linux-stable.git/drivers, branch linux-5.0.y

xen/pciback: Don't disable PCI_COMMAND on PCI device reset.

2019-06-04T06:01:30+00:00

commit 7681f31ec9cdacab4fd10570be924f2cef6669ba upstream.

There is no need for this at all. Worst it means that if
the guest tries to write to BARs it could lead (on certain
platforms) to PCI SERR errors.

Please note that with af6fc858a35b90e89ea7a7ee58e66628c55c776b
"xen-pciback: limit guest control of command register"
a guest is still allowed to enable those control bits (safely), but
is not allowed to disable them and that therefore a well behaved
frontend which enables things before using them will still
function correctly.

This is done via an write to the configuration register 0x4 which
triggers on the backend side:
command_write
  \- pci_enable_device
     \- pci_enable_device_flags
        \- do_pci_enable_device
           \- pcibios_enable_device
              \-pci_enable_resourcess
                [which enables the PCI_COMMAND_MEMORY|PCI_COMMAND_IO]

However guests (and drivers) which don't do this could cause
problems, including the security issues which XSA-120 sought
to address.

Reported-by: Jan Beulich 
Signed-off-by: Konrad Rzeszutek Wilk 
Reviewed-by: Prarit Bhargava 
Signed-off-by: Juergen Gross 
Cc: Ben Hutchings 
Signed-off-by: Greg Kroah-Hartman

crypto: vmx - ghash: do nosimd fallback manually

2019-06-04T06:01:30+00:00

commit 357d065a44cdd77ed5ff35155a989f2a763e96ef upstream.

VMX ghash was using a fallback that did not support interleaving simd
and nosimd operations, leading to failures in the extended test suite.

If I understood correctly, Eric's suggestion was to use the same
data format that the generic code uses, allowing us to call into it
with the same contexts. I wasn't able to get that to work - I think
there's a very different key structure and data layout being used.

So instead steal the arm64 approach and perform the fallback
operations directly if required.

Fixes: cc333cd68dfa ("crypto: vmx - Adding GHASH routines for VMX module")
Cc: stable@vger.kernel.org # v4.1+
Reported-by: Eric Biggers 
Signed-off-by: Daniel Axtens 
Acked-by: Ard Biesheuvel 
Tested-by: Michael Ellerman 
Signed-off-by: Herbert Xu 
Signed-off-by: Daniel Axtens 
Signed-off-by: Greg Kroah-Hartman

cxgb4: Revert "cxgb4: Remove SGE_HOST_PAGE_SIZE dependency on page size"

2019-06-04T06:01:30+00:00

[ Upstream commit ab0610efabb4c4f419a531455708caf1dd29357e ]

This reverts commit 2391b0030e241386d710df10e53e2cfc3c5d4fc1 which has
introduced regression. Now SGE's BAR2 Doorbell/GTS Page Size is
interpreted correctly in the firmware itself by using actual host
page size. Hence previous commit needs to be reverted.

Signed-off-by: Vishal Kulkarni 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

bnxt_en: Reduce memory usage when running in kdump kernel.

2019-06-04T06:01:29+00:00

[ Upstream commit d629522e1d66561f38e5c8d4f52bb6d254ec0707 ]

Skip RDMA context memory allocations, reduce to 1 ring, and disable
TPA when running in the kdump kernel.  Without this patch, the driver
fails to initialize with memory allocation errors when running in a
typical kdump kernel.

Fixes: cf6daed098d1 ("bnxt_en: Increase context memory allocations on 57500 chips for RDMA.")
Signed-off-by: Michael Chan 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

bnxt_en: Fix possible BUG() condition when calling pci_disable_msix().

2019-06-04T06:01:29+00:00

[ Upstream commit 1b3f0b75c39f534278a895c117282014e9d0ae1f ]

When making configuration changes, the driver calls bnxt_close_nic()
and then bnxt_open_nic() for the changes to take effect.  A parameter
irq_re_init is passed to the call sequence to indicate if IRQ
should be re-initialized.  This irq_re_init parameter needs to
be included in the bnxt_reserve_rings() call.  bnxt_reserve_rings()
can only call pci_disable_msix() if the irq_re_init parameter is
true, otherwise it may hit BUG() because some IRQs may not have been
freed yet.

