From: Kan Liang
Multiplexing overhead is one of the key overhead when the number of
events is more than available counters.
Signed-off-by: Kan Liang
---
include/linux/perf_event.h | 2 ++
include/uapi/linux/perf_event.h | 1 +
kernel/events/core.c| 16
3 fil
From: Kan Liang
The rough overhead rate can be caculated by the sum of all kinds of
overhead / elapsed time.
If the overhead rate is higher than 10%, warning the user.
Signed-off-by: Kan Liang
---
tools/perf/util/session.c | 26 ++
1 file changed, 26 insertions(+)
diff
From: Kan Liang
get_nsecs will be used to get time in following patches.
Signed-off-by: Kan Liang
---
tools/perf/builtin-sched.c | 2 +-
tools/perf/builtin.h | 1 +
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
index
From: Kan Liang
Record the elapsed time of perf record, and display it in perf report
Signed-off-by: Kan Liang
---
tools/perf/builtin-record.c | 10 ++
tools/perf/builtin-report.c | 1 +
tools/perf/util/event.h | 3 +++
tools/perf/util/machine.c | 3 +++
tools/perf/util/sessio
From: Kan Liang
Iterating all events which need to receive side-band events also bring
some overhead.
Save the overhead information in task context or CPU context, whichever
context is available.
Signed-off-by: Kan Liang
---
include/linux/perf_event.h | 2 ++
include/uapi/linux/perf_even
From: Kan Liang
The infrastructure to handle PERF_RECORD_OVERHEAD record type. A new
perf report option is also introduced as a knob to show the overhead
information.
Signed-off-by: Kan Liang
---
tools/include/uapi/linux/perf_event.h| 33
tools/perf/Documen
From: Kan Liang
Caculate the total NMI overhead on each CPU, and display them in perf
report
Signed-off-by: Kan Liang
---
tools/perf/builtin-report.c | 11 +++
tools/perf/util/event.h | 4
tools/perf/util/machine.c | 9 +
tools/perf/util/session.c | 18 ++
From: Kan Liang
A new perf record is introduced to export perf overhead information to
userspace. So we can measure the overhead of sampling directly.
To output the overhead information, it takes advantage of the existing
event log mechanism. But the overhead information is the system
overhead,
From: Kan Liang
Profiling brings additional overhead. High overhead may impacts the
behavior of the profiling object, impacts the accuracy of the
profiling result, and even hang the system.
Currently, perf has dynamic interrupt throttle mechanism to lower the
sample rate and overhead. But it has
From: Kan Liang
Caculate the total multiplexing overhead on each CPU, and display them
in perf report
Signed-off-by: Kan Liang
---
tools/perf/builtin-report.c | 8 ++--
tools/perf/util/event.h | 3 +++
tools/perf/util/machine.c | 5 +
tools/perf/util/session.c | 4
4 files
From: Kan Liang
Caculate the total overhead from accessing side-band events handler
function, and display them in perf report
Signed-off-by: Kan Liang
---
tools/perf/builtin-report.c | 3 +++
tools/perf/util/event.h | 5 +
tools/perf/util/machine.c | 5 +
tools/perf/util/session.
We shouldn't free cert->pub->key in x509_cert_parse() because
x509_free_certificate() also does this:
BUG: Double free or freeing an invalid pointer
...
Call Trace:
[] dump_stack+0x63/0x83
[] kasan_object_err+0x21/0x70
[] kasan_report_double_free+0
Hi Ryan,
[auto build test ERROR on asoc/for-next]
[also build test ERROR on v4.9-rc6 next-20161123]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
https://github.com/0day-ci/linux/commits/Ryan-Lee/ALSA-SoC-MAX98927-driver-Initial
On Nov 22, 2016, at 5:55 PM, David Howells wrote:
>
> Return enhanced file attributes from the Ext4 filesystem. This includes
> the following:
>
> (1) The inode creation time (i_crtime) as stx_btime, setting STATX_BTIME.
>
> (2) Certain FS_xxx_FL flags are mapped to stx_attribute flags.
