On Tue, Sep 24, 2013 at 5:32 PM, Viresh Kumar wrote:
> On 25 September 2013 05:51, Jesse Gross wrote:
>> On Mon, Sep 23, 2013 at 11:20 PM, Viresh Kumar
>> wrote:
>
>>> static inline void rcu_read_lock_bh(void)
>>> {
>>> local_bh_disable();
>>> #ifdef CONFIG_PREEMPT_RT_FULL
>>> r
On 25 September 2013 05:51, Jesse Gross wrote:
> On Mon, Sep 23, 2013 at 11:20 PM, Viresh Kumar
> wrote:
>> static inline void rcu_read_lock_bh(void)
>> {
>> local_bh_disable();
>> #ifdef CONFIG_PREEMPT_RT_FULL
>> rcu_read_lock();
>> #else
>> __acquire(RCU_BH);
>>
On Mon, Sep 23, 2013 at 11:20 PM, Viresh Kumar wrote:
> On 24 September 2013 02:00, Jesse Gross wrote:
>> Search net/core/dev.c for RECURSION_LIMIT.
>
> Ahh, thanks for the pointers..
>
> This stuff was added in 2.6.37 And you said this earlier:
>
> "This loop checker is just compatibility code s
On 24 September 2013 02:00, Jesse Gross wrote:
> Search net/core/dev.c for RECURSION_LIMIT.
Ahh, thanks for the pointers..
This stuff was added in 2.6.37 And you said this earlier:
"This loop checker is just compatibility code since the exact same
logic exists in net/core/dev.c on newer kernels
On Mon, Sep 23, 2013 at 3:51 AM, Viresh Kumar wrote:
> On 21 September 2013 00:28, Jesse Gross wrote:
>> OVS is in the upstream kernel as well. This loop checker is just
>> compatibility code since the exact same logic exists in net/core/dev.c
>> on newer kernels. The same transformations should
On 21 September 2013 00:28, Jesse Gross wrote:
> On Mon, Sep 16, 2013 at 2:13 AM, Viresh Kumar wrote:
>> Is this piece of code reentrant? I don't think so, otherwise there would
>> have been races to update loop_counters..
> It certainly can be accessed simultaneously on multiple CPUs.
> Obviou
On Mon, Sep 16, 2013 at 2:13 AM, Viresh Kumar wrote:
> On 31 August 2013 06:27, Jesse Gross wrote:
>> On Wed, Aug 28, 2013 at 2:57 AM, Viresh Kumar
>> wrote:
>>> Which should have worked for both RT and non-RT kernels..
>>
>> I think that will work for the loop checker but executing the actions
On 31 August 2013 06:27, Jesse Gross wrote:
> On Wed, Aug 28, 2013 at 2:57 AM, Viresh Kumar wrote:
Sorry for delayed response.. Was on holidays for a week and then
was busy with CPUFreq bugs last week..
>> Which should have worked for both RT and non-RT kernels..
>
> I think that will work for
On Wed, Aug 28, 2013 at 2:57 AM, Viresh Kumar wrote:
> On 28 August 2013 05:48, Jesse Gross wrote:
>
>> The implementation is actually pretty much exactly the same as before.
>> The only reason why there are no longer separate process/interrupt
>> counters is because we started disabling bottom h
On 28 August 2013 05:48, Jesse Gross wrote:
> The implementation is actually pretty much exactly the same as before.
> The only reason why there are no longer separate process/interrupt
> counters is because we started disabling bottom halves when processing
> packets for userspace. However, with
10 matches
Mail list logo
