svm_fifo rework to avoid contention on cursize

Problems Addressed:
- Contention of cursize by producer and consumer.
- Reduce the no of modulo operations.

Changes:
- Synchronization between producer and consumer changed from cursize
  to head and tail indexes
  Implications: reduces the usable size of fifo by 1.
- Using weaker memory ordering C++11 atomics to access head and tail
  based on producer and consumer role.
- Head and tail indexes are unsigned 32 bit integers. Additions and
  subtraction on them are implicit 32 bit Modulo operation.
- Adding weaker memory ordering variants of max_enq, max_deq, is_empty
  and is_full Using them appropriately in all places.

Perfomance improvement (iperf3 via Hoststack):

iperf3 Server: Marvell ThunderX2(AArch64) - iperf3 Client: Skylake(x86)
   ~6%(256 rxd/txd) - ~11%(2048 rxd/txd)

Change-Id: I1d484e000e437430fdd5a819657d1c6b62443018
Signed-off-by: Sirshak Das <sirshak.das@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>

1 files changed, 6 insertions, 1 deletions

diff --git a/src/svm/svm_fifo_segment.c b/src/svm/svm_fifo_segment.c
index e377160f6d8..94bf6a5e73e 100644
--- a/src/svm/svm_fifo_segment.c
+++ b/src/svm/svm_fifo_segment.c
@@ -381,7 +381,12 @@ svm_fifo_segment_alloc_fifo (svm_fifo_segment_private_t * fs,
 	  fsh->free_fifos[freelist_index] = f->next;
 	  /* (re)initialize the fifo, as in svm_fifo_create */
 	  clib_memset (f, 0, sizeof (*f));
-	  f->nitems = data_size_in_bytes;
+	  f->size = (1 << (max_log2 (data_size_in_bytes)));
+	  /*
+	   * usable size of the fifo set to rounded_data_size - 1
+	   * to differentiate between free fifo and empty fifo.
+	   */
+	  f->nitems = f->size - 1;
 	  f->ooos_list_head = OOO_SEGMENT_INVALID_INDEX;
 	  f->ct_session_index = SVM_FIFO_INVALID_SESSION_INDEX;
 	  f->refcnt = 1;