iommu.lists.linux-foundation.org archive mirror
 help / color / mirror / Atom feed
* [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails
@ 2020-08-20 12:49 vjitta
  2020-08-20 12:49 ` [PATCH v2 2/2] iommu/iova: Free global iova rcache on iova alloc failure vjitta
                   ` (2 more replies)
  0 siblings, 3 replies; 10+ messages in thread
From: vjitta @ 2020-08-20 12:49 UTC (permalink / raw)
  To: joro, iommu, linux-kernel; +Cc: vjitta, robin.murphy, vinmenon, kernel-team

From: Vijayanand Jitta <vjitta@codeaurora.org>

When ever a new iova alloc request comes iova is always searched
from the cached node and the nodes which are previous to cached
node. So, even if there is free iova space available in the nodes
which are next to the cached node iova allocation can still fail
because of this approach.

Consider the following sequence of iova alloc and frees on
1GB of iova space

1) alloc - 500MB
2) alloc - 12MB
3) alloc - 499MB
4) free -  12MB which was allocated in step 2
5) alloc - 13MB

After the above sequence we will have 12MB of free iova space and
cached node will be pointing to the iova pfn of last alloc of 13MB
which will be the lowest iova pfn of that iova space. Now if we get an
alloc request of 2MB we just search from cached node and then look
for lower iova pfn's for free iova and as they aren't any, iova alloc
fails though there is 12MB of free iova space.

To avoid such iova search failures do a retry from the last rb tree node
when iova search fails, this will search the entire tree and get an iova
if its available.

Signed-off-by: Vijayanand Jitta <vjitta@codeaurora.org>
---
 drivers/iommu/iova.c | 23 +++++++++++++++++------
 1 file changed, 17 insertions(+), 6 deletions(-)

diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
index 49fc01f..4e77116 100644
--- a/drivers/iommu/iova.c
+++ b/drivers/iommu/iova.c
@@ -184,8 +184,9 @@ static int __alloc_and_insert_iova_range(struct iova_domain *iovad,
 	struct rb_node *curr, *prev;
 	struct iova *curr_iova;
 	unsigned long flags;
-	unsigned long new_pfn;
+	unsigned long new_pfn, low_pfn_new;
 	unsigned long align_mask = ~0UL;
+	unsigned long high_pfn = limit_pfn, low_pfn = iovad->start_pfn;
 
 	if (size_aligned)
 		align_mask <<= fls_long(size - 1);
@@ -198,15 +199,25 @@ static int __alloc_and_insert_iova_range(struct iova_domain *iovad,
 
 	curr = __get_cached_rbnode(iovad, limit_pfn);
 	curr_iova = rb_entry(curr, struct iova, node);
+	low_pfn_new = curr_iova->pfn_hi + 1;
+
+retry:
 	do {
-		limit_pfn = min(limit_pfn, curr_iova->pfn_lo);
-		new_pfn = (limit_pfn - size) & align_mask;
+		high_pfn = min(high_pfn, curr_iova->pfn_lo);
+		new_pfn = (high_pfn - size) & align_mask;
 		prev = curr;
 		curr = rb_prev(curr);
 		curr_iova = rb_entry(curr, struct iova, node);
-	} while (curr && new_pfn <= curr_iova->pfn_hi);
-
-	if (limit_pfn < size || new_pfn < iovad->start_pfn) {
+	} while (curr && new_pfn <= curr_iova->pfn_hi && new_pfn >= low_pfn);
+
+	if (high_pfn < size || new_pfn < low_pfn) {
+		if (low_pfn == iovad->start_pfn && low_pfn_new < limit_pfn) {
+			high_pfn = limit_pfn;
+			low_pfn = low_pfn_new;
+			curr = &iovad->anchor.node;
+			curr_iova = rb_entry(curr, struct iova, node);
+			goto retry;
+		}
 		iovad->max32_alloc_size = size;
 		goto iova32_full;
 	}
-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation
1.9.1

_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [PATCH v2 2/2] iommu/iova: Free global iova rcache on iova alloc failure
  2020-08-20 12:49 [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails vjitta
@ 2020-08-20 12:49 ` vjitta
  2020-09-18 14:41   ` Robin Murphy
  2020-08-28  7:31 ` [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails Vijayanand Jitta
  2020-09-18 14:18 ` Robin Murphy
  2 siblings, 1 reply; 10+ messages in thread
From: vjitta @ 2020-08-20 12:49 UTC (permalink / raw)
  To: joro, iommu, linux-kernel; +Cc: vjitta, robin.murphy, vinmenon, kernel-team

From: Vijayanand Jitta <vjitta@codeaurora.org>

When ever an iova alloc request fails we free the iova
ranges present in the percpu iova rcaches and then retry
but the global iova rcache is not freed as a result we could
still see iova alloc failure even after retry as global
rcache is holding the iova's which can cause fragmentation.
So, free the global iova rcache as well and then go for the
retry.

