currently value
in variable mask is bitwise-masked and the variable is being
updated and then returned. Just updating the mask is all that is
required as the following statement is a return.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
3 years agoata: ahci: Add support for AMD A85 FCH (Hudson D4)
Paul Menzel [Wed, 5 Jan 2022 15:36:18 +0000 (16:36 +0100)]
ata: ahci: Add support for AMD A85 FCH (Hudson D4)

Add support for the AMD A85 FCH (Hudson D4) AHCI adapter.

Since this adapter does not require the default 200 ms debounce delay
in sata_link_resume(), create a new board board_ahci_no_debounce_delay
with the link flag ATA_LFLAG_NO_DEBOUNCE_DELAY, and, for now, configure
the AMD A85 FCH (Hudson D4) to use it. On the ASUS F2A85-M PRO it
reduces the Linux kernel boot time by the expected 200 ms from 787 ms
to 585 ms.

Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
3 years agoata: libata: Rename link flag ATA_LFLAG_NO_DB_DELAY
Paul Menzel [Wed, 5 Jan 2022 15:36:16 +0000 (16:36 +0100)]
ata: libata: Rename link flag ATA_LFLAG_NO_DB_DELAY

Rename the link flag ATA_LFLAG_NO_DB_DELAY to
ATA_LFLAG_NO_DEBOUNCE_DELAY. The new name is longer, but clearer.

Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
3 years agoata: libata-scsi: simplify __ata_scsi_queuecmd()
Wenchao Hao [Thu, 6 Jan 2022 00:13:54 +0000 (19:13 -0500)]
ata: libata-scsi: simplify __ata_scsi_queuecmd()

This patch cleans up the code of __ata_scsi_queuecmd(). Since each
branch of the "if" condition check that scmd->cmd_len is not zero, move
this check out of the "if" to simplify the conditions being checked in
the "else" branch.

While at it, avoid the if-else-if-else structure using if-else if
structure and remove the redundant rc local variable.

This patch does not change the function logic.

Signed-off-by: Wenchao Hao <haowenchao@huawei.com>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
3 years agoata: pata_of_platform: Use platform_get_irq_optional() to get the interrupt
Lad Prabhakar [Wed, 5 Jan 2022 18:17:21 +0000 (18:17 +0000)]
ata: pata_of_platform: Use platform_get_irq_optional() to get the interrupt

platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
allocation of IRQ resources in DT core code, this causes an issue
when using hierarchical interrupt domains using "interrupts" property
in the node as this bypasses the hierarchical setup and messes up the
irq chaining.

In preparation for removal of static setup of IRQ resource from DT core
code use platform_get_irq_optional().

Note the code does not set the IRQ flags as this is handled
automatically for DT.

Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
3 years agoata: pata_samsung_cf: add compile test support
Damien Le Moal [Tue, 4 Jan 2022 06:48:17 +0000 (15:48 +0900)]
ata: pata_samsung_cf: add compile test support

Add Kconfig dependendy on COMPILE_TEST to allow compile tests with
configs that do not enable SAMSUNG_DEV_IDE.

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
3 years agoata: pata_pxa: add compile test support
Damien Le Moal [Tue, 4 Jan 2022 06:35:57 +0000 (15:35 +0900)]
ata: pata_pxa: add compile test support

Add Kconfig dependendy on COMPILE_TEST to allow compile tests with
configs that do not enable ARCH_PXA.

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
3 years agoata: pata_imx: add compile test support
Damien Le Moal [Tue, 4 Jan 2022 06:31:34 +0000 (15:31 +0900)]
ata: pata_imx: add compile test support

Add Kconfig dependendy on COMPILE_TEST to allow compile tests with
configs that do not enable ARCH_MXC.

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
3 years agoata: pata_ftide010: add compile test support
Damien Le Moal [Tue, 4 Jan 2022 06:24:52 +0000 (15:24 +0900)]
ata: pata_ftide010: add compile test support

Add Kconfig dependendy on COMPILE_TEST to allow compile tests with
configs that do not enable ARM.

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
3 years agoata: pata_cs5535: add compile test support
Damien Le Moal [Tue, 4 Jan 2022 06:19:30 +0000 (15:19 +0900)]
ata: pata_cs5535: add compile test support

Add Kconfig dependendy on X86_64 && COMPILE_TEST to allow compile tests
with configs that do not have X86_32 enabled on X86_64 hosts.

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
3 years agoata: pata_octeon_cf: remove redundant val variable
Minghao Chi [Tue, 4 Jan 2022 11:25:45 +0000 (11:25 +0000)]
ata: pata_octeon_cf: remove redundant val variable

Return value from DIV_ROUND_UP() directly instead
of taking this in another redundant variable.

Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi <chi.minghao@zte.com.cn>
Signed-off-by: CGEL ZTE <cgel.zte@gmail.com>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
3 years agoata: fix read_id() ata port operation interface
Damien Le Moal [Tue, 4 Jan 2022 08:54:18 +0000 (17:54 +0900)]
ata: fix read_id() ata port operation interface

Drivers that need to tweak a device IDENTIFY data implement the
read_id() port operation. The IDENTIFY data buffer is passed as an
argument to the read_id() operation for drivers to use. However, when
this operation is called, the IDENTIFY data is not yet converted to CPU
endian and contains le16 words.

Change the interface of the read_id operation to pass a __le16 * pointer
to the IDENTIFY data buffer to clarify the buffer endianness. Fix the
pata_netcell, pata_it821x, ahci_xgene, ahci_ceva and ahci_brcm drivers
implementation of this operation and modify the code to corretly deal
with identify data words manipulation to avoid sparse warnings such as:

drivers/ata/ahci_xgene.c:262:33: warning: invalid assignment: &=
drivers/ata/ahci_xgene.c:262:33:    left side has type unsigned short
drivers/ata/ahci_xgene.c:262:33:    right side has type restricted __le16

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
3 years agoata: ahci_xgene: use correct type for port mmio address
Damien Le Moal [Tue, 4 Jan 2022 08:49:54 +0000 (17:49 +0900)]
ata: ahci_xgene: use correct type for port mmio address

Sparse complains about an incorrect type for port_mmio pointer
variables:

drivers/ata/ahci_xgene.c:196:41: warning: incorrect type in initializer
(different address spaces)
drivers/ata/ahci_xgene.c:196:41:    expected void *port_mmio
drivers/ata/ahci_xgene.c:196:41:    got void [noderef] __iomem *

Fix this by declaring port_mmio as "void __iomem *" instead of "void *".

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
3 years agoata: sata_fsl: fix cmdhdr_tbl_entry and prde struct definitions
Damien Le Moal [Tue, 4 Jan 2022 10:04:49 +0000 (19:04 +0900)]
ata: sata_fsl: fix cmdhdr_tbl_entry and prde struct definitions

The fields of the cmdhdr_tbl_entry structure all store __le32 values,
and so are the dba and ddc_and_ext fields of the prde structure. Define
these fields using the __le32 type to avoid sparse warnings about
incorrect type in assignment.

The debug message in sata_fsl_setup_cmd_hdr_entry() is changed to
display the correct values of the cmdhdr_tbl_entry fields on big endian
systems.

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
3 years agoata: sata_fsl: fix scsi host initialization
Damien Le Moal [Tue, 4 Jan 2022 07:14:46 +0000 (16:14 +0900)]
ata: sata_fsl: fix scsi host initialization

When compiling with W=1, the sata_fsl driver compilation throws the
warning:

drivers/ata/sata_fsl.c:1385:22: error: initialized field overwritten
[-Werror=override-init]
 1385 |         .can_queue = SATA_FSL_QUEUE_DEPTH,

This is due to the driver scsi host template initialization overwriting
the can_queue field that is already set using the ATA_NCQ_SHT()
initializer macro, resulting in the same field being initialized twice
in the host template declaration.

To remove this warning, introduce the ATA_SUBBASE_SHT_QD() and
ATA_NCQ_SHT_QD() initialization macros to allow specifying a queue depth
different from the default ATA_DEF_QUEUE using an additional argument to
the macro.

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
3 years agoata: pata_bk3710: add compile test support
Damien Le Moal [Tue, 4 Jan 2022 06:17:42 +0000 (15:17 +0900)]
ata: pata_bk3710: add compile test support

Add Kconfig dependendy on COMPILE_TEST to allow compile tests with
configs that do not enable ARCH_DAVINCI.

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
3 years agoata: ahci_seattle: add compile test support
Damien Le Moal [Tue, 4 Jan 2022 06:11:15 +0000 (15:11 +0900)]
ata: ahci_seattle: add compile test support

Add Kconfig dependendy on COMPILE_TEST to allow compile tests with
configs that do not enable ARCH_SEATTLE.

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
3 years agoata: ahci_xgene: add compile test support
Damien Le Moal [Tue, 4 Jan 2022 06:09:23 +0000 (15:09 +0900)]
ata: ahci_xgene: add compile test support

Add Kconfig dependendy on COMPILE_TEST to allow compile tests with
configs that do not enable PHY_XGENE.

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
3 years agoata: ahci_tegra: add compile test support
Damien Le Moal [Tue, 4 Jan 2022 06:08:01 +0000 (15:08 +0900)]
ata: ahci_tegra: add compile test support

Add Kconfig dependendy on COMPILE_TEST to allow compile tests with
configs that do not enable ARCH_TEGRA.

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
3 years agoata: ahci_sunxi: add compile test support
Damien Le Moal [Tue, 4 Jan 2022 06:06:12 +0000 (15:06 +0900)]
ata: ahci_sunxi: add compile test support

Add Kconfig dependendy on COMPILE_TEST to allow compile tests with
configs that do not enable ARCH_SUNXI.

Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>