Paolo Abeni [Thu, 9 Mar 2017 12:54:08 +0000 (13:54 +0100)]
net/socket: use per af lockdep classes for sk queues
Currently the sock queue's spin locks get their lockdep
classes by the default init_spin_lock() initializer:
all socket families get - usually, see below - a single
class for rx, another specific class for tx, etc.
This can lead to false positive lockdep splat, as
reported by Andrey.
Moreover there are two separate initialization points
for the sock queues, one in sk_clone_lock() and one
in sock_init_data(), so that e.g. the rx queue lock
can get one of two possible, different classes, depending
on the socket being cloned or not.
This change tries to address the above, setting explicitly
a per address family lockdep class for each queue's
spinlock. Also, move the duplicated initialization code to a
single location.
v1 -> v2:
- renamed the init helper
rfc -> v1:
- no changes, tested with several different workload
Suggested-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jie Deng [Wed, 8 Mar 2017 06:06:18 +0000 (14:06 +0800)]
net: dwc-xlgmac: Initial driver for DesignWare Enterprise Ethernet
Synopsys provides a new DesignWare Core Enterprise Ethernet MAC
IP (DWC-XLGMAC) for Ethernet designs. It is compliant with the
IEEE 802.3-2012 specifications, including IEEE 802.3ba and
consortium specifications.
This patch provides the initial 25G/40G/50G/100G Ethernet driver
for Synopsys XLGMAC IP Prototyping Kit.
Signed-off-by: Jie Deng <jiedeng@synopsys.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 9 Mar 2017 21:25:05 +0000 (13:25 -0800)]
Merge branch 'xgene-v2'
Iyappan Subramanian says:
====================
drivers: net: xgene-v2: Add RGMII based 1G driver
This patch set adds support for RGMII based 1GbE hardware which uses a linked
list of DMA descriptor architecture (v2) for APM X-Gene SoCs.
v4: Address review comments from v3
- fixed local variable declarations to reverse christmas tree order
v3: Address review comments from v2
- fixed kbuild warnings (this 'if' clause does not guard)
v2: Address review comments from v1
- moved create_desc_ring and delete_desc_ring to open() and close()
respectively
- changed to use dma_zalloc APIs
- fixed tx_timeout()
- removed tx completion polling upper bound
- added error checking on rx packets
- added netif_stop_queue() and netif_wake_queue()
v1:
- Initial version
====================
Signed-off-by: Iyappan Subramanian <isubramanian@apm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
MAINTAINERS: Add entry for APM X-Gene SoC Ethernet (v2) driver
This patch adds a MAINTAINERS entry for the ethernet driver for
the on-chip ethernet interface which uses a linked list of DMA
descriptor architecture (v2) for APM X-Gene SoCs.
Signed-off-by: Iyappan Subramanian <isubramanian@apm.com> Signed-off-by: Keyur Chudgar <kchudgar@apm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds,
- Transmit
- Transmit completion poll
- Receive poll
- NAPI handler
and enables the driver.
Signed-off-by: Iyappan Subramanian <isubramanian@apm.com> Signed-off-by: Keyur Chudgar <kchudgar@apm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
- probe, remove, shutdown
- open, close and stats
- create and delete ring
- request and delete irq
Signed-off-by: Iyappan Subramanian <isubramanian@apm.com> Signed-off-by: Keyur Chudgar <kchudgar@apm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds functions to configure ethernet hardware.
Signed-off-by: Iyappan Subramanian <isubramanian@apm.com> Signed-off-by: Keyur Chudgar <kchudgar@apm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds functions to configure and control mac. This
patch also adds helper functions to get/set registers.
Signed-off-by: Iyappan Subramanian <isubramanian@apm.com> Signed-off-by: Keyur Chudgar <kchudgar@apm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds DMA descriptor setup and interrupt enable/disable
functions.
Signed-off-by: Iyappan Subramanian <isubramanian@apm.com> Signed-off-by: Keyur Chudgar <kchudgar@apm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Rick Farrington [Tue, 7 Mar 2017 19:40:41 +0000 (11:40 -0800)]
liquidio: add support for XPS
Add support for XPS.
