[PATCH v4 1/3] malloc: add biggest free IOVA-contiguous element to stats

[Anatoly Burakov <anatoly.burakov@intel.com>]

User might be interested to find out what is the biggest chunk of
IOVA-contiguous free space that can be allocated from malloc. Add
relevant malloc-internal functions and expose this through malloc
stats calculation call.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Remy Horton <remy.horton@intel.com>
---

Notes:
    v4:
    - Fix comments to be more up to date with v4 code
    - Add comments explaining trailer handling
    
    v2:
    - Add header to newly recalculated element start
    
    v2:
    - Add header to newly recalculated element start

 lib/librte_eal/common/include/rte_malloc.h |  1 +
 lib/librte_eal/common/malloc_elem.c        | 79 ++++++++++++++++++++++++++++++
 lib/librte_eal/common/malloc_elem.h        |  6 +++
 lib/librte_eal/common/malloc_heap.c        | 11 +++++
 lib/librte_eal/common/rte_malloc.c         |  2 +
 5 files changed, 99 insertions(+)

diff --git a/lib/librte_eal/common/include/rte_malloc.h b/lib/librte_eal/common/include/rte_malloc.h
index a9fb7e4..2553201 100644
--- a/lib/librte_eal/common/include/rte_malloc.h
+++ b/lib/librte_eal/common/include/rte_malloc.h
@@ -27,6 +27,7 @@ struct rte_malloc_socket_stats {
 	size_t heap_totalsz_bytes; /**< Total bytes on heap */
 	size_t heap_freesz_bytes;  /**< Total free bytes on heap */
 	size_t greatest_free_size; /**< Size in bytes of largest free block */
+	size_t greatest_free_iova_contig_size; /**< Size in bytes of largest IOVA-contiguous block */
 	unsigned free_count;       /**< Number of free elements on heap */
 	unsigned alloc_count;      /**< Number of allocated elements on heap */
 	size_t heap_allocsz_bytes; /**< Total allocated bytes on heap */
diff --git a/lib/librte_eal/common/malloc_elem.c b/lib/librte_eal/common/malloc_elem.c
index 9bfe9b9..f1bb4fe 100644
--- a/lib/librte_eal/common/malloc_elem.c
+++ b/lib/librte_eal/common/malloc_elem.c
@@ -18,10 +18,89 @@
 #include <rte_common.h>
 #include <rte_spinlock.h>
 
+#include "eal_internal_cfg.h"
 #include "eal_memalloc.h"
 #include "malloc_elem.h"
 #include "malloc_heap.h"
 
+size_t
+malloc_elem_find_max_iova_contig(struct malloc_elem *elem, size_t align)
+{
+	void *cur_page, *contig_seg_start, *page_end, *cur_seg_end;
+	void *data_start, *data_end;
+	rte_iova_t expected_iova;
+	struct rte_memseg *ms;
+	size_t page_sz, cur, max;
+
+	page_sz = (size_t)elem->msl->page_sz;
+	data_start = RTE_PTR_ADD(elem, MALLOC_ELEM_HEADER_LEN);
+	data_end = RTE_PTR_ADD(elem, elem->size - MALLOC_ELEM_TRAILER_LEN);
+	/* segment must start after header and with specified alignment */
+	contig_seg_start = RTE_PTR_ALIGN_CEIL(data_start, align);
+
+	/* if we're in IOVA as VA mode, or if we're in legacy mode with
+	 * hugepages, all elements are IOVA-contiguous.
+	 */
+	if (rte_eal_iova_mode() == RTE_IOVA_VA ||
+			(internal_config.legacy_mem && rte_eal_has_hugepages()))
+		return RTE_PTR_DIFF(data_end, contig_seg_start);
+
+	cur_page = RTE_PTR_ALIGN_FLOOR(contig_seg_start, page_sz);
+	ms = rte_mem_virt2memseg(cur_page, elem->msl);
+
+	/* do first iteration outside the loop */
+	page_end = RTE_PTR_ADD(cur_page, page_sz);
+	cur_seg_end = RTE_MIN(page_end, data_end);
+	cur = RTE_PTR_DIFF(cur_seg_end, contig_seg_start) -
+			MALLOC_ELEM_TRAILER_LEN;
+	max = cur;
+	expected_iova = ms->iova + page_sz;
+	/* memsegs are contiguous in memory */
+	ms++;
+
+	cur_page = RTE_PTR_ADD(cur_page, page_sz);
+
+	while (cur_page < data_end) {
+		page_end = RTE_PTR_ADD(cur_page, page_sz);
+		cur_seg_end = RTE_MIN(page_end, data_end);
+
+		/* reset start of contiguous segment if unexpected iova */
+		if (ms->iova != expected_iova) {
+			/* next contiguous segment must start at specified
+			 * alignment.
+			 */
+			contig_seg_start = RTE_PTR_ALIGN(cur_page, align);
+			/* new segment start may be on a different page, so find
+			 * the page and skip to next iteration to make sure
+			 * we're not blowing past data end.
+			 */
+			ms = rte_mem_virt2memseg(contig_seg_start, elem->msl);
+			cur_page = ms->addr;
+			/* don't trigger another recalculation */
+			expected_iova = ms->iova;
+			continue;
+		}
+		/* cur_seg_end ends on a page boundary or on data end. if we're
+		 * looking at data end, then malloc trailer is already included
+		 * in the calculations. if we're looking at page end, then we
+		 * know there's more data past this page and thus there's space
+		 * for malloc element trailer, so don't count it here.
+		 */
+		cur = RTE_PTR_DIFF(cur_seg_end, contig_seg_start);
+		/* update max if cur value is bigger */
+		if (cur > max)
+			max = cur;
+
+		/* move to next page */
+		cur_page = page_end;
+		expected_iova = ms->iova + page_sz;
+		/* memsegs are contiguous in memory */
+		ms++;
+	}
+
+	return max;
+}
+
 /*
  * Initialize a general malloc_elem header structure
  */
diff --git a/lib/librte_eal/common/malloc_elem.h b/lib/librte_eal/common/malloc_elem.h
index 7331af9..e2bda4c 100644
--- a/lib/librte_eal/common/malloc_elem.h
+++ b/lib/librte_eal/common/malloc_elem.h
@@ -179,4 +179,10 @@ malloc_elem_free_list_index(size_t size);
 void
 malloc_elem_free_list_insert(struct malloc_elem *elem);
 