Fixes: 41e8d7983752 ("bnxt_en: Modify the ring reservation functions for 57500 series chips.")
Signed-off-by: Michael Chan 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

bnxt_en: Fix aggregation buffer leak under OOM condition.

2019-06-04T06:01:29+00:00

[ Upstream commit 296d5b54163964b7ae536b8b57dfbd21d4e868e1 ]

For every RX packet, the driver replenishes all buffers used for that
packet and puts them back into the RX ring and RX aggregation ring.
In one code path where the RX packet has one RX buffer and one or more
aggregation buffers, we missed recycling the aggregation buffer(s) if
we are unable to allocate a new SKB buffer.  This leads to the
aggregation ring slowly running out of buffers over time.  Fix it
by properly recycling the aggregation buffers.

Fixes: c0c050c58d84 ("bnxt_en: New Broadcom ethernet driver.")
Reported-by: Rakesh Hemnani 
Signed-off-by: Michael Chan 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

net: stmmac: dma channel control register need to be init first

2019-06-04T06:01:29+00:00

stmmac_init_chan() needs to be called before stmmac_init_rx_chan() and
stmmac_init_tx_chan(). This is because if PBLx8 is to be used,
"DMA_CH(#i)_Control.PBLx8" needs to be set before programming
"DMA_CH(#i)_TX_Control.TxPBL" and "DMA_CH(#i)_RX_Control.RxPBL".

Fixes: 47f2a9ce527a ("net: stmmac: dma channel init prepared for multiple queues")
Reviewed-by: Zhang, Baoli 
Signed-off-by: Ong Boon Leong 
Signed-off-by: Weifeng Voon 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

net: stmmac: fix ethtool flow control not able to get/set

2019-06-04T06:01:29+00:00

Currently ethtool was not able to get/set the flow control due to a
missing "!". It will always return -EOPNOTSUPP even the device is
flow control supported.

This patch fixes the condition check for ethtool flow control get/set
function for ETHTOOL_LINK_MODE_Asym_Pause_BIT.

Fixes: 3c1bcc8614db (“net: ethernet: Convert phydev advertize and supported from u32 to link mode”)
Signed-off-by: Tan, Tee Min 
Reviewed-by: Ong Boon Leong 
Signed-off-by: Voon, Weifeng 
Reviewed-by: Andrew Lunn 
Signed-off-by: David S. Miller 
Signed-off-by: Greg Kroah-Hartman

net/mlx5e: Disable rxhash when CQE compress is enabled

2019-06-04T06:01:29+00:00

[ Upstream commit c0194e2d0ef0e5ce5e21a35640d23a706827ae28 ]

When CQE compression is enabled (Multi-host systems), compressed CQEs
might arrive to the driver rx, compressed CQEs don't have a valid hash
offload and the driver already reports a hash value of 0 and invalid hash
type on the skb for compressed CQEs, but this is not good enough.

On a congested PCIe, where CQE compression will kick in aggressively,
gro will deliver lots of out of order packets due to the invalid hash
and this might cause a serious performance drop.

The only valid solution, is to disable rxhash offload at all when CQE
compression is favorable (Multi-host systems).

Fixes: 7219ab34f184 ("net/mlx5e: CQE compression")
Signed-off-by: Saeed Mahameed 
Signed-off-by: Greg Kroah-Hartman

net/mlx5: Allocate root ns memory using kzalloc to match kfree

2019-06-04T06:01:29+00:00

[ Upstream commit 25fa506b70cadb580c1e9cbd836d6417276d4bcd ]

root ns is yet another fs core node which is freed using kfree() by
tree_put_node().
Rest of the other fs core objects are also allocated using kmalloc
variants.

However, root ns memory is allocated using kvzalloc().
Hence allocate root ns memory using kzalloc().

Fixes: 2530236303d9e ("net/mlx5_core: Flow steering tree initialization")
Signed-off-by: Parav Pandit 
Reviewed-by: Daniel Jurgens 
Reviewed-by: Mark Bloch 
Signed-off-by: Saeed Mahameed 
Signed-off-by: Greg Kroah-Hartman