linux-stable.git/drivers/hv, branch linux-4.9.y

Drivers: hv: Never allocate anything besides framebuffer from framebuffer memory region

2022-09-28T08:55:47+00:00

[ Upstream commit f0880e2cb7e1f8039a048fdd01ce45ab77247221 ]

Passed through PCI device sometimes misbehave on Gen1 VMs when Hyper-V
DRM driver is also loaded. Looking at IOMEM assignment, we can see e.g.

$ cat /proc/iomem
...
f8000000-fffbffff : PCI Bus 0000:00
  f8000000-fbffffff : 0000:00:08.0
    f8000000-f8001fff : bb8c4f33-2ba2-4808-9f7f-02f3b4da22fe
...
fe0000000-fffffffff : PCI Bus 0000:00
  fe0000000-fe07fffff : bb8c4f33-2ba2-4808-9f7f-02f3b4da22fe
    fe0000000-fe07fffff : 2ba2:00:02.0
      fe0000000-fe07fffff : mlx4_core

the interesting part is the 'f8000000' region as it is actually the
VM's framebuffer:

$ lspci -v
...
0000:00:08.0 VGA compatible controller: Microsoft Corporation Hyper-V virtual VGA (prog-if 00 [VGA controller])
	Flags: bus master, fast devsel, latency 0, IRQ 11
	Memory at f8000000 (32-bit, non-prefetchable) [size=64M]
...

 hv_vmbus: registering driver hyperv_drm
 hyperv_drm 5620e0c7-8062-4dce-aeb7-520c7ef76171: [drm] Synthvid Version major 3, minor 5
 hyperv_drm 0000:00:08.0: vgaarb: deactivate vga console
 hyperv_drm 0000:00:08.0: BAR 0: can't reserve [mem 0xf8000000-0xfbffffff]
 hyperv_drm 5620e0c7-8062-4dce-aeb7-520c7ef76171: [drm] Cannot request framebuffer, boot fb still active?

Note: "Cannot request framebuffer" is not a fatal error in
hyperv_setup_gen1() as the code assumes there's some other framebuffer
device there but we actually have some other PCI device (mlx4 in this
case) config space there!

The problem appears to be that vmbus_allocate_mmio() can use dedicated
framebuffer region to serve any MMIO request from any device. The
semantics one might assume of a parameter named "fb_overlap_ok"
aren't implemented because !fb_overlap_ok essentially has no effect.
The existing semantics are really "prefer_fb_overlap". This patch
implements the expected and needed semantics, which is to not allocate
from the frame buffer space when !fb_overlap_ok.

Note, Gen2 VMs are usually unaffected by the issue because
framebuffer region is already taken by EFI fb (in case kernel supports
it) but Gen1 VMs may have this region unclaimed by the time Hyper-V PCI
pass-through driver tries allocating MMIO space if Hyper-V DRM/FB drivers
load after it. Devices can be brought up in any sequence so let's
resolve the issue by always ignoring 'fb_mmio' region for non-FB
requests, even if the region is unclaimed.

Reviewed-by: Michael Kelley 
Signed-off-by: Vitaly Kuznetsov 
Link: https://lore.kernel.org/r/20220827130345.1320254-4-vkuznets@redhat.com
Signed-off-by: Wei Liu 
Signed-off-by: Sasha Levin

random: remove unused irq_flags argument from add_interrupt_randomness()

2022-06-25T09:45:04+00:00

commit 703f7066f40599c290babdb79dd61319264987e9 upstream.

Since commit
   ee3e00e9e7101 ("random: use registers from interrupted code for CPU's w/o a cycle counter")

the irq_flags argument is no longer used.

Remove unused irq_flags.

Cc: Borislav Petkov 
Cc: Dave Hansen 
Cc: Dexuan Cui 
Cc: H. Peter Anvin 
Cc: Haiyang Zhang 
Cc: Ingo Molnar 
Cc: K. Y. Srinivasan 
Cc: Stephen Hemminger 
Cc: Thomas Gleixner 
Cc: Wei Liu 
Cc: linux-hyperv@vger.kernel.org
Cc: x86@kernel.org
Signed-off-by: Sebastian Andrzej Siewior 
Acked-by: Wei Liu 
Signed-off-by: Jason A. Donenfeld 
Signed-off-by: Greg Kroah-Hartman

hyperv/vmbus: include linux/bitops.h

2021-11-26T10:48:20+00:00

[ Upstream commit 8017c99680fa65e1e8d999df1583de476a187830 ]

