| Age | Commit message (Collapse) | Author |
|---|
|
commit da87ca4d4ca101f177fffd84f1f0a5e4c0343557 upstream.
Since commit 77873803363c "net_dma: mark broken" we no longer pin dma
engines active for the network-receive-offload use case. As a result
the ->free_chan_resources() that occurs after the driver self test no
longer has a NET_DMA induced ->alloc_chan_resources() to back it up. A
late firing irq can lead to ksoftirqd spinning indefinitely due to the
tasklet_disable() performed by ->free_chan_resources(). Only
->alloc_chan_resources() can clear this condition in affected kernels.
This problem has been present since commit 3e037454bcfa "I/OAT: Add
support for MSI and MSI-X" in 2.6.24, but is now exposed. Given the
NET_DMA use case is deprecated we can revisit moving the driver to use
threaded irqs. For now, just tear down the irq and tasklet properly by:
1/ Disable the irq from triggering the tasklet
2/ Disable the irq from re-arming
3/ Flush inflight interrupts
4/ Flush the timer
5/ Flush inflight tasklets
References:
https://lkml.org/lkml/2014/1/27/282
https://lkml.org/lkml/2014/2/19/672
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Steven Rostedt <rostedt@goodmis.org>
Reported-by: Mike Galbraith <bitbucket@online.de>
Reported-by: Stanislav Fomichev <stfomichev@yandex-team.ru>
Tested-by: Mike Galbraith <bitbucket@online.de>
Tested-by: Stanislav Fomichev <stfomichev@yandex-team.ru>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 21e96c7313486390c694919522a76dfea0a86c59 upstream.
When performing continuations there are implied sources that need to be
added to the source count. Quoting dma_set_maxpq:
/* dma_maxpq - reduce maxpq in the face of continued operations
* @dma - dma device with PQ capability
* @flags - to check if DMA_PREP_CONTINUE and DMA_PREP_PQ_DISABLE_P are set
*
* When an engine does not support native continuation we need 3 extra
* source slots to reuse P and Q with the following coefficients:
* 1/ {00} * P : remove P from Q', but use it as a source for P'
* 2/ {01} * Q : use Q to continue Q' calculation
* 3/ {00} * Q : subtract Q from P' to cancel (2)
*
* In the case where P is disabled we only need 1 extra source:
* 1/ {01} * Q : use Q to continue Q' calculation
*/
...fix the selection of the 16 source path to take these implied sources
into account.
Note this also kills the BUG_ON(src_cnt < 9) check in
__ioat3_prep_pq16_lock(). Besides not accounting for implied sources
the check is redundant given we already made the path selection.
Cc: Dave Jiang <dave.jiang@intel.com>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit 5d48b9b5d80e3aa38a5161565398b1e48a650573 upstream.
The array to lookup the sed pool based on the number of sources
(pq16_idx_to_sedi) is 16 entries and expects a max source index.
However, we pass the total source count which runs off the end of the
array when src_cnt == 16. The minimal fix is to just pass src_cnt-1,
but given we know the source count is > 8 we can just calculate the sed
pool by (src_cnt - 2) >> 3.
Cc: Dave Jiang <dave.jiang@intel.com>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
v3.3 provides support for write back descriptor error status. This allows
reporting of errors in a descriptor field. In supporting this, certain
errors such as P/Q validation errors no longer halts the channel. The DMA
engine can continue to execute until the end of the chain and allow software
to report the "errors" up the stack. We are also going to mask those error
interrupts and handle them when the "chain" has completed at the end.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked-by: Dan Williams <djbw@fb.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
This workaround checks for channel 2&3 and remove RAID cap.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked-by: Dan Williams <djbw@fb.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
v3.3 introduced 16 sources PQ operations. This also introduced super extended
descriptors to support the 16 srcs operations. This patch adds support for
the 16 sources ops and in turn adds the super extended descriptors for those
ops.