Signed-off-by: Rick Farrington <ricardo.farrington@cavium.com> Signed-off-by: Felix Manlunas <felix.manlunas@cavium.com> Signed-off-by: Derek Chickles <derek.chickles@cavium.com> Signed-off-by: Satanand Burla <satananda.burla@cavium.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
net: mvpp2: add initial support for PPv2.2
The goal of this patch series is to add basic support for PPv2.2 in
the existing mvpp2 driver. mvpp2 currently supported the PPv2.1
version of the IP, used in the 32 bits Marvell Armada 375 SoC. PPv2.2
is an evolution of this IP block, used in the 64 bits Marvell Armada
7K/8K SoCs.
In order to ease the review, the introduction of PPv2.2 support has
been made into multiple small commits, with the final commit adding
the compatible string that makes the PPv2.2 support actually
usable. The series remain fully bisectable.
People interested in testing the code will find the full series (plus
a few Device Tree patches) at:
I'd like to thank Stefan Chulski and Marcin Wojtas, who helped me a
lot in the development of this patch series, by reviewing the patches,
and giving lots of useful hints to debug the driver on PPv2.2. Thanks
as well to Russell King for reviewing previous iterations of this
series, and providing suggestions and fixes.
Changes between v2 and v3:
- Rebased on v4.11-rc1.
- Add patch "net: mvpp2: fix DMA address calculation in
mvpp2_txq_inc_put()", to properly take into account the "packet
offset" field of the TX descriptors. Without this, we were getting
DMA_API_DEBUG warnings that we are unmapping DMA mappings with a
non-mapped DMA address.
- In patch "net: mvpp2: add and use accessors for TX/RX descriptors",
add a function named mvpp2_txdesc_offset_get(), which is needed for
the DMA address calculation fix.
- In patch "net: mvpp2: add and use accessors for TX/RX descriptors",
fix the calculation of tx_desc physical address and packet offset
in mvpp2_tx_frag_process(). The offset was assigned into the buffer
physical address, and the physical address to the packet offset,
which meant the fragment process was completely broken.
- In patch "net: mvpp2: adjust the allocation/free of BM pools for
PPv2.2" fix how MVPP22_BM_ADDR_HIGH_VIRT_RLS_MASK is used. This
mask is already shifted. So the value should be shifted before
being masked and not the opposite.
- Add a new patch "net: mvpp2: set dma mask and coherent dma mask on
PPv2.2", to set the DMA mask and DMA coherent mask. By setting the
DMA mask to 40 bits we avoid using bounce buffers when network
packets are above the 4 GB limit. The coherent mask remains set to
32 bits, because the BM pools must all have the same high 32 bits
in their addresses.
- Use "dma" instead of "phys" where appropriate, as suggested by
Russell King.
- Use the "cookie" field of the RX descriptor to store the physical
address instead of the virtual address, and then use phys_to_virt()
to get the virtual address. This allows to work around the limit
that the "cookie" field only has 40 bits, which is not sufficient
to store a virtual address on 64 bits platforms. This was suggested
by Russell King.
As part of this change, also got rid of all the compile time
conditionals on CONFIG_ARCH_DMA_ADDR_T_64BIT, to get better
compile-time coverage.
- In patch "net: mvpp2: handle misc PPv2.1/PPv2.2 differences":
* Instead of calling mvpp21_port_power_up(port) only on PPv2.1,
remove this function, and call its relevant parts directly from
->probe(). Only mvpp2_port_fc_adv_enable() is PPv2.1
specific. Reported by Russell King.
* Add a mvpp22_port_mii_set() function that properly initializes
SGMII support on PPv2.2. Code provided by Russell King.
- In patch "net: mvpp2: handle register mapping and access for PPv2.2":
* Adjust the code to match the change of the DT binding in terms
of mapping the second register area on PPv2.2.
* Rework the register accessors to remove the get_cpu()/put_cpu(),
and instead use separate accessors for global registers
vs. per-CPU registers.