On arm64 randconfig builds, hyperv sometimes fails with this
error:

In file included from drivers/hv/hv_trace.c:3:
In file included from drivers/hv/hyperv_vmbus.h:16:
In file included from arch/arm64/include/asm/sync_bitops.h:5:
arch/arm64/include/asm/bitops.h:11:2: error: only  can be included directly
In file included from include/asm-generic/bitops/hweight.h:5:
include/asm-generic/bitops/arch_hweight.h:9:9: error: implicit declaration of function '__sw_hweight32' [-Werror,-Wimplicit-function-declaration]
include/asm-generic/bitops/atomic.h:17:7: error: implicit declaration of function 'BIT_WORD' [-Werror,-Wimplicit-function-declaration]

Include the correct header first.

Signed-off-by: Arnd Bergmann 
Link: https://lore.kernel.org/r/20211018131929.2260087-1-arnd@kernel.org
Signed-off-by: Wei Liu 
Signed-off-by: Sasha Levin

Drivers: hv: vmbus: Increase wait time for VMbus unload

2021-05-22T08:40:24+00:00

[ Upstream commit 77db0ec8b7764cb9b09b78066ebfd47b2c0c1909 ]

When running in Azure, disks may be connected to a Linux VM with
read/write caching enabled. If a VM panics and issues a VMbus
UNLOAD request to Hyper-V, the response is delayed until all dirty
data in the disk cache is flushed.  In extreme cases, this flushing
can take 10's of seconds, depending on the disk speed and the amount
of dirty data. If kdump is configured for the VM, the current 10 second
timeout in vmbus_wait_for_unload() may be exceeded, and the UNLOAD
complete message may arrive well after the kdump kernel is already
running, causing problems.  Note that no problem occurs if kdump is
not enabled because Hyper-V waits for the cache flush before doing
a reboot through the BIOS/UEFI code.

Fix this problem by increasing the timeout in vmbus_wait_for_unload()
to 100 seconds. Also output periodic messages so that if anyone is
watching the serial console, they won't think the VM is completely
hung.

Fixes: 911e1987efc8 ("Drivers: hv: vmbus: Add timeout to vmbus_wait_for_unload")
Signed-off-by: Michael Kelley 
Reviewed-by: Vitaly Kuznetsov 
Link: https://lore.kernel.org/r/1618894089-126662-1-git-send-email-mikelley@microsoft.com
Signed-off-by: Wei Liu 
Signed-off-by: Sasha Levin

Drivers: hv: vmbus: Add timeout to vmbus_wait_for_unload

2020-09-23T06:46:15+00:00

[ Upstream commit 911e1987efc8f3e6445955fbae7f54b428b92bd3 ]

vmbus_wait_for_unload() looks for a CHANNELMSG_UNLOAD_RESPONSE message
coming from Hyper-V.  But if the message isn't found for some reason,
the panic path gets hung forever.  Add a timeout of 10 seconds to prevent
this.

Fixes: 415719160de3 ("Drivers: hv: vmbus: avoid scheduling in interrupt context in vmbus_initiate_unload()")
Signed-off-by: Michael Kelley 
Reviewed-by: Dexuan Cui 
Reviewed-by: Vitaly Kuznetsov 
Link: https://lore.kernel.org/r/1600026449-23651-1-git-send-email-mikelley@microsoft.com
Signed-off-by: Wei Liu 
Signed-off-by: Sasha Levin

Drivers: hv: vmbus: Return -EINVAL for the sys files for unopened channels

2018-12-29T12:40:15+00:00

commit fc96df16a1ce80cbb3c316ab7d4dc8cd5c2852ce upstream.

Before 98f4c651762c, we returned zeros for unopened channels.
With 98f4c651762c, we started to return random on-stack values.

We'd better return -EINVAL instead.

Fixes: 98f4c651762c ("hv: move ringbuffer bus attributes to dev_groups")
Cc: stable@vger.kernel.org
Cc: K. Y. Srinivasan 
Cc: Haiyang Zhang 
Cc: Stephen Hemminger 
Signed-off-by: Dexuan Cui 
Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

Drivers: hv: vmbus: check the creation_status in vmbus_establish_gpadl()

2018-12-05T18:42:41+00:00

commit eceb05965489784f24bbf4d61ba60e475a983016 upstream.