Signed-off-by: Vijayanand Jitta <vjitta@codeaurora.org>
---
 drivers/iommu/iova.c | 23 +++++++++++++++++++++++
 include/linux/iova.h |  6 ++++++
 2 files changed, 29 insertions(+)

diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
index 4e77116..5836c87 100644
--- a/drivers/iommu/iova.c
+++ b/drivers/iommu/iova.c
@@ -442,6 +442,7 @@ struct iova *find_iova(struct iova_domain *iovad, unsigned long pfn)
 		flush_rcache = false;
 		for_each_online_cpu(cpu)
 			free_cpu_cached_iovas(cpu, iovad);
+		free_global_cached_iovas(iovad);
 		goto retry;
 	}
 
@@ -1055,5 +1056,27 @@ void free_cpu_cached_iovas(unsigned int cpu, struct iova_domain *iovad)
 	}
 }
 
+/*
+ * free all the IOVA ranges of global cache
+ */
+void free_global_cached_iovas(struct iova_domain *iovad)
+{
+	struct iova_rcache *rcache;
+	unsigned long flags;
+	int i, j;
+
+	for (i = 0; i < IOVA_RANGE_CACHE_MAX_SIZE; ++i) {
+		rcache = &iovad->rcaches[i];
+		spin_lock_irqsave(&rcache->lock, flags);
+		for (j = 0; j < rcache->depot_size; ++j) {
+			iova_magazine_free_pfns(rcache->depot[j], iovad);
+			iova_magazine_free(rcache->depot[j]);
+			rcache->depot[j] = NULL;
+		}
+		rcache->depot_size = 0;
+		spin_unlock_irqrestore(&rcache->lock, flags);
+	}
+}
+
 MODULE_AUTHOR("Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>");
 MODULE_LICENSE("GPL");
diff --git a/include/linux/iova.h b/include/linux/iova.h
index a0637ab..a905726 100644
--- a/include/linux/iova.h
+++ b/include/linux/iova.h
@@ -163,6 +163,7 @@ int init_iova_flush_queue(struct iova_domain *iovad,
 struct iova *split_and_remove_iova(struct iova_domain *iovad,
 	struct iova *iova, unsigned long pfn_lo, unsigned long pfn_hi);
 void free_cpu_cached_iovas(unsigned int cpu, struct iova_domain *iovad);
+void free_global_cached_iovas(struct iova_domain *iovad);
 #else
 static inline int iova_cache_get(void)
 {
@@ -270,6 +271,11 @@ static inline void free_cpu_cached_iovas(unsigned int cpu,
 					 struct iova_domain *iovad)
 {
 }
+
+static inline void free_global_cached_iovas(struct iova_domain *iovad)
+{
+}
+
 #endif
 
 #endif
-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation
1.9.1

_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails
  2020-08-20 12:49 [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails vjitta
  2020-08-20 12:49 ` [PATCH v2 2/2] iommu/iova: Free global iova rcache on iova alloc failure vjitta
@ 2020-08-28  7:31 ` Vijayanand Jitta
  2020-09-14  4:50   ` Vijayanand Jitta
  2020-09-18 14:18 ` Robin Murphy
  2 siblings, 1 reply; 10+ messages in thread
From: Vijayanand Jitta @ 2020-08-28  7:31 UTC (permalink / raw)
  To: joro, iommu, linux-kernel; +Cc: robin.murphy, vinmenon, kernel-team



On 8/20/2020 6:19 PM, vjitta@codeaurora.org wrote:
> From: Vijayanand Jitta <vjitta@codeaurora.org>
> 
> When ever a new iova alloc request comes iova is always searched
> from the cached node and the nodes which are previous to cached
> node. So, even if there is free iova space available in the nodes
> which are next to the cached node iova allocation can still fail
> because of this approach.
> 
> Consider the following sequence of iova alloc and frees on
> 1GB of iova space
> 
> 1) alloc - 500MB
> 2) alloc - 12MB
> 3) alloc - 499MB
> 4) free -  12MB which was allocated in step 2
> 5) alloc - 13MB
> 
> After the above sequence we will have 12MB of free iova space and
> cached node will be pointing to the iova pfn of last alloc of 13MB
> which will be the lowest iova pfn of that iova space. Now if we get an
> alloc request of 2MB we just search from cached node and then look
> for lower iova pfn's for free iova and as they aren't any, iova alloc
> fails though there is 12MB of free iova space.
> 
> To avoid such iova search failures do a retry from the last rb tree node
> when iova search fails, this will search the entire tree and get an iova
> if its available.
> 
> Signed-off-by: Vijayanand Jitta <vjitta@codeaurora.org>
> ---
>  drivers/iommu/iova.c | 23 +++++++++++++++++------
>  1 file changed, 17 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
> index 49fc01f..4e77116 100644
> --- a/drivers/iommu/iova.c
> +++ b/drivers/iommu/iova.c
> @@ -184,8 +184,9 @@ static int __alloc_and_insert_iova_range(struct iova_domain *iovad,
>  	struct rb_node *curr, *prev;
>  	struct iova *curr_iova;
>  	unsigned long flags;
> -	unsigned long new_pfn;
> +	unsigned long new_pfn, low_pfn_new;
>  	unsigned long align_mask = ~0UL;
> +	unsigned long high_pfn = limit_pfn, low_pfn = iovad->start_pfn;
>  
>  	if (size_aligned)
>  		align_mask <<= fls_long(size - 1);
> @@ -198,15 +199,25 @@ static int __alloc_and_insert_iova_range(struct iova_domain *iovad,
>  
>  	curr = __get_cached_rbnode(iovad, limit_pfn);
>  	curr_iova = rb_entry(curr, struct iova, node);
> +	low_pfn_new = curr_iova->pfn_hi + 1;
> +
> +retry:
>  	do {
> -		limit_pfn = min(limit_pfn, curr_iova->pfn_lo);
> -		new_pfn = (limit_pfn - size) & align_mask;
> +		high_pfn = min(high_pfn, curr_iova->pfn_lo);
> +		new_pfn = (high_pfn - size) & align_mask;
>  		prev = curr;
>  		curr = rb_prev(curr);
>  		curr_iova = rb_entry(curr, struct iova, node);
> -	} while (curr && new_pfn <= curr_iova->pfn_hi);
> -
> -	if (limit_pfn < size || new_pfn < iovad->start_pfn) {
> +	} while (curr && new_pfn <= curr_iova->pfn_hi && new_pfn >= low_pfn);
> +
> +	if (high_pfn < size || new_pfn < low_pfn) {
> +		if (low_pfn == iovad->start_pfn && low_pfn_new < limit_pfn) {
> +			high_pfn = limit_pfn;
> +			low_pfn = low_pfn_new;
> +			curr = &iovad->anchor.node;
> +			curr_iova = rb_entry(curr, struct iova, node);
> +			goto retry;
> +		}
>  		iovad->max32_alloc_size = size;
>  		goto iova32_full;
>  	}
> 