5 SED pools are created depending on the descriptor sizes. An SED can be a 64
bytes sized descriptor or larger and must be physically contiguous. A kmem
cache pool is created for allocating the software descriptor that manages the
hardware descriptor. The super extended descriptor will take place of extended
descriptor under certain operations and be "attached" to the op descriptor
during operation. This is a new feature for ioatdma v3.3.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked-by: Dan Williams <djbw@fb.com>
Acked-by: Dan Williams <djbw@fb.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
CB3.2 and earlier hardware has silicon bugs that are no longer needed with
the new hardware. We don't have to use a NULL op to signal interrupt for
RAID ops any longer. This code make sure the legacy workarounds only happen on
legacy hardware.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked-by: Dan Williams <djbw@fb.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
The alignment workaround is only necessary for cb3.2 or earlier platforms.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked-by: Dan Williams <djbw@fb.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
The PQ Val ops work on the newer hardware so we should actually provide support
for it and remove the disabling bits.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked-by: Dan Williams <djbw@fb.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
Make it so only 3.2 and earlier platform need the PCI config register
clearings since this implementation does not have the registers.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked-by: Dan Williams <djbw@fb.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
The Intel Atom S1200 family ioatdma changed the channel reset behavior.
It does a reset similar to PCI FLR by resetting all the MSIX
registers. We have to re-init msix interrupts because of this. This
workaround is only specific to this platform and is not expected to carry
over to the later generations.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked-by: Dan Williams <djbw@fb.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
Adding Haswell PCI device IDs for ioatdma and simplify the detection of
certain Xeon CPUs that has alignment bugs so that modifications can be
changed at a single place going forward.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked-by: Dan Williams <djbw@fb.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
Looks like only the RAID channels are allowed to have irq coalescing support
in the existing code. Fixing that. The ioat3 cleanup code can handle memcpy
ops anyways
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked-by: Dan Williams <djbw@fb.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
Making OP field a hex instead of integer to make it more readable. Also add
the dump out of the NEXT field.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked-by: Dan Williams <djbw@fb.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
Pull slave-dmaengine updates from Vinod Koul:
"This is fairly big pull by my standards as I had missed last merge
window. So we have the support for device tree for slave-dmaengine,
large updates to dw_dmac driver from Andy for reusing on different
architectures. Along with this we have fixes on bunch of the drivers"
Fix up trivial conflicts, usually due to #include line movement next to
each other.
* 'next' of git://git.infradead.org/users/vkoul/slave-dma: (111 commits)
Revert "ARM: SPEAr13xx: Pass DW DMAC platform data from DT"
ARM: dts: pl330: Add #dma-cells for generic dma binding support
DMA: PL330: Register the DMA controller with the generic DMA helpers
DMA: PL330: Add xlate function
DMA: PL330: Add new pl330 filter for DT case.
dma: tegra20-apb-dma: remove unnecessary assignment
edma: do not waste memory for dma_mask
dma: coh901318: set residue only if dma is in progress
dma: coh901318: avoid unbalanced locking
dmaengine.h: remove redundant else keyword
dma: of-dma: protect list write operation by spin_lock
dmaengine: ste_dma40: do not remove descriptors for cyclic transfers
dma: of-dma.c: fix memory leakage
dw_dmac: apply default dma_mask if needed
dmaengine: ioat - fix spare sparse complain
dmaengine: move drivers/of/dma.c -> drivers/dma/of-dma.c
ioatdma: fix race between updating ioat->head and IOAT_COMPLETION_PENDING
dw_dmac: add support for Lynxpoint DMA controllers
dw_dmac: return proper residue value
dw_dmac: fill individual length of descriptor
...
|
|
>> drivers/dma/ioat/dma_v3.c:371:6: sparse: symbol 'ioat3_timer_event' was not declared.
