diff --git a/rtl/axis_arb_mux.v b/rtl/axis_arb_mux.v
index 5f8eb2ddc6d16d8f53aad302167ef3c8b7ffa7b3..a9f5c10551a33019e60598edc0063429d7698e06 100644
--- a/rtl/axis_arb_mux.v
+++ b/rtl/axis_arb_mux.v
@@ -43,8 +43,10 @@ module axis_arb_mux #
     parameter KEEP_WIDTH = (DATA_WIDTH/8),
     // Propagate tid signal
     parameter ID_ENABLE = 0,
-    // tid signal width
-    parameter ID_WIDTH = 8,
+    // input tid signal width
+    parameter S_ID_WIDTH = 8,
+    // output tid signal width
+    parameter M_ID_WIDTH = S_ID_WIDTH+$clog2(S_COUNT),
     // Propagate tdest signal
     parameter DEST_ENABLE = 0,
     // tdest signal width
@@ -67,26 +69,26 @@ module axis_arb_mux #
     /*
      * AXI Stream inputs
      */
-    input  wire [S_COUNT*DATA_WIDTH-1:0] s_axis_tdata,
-    input  wire [S_COUNT*KEEP_WIDTH-1:0] s_axis_tkeep,
-    input  wire [S_COUNT-1:0]            s_axis_tvalid,
-    output wire [S_COUNT-1:0]            s_axis_tready,
-    input  wire [S_COUNT-1:0]            s_axis_tlast,
-    input  wire [S_COUNT*ID_WIDTH-1:0]   s_axis_tid,
-    input  wire [S_COUNT*DEST_WIDTH-1:0] s_axis_tdest,
-    input  wire [S_COUNT*USER_WIDTH-1:0] s_axis_tuser,
+    input  wire [S_COUNT*DATA_WIDTH-1:0]  s_axis_tdata,
+    input  wire [S_COUNT*KEEP_WIDTH-1:0]  s_axis_tkeep,
+    input  wire [S_COUNT-1:0]             s_axis_tvalid,
+    output wire [S_COUNT-1:0]             s_axis_tready,
+    input  wire [S_COUNT-1:0]             s_axis_tlast,
+    input  wire [S_COUNT*S_ID_WIDTH-1:0]  s_axis_tid,
+    input  wire [S_COUNT*DEST_WIDTH-1:0]  s_axis_tdest,
+    input  wire [S_COUNT*USER_WIDTH-1:0]  s_axis_tuser,
 
     /*
      * AXI Stream output
      */
-    output wire [DATA_WIDTH-1:0]         m_axis_tdata,
-    output wire [KEEP_WIDTH-1:0]         m_axis_tkeep,
-    output wire                          m_axis_tvalid,
-    input  wire                          m_axis_tready,
-    output wire                          m_axis_tlast,
-    output wire [ID_WIDTH-1:0]           m_axis_tid,
-    output wire [DEST_WIDTH-1:0]         m_axis_tdest,
-    output wire [USER_WIDTH-1:0]         m_axis_tuser
+    output wire [DATA_WIDTH-1:0]          m_axis_tdata,
+    output wire [KEEP_WIDTH-1:0]          m_axis_tkeep,
+    output wire                           m_axis_tvalid,
+    input  wire                           m_axis_tready,
+    output wire                           m_axis_tlast,
+    output wire [M_ID_WIDTH-1:0]          m_axis_tid,
+    output wire [DEST_WIDTH-1:0]          m_axis_tdest,
+    output wire [USER_WIDTH-1:0]          m_axis_tuser
 );
 
 parameter CL_S_COUNT = $clog2(S_COUNT);
@@ -103,7 +105,7 @@ reg  [KEEP_WIDTH-1:0] m_axis_tkeep_int;
 reg                   m_axis_tvalid_int;
 reg                   m_axis_tready_int_reg = 1'b0;
 reg                   m_axis_tlast_int;
-reg  [ID_WIDTH-1:0]   m_axis_tid_int;
+reg  [M_ID_WIDTH-1:0] m_axis_tid_int;
 reg  [DEST_WIDTH-1:0] m_axis_tdest_int;
 reg  [USER_WIDTH-1:0] m_axis_tuser_int;
 wire                  m_axis_tready_int_early;
@@ -116,7 +118,7 @@ wire [KEEP_WIDTH-1:0] current_s_tkeep  = s_axis_tkeep[grant_encoded*KEEP_WIDTH +
 wire                  current_s_tvalid = s_axis_tvalid[grant_encoded];
 wire                  current_s_tready = s_axis_tready[grant_encoded];
 wire                  current_s_tlast  = s_axis_tlast[grant_encoded];
-wire [ID_WIDTH-1:0]   current_s_tid    = s_axis_tid[grant_encoded*ID_WIDTH +: ID_WIDTH];
+wire [S_ID_WIDTH-1:0] current_s_tid    = s_axis_tid[grant_encoded*S_ID_WIDTH +: S_ID_WIDTH];
 wire [DEST_WIDTH-1:0] current_s_tdest  = s_axis_tdest[grant_encoded*DEST_WIDTH +: DEST_WIDTH];
 wire [USER_WIDTH-1:0] current_s_tuser  = s_axis_tuser[grant_encoded*USER_WIDTH +: USER_WIDTH];
 
@@ -157,7 +159,7 @@ reg [DATA_WIDTH-1:0] m_axis_tdata_reg  = {DATA_WIDTH{1'b0}};
 reg [KEEP_WIDTH-1:0] m_axis_tkeep_reg  = {KEEP_WIDTH{1'b0}};
 reg                  m_axis_tvalid_reg = 1'b0, m_axis_tvalid_next;
 reg                  m_axis_tlast_reg  = 1'b0;
-reg [ID_WIDTH-1:0]   m_axis_tid_reg    = {ID_WIDTH{1'b0}};
+reg [M_ID_WIDTH-1:0] m_axis_tid_reg    = {M_ID_WIDTH{1'b0}};
 reg [DEST_WIDTH-1:0] m_axis_tdest_reg  = {DEST_WIDTH{1'b0}};
 reg [USER_WIDTH-1:0] m_axis_tuser_reg  = {USER_WIDTH{1'b0}};
 