>
> T
This adds the compatible string to the ohci driver
to be able to probe from DT
Signed-off-by: Axel Haslam
---
drivers/usb/host/ohci-da8xx.c | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/usb/host/ohci-da8xx.c b/drivers/usb/host/ohci-da8xx.c
index 07366ae..1818206 100644
---
Using a regulator to handle VBUS will eliminate the need for
platform data and callbacks, and make the driver more generic
allowing different types of regulators to handle VBUS.
The regulator equivalents to the platform callbacks are:
set_power -> regulator_enable/regulator_disable
get_
When booting using device tree, we can not make use of
platform callbacks to handle vbus and over current gpios.
This series allows the ohci-da8xx driver to use a regulator
instead of the platform callbacks to control vbus and adds
the device tree bindings to be able to probe using DT.
Once all u
To migrate to a DT based boot, we will remove the use of platform
callbacks, in favor of using the regulator framework to handle
vbus and over current.
In preparation to use a regulator instead of callbacks, move the platform
data callbacks into separate functions. This provides well defined place
This patch documents the device tree bindings required for
the ohci controller found in TI da8xx family of SoC's
Cc: robh...@kernel.org
Cc: mark.rutl...@arm.com
Cc: devicet...@vger.kernel.org
Acked-by: Rob Herring
Signed-off-by: Axel Haslam
---
.../devicetree/bindings/usb/ohci-da8xx.txt
Instead of global variables, use the extra_priv_size of
the ohci driver.
We cannot yet move the ocic mask because this is used on
the interrupt handler which is registered through platform
data and does not have an hcd pointer. This will be moved
on a later patch.
Tested-by: David Lechner
Signed
sound/soc/codecs/max98927.c:941:3-8: No need to set .owner here. The core will
do it.
Remove .owner field if calls are used which set it automatically
Generated by: scripts/coccinelle/api/platform_no_drv_owner.cocci
CC: Ryan Lee
Signed-off-by: Fengguang Wu
---
max98927.c |1 -
1 file c
Hi Ryan,
[auto build test WARNING on asoc/for-next]
[also build test WARNING on v4.9-rc6 next-20161123]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
https://github.com/0day-ci/linux/commits/Ryan-Lee/ALSA-SoC-MAX98927-driver-Initial
This patch remove UPDATE_STATS macro from
header slic.h which is not being used.
Signed-off-by: Sergio Paracuellos
---
drivers/staging/slicoss/slic.h | 8
1 file changed, 8 deletions(-)
diff --git a/drivers/staging/slicoss/slic.h b/drivers/staging/slicoss/slic.h
index 1f6562c..2c05868
Hello,
On Wed, Nov 23, 2016 at 08:00:57AM -0800, Randy Dunlap wrote:
> On 11/22/16 22:58, Stephen Rothwell wrote:
> > Hi all,
> >
> > Changes since 20161122:
> >
>
> on i386:
>
> fs/built-in.o: In function `__issue_discard_async':
> segment.c:(.text+0x5e2135): undefined reference to `__umoddi3
On 11/22/16 21:55, Sekhar Nori wrote:
> On Tuesday 22 November 2016 11:51 PM, Frank Rowand wrote:
>> Please note that the compatible property might contain several strings, not
>> just
>> a single string.
>
> So I guess the best thing to do is to use
> of_property_read_string_index() and print th
This patchset clean some code in slicoss driver:
* Removes not used macro.
* Replaces a macro with an inline function.
Sergio Paracuellos (2):
staging: slicoss: remove not used UPDATE_STATS macro
staging: slicoss: replace UPDATE_STATS_GB macro into an inline
function
drivers/staging/slic
Hi Eric,
On Tue, Nov 22, 2016 at 08:09:58AM -0800, Eric Dumazet wrote:
Looks like there's a few similarly named devices and drivers. Mine is
an SMSC LAN91C111 using the smc91x driver in
drivers/net/ethernet/smsc/smc91x.c, rather than smc911x.c. So the
interrupt handler is smc_interrupt()
CONFI
On Wed, Nov 23, 2016 at 12:21:03PM +0100, Tomasz Nowicki wrote:
> Hi Bjorn,
>
> On 23.11.2016 00:13, Bjorn Helgaas wrote:
> >Hi Tomasz,
> >
> >On Fri, Jun 10, 2016 at 09:55:19PM +0200, Tomasz Nowicki wrote:
> >>Implement pci_acpi_scan_root and other arch-specific call so that ARM64
> >>can start u
> A recent cleanup had the right idea to remove the initialization
> of the error variable, but missed the actual benefit of that,
> which is that we get warnings if there is a bug in it.
