diff mbox series

[12/15] swiotlb: provide swiotlb_init variants that remap the buffer

Message ID 20220314073129.1862284-13-hch@lst.de (mailing list archive)
State Superseded
Headers show
Series [01/15] dma-direct: use is_swiotlb_active in dma_direct_map_page | expand

Commit Message

Christoph Hellwig March 14, 2022, 7:31 a.m. UTC 
To shared more code between swiotlb and xen-swiotlb, offer a
swiotlb_init_remap interface and add a remap callback to
swiotlb_init_late that will allow Xen to remap the buffer the
buffer without duplicating much of the logic.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 arch/x86/pci/sta2x11-fixup.c |  2 +-
 include/linux/swiotlb.h      |  5 ++++-
 kernel/dma/swiotlb.c         | 38 +++++++++++++++++++++++++++++++++---
 3 files changed, 40 insertions(+), 5 deletions(-)

Comments

Boris Ostrovsky March 14, 2022, 10:39 p.m. UTC | #1 
On 3/14/22 3:31 AM, Christoph Hellwig wrote:
> -void __init swiotlb_init(bool addressing_limit, unsigned int flags)
> +void __init swiotlb_init_remap(bool addressing_limit, unsigned int flags,
> +		int (*remap)(void *tlb, unsigned long nslabs))
>   {
> -	size_t bytes = PAGE_ALIGN(default_nslabs << IO_TLB_SHIFT);
> +	unsigned long nslabs = default_nslabs;
> +	size_t bytes;
>   	void *tlb;
>   
>   	if (!addressing_limit && !swiotlb_force_bounce)
> @@ -271,12 +273,24 @@ void __init swiotlb_init(bool addressing_limit, unsigned int flags)
>   	 * allow to pick a location everywhere for hypervisors with guest
>   	 * memory encryption.
>   	 */
> +retry:
> +	bytes = PAGE_ALIGN(default_nslabs << IO_TLB_SHIFT);
>   	if (flags & SWIOTLB_ANY)
>   		tlb = memblock_alloc(bytes, PAGE_SIZE);
>   	else
>   		tlb = memblock_alloc_low(bytes, PAGE_SIZE);
>   	if (!tlb)
>   		goto fail;
> +	if (remap && remap(tlb, nslabs) < 0) {
> +		memblock_free(tlb, PAGE_ALIGN(bytes));
> +
> +		/* Min is 2MB */
> +		if (nslabs <= 1024)


This is IO_TLB_MIN_SLABS, isn't it? (Xen code didn't say so but that's what it meant to say I believe)


> +			panic("%s: Failed to remap %zu bytes\n",
> +			      __func__, bytes);
> +		nslabs = max(1024UL, ALIGN(nslabs >> 1, IO_TLB_SEGSIZE));
> +		goto retry;
> +	}
>
> @@ -303,6 +323,7 @@ int swiotlb_init_late(size_t size, gfp_t gfp_mask)
>   	if (swiotlb_force_disable)
>   		return 0;
>   
> +retry:
>   	order = get_order(nslabs << IO_TLB_SHIFT);
>   	nslabs = SLABS_PER_PAGE << order;
>   	bytes = nslabs << IO_TLB_SHIFT;
> @@ -317,6 +338,17 @@ int swiotlb_init_late(size_t size, gfp_t gfp_mask)
>   
>   	if (!vstart)
>   		return -ENOMEM;
> +	if (remap)
> +		rc = remap(vstart, nslabs);
> +	if (rc) {
> +		free_pages((unsigned long)vstart, order);
> +
> +		/* Min is 2MB */
> +		if (nslabs <= 1024)


Same here.


-boris


> +			return rc;
> +		nslabs = max(1024UL, ALIGN(nslabs >> 1, IO_TLB_SEGSIZE));
> +		goto retry;
> +	}
>   
>   	if (order != get_order(bytes)) {
>   		pr_warn("only able to allocate %ld MB\n",
Christoph Hellwig March 15, 2022, 6:36 a.m. UTC | #2 
On Mon, Mar 14, 2022 at 06:39:21PM -0400, Boris Ostrovsky wrote:
> This is IO_TLB_MIN_SLABS, isn't it? (Xen code didn't say so but that's what it meant to say I believe)

Yes, that makes much more sense.  I've switched the patch to use
IO_TLB_MIN_SLABS and drop the 2MB comment in both places.