ping ?

Thanks,
Vijay
-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a
member of Code Aurora Forum, hosted by The Linux Foundation
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails
  2020-08-28  7:31 ` [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails Vijayanand Jitta
@ 2020-09-14  4:50   ` Vijayanand Jitta
  2020-09-18  8:18     ` Joerg Roedel
  0 siblings, 1 reply; 10+ messages in thread
From: Vijayanand Jitta @ 2020-09-14  4:50 UTC (permalink / raw)
  To: joro, iommu, linux-kernel; +Cc: robin.murphy, vinmenon, kernel-team



On 8/28/2020 1:01 PM, Vijayanand Jitta wrote:
> 
> 
> On 8/20/2020 6:19 PM, vjitta@codeaurora.org wrote:
>> From: Vijayanand Jitta <vjitta@codeaurora.org>
>>
>> When ever a new iova alloc request comes iova is always searched
>> from the cached node and the nodes which are previous to cached
>> node. So, even if there is free iova space available in the nodes
>> which are next to the cached node iova allocation can still fail
>> because of this approach.
>>
>> Consider the following sequence of iova alloc and frees on
>> 1GB of iova space
>>
>> 1) alloc - 500MB
>> 2) alloc - 12MB
>> 3) alloc - 499MB
>> 4) free -  12MB which was allocated in step 2
>> 5) alloc - 13MB
>>
>> After the above sequence we will have 12MB of free iova space and
>> cached node will be pointing to the iova pfn of last alloc of 13MB
>> which will be the lowest iova pfn of that iova space. Now if we get an
>> alloc request of 2MB we just search from cached node and then look
>> for lower iova pfn's for free iova and as they aren't any, iova alloc
>> fails though there is 12MB of free iova space.
>>
>> To avoid such iova search failures do a retry from the last rb tree node
>> when iova search fails, this will search the entire tree and get an iova
>> if its available.
>>
>> Signed-off-by: Vijayanand Jitta <vjitta@codeaurora.org>
>> ---
>>  drivers/iommu/iova.c | 23 +++++++++++++++++------
>>  1 file changed, 17 insertions(+), 6 deletions(-)
>>
>> diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
>> index 49fc01f..4e77116 100644
>> --- a/drivers/iommu/iova.c
>> +++ b/drivers/iommu/iova.c
>> @@ -184,8 +184,9 @@ static int __alloc_and_insert_iova_range(struct iova_domain *iovad,
>>  	struct rb_node *curr, *prev;
>>  	struct iova *curr_iova;
>>  	unsigned long flags;
>> -	unsigned long new_pfn;
>> +	unsigned long new_pfn, low_pfn_new;
>>  	unsigned long align_mask = ~0UL;
>> +	unsigned long high_pfn = limit_pfn, low_pfn = iovad->start_pfn;
>>  
>>  	if (size_aligned)
>>  		align_mask <<= fls_long(size - 1);
>> @@ -198,15 +199,25 @@ static int __alloc_and_insert_iova_range(struct iova_domain *iovad,
>>  
>>  	curr = __get_cached_rbnode(iovad, limit_pfn);
>>  	curr_iova = rb_entry(curr, struct iova, node);
>> +	low_pfn_new = curr_iova->pfn_hi + 1;
>> +
>> +retry:
>>  	do {
>> -		limit_pfn = min(limit_pfn, curr_iova->pfn_lo);
>> -		new_pfn = (limit_pfn - size) & align_mask;
>> +		high_pfn = min(high_pfn, curr_iova->pfn_lo);
>> +		new_pfn = (high_pfn - size) & align_mask;
>>  		prev = curr;
>>  		curr = rb_prev(curr);
>>  		curr_iova = rb_entry(curr, struct iova, node);
>> -	} while (curr && new_pfn <= curr_iova->pfn_hi);
>> -
>> -	if (limit_pfn < size || new_pfn < iovad->start_pfn) {
>> +	} while (curr && new_pfn <= curr_iova->pfn_hi && new_pfn >= low_pfn);
>> +
>> +	if (high_pfn < size || new_pfn < low_pfn) {
>> +		if (low_pfn == iovad->start_pfn && low_pfn_new < limit_pfn) {
>> +			high_pfn = limit_pfn;
>> +			low_pfn = low_pfn_new;
>> +			curr = &iovad->anchor.node;
>> +			curr_iova = rb_entry(curr, struct iova, node);
>> +			goto retry;
>> +		}
>>  		iovad->max32_alloc_size = size;
>>  		goto iova32_full;
>>  	}
>>
> 
> ping ?
> 
> Thanks,
> Vijay
> 

