Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1684380 > unrolled thread

[PATCH v2] irqchip: gicv3-its: Use NUMA aware memory allocation for ITS tables

Started byShanker Donthineni <shankerd@codeaurora.org>
First post2017-07-10 18:00 +0200
Last post2017-07-10 18:10 +0200
Articles 2 — 1 participant

Back to article view | Back to linux.kernel

Contents

  [PATCH v2] irqchip: gicv3-its: Use NUMA aware memory allocation for ITS tables Shanker Donthineni <shankerd@codeaurora.org> - 2017-07-10 18:00 +0200
    Re: [PATCH v2] irqchip: gicv3-its: Use NUMA aware memory allocation  for ITS tables Shanker Donthineni <shankerd@codeaurora.org> - 2017-07-10 18:10 +0200

#1684380 — [PATCH v2] irqchip: gicv3-its: Use NUMA aware memory allocation for ITS tables

FromShanker Donthineni <shankerd@codeaurora.org>
Date2017-07-10 18:00 +0200
Subject[PATCH v2] irqchip: gicv3-its: Use NUMA aware memory allocation for ITS tables
Message-ID<u1Jgl-6Vr-11@gated-at.bofh.it>
The NUMA node information is visible to ITS driver but not being used
other than handling hardware errata. ITS/GICR hardware accesses to the
local NUMA node is usually quicker than the remote NUMA node. How slow
the remote NUMA accesses are depends on the implementation details.

This patch allocates memory for ITS management tables and command
queue from the corresponding NUMA node using the appropriate NUMA
aware functions. This change improves the performance of the ITS
tables read latency on systems where it has more than one ITS block,
and with the slower inter node accesses.

Signed-off-by: Shanker Donthineni <shankerd@codeaurora.org>
Tested-by: Ganapatrao Kulkarni <ganapatrao.kulkarni@cavium.com>
---
 drivers/irqchip/irq-gic-v3-its.c | 36 ++++++++++++++++++++----------------
 1 file changed, 20 insertions(+), 16 deletions(-)

diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c
index 45ea1933..40442fb 100644
--- a/drivers/irqchip/irq-gic-v3-its.c
+++ b/drivers/irqchip/irq-gic-v3-its.c
@@ -858,8 +858,8 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser,
 	u64 val = its_read_baser(its, baser);
 	u64 esz = GITS_BASER_ENTRY_SIZE(val);
 	u64 type = GITS_BASER_TYPE(val);
+	struct page *page;
 	u32 alloc_pages;
-	void *base;
 	u64 tmp;
 
 retry_alloc_baser:
@@ -872,12 +872,12 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser,
 		order = get_order(GITS_BASER_PAGES_MAX * psz);
 	}
 
-	base = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, order);
-	if (!base)
+	page = alloc_pages_node(its->numa_node, GFP_KERNEL | __GFP_ZERO, order);
+	if (!page)
 		return -ENOMEM;
 
 retry_baser:
-	val = (virt_to_phys(base)				 |
+	val = (page_to_phys(page)				 |
 		(type << GITS_BASER_TYPE_SHIFT)			 |
 		((esz - 1) << GITS_BASER_ENTRY_SIZE_SHIFT)	 |
 		((alloc_pages - 1) << GITS_BASER_PAGES_SHIFT)	 |
@@ -913,7 +913,8 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser,
 		shr = tmp & GITS_BASER_SHAREABILITY_MASK;
 		if (!shr) {
 			cache = GITS_BASER_nC;
-			gic_flush_dcache_to_poc(base, PAGE_ORDER_TO_SIZE(order));
+			gic_flush_dcache_to_poc(page_to_virt(page),
+						PAGE_ORDER_TO_SIZE(order));
 		}
 		goto retry_baser;
 	}
@@ -924,7 +925,7 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser,
 		 * size and retry. If we reach 4K, then
 		 * something is horribly wrong...
 		 */
-		free_pages((unsigned long)base, order);
+		__free_pages(page, order);
 		baser->base = NULL;
 
