Pull DMA mask updates from Russell King:
"This series cleans up the handling of DMA masks in a lot of drivers,
fixing some bugs as we go.
Some of the more serious errors include:
- drivers which only set their coherent DMA mask if the attempt to
set the streaming mask fails.
- drivers which test for a NULL dma mask pointer, and then set the
dma mask pointer to a location in their module .data section -
which will cause problems if the module is reloaded.
To counter these, I have introduced two helper functions:
- dma_set_mask_and_coherent() takes care of setting both the
streaming and coherent masks at the same time, with the correct
error handling as specified by the API.
- dma_coerce_mask_and_coherent() which resolves the problem of
drivers forcefully setting DMA masks. This is more a marker for
future work to further clean these locations up - the code which
creates the devices really should be initialising these, but to fix
that in one go along with this change could potentially be very
disruptive.
The last thing this series does is prise away some of Linux's addition
to "DMA addresses are physical addresses and RAM always starts at
zero". We have ARM LPAE systems where all system memory is above 4GB
physical, hence having DMA masks interpreted by (eg) the block layers
as describing physical addresses in the range 0..DMAMASK fails on
these platforms. Santosh Shilimkar addresses this in this series; the
patches were copied to the appropriate people multiple times but were
ignored.
Fixing this also gets rid of some ARM weirdness in the setup of the
max*pfn variables, and brings ARM into line with every other Linux
architecture as far as those go"
* 'for-linus-dma-masks' of git://git.linaro.org/people/rmk/linux-arm: (52 commits)
ARM: 7805/1: mm: change max*pfn to include the physical offset of memory
ARM: 7797/1: mmc: Use dma_max_pfn(dev) helper for bounce_limit calculations
ARM: 7796/1: scsi: Use dma_max_pfn(dev) helper for bounce_limit calculations
ARM: 7795/1: mm: dma-mapping: Add dma_max_pfn(dev) helper function
ARM: 7794/1: block: Rename parameter dma_mask to max_addr for blk_queue_bounce_limit()
ARM: DMA-API: better handing of DMA masks for coherent allocations
ARM: 7857/1: dma: imx-sdma: setup dma mask
DMA-API: firmware/google/gsmi.c: avoid direct access to DMA masks
DMA-API: dcdbas: update DMA mask handing
DMA-API: dma: edma.c: no need to explicitly initialize DMA masks
DMA-API: usb: musb: use platform_device_register_full() to avoid directly messing with dma masks
DMA-API: crypto: remove last references to 'static struct device *dev'
DMA-API: crypto: fix ixp4xx crypto platform device support
DMA-API: others: use dma_set_coherent_mask()
DMA-API: staging: use dma_set_coherent_mask()
DMA-API: usb: use new dma_coerce_mask_and_coherent()
DMA-API: usb: use dma_set_coherent_mask()
DMA-API: parport: parport_pc.c: use dma_coerce_mask_and_coherent()
DMA-API: net: octeon: use dma_coerce_mask_and_coherent()
DMA-API: net: nxp/lpc_eth: use dma_coerce_mask_and_coherent()
...
The PCI core caches the "PCIe Max Payload Size Supported" in
pci_dev->pcie_mpss, so use that instead of pcie_capability_read_dword().
Also use pcie_set_readrq() instead of pcie_capability_clear_and_set_word()
to simplify code.
Signed-off-by: Yijing Wang <wangyijing@huawei.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Acked-by: Mark Einon <mark.einon@gmail.com>
Replace the following sequence:
dma_set_mask(dev, mask);
dma_set_coherent_mask(dev, mask);
with a call to the new helper dma_set_mask_and_coherent().
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Adding some trivial formatting suggestions made by Dan Carpenter to
the TODO.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove braces from a few single line if statements.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The rcv_pend_lock spinlock isn't used anymore. remove it.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
phy_lock is no longer used in any useful code, it's all been moved into
a phy_device. Remove the lock definition and init.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
After prolonged testing for a few days of normal use with new et131x
hardware, I've concluded that this was a hardware issue with the older
hardware I had. Removing this item from the TODO.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The work item has been added to the queue using INIT_WORK and scheduled
in interrupt handler. when module unloads that work item has not been
removed from the queue. remove and stop its further execution when the
module unloaded
Cc: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
should be err < 0 instead of if (err) which actually the read register
value can be a positive number
Acked-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Here's the big staging tree merge for 3.9-rc1
Lots of cleanups and updates for drivers all through the staging tree.
We are pretty much "code neutral" here, adding just about as many lines
as we removed.
