At the end of the nexthop initialization process we determine whether
the nexthop should be offloaded or not based on the NUD state of the
neighbour representing it. After all the nexthops were initialized we
refresh the nexthop group and potentially offload it to the device, in
case some of the nexthops were resolved.

Make the destruction of a nexthop group symmetric with its creation by
marking all nexthops as invalid and then refresh the nexthop group to
make sure it was removed from the device's tables.

Fixes: b2157149 ("mlxsw: spectrum_router: Add the nexthop neigh activity update")
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When a neighbour is considered to be dead, we should remove it from the
device's table regardless of its NUD state.

Without this patch, after setting a port to be administratively down we
get the following errors when we periodically try to update the kernel
about neighbours activity:

[  461.947268] mlxsw_spectrum 0000:03:00.0 sw1p3: Failed to find
matching neighbour for IP=192.168.100.2

Fixes: a6bf9e93 ("mlxsw: spectrum_router: Offload neighbours based on NUD state change")
Signed-off-by: NIdo Schimmel <idosch@mellanox.com>
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When testing stmmac with my QoS reference design I checked a problem in the
CSR clock configuration that was impossibilitating the phy discovery, since
every read operation returned 0x0000ffff. This patch fixes the issue.
Signed-off-by: NJoao Pinto <jpinto@synopsys.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Trivial fix: Addresses should be printed using the %p format specifier
rather than using %x.
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since commit 71ce391d ("net: mvpp2: enable proper per-CPU TX
buffers unmapping"), we are not correctly DMA unmapping TX buffers for
fragments.

Indeed, the mvpp2_txq_inc_put() function only stores in the
txq_cpu->tx_buffs[] array the physical address of the buffer to be
DMA-unmapped when skb != NULL. In addition, when DMA-unmapping, we use
skb_headlen(skb) to get the size to be unmapped. Both of this works fine
for TX descriptors that are associated directly to a SKB, but not the
ones that are used for fragments, with a NULL pointer as skb:

 - We have a NULL physical address when calling DMA unmap
 - skb_headlen(skb) crashes because skb is NULL

This causes random crashes when fragments are used.

To solve this problem, we need to:

 - Store the physical address of the buffer to be unmapped
   unconditionally, regardless of whether it is tied to a SKB or not.

 - Store the length of the buffer to be unmapped, which requires a new
   field.

Instead of adding a third array to store the length of the buffer to be
unmapped, and as suggested by David Miller, this commit refactors the
tx_buffs[] and tx_skb[] arrays of 'struct mvpp2_txq_pcpu' into a
separate structure 'mvpp2_txq_pcpu_buf', to which a 'size' field is
added. Therefore, instead of having three arrays to allocate/free, we
have a single one, which also improve data locality, reducing the
impact on the CPU cache.

Fixes: 71ce391d ("net: mvpp2: enable proper per-CPU TX buffers unmapping")
Reported-by: NRaphael G <raphael.glon@corp.ovh.com>
Cc: Raphael G <raphael.glon@corp.ovh.com>
Cc: stable@vger.kernel.org
Signed-off-by: NThomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Right now the dwmac-rk tries to set up the GRF-specific speed and link
options before enabling clocks, phys etc and on previous socs this works
because the GRF is supplied on the whole by one clock.

On the rk3399 however the GRF (General Register Files) clock-supply
has been split into multiple clocks and while there is no specific
grf-gmac clock like for other sub-blocks, it seems the mac-specific
portions are actually supplied by the general mac clock.

This results in hangs on rk3399 boards if the driver is build as module.
When built in te problem of course doesn't surface, as the clocks
are of course still on at the stage before clock_disable_unused.

To solve this, simply move the clock enablement to the first position
in the powerup callback. This is also a good idea in general to
enable clocks before everything else.

Tested on rk3288, rk3368 and rk3399 the dwmac still works on all of them.
Signed-off-by: NHeiko Stuebner <heiko@sntech.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The driver currently allocates 128 bytes of skb headroom.
This was found to be insufficient with some configurations
like Geneve tunnels, which resulted in skb head reallocations.

Increase the headroom to 256 bytes to fix this.
Signed-off-by: NKalesh A P <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: NSuresh Reddy <suresh.reddy@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

With commit e4965614 {"rtlwifi: Use dev_kfree_skb_irq instead of
kfree_skb"), the method used to free an skb was changed because the
kfree_skb() was inside a spinlock. What was forgotten is that kfree_skb()
guards against a NULL value for the argument. Routine dev_kfree_skb_irq()
does not, and a test is needed to prevent kernel panics.

