From 3ae30cd6b9407cea58aa9d32c3f8e8916153cf8e Mon Sep 17 00:00:00 2001
From: Jerome Robert <jeromerobert@gmx.com>
Date: Fri, 15 Jan 2016 17:12:04 +0100
Subject: [PATCH 1/3] Disable multi-threading for small matrices in [z]ger

Ref #731
---
 interface/ger.c  | 7 +++++--
 interface/zger.c | 6 +++++-
 2 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/interface/ger.c b/interface/ger.c
index 2d19e3bc0..b2f803dec 100644
--- a/interface/ger.c
+++ b/interface/ger.c
@@ -174,8 +174,11 @@ void CNAME(enum CBLAS_ORDER order,
   STACK_ALLOC(m, FLOAT, buffer);
 
 #ifdef SMPTEST
-  nthreads = num_cpu_avail(2);
-
+  // Threshold chosen so that speed-up is > 1 on a Xeon E5-2630
+  if(1L * m * n > 24L * GEMM_MULTITHREAD_THRESHOLD)
+    nthreads = num_cpu_avail(2);
+  else
+    nthreads = 1;
 
   if (nthreads == 1) {
 #endif
diff --git a/interface/zger.c b/interface/zger.c
index f46a462e2..5bcbfc09b 100644
--- a/interface/zger.c
+++ b/interface/zger.c
@@ -213,7 +213,11 @@ void CNAME(enum CBLAS_ORDER order,
   buffer = (FLOAT *)blas_memory_alloc(1);
 
 #ifdef SMPTEST
-  nthreads = num_cpu_avail(2);
+  // Threshold chosen so that speed-up is > 1 on a Xeon E5-2630
+  if(1L * m * n > 3L * sizeof(FLOAT) * GEMM_MULTITHREAD_THRESHOLD)
+    nthreads = num_cpu_avail(2);
+  else
+    nthreads = 1;
 
   if (nthreads == 1) {
 #endif

From 66eafb16cfaf7e119afff96ff3e44663134141fa Mon Sep 17 00:00:00 2001
From: Jerome Robert <jeromerobert@gmx.com>
Date: Mon, 18 Jan 2016 09:12:37 +0100
Subject: [PATCH 2/3] swap: disable multi-threading for small matrices

Close #731
---
 interface/swap.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/interface/swap.c b/interface/swap.c
index 3baeb27c4..23b2e4ec8 100644
--- a/interface/swap.c
+++ b/interface/swap.c
@@ -77,12 +77,13 @@ void CNAME(blasint n, FLOAT *x, blasint incx, FLOAT *y, blasint incy){
   if (incy < 0) y -= (n - 1) * incy;
 
 #ifdef SMP
-  nthreads = num_cpu_avail(1);
 
   //disable multi-thread when incx==0 or incy==0
   //In that case, the threads would be dependent.
-  if (incx == 0 || incy == 0)
-	  nthreads = 1;
+  if (incx == 0 || incy == 0 || n < 2097152 * GEMM_MULTITHREAD_THRESHOLD / sizeof(FLOAT))
+    nthreads = 1;
+  else
+    nthreads = num_cpu_avail(1);
 
   if (nthreads == 1) {
 #endif

From 14db1ca508575835bf4dffb8003df1b88adf4f32 Mon Sep 17 00:00:00 2001
From: Jerome Robert <jeromerobert@gmx.com>
Date: Tue, 19 Jan 2016 17:15:31 +0100
Subject: [PATCH 3/3] update CONTRIBUTORS.md

---
 CONTRIBUTORS.md | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/CONTRIBUTORS.md b/CONTRIBUTORS.md
index 88e461dc4..b1c8dd140 100644
--- a/CONTRIBUTORS.md
+++ b/CONTRIBUTORS.md
@@ -121,6 +121,11 @@ In chronological order:
   * [2014-10-10] trmm and sgemm kernels (optimized for APM's X-Gene 1).
                  ARMv8 support.
 
+* Jerome Robert <jeromerobert@gmx.com>
+  * [2015-01-01] Speed-up small `ger` and `gemv` using stack allocation (bug #478)
+  * [2015-12-23] `stack_check` in `gemv.c` (bug #722)
+  * [2016-01-19] Disable multi-threading in `ger` and `swap` for small matrices (bug #731)
+
 * Dan Kortschak
   * [2015-01-07] Added test for drotmg bug #484.