When CONFIG_RFS_ACCEL or SMP aren't set, we fail to build, fix it.
Also, avoid build warning as of unused function on that setup.
Fixes: c66fa19c40 ('net/mlx4: Add EQ pool')
Reported-by: Michael Ellerman <[email protected]>
Signed-off-by: Matan Barak <[email protected]>
Signed-off-by: Or Gerlitz <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
Currently the bpf frame pointer is set to the old r15. This is
wrong because of packed stack. Fix this and adjust the frame pointer
to respect packed stack. This now generates a prolog like the following:
3ff8001c3fa: eb67f0480024 stmg %r6,%r7,72(%r15)
3ff8001c400: ebcff0780024 stmg %r12,%r15,120(%r15)
3ff8001c406: b904001f lgr %r1,%r15 <- load backchain
3ff8001c40a: 41d0f048 la %r13,72(%r15) <- load adjusted bfp
3ff8001c40e: a7fbfd98 aghi %r15,-616
3ff8001c412: e310f0980024 stg %r1,152(%r15) <- save backchain
Fixes: 0546231057 ("s390/bpf: Add s390x eBPF JIT compiler backend")
Signed-off-by: Michael Holzheu <[email protected]>
Acked-by: Heiko Carstens <[email protected]>
Signed-off-by: Alexei Starovoitov <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
On s390x we have to provide 160 bytes stack space before we can call
the next function. From the 160 bytes that we got from the previous
function we only use 11 * 8 bytes and have 160 - 11 * 8 bytes left.
Currently for BPF we allocate additional 160 - 11 * 8 bytes for the
next function. This is wrong because then the next function only gets:
(160 - 11 * 8) + (160 - 11 * 8) = 2 * 72 = 144 bytes
Fix this and allocate enough memory for the next function.
Fixes: 0546231057 ("s390/bpf: Add s390x eBPF JIT compiler backend")
Signed-off-by: Michael Holzheu <[email protected]>
Acked-by: Heiko Carstens <[email protected]>
Signed-off-by: Alexei Starovoitov <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
One minor fix for last -rc of 4.1. The fix moves the update of an attribute in
topology to *after* it is actually initialized.
* tag 'drm-amdkfd-fixes-2015-06-03' of git://people.freedesktop.org/~gabbayo/linux:
drm/amdkfd: fix topology bug with capability attr.
One more small r7xx audio regression fix for 4.1
* 'drm-fixes-4.1' of git://people.freedesktop.org/~agd5f/linux:
drm/radeon: use proper ACR regisiter for DCE3.2
Pull block layer fixes from Jens Axboe:
"Sending this off now, as I'm not aware of other current bugs, nor do I
expect further fixes before 4.1 final. This contains two fixes:
- a fix for a bdi unregister warning that gets spewed on md, due to a
regression introduced earlier in this cycle. From Neil Brown.
- a fix for a compile warning for NVMe on 32-bit platforms, also a
regression introduced in this cycle. From Arnd Bergmann"
* 'for-linus' of git://git.kernel.dk/linux-block:
NVMe: fix type warning on 32-bit
block: discard bdi_unregister() in favour of bdi_destroy()
OSDL has been gone for many years, looks like there still was
one reference to it.
Signed-off-by: Stephen Hemminger <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
This patch fixes a bug where the number of watch points
was shown before it was actually calculated
Signed-off-by: Alexey Skidanov <[email protected]>
Cc: [email protected]
Signed-off-by: Oded Gabbay <[email protected]>
strnlen_user() can return a number in a range 0 to count +
sizeof(unsigned long) - 1. Clarify the comment at the top of the
function so that users don't think the function returns at most count+1.
Signed-off-by: Jan Kara <[email protected]>
[ Also added commentary about preferably not using this function ]
Signed-off-by: Linus Torvalds <[email protected]>
We can avoid this indirect return variable by directly returning the
error values.
Signed-off-by: Markus Pargmann <[email protected]>
Signed-off-by: Marek Lindner <[email protected]>
Declare the returntype of batadv_compare_eth as bool.
The function called inside this helper function
(ether_addr_equal_unaligned) also uses bool as return value, so there is
no need to return int.
Signed-off-by: Markus Pargmann <[email protected]>
Signed-off-by: Marek Lindner <[email protected]>
It is much clearer to see a bool type as return value than 'int' for
functions that are supposed to return true or false.
Signed-off-by: Markus Pargmann <[email protected]>
Signed-off-by: Marek Lindner <[email protected]>
orig_ifinfo is dereferenced multiple times in batadv_iv_ogm_update_seqnos
before the check for NULL is done. The function also exists at the
beginning when orig_ifinfo would have been NULL. This makes the check at
the end unnecessary and only confuses the reader/code analyzers.
Signed-off-by: Sven Eckelmann <[email protected]>
Signed-off-by: Marek Lindner <[email protected]>
batadv_orig_bat_iv->bcast_own is actually not a bitfield, it is an
array. Adjust the comment accordingly.