Fixes: e4965614 ("rtlwifi: Use dev_kfree_skb_irq instead of kfree_skb")
Signed-off-by: NLarry Finger <Larry.Finger@lwfinger.net>
Cc: Stable <stable@vger.kernel.org> # 4.9+
Cc: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: NKalle Valo <kvalo@codeaurora.org>

Starting with commit d94a461d ("ath9k: use ieee80211_tx_status_noskb
where possible") the driver uses rcu_read_lock() && rcu_read_unlock(), yet on
returning early in ath_tx_edma_tasklet() the unlock is missing leading to stalls
and suspicious RCU usage:

 ===============================
 [ INFO: suspicious RCU usage. ]
 4.9.0-rc8 #11 Not tainted
 -------------------------------
 kernel/rcu/tree.c:705 Illegal idle entry in RCU read-side critical section.!

 other info that might help us debug this:

 RCU used illegally from idle CPU!
 rcu_scheduler_active = 1, debug_locks = 0
 RCU used illegally from extended quiescent state!
 1 lock held by swapper/7/0:
 #0:
  (
 rcu_read_lock
 ){......}
 , at:
 [<ffffffffa06ed110>] ath_tx_edma_tasklet+0x0/0x450 [ath9k]

 stack backtrace:
 CPU: 7 PID: 0 Comm: swapper/7 Not tainted 4.9.0-rc8 #11
 Hardware name: Acer Aspire V3-571G/VA50_HC_CR, BIOS V2.21 12/16/2013
  ffff88025efc3f38 ffffffff8132b1e5 ffff88017ede4540 0000000000000001
  ffff88025efc3f68 ffffffff810a25f7 ffff88025efcee60 ffff88017edebdd8
  ffff88025eeb5400 0000000000000091 ffff88025efc3f88 ffffffff810c3cd4
 Call Trace:
  <IRQ>
  [<ffffffff8132b1e5>] dump_stack+0x68/0x93
  [<ffffffff810a25f7>] lockdep_rcu_suspicious+0xd7/0x110
  [<ffffffff810c3cd4>] rcu_eqs_enter_common.constprop.85+0x154/0x200
  [<ffffffff810c5a54>] rcu_irq_exit+0x44/0xa0
  [<ffffffff81058631>] irq_exit+0x61/0xd0
  [<ffffffff81018d25>] do_IRQ+0x65/0x110
  [<ffffffff81672189>] common_interrupt+0x89/0x89
  <EOI>
  [<ffffffff814ffe11>] ? cpuidle_enter_state+0x151/0x200
  [<ffffffff814ffee2>] cpuidle_enter+0x12/0x20
  [<ffffffff8109a6ae>] call_cpuidle+0x1e/0x40
  [<ffffffff8109a8f6>] cpu_startup_entry+0x146/0x220
  [<ffffffff810336f8>] start_secondary+0x148/0x170
Signed-off-by: NTobias Klausmann <tobias.johannes.klausmann@mni.thm.de>
Fixes: d94a461d ("ath9k: use ieee80211_tx_status_noskb where possible")
Cc: <stable@vger.kernel.org> # v4.9
Acked-by: NFelix Fietkau <nbd@nbd.name>
Acked-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: NGabriel Craciunescu <nix.or.die@gmail.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

To make the code clearer, use rb_entry() instead of container_of() to
deal with rbtree.
Signed-off-by: NGeliang Tang <geliangtang@gmail.com>
Acked-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since commit

  5a6681e2 ("sfc: separate out SFC4000 ("Falcon") support into new sfc-falcon driver")

there are two drivers for Solarflare devices, but both still show up
directly beneath "Ethernet driver support" in the Kconfig. Follow the
pattern of other vendors and group them beneath an own vendor Kconfig
entry for Solarflare.

Cc: Edward Cree <ecree@solarflare.com>
Signed-off-by: NTobias Klauser <tklauser@distanz.ch>
Acked-by: NEdward Cree <ecree@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The SoC hix5hd2 compatible string has the suffix "-gmac" and
we should not change its compatible string.
So we should name all the compatible string with the suffix "-gmac".
Creating a new name suffix "-gemac" is unnecessary.

We also add another SoC compatible string in dt binding documentation
and describe which generic version the SoC belongs to.