> Now we get a warning about a bug that was introduced by this cleanup:
>
> drivers/media/platform/davinci/vpf
On 11/23/16 10:13, Frank Rowand wrote:
> On 11/22/16 21:55, Sekhar Nori wrote:
>> On Tuesday 22 November 2016 11:51 PM, Frank Rowand wrote:
>>> Please note that the compatible property might contain several strings, not
>>> just
>>> a single string.
>>
>> So I guess the best thing to do is to use
This patch replaces UPDATE_STATS_GB macro in slic.h header file
into an inline function. This provides type safety and readability.
Signed-off-by: Sergio Paracuellos
---
drivers/staging/slicoss/slic.h| 6 ++---
drivers/staging/slicoss/slicoss.c | 56 ---
On resume from suspend a failure with -ESHUTDOWN is returned
from ohci_bus_resume, and the usb is inoperable.
This happens because ohci_suspend disables the master interrupt
and sets an hcd flag to say that the hw is no longer accessible.
Calling ohci_resume reverts the steps taken on ohci_suspen
Hi Hongzhou,
On 12/05/16 04:55, Hongzhou Yang wrote:
On Wed, 2016-05-11 at 19:09 -0700, Hongzhou Yang wrote:
On Thu, 2016-05-12 at 09:41 +0800, chunfeng yun wrote:
Hi,
On Wed, 2016-05-11 at 11:32 -0700, Hongzhou Yang wrote:
On Wed, 2016-05-11 at 13:56 +0200, Linus Walleij wrote:
On Tue, May
+++ Steven Rostedt [23/11/16 11:00 -0500]:
On Wed, 9 Nov 2016 05:40:58 -0500
Jessica Yu wrote:
+++ Aaron Tomlin [07/11/16 11:46 +]:
>Hi Jessica,
>
>Any thoughts?
Hi Aaron,
Thanks for your patience as I slowly work through a large swath of emails :-)
Anyway, this looks fine to me. A goin
Hi Caesar,
On Wed, Nov 23, 2016 at 10:29:32PM +0800, Caesar Wang wrote:
> The temp_to_code function will return 0 when we set the temperature to a
> invalid value (e.g. 61C, 62C, 63C), that's unpractical. This patch
> will prevent this case happening. That will return the max analog value to
>
Several versions of DW DMAC have multi block transfers hardware
support. Hardware support of multi block transfers is disabled
by default if we use DT to configure DMAC and software emulation
of multi block transfers used instead.
Add multi-block property, so it is possible to enable hardware
multi
It wasn't possible to enable some features like
memory-to-memory transfers or multi block transfers via DT.
It is fixed by these patches.
Changes for v4:
* Fix setting inverted value to "dwc->nollp". My fault - I
tested with wrong DTS, so DMAC was configured from autoconfig
instead of devi
All known devices, which use DT for configuration, support
memory-to-memory transfers. So enable it by default, if we read
configuration from DT.
Acked-by: Andy Shevchenko
Signed-off-by: Eugeniy Paltsev
---
drivers/dma/dw/platform.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/driv
On Wed, Nov 23, 2016 at 10:29:34PM +0800, Caesar Wang wrote:
> In some cases, some sensors didn't need the trip points, the
> set_trips will return {-INT_MAX, INT_MAX} to trigger thermal alarm.