ping ?

Thanks,
Vijay
-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a
member of Code Aurora Forum, hosted by The Linux Foundation
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails
  2020-09-14  4:50   ` Vijayanand Jitta
@ 2020-09-18  8:18     ` Joerg Roedel
  0 siblings, 0 replies; 10+ messages in thread
From: Joerg Roedel @ 2020-09-18  8:18 UTC (permalink / raw)
  To: Vijayanand Jitta; +Cc: vinmenon, iommu, kernel-team, linux-kernel, robin.murphy

On Mon, Sep 14, 2020 at 10:20:55AM +0530, Vijayanand Jitta wrote:
> ping ?

Robin needs to have a look first.

Regards,

	Joerg
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails
  2020-08-20 12:49 [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails vjitta
  2020-08-20 12:49 ` [PATCH v2 2/2] iommu/iova: Free global iova rcache on iova alloc failure vjitta
  2020-08-28  7:31 ` [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails Vijayanand Jitta
@ 2020-09-18 14:18 ` Robin Murphy
  2020-09-30  5:48   ` Vijayanand Jitta
  2 siblings, 1 reply; 10+ messages in thread
From: Robin Murphy @ 2020-09-18 14:18 UTC (permalink / raw)
  To: vjitta, joro, iommu, linux-kernel; +Cc: vinmenon, kernel-team

On 2020-08-20 13:49, vjitta@codeaurora.org wrote:
> From: Vijayanand Jitta <vjitta@codeaurora.org>
> 
> When ever a new iova alloc request comes iova is always searched
> from the cached node and the nodes which are previous to cached
> node. So, even if there is free iova space available in the nodes
> which are next to the cached node iova allocation can still fail
> because of this approach.
> 
> Consider the following sequence of iova alloc and frees on
> 1GB of iova space
> 
> 1) alloc - 500MB
> 2) alloc - 12MB
> 3) alloc - 499MB
> 4) free -  12MB which was allocated in step 2
> 5) alloc - 13MB
> 
> After the above sequence we will have 12MB of free iova space and
> cached node will be pointing to the iova pfn of last alloc of 13MB
> which will be the lowest iova pfn of that iova space. Now if we get an
> alloc request of 2MB we just search from cached node and then look
> for lower iova pfn's for free iova and as they aren't any, iova alloc
> fails though there is 12MB of free iova space.
> 
> To avoid such iova search failures do a retry from the last rb tree node
> when iova search fails, this will search the entire tree and get an iova
> if its available.
> 
> Signed-off-by: Vijayanand Jitta <vjitta@codeaurora.org>
> ---
>   drivers/iommu/iova.c | 23 +++++++++++++++++------
>   1 file changed, 17 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
> index 49fc01f..4e77116 100644
> --- a/drivers/iommu/iova.c
> +++ b/drivers/iommu/iova.c
> @@ -184,8 +184,9 @@ static int __alloc_and_insert_iova_range(struct iova_domain *iovad,
>   	struct rb_node *curr, *prev;
>   	struct iova *curr_iova;
>   	unsigned long flags;
> -	unsigned long new_pfn;
> +	unsigned long new_pfn, low_pfn_new;
>   	unsigned long align_mask = ~0UL;
> +	unsigned long high_pfn = limit_pfn, low_pfn = iovad->start_pfn;
>   
>   	if (size_aligned)
>   		align_mask <<= fls_long(size - 1);
> @@ -198,15 +199,25 @@ static int __alloc_and_insert_iova_range(struct iova_domain *iovad,
>   
>   	curr = __get_cached_rbnode(iovad, limit_pfn);
>   	curr_iova = rb_entry(curr, struct iova, node);
> +	low_pfn_new = curr_iova->pfn_hi + 1;

Could we call "low_pfn_new" something like "retry_pfn" instead? This 
code already has unavoidable readability struggles with so many 
different "pfn"s in play, so having two different meanings of "new" 
really doesn't help.