All of these have been in linux-next for a while.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)
iEUEABECAAYFAlEmW7QACgkQMUfUDdst+ymlIACXT5mv8Y5A/KJa+QLTNNsofI8u
aACgq9hNZxJzX6VQMLXUV8+2SILOqYo=
=5wIj
-----END PGP SIGNATURE-----
Merge tag 'staging-3.9-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging
Pull staging tree update from Greg Kroah-Hartman:
"Here's the big staging tree merge for 3.9-rc1
Lots of cleanups and updates for drivers all through the staging tree.
We are pretty much "code neutral" here, adding just about as many
lines as we removed.
All of these have been in linux-next for a while."
* tag 'staging-3.9-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging: (804 commits)
staging: comedi: vmk80xx: wait for URBs to complete
staging: comedi: drivers: addi-data: hwdrv_apci3200.c: Add a missing semicolon
staging: et131x: Update TODO list
staging: et131x: Remove assignment of skb->dev
staging: wlan-ng: hfa384x.h: fix for error reported by smatch
staging/zache checkpatch ERROR: spaces prohibited around that
staging/ozwpan: Mark read only parameters and structs as const
staging/ozwpan: Remove empty and unused function oz_cdev_heartbeat
staging/ozwpan: Mark local functions as static (fix sparse warnings)
staging/ozwpan: Add missing header includes
staging/usbip: Mark local functions as static (fix sparse warnings)
staging/xgifb: Remove duplicated code in loops.
staging/xgifb: Consolidate return paths
staging/xgifb: Remove code without effect
staging/xgifb: Remove unnecessary casts
staging/xgifb: Consolidate if/else if with identical code branches
staging: vt6656: replaced custom TRUE definition with true
staging: vt6656: replaced custom FALSE definition with false
staging: vt6656: replace custom BOOL definition with bool
staging/rtl8187se: Mark functions as static to silence sparse
...
David Miller <davem@davemloft.net> had provided some review
comments for this driver. Update the TODO list with some action
points from his review.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The call eth_type_trans() sets skb->dev to netdev, so there's
no needto set it before the call.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
alloc failures already get standardized OOM
messages and a dump_stack.
For the affected mallocs around these OOM messages:
Converted kzallocs with multiplies to kcalloc.
Converted kmallocs with multiplies to kmalloc_array.
Converted a kmalloc/strlen/strncpy to kstrdup.
Moved a spin_lock below a removed OOM message and
removed a now unnecessary spin_unlock.
Neatened alignment and whitespace.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In preparation for moving et131x to drivers/net, fixup the block
comments to match the preferred networking style - no /* on separate
line, but */ are on a seperate line.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Checkpatch now highlights some camel case flag names. Fix these issues.
Also fix some remaining lines > 80chars issues for completeness.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
et131x_get_regs() calls et131x_mii_read(), passing the address of a u32
which is cast to a (u16 *). This works fine for little endian systems,
but not for big endian. Change so that the types are cast, not pointers
to the types.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some config regs get set whenever adjust_link() is called. Change this
so that these registers only get set when the link state changes to UP.
Also remove unnecessary second assignment of boot_coma.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The flags argument of the phy_{attach,connect,connect_direct} functions
is then used to assign a struct phy_device dev_flags with its value.
All callers but the tg3 driver pass the flag 0, which results in the
underlying PHY drivers in drivers/net/phy/ not being able to actually
use any of the flags they would set in dev_flags. This patch gets rid of
the flags argument, and passes phydev->dev_flags to the internal PHY
library call phy_attach_direct() such that drivers which actually modify
a phy device dev_flags get the value preserved for use by the underlying
phy driver.
Acked-by: Kosta Zertsekel <konszert@marvell.com>
Signed-off-by: Florian Fainelli <florian@openwrt.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use register name defines instead of magic numbers where a bit of
clarity would be useful.
Also a small typo fix and some register prefixes added, for further
clarity.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Several magic numbers were used to represent rxdma csr register bitmasks.
Replace them with descriptive defines.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Use strlcpy where possible to ensure the string is \0 terminated.
Use always sizeof(string) instead of 32, ETHTOOL_BUSINFO_LEN
and custom defines.
Use snprintf instead of sprint.
Remove unnecessary inits of ->fw_version
Remove unnecessary inits of drvinfo struct.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
Trivial Whitespace changes only. No checkpatch errors exist in et131x
after this change.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The use of a kmem_cache was noted as being unusual in the TODO. Replace
the kmem_cache with kmalloc/kfree so that the code is less suprising.
Also tidy up the mess that was the et131x_init_recv() out of memory
error path.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A previous change removed code that aligned memory returned from
dma_alloc_coherent() to a 4k boundary, which was not necessary.