>
> Signed-off-by: Caesar Wang
> ---
>
> Changes in v2:
> - Fixes something as Brian comments on
> h
On Wed, Nov 23, 2016 at 9:27 AM, Bart Van Assche
wrote:
> On 11/23/2016 09:13 AM, Logan Gunthorpe wrote:
>>
>> IMO any memory that has been registered for a P2P transaction should be
>> locked from being evicted. So if there's a get_user_pages call it needs
>> to be pinned until the put_page. The
I pulled Linus' tree this morning to make sure there wasn't a fix
since -rc6, but I still get "no symbol version" for modules on boot.
I'm running Debian stretch (gcc 6.2.0-13) on amd64 and build the
kernel with "make -j 8 bindeb-pkg".
I'm using CONFIG_VMAP_STACK=y
If I set CONFIG_MODVERSIONS=n,
Hi Arnd,
As Damien mentioned, is this fine to go, if you don't mind?
I've confirmed that sd_zbc.c allows a power-of-two number only.
Thanks,
>From b709ff532daedc42526a46cb26860127f4959e01 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann
Date: Tue, 22 Nov 2016 15:20:16 +0100
Subject: [PATCH] f2fs: f
Add tracepoints to dax_pmd_load_hole(), following the same logging
conventions as the tracepoints in dax_iomap_pmd_fault().
Here is an example PMD fault showing the new tracepoints:
read_big-1393 [007] 32.133809: dax_pmd_fault: shared mapping read
address 0x1040 vm_start 0x1020 v
Add the new include/trace/events/fs_dax.h tracepoint header, update
Matthew's email address and add myself as a maintainer for filesystem DAX.
Signed-off-by: Ross Zwisler
Suggested-by: Matthew Wilcox
---
MAINTAINERS | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/MAINTAIN
Tracepoints are the standard way to capture debugging and tracing
information in many parts of the kernel, including the XFS and ext4
filesystems. Create a tracepoint header for FS DAX and add the first DAX
tracepoints to the PMD fault handler. This allows the tracing for DAX to
be done in the sa
Add tracepoints to dax_pmd_insert_mapping(), following the same logging
conventions as the tracepoints in dax_iomap_pmd_fault().
Here is an example PMD fault showing the new tracepoints:
big-1544 [006] 48.153479: dax_pmd_fault: shared mapping write
address 0x10505000 vm_start 0x1020
Tracepoints are the standard way to capture debugging and tracing
information in many parts of the kernel, including the XFS and ext4
filesystems. This series creates a tracepoint header for FS DAX and add
the first few DAX tracepoints to the PMD fault handler. This allows the
tracing for DAX to
No functional change.
As of this commit:
commit 218dd85887da (".gitattributes: set git diff driver for C source code
files")
git-diff and git-format-patch both generate diffs whose hunks are correctly
prefixed by function names instead of labels, even if those labels aren't
indented with spaces.
With the current Kconfig setup it is possible to have the following:
CONFIG_EXT4_FS=y
CONFIG_FS_DAX=y
CONFIG_FS_IOMAP=n # this is in fs/Kconfig & isn't user accessible
With this config we get build failures in ext4_dax_fault() because the
iomap functions in fs/dax.c are missing:
fs/built-i
On Wed, 23 Nov 2016 10:33:41 +0100 Dmitry Vyukov wrote:
> On Wed, Nov 23, 2016 at 9:46 AM, Arnd Bergmann wrote:
> > On Wednesday, November 23, 2016 11:25:49 AM CET Andrey Ryabinin wrote:
> >>
> >> We have implementation for this feature:
> >> "[PATCH] kasan: support use-after-scope detec
On 11/22/2016 10:31 PM, Tetsuo Handa wrote:
> Tetsuo Handa wrote:
>> John Johansen wrote:
In order to minimize the burden of reviewing, this patchset implements
only functionality of checking program execution requests (i.e. execve()
system call) using pathnames. I'm planning to add
On 11/23/2016 02:01 AM, Sascha Hauer wrote:
> With this patch the serial core provides LED triggers for RX and TX.
>
> As the serial core layer does not know when the hardware actually sends
> or receives characters, this needs help from the UART drivers.