@@ -165,7 +167,7 @@ reg [DATA_WIDTH-1:0] temp_m_axis_tdata_reg  = {DATA_WIDTH{1'b0}};
 reg [KEEP_WIDTH-1:0] temp_m_axis_tkeep_reg  = {KEEP_WIDTH{1'b0}};
 reg                  temp_m_axis_tvalid_reg = 1'b0, temp_m_axis_tvalid_next;
 reg                  temp_m_axis_tlast_reg  = 1'b0;
-reg [ID_WIDTH-1:0]   temp_m_axis_tid_reg    = {ID_WIDTH{1'b0}};
+reg [M_ID_WIDTH-1:0] temp_m_axis_tid_reg    = {M_ID_WIDTH{1'b0}};
 reg [DEST_WIDTH-1:0] temp_m_axis_tdest_reg  = {DEST_WIDTH{1'b0}};
 reg [USER_WIDTH-1:0] temp_m_axis_tuser_reg  = {USER_WIDTH{1'b0}};
 
@@ -178,7 +180,7 @@ assign m_axis_tdata  = m_axis_tdata_reg;
 assign m_axis_tkeep  = KEEP_ENABLE ? m_axis_tkeep_reg : {KEEP_WIDTH{1'b1}};
 assign m_axis_tvalid = m_axis_tvalid_reg;
 assign m_axis_tlast  = LAST_ENABLE ? m_axis_tlast_reg : 1'b1;
-assign m_axis_tid    = ID_ENABLE   ? m_axis_tid_reg   : {ID_WIDTH{1'b0}};
+assign m_axis_tid    = ID_ENABLE   ? m_axis_tid_reg   : {M_ID_WIDTH{1'b0}};
 assign m_axis_tdest  = DEST_ENABLE ? m_axis_tdest_reg : {DEST_WIDTH{1'b0}};
 assign m_axis_tuser  = USER_ENABLE ? m_axis_tuser_reg : {USER_WIDTH{1'b0}};
 
diff --git a/rtl/axis_arb_mux_wrap.py b/rtl/axis_arb_mux_wrap.py
index 4cfe2b8fc5aeddd68719b6b6df55fd92811233eb..7e02d207fbb6028bb745e4f8f5c27e73ac298bc0 100755
--- a/rtl/axis_arb_mux_wrap.py
+++ b/rtl/axis_arb_mux_wrap.py
@@ -78,8 +78,10 @@ module {{name}} #
     parameter KEEP_WIDTH = (DATA_WIDTH/8),
     // Propagate tid signal
     parameter ID_ENABLE = 0,
-    // tid signal width
-    parameter ID_WIDTH = 8,
+    // input tid signal width
+    parameter S_ID_WIDTH = 8,
+    // output tid signal width
+    parameter M_ID_WIDTH = S_ID_WIDTH+{{cn}},
     // Propagate tdest signal
     parameter DEST_ENABLE = 0,
     // tdest signal width
@@ -96,33 +98,33 @@ module {{name}} #
     parameter ARB_LSB_HIGH_PRIORITY = 1
 )
 (
-    input  wire                  clk,
-    input  wire                  rst,
+    input  wire                   clk,
+    input  wire                   rst,
 
     /*
      * AXI Stream inputs
      */
 {%- for p in range(n) %}
-    input  wire [DATA_WIDTH-1:0] s{{'%02d'%p}}_axis_tdata,
-    input  wire [KEEP_WIDTH-1:0] s{{'%02d'%p}}_axis_tkeep,
-    input  wire                  s{{'%02d'%p}}_axis_tvalid,
-    output wire                  s{{'%02d'%p}}_axis_tready,
-    input  wire                  s{{'%02d'%p}}_axis_tlast,
-    input  wire [ID_WIDTH-1:0]   s{{'%02d'%p}}_axis_tid,
-    input  wire [DEST_WIDTH-1:0] s{{'%02d'%p}}_axis_tdest,
-    input  wire [USER_WIDTH-1:0] s{{'%02d'%p}}_axis_tuser,
+    input  wire [DATA_WIDTH-1:0]  s{{'%02d'%p}}_axis_tdata,
+    input  wire [KEEP_WIDTH-1:0]  s{{'%02d'%p}}_axis_tkeep,
+    input  wire                   s{{'%02d'%p}}_axis_tvalid,
+    output wire                   s{{'%02d'%p}}_axis_tready,
+    input  wire                   s{{'%02d'%p}}_axis_tlast,
+    input  wire [S_ID_WIDTH-1:0]  s{{'%02d'%p}}_axis_tid,
+    input  wire [DEST_WIDTH-1:0]  s{{'%02d'%p}}_axis_tdest,
+    input  wire [USER_WIDTH-1:0]  s{{'%02d'%p}}_axis_tuser,
 {% endfor %}
     /*
      * AXI Stream output
      */
-    output wire [DATA_WIDTH-1:0] m_axis_tdata,
-    output wire [KEEP_WIDTH-1:0] m_axis_tkeep,
-    output wire                  m_axis_tvalid,
-    input  wire                  m_axis_tready,
-    output wire                  m_axis_tlast,
-    output wire [ID_WIDTH-1:0]   m_axis_tid,
-    output wire [DEST_WIDTH-1:0] m_axis_tdest,
-    output wire [USER_WIDTH-1:0] m_axis_tuser
+    output wire [DATA_WIDTH-1:0]  m_axis_tdata,
+    output wire [KEEP_WIDTH-1:0]  m_axis_tkeep,
+    output wire                   m_axis_tvalid,
+    input  wire                   m_axis_tready,
+    output wire                   m_axis_tlast,
+    output wire [M_ID_WIDTH-1:0]  m_axis_tid,
+    output wire [DEST_WIDTH-1:0]  m_axis_tdest,
+    output wire [USER_WIDTH-1:0]  m_axis_tuser
 );
 
