le CPU
resources.
Pavel Tatashin (2):
ixgbe: release lock for the duration of ixgbe_suspend_close()
drivers core: multi-threading device shutdown
drivers/base/core.c | 238 ++
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 9 +-
2 files change
down parents, so parent
device must wait for its children to finish.
With this patch, on the same machine devices_shutdown() takes 1.142s, and
without mlx4 one second delay only 0.38s
Signed-off-by: Pavel Tatashin
---
drivers/base/core.c | 238 +++-
1 file
is not held when ixgbe_close() is called also during shutdown but for
kexec case.
Signed-off-by: Pavel Tatashin
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
b/drivers/net
Hi Alex,
> I'm not a fan of dropping the mutex while we go through
> ixgbe_close_suspend. I'm concerned it will result in us having a
> number of races on shutdown.
I would agree, but ixgbe_close_suspend() is already called without this
mutex from ixgbe_close(). This path is executed also during
On Thu, May 3, 2018 at 1:54 AM Tobin C. Harding wrote:
> This code was a pleasure to read, super clean.
Hi Tobin,
Thank you very much for your review, I will address all of your comments in
the next revision.
BTW, I found a lock ordering issue in my work that that I will need to
fix:
In de
> Hi Pavel, you might want to pull lock optimization out of this patch
series.
> The parallelization by itself is valuable, and optimizations for
individual
> devices including locks can come later.
Hi Steve,
Yes, I will pull this patch out of the series. Thank you for the suggestion.
Pavel
4 one second delay only 0.38s
Signed-off-by: Pavel Tatashin
---
drivers/base/core.c | 275
1 file changed, 225 insertions(+), 50 deletions(-)
diff --git a/drivers/base/core.c b/drivers/base/core.c
index b610816eb887..110d7970deef 100644
--- a/drivers
these functions for every single device but
only using one thread.
Since, nothing else is running on the machine by the device_shutdown()
s called, there is no reason not to utilize all the available CPU
resources.
Pavel Tatashin (1):
drivers core: multi-threading device shutdown
drivers/base/core.c
4 one second delay only 0.38s
Signed-off-by: Pavel Tatashin
---
drivers/base/core.c | 275
1 file changed, 225 insertions(+), 50 deletions(-)
diff --git a/drivers/base/core.c b/drivers/base/core.c
index b610816eb887..1f9fda42a968 100644
--- a/drivers
ll the available CPU
resources.
Pavel Tatashin (1):
drivers core: multi-threading device shutdown
drivers/base/core.c | 275
1 file changed, 225 insertions(+), 50 deletions(-)
--
2.17.0
On 05/14/2018 11:03 AM, Greg KH wrote:
> On Mon, May 07, 2018 at 11:54:01AM -0400, Pavel Tatashin wrote:
>> Changelog
>> v2 - v3
>> - Fixed warning from kbuild test.
>> - Moved device_lock/device_unlock inside device_shutdown_tree().
>>
>> v1 - v
4 one second delay only 0.38s
This feature can be optionally disabled via kernel parameter:
device_shutdown_serial. When booted with this parameter, device_shutdown()
will shutdown devices one by one.
Signed-off-by: Pavel Tatashin
---
drivers/base/core.c | 292 +++
using one thread.
Since, nothing else is running on the machine by the time
device_shutdown() is called, there is no reason not to utilize all the
available CPU resources.
Pavel Tatashin (1):
drivers core: multi-threading device shutdown
drivers/base/core.c
Hi Andy,
Thank you for your comments. I will send an updated patch soon. My replies are
below:
On 05/14/2018 04:04 PM, Andy Shevchenko wrote:
> Can we still preserve an order here? (Yes, even if the entire list is
> not fully ordered)
> In the context I see it would go before netdevice.h.
Sure
Hi Greg,
> Can you refactor this to be at least 2 patches? One that moves code
> around to comon functions to make the second patch, that adds the new
> functionality, easier to review and understand?
Yes, I will split the patch into a two-three patches.
> And I echo the "don't use kerneldoc f
ion on one device.
Signed-off-by: Pavel Tatashin
---
drivers/base/core.c | 50 +++--
1 file changed, 30 insertions(+), 20 deletions(-)
diff --git a/drivers/base/core.c b/drivers/base/core.c
index b610816eb887..ed189f6d1a2f 100644
--- a/drivers/base/core.c
multi-threaded have _task() in their name.
Signed-off-by: Pavel Tatashin
---
drivers/base/core.c | 178
1 file changed, 149 insertions(+), 29 deletions(-)
diff --git a/drivers/base/core.c b/drivers/base/core.c
index ed189f6d1a2f..210b619931bc 100644
4 one second delay only 0.38s
This feature can be optionally disabled via kernel parameter:
device_shutdown_serial. When booted with this parameter, device_shutdown()
will shutdown devices one by one.
Signed-off-by: Pavel Tatashin
---
drivers/base/core.c | 70
e time
device_shutdown() is called, there is no reason not to utilize all the
available CPU resources.
Pavel Tatashin (3):
drivers core: refactor device_shutdown
drivers core: prepare device_shutdown for multi-threading
drivers core: multi-threading device shutdown
drivers/base/core.c
* parallelized this function, so drop lock for the
>
> Parallelizing? Else the sentence doesn't parse for me. :-)
Hi Sergei,
In a separate series I parallelized device_shutdown(), see:
http://lkml.kernel.org/r/20180516024004.28977-1-pasha.tatas...@oracle.com
But, this particular patch
On 05/18/2018 01:28 PM, Sergei Shtylyov wrote:
> On 05/18/2018 02:37 PM, Pavel Tatashin wrote:
>
>> * parallelized this function, so drop lock for the
>>>
>>> Parallelizing? Else the sentence doesn't parse for me. :-)
>
>My comment ha
Hi Pavel,
Thank you for looking at this patch. BTW, the version 5 is out. The latest
thread is anchered here:
http://lkml.kernel.org/r/20180516024004.28977-1-pasha.tatas...@oracle.com
> ixgbe is network card, right? So ... it does not have any persistent
> state and no moving parts, and there's n
22 matches
Mail list logo