The driver Kconfig symbol is tristate which means that it can be built as
a module but the module alias information is not added to the module info
so module autoload won't work since user-space won't have the information.
Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A driver's SPI id table is expected to be an array of struct spi_device_id
that ends with a zero-initialized sentinel entry. But this driver defines
the table as a single struct spi_device_id and sets .id_table to a pointer
to this struct.
But spi_match_id() has a loop that iterates while the struct spi_device_id
.name[0] is not NULL, so not having a sentinel can cause a NULL pointer
deference error.
This patch defines the SPI id table correctly as all other SPI drivers do.
Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
* L3 and SoC support for xgene_edac. (Loc Ho)
* AMD F15h, models 0x60-6f support to amd64_edac. (Aravind Gopalakrishnan)
* Fixes and cleanups all over the place.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=Ek5l
-----END PGP SIGNATURE-----
Merge tag 'edac_for_4.4' of git://git.kernel.org/pub/scm/linux/kernel/git/bp/bp
Pull EDAC updates from Borislav Petkov:
"A bunch of fixes all over the place and some hw enablement this time.
- Convert EDAC to debugfs wrappers and make drivers use those
(Borislav Petkov)
- L3 and SoC support for xgene_edac (Loc Ho)
- AMD F15h, models 0x60-6f support to amd64_edac (Aravind
Gopalakrishnan)
- Fixes and cleanups all over the place"
* tag 'edac_for_4.4' of git://git.kernel.org/pub/scm/linux/kernel/git/bp/bp: (22 commits)
EDAC: Fix PAGES_TO_MiB macro misuse
EDAC, altera: SoCFPGA EDAC should not look for ECC_CORR_EN
EDAC: Use edac_debugfs_remove_recursive()
EDAC, ppc4xx_edac: Fix module autoload for OF platform driver
Documentation/EDAC: Add reference documents section for amd64_edac
EDAC, amd64_edac: Update copyright and remove changelog
EDAC, amd64_edac: Extend scrub rate support to F15hM60h
EDAC: Don't allow empty DIMM labels
EDAC: Fix sysfs dimm_label store operation
EDAC: Fix sysfs dimm_label show operation
arm64, EDAC: Add L3/SoC DT subnodes to the APM X-Gene SoC EDAC node
EDAC, xgene: Add SoC support
EDAC, xgene: Fix possible sprintf() overflow issue
EDAC, xgene: Add L3 support
EDAC, Documentation: Update X-Gene EDAC binding for L3/SoC subnodes
EDAC, sb_edac: Fix TAD presence check for sbridge_mci_bind_devs()
EDAC, ghes_edac: Remove redundant memory_type array
EDAC, xgene: Convert to debugfs wrappers
EDAC, i5100: Convert to debugfs wrappers
EDAC, altera: Convert to debugfs wrappers
...
The affinity hint is used by the user space daemon, irqbalancer, to
indicate a preferred CPU mask for irqs. This patch sets the irq affinity
hint to local numa core first, when exausted we try non-local numa cores.
Also set tx xps cpus mask bassed on affinity hint.
v2: remove the global affinity policy.