Other than that, I think this looks OK (IIRC it's basically what I 
originally suggested), so with the naming tweaked,

Reviewed-by: Robin Murphy <robin.murphy@arm.com>

> +
> +retry:
>   	do {
> -		limit_pfn = min(limit_pfn, curr_iova->pfn_lo);
> -		new_pfn = (limit_pfn - size) & align_mask;
> +		high_pfn = min(high_pfn, curr_iova->pfn_lo);
> +		new_pfn = (high_pfn - size) & align_mask;
>   		prev = curr;
>   		curr = rb_prev(curr);
>   		curr_iova = rb_entry(curr, struct iova, node);
> -	} while (curr && new_pfn <= curr_iova->pfn_hi);
> -
> -	if (limit_pfn < size || new_pfn < iovad->start_pfn) {
> +	} while (curr && new_pfn <= curr_iova->pfn_hi && new_pfn >= low_pfn);
> +
> +	if (high_pfn < size || new_pfn < low_pfn) {
> +		if (low_pfn == iovad->start_pfn && low_pfn_new < limit_pfn) {
> +			high_pfn = limit_pfn;
> +			low_pfn = low_pfn_new;
> +			curr = &iovad->anchor.node;
> +			curr_iova = rb_entry(curr, struct iova, node);
> +			goto retry;
> +		}
>   		iovad->max32_alloc_size = size;
>   		goto iova32_full;
>   	}
> 
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH v2 2/2] iommu/iova: Free global iova rcache on iova alloc failure
  2020-08-20 12:49 ` [PATCH v2 2/2] iommu/iova: Free global iova rcache on iova alloc failure vjitta
@ 2020-09-18 14:41   ` Robin Murphy
  2020-09-28 12:41     ` Vijayanand Jitta
  0 siblings, 1 reply; 10+ messages in thread
From: Robin Murphy @ 2020-09-18 14:41 UTC (permalink / raw)
  To: vjitta, joro, iommu, linux-kernel; +Cc: vinmenon, kernel-team

On 2020-08-20 13:49, vjitta@codeaurora.org wrote:
> From: Vijayanand Jitta <vjitta@codeaurora.org>
> 
> When ever an iova alloc request fails we free the iova
> ranges present in the percpu iova rcaches and then retry
> but the global iova rcache is not freed as a result we could
> still see iova alloc failure even after retry as global
> rcache is holding the iova's which can cause fragmentation.
> So, free the global iova rcache as well and then go for the
> retry.
> 
> Signed-off-by: Vijayanand Jitta <vjitta@codeaurora.org>
> ---
>   drivers/iommu/iova.c | 23 +++++++++++++++++++++++
>   include/linux/iova.h |  6 ++++++
>   2 files changed, 29 insertions(+)
> 
> diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
> index 4e77116..5836c87 100644
> --- a/drivers/iommu/iova.c
> +++ b/drivers/iommu/iova.c
> @@ -442,6 +442,7 @@ struct iova *find_iova(struct iova_domain *iovad, unsigned long pfn)
>   		flush_rcache = false;
>   		for_each_online_cpu(cpu)
>   			free_cpu_cached_iovas(cpu, iovad);
> +		free_global_cached_iovas(iovad);
>   		goto retry;
>   	}
>   
> @@ -1055,5 +1056,27 @@ void free_cpu_cached_iovas(unsigned int cpu, struct iova_domain *iovad)
>   	}
>   }
>   
> +/*
> + * free all the IOVA ranges of global cache
> + */
> +void free_global_cached_iovas(struct iova_domain *iovad)

As John pointed out last time, this should be static and the header 
changes dropped.

(TBH we should probably register our own hotplug notifier instance for a 
flush queue, so that external code has no need to poke at the per-CPU 
caches either)

Robin.

> +{
> +	struct iova_rcache *rcache;
> +	unsigned long flags;
> +	int i, j;
> +
> +	for (i = 0; i < IOVA_RANGE_CACHE_MAX_SIZE; ++i) {
> +		rcache = &iovad->rcaches[i];
> +		spin_lock_irqsave(&rcache->lock, flags);
> +		for (j = 0; j < rcache->depot_size; ++j) {
> +			iova_magazine_free_pfns(rcache->depot[j], iovad);
> +			iova_magazine_free(rcache->depot[j]);
> +			rcache->depot[j] = NULL;
> +		}
> +		rcache->depot_size = 0;
> +		spin_unlock_irqrestore(&rcache->lock, flags);
> +	}
> +}
> +
>   MODULE_AUTHOR("Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>");
>   MODULE_LICENSE("GPL");
> diff --git a/include/linux/iova.h b/include/linux/iova.h
> index a0637ab..a905726 100644
> --- a/include/linux/iova.h
> +++ b/include/linux/iova.h
> @@ -163,6 +163,7 @@ int init_iova_flush_queue(struct iova_domain *iovad,
>   struct iova *split_and_remove_iova(struct iova_domain *iovad,
>   	struct iova *iova, unsigned long pfn_lo, unsigned long pfn_hi);
>   void free_cpu_cached_iovas(unsigned int cpu, struct iova_domain *iovad);
> +void free_global_cached_iovas(struct iova_domain *iovad);
>   #else
>   static inline int iova_cache_get(void)
>   {
> @@ -270,6 +271,11 @@ static inline void free_cpu_cached_iovas(unsigned int cpu,
>   					 struct iova_domain *iovad)
>   {
>   }
> +
> +static inline void free_global_cached_iovas(struct iova_domain *iovad)
> +{
> +}
> +
>   #endif
>   
>   #endif
> 
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH v2 2/2] iommu/iova: Free global iova rcache on iova alloc failure
  2020-09-18 14:41   ` Robin Murphy
@ 2020-09-28 12:41     ` Vijayanand Jitta
  2020-09-30  5:59       ` Vijayanand Jitta
  0 siblings, 1 reply; 10+ messages in thread
From: Vijayanand Jitta @ 2020-09-28 12:41 UTC (permalink / raw)
  To: Robin Murphy, joro, iommu, linux-kernel; +Cc: vinmenon, kernel-team