Signed-off-by: Markus Pargmann <[email protected]>
Signed-off-by: Antonio Quartulli <[email protected]>
Signed-off-by: Marek Lindner <[email protected]>
Bump the version to reflect the driver changes and bug fixes for i219.
Also update the copyright, while we are at it.
Signed-off-by: Yanir Lubetkin <[email protected]>
Tested-by: Aaron Brown <[email protected]>
Signed-off-by: Jeff Kirsher <[email protected]>
System would hang during execution of "ethtool -t <NIC>" for the same
reason that required flushing the descriptor rings. This fix disables
MULR for the loopback test to avoid the hang state.
Signed-off-by: Yanir Lubetkin <[email protected]>
Tested-by: Aaron Brown <[email protected]>
Signed-off-by: Jeff Kirsher <[email protected]>
Two issues involving systim were reported.
1. Clock is not running in the correct frequency
2. In some situations, systim values were not incremented linearly
This patch fixes the hardware clock configuration and the spurious
non-linear increment.
Signed-off-by: Yanir Lubetkin <[email protected]>
Tested-by: Aaron Brown <[email protected]>
Signed-off-by: Jeff Kirsher <[email protected]>
This fix handles a hardware issue that prevented i219 from
working in legacy interrupts mode (IntMode=0)
Signed-off-by: Yanir Lubetkin <[email protected]>
Tested-by: Aaron Brown <[email protected]>
Signed-off-by: Jeff Kirsher <[email protected]>
The indication that a descriptor ring flush is required was read from
FEXTNVM7 by mistake. It should be read from the PCI config space.
Signed-off-by: Yanir Lubetkin <[email protected]>
Tested-by: Aaron Brown <[email protected]>
Signed-off-by: Jeff Kirsher <[email protected]>
The condition under which the flush should occur was reversed. The fix
should be applied before any HW reset (unless followed by bus reset)
and before any power state transition from D0.
If E1000_FEXTNVM7_NEED_DESCRING_FLUSH bit is set in FEXTNVM7 and TDLEN > 0
the Tx ring should be flushed. (fixes ~95% of the hang states).
If the E1000_FEXTNVM7_NEED_DESCRING_FLUSH did not clear, we should also
flush the RX ring. Bug was caught by Alexander Duyck during a code review
when examining this fix.
Signed-off-by: Yanir Lubetkin <[email protected]>
Tested-by: Aaron Brown <[email protected]>
Signed-off-by: Jeff Kirsher <[email protected]>
After testing various cases, the conclusion is that the fix MUST be
executed BEFORE any event that the HW is reset or transition to D3.
To fix that I moved the execution to the relevant places but per
Alexander Duyck's review, ensure now that the DMA is valid and was not
freed before manipulating the ring.
Signed-off-by: Yanir Lubetkin <[email protected]>
Tested-by: Aaron Brown <[email protected]>
Signed-off-by: Jeff Kirsher <[email protected]>
Unit hang may occur if multiple descriptors are available in the rings
during reset or runtime suspend. This state can be detected by testing
bit 8 in the FEXTNVM7 register. If this bit is set and there are pending
descriptors in one of the rings, we must flush them prior to reset. Same
applies entering runtime suspend.
Signed-off-by: Yanir Lubetkin <[email protected]>
Reviewed-by: Alexander Duyck <[email protected]>
Signed-off-by: Jeff Kirsher <[email protected]>
The driver worked around an error in the MAYA44 USB(+)'s mixer unit
descriptor by aborting before parsing the missing field. However,
aborting parsing too early prevented parsing of the other units
connected to this unit, so the capture mixer controls would be missing.
Fix this by moving the check for this descriptor error after the parsing
of the unit's input pins.
Reported-by: nightmixes <[email protected]>
Tested-by: nightmixes <[email protected]>
Signed-off-by: Clemens Ladisch <[email protected]>
Cc: <[email protected]>
Signed-off-by: Takashi Iwai <[email protected]>
Add mixer control names for the ESI Maya44 USB+ (which appears to be
identical width the AudioTrak Maya44 USB).
Reported-by: nightmixes <[email protected]>
Signed-off-by: Clemens Ladisch <[email protected]>
Cc: <[email protected]>
Signed-off-by: Takashi Iwai <[email protected]>
platforms that have special power states)
* Rework completely the RBD allocation model towards new
multi RX hardware.
* cleanups
* scan reworks continuation (Luca)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=9GDL
-----END PGP SIGNATURE-----
Merge tag 'iwlwifi-next-for-kalle-2015-06-03' of https://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/iwlwifi-next
* a few fixes (re-enablement of interrupts for certain new
platforms that have special power states)
* Rework completely the RBD allocation model towards new
multi RX hardware.
* cleanups
* scan reworks continuation (Luca)
This is a small copy paste fix for batadv_ing_buffer_avg.
Signed-off-by: Markus Pargmann <[email protected]>
Signed-off-by: Marek Lindner <[email protected]>
The kernel coding style says, that there should not be multiple
assignments in one row.