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
There is a race that can hit during __cleanup() when the ioat->head pointer is
incremented during descriptor submission. The __cleanup() can clear the
PENDING flag when it does not see any active descriptors. This causes new
submitted descriptors to be ignored because the COMPLETION_PENDING flag is
cleared. This was introduced when code was adapted from ioatdma v1 to ioatdma
v2. For v2 and v3, IOAT_COMPLETION_PENDING flag will be abandoned and a new
flag IOAT_CHAN_ACTIVE will be utilized. This flag will also be protected under
the prep_lock when being modified in order to avoid the race.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Dan Williams <djbw@fb.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
|
|
Pull slave-dmaengine fixes from Vinod Koul:
"A few fixes on slave dmanengine. There are trivial fixes in imx-dma,
tegra-dma & ioat driver"
* 'fixes' of git://git.infradead.org/users/vkoul/slave-dma:
dma: tegra: implement flags parameters for cyclic transfer
dmaengine: imx-dma: Disable use of hw_chain to fix sg_dma transfers.
ioat: Fix DMA memory sync direction correct flag
|
|
The existing code set a value in the PCI_CHANERRMSK_INT register
for a workaround to address a pre-silicon bug on the Intel 5520 IO hub that
has been fixed when the hardware was released. There is no need for this
code.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Dan Williams <djbw@fb.com>
|
|
The PCI IDs for IvyBridge IOAT DMA needs to go into a header file since
dma_v3.c looks them up for certain hardware workarounds. Need to add to the
alignment workaround for IOAT 3.2 since it wasn't fixed in IVB.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Dan Williams <djbw@fb.com>
|
|
Make ioat_xor_val_self_test() do DMA unmapping itself and fix handling
of failure cases.
Cc: Dan Williams <djbw@fb.com>
Cc: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Dan Williams <djbw@fb.com>
|
|
ioat does DMA memory sync with DMA_TO_DEVICE direction on a buffer allocated
for DMA_FROM_DEVICE dma, resulting in the following warning from dma debug.
Fixed the dma_sync_single_for_device() call to use the correct direction.
[ 226.288947] WARNING: at lib/dma-debug.c:990 check_sync+0x132/0x550()
[ 226.288948] Hardware name: ProLiant DL380p Gen8
[ 226.288951] ioatdma 0000:00:04.0: DMA-API: device driver syncs DMA memory with different direction [device address=0x00000000ffff7000] [size=4096 bytes] [mapped with DMA_FROM_DEVICE] [synced with DMA_TO_DEVICE]
[ 226.288953] Modules linked in: iTCO_wdt(+) sb_edac(+) ioatdma(+) microcode serio_raw pcspkr edac_core hpwdt(+) iTCO_vendor_support hpilo(+) dca acpi_power_meter ata_generic pata_acpi sd_mod crc_t10dif ata_piix libata hpsa tg3 netxen_nic(+) sunrpc dm_mirror dm_region_hash dm_log dm_mod
[ 226.288967] Pid: 1055, comm: work_for_cpu Tainted: G W 3.3.0-0.20.el7.x86_64 #1
[ 226.288968] Call Trace:
[ 226.288974] [<ffffffff810644cf>] warn_slowpath_common+0x7f/0xc0
[ 226.288977] [<ffffffff810645c6>] warn_slowpath_fmt+0x46/0x50
[ 226.288980] [<ffffffff81345502>] check_sync+0x132/0x550
[ 226.288983] [<ffffffff81345c9f>] debug_dma_sync_single_for_device+0x3f/0x50
[ 226.288988] [<ffffffff81661002>] ? wait_for_common+0x72/0x180
[ 226.288995] [<ffffffffa019590f>] ioat_xor_val_self_test+0x3e5/0x832 [ioatdma]
[ 226.288999] [<ffffffff811a5739>] ? kfree+0x259/0x270
[ 226.289004] [<ffffffffa0195d77>] ioat3_dma_self_test+0x1b/0x20 [ioatdma]
[ 226.289008] [<ffffffffa01952c3>] ioat_probe+0x2f8/0x348 [ioatdma]
[ 226.289011] [<ffffffffa0195f51>] ioat3_dma_probe+0x1d5/0x2aa [ioatdma]
[ 226.289016] [<ffffffffa0194d12>] ioat_pci_probe+0x139/0x17c [ioatdma]