Fixes: d0fb6ba7 ("net: hix5hd2_gmac: add generic compatible string")
Signed-off-by: NDongpo Li <lidongpo@hisilicon.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds the missing 10gbe host port tx priority map
configurations.
Signed-off-by: NWingMan Kwok <w-kwok2@ti.com>
Signed-off-by: NMurali Karicheri <m-karicheri2@ti.com>
Signed-off-by: NSekhar Nori <nsekhar@ti.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In ethtool ops, it needs to retrieve the corresponding
ethss module (gbe or xgbe) from the net_device structure.
Prior to this patch, the retrieving procedure only
checks for the gbe module.  This patch fixes the issue
by checking the xgbe module if the net_device structure
does not correspond to the gbe module.
Signed-off-by: NWingMan Kwok <w-kwok2@ti.com>
Signed-off-by: NMurali Karicheri <m-karicheri2@ti.com>
Signed-off-by: NSekhar Nori <nsekhar@ti.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NMadalin Bucur <madalin.bucur@nxp.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NMadalin Bucur <madalin.bucur@nxp.com>
Reviewed-by: NCamelia Groza <camelia.groza@nxp.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The fsl/fman drivers will use of_platform_populate() on all
supported platforms. Call of_platform_populate() to probe the
FMan sub-nodes.
Signed-off-by: NIgal Liberman <igal.liberman@freescale.com>
Signed-off-by: NMadalin Bucur <madalin.bucur@nxp.com>
Acked-by: NScott Wood <oss@buserror.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

QSGMII ports were not advertising 1G speed.
Signed-off-by: NMadalin Bucur <madalin.bucur@nxp.com>
Reviewed-by: NCamelia Groza <camelia.groza@nxp.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The patches regarding eee-broken-modes was merged before all people
involved could find an agreement on the best way to move forward.

While we agreed on having a DT property to mark particular modes as broken,
the value used for eee-broken-modes mapped the phy register in very direct
way. Because of this, the concern is that it could be used to implement
configuration policies instead of describing a broken HW.

In the end, having a boolean property for each mode seems to be preferred
over one bit field value mapping the register (too) directly.

Cc: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NJerome Brunet <jbrunet@baylibre.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In genphy_config_eee_advert, the return value of phy_read_mmd_indirect is
checked to know if the register could be accessed but the result is
assigned to a 'u32'.
Changing to 'int' to correctly get errors from phy_read_mmd_indirect.

Fixes: d853d145 ("net: phy: add an option to disable EEE advertisement")
Reported-by: NJulia Lawall <julia.lawall@lip6.fr>
Signed-off-by: NJerome Brunet <jbrunet@baylibre.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Hopefully this fixes the problem reported by Kalle:

Noticed this in my log, but I don't have time to investigate this in
detail right now:

[  413.795346] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready
[  414.158755] IPv6: ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
[  477.439659] ath10k_pci 0000:02:00.0: could not get mac80211 beacon
[  481.666630] ------------[ cut here ]------------
[  481.666669] WARNING: CPU: 0 PID: 1978 at lib/dma-debug.c:1155 check_unmap+0x320/0x8e0
[  481.666688] ath10k_pci 0000:02:00.0: DMA-API: device driver frees DMA memory with different direction [device address=0x000000002d130000] [size=63800 bytes] [mapped with DMA_BIDIRECTIONAL] [unmapped with DMA_TO_DEVICE]
[  481.666703] Modules linked in: ctr ccm ath10k_pci(E-) ath10k_core(E) ath(E) mac80211(E) cfg80211(E) snd_hda_codec_hdmi snd_hda_codec_idt snd_hda_codec_generic snd_hda_intel snd_hda_codec snd_hda_core snd_hwdep snd_pcm snd_seq_midi arc4 snd_rawmidi snd_seq_midi_event snd_seq btusb btintel snd_seq_device joydev coret
[  481.671468] CPU: 0 PID: 1978 Comm: rmmod Tainted: G            E   4.9.0-rc7-wt+ #54
[  481.671478] Hardware name: Hewlett-Packard HP ProBook 6540b/1722, BIOS 68CDD Ver. F.04 01/27/2010
[  481.671489]  ef49dcec c842ee92 c8b5830e ef49dd34 ef49dd20 c80850f5 c8b5a13c ef49dd50
[  481.671560]  000007ba c8b5830e 00000483 c8461830 c8461830 00000483 ef49ddcc f34e64b8
[  481.671641]  c8b58360 ef49dd3c c80851bb 00000009 00000000 ef49dd34 c8b5a13c ef49dd50
[  481.671716] Call Trace:
[  481.671731]  [<c842ee92>] dump_stack+0x76/0xb4
[  481.671745]  [<c80850f5>] __warn+0xe5/0x100
[  481.671757]  [<c8461830>] ? check_unmap+0x320/0x8e0
[  481.671769]  [<c8461830>] ? check_unmap+0x320/0x8e0
[  481.671780]  [<c80851bb>] warn_slowpath_fmt+0x3b/0x40
[  481.671791]  [<c8461830>] check_unmap+0x320/0x8e0
[  481.671804]  [<c8462054>] debug_dma_unmap_page+0x84/0xa0
[  481.671835]  [<f937cd7a>] ath10k_wmi_free_host_mem+0x9a/0xe0 [ath10k_core]
[  481.671861]  [<f9363400>] ath10k_core_destroy+0x50/0x60 [ath10k_core]
[  481.671875]  [<f8e13969>] ath10k_pci_remove+0x79/0xa0 [ath10k_pci]
[  481.671889]  [<c848d8d8>] pci_device_remove+0x38/0xb0
[  481.671901]  [<c859fe4b>] __device_release_driver+0x7b/0x110
[  481.671913]  [<c85a00e7>] driver_detach+0x97/0xa0
[  481.671923]  [<c859ef8b>] bus_remove_driver+0x4b/0xb0
[  481.671934]  [<c85a0cda>] driver_unregister+0x2a/0x60
[  481.671949]  [<c848c888>] pci_unregister_driver+0x18/0x70
[  481.671965]  [<f8e14dae>] ath10k_pci_exit+0xd/0x25f [ath10k_pci]
[  481.671979]  [<c812bb84>] SyS_delete_module+0xf4/0x180
[  481.671995]  [<c81f801b>] ? __might_fault+0x8b/0xa0
[  481.672009]  [<c80037d0>] do_fast_syscall_32+0xa0/0x1e0
[  481.672025]  [<c88d4c88>] sysenter_past_esp+0x45/0x74
[  481.672037] ---[ end trace 3fd23759e17e1622 ]---
[  481.672049] Mapped at:
[  481.672060]  [  481.672072] [<c846062c>] debug_dma_map_page.part.25+0x1c/0xf0
[  481.672083]  [  481.672095] [<c8460799>] debug_dma_map_page+0x99/0xc0
[  481.672106]  [  481.672132] [<f93745ec>] ath10k_wmi_alloc_chunk+0x12c/0x1f0 [ath10k_core]
[  481.672142]  [  481.672168] [<f937d0c4>] ath10k_wmi_event_service_ready_work+0x304/0x540 [ath10k_core]
[  481.672178]  [  481.672190] [<c80a3643>] process_one_work+0x1c3/0x670
[  482.137134] ath10k_pci 0000:02:00.0: pci irq msi oper_irq_mode 2 irq_mode 0 reset_mode 0
[  482.313144] ath10k_pci 0000:02:00.0: Direct firmware load for ath10k/pre-cal-pci-0000:02:00.0.bin failed with error -2
[  482.313274] ath10k_pci 0000:02:00.0: Direct firmware load for ath10k/cal-pci-0000:02:00.0.bin failed with error -2
[  482.313768] ath10k_pci 0000:02:00.0: qca988x hw2.0 target 0x4100016c chip_id 0x043202ff sub 0000:0000
[  482.313777] ath10k_pci 0000:02:00.0: kconfig debug 1 debugfs 1 tracing 1 dfs 0 testmode 1
[  482.313974] ath10k_pci 0000:02:00.0: firmware ver 10.2.4.70.59-2 api 5 features no-p2p,raw-mode,mfp,allows-mesh-bcast crc32 4159f498
[  482.369858] ath10k_pci 0000:02:00.0: Direct firmware load for ath10k/QCA988X/hw2.0/board-2.bin failed with error -2
[  482.370011] ath10k_pci 0000:02:00.0: board_file api 1 bmi_id N/A crc32 bebc7c08
[  483.596770] ath10k_pci 0000:02:00.0: htt-ver 2.1 wmi-op 5 htt-op 2 cal otp max-sta 128 raw 0 hwcrypto 1
[  483.701686] ath: EEPROM regdomain: 0x0
[  483.701706] ath: EEPROM indicates default country code should be used
[  483.701713] ath: doing EEPROM country->regdmn map search
[  483.701721] ath: country maps to regdmn code: 0x3a
[  483.701730] ath: Country alpha2 being used: US
[  483.701737] ath: Regpair used: 0x3a
Reported-by: NKalle Valo <kvalo@qca.qualcomm.com>
Signed-off-by: NBen Greear <greearb@candelatech.com>
Signed-off-by: NKalle Valo <kvalo@qca.qualcomm.com>

The scheduled scan ssid configuration in firmware has a flags field that
was not initialized resulting in unexpected behaviour.