- Add a few new patches removing dead/unused/useless code:
net: mvpp2: remove support for buffer header
net: mvpp2: remove unused register definition MVPP2_TXQ_THRESH_REG
net: mvpp2: remove mvpp2_txq_pend_desc_num_get() function
- Fix a number of checkpatch warnings.
Changes between v1 and v2:
- Made a separate series from the set of patches doing preparation
changes/fixes to the mvpp2 driver.
- Rebased on top of v4.10-rc1.
- Update Kconfig text of the mvpp2 driver to mention the support for
Armada 7K and 8K (PPv2.2).
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: finally add the PPv2.2 compatible string
Now that the mvpp2 driver has been modified to accommodate the support
for PPv2.2, we can finally advertise this support by adding the
appropriate compatible string.
At the same time, we update the Kconfig description of the MVPP2 driver.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: set dma mask and coherent dma mask on PPv2.2
On PPv2.2, the streaming mappings can be anywhere in the first 40 bits
of the physical address space. However, for the coherent mappings, we
still need them to be in the first 32 bits of the address space,
because all BM pools share a single register to store the high 32 bits
of the BM pool address, which means all BM pools must be allocated in
the same 4GB memory area.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: add support for an additional clock needed for PPv2.2
The PPv2.2 variant of the network controller needs an additional
clock, the "MG clock" in order for the IP block to operate
properly. This commit adds support for this additional clock to the
driver, reworking as needed the error handling path.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
In PPv2.1, we have a maximum of 8 RXQs per port, with a default of 4
RXQs per port, and we were assigning RXQs 0->3 to the first port, 4->7
to the second port, 8->11 to the third port, etc.
In PPv2.2, we have a maximum of 32 RXQs per port, and we must allocate
RXQs from the range of 32 RXQs available for each port. So port 0 must
use RXQs in the range 0->31, port 1 in the range 32->63, etc.
This commit adapts the mvpp2 to this difference between PPv2.1 and
PPv2.2:
- The constant definition MVPP2_MAX_RXQ is replaced by a new field
'max_port_rxqs' in 'struct mvpp2', which stores the maximum number of
RXQs per port. This field is initialized during ->probe() depending
on the IP version.
- MVPP2_RXQ_TOTAL_NUM is removed, and instead we calculate the total
number of RXQs by multiplying the number of ports by the maximum of
RXQs per port. This was anyway used in only one place.
- In mvpp2_port_probe(), the calculation of port->first_rxq is adjusted
to cope with the different allocation strategy between PPv2.1 and
PPv2.2. Due to this change, the 'next_first_rxq' argument of this
function is no longer needed and is removed.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
This commit handles a few miscellaneous differences between PPv2.1 and
PPv2.2 in different areas, where code done for PPv2.1 doesn't apply for
PPv2.2 or needs to be adjusted (getting the MAC address, disabling PHY
polling, etc.).
Thanks to Russell King for providing the initial implementation of
mvpp22_port_mii_set().
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: handle register mapping and access for PPv2.2
This commit adjusts the mvpp2 driver register mapping and access logic
to support PPv2.2, to handle a number of differences.
Due to how the registers are laid out in memory, the Device Tree binding
for the "reg" property is different:
- On PPv2.1, we had a first area for the packet processor
registers (common to all ports), and then one area per port.
- On PPv2.2, we have a first area for the packet processor
registers (common to all ports), and a second area for numerous other
registers, including a large number of per-port registers
In addition, on PPv2.2, the area for the common registers is split into
so-called "address spaces" of 64 KB each. They allow to access per-CPU
registers, where each CPU has its own copy of some registers. A few
other registers, which have a single copy, also need to be accessed from
those per-CPU windows if they are related to a per-CPU register. For
example:
- Writing to MVPP2_TXQ_NUM_REG selects a TX queue. This register is a
per-CPU register, it must be accessed from the current CPU register
window.