[ 226.289020] [<ffffffff81354b8c>] local_pci_probe+0x5c/0xd0
[ 226.289023] [<ffffffff81083e50>] ? destroy_work_on_stack+0x20/0x20
[ 226.289025] [<ffffffff81083e68>] do_work_for_cpu+0x18/0x30
[ 226.289029] [<ffffffff8108d997>] kthread+0xb7/0xc0
[ 226.289033] [<ffffffff8166cef4>] kernel_thread_helper+0x4/0x10
[ 226.289036] [<ffffffff81662d20>] ? _raw_spin_unlock_irq+0x30/0x50
[ 226.289038] [<ffffffff81663234>] ? retint_restore_args+0x13/0x13
[ 226.289041] [<ffffffff8108d8e0>] ? kthread_worker_fn+0x1a0/0x1a0
[ 226.289044] [<ffffffff8166cef0>] ? gs_change+0x13/0x13
[ 226.289045] ---[ end trace e1618afc7a606089 ]---
[ 226.289047] Mapped at:
[ 226.289048] [<ffffffff81345307>] debug_dma_map_page+0x87/0x150
[ 226.289050] [<ffffffffa019653c>] dma_map_page.constprop.18+0x70/0xb34 [ioatdma]
[ 226.289054] [<ffffffffa0195702>] ioat_xor_val_self_test+0x1d8/0x832 [ioatdma]
[ 226.289058] [<ffffffffa0195d77>] ioat3_dma_self_test+0x1b/0x20 [ioatdma]
[ 226.289061] [<ffffffffa01952c3>] ioat_probe+0x2f8/0x348 [ioatdma]
Signed-off-by: Shuah Khan <shuah.khan@hp.com>
CC: <stable@vger.kernel.org>
Signed-off-by: Vinod Koul <vinod.koul@linux.intel.com>
|
|
CONFIG_HOTPLUG is going away as an option. As a result, the __dev*
markings need to be removed.
This change removes the use of __devinit, __devexit_p, __devinitconst,
and __devexit from these drivers.
Based on patches originally written by Bill Pemberton, but redone by me
in order to handle some of the coding style issues better, by hand.
Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Viresh Kumar <viresh.linux@gmail.com>
Cc: Dan Williams <djbw@fb.com>
Cc: Vinod Koul <vinod.koul@intel.com>
Cc: Barry Song <baohua.song@csr.com>
Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Cc: Alexander Duyck <alexander.h.duyck@intel.com>
Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: Jassi Brar <jassisinghbrar@gmail.com>
Cc: Dave Jiang <dave.jiang@intel.com>
Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/djbw/dmaengine
Pull dmaengine fixes from Dan Williams:
1/ regression fix for Xen as it now trips over a broken assumption
about the dma address size on 32-bit builds
2/ new quirk for netdma to ignore dma channels that cannot meet
netdma alignment requirements
3/ fixes for two long standing issues in ioatdma (ring size overflow)
and iop-adma (potential stack corruption)
* tag 'dmaengine-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/dmaengine:
netdma: adding alignment check for NETDMA ops
ioatdma: DMA copy alignment needed to address IOAT DMA silicon errata
ioat: ring size variables need to be 32bit to avoid overflow
iop-adma: Corrected array overflow in RAID6 Xscale(R) test.
ioat: fix size of 'completion' for Xen
|
|
Silicon errata where when RAID and legacy descriptors are mixed, the legacy
(memcpy and friends) operation must have alignment of 64 bytes to avoid
hanging. This effects Intel Xeon C55xx, C35xx, E5-2600.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
Starting with v3.2 Jonathan reports that Xen crashes loading the ioatdma
driver. A debug run shows:
ioatdma 0000:00:16.4: desc[0]: (0x300cc7000->0x300cc7040) cookie: 0 flags: 0x2 ctl: 0x29 (op: 0 int_en: 1 compl: 1)
...
ioatdma 0000:00:16.4: ioat_get_current_completion: phys_complete: 0xcc7000
...which shows that in this environment GFP_KERNEL memory may be backed
by a 64-bit dma address. This breaks the driver's assumption that an
unsigned long should be able to contain the physical address for
descriptor memory. Switch to dma_addr_t which beyond being the right
size, is the true type for the data i.e. an io-virtual address
inidicating the engine's last processed descriptor.
