[PATCH v3 net] octeontx2-af: allocate qmem from page allocator to avoid CMA

[PATCH v3 net] octeontx2-af: allocate qmem from page allocator to avoid CMA

[Ratheesh Kannoth]

qmem_alloc() uses dma_alloc_attrs() with DMA_ATTR_FORCE_CONTIGUOUS for
physically contiguous DMA memory (e.g. CN10K LMTST regions spanning page
boundaries). With CMA enabled, those allocations are drawn from the CMA
pool. qmem backs NIX/NPA queue contexts, admin queues, and LMTST regions,
so total usage scales with enabled interfaces and is hard to provision in
CMA.

Add otx2_dma_alloc_coherent() and otx2_dma_free_coherent() helpers that
allocate compound pages from the buddy allocator via __get_free_pages()
with __GFP_COMP, __GFP_ZERO, and __GFP_RECLAIM (qmem_alloc() runs in
GFP_KERNEL context), select DMA-reachable memory with dma_coherent_ok()
(retrying with GFP_DMA32), and map the contiguous physical range for DMA
through dma_map_phys() and dma_unmap_phys() with DMA_ATTR_REQUIRE_COHERENT
so SMMU/IOMMU page table entries are installed on IOMMU-backed platforms.
Switch qmem_alloc() and qmem_free() to these helpers instead of
dma_alloc_attrs()/dma_free_attrs() with DMA_ATTR_FORCE_CONTIGUOUS.

Allocations requiring more than MAX_PAGE_ORDER pages are still rejected,
as the buddy allocator cannot serve them without CMA.

Fixes: 73d33dbc0723 ("octeontx2-af: Use DMA_ATTR_FORCE_CONTIGUOUS attribute in DMA alloc")
Signed-off-by: Ratheesh Kannoth <rkannoth@marvell.com>

---
v2 -> v3: Addressed sashiko comments
	https://sashiko.dev/#/patchset/20260825045616.3723078-1-rkannoth%40marvell.com
---
 .../ethernet/marvell/octeontx2/af/common.h    | 81 +++++++++++++++++--
 1 file changed, 75 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/af/common.h b/drivers/net/ethernet/marvell/octeontx2/af/common.h
index 779413a383b7..c6a6aae436a3 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/common.h
@@ -7,6 +7,10 @@
 #ifndef COMMON_H
 #define COMMON_H
 
+#include <linux/dma-mapping.h>
+#include <linux/gfp.h>
+#include <linux/mm.h>
+
 #include "rvu_struct.h"
 
 #define OTX2_ALIGN			128  /* Align to cacheline */
@@ -44,6 +48,72 @@ struct qmem {
 	u32		qsize;
 };
 
+/* Buddy-backed coherent DMA alloc (Option 3): pages from __get_free_pages(),
+ * DMA-reachable RAM via dma_coherent_ok(), and bus/SMMU mappings via
+ * dma_map_phys() -> iommu_dma_map_phys() -> iommu_map() on SMMU systems.
+ */
+#define OTX2_DMA_COHERENT_ATTRS	DMA_ATTR_REQUIRE_COHERENT
+
+static inline void *otx2_dma_alloc_coherent(struct device *dev, size_t size,
+					    dma_addr_t *dma_handle, gfp_t gfp)
+{
+	dma_addr_t dma_addr;
+	unsigned int order;
+	phys_addr_t paddr;
+	void *vaddr;
+	gfp_t alloc_gfp;
+
+	if (!dev || !dma_handle || !size)
+		return NULL;
+
+	size = PAGE_ALIGN(size);
+	order = get_order(size);
+	if (order > MAX_PAGE_ORDER)
+		return NULL;
+
+	alloc_gfp = (gfp & ~(__GFP_DMA | __GFP_DMA32 | __GFP_HIGHMEM)) |
+		    __GFP_ZERO | __GFP_COMP | __GFP_RECLAIM;
+
+	vaddr = (void *)__get_free_pages(alloc_gfp, order);
+	while (vaddr &&
+	       !dma_coherent_ok(dev, page_to_phys(virt_to_page(vaddr)), size)) {
+		free_pages((unsigned long)vaddr, order);
+		if (alloc_gfp & GFP_DMA32)
+			return NULL;
+		alloc_gfp |= GFP_DMA32;
+		vaddr = (void *)__get_free_pages(alloc_gfp, order);
+	}
+	if (!vaddr)
+		return NULL;
+
+	paddr = page_to_phys(virt_to_page(vaddr));
+	dma_addr = dma_map_phys(dev, paddr, size, DMA_BIDIRECTIONAL,
+				OTX2_DMA_COHERENT_ATTRS);
+	if (dma_mapping_error(dev, dma_addr)) {
+		free_pages((unsigned long)vaddr, order);
+		return NULL;
+	}
+
+	*dma_handle = dma_addr;
+	return vaddr;
+}
+
+static inline void otx2_dma_free_coherent(struct device *dev, size_t size,
+					  void *vaddr, dma_addr_t dma_handle)
+{
+	unsigned int order;
+
+	if (!dev || !vaddr)
+		return;
+
+	size = PAGE_ALIGN(size);
+	order = get_order(size);
+
+	dma_unmap_phys(dev, dma_handle, size, DMA_BIDIRECTIONAL,
+		       OTX2_DMA_COHERENT_ATTRS);
+	free_pages((unsigned long)vaddr, order);
+}
+
 static inline int qmem_alloc(struct device *dev, struct qmem **q,
 			     int qsize, int entry_sz)
 {
@@ -60,8 +130,8 @@ static inline int qmem_alloc(struct device *dev, struct qmem **q,
 
 	qmem->entry_sz = entry_sz;
 	qmem->alloc_sz = (qsize * entry_sz) + OTX2_ALIGN;
-	qmem->base = dma_alloc_attrs(dev, qmem->alloc_sz, &qmem->iova,
-				     GFP_KERNEL, DMA_ATTR_FORCE_CONTIGUOUS);
+	qmem->base = otx2_dma_alloc_coherent(dev, qmem->alloc_sz, &qmem->iova,
+					     GFP_KERNEL);
 	if (!qmem->base)
 		return -ENOMEM;
 
@@ -80,10 +150,9 @@ static inline void qmem_free(struct device *dev, struct qmem *qmem)
 		return;
 
 	if (qmem->base)
-		dma_free_attrs(dev, qmem->alloc_sz,
-			       qmem->base - qmem->align,
-			       qmem->iova - qmem->align,
-			       DMA_ATTR_FORCE_CONTIGUOUS);
+		otx2_dma_free_coherent(dev, qmem->alloc_sz,
+				       qmem->base - qmem->align,
+				       qmem->iova - qmem->align);
 	devm_kfree(dev, qmem);
 }
 
-- 
2.43.0