Signed-off-by: Govindarajulu Varadarajan <_govind@gmx.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- Add new API to set VCCQ voltage - mmc_regulator_set_vqmmc()
- Add new ioctl to allow userspace to send multi commands
- Wait for card busy signalling before starting SDIO requests
- Remove MMC_CLKGATE
- Enable tuning for DDR50 mode
- Some code clean-up/improvements to mmc pwrseq
- Use highest priority for eMMC restart handler
- Add DT bindings for eMMC hardware reset support
- Extend the mmc_send_tuning() API
- Improve ios show for debugfs
- A couple of code optimizations
MMC host:
- Some generic OF improvements
- Various code clean-ups
- sirf: Add support for DDR50
- sunxi: Add support for card busy detection
- mediatek: Use MMC_CAP_RUNTIME_RESUME
- mediatek: Add support for eMMC HW-reset
- mediatek: Add support for HS400
- dw_mmc: Convert to use the new mmc_regulator_set_vqmmc() API
- dw_mmc: Add external DMA interface support
- dw_mmc: Some various improvements
- dw_mmc-rockchip: MMC tuning with the clock phase framework
- sdhci: Properly clear IRQs during resume
- sdhci: Enable tuning for DDR50 mode
- sdhci-of-esdhc: Use IRQ mode for card detection
- sdhci-of-esdhc: Support both BE and LE host controller
- sdhci-pci: Build o2micro support in the same module
- sdhci-pci: Support for new Intel host controllers
- sdhci-acpi: Support for new Intel host controllers
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=gn/m
-----END PGP SIGNATURE-----
Merge tag 'mmc-v4.4' of git://git.linaro.org/people/ulf.hansson/mmc
Pull MMC updates from Ulf Hansson:
"MMC core:
- Add new API to set VCCQ voltage - mmc_regulator_set_vqmmc()
- Add new ioctl to allow userspace to send multi commands
- Wait for card busy signalling before starting SDIO requests
- Remove MMC_CLKGATE
- Enable tuning for DDR50 mode
- Some code clean-up/improvements to mmc pwrseq
- Use highest priority for eMMC restart handler
- Add DT bindings for eMMC hardware reset support
- Extend the mmc_send_tuning() API
- Improve ios show for debugfs
- A couple of code optimizations
MMC host:
- Some generic OF improvements
- Various code clean-ups
- sirf: Add support for DDR50
- sunxi: Add support for card busy detection
- mediatek: Use MMC_CAP_RUNTIME_RESUME
- mediatek: Add support for eMMC HW-reset
- mediatek: Add support for HS400
- dw_mmc: Convert to use the new mmc_regulator_set_vqmmc() API
- dw_mmc: Add external DMA interface support
- dw_mmc: Some various improvements
- dw_mmc-rockchip: MMC tuning with the clock phase framework
- sdhci: Properly clear IRQs during resume
- sdhci: Enable tuning for DDR50 mode
- sdhci-of-esdhc: Use IRQ mode for card detection
- sdhci-of-esdhc: Support both BE and LE host controller
- sdhci-pci: Build o2micro support in the same module
- sdhci-pci: Support for new Intel host controllers
- sdhci-acpi: Support for new Intel host controllers"
* tag 'mmc-v4.4' of git://git.linaro.org/people/ulf.hansson/mmc: (73 commits)
mmc: dw_mmc: fix the wrong setting for UHS-DDR50 mode
mmc: dw_mmc: fix the CardThreshold boundary at CardThrCtl register
mmc: dw_mmc: NULL dereference in error message
mmc: pwrseq: Use highest priority for eMMC restart handler
mmc: mediatek: add HS400 support
mmc: mmc: extend the mmc_send_tuning()
mmc: mediatek: add implement of ops->hw_reset()
mmc: mediatek: fix got GPD checksum error interrupt when data transfer
mmc: mediatek: change the argument "ddr" to "timing"
mmc: mediatek: make cmd_ints_mask to const
mmc: dt-bindings: update Mediatek MMC bindings
mmc: core: Add DT bindings for eMMC hardware reset support
mmc: omap_hsmmc: Enable omap_hsmmc for Keystone 2
mmc: sdhci-acpi: Add more ACPI HIDs for Intel controllers
mmc: sdhci-pci: Add more PCI IDs for Intel controllers
arm: lpc18xx_defconfig: remove CONFIG_MMC_DW_IDMAC
arm: hisi_defconfig: remove CONFIG_MMC_DW_IDMAC
arm: exynos_defconfig: remove CONFIG_MMC_DW_IDMAC
arc: axs10x_defconfig: remove CONFIG_MMC_DW_IDMAC
mips: pistachio_defconfig: remove CONFIG_MMC_DW_IDMAC
...
This patch changes how the multipath hash is computed for locally
generated flows: now the hash comprises l4 information.