- Then a write to MVPP2_TXQ_PENDING_REG, MVPP2_TXQ_DESC_ADDR_REG (and
a few others) will affect the TX queue that was selected by the
write to MVPP2_TXQ_NUM_REG. It must be accessed from the same CPU
window as the write to the TXQ_NUM_REG.
Therefore, the ->base member of 'struct mvpp2' is replaced with a
->cpu_base[] array, each entry pointing to a mapping of the per-CPU
area. Since PPv2.1 doesn't have this concept of per-CPU windows, all
entries in ->cpu_base[] point to the same io-remapped area.
The existing mvpp2_read() and mvpp2_write() accessors use cpu_base[0],
they are used for registers for which the CPU window doesn't matter.
mvpp2_percpu_read() and mvpp2_percpu_write() are new accessors added to
access the registers for which the CPU window does matter, which is why
they take a "cpu" as argument.
The driver is then changed to use mvpp2_percpu_read() and
mvpp2_percpu_write() where it matters.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: adjust mvpp2_{rxq, txq}_init for PPv2.2
In PPv2.2, the MVPP2_RXQ_DESC_ADDR_REG and MVPP2_TXQ_DESC_ADDR_REG
registers have a slightly different layout, because they need to contain
a 64-bit address for the RX and TX descriptor arrays. This commit
adjusts those functions accordingly.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: adapt the mvpp2_rxq_*_pool_set functions to PPv2.2
The MVPP2_RXQ_CONFIG_REG register has a slightly different layout
between PPv2.1 and PPv2.2, so this commit adapts the functions modifying
this register to accommodate for both the PPv2.1 and PPv2.2 cases.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: adjust the allocation/free of BM pools for PPv2.2
This commit adjusts the allocation and freeing of BM pools to support
PPv2.2. This involves:
- Checking that the number of buffer pointers is a multiple of 16, as
required by the hardware.
- Adjusting the size of the DMA coherent area allocated for buffer
pointers. Indeed, PPv2.2 needs space for 2 pointers of 64-bits per
buffer, as opposed to 2 pointers of 32-bits per buffer in
PPv2.1. The size in bytes is now stored in a new field of the
mvpp2_bm_pool structure.
- On PPv2.2, getting the DMA address and cookie (used for the physical
address) of each buffer requires reading the
MVPP22_BM_ADDR_HIGH_ALLOC to get the high order bits of those
addresses. A new utility function mvpp2_bm_bufs_get_addrs() is
introduced to handle this.
- On PPv2.2, releasing a buffer requires writing the high order 32 bits
of the DMA address and cookie to MVPP22_BM_PHY_VIRT_HIGH_RLS_REG.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: introduce PPv2.2 HW descriptors and adapt accessors
This commit adds the definition of the PPv2.2 HW descriptors, adjusts
the mvpp2_tx_desc and mvpp2_rx_desc structures accordingly, and adapts
the accessors to work on both PPv2.1 and PPv2.2.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: introduce an intermediate union for the TX/RX descriptors
Since the format of the HW descriptors is different between PPv2.1 and
PPv2.2, this commit introduces an intermediate union, with for now
only the PPv2.1 descriptors. The bulk of the driver code only
manipulates opaque mvpp2_tx_desc and mvpp2_rx_desc pointers, and the
descriptors can only be accessed and modified through the accessor
functions. A follow-up commit will add the descriptor definitions for
PPv2.2.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: add hw_version field in "struct mvpp2"
In preparation to the introduction for the support of PPv2.2 in the
mvpp2 driver, this commit adds a hw_version field to the struct
mvpp2, and uses the .data field of the DT match table to fill it in.
Having the MVPP21 and MVPP22 definitions available will allow to start
adding the necessary conditional code to support PPv2.2.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: add and use accessors for TX/RX descriptors
The PPv2.2 IP has a different TX and RX descriptor layout compared to
PPv2.1. In order to prepare for the introduction of PPv2.2 support in
mvpp2, this commit adds accessors for the different fields of the TX
and RX descriptors, and changes the code to use them.
For now, the mvpp2_port argument passed to the accessors is not used,
but it will be used in follow-up to update the descriptor according to
the version of the IP being used.