Fixes: e3bdb7cc0300 ("brcmfmac: fix handling ssids in .sched_scan_start() callback")
Reviewed-by: NHante Meuleman <hante.meuleman@broadcom.com>
Reviewed-by: NPieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com>
Reviewed-by: NFranky Lin <franky.lin@broadcom.com>
Signed-off-by: NArend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: NKalle Valo <kvalo@codeaurora.org>

In brcmf_cfg80211_attach() there was one error path not properly
handled as it leaked memory allocated in brcmf_btcoex_attach().
Reviewed-by: NHante Meuleman <hante.meuleman@broadcom.com>
Reviewed-by: NPieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com>
Reviewed-by: NFranky Lin <franky.lin@broadcom.com>
Signed-off-by: NArend van Spriel <arend.vanspriel@broadcom.com>
Signed-off-by: NKalle Valo <kvalo@codeaurora.org>

Fix up memory barriers in stmmac driver. They are meant to protect
against DMA engine, so smp_ variants are certainly wrong, and dma_
variants are preferable.
Signed-off-by: NPavel Machek <pavel@denx.de>
Tested-by: NNiklas Cassel <niklas.cassel@axis.com>
Acked-by: NGiuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Here, If devm_ioremap will fail. It will return NULL.
Kernel can run into a NULL-pointer dereference.
This error check will avoid NULL pointer dereference.
Signed-off-by: NArvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When CONFIG_SRAM is enable and the SRAM region is found, the entire SRAM
region resource is requested and marked as occupied by SRAM driver even
if certain parts of regions is marked reserved.

It's quite possible that a small region of the SRAM is reserved for all
the mailbox communication and hence it may fail to request the region
as it's already marked busy region.

This patch tries to just do a ioremap of this mailbox memory region if
it finds it busy.

Cc: Lee Jones <lee.jones@linaro.org>
Signed-off-by: NSudeep Holla <sudeep.holla@arm.com>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

Currently the read operation on the message debug file returns error if
there's no data ready to be read. It expects the userspace to retry if
it fails. Since the mailbox response could be asynchronous, it would be
good to add support to block the read until the data is available.

We can also implement poll file operations so that the userspace can
wait to become ready to perform any I/O.

This patch implements the poll and fasync file operation callback for
the test mailbox device.

Cc: Lee Jones <lee.jones@linaro.org>
Signed-off-by: NSudeep Holla <sudeep.holla@arm.com>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

Remove unnecessary void* casts in register writes. Fix two other
minor formatting issues.
Signed-off-by: NRob Rice <rob.rice@broadcom.com>
Reviewed-by: NAndy Gospodarek <gospo@broadcom.com>
Reviewed-by: NJon Mason <jon.mason@broadcom.com>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

Earlier versions of the PDC driver registered for both
transmit and receive interrupts. The hard IRQ handler had to
communicate to the soft handler which interrupt(s) had occurred.
The PDC driver no longer registers for tx interrupts. So there is
no reason to save the intstatus. So remove the intstatus member
of the PDC state.
Signed-off-by: NRob Rice <rob.rice@broadcom.com>
Reviewed-by: NAndy Gospodarek <gospo@broadcom.com>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

Three changes to improve performance in the PDC driver:
- disable and reenable interrupts while the interrupt handler is
running
- update rxin and txin descriptor indexes more efficiently
- group receive descriptor context into a structure and keep
context in a single array rather than five to improve locality
of reference
Signed-off-by: NRob Rice <rob.rice@broadcom.com>
Reviewed-by: NAndy Gospodarek <gospo@broadcom.com>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

In PDC driver, it is not necessary to use iowrite32()
when writing DMA descriptors to the transmit and receive rings.
The ring memory is in host memory. So convert to normal
assignment statements.
Signed-off-by: NRob Rice <rob.rice@broadcom.com>
Reviewed-by: NAndy Gospodarek <gospo@broadcom.com>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

Previously used threaded IRQs in the PDC driver to defer
processing the rx DMA ring after getting an rx done interrupt.
Instead, use a tasklet at normal priority for deferred processing.
Signed-off-by: NRob Rice <rob.rice@broadcom.com>
Reviewed-by: NAndy Gospodarek <gospo@broadcom.com>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

Use likely/unlikely directives to improve branch prediction.
Signed-off-by: NRob Rice <rob.rice@broadcom.com>
Reviewed-by: NAndy Gospodarek <gospo@broadcom.com>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

Remove the unnecessary rmb() from the receive path.