Looking at 8250, we call serial8250_tx_ch
Hi Ryan,
[auto build test WARNING on asoc/for-next]
[also build test WARNING on v4.9-rc6 next-20161123]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
https://github.com/0day-ci/linux/commits/Ryan-Lee/ALSA-SoC-MAX98927-driver-Initial
Commit e1399ba20eee ("powercap / RAPL: handle missing MSRs") added
contraint_to_pl() function to return index into an array. But it
can potentially return -EINVAL if powercap layer sends an out of
range constraint ID. This patch adds sanity check.
Unnecessary RAPL domain pointer check is removed s
On Wed, Nov 23, 2016 at 10:13:03AM -0700, Logan Gunthorpe wrote:
> an MR would be very tricky. The MR may be relied upon by another host
> and the kernel would have to inform user-space the MR was invalid then
> user-space would have to tell the remote application.
As Bart says, it would be best
Modular drivers may want to use of_for_each_phandle() - export its
constituent functions.
CC: Rob Herring
CC: Frank Rowand
Signed-off-by: Robin Murphy
---
drivers/of/base.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/of/base.c b/drivers/of/base.c
index a0bccb54a9bd..92e35fe0
On 11/23/2016 07:10 PM, Sergio Paracuellos wrote:
> This patch replaces UPDATE_STATS_GB macro in slic.h header file
> into an inline function. This provides type safety and readability.
>
> Signed-off-by: Sergio Paracuellos
> ---
> drivers/staging/slicoss/slic.h| 6 ++---
> drivers/staging/
On Tue, 22 Nov 2016 21:15:56 -
Thomas Gleixner wrote:
> The driver fails to:
>
> - initialize packages which are not available at driver init time,
> though the value of that initialization is completely unclear as
> nothing ever uses these values. I fixed it up nevertheless and leave
> it
On 11/16, Kuninori Morimoto wrote:
>
> Hi Rob, Michael, Russell
>
>
> What is the conclusion of this patch ?
> We shouldn't add devm_of_clk_get() ? or can I continue ?
>
> The problem of current [devm_]clk_get() handles *dev only,
> but I need to get clocks from DT node, not dev
>
> soun
On Wed, Nov 23, 2016 at 10:40:47AM -0800, Dan Williams wrote:
> I don't think that was designed for the case where the backing memory
> is a special/static physical address range rather than anonymous
> "System RAM", right?
The hardware doesn't care where the memory is. ODP is just a generic
mech
On 2016-11-23 02:05 PM, Jason Gunthorpe wrote:
On Wed, Nov 23, 2016 at 10:13:03AM -0700, Logan Gunthorpe wrote:
an MR would be very tricky. The MR may be relied upon by another host
and the kernel would have to inform user-space the MR was invalid then
user-space would have to tell the remote
When running certain workload on a debug kernel with lockdep turned on,
a ppc64 kvm guest could sometimes hit the following lockdep warning:
[ INFO: possible circular locking dependency detected ]
Possible unsafe locking scenario:
CPU0CPU1
On Wed, Nov 23, 2016 at 1:06 AM, Paolo Bonzini wrote:
>
>> Set MSR_IA32_CR{0,4}_FIXED1 to match the CPU's MSRs.
>>
>> In addition, MSR_IA32_CR4_FIXED1 should reflect the available CR4 bits
>> according to CPUID. Whenever guest CPUID is updated by userspace,
>> regenerate MSR_IA32_CR4_FIXED1 to mat
On Wed, Nov 23, 2016 at 8:30 AM, Laurent Pinchart
wrote:
> Hi Matt,
>
> Thank you for the patch.
>
> On Tuesday 22 Nov 2016 17:18:40 Matt Ranostay wrote:
>> There are several thermal sensors that only have a low-speed bus
>> interface but output valid video data. This patchset enables support
>> f
On 2016-11-23 12:27 PM, Bart Van Assche wrote:
On 11/23/2016 09:13 AM, Logan Gunthorpe wrote:
IMO any memory that has been registered for a P2P transaction should be
locked from being evicted. So if there's a get_user_pages call it needs
to be pinned until the put_page. The main issue being with
On 23/11/2016 20:16, David Matlack wrote:
> > Oh, I thought userspace would do that! Doing it in KVM is fine as well,
> > but then do we need to give userspace access to CR{0,4}_FIXED{0,1} at all?