This allows better utilization of the available paths when the existing
flows have the same source IP and the same destination IP: with l3 hash,
even when multiple connections are in place simultaneously, a single path
will be used, while with l4 hash we can use all the available paths.
v2 changes:
- use get_hash_from_flowi4() instead of implementing just another l4 hash
function
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
New driver for MAX31790, added support for TMP75C, as well as cleanups
and minor improvements in various drivers.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=Fa0c
-----END PGP SIGNATURE-----
Merge tag 'hwmon-for-linus-v4.4' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging
Pull hwmon updates from Guenter Roeck:
"New driver for MAX31790, added support for TMP75C, as well as cleanups
and minor improvements in various drivers"
* tag 'hwmon-for-linus-v4.4' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
hwmon: (fam15h_power) Add max compute unit accumulated power
hwmon: (fam15h_power) Enable power1_input on AMD Carrizo
hwmon: (fam15h_power) Refactor attributes for dynamically added
hwmon: (ina2xx) remove no longer used variable 'kind'
hwmon: (nct6775) Introduce separate temperature labels for NCT6792 and NCT6793
hwmon: (nct6775) NCT6791D and NCT6792D have an additional temperature source
hwmon: (ina2xx) give precedence to DT over checking for platform data.
hwmon: (ina2xx) convert driver to using regmap
hwmon: (coretemp) Increase limit of maximum core ID from 32 to 128.
hwmon: (lm75) Add support for TMP75C
hwmon: (ibmpowernv) Add OF compatibility table entry
hwmon: (abx500) drop the use of IRQF_NO_SUSPEND
hwmon: (max31790) Fix dereference of ERR_PTR
hwmon: Driver for Maxim MAX31790
Make ftrace_event_is_function() return bool to improve readability
due to this particular function only using either one or zero as its
return value.
No functional change.
Link: http://lkml.kernel.org/r/1443537816-5788-9-git-send-email-bywxiaobai@163.com
Signed-off-by: Yaowei Bai <bywxiaobai@163.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Make is_legal_op() return bool to improve readability due to this particular
function only using either one or zero as its return value.
No functional change.
Link: http://lkml.kernel.org/r/1443537816-5788-8-git-send-email-bywxiaobai@163.com
Signed-off-by: Yaowei Bai <bywxiaobai@163.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Make rb_event_is_commit() return bool to improve readability
due to this particular function only using either one or zero as its
return value.
No functional change.
Link: http://lkml.kernel.org/r/1443537816-5788-7-git-send-email-bywxiaobai@163.com
Signed-off-by: Yaowei Bai <bywxiaobai@163.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Make rb_is_reader_page() return bool to improve readability due to this
particular function only using either true or false as its return value.
No functional change.
Link: http://lkml.kernel.org/r/1443537816-5788-4-git-send-email-bywxiaobai@163.com
Signed-off-by: Yaowei Bai <bywxiaobai@163.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
This patch makes report_latency return bool due to this
particular function only using either one or zero as its
return value.
No functional change.
Link: http://lkml.kernel.org/r/1443537816-5788-3-git-send-email-bywxiaobai@163.com
Signed-off-by: Yaowei Bai <bywxiaobai@163.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
This patch makes report_latency return bool to improve readability,
indicating whether this new latency should be reported/recorded.
No functional change.
Link: http://lkml.kernel.org/r/1443537816-5788-2-git-send-email-bywxiaobai@163.com
Signed-off-by: Yaowei Bai <bywxiaobai@163.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The commit 889204278c ("tracing: Update trace-event-sample with
TRACE_SYSTEM_VAR documentation") changed TRACE_SYSTEM to 'sample-trace',
but didn't make the according change of its name in the comments.
Link: http://lkml.kernel.org/r/1443599650-23680-1-git-send-email-zhang.chunyan@linaro.org
Signed-off-by: Chunyan Zhang <zhang.chunyan@linaro.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Function get_cmos_time() was removed with commit 657926a83df9 ("cris:
time: Cleanup of persistent clock stuff"). The remaining reference to
it may cause the following build error.
arch/cris/kernel/built-in.o:(___ksymtab+get_cmos_time+0x0):
undefined reference to `get_cmos_time'
Makefile:946: recipe for target 'vmlinux' failed
Fixes: 657926a83df9 ("cris: time: Cleanup of persistent clock stuff")
Cc: Xunlei Pang <pang.xunlei@linaro.org>
Cc: Rabin Vincent <rabin@rab.in>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Jesper Nilsson <jespern@axis.com>
Drop all code related to Kconfigs that don't exist.