Can I get a review with that fixed up?

---
From 153085bf3e6e69d676bef0fb96395a86fb8122f5 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Mon, 14 Mar 2022 08:02:57 +0100
Subject: swiotlb: provide swiotlb_init variants that remap the buffer

To shared more code between swiotlb and xen-swiotlb, offer a
swiotlb_init_remap interface and add a remap callback to
swiotlb_init_late that will allow Xen to remap the buffer the
buffer without duplicating much of the logic.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 arch/x86/pci/sta2x11-fixup.c |  2 +-
 include/linux/swiotlb.h      |  5 ++++-
 kernel/dma/swiotlb.c         | 36 +++++++++++++++++++++++++++++++++---
 3 files changed, 38 insertions(+), 5 deletions(-)

diff --git a/arch/x86/pci/sta2x11-fixup.c b/arch/x86/pci/sta2x11-fixup.c
index c7e6faf59a861..7368afc039987 100644
--- a/arch/x86/pci/sta2x11-fixup.c
+++ b/arch/x86/pci/sta2x11-fixup.c
@@ -57,7 +57,7 @@ static void sta2x11_new_instance(struct pci_dev *pdev)
 		int size = STA2X11_SWIOTLB_SIZE;
 		/* First instance: register your own swiotlb area */
 		dev_info(&pdev->dev, "Using SWIOTLB (size %i)\n", size);
-		if (swiotlb_init_late(size, GFP_DMA))
+		if (swiotlb_init_late(size, GFP_DMA, NULL))
 			dev_emerg(&pdev->dev, "init swiotlb failed\n");
 	}
 	list_add(&instance->list, &sta2x11_instance_list);
diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h
index ee655f2e4d28b..7b50c82f84ce9 100644
--- a/include/linux/swiotlb.h
+++ b/include/linux/swiotlb.h
@@ -36,8 +36,11 @@ struct scatterlist;
 
 int swiotlb_init_with_tbl(char *tlb, unsigned long nslabs, unsigned int flags);
 unsigned long swiotlb_size_or_default(void);
+void __init swiotlb_init_remap(bool addressing_limit, unsigned int flags,
+	int (*remap)(void *tlb, unsigned long nslabs));
+int swiotlb_init_late(size_t size, gfp_t gfp_mask,
+	int (*remap)(void *tlb, unsigned long nslabs));
 extern int swiotlb_late_init_with_tbl(char *tlb, unsigned long nslabs);
-int swiotlb_init_late(size_t size, gfp_t gfp_mask);
 extern void __init swiotlb_update_mem_attributes(void);
 
 phys_addr_t swiotlb_tbl_map_single(struct device *hwdev, phys_addr_t phys,
diff --git a/kernel/dma/swiotlb.c b/kernel/dma/swiotlb.c
index 79641c446d284..b3d4f24fb5f5e 100644
--- a/kernel/dma/swiotlb.c
+++ b/kernel/dma/swiotlb.c
@@ -256,9 +256,11 @@ int __init swiotlb_init_with_tbl(char *tlb, unsigned long nslabs,
  * Statically reserve bounce buffer space and initialize bounce buffer data
  * structures for the software IO TLB used to implement the DMA API.
  */
-void __init swiotlb_init(bool addressing_limit, unsigned int flags)
+void __init swiotlb_init_remap(bool addressing_limit, unsigned int flags,
+		int (*remap)(void *tlb, unsigned long nslabs))
 {
-	size_t bytes = PAGE_ALIGN(default_nslabs << IO_TLB_SHIFT);
+	unsigned long nslabs = default_nslabs;
+	size_t bytes;
 	void *tlb;
 