[stable: 3.2+]
Cc: <stable@vger.kernel.org>
Reported-by: Jonathan Nieder <jrnieder@gmail.com>
Reported-by: William Dauchy <wdauchy@gmail.com>
Tested-by: William Dauchy <wdauchy@gmail.com>
Tested-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
Fixed trivial issues in drivers:
drivers/dma/imx-sdma.c
drivers/dma/intel_mid_dma.c
drivers/dma/ioat/dma_v3.c
drivers/dma/iop-adma.c
drivers/dma/sirf-dma.c
drivers/dma/timb_dma.c
Signed-off-by: Vinod Koul <vinod.koul@linux.intel.com>
|
|
Now that we have the completed cookie in the dma_chan structure, we
can consolidate the tx_status functions by providing a function to set
the txstate structure and returning the DMA status. We also provide
a separate helper to set the residue for cookies which are still in
progress.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Tested-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Acked-by: Jassi Brar <jassisinghbrar@gmail.com>
[imx-sdma.c & mxs-dma.c]
Tested-by: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Vinod Koul <vinod.koul@linux.intel.com>
|
|
Provide a common function to do the cookie mechanics for completing
a DMA descriptor.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Tested-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Acked-by: Jassi Brar <jassisinghbrar@gmail.com>
[imx-sdma.c & mxs-dma.c]
Tested-by: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Vinod Koul <vinod.koul@linux.intel.com>
|
|
Every DMA engine implementation declares a last completed dma cookie
in their private dma channel structures. This is pointless, and
forces driver specific code. Move this out into the common dma_chan
structure.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Tested-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Acked-by: Jassi Brar <jassisinghbrar@gmail.com>
[imx-sdma.c & mxs-dma.c]
Tested-by: Shawn Guo <shawn.guo@linaro.org>
Signed-off-by: Vinod Koul <vinod.koul@linux.intel.com>
|
|
For versions of the device that implement operation-types 0x87, 0x88
(IOAT_OP_XOR, IOAT_OP_XOR_VAL) this map determines whether a given
source is located in the base or extended descriptor. Source addresses
6 through 8 require an extended descriptor, hence 0xe0, not 0xd0. No
shipping hardware currently implements these operation types.
Reported-by: Evgueni Smogailov <evgueni.smogailov@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
const __read_mostly is not legal and causes section type conflicts.
That's because the read.mostly section is not read only.
Simply drop the __read_mostly designation.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
[drop __read_mostly instead of const]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
After discovering that wide use of prefetch on modern CPUs
could be a net loss instead of a win, net drivers which were
relying on the implicit inclusion of prefetch.h via the list
headers showed up in the resulting cleanup fallout. Give
them an explicit include via the following $0.02 script.
=========================================
#!/bin/bash
MANUAL=""
for i in `git grep -l 'prefetch(.*)' .` ; do
grep -q '<linux/prefetch.h>' $i
if [ $? = 0 ] ; then
continue
fi
( echo '?^#include <linux/?a'
echo '#include <linux/prefetch.h>'
echo .
echo w
echo q
) | ed -s $i > /dev/null 2>&1
if [ $? != 0 ]; then
echo $i needs manual fixup
MANUAL="$i $MANUAL"
fi
done
echo ------------------- 8\<----------------------
echo vi $MANUAL
=========================================
Signed-off-by: Paul <paul.gortmaker@windriver.com>
[ Fixed up some incorrect #include placements, and added some
non-network drivers and the fib_trie.c case - Linus ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
On some platforms (MacPro3,1) the BIOS assigns the ioatdma device to the
incorrect iommu causing faults when the driver initializes. Add a quirk
to catch this misconfiguration and try falling back to untranslated
operation (which works in the MacPro3,1 case).
Assuming there are other platforms with misconfigured iommus teach the
ioatdma driver to treat initialization failures as non-fatal (just fail
the driver load and emit a warning instead of triggering a BUG_ON).
This can be classified as a boot regression since 2.6.32 on affected
platforms since the ioatdma module did not autoload prior to that
kernel.