 axis_arb_mux #(
@@ -131,7 +133,8 @@ axis_arb_mux #(
     .KEEP_ENABLE(KEEP_ENABLE),
     .KEEP_WIDTH(KEEP_WIDTH),
     .ID_ENABLE(ID_ENABLE),
-    .ID_WIDTH(ID_WIDTH),
+    .S_ID_WIDTH(S_ID_WIDTH),
+    .M_ID_WIDTH(M_ID_WIDTH),
     .DEST_ENABLE(DEST_ENABLE),
     .DEST_WIDTH(DEST_WIDTH),
     .USER_ENABLE(USER_ENABLE),
diff --git a/rtl/axis_demux.v b/rtl/axis_demux.v
index fbfc748d2b644afc33a42fa931c7b62ecd396f36..c49fe425b8c0420b602ef689a938619f7e657c01 100644
--- a/rtl/axis_demux.v
+++ b/rtl/axis_demux.v
@@ -47,47 +47,49 @@ module axis_demux #
     parameter ID_WIDTH = 8,
     // Propagate tdest signal
     parameter DEST_ENABLE = 0,
-    // tdest signal width
-    parameter DEST_WIDTH = 8,
+    // output tdest signal width
+    parameter M_DEST_WIDTH = 8,
+    // input tdest signal width
+    parameter S_DEST_WIDTH = M_DEST_WIDTH+$clog2(M_COUNT),
     // Propagate tuser signal
     parameter USER_ENABLE = 1,
     // tuser signal width
     parameter USER_WIDTH = 1
 )
 (
-    input  wire                          clk,
-    input  wire                          rst,
+    input  wire                             clk,
+    input  wire                             rst,
 
     /*
      * AXI input
      */
-    input  wire [DATA_WIDTH-1:0]         s_axis_tdata,
-    input  wire [KEEP_WIDTH-1:0]         s_axis_tkeep,
-    input  wire                          s_axis_tvalid,
-    output wire                          s_axis_tready,
-    input  wire                          s_axis_tlast,
-    input  wire [ID_WIDTH-1:0]           s_axis_tid,
-    input  wire [DEST_WIDTH-1:0]         s_axis_tdest,
-    input  wire [USER_WIDTH-1:0]         s_axis_tuser,
+    input  wire [DATA_WIDTH-1:0]            s_axis_tdata,
+    input  wire [KEEP_WIDTH-1:0]            s_axis_tkeep,
+    input  wire                             s_axis_tvalid,
+    output wire                             s_axis_tready,
+    input  wire                             s_axis_tlast,
+    input  wire [ID_WIDTH-1:0]              s_axis_tid,
+    input  wire [S_DEST_WIDTH-1:0]          s_axis_tdest,
+    input  wire [USER_WIDTH-1:0]            s_axis_tuser,
 
     /*
      * AXI outputs
      */
-    output wire [M_COUNT*DATA_WIDTH-1:0] m_axis_tdata,
-    output wire [M_COUNT*KEEP_WIDTH-1:0] m_axis_tkeep,
-    output wire [M_COUNT-1:0]            m_axis_tvalid,
-    input  wire [M_COUNT-1:0]            m_axis_tready,
-    output wire [M_COUNT-1:0]            m_axis_tlast,
-    output wire [M_COUNT*ID_WIDTH-1:0]   m_axis_tid,
-    output wire [M_COUNT*DEST_WIDTH-1:0] m_axis_tdest,
-    output wire [M_COUNT*USER_WIDTH-1:0] m_axis_tuser,
+    output wire [M_COUNT*DATA_WIDTH-1:0]    m_axis_tdata,
+    output wire [M_COUNT*KEEP_WIDTH-1:0]    m_axis_tkeep,
+    output wire [M_COUNT-1:0]               m_axis_tvalid,
+    input  wire [M_COUNT-1:0]               m_axis_tready,
+    output wire [M_COUNT-1:0]               m_axis_tlast,
+    output wire [M_COUNT*ID_WIDTH-1:0]      m_axis_tid,
+    output wire [M_COUNT*M_DEST_WIDTH-1:0]  m_axis_tdest,
+    output wire [M_COUNT*USER_WIDTH-1:0]    m_axis_tuser,
 
     /*
      * Control
      */
-    input  wire                          enable,
-    input  wire                          drop,
-    input  wire [$clog2(M_COUNT)-1:0]    select
+    input  wire                             enable,
+    input  wire                             drop,
+    input  wire [$clog2(M_COUNT)-1:0]       select
 );
 
 parameter CL_M_COUNT = $clog2(M_COUNT);
@@ -99,15 +101,15 @@ reg frame_reg = 1'b0, frame_ctl, frame_next;
 reg s_axis_tready_reg = 1'b0, s_axis_tready_next;
 