 	if (!addressing_limit && !swiotlb_force_bounce)
@@ -271,12 +273,23 @@ void __init swiotlb_init(bool addressing_limit, unsigned int flags)
 	 * allow to pick a location everywhere for hypervisors with guest
 	 * memory encryption.
 	 */
+retry:
+	bytes = PAGE_ALIGN(default_nslabs << IO_TLB_SHIFT);
 	if (flags & SWIOTLB_ANY)
 		tlb = memblock_alloc(bytes, PAGE_SIZE);
 	else
 		tlb = memblock_alloc_low(bytes, PAGE_SIZE);
 	if (!tlb)
 		goto fail;
+	if (remap && remap(tlb, nslabs) < 0) {
+		memblock_free(tlb, PAGE_ALIGN(bytes));
+
+		if (nslabs <= IO_TLB_MIN_SLABS)
+			panic("%s: Failed to remap %zu bytes\n",
+			      __func__, bytes);
+		nslabs = max(1024UL, ALIGN(nslabs >> 1, IO_TLB_SEGSIZE));
+		goto retry;
+	}
 	if (swiotlb_init_with_tbl(tlb, default_nslabs, flags))
 		goto fail_free_mem;
 	return;
@@ -287,12 +300,18 @@ void __init swiotlb_init(bool addressing_limit, unsigned int flags)
 	pr_warn("Cannot allocate buffer");
 }
 
+void __init swiotlb_init(bool addressing_limit, unsigned int flags)
+{
+	return swiotlb_init_remap(addressing_limit, flags, NULL);
+}
+
 /*
  * Systems with larger DMA zones (those that don't support ISA) can
  * initialize the swiotlb later using the slab allocator if needed.
  * This should be just like above, but with some error catching.
  */
-int swiotlb_init_late(size_t size, gfp_t gfp_mask)
+int swiotlb_init_late(size_t size, gfp_t gfp_mask,
+		int (*remap)(void *tlb, unsigned long nslabs))
 {
 	unsigned long nslabs = ALIGN(size >> IO_TLB_SHIFT, IO_TLB_SEGSIZE);
 	unsigned long bytes;
@@ -303,6 +322,7 @@ int swiotlb_init_late(size_t size, gfp_t gfp_mask)
 	if (swiotlb_force_disable)
 		return 0;
 
+retry:
 	order = get_order(nslabs << IO_TLB_SHIFT);
 	nslabs = SLABS_PER_PAGE << order;
 	bytes = nslabs << IO_TLB_SHIFT;
@@ -317,6 +337,16 @@ int swiotlb_init_late(size_t size, gfp_t gfp_mask)
 
 	if (!vstart)
 		return -ENOMEM;
+	if (remap)
+		rc = remap(vstart, nslabs);
+	if (rc) {
+		free_pages((unsigned long)vstart, order);
+ 
+		if (IO_TLB_MIN_SLABS <= 1024)
+			return rc;
+		nslabs = max(1024UL, ALIGN(nslabs >> 1, IO_TLB_SEGSIZE));
+		goto retry;
+	}
 
 	if (order != get_order(bytes)) {
 		pr_warn("only able to allocate %ld MB\n",
Boris Ostrovsky March 16, 2022, 12:39 a.m. UTC | #3 
On 3/15/22 2:36 AM, Christoph Hellwig wrote:

> @@ -271,12 +273,23 @@ void __init swiotlb_init(bool addressing_limit, unsigned int flags)
>   	 * allow to pick a location everywhere for hypervisors with guest
>   	 * memory encryption.
>   	 */
> +retry:
> +	bytes = PAGE_ALIGN(default_nslabs << IO_TLB_SHIFT);
>   	if (flags & SWIOTLB_ANY)
>   		tlb = memblock_alloc(bytes, PAGE_SIZE);
>   	else
>   		tlb = memblock_alloc_low(bytes, PAGE_SIZE);
>   	if (!tlb)
>   		goto fail;
> +	if (remap && remap(tlb, nslabs) < 0) {
> +		memblock_free(tlb, PAGE_ALIGN(bytes));
> +
> +		if (nslabs <= IO_TLB_MIN_SLABS)
> +			panic("%s: Failed to remap %zu bytes\n",
> +			      __func__, bytes);
> +		nslabs = max(1024UL, ALIGN(nslabs >> 1, IO_TLB_SEGSIZE));


I spoke with Konrad (who wrote the original patch --- f4b2f07b2ed9b469ead87e06fc2fc3d12663a725) and apparently the reason for 2MB was to optimize for Xen's slab allocator, it had nothing to do with IO_TLB_MIN_SLABS. Since this is now common code we should not expose Xen-specific optimizations here and smaller values will still work so IO_TLB_MIN_SLABS is fine.

I think this should be mentioned in the commit message though, probably best in the next patch where you switch to this code.