On 9/18/2020 8:11 PM, Robin Murphy wrote:
> On 2020-08-20 13:49, vjitta@codeaurora.org wrote:
>> From: Vijayanand Jitta <vjitta@codeaurora.org>
>>
>> When ever an iova alloc request fails we free the iova
>> ranges present in the percpu iova rcaches and then retry
>> but the global iova rcache is not freed as a result we could
>> still see iova alloc failure even after retry as global
>> rcache is holding the iova's which can cause fragmentation.
>> So, free the global iova rcache as well and then go for the
>> retry.
>>
>> Signed-off-by: Vijayanand Jitta <vjitta@codeaurora.org>
>> ---
>>   drivers/iommu/iova.c | 23 +++++++++++++++++++++++
>>   include/linux/iova.h |  6 ++++++
>>   2 files changed, 29 insertions(+)
>>
>> diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
>> index 4e77116..5836c87 100644
>> --- a/drivers/iommu/iova.c
>> +++ b/drivers/iommu/iova.c
>> @@ -442,6 +442,7 @@ struct iova *find_iova(struct iova_domain *iovad,
>> unsigned long pfn)
>>           flush_rcache = false;
>>           for_each_online_cpu(cpu)
>>               free_cpu_cached_iovas(cpu, iovad);
>> +        free_global_cached_iovas(iovad);
>>           goto retry;
>>       }
>>   @@ -1055,5 +1056,27 @@ void free_cpu_cached_iovas(unsigned int cpu,
>> struct iova_domain *iovad)
>>       }
>>   }
>>   +/*
>> + * free all the IOVA ranges of global cache
>> + */
>> +void free_global_cached_iovas(struct iova_domain *iovad)
> 
> As John pointed out last time, this should be static and the header
> changes dropped.
> 
> (TBH we should probably register our own hotplug notifier instance for a
> flush queue, so that external code has no need to poke at the per-CPU
> caches either)
> 
> Robin.
> 

Right, I have made it static and dropped header changes in v3.
can you please review that.

Thanks,
Vijay
>> +{
>> +    struct iova_rcache *rcache;
>> +    unsigned long flags;
>> +    int i, j;
>> +
>> +    for (i = 0; i < IOVA_RANGE_CACHE_MAX_SIZE; ++i) {
>> +        rcache = &iovad->rcaches[i];
>> +        spin_lock_irqsave(&rcache->lock, flags);
>> +        for (j = 0; j < rcache->depot_size; ++j) {
>> +            iova_magazine_free_pfns(rcache->depot[j], iovad);
>> +            iova_magazine_free(rcache->depot[j]);
>> +            rcache->depot[j] = NULL;
>> +        }
>> +        rcache->depot_size = 0;
>> +        spin_unlock_irqrestore(&rcache->lock, flags);
>> +    }
>> +}
>> +
>>   MODULE_AUTHOR("Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>");
>>   MODULE_LICENSE("GPL");
>> diff --git a/include/linux/iova.h b/include/linux/iova.h
>> index a0637ab..a905726 100644
>> --- a/include/linux/iova.h
>> +++ b/include/linux/iova.h
>> @@ -163,6 +163,7 @@ int init_iova_flush_queue(struct iova_domain *iovad,
>>   struct iova *split_and_remove_iova(struct iova_domain *iovad,
>>       struct iova *iova, unsigned long pfn_lo, unsigned long pfn_hi);
>>   void free_cpu_cached_iovas(unsigned int cpu, struct iova_domain
>> *iovad);
>> +void free_global_cached_iovas(struct iova_domain *iovad);
>>   #else
>>   static inline int iova_cache_get(void)
>>   {
>> @@ -270,6 +271,11 @@ static inline void free_cpu_cached_iovas(unsigned
>> int cpu,
>>                        struct iova_domain *iovad)
>>   {
>>   }
>> +
>> +static inline void free_global_cached_iovas(struct iova_domain *iovad)
>> +{
>> +}
>> +
>>   #endif
>>     #endif
>>

-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a
member of Code Aurora Forum, hosted by The Linux Foundation
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails
  2020-09-18 14:18 ` Robin Murphy
@ 2020-09-30  5:48   ` Vijayanand Jitta
  0 siblings, 0 replies; 10+ messages in thread
From: Vijayanand Jitta @ 2020-09-30  5:48 UTC (permalink / raw)
  To: Robin Murphy, joro, iommu, linux-kernel; +Cc: vinmenon, kernel-team