Signed-off-by: Markus Pargmann <[email protected]>
Signed-off-by: Marek Lindner <[email protected]>
CodingStyle describes that either none or both branches of a conditional
have to have brackets.
Signed-off-by: Markus Pargmann <[email protected]>
Signed-off-by: Marek Lindner <[email protected]>
It is just a bit easier to put the error handling at one place and let
multiple error paths use the same calls.
Signed-off-by: Markus Pargmann <[email protected]>
Signed-off-by: Marek Lindner <[email protected]>
Use the default FCSR value in mask probing, avoiding an FPE exception
where reset has left any exception enable and their corresponding cause
bits set and the register is then rewritten with these bits active.
Signed-off-by: Maciej W. Rozycki <[email protected]>
Cc: Joshua Kinard <[email protected]>
Cc: [email protected]
Signed-off-by: Ralf Baechle <[email protected]>
After the new ciphers CCMP-256 and GCMP-128/256 were implemented,
wpa_supplicant could start negotiating them and use the software
implementation. This, however, breaks D3 behaviour in the driver
since it means that WoWLAN will not be possible.
To avoid breaking that feature, advertise only ciphers that the
hardware supports.
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
Add DC2DC_CONFIG_CMD (0x83) cmd.
Add IWL_UCODE_TLV_CAPA_DC2DC_CONFIG_SUPPORT tlv.
The command allows the driver get & set the DCDC's frequency tune.
(freq_tune is the divider that is used to calculate the actual DCDC's
clock rate)
The command always returns the current/updated frequency tune values of
the DCDC.
Signed-off-by: Matti Gottlieb <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
The firmwares that used these commands is not supported
anymore. Remove them.
Signed-off-by: Matti Gottlieb <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
Currently, loading the firmware fails when it has higher API or CAPA
bits than the driver supports. That's an issue with integration.
At the same time, actually using api[0] and capa[0] will become
confusing when we also have api[1] and capa[1], and it's almost
certain that we'll mix up the bits and use the bits for api[1] with
api[0] by accident.
Avoid all this by translating the API/CAPA bits to the regular kernel
test_bit() format, and also providing wrapper functions. Also use the
__bitwise__ facility of sparse to check that we're testing the right
one.
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
As pointed out by smatch, there's no need for a loop that always
immediately terminates. Use an if statement instead and while at
it clean up the mvmvif initialization.
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
Since the firmware is responsible for duration calculation, the
driver can easily support fast-xmit.
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
Since the firmware is responsible for duration calculation, the
driver can easily support fast-xmit.
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
As a preperation for multiple RX queues change the RBD
allocation model.
The new model includes a background allocator. The allocator is
called by the interrupt handler when there are two released
buffers by the queue, and the allocator starts allocating eight
pages per request.
When the queue has released 8 pages it tries claiming the
request. If the pages are not ready - it keeps claiming.
This new model should make sure that RBDs are always available
across the multiple queues.
The RBDs are transferred between the allocator and the queue.
The queue moves the free RBDs upon freeing them to the allocator.
The allocator moves them back to the queue's possession when the
request is claimed.
The allocator has an initial pool to make sure there are always RBDs
available for the request completion.
Release of the buffers at exit is done per pools - the allocator
frees its own initial pool and the queue frees its own pool.
Existing code refactor -
-Queue's initial pool is the size of the queue only as the allocation
of the new buffers no longer uses this pool.
-Removal of replenish background work, and replenish calls in the
interrupt handler and restock().
-The replenish() and the rxq used_list are used only during
initialization.
-Moved page allocation to a new function for code reuse.
New code -
Allocator code - new structure and functions.
Interrupt handler uses the allocator functions for replenishing buffers.
Reuse of the restock() method.
Signed-off-by: Sara Sharon <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
On resume, all the interrupts are masked (CSR_INT_MASK is 0),
and ict is disabled.
Re-configure them both.
Signed-off-by: Eliad Peller <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
Don't use EBS for P2P find to make sure we find all GOs in
our only attempt.
Signed-off-by: David Spinadel <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
Currently EBS status in scan complete notifications is set to
success if EBS wasn't activated. FW will add a special return
value for cases when EBS wasn't activated and we add a print
of this status.
This change is needed for debug only, no behavior changes.
Signed-off-by: David Spinadel <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
Allow a cleaner way to access those hw-dependent registers,
instead of using the product family type etc.
Signed-off-by: Avri Altman <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
This adds the MIPI mode type to the types declared supported
by the driver. Without this patch, when using MIPI mode and
looking at the logs the user would see the debug destination
"UNKNOWN".
Signed-off-by: Liad Kaufman <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
A minor refactoring for following patches.
This enables the reuse of the checks functions.
type=cleanup
Signed-off-by: Eyal Shapira <[email protected]>
Signed-off-by: Emmanuel Grumbach <[email protected]>
The variable translate is bool type. So assigning true instead of 1.
Signed-off-by: Shailendra Verma <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