Apart from the mechanical changes to use the newly introduced
accessors, a few other changes, needed to use the accessors, are made:
- The mvpp2_txq_inc_put() function now takes a mvpp2_port as first
argument, as it is needed to use the accessors.
- Similarly, the mvpp2_bm_cookie_build() gains a mvpp2_port first
argument, for the same reason.
- In mvpp2_rx_error(), instead of accessing the RX descriptor in each
case of the switch, we introduce a local variable to store the
packet size.
- In mvpp2_tx_frag_process() and mvpp2_tx() instead of accessing the
packet size from the TX descriptor, we use the actual value
available in the function, which is used to set the TX descriptor
packet size a few lines before.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: store physical address of buffer in rx_desc->buf_cookie
The RX descriptors of the PPv2 hardware allow to store several
information, amongst which:
- the DMA address of the buffer in which the data has been received
- a "cookie" field, left to the use of the driver, and not used by the
hardware
In the current implementation, the "cookie" field is used to store the
virtual address of the buffer, so that in the receive completion path,
we can easily get the virtual address of the buffer that corresponds to
a completed RX descriptors.
On PPv2.1, used on 32-bit platforms, those two fields are 32-bit wide,
which is enough to store a DMA address in the first field, and a virtual
address in the second field.
On PPv2.2, used on 64-bit platforms, these two fields have been extended
to 40 bits. While 40 bits is enough to store a DMA address (as long as
the DMA mask is 40 bits or lower), it is not enough to store a virtual
address. Therefore, the "cookie" field can no longer be used to store
the virtual address of the buffer.
However, as Russell King pointed out, the RX buffers are always
allocated in the kernel linear mapping, and therefore using
phys_to_virt() on the physical address of the RX buffer is possible and
correct.
Therefore, this commit changes the driver to use the "cookie" field to
store the physical address instead of the virtual
address. phys_to_virt() is used in the receive completion path to
retrieve the virtual address from the physical address.
It is obviously important to realize that the DMA address and physical
address are two different things, which is why we store both in the RX
descriptors. While those addresses may be identical in some situations,
it remains two distinct concepts, and both addresses should be handled
separately.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: remove mvpp2_txq_pend_desc_num_get() function
The mvpp2_txq_pend_desc_num_get() function only selects a TX queue, and
reads the number of pending descriptors. It is used in only one place,
in mvpp2_txq_clean(), where the TX queue has already been selected by a
write to MVPP2_TXQ_NUM_REG.
Therefore, this function is useless, and the caller can simply read the
value of the MVPP2_TXQ_PENDING_REG register instead.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
The "buffer header" functionality is a functionality used by the
hardware to split an incoming packets over multiple BM buffers if they
are not large enough. However, the mvpp2 driver guarantees that a pool
of BM buffers has buffers with a size large enough to store MTU-sized
packets. Therefore, this functionality is completely unused, and the
code can be removed, and we should never get a descriptor with bit
MVPP2_RXD_BUF_HDR set.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: mvpp2: use "dma" instead of "phys" where appropriate
As indicated by Russell King, the mvpp2 driver currently uses a lot
"phys" or "phys_addr" to store what really is a DMA address. This commit
clarifies this by using "dma" or "dma_addr" where appropriate.
This is especially important as we are going to introduce more changes
where the distinction between physical address and DMA address will be
key.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
dt-bindings: net: update Marvell PPv2 binding for PPv2.2 support
The Marvell PPv2 Device Tree binding was so far only used to describe
the PPv2.1 network controller, used in the Marvell Armada 375.
A new version of this IP block, PPv2.2 is used in the Marvell Armada
7K/8K processor. This commit extends the existing binding so that it can
also be used to describe PPv2.2 hardware.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
mlx4: order-0 allocations and page recycling
As mentioned half a year ago, we better switch mlx4 driver to order-0
allocations and page recycling.
This reduces vulnerability surface thanks to better skb->truesize
tracking and provides better performance in most cases.