As far as the hunk above, I don't think we need the max() here: with IO_TLB_MIN_SLABS being 512 we may get stuck in an infinite loop. Something like

	nslabs = ALIGN(nslabs >> 1, IO_TLB_SEGSIZE);
	if (nslabs <= IO_TLB_MIN_SLABS)
		panic()

should be sufficient.


> +		goto retry;
> +	}
>   	if (swiotlb_init_with_tbl(tlb, default_nslabs, flags))
>   		goto fail_free_mem;
>   	return;
> @@ -287,12 +300,18 @@ void __init swiotlb_init(bool addressing_limit, unsigned int flags)
>   	pr_warn("Cannot allocate buffer");
>   }
>   
> +void __init swiotlb_init(bool addressing_limit, unsigned int flags)
> +{
> +	return swiotlb_init_remap(addressing_limit, flags, NULL);
> +}
> +
>   /*
>    * Systems with larger DMA zones (those that don't support ISA) can
>    * initialize the swiotlb later using the slab allocator if needed.
>    * This should be just like above, but with some error catching.
>    */
> -int swiotlb_init_late(size_t size, gfp_t gfp_mask)
> +int swiotlb_init_late(size_t size, gfp_t gfp_mask,
> +		int (*remap)(void *tlb, unsigned long nslabs))
>   {
>   	unsigned long nslabs = ALIGN(size >> IO_TLB_SHIFT, IO_TLB_SEGSIZE);
>   	unsigned long bytes;
> @@ -303,6 +322,7 @@ int swiotlb_init_late(size_t size, gfp_t gfp_mask)
>   	if (swiotlb_force_disable)
>   		return 0;
>   
> +retry:
>   	order = get_order(nslabs << IO_TLB_SHIFT);
>   	nslabs = SLABS_PER_PAGE << order;
>   	bytes = nslabs << IO_TLB_SHIFT;
> @@ -317,6 +337,16 @@ int swiotlb_init_late(size_t size, gfp_t gfp_mask)
>   
>   	if (!vstart)
>   		return -ENOMEM;
> +	if (remap)
> +		rc = remap(vstart, nslabs);
> +	if (rc) {
> +		free_pages((unsigned long)vstart, order);
> +
> +		if (IO_TLB_MIN_SLABS <= 1024)
> +			return rc;
> +		nslabs = max(1024UL, ALIGN(nslabs >> 1, IO_TLB_SEGSIZE));


Same here. (The 'if' check above is wrong anyway).

Patches 13 and 14 look good.


-boris



> +		goto retry;
> +	}
>   
>   	if (order != get_order(bytes)) {
>   		pr_warn("only able to allocate %ld MB\n",
diff mbox series

Patch

diff --git a/arch/x86/pci/sta2x11-fixup.c b/arch/x86/pci/sta2x11-fixup.c
index c7e6faf59a861..7368afc039987 100644
--- a/arch/x86/pci/sta2x11-fixup.c
+++ b/arch/x86/pci/sta2x11-fixup.c
@@ -57,7 +57,7 @@  static void sta2x11_new_instance(struct pci_dev *pdev)
 		int size = STA2X11_SWIOTLB_SIZE;
 		/* First instance: register your own swiotlb area */
 		dev_info(&pdev->dev, "Using SWIOTLB (size %i)\n", size);
-		if (swiotlb_init_late(size, GFP_DMA))
+		if (swiotlb_init_late(size, GFP_DMA, NULL))
 			dev_emerg(&pdev->dev, "init swiotlb failed\n");
 	}
 	list_add(&instance->list, &sta2x11_instance_list);
diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h
index ee655f2e4d28b..7b50c82f84ce9 100644
--- a/include/linux/swiotlb.h
+++ b/include/linux/swiotlb.h
@@ -36,8 +36,11 @@  struct scatterlist;
 
 int swiotlb_init_with_tbl(char *tlb, unsigned long nslabs, unsigned int flags);
 unsigned long swiotlb_size_or_default(void);
+void __init swiotlb_init_remap(bool addressing_limit, unsigned int flags,
+	int (*remap)(void *tlb, unsigned long nslabs));
+int swiotlb_init_late(size_t size, gfp_t gfp_mask,
+	int (*remap)(void *tlb, unsigned long nslabs));
 extern int swiotlb_late_init_with_tbl(char *tlb, unsigned long nslabs);
-int swiotlb_init_late(size_t size, gfp_t gfp_mask);
 extern void __init swiotlb_update_mem_attributes(void);
 