>
> I think it should be safe for userspace to skip restoring CR4_FIXED1,
> since it is 100% genera
On Wed, Nov 23, 2016 at 12:58:26PM +, David Howells wrote:
> Dominik Brodowski wrote:
>
> > You might also need to disable CIS overrides and CIS firmware loading for
> > PCMCIA drivers, I presume. That needs two changes:
> >
> > - Abort in drivers/pcmcia/ds.c::pcmcia_load_firmware() or d
On 16-11-23 10:12 AM, Hayes Wang wrote:
Mark Lord [ml...@pobox.com]
[...]
What does this code do:
static void r8153_set_rx_early_size(struct r8152 *tp)
{
u32 mtu = tp->netdev->mtu;
u32 ocp_data = (agg_buf_sz - mtu - VLAN_ETH_HLEN - VLAN_HLEN) / 4;
ocp_write_word(tp, MCU_
Hi
On Wed, Nov 23, 2016 at 5:57 AM, Ryan Lee wrote:
> Signed-off-by: Ryan Lee
> ---
> .../devicetree/bindings/sound/max98927.txt | 32 +
> sound/soc/codecs/Kconfig |5 +
> sound/soc/codecs/Makefile |2 +
> sound/soc/codecs/ma
On Wed, Nov 23, 2016 at 02:14:40PM -0500, Serguei Sagalovitch wrote:
>
> On 2016-11-23 02:05 PM, Jason Gunthorpe wrote:
> >As Bart says, it would be best to be combined with something like
> >Mellanox's ODP MRs, which allows a page to be evicted and then trigger
> >a CPU interrupt if a DMA is att
Sebastian,
your patch looks good to me. I run successfully some small tests with it.
I want to suggest a small change in iucv_init() to keep the uniform technique
of undo labels below. Do you agree?
Kind regards, Ursula
On 11/17/2016 07:35 PM, Sebastian Andrzej Siewior wrote:
> Install the callb
On Mon, Oct 31, 2016 at 01:44:55PM -0600, Chris Mason wrote:
> On Mon, Oct 31, 2016 at 12:35:16PM -0700, Linus Torvalds wrote:
> >On Mon, Oct 31, 2016 at 11:55 AM, Dave Jones
> >wrote:
> >>
> >> BUG: Bad page state in process kworker/u8:12 pfn:4e0e39
> >> page:ea0013838e40 count:0 mapc
On Wed, Nov 23, 2016 at 12:27:36PM -0500, Nayna Jain wrote:
> sizep = of_get_property(np, "linux,sml-size", NULL);
> + if (of_property_match_string(np, "compatible", "IBM,vtpm") < 0)
> + log_size = be32_to_cpup(sizep);
> + else
> + log_size = *sizep;
> +
Uh, n
On 2016-11-23 00:38, Boris Brezillon wrote:
> On Tue, 22 Nov 2016 13:55:33 -0800
> Stefan Agner wrote:
>
>> On 2016-11-01 00:10, Lukasz Majewski wrote:
>> > This commit provides apply() callback implementation for i.MX's PWMv2.
>> >
>> > Suggested-by: Stefan Agner
>> > Suggested-by: Boris Brezil
From: "Wei Hu (Xavier)"
This patch modified the logic of allocating memory using APIs in
hns RoCE driver. We used kcalloc instead of kmalloc_array and
bitmap_zero. And When kcalloc failed, call vzalloc to alloc
memory.
Signed-off-by: Wei Hu (Xavier)
Signed-off-by: Ping Zhang
Signed-off-by: Sal
From: "Wei Hu (Xavier)"
This patch added the code for refreshing CQ CI using TPTR in hip06
SoC.