Fix one Kconfig where it was actually typo:ed (ETRAX_KGB_PORT2)
Drop content related to CRIS v32 SoCs from etraxgpio.h headerfile,
all use of GPIO for both ETRAX FS and ARTPEC-3 should now be through
standard gpiolib instead.
Signed-off-by: Jesper Nilsson <jesper.nilsson@axis.com>
- Remove update_persistent_clock(), as it does nothing now.
- Remove read_persistent_clock(), let it fall back to the weak version.
Cc: Mikael Starvik <starvik@axis.com>
Cc: John Stultz <john.stultz@linaro.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Xunlei Pang <pang.xunlei@linaro.org>
Acked-by: Rabin Vincent <rabin@rab.in>
Signed-off-by: Jesper Nilsson <jespern@axis.com>
There are native helpers such as print_hex_byte() and %*ph specifier to dump
data in hex format. Re-use them instead of a custom approach.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Jesper Nilsson <jespern@axis.com>
Since we now have a gpiolib driver, remove this code:
The gpio-etraxfs driver (along with things like gpio-keys-polled for
polling support) replaces the GIO driver implementations in mach-a3 and
mach-fs. The various generic external chip drivers replace the "virtual
gpio" parts.
The generic gpio-leds driver replaces the LED handling.
Signed-off-by: Rabin Vincent <rabin@rab.in>
Signed-off-by: Jesper Nilsson <jespern@axis.com>
Now that we have a gpiolib GPIO driver, the generic i2c-gpio driver
provides this functionality.
Signed-off-by: Rabin Vincent <rabin@rab.in>
Signed-off-by: Jesper Nilsson <jespern@axis.com>
Add a device tree for the Axis P1343 with the ARTPEC-3 SoC and on-board
LEDs and RTC.
Signed-off-by: Rabin Vincent <rabin@rab.in>
Signed-off-by: Jesper Nilsson <jespern@axis.com>
Add the GPIO driver to the device tree and, using it, support for the
LEDs and the RTC chip (via I2C-GPIO), as well as the temperature sensor
(via SPI-GPIO).
Signed-off-by: Rabin Vincent <rabin@rab.in>
Signed-off-by: Jesper Nilsson <jespern@axis.com>
Add a dt-bindings symlink to get DT include files, as on other
architectures. See c58299a ("kbuild: create an "include chroot" for DT
bindings") for the details.
Signed-off-by: Rabin Vincent <rabin@rab.in>
Signed-off-by: Jesper Nilsson <jespern@axis.com>
Within one C file, current gcc can optimize the global static variables
according to the C code, but it will skip assembly code -- it will pass
them to gas directly.
if the static variable is used between C code and assembly code in one C
file (e.g. is_dyn_brkp in kgdb.c), it needs '__used' to let gcc know it
should be still used, or gcc may remove it for optimization.
The related error in this case:
LD init/built-in.o
arch/cris/arch-v10/kernel/built-in.o: In function `kgdb_handle_breakpoint':
(.text+0x2aca): undefined reference to `is_dyn_brkp'
arch/cris/arch-v10/kernel/built-in.o: In function `is_static':
kgdb.c:(.text+0x2ada): undefined reference to `is_dyn_brkp'
Signed-off-by: Chen Gang <gang.chen.5i5j@gmail.com>
Signed-off-by: Jesper Nilsson <jespern@axis.com>
For arch-v10, there is no DTP0 register, and at present, assembler know
BAR, so use BAR instead of DTP0, the related error (with allmodconfig):
CC arch/cris/arch-v10/kernel/kgdb.o
{standard input}: Assembler messages:
{standard input}:6: Error: Illegal operands
{standard input}:6: Error: Illegal operands
Signed-off-by: Chen Gang <gang.chen.5i5j@gmail.com>
Acked-by: Hans-Peter Nilsson <hp@axis.com>
Signed-off-by: Jesper Nilsson <jespern@axis.com>
There are kernel native helpers to convert hex ascii to the binary format:
hex_to_bin() and hex2bin(). Thus, no need to reimplement them customly.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Jesper Nilsson <jespern@axis.com>
Update instancd_rmdir to use tracefs_remove_recursive instead of
debugfs_remove_recursive.This was left in the transition from debugfs
to tracefs.