Some comments regarding this alignment still exist, so remove them
as they no longer apply.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This change negates an 'if' statement, allowing a large block of code
to be un-indented, making the code more readable.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This padding was used to align buffers to a 4k boundary when returned
from dma_alloc_coherent(). As the buffers are already 4k aligned, and
the alignment no longer performed, the padding is not needed.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
By negating a 'status' variable check in et131x_isr(), we can remove
the indenting of a large block of code, increasing the readability.
This patch does exactly that.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CONFIG_HOTPLUG is going away as an option so __devexit_p is no longer
needed.
Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: devel@driverdev.osuosl.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CONFIG_HOTPLUG is going away as an option so __devinit is no longer
needed.
Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Acked-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CONFIG_HOTPLUG is going away as an option so __devexit is no
longer needed.
Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Acked-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
"The cpu return address and the DMA bus master address are both
guaranteed to be aligned to the smallest PAGE_SIZE order which
is greater than or equal to the requested size."
There are several places in the et131x.c code where these addresses are
aligned to a 4k boundary after a call to dma_alloc_coherent(),
needlessly.
Remove these alignment offset calculations, and the
et131x_align_allocated_memory() call which is only used for 4k
alignments.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In et131x_rx_dma_memory_alloc(), we loop over fbr_entries - the values of
fbr_align and fbr_chunksize calculated in the loop do not depend on the
loop counter and are the same for all loop iterations.
Take the calculations of these values out of the loop, as we only
need to calculate them once per loop, not once per loop iteration.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A ring_index value of 0 is being used to reference the free buffer ring
fbr[1] and ring_index=1 is used to reference fbr[0]. Avoid this
confusing referencing by swapping the indicies so that ring_index=0
refers to fbr[0] and vice versa.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The fbr_lookup struct variables buffer1 and buffer2 are only assigned
and never used, so remove them.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In the ethtool call et131x_get_regs(), some MII regs are missing.
Add them to the regs returned.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The PHY registers are now being controlled from the connected phydev,
so there shouldn't be any reason for the et131x code to perform any
extra setup. Removing the interrupt setup code, and register defines
that are now unused.
On testing, no changes in behaviour were experienced.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
During a previous refactoring excerise, two NULL pointer assingments
were moved to their respective pointer declarations. Dan Carpenter
correctly points out that these assignments are not needed.
Removing them.
Also two small whitespace changes - rejoining split lines, as they are
now less than 80 chars.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-Added some extra items to the TODO list.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In nic_rx_pkts() some large chunks of code are indented in 'if (len)'
sections. Refactor the code to remove these indents and re-join longer
split lines.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Several places in et131x.c code is duplicated for fbr[0] and fbr[1].
Remove the duplicate lines and use loops to run over both indicies.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
dma_addr_t can be of size 64 or 32bits, depending on the architecture.
This fixes these build warnings for ARCH=i386, and also tested on x86_64:
drivers/staging/et131x/et131x.c: In function ‘et131x_rx_dma_memory_alloc’:
drivers/staging/et131x/et131x.c:2356:11: warning: passing argument 2 of ‘et131x_align_allocated_memory’ from incompatible pointer type [enabled by default]
drivers/staging/et131x/et131x.c:2260:13: note: expected ‘u64 *’ but argument is of type ‘dma_addr_t *’
drivers/staging/et131x/et131x.c:2378:11: warning: passing argument 2 of ‘et131x_align_allocated_memory’ from incompatible pointer type [enabled by default]
drivers/staging/et131x/et131x.c:2260:13: note: expected ‘u64 *’ but argument is of type ‘dma_addr_t *’
drivers/staging/et131x/et131x.c: In function ‘free_send_packet’:
drivers/staging/et131x/et131x.c:3540:5: warning: left shift count >= width of type [enabled by default]
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
USE_FBR0 has always been defined, even in the original driver code.
Remove the define and #ifdef code to leave the code in the same state.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
'>> 32 of a 32bit value is undefined in C. The compiler is free
to do what it likes with this...'
Change all uses of '>> 32' to use upper_32_bits() and use
the corresponding lower_32_bits() to match.
Also remove an incorrect comment about dma alloc always returning 32bit
addresses.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
fbr_lookup.real_physaddr is only being used as a pseudonym for
fbr_lookup.ring_physaddr, so remove it and rename all instances to
ring_physaddr.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The driver checks that the device can handle 64bit DMA addressing in
et131x_pci_setup(), but then assumes that the top dword of a tx dma
address is always zero when creating a dma mapping in nic_send_packet().
Fix the mapping to use the higher dword of the dma_addr_t returned by
dma_map_single() and skb_frag_dma_map().
Also remove incorrect comments stating that dma_map_single() only returns
a 32 bit address.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(skb->len - skb->data_len) is used in several places in the et131x
driver code. Converted all instances of this to use skb_headlen()
which is more readable.
Signed-off-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Linus Torvalds