We will send a doorbell to hardware for refreshing CQ CI when user
succeed to poll a cqe. But it will be failed if the doorbell has
been blocked. So hardware will read a special buffer called TPTR
to
Hi Sascha,
[auto build test ERROR on tty/tty-testing]
[also build test ERROR on v4.9-rc6 next-20161123]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
https://github.com/0day-ci/linux/commits/Sascha-Hauer/serial-core-Add-LED-trigger
From: Shaobo Xu
IB core has implemented the calculation of GIDs and the management
of GID tables, and it is now responsible to supply query function
for GIDs. So the calculation of GIDs and the management of GID
tables in the RoCE driver is redundant.
The patch is to implement the add_gid/del_gi
From: Lijun Ou
This patch modified the condition of notifying hardware loopback.
In hip06, RoCE Engine has several ports, one QP is related
to one port. hardware only support loopback in the same port,
not in the different ports.
So, If QP related to port N, the dmac in the QP context equals
th
From: "Wei Hu (Xavier)"
This patch modified the output query info qp_attr->port_num
to fix bug in hip06.
Signed-off-by: Wei Hu (Xavier)
Signed-off-by: Salil Mehta
---
drivers/infiniband/hw/hns/hns_roce_hw_v1.c |4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/
This patch correct the comment style errors caught by
checkpatch.pl script
Signed-off-by: Salil Mehta
---
drivers/infiniband/hw/hns/hns_roce_cmd.c|8 ++--
drivers/infiniband/hw/hns/hns_roce_device.h | 28 ++---
drivers/infiniband/hw/hns/hns_roce_eq.c |6 +--
drivers/in
On 2016-11-23 03:51 AM, Christian König wrote:
Am 23.11.2016 um 08:49 schrieb Daniel Vetter:
On Tue, Nov 22, 2016 at 01:21:03PM -0800, Dan Williams wrote:
On Tue, Nov 22, 2016 at 1:03 PM, Daniel Vetter wrote:
On Tue, Nov 22, 2016 at 9:35 PM, Serguei Sagalovitch
wrote:
On 2016-11-22 03:10 P
From: Lijun Ou
In old code, the value of qp state from qpc was assigned for
attr->qp_state. The value may be an error while attr_mask &
IB_QP_STATE is zero.
Signed-off-by: Lijun Ou
Reviewed-by: Wei Hu (Xavier)
Signed-off-by: Salil Mehta
---
drivers/infiniband/hw/hns/hns_roce_hw_v1.c |2
From: "Wei Hu (Xavier)"
This patch modified the macro for the timeout when cmd is
processing as follows:
Before modification:
enum {
HNS_ROCE_CMD_TIME_CLASS_A = 1,
HNS_ROCE_CMD_TIME_CLASS_B = 1,
HNS_ROCE_CMD_TIME_CLASS_C = 1,
};
After modifi
On Wed, Nov 23, 2016 at 02:30:45PM +0100, Linus Walleij wrote:
> On Tue, Nov 22, 2016 at 6:55 PM, Thierry Reding
> wrote:
>
> > From: Thierry Reding
> >
> > Tegra186 has two GPIO controllers that are largely register compatible
> > between one another but are completely different from the contro
From: "Wei Hu (Xavier)"
When using CM to establish connections, qp number that was freed
just now will be rejected by ib core. To fix these problem, We
change qpn allocation to round-robin mode. We added the round-robin
mode for allocating resources using bitmap. We use round-robin mode
for qp nu
From: Lijun Ou
This patch mainly adds self loopback support for CM.
Signed-off-by: Lijun Ou
Signed-off-by: Peter Chen
Reviewed-by: Wei Hu (Xavier)
Signed-off-by: Salil Mehta
---
drivers/infiniband/hw/hns/hns_roce_hw_v1.c | 11 +++
drivers/infiniband/hw/hns/hns_roce_hw_v1.h |2
This patchset introduces some code improvements and fixes
for the identified problems in the HNS RoCE driver.
Lijun Ou (4):
IB/hns: Add the interface for querying QP1
IB/hns: add self loopback for CM
IB/hns: Modify the condition of notifying hardware loopback
IB/hns: Fix the bug for qp sta
From: Lijun Ou
In old code, It only added the interface for querying non-specific
QP. This patch mainly adds an interface for querying QP1.