On 9/18/2020 7:48 PM, Robin Murphy wrote:
> On 2020-08-20 13:49, vjitta@codeaurora.org wrote:
>> From: Vijayanand Jitta <vjitta@codeaurora.org>
>>
>> When ever a new iova alloc request comes iova is always searched
>> from the cached node and the nodes which are previous to cached
>> node. So, even if there is free iova space available in the nodes
>> which are next to the cached node iova allocation can still fail
>> because of this approach.
>>
>> Consider the following sequence of iova alloc and frees on
>> 1GB of iova space
>>
>> 1) alloc - 500MB
>> 2) alloc - 12MB
>> 3) alloc - 499MB
>> 4) free -  12MB which was allocated in step 2
>> 5) alloc - 13MB
>>
>> After the above sequence we will have 12MB of free iova space and
>> cached node will be pointing to the iova pfn of last alloc of 13MB
>> which will be the lowest iova pfn of that iova space. Now if we get an
>> alloc request of 2MB we just search from cached node and then look
>> for lower iova pfn's for free iova and as they aren't any, iova alloc
>> fails though there is 12MB of free iova space.
>>
>> To avoid such iova search failures do a retry from the last rb tree node
>> when iova search fails, this will search the entire tree and get an iova
>> if its available.
>>
>> Signed-off-by: Vijayanand Jitta <vjitta@codeaurora.org>
>> ---
>>   drivers/iommu/iova.c | 23 +++++++++++++++++------
>>   1 file changed, 17 insertions(+), 6 deletions(-)
>>
>> diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
>> index 49fc01f..4e77116 100644
>> --- a/drivers/iommu/iova.c
>> +++ b/drivers/iommu/iova.c
>> @@ -184,8 +184,9 @@ static int __alloc_and_insert_iova_range(struct
>> iova_domain *iovad,
>>       struct rb_node *curr, *prev;
>>       struct iova *curr_iova;
>>       unsigned long flags;
>> -    unsigned long new_pfn;
>> +    unsigned long new_pfn, low_pfn_new;
>>       unsigned long align_mask = ~0UL;
>> +    unsigned long high_pfn = limit_pfn, low_pfn = iovad->start_pfn;
>>         if (size_aligned)
>>           align_mask <<= fls_long(size - 1);
>> @@ -198,15 +199,25 @@ static int __alloc_and_insert_iova_range(struct
>> iova_domain *iovad,
>>         curr = __get_cached_rbnode(iovad, limit_pfn);
>>       curr_iova = rb_entry(curr, struct iova, node);
>> +    low_pfn_new = curr_iova->pfn_hi + 1;
> 
> Could we call "low_pfn_new" something like "retry_pfn" instead? This
> code already has unavoidable readability struggles with so many
> different "pfn"s in play, so having two different meanings of "new"
> really doesn't help.
> 
> Other than that, I think this looks OK (IIRC it's basically what I
> originally suggested), so with the naming tweaked,
> 
> Reviewed-by: Robin Murphy <robin.murphy@arm.com>
> 

Thanks for review, I have renamed it to retry_pfn in v4.

Thanks,
Vijay
>> +
>> +retry:
>>       do {
>> -        limit_pfn = min(limit_pfn, curr_iova->pfn_lo);
>> -        new_pfn = (limit_pfn - size) & align_mask;
>> +        high_pfn = min(high_pfn, curr_iova->pfn_lo);
>> +        new_pfn = (high_pfn - size) & align_mask;
>>           prev = curr;
>>           curr = rb_prev(curr);
>>           curr_iova = rb_entry(curr, struct iova, node);
>> -    } while (curr && new_pfn <= curr_iova->pfn_hi);
>> -
>> -    if (limit_pfn < size || new_pfn < iovad->start_pfn) {
>> +    } while (curr && new_pfn <= curr_iova->pfn_hi && new_pfn >=
>> low_pfn);
>> +
>> +    if (high_pfn < size || new_pfn < low_pfn) {
>> +        if (low_pfn == iovad->start_pfn && low_pfn_new < limit_pfn) {
>> +            high_pfn = limit_pfn;
>> +            low_pfn = low_pfn_new;
>> +            curr = &iovad->anchor.node;
>> +            curr_iova = rb_entry(curr, struct iova, node);
>> +            goto retry;
>> +        }
>>           iovad->max32_alloc_size = size;
>>           goto iova32_full;
>>       }
>>

-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a
member of Code Aurora Forum, hosted by The Linux Foundation
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH v2 2/2] iommu/iova: Free global iova rcache on iova alloc failure
  2020-09-28 12:41     ` Vijayanand Jitta
@ 2020-09-30  5:59       ` Vijayanand Jitta
  0 siblings, 0 replies; 10+ messages in thread
From: Vijayanand Jitta @ 2020-09-30  5:59 UTC (permalink / raw)
  To: Robin Murphy, joro, iommu, linux-kernel; +Cc: vinmenon, kernel-team