Cc: <stable@kernel.org>
Acked-by: David Woodhouse <David.Woodhouse@intel.com>
Reported-by: Chris Li <lkml@chrisli.org>
Tested-by: Chris Li <lkml@chrisli.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
|
|
There are cases where cacheline-unaligned raid operations can hang the
dma channel. Simply disable these operations by increasing the
alignment constraints published to async_tx. The raid456 driver always
issues page aligned requests, so the only in-kernel user of the ioatdma
driver that is affected by this change is dmatest.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
Use separate locks for the descriptor prep (producer) and descriptor
cleanup (consumer) paths. Allows the producer path to run concurrently
with the cleanup path. Inspired by Documentation/circular-buffer.txt.
Cc: David Howells <dhowells@redhat.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Maciej Sosnowski <maciej.sosnowski@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
|
|
Convert the device_is_tx_complete() operation on the
DMA engine to a generic device_tx_status()operation which
can return three states, DMA_TX_RUNNING, DMA_TX_COMPLETE,
DMA_TX_PAUSED.
[dan.j.williams@intel.com: update for timberdale]
Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: Maciej Sosnowski <maciej.sosnowski@intel.com>
Cc: Nicolas Ferre <nicolas.ferre@atmel.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Li Yang <leoli@freescale.com>
Cc: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Cc: Magnus Damm <damm@opensource.se>
Cc: Liam Girdwood <lrg@slimlogic.co.uk>
Cc: Joe Perches <joe@perches.com>
Cc: Roland Dreier <rdreier@cisco.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
If the calling convention of ->timer_fn() and ->cleanup_fn() are unified
across hardware versions we can drop parameters to ioat_init_channel() and
unify ioat_is_dma_complete() implementations.
Both ->timer_fn() and ->cleanup_fn() are modified to expect a struct
dma_chan pointer.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
The hardware automatically disables further interrupts after each event
until rearmed. This allows a delay to be injected between the occurence
of the interrupt and the running of the cleanup routine. The delay is
scaled by the descriptor backlog and then written to the INTRDELAY
register which specifies the number of microseconds to hold off
interrupt delivery after an interrupt event occurs. According to
powertop this reduces the interrupt rate from ~5000 intr/s to ~150
intr/s per without affecting throughput (simple dd to a raid6 array).
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
Since ioat_cleanup_preamble() and the update of the last completed
descriptor are not synchronized there is a chance that two cleanup threads
can see descriptors to clean. If the first cleans up all pending
descriptors then the second will trigger the BUG_ON.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
Replace open coded ioat2_quiesce() call in ioat3_restart_channel
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
We already disallow raid operations while DCA is globally enabled, so
having it locally enabled is a nop and confusing when reading the code.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
Put the ioat2 and ioat3 state machines in the halted state with all
errors cleared.
The ioat1 init path is not disturbed for stability, there are no
reported ioat1 initiaization issues.
Cc: <stable@kernel.org>
Reported-by: Roland Dreier <rdreier@cisco.com>
Tested-by: Roland Dreier <rdreier@cisco.com>
Acked-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
When continuing a pq calculation the driver needs 3 extra sources. The
driver can perform a 3 source calculation with a single descriptor, but
needs an extended descriptor to process up to 8 sources in one
operation. However, in the p-disabled case only one extra source is
needed. When continuing a p-disabled operation there are occasions
(i.e. 0 < src_cnt % 8 < 3) where the tail operation does not need an
extended descriptor. Properly account for this fact otherwise invalid
'dmacount' values will be written to hardware usually causing the
channel to halt with 'invalid descriptor' errors.
Cc: <stable@kernel.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
The completion of a pq operation is notified with a null descriptor
appended to the end of the chain. This descriptor needs to be visible
to dma clients otherwise the client is precluded from ensuring all
operations are quiesced before freeing channel resources, i.e. due to
descriptor polling it may get the completion notification ahead of the
interrupt delivered by the null descriptor.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
ioat3.2 does not support asynchronous error notifications which makes
the driver experience latencies when non-zero pq validate results are
expected. Provide a mechanism for turning off async_xor_val and
async_syndrome_val via Kconfig. This approach is generally useful for
any driver that specifies ASYNC_TX_DISABLE_CHANNEL_SWITCH and would like
to force the async_tx api to fall back to the synchronous path for
certain operations.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
|
Modify is_ioat_bug() to catch all errors that are uncorrectable, or not
currently handled.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
|