Signed-off-by: Lijun Ou
Reviewed-by: Wei Hu (Xavier)
Signed-off-by: Salil Mehta
---
Change Log
Patch V2: Addressed the comment provided by Anurup M
Hi Michal,
On Tue, 2016-11-22 at 22:34 +0100, Michal Marek wrote:
> The KBUILD_IMAGE variable is used by the rpm and deb-pkg targets, which
> expect it to point to the image file in the build directory. The
> builddeb script has a workaround for architectures which only provide
> the basename, but
>From commit 90954e7b9407 ("x86/coredump: Use pr_reg size, rather that
TIF_IA32 flag") elf coredump file is constructed according to register
set size - and that's good: if binary crashes with 32-bit code selector,
generate 32-bit ELF core, otherwise - 64-bit core.
That was made for restoring 32-bi
On Wed, Nov 23, 2016 at 11:31:52AM -0600, Bjorn Helgaas wrote:
> On Mon, Nov 21, 2016 at 10:53:52AM -0600, Bjorn Helgaas wrote:
> > On Wed, Nov 16, 2016 at 12:11:58PM -0600, Bjorn Helgaas wrote:
> > > Hi Johannes,
> > >
> > > On Wed, Nov 02, 2016 at 04:35:52PM -0600, Johannes Thumshirn wrote:
> >
On 2016-11-23 02:12 PM, Jason Gunthorpe wrote:
On Wed, Nov 23, 2016 at 10:40:47AM -0800, Dan Williams wrote:
I don't think that was designed for the case where the backing memory
is a special/static physical address range rather than anonymous
"System RAM", right?
The hardware doesn't care w
On Wed, Nov 23, 2016 at 02:25:51PM +0100, Linus Walleij wrote:
> On Tue, Nov 22, 2016 at 6:30 PM, Thierry Reding
> wrote:
>
> > So I don't really know how to go about merging both. I'll reply to this
> > email later with a copy of the patch that I wrote, maybe we can take it
> > from there.
>
>
On Wed, Nov 23, 2016 at 02:34:19PM -0500, Dave Jones wrote:
> [ 317.689216] BUG: Bad page state in process kworker/u8:8 pfn:4d8fd4
> trace from just before this happened. Does this shed any light ?
>
> https://codemonkey.org.uk/junk/trace.txt
crap, I just noticed the timestamps in the trac
On Wed, Nov 23, 2016 at 12:27:35PM -0500, Nayna Jain wrote:
> The device driver code for the event log has the init functions and
> TPM 1.2 parsing logic both defined in same file(tpm_eventlog.c).
>
> Since the initialization functions are common with the TPM 2.0 event
> log support, this patch sp
On Wed, Nov 23, 2016 at 10:21 AM, Brian Starkey wrote:
> This patch didn't help.
>
> I did get some new traces though - I've attached the diff for the
> trace_printks I added.
>
> Before 4cd13c21b207:
> https://drive.google.com/open?id=0B8siaK6ZjvEwcEtOeFQzTmY0Nnc
> After 4cd13c21b207:
> https://
On 11/23, Georgi Djakov wrote:
> Fix the clk_hw references to the actual clocks and add a xlate function
> to return the hw pointers from the already existing static array.
>
> Reported-by: Michael Scott
> Signed-off-by: Georgi Djakov
> ---
Applied to clk-next
--
Qualcomm Innovation Center, I
On 11/23, Georgi Djakov wrote:
> Fix the clk_hw references to the actual clocks and add a xlate function
> to return the hw pointers from the already existing static array.
>
> Reported-by: Michael Scott
> Signed-off-by: Georgi Djakov
> ---
Applied to clk-next
--
Qualcomm Innovation Center, I
On Wed, Nov 23, 2016 at 04:44:40AM -0500, kan.li...@intel.com wrote:
> From: Kan Liang
>
> NMI handler is one of the most important part which brings overhead.
>
> There are lots of NMI during sampling. It's very expensive to log each
> NMI. So the accumulated time and NMI# will be output when e
501 - 600 of 925 matches
Mail list logo