If the rx ring has multiple messages ready, avoid reading
last_rx_curr multiple times from the register.
Signed-off-by: NRob Rice <rob.rice@broadcom.com>
Reviewed-by: NAndy Gospodarek <gospo@broadcom.com>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

The PDC driver is a mailbox controller. A mailbox controller
can report that a mailbox message has been "transmitted" either when
a tx interrupt fires or by having the mailbox framework poll. This
commit converts the PDC driver to the poll method. We found that the
tx interrupt happens when the descriptors are read by the SPU hw. Thus,
the interrupt method does not allow more than one tx message in the PDC
tx DMA ring at a time. To keep the SPU hw busy, we would like to keep
the tx ring full under heavy load.

With the poll method, the PDC driver responds that the previous message
has been transmitted if the tx ring has space for another message.
SPU request messages take a variable number of descriptors. If 15
descriptors are available, there is a good chance another message will
fit. Also increased the ring size from 128 to 512 descriptors.

With this change, I found the PDC driver hangs on its spinlock under
heavy load. The PDC spinlock is not required; so I removed it. Calls
to pdc_send_data() are already synchronized because of the channel
spinlock in the mailbox framework. Other references to ring indexes
should not require locking because they only written on either the
tx or rx side.
Signed-off-by: NRob Rice <rob.rice@broadcom.com>
Reviewed-by: NAndy Gospodarek <gospo@broadcom.com>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

Minor fix to ensure that debugfs stats pseudo-files are
removed when driver module is unloaded.  Previously, the call to
debugfs_remove_recursive() was never being called since the
directory was not empty, and a seg fault would occur if another
process tried to access these leftover files.
Signed-off-by: NSteve Lin <steven.lin1@broadcom.com>
Signed-off-by: NRob Rice <rob.rice@broadcom.com>
Reviewed-by: NAndy Gospodarek <gospo@broadcom.com>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

Ensure that DMA is disabled, and pointers reset, when changing
DMA base addresses in pdc_ring_init().  This allows a mailbox client
to be re-inserted after being removed.  Otherwise, the DMA doesn't
restart so the client hangs while being reinserted.
Signed-off-by: NSteve Lin <steven.lin1@broadcom.com>
Signed-off-by: NRob Rice <rob.rice@broadcom.com>
Reviewed-by: NAndy Gospodarek <gospo@broadcom.com>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

When creating the debugfs files for the PDC driver, use
octal file permissions rather than symbolic file permissions.
Signed-off-by: NRob Rice <rob.rice@broadcom.com>
Reviewed-by: NAndy Gospodarek <gospo@broadcom.com>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

If the driver is built as a module, autoload won't work because the module
alias information is not filled. So user-space can't match the registered
device with the corresponding module.

Export the module alias information using the MODULE_DEVICE_TABLE() macro.

Before this patch:

$ modinfo drivers/mailbox/mailbox-sti.ko | grep alias
alias:          platform:mailbox-sti

After this patch:

$ modinfo drivers/mailbox/mailbox-sti.ko | grep alias
alias:          platform:mailbox-sti
alias:          of:N*T*Cst,stih407-mailboxC*
alias:          of:N*T*Cst,stih407-mailbox
Signed-off-by: NJavier Martinez Canillas <javier@osg.samsung.com>
Acked-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>

If the driver is built as a module, autoload won't work because the module
alias information is not filled. So user-space can't match the registered
device with the corresponding module.

Export the module alias information using the MODULE_DEVICE_TABLE() macro.

Before this patch:

$ modinfo drivers/mailbox/mailbox-test.ko | grep alias
$

After this patch:

$ modinfo drivers/mailbox/mailbox-test.ko | grep alias
alias:          of:N*T*Cmailbox-testC*
alias:          of:N*T*Cmailbox-test
Signed-off-by: NJavier Martinez Canillas <javier@osg.samsung.com>
Acked-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NJassi Brar <jaswinder.singh@linaro.org>