 // internal datapath
-reg  [DATA_WIDTH-1:0] m_axis_tdata_int;
-reg  [KEEP_WIDTH-1:0] m_axis_tkeep_int;
-reg  [M_COUNT-1:0]    m_axis_tvalid_int;
-reg                   m_axis_tready_int_reg = 1'b0;
-reg                   m_axis_tlast_int;
-reg  [ID_WIDTH-1:0]   m_axis_tid_int;
-reg  [DEST_WIDTH-1:0] m_axis_tdest_int;
-reg  [USER_WIDTH-1:0] m_axis_tuser_int;
-wire                  m_axis_tready_int_early;
+reg  [DATA_WIDTH-1:0]    m_axis_tdata_int;
+reg  [KEEP_WIDTH-1:0]    m_axis_tkeep_int;
+reg  [M_COUNT-1:0]       m_axis_tvalid_int;
+reg                      m_axis_tready_int_reg = 1'b0;
+reg                      m_axis_tlast_int;
+reg  [ID_WIDTH-1:0]      m_axis_tid_int;
+reg  [M_DEST_WIDTH-1:0]  m_axis_tdest_int;
+reg  [USER_WIDTH-1:0]    m_axis_tuser_int;
+wire                     m_axis_tready_int_early;
 
 assign s_axis_tready = s_axis_tready_reg && enable;
 
@@ -167,21 +169,21 @@ always @(posedge clk) begin
 end
 
 // output datapath logic
-reg [DATA_WIDTH-1:0] m_axis_tdata_reg  = {DATA_WIDTH{1'b0}};
-reg [KEEP_WIDTH-1:0] m_axis_tkeep_reg  = {KEEP_WIDTH{1'b0}};
-reg [M_COUNT-1:0]    m_axis_tvalid_reg = {M_COUNT{1'b0}}, m_axis_tvalid_next;
-reg                  m_axis_tlast_reg  = 1'b0;
-reg [ID_WIDTH-1:0]   m_axis_tid_reg    = {ID_WIDTH{1'b0}};
-reg [DEST_WIDTH-1:0] m_axis_tdest_reg  = {DEST_WIDTH{1'b0}};
-reg [USER_WIDTH-1:0] m_axis_tuser_reg  = {USER_WIDTH{1'b0}};
-
-reg [DATA_WIDTH-1:0] temp_m_axis_tdata_reg  = {DATA_WIDTH{1'b0}};
-reg [KEEP_WIDTH-1:0] temp_m_axis_tkeep_reg  = {KEEP_WIDTH{1'b0}};
-reg [M_COUNT-1:0]    temp_m_axis_tvalid_reg = {M_COUNT{1'b0}}, temp_m_axis_tvalid_next;
-reg                  temp_m_axis_tlast_reg  = 1'b0;
-reg [ID_WIDTH-1:0]   temp_m_axis_tid_reg    = {ID_WIDTH{1'b0}};
-reg [DEST_WIDTH-1:0] temp_m_axis_tdest_reg  = {DEST_WIDTH{1'b0}};
-reg [USER_WIDTH-1:0] temp_m_axis_tuser_reg  = {USER_WIDTH{1'b0}};
+reg [DATA_WIDTH-1:0]    m_axis_tdata_reg  = {DATA_WIDTH{1'b0}};
+reg [KEEP_WIDTH-1:0]    m_axis_tkeep_reg  = {KEEP_WIDTH{1'b0}};
+reg [M_COUNT-1:0]       m_axis_tvalid_reg = {M_COUNT{1'b0}}, m_axis_tvalid_next;
+reg                     m_axis_tlast_reg  = 1'b0;
+reg [ID_WIDTH-1:0]      m_axis_tid_reg    = {ID_WIDTH{1'b0}};
+reg [M_DEST_WIDTH-1:0]  m_axis_tdest_reg  = {M_DEST_WIDTH{1'b0}};
+reg [USER_WIDTH-1:0]    m_axis_tuser_reg  = {USER_WIDTH{1'b0}};
+
+reg [DATA_WIDTH-1:0]    temp_m_axis_tdata_reg  = {DATA_WIDTH{1'b0}};
+reg [KEEP_WIDTH-1:0]    temp_m_axis_tkeep_reg  = {KEEP_WIDTH{1'b0}};
+reg [M_COUNT-1:0]       temp_m_axis_tvalid_reg = {M_COUNT{1'b0}}, temp_m_axis_tvalid_next;
+reg                     temp_m_axis_tlast_reg  = 1'b0;
+reg [ID_WIDTH-1:0]      temp_m_axis_tid_reg    = {ID_WIDTH{1'b0}};
+reg [M_DEST_WIDTH-1:0]  temp_m_axis_tdest_reg  = {M_DEST_WIDTH{1'b0}};
+reg [USER_WIDTH-1:0]    temp_m_axis_tuser_reg  = {USER_WIDTH{1'b0}};
 
 // datapath control
 reg store_axis_int_to_output;
@@ -193,7 +195,7 @@ assign m_axis_tkeep  = KEEP_ENABLE ? {M_COUNT{m_axis_tkeep_reg}} : {M_COUNT*KEEP
 assign m_axis_tvalid = m_axis_tvalid_reg;
 assign m_axis_tlast  = {M_COUNT{m_axis_tlast_reg}};
 assign m_axis_tid    = ID_ENABLE   ? {M_COUNT{m_axis_tid_reg}}   : {M_COUNT*ID_WIDTH{1'b0}};
-assign m_axis_tdest  = DEST_ENABLE ? {M_COUNT{m_axis_tdest_reg}} : {M_COUNT*DEST_WIDTH{1'b0}};
+assign m_axis_tdest  = DEST_ENABLE ? {M_COUNT{m_axis_tdest_reg}} : {M_COUNT*M_DEST_WIDTH{1'b0}};
 assign m_axis_tuser  = USER_ENABLE ? {M_COUNT{m_axis_tuser_reg}} : {M_COUNT*USER_WIDTH{1'b0}};
 