+/*
+ * Find biggest IOVA-contiguous zone within an element with specified alignment.
+ */
+size_t
+malloc_elem_find_max_iova_contig(struct malloc_elem *elem, size_t align);
+
 #endif /* MALLOC_ELEM_H_ */
diff --git a/lib/librte_eal/common/malloc_heap.c b/lib/librte_eal/common/malloc_heap.c
index d6cf3af..9305b38 100644
--- a/lib/librte_eal/common/malloc_heap.c
+++ b/lib/librte_eal/common/malloc_heap.c
@@ -803,16 +803,27 @@ malloc_heap_get_stats(struct malloc_heap *heap,
 	socket_stats->free_count = 0;
 	socket_stats->heap_freesz_bytes = 0;
 	socket_stats->greatest_free_size = 0;
+	socket_stats->greatest_free_iova_contig_size = 0;
 
 	/* Iterate through free list */
 	for (idx = 0; idx < RTE_HEAP_NUM_FREELISTS; idx++) {
 		for (elem = LIST_FIRST(&heap->free_head[idx]);
 			!!elem; elem = LIST_NEXT(elem, free_list))
 		{
+			size_t iova_contig_sz;
 			socket_stats->free_count++;
 			socket_stats->heap_freesz_bytes += elem->size;
 			if (elem->size > socket_stats->greatest_free_size)
 				socket_stats->greatest_free_size = elem->size;
+			iova_contig_sz =
+					malloc_elem_find_max_iova_contig(elem,
+							RTE_CACHE_LINE_SIZE);
+			/* find_max_iova_contig doesn't include overhead */
+			iova_contig_sz += MALLOC_ELEM_OVERHEAD;
+			if (iova_contig_sz >
+					socket_stats->greatest_free_iova_contig_size)
+				socket_stats->greatest_free_iova_contig_size =
+						iova_contig_sz;
 		}
 	}
 	/* Get stats on overall heap and allocated memory on this heap */
diff --git a/lib/librte_eal/common/rte_malloc.c b/lib/librte_eal/common/rte_malloc.c
index b51a6d1..b4e87a3 100644
--- a/lib/librte_eal/common/rte_malloc.c
+++ b/lib/librte_eal/common/rte_malloc.c
@@ -195,6 +195,8 @@ rte_malloc_dump_stats(FILE *f, __rte_unused const char *type)
 		fprintf(f, "\tAlloc_size:%zu,\n", sock_stats.heap_allocsz_bytes);
 		fprintf(f, "\tGreatest_free_size:%zu,\n",
 				sock_stats.greatest_free_size);
+		fprintf(f, "\tGreatest_free_iova_contig_size:%zu,\n",
+				sock_stats.greatest_free_iova_contig_size);
 		fprintf(f, "\tAlloc_count:%u,\n",sock_stats.alloc_count);
 		fprintf(f, "\tFree_count:%u,\n", sock_stats.free_count);
 	}
-- 
2.7.4