On 9/28/2020 6:11 PM, Vijayanand Jitta wrote:
> 
> 
> On 9/18/2020 8:11 PM, Robin Murphy wrote:
>> On 2020-08-20 13:49, vjitta@codeaurora.org wrote:
>>> From: Vijayanand Jitta <vjitta@codeaurora.org>
>>>
>>> When ever an iova alloc request fails we free the iova
>>> ranges present in the percpu iova rcaches and then retry
>>> but the global iova rcache is not freed as a result we could
>>> still see iova alloc failure even after retry as global
>>> rcache is holding the iova's which can cause fragmentation.
>>> So, free the global iova rcache as well and then go for the
>>> retry.
>>>
>>> Signed-off-by: Vijayanand Jitta <vjitta@codeaurora.org>
>>> ---
>>>   drivers/iommu/iova.c | 23 +++++++++++++++++++++++
>>>   include/linux/iova.h |  6 ++++++
>>>   2 files changed, 29 insertions(+)
>>>
>>> diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
>>> index 4e77116..5836c87 100644
>>> --- a/drivers/iommu/iova.c
>>> +++ b/drivers/iommu/iova.c
>>> @@ -442,6 +442,7 @@ struct iova *find_iova(struct iova_domain *iovad,
>>> unsigned long pfn)
>>>           flush_rcache = false;
>>>           for_each_online_cpu(cpu)
>>>               free_cpu_cached_iovas(cpu, iovad);
>>> +        free_global_cached_iovas(iovad);
>>>           goto retry;
>>>       }
>>>   @@ -1055,5 +1056,27 @@ void free_cpu_cached_iovas(unsigned int cpu,
>>> struct iova_domain *iovad)
>>>       }
>>>   }
>>>   +/*
>>> + * free all the IOVA ranges of global cache
>>> + */
>>> +void free_global_cached_iovas(struct iova_domain *iovad)
>>
>> As John pointed out last time, this should be static and the header
>> changes dropped.
>>
>> (TBH we should probably register our own hotplug notifier instance for a
>> flush queue, so that external code has no need to poke at the per-CPU
>> caches either)
>>
>> Robin.
>>
> 
> Right, I have made it static and dropped header changes in v3.
> can you please review that.
> 
> Thanks,
> Vijay

Please review v4 instead of v3, I have updated other patch as well in v4.

Thanks,
Vijay
>>> +{
>>> +    struct iova_rcache *rcache;
>>> +    unsigned long flags;
>>> +    int i, j;
>>> +
>>> +    for (i = 0; i < IOVA_RANGE_CACHE_MAX_SIZE; ++i) {
>>> +        rcache = &iovad->rcaches[i];
>>> +        spin_lock_irqsave(&rcache->lock, flags);
>>> +        for (j = 0; j < rcache->depot_size; ++j) {
>>> +            iova_magazine_free_pfns(rcache->depot[j], iovad);
>>> +            iova_magazine_free(rcache->depot[j]);
>>> +            rcache->depot[j] = NULL;
>>> +        }
>>> +        rcache->depot_size = 0;
>>> +        spin_unlock_irqrestore(&rcache->lock, flags);
>>> +    }
>>> +}
>>> +
>>>   MODULE_AUTHOR("Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>");
>>>   MODULE_LICENSE("GPL");
>>> diff --git a/include/linux/iova.h b/include/linux/iova.h
>>> index a0637ab..a905726 100644
>>> --- a/include/linux/iova.h
>>> +++ b/include/linux/iova.h
>>> @@ -163,6 +163,7 @@ int init_iova_flush_queue(struct iova_domain *iovad,
>>>   struct iova *split_and_remove_iova(struct iova_domain *iovad,
>>>       struct iova *iova, unsigned long pfn_lo, unsigned long pfn_hi);
>>>   void free_cpu_cached_iovas(unsigned int cpu, struct iova_domain
>>> *iovad);
>>> +void free_global_cached_iovas(struct iova_domain *iovad);
>>>   #else
>>>   static inline int iova_cache_get(void)
>>>   {
>>> @@ -270,6 +271,11 @@ static inline void free_cpu_cached_iovas(unsigned
>>> int cpu,
>>>                        struct iova_domain *iovad)
>>>   {
>>>   }
>>> +
>>> +static inline void free_global_cached_iovas(struct iova_domain *iovad)
>>> +{
>>> +}
>>> +
>>>   #endif
>>>     #endif
>>>
> 

-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a
member of Code Aurora Forum, hosted by The Linux Foundation
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2020-09-30  5:59 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-08-20 12:49 [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails vjitta
2020-08-20 12:49 ` [PATCH v2 2/2] iommu/iova: Free global iova rcache on iova alloc failure vjitta
2020-09-18 14:41   ` Robin Murphy
2020-09-28 12:41     ` Vijayanand Jitta
2020-09-30  5:59       ` Vijayanand Jitta
2020-08-28  7:31 ` [PATCH v2 1/2] iommu/iova: Retry from last rb tree node if iova search fails Vijayanand Jitta
2020-09-14  4:50   ` Vijayanand Jitta
2020-09-18  8:18     ` Joerg Roedel
2020-09-18 14:18 ` Robin Murphy
2020-09-30  5:48   ` Vijayanand Jitta

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).