 // enable ready input next cycle if output is ready or the temp reg will not be filled on the next cycle (output reg empty or no input)
diff --git a/rtl/axis_demux_wrap.py b/rtl/axis_demux_wrap.py
index 8f69561db4b91825ab739416399ddbfe37a9691c..682e31b0ebac36c1c01ae772351fad129aac7141 100755
--- a/rtl/axis_demux_wrap.py
+++ b/rtl/axis_demux_wrap.py
@@ -82,49 +82,50 @@ module {{name}} #
     parameter ID_WIDTH = 8,
     // Propagate tdest signal
     parameter DEST_ENABLE = 0,
-    // tdest signal width
-    parameter DEST_WIDTH = 8,
+    // output tdest signal width
+    parameter M_DEST_WIDTH = 1,
+    // input tdest signal width
+    parameter S_DEST_WIDTH = M_DEST_WIDTH+{{cn}},
     // Propagate tuser signal
     parameter USER_ENABLE = 1,
     // tuser signal width
     parameter USER_WIDTH = 1
 )
 (
-    input  wire                  clk,
-    input  wire                  rst,
+    input  wire                     clk,
+    input  wire                     rst,
 
     /*
      * AXI Stream input
      */
-    input  wire [DATA_WIDTH-1:0] s_axis_tdata,
-    input  wire [KEEP_WIDTH-1:0] s_axis_tkeep,
-    input  wire                  s_axis_tvalid,
-    output wire                  s_axis_tready,
-    input  wire                  s_axis_tlast,
-    input  wire [ID_WIDTH-1:0]   s_axis_tid,
-    input  wire [DEST_WIDTH-1:0] s_axis_tdest,
-    input  wire [USER_WIDTH-1:0] s_axis_tuser,
+    input  wire [DATA_WIDTH-1:0]    s_axis_tdata,
+    input  wire [KEEP_WIDTH-1:0]    s_axis_tkeep,
+    input  wire                     s_axis_tvalid,
+    output wire                     s_axis_tready,
+    input  wire                     s_axis_tlast,
+    input  wire [ID_WIDTH-1:0]      s_axis_tid,
+    input  wire [S_DEST_WIDTH-1:0]  s_axis_tdest,
+    input  wire [USER_WIDTH-1:0]    s_axis_tuser,
 
     /*
      * AXI Stream outputs
      */
 {%- for p in range(n) %}
-    output wire [DATA_WIDTH-1:0] m{{'%02d'%p}}_axis_tdata,
-    output wire [KEEP_WIDTH-1:0] m{{'%02d'%p}}_axis_tkeep,
-    output wire                  m{{'%02d'%p}}_axis_tvalid,
-    input  wire                  m{{'%02d'%p}}_axis_tready,
-    output wire                  m{{'%02d'%p}}_axis_tlast,
-    output wire [ID_WIDTH-1:0]   m{{'%02d'%p}}_axis_tid,
-    output wire [DEST_WIDTH-1:0] m{{'%02d'%p}}_axis_tdest,
-    output wire [USER_WIDTH-1:0] m{{'%02d'%p}}_axis_tuser,
-{% endfor -%}
-
+    output wire [DATA_WIDTH-1:0]    m{{'%02d'%p}}_axis_tdata,
+    output wire [KEEP_WIDTH-1:0]    m{{'%02d'%p}}_axis_tkeep,
+    output wire                     m{{'%02d'%p}}_axis_tvalid,
+    input  wire                     m{{'%02d'%p}}_axis_tready,
+    output wire                     m{{'%02d'%p}}_axis_tlast,
+    output wire [ID_WIDTH-1:0]      m{{'%02d'%p}}_axis_tid,
+    output wire [M_DEST_WIDTH-1:0]  m{{'%02d'%p}}_axis_tdest,
+    output wire [USER_WIDTH-1:0]    m{{'%02d'%p}}_axis_tuser,
+{% endfor %}
     /*
      * Control
      */
-    input  wire                  enable,
-    input  wire                  drop,
-    input  wire [{{cn-1}}:0]            select
+    input  wire                     enable,
+    input  wire                     drop,
+    input  wire [{{cn-1}}:0]               select
 );
 
 axis_demux #(
@@ -135,7 +136,8 @@ axis_demux #(
     .ID_ENABLE(ID_ENABLE),
     .ID_WIDTH(ID_WIDTH),
     .DEST_ENABLE(DEST_ENABLE),
-    .DEST_WIDTH(DEST_WIDTH),
+    .S_DEST_WIDTH(S_DEST_WIDTH),
+    .M_DEST_WIDTH(M_DEST_WIDTH),
     .USER_ENABLE(USER_ENABLE),
     .USER_WIDTH(USER_WIDTH)
 )
diff --git a/rtl/axis_ram_switch.v b/rtl/axis_ram_switch.v
index c936232c55fb13d11be96b88852628aa957e4953..a88184161b6a4490f8a5a96aae9f75cb1b845ccb 100644
--- a/rtl/axis_ram_switch.v
+++ b/rtl/axis_ram_switch.v
@@ -61,11 +61,15 @@ module axis_ram_switch #
     parameter M_KEEP_WIDTH = (M_DATA_WIDTH/8),
     // Propagate tid signal
     parameter ID_ENABLE = 0,
-    // tid signal width
-    parameter ID_WIDTH = 8,
-    // tdest signal width
+    // input tid signal width
+    parameter S_ID_WIDTH = 8,
+    // output tid signal width
+    parameter M_ID_WIDTH = S_ID_WIDTH+$clog2(S_COUNT),
+    // output tdest signal width
+    parameter M_DEST_WIDTH = 1,
+    // input tdest signal width
     // must be wide enough to uniquely address outputs
-    parameter DEST_WIDTH = $clog2(M_COUNT),
+    parameter S_DEST_WIDTH = M_DEST_WIDTH+$clog2(M_COUNT),
     // Propagate tuser signal
     parameter USER_ENABLE = 1,
     // tuser signal width
@@ -80,12 +84,12 @@ module axis_ram_switch #
     // When set, s_axis_tready is always asserted
     parameter DROP_WHEN_FULL = 0,
     // Output interface routing base tdest selection
-    // Concatenate M_COUNT DEST_WIDTH sized constants
+    // Concatenate M_COUNT S_DEST_WIDTH sized constants
     // Port selected if M_BASE <= tdest <= M_TOP
     // set to zero for default routing with tdest MSBs as port index
     parameter M_BASE = 0,
     // Output interface routing top tdest selection
-    // Concatenate M_COUNT DEST_WIDTH sized constants
+    // Concatenate M_COUNT S_DEST_WIDTH sized constants
     // Port selected if M_BASE <= tdest <= M_TOP
     // set to zero to inherit from M_BASE
     parameter M_TOP = 0,
@@ -100,39 +104,39 @@ module axis_ram_switch #
     parameter RAM_PIPELINE = 2
 )
 (
-    input  wire                            clk,
-    input  wire                            rst,
+    input  wire                             clk,
+    input  wire                             rst,
 