Link: http://lkml.kernel.org/r/1445169490-18315-2-git-send-email-hello.wjx@gmail.com
Cc: stable@vger.kernel.org # 4.1+
Fixes: 8434dc9340 ("tracing: Convert the tracing facility over to use tracefs")
Signed-off-by: Jiaxing Wang <hello.wjx@gmail.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
For loosely coupled pNFS/flexfiles systems, there is often no advantage
at all in going through the MDS for I/O, since the MDS is subject to
the same limitations as all other clients when talking to DSes. If a
DS is unresponsive, I/O through the MDS will fail.
For such systems, the only scalable solution is to have the pNFS clients
retry doing pNFS, and so the protocol now provides a flag that allows
the pNFS server to signal this.
If LAYOUTGET returns FF_FLAGS_NO_IO_THRU_MDS, then we should assume that
the MDS wants the client to retry using these devices, even if they were
previously marked as being unavailable. To do so, we add a helper,
ff_layout_mark_devices_valid() that will be called from layoutget.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
If the pNFS/flexfiles file is mirrored, and a read to one mirror fails,
then we should bump the mirror index, so that we retry to a different
mirror. Once we've iterated through all mirrors and all failed, we can
return the layout and issue a new LAYOUTGET.
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Allow the use of other transport classes when handling a backward
direction RPC call.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
On NFSv4.1 mount points, the Linux NFS client uses this transport
endpoint to receive backward direction calls and route replies back
to the NFSv4.1 server.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Acked-by: "J. Bruce Fields" <bfields@fieldses.org>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Introduce a code path in the rpcrdma_reply_handler() to catch
incoming backward direction RPC calls and route them to the ULP's
backchannel server.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Backward direction RPC replies are sent via the client transport's
send_request method, the same way forward direction RPC calls are
sent.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Pre-allocate extra send and receive Work Requests needed to handle
backchannel receives and sends.
The transport doesn't know how many extra WRs to pre-allocate until
the xprt_setup_backchannel() call, but that's long after the WRs are
allocated during forechannel setup.
So, use a fixed value for now.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
xprtrdma's backward direction send and receive buffers are the same
size as the forechannel's inline threshold, and must be pre-
registered.
The consumer has no control over which receive buffer the adapter
chooses to catch an incoming backwards-direction call. Any receive
buffer can be used for either a forward reply or a backward call.
Thus both types of RPC message must all be the same size.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
xprt_{setup,destroy}_backchannel() won't be adequate for RPC/RMDA
bi-direction. In particular, receive buffers have to be pre-
registered and posted in order to receive incoming backchannel
requests.
Add a virtual function call to allow the insertion of appropriate
backchannel setup and destruction methods for each transport.
In addition, freeing a backchannel request is a little different
for RPC/RDMA. Introduce an rpc_xprt_op to handle the difference.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Now that RPC replies are processed in a workqueue, there's no need
to disable IRQs when managing send and receive buffers. This saves
noticeable overhead per RPC.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: The reply tasklet is no longer used.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
The reply tasklet is fast, but it's single threaded. After reply
traffic saturates a single CPU, there's no more reply processing
capacity.
Replace the tasklet with a workqueue to spread reply handling across
all CPUs. This also moves RPC/RDMA reply handling out of the soft
IRQ context and into a context that allows sleeps.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
The rb_send_bufs and rb_recv_bufs arrays are used to implement a
pair of stacks for keeping track of free rpcrdma_req and rpcrdma_rep
structs. Replace those arrays with free lists.
To allow more than 512 RPCs in-flight at once, each of these arrays
would be larger than a page (assuming 8-byte addresses and 4KB
pages). Allowing up to 64K in-flight RPCs (as TCP now does), each
buffer array would have to be 128 pages. That's an order-6
allocation. (Not that we're going there.)