 		switch (psz) {
@@ -941,19 +942,19 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser,
 		pr_err("ITS@%pa: %s doesn't stick: %llx %llx\n",
 		       &its->phys_base, its_base_type_string[type],
 		       val, tmp);
-		free_pages((unsigned long)base, order);
+		__free_pages(page, order);
 		return -ENXIO;
 	}
 
 	baser->order = order;
-	baser->base = base;
+	baser->base = page_to_virt(page);
 	baser->psz = psz;
 	tmp = indirect ? GITS_LVL1_ENTRY_SIZE : esz;
 
 	pr_info("ITS@%pa: allocated %d %s @%lx (%s, esz %d, psz %dK, shr %d)\n",
 		&its->phys_base, (int)(PAGE_ORDER_TO_SIZE(order) / (int)tmp),
 		its_base_type_string[type],
-		(unsigned long)virt_to_phys(base),
+		(unsigned long)page_to_phys(page),
 		indirect ? "indirect" : "flat", (int)esz,
 		psz / SZ_1K, (int)shr >> GITS_BASER_SHAREABILITY_SHIFT);
 
@@ -1017,7 +1018,7 @@ static void its_free_tables(struct its_node *its)
 
 	for (i = 0; i < GITS_BASER_NR_REGS; i++) {
 		if (its->tables[i].base) {
-			free_pages((unsigned long)its->tables[i].base,
+			__free_pages(virt_to_page(its->tables[i].base),
 				   its->tables[i].order);
 			its->tables[i].base = NULL;
 		}
@@ -1284,7 +1285,8 @@ static bool its_alloc_device_table(struct its_node *its, u32 dev_id)
 
 	/* Allocate memory for 2nd level table */
 	if (!table[idx]) {
-		page = alloc_pages(GFP_KERNEL | __GFP_ZERO, get_order(baser->psz));
+		page = alloc_pages_node(its->numa_node, GFP_KERNEL | __GFP_ZERO,
+					get_order(baser->psz));
 		if (!page)
 			return false;
 
@@ -1330,7 +1332,7 @@ static struct its_device *its_create_device(struct its_node *its, u32 dev_id,
 	nr_ites = max(2UL, roundup_pow_of_two(nvecs));
 	sz = nr_ites * its->ite_size;
 	sz = max(sz, ITS_ITT_ALIGN) + ITS_ITT_ALIGN - 1;
-	itt = kzalloc(sz, GFP_KERNEL);
+	itt = kzalloc_node(sz, GFP_KERNEL, its->numa_node);
 	lpi_map = its_lpi_alloc_chunks(nvecs, &lpi_base, &nr_lpis);
 	if (lpi_map)
 		col_map = kzalloc(sizeof(*col_map) * nr_lpis, GFP_KERNEL);
@@ -1675,6 +1677,7 @@ static int __init its_probe_one(struct resource *res,
 {
 	struct its_node *its;
 	void __iomem *its_base;
+	struct page *page;
 	u32 val;
 	u64 baser, tmp;
 	int err;
@@ -1714,12 +1717,13 @@ static int __init its_probe_one(struct resource *res,
 	its->ite_size = ((gic_read_typer(its_base + GITS_TYPER) >> 4) & 0xf) + 1;
 	its->numa_node = numa_node;
 
-	its->cmd_base = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO,
-						get_order(ITS_CMD_QUEUE_SZ));
-	if (!its->cmd_base) {
+	page = alloc_pages_node(its->numa_node, GFP_KERNEL | __GFP_ZERO,
+				get_order(ITS_CMD_QUEUE_SZ));
+	if (!page) {
 		err = -ENOMEM;
 		goto out_free_its;
 	}
+	its->cmd_base = page_to_virt(page);
 	its->cmd_write = its->cmd_base;
 
 	its_enable_quirks(its);
@@ -1773,7 +1777,7 @@ static int __init its_probe_one(struct resource *res,
 out_free_tables:
 	its_free_tables(its);
 out_free_cmd:
-	free_pages((unsigned long)its->cmd_base, get_order(ITS_CMD_QUEUE_SZ));
+	__free_pages(virt_to_page(its->cmd_base), get_order(ITS_CMD_QUEUE_SZ));
 out_free_its:
 	kfree(its);
 out_unmap:
-- 
Qualcomm Datacenter Technologies, Inc. on behalf of the Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.

