[Intel-wired-lan] [patch 11/35] net: ionic: Replace in_interrupt() usage.

Shannon Nelson snelson at pensando.io
Mon Sep 28 19:51:14 UTC 2020


On 9/28/20 10:24 AM, Shannon Nelson wrote:
> On 9/27/20 12:48 PM, Thomas Gleixner wrote:
>> From: Sebastian Andrzej Siewior <bigeasy at linutronix.de>
>>
>> The in_interrupt() usage in this driver tries to figure out which 
>> context
>> may sleep and which context may not sleep. in_interrupt() is not really
>> suitable as it misses both preemption disabled and interrupt disabled
>> invocations from task context.
>>
>> Conditionals like that in driver code are frowned upon in general 
>> because
>> invocations of functions from invalid contexts might not be detected
>> as the conditional papers over it.
>>
>> ionic_lif_addr() can be called from:
>>
>>   1) ->ndo_set_rx_mode() which is under netif_addr_lock_bh()) so it 
>> must not
>>      sleep.
>>
>>   2) Init and setup functions which are in fully preemptible task 
>> context.
>>
>> _ionic_lif_rx_mode() has only one call path with BH disabled.

Now that I've had my coffee, let's look at this again - there are 
multiple paths that get us to _ionic_lif_rx_mode():

.ndo_set_rx_mode
   ionic_set_rx_mode,
     _ionic_lif_rx_mode

{ ionic_open, ionic_lif_handle_fw_up, ionic_start_queues_reconfig }
     ionic_txrx_init
       ionic_set_rx_mode
         _ionic_lif_rx_mode

We may not want to change this one.

sln



>>
>> ionic_link_status_check_request() has two call paths:
>>
>>   1) NAPI which obviously cannot sleep
>>
>>   2) Setup which is again fully preemptible task context
>>
>> Add 'can_sleep' arguments to the affected functions and let the callers
>> provide the context instead of letting the functions deduce it.
>>
>> Signed-off-by: Sebastian Andrzej Siewior <bigeasy at linutronix.de>
>> Signed-off-by: Thomas Gleixner <tglx at linutronix.de>
>> Cc: Shannon Nelson <snelson at pensando.io>
>> Cc: Pensando Drivers <drivers at pensando.io>
>> Cc: "David S. Miller" <davem at davemloft.net>
>> Cc: Jakub Kicinski <kuba at kernel.org>
>> Cc: netdev at vger.kernel.org
>
> Acked-by: Shannon Nelson <snelson at pensando.io>
>
>> ---
>>
>> While reviewing the callpaths, a couple of things were observed which 
>> could
>> be improved:
>>
>> - ionic_lif_deferred_work() can iterate over the list. There is no need
>>    to schedule the work item after each iteration
>
> I think the original writer's intent was to avoid monopolizing the 
> work thread for very long on any one cycle, with the thought that we'd 
> be making more use of this than we currently are.  I'll address this.
>
>>
>> - ionic_link_status_check_request() could have ionic_deferred_work 
>> within
>>    ionic_lif(). This would avoid memory allocation from NAPI. More
>>    important, once IONIC_LIF_F_LINK_CHECK_REQUESTED is set and that 
>> alloc
>>    fails, the link check never happens.
>
> Thanks, I'll fix up that error condition.
>
>>
>> - ionic_lif_handle_fw_down() sets IONIC_LIF_F_FW_RESET. Invokes then
>>    ionic_lif_deinit() which only invokes cancel_work_sync() if
>>    IONIC_LIF_F_FW_RESET is not set. I think the logic is wrong here as
>>    the work must always be cancled. Also the list with ionic_deferred
>>    work items needs a clean up.
>
> I'll look at that, thanks.
>
> sln
>
>



More information about the Intel-wired-lan mailing list