A list is easier to expand dynamically. Instead of allocating a
larger array of pointers and copying the existing pointers to the
new array, simply append more buffers to each list.
This also makes it simpler to manage receive buffers that might
catch backwards-direction calls, or to post receive buffers in
bulk to amortize the overhead of ib_post_recv.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Reviewed-by: Devesh Sharma <devesh.sharma@avagotech.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Clean up: The error cases in rpcrdma_reply_handler() almost never
execute. Ensure the compiler places them out of the hot path.
No behavior change expected.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Reviewed-by: Devesh Sharma <devesh.sharma@avagotech.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Commit 8301a2c047 ("xprtrdma: Limit work done by completion
handler") was supposed to prevent xprtrdma's upcall handlers from
starving other softIRQ work by letting them return to the provider
before all CQEs have been polled.
The logic assumes the provider will call the upcall handler again
immediately if the CQ is re-armed while there are still queued CQEs.
This assumption is invalid. The IBTA spec says that after a CQ is
armed, the hardware must interrupt only when a new CQE is inserted.
xprtrdma can't rely on the provider calling again, even though some
providers do.
Therefore, leaving CQEs on queue makes sense only when there is
another mechanism that ensures all remaining CQEs are consumed in a
timely fashion. xprtrdma does not have such a mechanism. If a CQE
remains queued, the transport can wait forever to send the next RPC.
Finally, move the wcs array back onto the stack to ensure that the
poll array is always local to the CPU where the completion upcall is
running.
Fixes: 8301a2c047 ("xprtrdma: Limit work done by completion ...")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Reviewed-by: Devesh Sharma <devesh.sharma@avagotech.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
ib_req_notify_cq(IB_CQ_REPORT_MISSED_EVENTS) returns a positive
value if WCs were added to a CQ after the last completion upcall
but before the CQ has been re-armed.
Commit 7f23f6f6e3 ("xprtrmda: Reduce lock contention in
completion handlers") assumed that when ib_req_notify_cq() returned
a positive RC, the CQ had also been successfully re-armed, making
it safe to return control to the provider without losing any
completion signals. That is an invalid assumption.
Change both completion handlers to continue polling while
ib_req_notify_cq() returns a positive value.
Fixes: 7f23f6f6e3 ("xprtrmda: Reduce lock contention in ...")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Reviewed-by: Devesh Sharma <devesh.sharma@avagotech.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
After adding a swapfile on an NFS/RDMA mount and removing the
normal swap partition, I was able to push the NFS client well
into swap without any issue.
I forgot to swapoff the NFS file before rebooting. This pinned
the NFS mount and the IB core and provider, causing shutdown to
hang. I think this is expected and safe behavior. Probably
shutdown scripts should "swapoff -a" before unmounting any
filesystems.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Tested-By: Devesh Sharma <devesh.sharma@avagotech.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Unsignaled send WRs can get flushed as part of normal unmount, so don't
log them as warnings.
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
There's no need to record the time tracepoints take when tracing is off.
This is because:
1) We cannot see these records since ring_buffer record is off at that
moment.
2) If tracing is off and benchmark tracepoint is enabled, the time
tracepoint takes is fewer than the same situation when tracing is on,
since the tracepoints need to be wrote into ring_buffer, it would
take more time. If turn on tracing at this moment, the average and
standard deviation cannot exactly present the time that tracepoints
take to write data into ring_buffer.
Link: http://lkml.kernel.org/r/1445947933-27955-1-git-send-email-zhang.chunyan@linaro.org
Signed-off-by: Chunyan Zhang <zhang.chunyan@linaro.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Although, the default value of rel_type_nop is zero, and the value
of R_386_NONE/R_X86_64_NONE is zero too, but it should be assigned
a meaningful value explicitly, otherwise it looks confused.
Assign R_386_NONE to rel_type_nop for 386, assign R_X86_64_NONE
to rel_type_nop for x86_64.
Link: http://lkml.kernel.org/r/1446020606-16352-1-git-send-email-huawei.libin@huawei.com
Signed-off-by: Li Bin <huawei.libin@huawei.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