 phys_addr_t swiotlb_tbl_map_single(struct device *hwdev, phys_addr_t phys,
diff --git a/kernel/dma/swiotlb.c b/kernel/dma/swiotlb.c
index 79641c446d284..88ea7b9bce6e9 100644
--- a/kernel/dma/swiotlb.c
+++ b/kernel/dma/swiotlb.c
@@ -256,9 +256,11 @@  int __init swiotlb_init_with_tbl(char *tlb, unsigned long nslabs,
  * Statically reserve bounce buffer space and initialize bounce buffer data
  * structures for the software IO TLB used to implement the DMA API.
  */
-void __init swiotlb_init(bool addressing_limit, unsigned int flags)
+void __init swiotlb_init_remap(bool addressing_limit, unsigned int flags,
+		int (*remap)(void *tlb, unsigned long nslabs))
 {
-	size_t bytes = PAGE_ALIGN(default_nslabs << IO_TLB_SHIFT);
+	unsigned long nslabs = default_nslabs;
+	size_t bytes;
 	void *tlb;
 
 	if (!addressing_limit && !swiotlb_force_bounce)
@@ -271,12 +273,24 @@  void __init swiotlb_init(bool addressing_limit, unsigned int flags)
 	 * allow to pick a location everywhere for hypervisors with guest
 	 * memory encryption.
 	 */
+retry:
+	bytes = PAGE_ALIGN(default_nslabs << IO_TLB_SHIFT);
 	if (flags & SWIOTLB_ANY)
 		tlb = memblock_alloc(bytes, PAGE_SIZE);
 	else
 		tlb = memblock_alloc_low(bytes, PAGE_SIZE);
 	if (!tlb)
 		goto fail;
+	if (remap && remap(tlb, nslabs) < 0) {
+		memblock_free(tlb, PAGE_ALIGN(bytes));
+
+		/* Min is 2MB */
+		if (nslabs <= 1024)
+			panic("%s: Failed to remap %zu bytes\n",
+			      __func__, bytes);
+		nslabs = max(1024UL, ALIGN(nslabs >> 1, IO_TLB_SEGSIZE));
+		goto retry;
+	}
 	if (swiotlb_init_with_tbl(tlb, default_nslabs, flags))
 		goto fail_free_mem;
 	return;
@@ -287,12 +301,18 @@  void __init swiotlb_init(bool addressing_limit, unsigned int flags)
 	pr_warn("Cannot allocate buffer");
 }
 
+void __init swiotlb_init(bool addressing_limit, unsigned int flags)
+{
+	return swiotlb_init_remap(addressing_limit, flags, NULL);
+}
+
 /*
  * Systems with larger DMA zones (those that don't support ISA) can
  * initialize the swiotlb later using the slab allocator if needed.
  * This should be just like above, but with some error catching.
  */
-int swiotlb_init_late(size_t size, gfp_t gfp_mask)
+int swiotlb_init_late(size_t size, gfp_t gfp_mask,
+		int (*remap)(void *tlb, unsigned long nslabs))
 {
 	unsigned long nslabs = ALIGN(size >> IO_TLB_SHIFT, IO_TLB_SEGSIZE);
 	unsigned long bytes;
@@ -303,6 +323,7 @@  int swiotlb_init_late(size_t size, gfp_t gfp_mask)
 	if (swiotlb_force_disable)
 		return 0;
 
+retry:
 	order = get_order(nslabs << IO_TLB_SHIFT);
 	nslabs = SLABS_PER_PAGE << order;
 	bytes = nslabs << IO_TLB_SHIFT;
@@ -317,6 +338,17 @@  int swiotlb_init_late(size_t size, gfp_t gfp_mask)
 
 	if (!vstart)
 		return -ENOMEM;
+	if (remap)
+		rc = remap(vstart, nslabs);
+	if (rc) {
+		free_pages((unsigned long)vstart, order);
+ 
+		/* Min is 2MB */
+		if (nslabs <= 1024)
+			return rc;
+		nslabs = max(1024UL, ALIGN(nslabs >> 1, IO_TLB_SEGSIZE));
+		goto retry;
+	}
 
 	if (order != get_order(bytes)) {
 		pr_warn("only able to allocate %ld MB\n",