[toc] | [next] | [standalone]

#1684392 — Re: [PATCH v2] irqchip: gicv3-its: Use NUMA aware memory allocation for ITS tables

FromShanker Donthineni <shankerd@codeaurora.org>
Date2017-07-10 18:10 +0200
SubjectRe: [PATCH v2] irqchip: gicv3-its: Use NUMA aware memory allocation for ITS tables
Message-ID<u1Jq2-7dI-21@gated-at.bofh.it>
In reply to#1684380

On 07/10/2017 10:53 AM, Shanker Donthineni wrote:
> The NUMA node information is visible to ITS driver but not being used
> other than handling hardware errata. ITS/GICR hardware accesses to the
> local NUMA node is usually quicker than the remote NUMA node. How slow
> the remote NUMA accesses are depends on the implementation details.
> 
> This patch allocates memory for ITS management tables and command
> queue from the corresponding NUMA node using the appropriate NUMA
> aware functions. This change improves the performance of the ITS
> tables read latency on systems where it has more than one ITS block,
> and with the slower inter node accesses.
> 
> Signed-off-by: Shanker Donthineni <shankerd@codeaurora.org>
> Tested-by: Ganapatrao Kulkarni <ganapatrao.kulkarni@cavium.com>
> ---
Sorry forgot to include v2 changes:
  - Edited commit text.
  - Added Ganapatrao's tested-by.
 