     /*
      * AXI Stream inputs
      */
-    input  wire [S_COUNT*S_DATA_WIDTH-1:0] s_axis_tdata,
-    input  wire [S_COUNT*S_KEEP_WIDTH-1:0] s_axis_tkeep,
-    input  wire [S_COUNT-1:0]              s_axis_tvalid,
-    output wire [S_COUNT-1:0]              s_axis_tready,
-    input  wire [S_COUNT-1:0]              s_axis_tlast,
-    input  wire [S_COUNT*ID_WIDTH-1:0]     s_axis_tid,
-    input  wire [S_COUNT*DEST_WIDTH-1:0]   s_axis_tdest,
-    input  wire [S_COUNT*USER_WIDTH-1:0]   s_axis_tuser,
+    input  wire [S_COUNT*S_DATA_WIDTH-1:0]  s_axis_tdata,
+    input  wire [S_COUNT*S_KEEP_WIDTH-1:0]  s_axis_tkeep,
+    input  wire [S_COUNT-1:0]               s_axis_tvalid,
+    output wire [S_COUNT-1:0]               s_axis_tready,
+    input  wire [S_COUNT-1:0]               s_axis_tlast,
+    input  wire [S_COUNT*S_ID_WIDTH-1:0]    s_axis_tid,
+    input  wire [S_COUNT*S_DEST_WIDTH-1:0]  s_axis_tdest,
+    input  wire [S_COUNT*USER_WIDTH-1:0]    s_axis_tuser,
 
     /*
      * AXI Stream outputs
      */
-    output wire [M_COUNT*M_DATA_WIDTH-1:0] m_axis_tdata,
-    output wire [M_COUNT*M_KEEP_WIDTH-1:0] m_axis_tkeep,
-    output wire [M_COUNT-1:0]              m_axis_tvalid,
-    input  wire [M_COUNT-1:0]              m_axis_tready,
-    output wire [M_COUNT-1:0]              m_axis_tlast,
-    output wire [M_COUNT*ID_WIDTH-1:0]     m_axis_tid,
-    output wire [M_COUNT*DEST_WIDTH-1:0]   m_axis_tdest,
-    output wire [M_COUNT*USER_WIDTH-1:0]   m_axis_tuser,
+    output wire [M_COUNT*M_DATA_WIDTH-1:0]  m_axis_tdata,
+    output wire [M_COUNT*M_KEEP_WIDTH-1:0]  m_axis_tkeep,
+    output wire [M_COUNT-1:0]               m_axis_tvalid,
+    input  wire [M_COUNT-1:0]               m_axis_tready,
+    output wire [M_COUNT-1:0]               m_axis_tlast,
+    output wire [M_COUNT*M_ID_WIDTH-1:0]    m_axis_tid,
+    output wire [M_COUNT*M_DEST_WIDTH-1:0]  m_axis_tdest,
+    output wire [M_COUNT*USER_WIDTH-1:0]    m_axis_tuser,
 
     /*
      * Status
      */
-    output wire [S_COUNT-1:0]              status_overflow,
-    output wire [S_COUNT-1:0]              status_bad_frame,
-    output wire [S_COUNT-1:0]              status_good_frame
+    output wire [S_COUNT-1:0]               status_overflow,
+    output wire [S_COUNT-1:0]               status_bad_frame,
+    output wire [S_COUNT-1:0]               status_good_frame
 );
 
 parameter CL_S_COUNT = $clog2(S_COUNT);
@@ -179,8 +183,8 @@ initial begin
         $finish;
     end
 
-    if (DEST_WIDTH < CL_M_COUNT) begin
-        $error("Error: DEST_WIDTH too small for port count (instance %m)");
+    if (S_DEST_WIDTH < CL_M_COUNT) begin
+        $error("Error: S_DEST_WIDTH too small for port count (instance %m)");
         $finish;
     end
 
@@ -188,21 +192,21 @@ initial begin
         // M_BASE is zero, route with tdest as port index
         $display("Addressing configuration for axis_switch instance %m");
         for (i = 0; i < M_COUNT; i = i + 1) begin
-            $display("%d: %08x-%08x (connect mask %b)", i, i << (DEST_WIDTH-CL_M_COUNT), ((i+1) << (DEST_WIDTH-CL_M_COUNT))-1, M_CONNECT[i*S_COUNT +: S_COUNT]);
+            $display("%d: %08x-%08x (connect mask %b)", i, i << (S_DEST_WIDTH-CL_M_COUNT), ((i+1) << (S_DEST_WIDTH-CL_M_COUNT))-1, M_CONNECT[i*S_COUNT +: S_COUNT]);
         end
 
     end else if (M_TOP == 0) begin
         // M_TOP is zero, assume equal to M_BASE
         $display("Addressing configuration for axis_switch instance %m");
         for (i = 0; i < M_COUNT; i = i + 1) begin
-            $display("%d: %08x (connect mask %b)", i, M_BASE[i*DEST_WIDTH +: DEST_WIDTH], M_CONNECT[i*S_COUNT +: S_COUNT]);
+            $display("%d: %08x (connect mask %b)", i, M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH], M_CONNECT[i*S_COUNT +: S_COUNT]);
         end
 