This is a longstanding issue: if the vmbus upper-layer drivers try to
consume too many GPADLs, the host may return with an error
0xC0000044 (STATUS_QUOTA_EXCEEDED), but currently we forget to check
the creation_status, and hence we can pass an invalid GPADL handle
into the OPEN_CHANNEL message, and get an error code 0xc0000225 in
open_info->response.open_result.status, and finally we hang in
vmbus_open() -> "goto error_free_info" -> vmbus_teardown_gpadl().

With this patch, we can exit gracefully on STATUS_QUOTA_EXCEEDED.

Cc: Stephen Hemminger 
Cc: K. Y. Srinivasan 
Cc: Haiyang Zhang 
Cc: stable@vger.kernel.org
Signed-off-by: Dexuan Cui 
Signed-off-by: K. Y. Srinivasan 
Signed-off-by: Greg Kroah-Hartman

HV: properly delay KVP packets when negotiation is in progress

2018-10-20T07:51:33+00:00

commit a3ade8cc474d848676278660e65f5af1e9e094d9 upstream.

The host may send multiple negotiation packets
(due to timeout) before the KVP user-mode daemon
is connected. KVP user-mode daemon is connected.
We need to defer processing those packets
until the daemon is negotiated and connected.
It's okay for guest to respond
to all negotiation packets.

In addition, the host may send multiple staged
KVP requests as soon as negotiation is done.
We need to properly process those packets using one
tasklet for exclusive access to ring buffer.

This patch is based on the work of
Nick Meier .

The above is the original changelog of
a3ade8cc474d ("HV: properly delay KVP packets when negotiation is in progress"

Here I re-worked the original patch because the mainline version
can't work for the linux-4.4.y branch, on which channel->callback_event
doesn't exist yet. In the mainline, channel->callback_event was added by:
631e63a9f346 ("vmbus: change to per channel tasklet"). Here we don't want
to backport it to v4.4, as it requires extra supporting changes and fixes,
which are unnecessary as to the KVP bug we're trying to resolve.

NOTE: before this patch is used, we should cherry-pick the other related
3 patches from the mainline first:

The background of this backport request is that: recently Wang Jian reported
some KVP issues: https://github.com/LIS/lis-next/issues/593:
e.g. the /var/lib/hyperv/.kvp_pool_* files can not be updated, and sometimes
if the hv_kvp_daemon doesn't timely start, the host may not be able to query
the VM's IP address via KVP.

Reported-by: Wang Jian 
Tested-by: Wang Jian 
Signed-off-by: Dexuan Cui 
Signed-off-by: Long Li 
Signed-off-by: K. Y. Srinivasan 
Signed-off-by: Greg Kroah-Hartman

Drivers: hv: vmbus: do not mark HV_PCIE as perf_device

2018-04-20T06:20:41+00:00

commit 238064f13d057390a8c5e1a6a80f4f0a0ec46499 upstream.

The pci-hyperv driver's channel callback hv_pci_onchannelcallback() is not
really a hot path, so we don't need to mark it as a perf_device, meaning
with this patch all HV_PCIE channels' target_cpu will be CPU0.

Signed-off-by: Dexuan Cui 
Cc: stable@vger.kernel.org
Cc: Stephen Hemminger 
Signed-off-by: K. Y. Srinivasan 
Signed-off-by: Greg Kroah-Hartman 
Signed-off-by: Greg Kroah-Hartman

x86/retpoline/hyperv: Convert assembler indirect jumps

2018-01-17T08:38:59+00:00

commit e70e5892b28c18f517f29ab6e83bd57705104b31 upstream.

Convert all indirect jumps in hyperv inline asm code to use non-speculative
sequences when CONFIG_RETPOLINE is enabled.

Signed-off-by: David Woodhouse 
Signed-off-by: Thomas Gleixner 
Acked-by: Arjan van de Ven 
Acked-by: Ingo Molnar 
Cc: gnomes@lxorguk.ukuu.org.uk
Cc: Rik van Riel 
Cc: Andi Kleen 
Cc: Josh Poimboeuf 
Cc: thomas.lendacky@amd.com
Cc: Peter Zijlstra 
Cc: Linus Torvalds 
Cc: Jiri Kosina 
Cc: Andy Lutomirski 
Cc: Dave Hansen 
Cc: Kees Cook 
Cc: Tim Chen 
Cc: Greg Kroah-Hartman 
Cc: Paul Turner 
Link: https://lkml.kernel.org/r/1515707194-20531-9-git-send-email-dwmw@amazon.co.uk
[ backport to 4.9, hopefully correct, not tested... - gregkh ]
Signed-off-by: Greg Kroah-Hartman