>  drivers/irqchip/irq-gic-v3-its.c | 36 ++++++++++++++++++++----------------
>  1 file changed, 20 insertions(+), 16 deletions(-)
> 
> diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c
> index 45ea1933..40442fb 100644
> --- a/drivers/irqchip/irq-gic-v3-its.c
> +++ b/drivers/irqchip/irq-gic-v3-its.c
> @@ -858,8 +858,8 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser,
>  	u64 val = its_read_baser(its, baser);
>  	u64 esz = GITS_BASER_ENTRY_SIZE(val);
>  	u64 type = GITS_BASER_TYPE(val);
> +	struct page *page;
>  	u32 alloc_pages;
> -	void *base;
>  	u64 tmp;
>  
>  retry_alloc_baser:
> @@ -872,12 +872,12 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser,
>  		order = get_order(GITS_BASER_PAGES_MAX * psz);
>  	}
>  
> -	base = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, order);
> -	if (!base)
> +	page = alloc_pages_node(its->numa_node, GFP_KERNEL | __GFP_ZERO, order);
> +	if (!page)
>  		return -ENOMEM;
>  
>  retry_baser:
> -	val = (virt_to_phys(base)				 |
> +	val = (page_to_phys(page)				 |
>  		(type << GITS_BASER_TYPE_SHIFT)			 |
>  		((esz - 1) << GITS_BASER_ENTRY_SIZE_SHIFT)	 |
>  		((alloc_pages - 1) << GITS_BASER_PAGES_SHIFT)	 |
> @@ -913,7 +913,8 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser,
>  		shr = tmp & GITS_BASER_SHAREABILITY_MASK;
>  		if (!shr) {
>  			cache = GITS_BASER_nC;
> -			gic_flush_dcache_to_poc(base, PAGE_ORDER_TO_SIZE(order));
> +			gic_flush_dcache_to_poc(page_to_virt(page),
> +						PAGE_ORDER_TO_SIZE(order));
>  		}
>  		goto retry_baser;
>  	}
> @@ -924,7 +925,7 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser,
>  		 * size and retry. If we reach 4K, then
>  		 * something is horribly wrong...
>  		 */
> -		free_pages((unsigned long)base, order);
> +		__free_pages(page, order);
>  		baser->base = NULL;
>  
>  		switch (psz) {
> @@ -941,19 +942,19 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser,
>  		pr_err("ITS@%pa: %s doesn't stick: %llx %llx\n",
>  		       &its->phys_base, its_base_type_string[type],
>  		       val, tmp);
> -		free_pages((unsigned long)base, order);
> +		__free_pages(page, order);
>  		return -ENXIO;
>  	}
>  
>  	baser->order = order;
> -	baser->base = base;
> +	baser->base = page_to_virt(page);
>  	baser->psz = psz;
>  	tmp = indirect ? GITS_LVL1_ENTRY_SIZE : esz;
>  
>  	pr_info("ITS@%pa: allocated %d %s @%lx (%s, esz %d, psz %dK, shr %d)\n",
>  		&its->phys_base, (int)(PAGE_ORDER_TO_SIZE(order) / (int)tmp),
>  		its_base_type_string[type],
> -		(unsigned long)virt_to_phys(base),
> +		(unsigned long)page_to_phys(page),
>  		indirect ? "indirect" : "flat", (int)esz,
>  		psz / SZ_1K, (int)shr >> GITS_BASER_SHAREABILITY_SHIFT);
>  
> @@ -1017,7 +1018,7 @@ static void its_free_tables(struct its_node *its)
>  
>  	for (i = 0; i < GITS_BASER_NR_REGS; i++) {
>  		if (its->tables[i].base) {
> -			free_pages((unsigned long)its->tables[i].base,
> +			__free_pages(virt_to_page(its->tables[i].base),
>  				   its->tables[i].order);
>  			its->tables[i].base = NULL;
>  		}
> @@ -1284,7 +1285,8 @@ static bool its_alloc_device_table(struct its_node *its, u32 dev_id)
>  
>  	/* Allocate memory for 2nd level table */
>  	if (!table[idx]) {
> -		page = alloc_pages(GFP_KERNEL | __GFP_ZERO, get_order(baser->psz));
> +		page = alloc_pages_node(its->numa_node, GFP_KERNEL | __GFP_ZERO,
> +					get_order(baser->psz));
>  		if (!page)
>  			return false;
>  
> @@ -1330,7 +1332,7 @@ static struct its_device *its_create_device(struct its_node *its, u32 dev_id,
>  	nr_ites = max(2UL, roundup_pow_of_two(nvecs));
>  	sz = nr_ites * its->ite_size;
>  	sz = max(sz, ITS_ITT_ALIGN) + ITS_ITT_ALIGN - 1;
> -	itt = kzalloc(sz, GFP_KERNEL);
> +	itt = kzalloc_node(sz, GFP_KERNEL, its->numa_node);
>  	lpi_map = its_lpi_alloc_chunks(nvecs, &lpi_base, &nr_lpis);
>  	if (lpi_map)
>  		col_map = kzalloc(sizeof(*col_map) * nr_lpis, GFP_KERNEL);
> @@ -1675,6 +1677,7 @@ static int __init its_probe_one(struct resource *res,
>  {
>  	struct its_node *its;
>  	void __iomem *its_base;
> +	struct page *page;
>  	u32 val;
>  	u64 baser, tmp;
>  	int err;
> @@ -1714,12 +1717,13 @@ static int __init its_probe_one(struct resource *res,
>  	its->ite_size = ((gic_read_typer(its_base + GITS_TYPER) >> 4) & 0xf) + 1;
>  	its->numa_node = numa_node;
>  
> -	its->cmd_base = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO,
> -						get_order(ITS_CMD_QUEUE_SZ));
> -	if (!its->cmd_base) {
> +	page = alloc_pages_node(its->numa_node, GFP_KERNEL | __GFP_ZERO,
> +				get_order(ITS_CMD_QUEUE_SZ));
> +	if (!page) {
>  		err = -ENOMEM;
>  		goto out_free_its;
>  	}
> +	its->cmd_base = page_to_virt(page);
>  	its->cmd_write = its->cmd_base;
>  
>  	its_enable_quirks(its);
> @@ -1773,7 +1777,7 @@ static int __init its_probe_one(struct resource *res,
>  out_free_tables:
>  	its_free_tables(its);
>  out_free_cmd:
> -	free_pages((unsigned long)its->cmd_base, get_order(ITS_CMD_QUEUE_SZ));
> +	__free_pages(virt_to_page(its->cmd_base), get_order(ITS_CMD_QUEUE_SZ));
>  out_free_its:
>  	kfree(its);
>  out_unmap:
> 

-- 
Shanker Donthineni
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web