(33 Gbit for one TCP flow on my lab hosts)
I will provide for linux-4.13 a patch on top of this series,
trying to improve data locality as described in
https://www.spinics.net/lists/netdev/msg422258.html
v2 provides an ethtool -S new counter (rx_alloc_pages) and
code factorization, plus Tariq fix.
v3 includes various fixes based on Tariq tests and feedback
from Saeed and Tariq.
v4 rebased on net-next for inclusion in linux-4.12, as requested
by Tariq.
Worth noting this patch series deletes ~250 lines of code ;)
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Wed, 8 Mar 2017 16:17:18 +0000 (08:17 -0800)]
mlx4: remove duplicate code in mlx4_en_process_rx_cq()
We should keep one way to build skbs, regardless of GRO being on or off.
Note that I made sure to defer as much as possible the point we need to
pull data from the frame, so that future prefetch() we might add
are more effective.
These skb attributes derive from the CQE or ring :
ip_summed, csum
hash
vlan offload
hwtstamps
queue_mapping
As a bonus, this patch removes mlx4 dependency on eth_get_headlen()
which is very often broken enough to give us headaches.
Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: Tariq Toukan <tariqt@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Wed, 8 Mar 2017 16:17:13 +0000 (08:17 -0800)]
mlx4: add page recycling in receive path
Same technique than some Intel drivers, for arches where PAGE_SIZE = 4096
In most cases, pages are reused because they were consumed
before we could loop around the RX ring.
This brings back performance, and is even better,
a single TCP flow reaches 30Gbit on my hosts.
v2: added full memset() in mlx4_en_free_frag(), as Tariq found it was needed
if we switch to large MTU, as priv->log_rx_info can dynamically be changed.
Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: Tariq Toukan <tariqt@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Wed, 8 Mar 2017 16:17:12 +0000 (08:17 -0800)]
mlx4: use order-0 pages for RX
Use of order-3 pages is problematic in some cases.
This patch might add three kinds of regression :
1) a CPU performance regression, but we will add later page
recycling and performance should be back.
2) TCP receiver could grow its receive window slightly slower,
because skb->len/skb->truesize ratio will decrease.
This is mostly ok, we prefer being conservative to not risk OOM,
and eventually tune TCP better in the future.
This is consistent with other drivers using 2048 per ethernet frame.
3) Because we allocate one page per RX slot, we consume more
memory for the ring buffers. XDP already had this constraint anyway.
Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: Tariq Toukan <tariqt@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Mon, 6 Mar 2017 20:25:20 +0000 (21:25 +0100)]
mlxsw: spectrum: Fix helper function and port variable names
Commit dd82364c3ab9 ("mlxsw: Flip to the new dev walk API") did some
small changes in mlxsw code, but it did not respect the naming
conventions. So fix this now.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Reviewed-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 9 Mar 2017 07:08:59 +0000 (23:08 -0800)]
Merge branch 'flow_dissector-improvements'
Jiri Pirko says:
====================
flow dissector improvements
This patchset follows-up the discussion about future extensions of flow
dissector and tries to address the mentioned concerns. Some parts are
cut out into sub-functions. Also, the processing of the code (ARP, MPLS)
is made dependent on user actually requiring the bisected values.
This prepares the code for future extensions to bisect IPv6 ND messages,
TCP flags, etc.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Mon, 6 Mar 2017 15:39:53 +0000 (16:39 +0100)]
flow_dissector: Fix GRE header error path
Now, when an unexpected element in the GRE header appears, we break so
the l4 ports are processed. But since the ports are processed
unconditionally, there will be certainly random values dissected. Fix
this by just bailing out in such situations.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Mon, 6 Mar 2017 15:39:52 +0000 (16:39 +0100)]
flow_dissector: Move MPLS dissection into a separate function
Make the main flow_dissect function a bit smaller and move the MPLS
dissection into a separate function. Along with that, do the MPLS header
processing only in case the flow dissection user requires it.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Mon, 6 Mar 2017 15:39:51 +0000 (16:39 +0100)]
flow_dissector: Move ARP dissection into a separate function
Make the main flow_dissect function a bit smaller and move the ARP
dissection into a separate function. Along with that, do the ARP header
processing only in case the flow dissection user requires it.
Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Philippe Reynes [Sat, 4 Mar 2017 11:42:39 +0000 (12:42 +0100)]
net: smsc: smc91x: use new api ethtool_{get|set}_link_ksettings
The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.
As I don't have the hardware, I'd be very pleased if
someone may test this patch.
Signed-off-by: Philippe Reynes <tremyfr@gmail.com> Tested-by: Robert Jarzmik <robert.jarzmik@free.fr> Signed-off-by: David S. Miller <davem@davemloft.net>
Joey Zhong [Tue, 28 Feb 2017 18:51:01 +0000 (13:51 -0500)]
net: ethernet: bgmac: driver power manangement
Implement suspend/resume callbacks in the bgmac driver. This makes sure
that we de-initialize and re-initialize the hardware correctly before
entering suspend and when resuming.
Signed-off-by: Joey Zhong <zhongx@broadcom.com> Signed-off-by: Jon Mason <jon.mason@broadcom.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jon Mason [Tue, 28 Feb 2017 18:51:00 +0000 (13:51 -0500)]
net: ethernet: bgmac: unify code of the same family
BCM471X and BCM535X are of the same family (from what I can derive from
internal documents). Group them into the case statement together, which
results in more code reuse.
Also, use existing helper variables to make the code a little more
readable too.
Signed-off-by: Jon Mason <jon.mason@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jon Mason [Tue, 28 Feb 2017 18:50:59 +0000 (13:50 -0500)]
net: ethernet: bgmac: use #defines for MAX size
The maximum frame size is really just the standard ethernet frame size
and FCS. So use those existing defines to make the code a little more
beautiful.
Signed-off-by: Jon Mason <jon.mason@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
This is mostly just a refactoring of previous functions
(get_pkt_next_raw, put_pkt_raw and commit_rd_index) to make it easier
to use for other drivers and NAPI.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
There are two variables named packet in the same function. One is the
metadata descriptor from host (vmpacket_descriptor) and the other is
the control block in the skb used to hold metadata from send.
Change name to avoid possible confusion and bugs.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David Forster [Thu, 23 Feb 2017 16:27:18 +0000 (16:27 +0000)]
ipv6: Provide ipv6 version of "disable_policy" sysctl
This provides equivalent functionality to the existing ipv4
"disable_policy" systcl. ie. Allows IPsec processing to be skipped
on terminating packets on a per-interface basis.
Signed-off-by: David Forster <dforster@brocade.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: phy: Do not perform software reset for Generic PHY
The Generic PHY driver is a catch-all PHY driver and it should preserve
whatever prior initialization has been done by boot loader or firmware
agents. For specific PHY device configuration it is expected that a
specialized PHY driver would take over that role.
Resetting the generic PHY was a bad idea that has lead to several
complaints and downstream workarounds e.g: in OpenWrt/LEDE so restore
the behavior prior to 87aa9f9c61ad ("net: phy: consolidate PHY
reset in phy_init_hw()").
Reported-by: Felix Fietkau <nbd@nbd.name> Fixes: 87aa9f9c61ad ("net: phy: consolidate PHY reset in phy_init_hw()") Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
map_get_next_key callback is mandatory. Supply dummy handler.
Fixes: b95a5c4db09b ("bpf: add a longest prefix match trie map implementation") Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
4) Fix sendmsg deadlock in rxrpc, from David Howells.
5) Add missing RCU locking to transport hashtable scan, from Xin Long.
6) Fix potential packet loss in mlxsw driver, from Ido Schimmel.
7) Fix race in NAPI handling between poll handlers and busy polling,
from Eric Dumazet.
8) TX path in vxlan and geneve need proper RCU locking, from Jakub
Kicinski.
9) SYN processing in DCCP and TCP need to disable BH, from Eric
Dumazet.
10) Properly handle net_enable_timestamp() being invoked from IRQ
context, also from Eric Dumazet.
