BUGFIX: Workaround for ISE tools implementing memory as FFs in case of 1 stream

Changelog.md

@@ -1,3 +1,7 @@
+## 1.2.2.
+* Bugfixes
+  * Workaround for ISE tools implementing memory as FFs in case of 1 stream
+
 ## 1.2.1
 * Bugfixes
   * Optimized timing between input FIFO and DMA

hdl/psi_ms_daq_axi.vhd

@@ -398,7 +398,7 @@ begin
 			AxiDataWidth_g			=> AxiDataWidth_g,
 			AxiMaxBeats_g			=> AxiMaxBurstBeats_g,
 			AxiMaxOpenTrasactions_g	=> AxiMaxOpenTrasactions_g,
-			MaxOpenCommands_g		=> Streams_g,
+			MaxOpenCommands_g		=> max(2, Streams_g), -- ISE tools implement memory as FFs for one stream. Reason is unkown, so we always implement two streams for resource optimization reasons.
 			DataFifoDepth_g			=> 1024,
 			AxiFifoDepth_g			=> AxiFifoDepth_g,
 			RamBehavior_g			=> "RBW" -- Okay for Xilinx chips