Date:   Tue, 2 Jul 2019 13:27:07 +0300
From:   Ivan Khoronzhuk <ivan.khoronzhuk@...aro.org>
To:     Jesper Dangaard Brouer <brouer@...hat.com>
Cc:     grygorii.strashko@...com, hawk@...nel.org, davem@...emloft.net,
        ast@...nel.org, linux-kernel@...r.kernel.org,
        linux-omap@...r.kernel.org, xdp-newbies@...r.kernel.org,
        ilias.apalodimas@...aro.org, netdev@...r.kernel.org,
        daniel@...earbox.net, jakub.kicinski@...ronome.com,
        john.fastabend@...il.com
Subject: Re: [PATCH v5 net-next 1/6] xdp: allow same allocator usage

On Mon, Jul 01, 2019 at 01:40:59PM +0200, Jesper Dangaard Brouer wrote:
>
>I'm very skeptical about this approach.
>
>On Sun, 30 Jun 2019 20:23:43 +0300
>Ivan Khoronzhuk <ivan.khoronzhuk@...aro.org> wrote:
>
>> XDP rxqs can be same for ndevs running under same rx napi softirq.
>> But there is no ability to register same allocator for both rxqs,
>> by fact it can same rxq but has different ndev as a reference.
>
>This description is not very clear. It can easily be misunderstood.
>
>It is an absolute requirement that each RX-queue have their own
>page_pool object/allocator. (This where the performance comes from) as
>the page_pool have NAPI protected array for alloc and XDP_DROP recycle.
>
>Your driver/hardware seems to have special case, where a single
>RX-queue can receive packets for two different net_device'es.
>
>Do you violate this XDP devmap redirect assumption[1]?
>[1] https://github.com/torvalds/linux/blob/v5.2-rc7/kernel/bpf/devmap.c#L324-L329
Seems that yes, but that's used only for trace for now.
As it runs under napi and flush clear dev_rx i must do it right in the
rx_handler. So next patchset version will have one patch less.

Thanks!

>
>
>> Due to last changes allocator destroy can be defered till the moment
>> all packets are recycled by destination interface, afterwards it's
>> freed. In order to schedule allocator destroy only after all users are
>> unregistered, add refcnt to allocator object and schedule to destroy
>> only it reaches 0.
>
>The guiding principles when designing an API, is to make it easy to
>use, but also make it hard to misuse.
>
>Your API change makes it easy to misuse the API.  As it make it easy to
>(re)use the allocator pointer (likely page_pool) for multiple
>xdp_rxq_info structs.  It is only valid for your use-case, because you
>have hardware where a single RX-queue delivers to two different
>net_devices.  For other normal use-cases, this will be a violation.
>
>If I was a user of this API, and saw your xdp_allocator_get(), then I
>would assume that this was the normal case.  As minimum, we need to add
>a comment in the code, about this specific/intended use-case.  I
>through about detecting the misuse, by adding a queue_index to
>xdp_mem_allocator, that can be checked against, when calling
>xdp_rxq_info_reg_mem_model() with another xdp_rxq_info struct (to catch
>the obvious mistake where queue_index mismatch).

I can add, but not sure if it has or can have some conflicts with another
memory allocators, now or in future. Main here to not became a cornerstone
in some not obvious use-cases.

So, for now, let it be in this way:

--- a/include/net/xdp_priv.h
+++ b/include/net/xdp_priv.h
@@ -19,6 +19,7 @@ struct xdp_mem_allocator {
        struct delayed_work defer_wq;
        unsigned long defer_warn;
        unsigned long refcnt;
+       u32 queue_index;
 };

 #endif /* __LINUX_NET_XDP_PRIV_H__ */
diff --git a/net/core/xdp.c b/net/core/xdp.c
index a44621190fdc..c4bf29810f4d 100644
--- a/net/core/xdp.c
+++ b/net/core/xdp.c
@@ -324,7 +324,7 @@ static bool __is_supported_mem_type(enum xdp_mem_type type)
        return true;
 }

-static struct xdp_mem_allocator *xdp_allocator_get(void *allocator)
+static struct xdp_mem_allocator *xdp_allocator_find(void *allocator)
 {
        struct xdp_mem_allocator *xae, *xa = NULL;
        struct rhashtable_iter iter;
@@ -336,7 +336,6 @@ static struct xdp_mem_allocator *xdp_allocator_get(void *allocator)

                while ((xae = rhashtable_walk_next(&iter)) && !IS_ERR(xae)) {
                        if (xae->allocator == allocator) {
-                               xae->refcnt++;
                                xa = xae;
                                break;
                        }
@@ -386,9 +385,13 @@ int xdp_rxq_info_reg_mem_model(struct xdp_rxq_info *xdp_rxq,
                }
        }