         for (i = 0; i < M_COUNT; i = i + 1) begin
             for (j = i+1; j < M_COUNT; j = j + 1) begin
-                if (M_BASE[i*DEST_WIDTH +: DEST_WIDTH] == M_BASE[j*DEST_WIDTH +: DEST_WIDTH]) begin
-                    $display("%d: %08x", i, M_BASE[i*DEST_WIDTH +: DEST_WIDTH]);
-                    $display("%d: %08x", j, M_BASE[j*DEST_WIDTH +: DEST_WIDTH]);
+                if (M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH] == M_BASE[j*S_DEST_WIDTH +: S_DEST_WIDTH]) begin
+                    $display("%d: %08x", i, M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH]);
+                    $display("%d: %08x", j, M_BASE[j*S_DEST_WIDTH +: S_DEST_WIDTH]);
                     $error("Error: ranges overlap (instance %m)");
                     $finish;
                 end
@@ -211,11 +215,11 @@ initial begin
     end else begin
         $display("Addressing configuration for axis_switch instance %m");
         for (i = 0; i < M_COUNT; i = i + 1) begin
-            $display("%d: %08x-%08x (connect mask %b)", i, M_BASE[i*DEST_WIDTH +: DEST_WIDTH], M_TOP[i*DEST_WIDTH +: DEST_WIDTH], M_CONNECT[i*S_COUNT +: S_COUNT]);
+            $display("%d: %08x-%08x (connect mask %b)", i, M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH], M_TOP[i*S_DEST_WIDTH +: S_DEST_WIDTH], M_CONNECT[i*S_COUNT +: S_COUNT]);
         end
 
         for (i = 0; i < M_COUNT; i = i + 1) begin
-            if (M_BASE[i*DEST_WIDTH +: DEST_WIDTH] > M_TOP[i*DEST_WIDTH +: DEST_WIDTH]) begin
+            if (M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH] > M_TOP[i*S_DEST_WIDTH +: S_DEST_WIDTH]) begin
                 $error("Error: invalid range (instance %m)");
                 $finish;
             end
@@ -223,9 +227,9 @@ initial begin
 
         for (i = 0; i < M_COUNT; i = i + 1) begin
             for (j = i+1; j < M_COUNT; j = j + 1) begin
-                if (M_BASE[i*DEST_WIDTH +: DEST_WIDTH] <= M_TOP[j*DEST_WIDTH +: DEST_WIDTH] && M_BASE[j*DEST_WIDTH +: DEST_WIDTH] <= M_TOP[i*DEST_WIDTH +: DEST_WIDTH]) begin
-                    $display("%d: %08x-%08x", i, M_BASE[i*DEST_WIDTH +: DEST_WIDTH], M_TOP[i*DEST_WIDTH +: DEST_WIDTH]);
-                    $display("%d: %08x-%08x", j, M_BASE[j*DEST_WIDTH +: DEST_WIDTH], M_TOP[j*DEST_WIDTH +: DEST_WIDTH]);
+                if (M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH] <= M_TOP[j*S_DEST_WIDTH +: S_DEST_WIDTH] && M_BASE[j*S_DEST_WIDTH +: S_DEST_WIDTH] <= M_TOP[i*S_DEST_WIDTH +: S_DEST_WIDTH]) begin
+                    $display("%d: %08x-%08x", i, M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH], M_TOP[i*S_DEST_WIDTH +: S_DEST_WIDTH]);
+                    $display("%d: %08x-%08x", j, M_BASE[j*S_DEST_WIDTH +: S_DEST_WIDTH], M_TOP[j*S_DEST_WIDTH +: S_DEST_WIDTH]);
                     $error("Error: ranges overlap (instance %m)");
                     $finish;
                 end
@@ -349,21 +353,21 @@ always @(posedge clk) begin
 end
 
 // Interconnect
-wire [S_COUNT*RAM_ADDR_WIDTH-1:0] int_cmd_addr;
-wire [S_COUNT*ADDR_WIDTH-1:0]     int_cmd_len;
-wire [S_COUNT*CMD_ADDR_WIDTH-1:0] int_cmd_id;
-wire [S_COUNT*KEEP_WIDTH-1:0]     int_cmd_tkeep;
-wire [S_COUNT*ID_WIDTH-1:0]       int_cmd_tid;
-wire [S_COUNT*DEST_WIDTH-1:0]     int_cmd_tdest;
-wire [S_COUNT*USER_WIDTH-1:0]     int_cmd_tuser;
+wire [S_COUNT*RAM_ADDR_WIDTH-1:0]  int_cmd_addr;
+wire [S_COUNT*ADDR_WIDTH-1:0]      int_cmd_len;
+wire [S_COUNT*CMD_ADDR_WIDTH-1:0]  int_cmd_id;
+wire [S_COUNT*KEEP_WIDTH-1:0]      int_cmd_tkeep;
+wire [S_COUNT*S_ID_WIDTH-1:0]      int_cmd_tid;
+wire [S_COUNT*S_DEST_WIDTH-1:0]    int_cmd_tdest;
+wire [S_COUNT*USER_WIDTH-1:0]      int_cmd_tuser;
 
-wire [S_COUNT*M_COUNT-1:0]        int_cmd_valid;
-wire [M_COUNT*S_COUNT-1:0]        int_cmd_ready;
+wire [S_COUNT*M_COUNT-1:0]         int_cmd_valid;
+wire [M_COUNT*S_COUNT-1:0]         int_cmd_ready;
 
-wire [M_COUNT*CMD_ADDR_WIDTH-1:0] int_cmd_status_id;
+wire [M_COUNT*CMD_ADDR_WIDTH-1:0]  int_cmd_status_id;
 