11) Fix crash on device-tree systems in xgene driver, from Alban Bedel.
12) Do not call sk_free() on a locked socket, from Arnaldo Carvalho de
Melo.
13) Fix use-after-free in netvsc driver, from Dexuan Cui.
14) Fix max MTU setting in bonding driver, from WANG Cong.
15) xen-netback hash table can be allocated from softirq context, so use
GFP_ATOMIC. From Anoob Soman.
16) Fix MAC address change bug in bgmac driver, from Hari Vyas.
17) strparser needs to destroy strp_wq on module exit, from WANG Cong.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (69 commits)
strparser: destroy workqueue on module exit
sfc: fix IPID endianness in TSOv2
sfc: avoid max() in array size
rds: remove unnecessary returned value check
rxrpc: Fix potential NULL-pointer exception
nfp: correct DMA direction in XDP DMA sync
nfp: don't tell FW about the reserved buffer space
net: ethernet: bgmac: mac address change bug
net: ethernet: bgmac: init sequence bug
xen-netback: don't vfree() queues under spinlock
xen-netback: keep a local pointer for vif in backend_disconnect()
netfilter: nf_tables: don't call nfnetlink_set_err() if nfnetlink_send() fails
netfilter: nft_set_rbtree: incorrect assumption on lower interval lookups
netfilter: nf_conntrack_sip: fix wrong memory initialisation
can: flexcan: fix typo in comment
can: usb_8dev: Fix memory leak of priv->cmd_msg_buffer
can: gs_usb: fix coding style
can: gs_usb: Don't use stack memory for USB transfers
ixgbe: Limit use of 2K buffers on architectures with 256B or larger cache lines
ixgbe: update the rss key on h/w, when ethtool ask for it
...
Linus Torvalds [Sat, 4 Mar 2017 19:36:19 +0000 (11:36 -0800)]
Merge tag 'kvm-4.11-2' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull more KVM updates from Radim Krčmář:
"Second batch of KVM changes for the 4.11 merge window:
PPC:
- correct assumption about ASDR on POWER9
- fix MMIO emulation on POWER9
x86:
- add a simple test for ioperm
- cleanup TSS (going through KVM tree as the whole undertaking was
caused by VMX's use of TSS)
- fix nVMX interrupt delivery
- fix some performance counters in the guest
... and two cleanup patches"
* tag 'kvm-4.11-2' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: nVMX: Fix pending events injection
x86/kvm/vmx: remove unused variable in segment_base()
selftests/x86: Add a basic selftest for ioperm
x86/asm: Tidy up TSS limit code
kvm: convert kvm.users_count from atomic_t to refcount_t
KVM: x86: never specify a sample period for virtualized in_tx_cp counters
KVM: PPC: Book3S HV: Don't use ASDR for real-mode HPT faults on POWER9
KVM: PPC: Book3S HV: Fix software walk of guest process page tables
Linus Torvalds [Sat, 4 Mar 2017 19:26:18 +0000 (11:26 -0800)]
Merge tag 'staging-4.11-rc1-part2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging
Pull staging/IIO driver fixes from Greg KH:
"Here are a few small staging and IIO driver fixes for issues that
showed up after the big set if changes you merged last week.
Nothing major, just small bugs resolved in some IIO drivers, a lustre
allocation fix, and some RaspberryPi driver fixes for reported
problems, as well as a MAINTAINERS entry update.
All of these have been in linux-next for a week with no reported
issues"
* tag 'staging-4.11-rc1-part2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
staging: fsl-mc: fix warning in DT ranges parser
MAINTAINERS: Remove Noralf Trønnes as fbtft maintainer
staging: vchiq_2835_arm: Make cache-line-size a required DT property
staging: bcm2835/mmal-vchiq: unlock on error in buffer_from_host()
staging/lustre/lnet: Fix allocation size for sv_cpt_data
iio: adc: xilinx: Fix error handling
iio: 104-quad-8: Fix off-by-one error when addressing flag register
iio: adc: handle unknow of_device_id data
projects / mirror_ubuntu-artful-kernel.git / log