-       xdp_alloc = xdp_allocator_get(allocator);
+       xdp_alloc = xdp_allocator_find(allocator);
        if (xdp_alloc) {
+               if (xdp_alloc->queue_index != xdp_rxq->queue_index)
+                       return -EINVAL;
+
                xdp_rxq->mem.id = xdp_alloc->mem.id;
+               xdp_alloc->refcnt++;
                return 0;
        }

@@ -406,6 +409,7 @@ int xdp_rxq_info_reg_mem_model(struct xdp_rxq_info *xdp_rxq,
        xdp_alloc->mem  = xdp_rxq->mem;
        xdp_alloc->allocator = allocator;
        xdp_alloc->refcnt = 1;
+       xdp_alloc->queue_index = xdp_rxq->queue_index;

        /* Insert allocator into ID lookup table */
        ptr = rhashtable_insert_slow(mem_id_ht, &id, &xdp_alloc->node);

Jesper, are you Ok with this version?

>
>
>> Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@...aro.org>
>> ---
>>  include/net/xdp_priv.h |  1 +
>>  net/core/xdp.c         | 46 ++++++++++++++++++++++++++++++++++++++++++
>>  2 files changed, 47 insertions(+)
>>
>> diff --git a/include/net/xdp_priv.h b/include/net/xdp_priv.h
>> index 6a8cba6ea79a..995b21da2f27 100644
>> --- a/include/net/xdp_priv.h
>> +++ b/include/net/xdp_priv.h
>> @@ -18,6 +18,7 @@ struct xdp_mem_allocator {
>>  	struct rcu_head rcu;
>>  	struct delayed_work defer_wq;
>>  	unsigned long defer_warn;
>> +	unsigned long refcnt;
>>  };
>>
>>  #endif /* __LINUX_NET_XDP_PRIV_H__ */
>> diff --git a/net/core/xdp.c b/net/core/xdp.c
>> index b29d7b513a18..a44621190fdc 100644
>> --- a/net/core/xdp.c
>> +++ b/net/core/xdp.c
>> @@ -98,6 +98,18 @@ bool __mem_id_disconnect(int id, bool force)
>>  		WARN(1, "Request remove non-existing id(%d), driver bug?", id);
>>  		return true;
>>  	}
>> +
>> +	/* to avoid calling hash lookup twice, decrement refcnt here till it
>> +	 * reaches zero, then it can be called from workqueue afterwards.
>> +	 */
>> +	if (xa->refcnt)
>> +		xa->refcnt--;
>> +
>> +	if (xa->refcnt) {
>> +		mutex_unlock(&mem_id_lock);
>> +		return true;
>> +	}
>> +
>>  	xa->disconnect_cnt++;
>>
>>  	/* Detects in-flight packet-pages for page_pool */
>> @@ -312,6 +324,33 @@ static bool __is_supported_mem_type(enum xdp_mem_type type)
>>  	return true;
>>  }
>>
>> +static struct xdp_mem_allocator *xdp_allocator_get(void *allocator)
>
>API wise, when you have "get" operation, you usually also have a "put"
>operation...

It's not part of external API.
I propose to rename it on xdp_allocator_find() as in above diff.
What do you say?

>
>> +{
>> +	struct xdp_mem_allocator *xae, *xa = NULL;
>> +	struct rhashtable_iter iter;
>> +
>> +	mutex_lock(&mem_id_lock);
>> +	rhashtable_walk_enter(mem_id_ht, &iter);
>> +	do {
>> +		rhashtable_walk_start(&iter);
>> +
>> +		while ((xae = rhashtable_walk_next(&iter)) && !IS_ERR(xae)) {
>> +			if (xae->allocator == allocator) {
>> +				xae->refcnt++;
>> +				xa = xae;
>> +				break;
>> +			}
>> +		}
>> +
>> +		rhashtable_walk_stop(&iter);
>> +
>> +	} while (xae == ERR_PTR(-EAGAIN));
>> +	rhashtable_walk_exit(&iter);
>> +	mutex_unlock(&mem_id_lock);
>> +
>> +	return xa;
>> +}
>> +
>>  int xdp_rxq_info_reg_mem_model(struct xdp_rxq_info *xdp_rxq,
>>  			       enum xdp_mem_type type, void *allocator)
>>  {
>> @@ -347,6 +386,12 @@ int xdp_rxq_info_reg_mem_model(struct xdp_rxq_info *xdp_rxq,
>>  		}
>>  	}
>>
>> +	xdp_alloc = xdp_allocator_get(allocator);
>> +	if (xdp_alloc) {
>> +		xdp_rxq->mem.id = xdp_alloc->mem.id;
>> +		return 0;
>> +	}
>> +
>
>The allocator pointer (in-practice) becomes the identifier for the
>mem.id (which rhashtable points to xdp_mem_allocator object).

So, you have no obj against it?

[...]

-- 
Regards,
Ivan Khoronzhuk

Powered by blists - more mailing lists