-wire [M_COUNT*S_COUNT-1:0]        int_cmd_status_valid;
-wire [S_COUNT*M_COUNT-1:0]        int_cmd_status_ready;
+wire [M_COUNT*S_COUNT-1:0]         int_cmd_status_valid;
+wire [S_COUNT*M_COUNT-1:0]         int_cmd_status_ready;
 
 generate
 
@@ -371,14 +375,14 @@ generate
 
     for (m = 0; m < S_COUNT; m = m + 1) begin : s_ifaces
 
-        wire [DATA_WIDTH-1:0] port_axis_tdata;
-        wire [KEEP_WIDTH-1:0] port_axis_tkeep;
-        wire                  port_axis_tvalid;
-        wire                  port_axis_tready;
-        wire                  port_axis_tlast;
-        wire [ID_WIDTH-1:0]   port_axis_tid;
-        wire [DEST_WIDTH-1:0] port_axis_tdest;
-        wire [USER_WIDTH-1:0] port_axis_tuser;
+        wire [DATA_WIDTH-1:0]    port_axis_tdata;
+        wire [KEEP_WIDTH-1:0]    port_axis_tkeep;
+        wire                     port_axis_tvalid;
+        wire                     port_axis_tready;
+        wire                     port_axis_tlast;
+        wire [S_ID_WIDTH-1:0]    port_axis_tid;
+        wire [S_DEST_WIDTH-1:0]  port_axis_tdest;
+        wire [USER_WIDTH-1:0]    port_axis_tuser;
 
         axis_adapter #(
             .S_DATA_WIDTH(S_DATA_WIDTH),
@@ -388,9 +392,9 @@ generate
             .M_KEEP_ENABLE(1),
             .M_KEEP_WIDTH(KEEP_WIDTH),
             .ID_ENABLE(ID_ENABLE),
-            .ID_WIDTH(ID_WIDTH),
+            .ID_WIDTH(S_ID_WIDTH),
             .DEST_ENABLE(1),
-            .DEST_WIDTH(DEST_WIDTH),
+            .DEST_WIDTH(S_DEST_WIDTH),
             .USER_ENABLE(USER_ENABLE),
             .USER_WIDTH(USER_WIDTH)
         )
@@ -403,8 +407,8 @@ generate
             .s_axis_tvalid(s_axis_tvalid[m]),
             .s_axis_tready(s_axis_tready[m]),
             .s_axis_tlast(s_axis_tlast[m]),
-            .s_axis_tid(s_axis_tid[ID_WIDTH*m +: ID_WIDTH]),
-            .s_axis_tdest(s_axis_tdest[DEST_WIDTH*m +: DEST_WIDTH]),
+            .s_axis_tid(s_axis_tid[S_ID_WIDTH*m +: S_ID_WIDTH]),
+            .s_axis_tdest(s_axis_tdest[S_DEST_WIDTH*m +: S_DEST_WIDTH]),
             .s_axis_tuser(s_axis_tuser[USER_WIDTH*m +: USER_WIDTH]),
             // AXI output
             .m_axis_tdata(port_axis_tdata),
@@ -442,7 +446,7 @@ generate
                             drop_next = 1'b0;
                         end else begin
                             // M_BASE is zero, route with $clog2(M_COUNT) MSBs of tdest as port index
-                            if (port_axis_tdest[DEST_WIDTH-CL_M_COUNT +: CL_M_COUNT] == k && (M_CONNECT & (1 << (m+k*S_COUNT)))) begin
+                            if (port_axis_tdest[S_DEST_WIDTH-CL_M_COUNT +: CL_M_COUNT] == k && (M_CONNECT & (1 << (m+k*S_COUNT)))) begin
                                 select_next = k;
                                 select_valid_next = 1'b1;
                                 drop_next = 1'b0;
@@ -450,13 +454,13 @@ generate
                         end
                     end else if (M_TOP == 0) begin
                         // M_TOP is zero, assume equal to M_BASE
-                        if (port_axis_tdest == M_BASE[k*DEST_WIDTH +: DEST_WIDTH] && (M_CONNECT & (1 << (m+k*S_COUNT)))) begin
+                        if (port_axis_tdest == M_BASE[k*S_DEST_WIDTH +: S_DEST_WIDTH] && (M_CONNECT & (1 << (m+k*S_COUNT)))) begin
                             select_next = k;
                             select_valid_next = 1'b1;
                             drop_next = 1'b0;
                         end
                     end else begin
-                        if (port_axis_tdest >= M_BASE[k*DEST_WIDTH +: DEST_WIDTH] && port_axis_tdest <= M_TOP[k*DEST_WIDTH +: DEST_WIDTH] && (M_CONNECT & (1 << (m+k*S_COUNT)))) begin
+                        if (port_axis_tdest >= M_BASE[k*S_DEST_WIDTH +: S_DEST_WIDTH] && port_axis_tdest <= M_TOP[k*S_DEST_WIDTH +: S_DEST_WIDTH] && (M_CONNECT & (1 << (m+k*S_COUNT)))) begin
                             select_next = k;
                             select_valid_next = 1'b1;
                             drop_next = 1'b0;
@@ -542,8 +546,8 @@ generate
         reg [ADDR_WIDTH-1:0] cmd_table_len[2**CMD_ADDR_WIDTH-1:0];
         reg [CL_M_COUNT-1:0] cmd_table_select[2**CMD_ADDR_WIDTH-1:0];
         reg [KEEP_WIDTH-1:0] cmd_table_tkeep[2**CMD_ADDR_WIDTH-1:0];
-        reg [ID_WIDTH-1:0] cmd_table_tid[2**CMD_ADDR_WIDTH-1:0];
-        reg [DEST_WIDTH-1:0] cmd_table_tdest[2**CMD_ADDR_WIDTH-1:0];
+        reg [S_ID_WIDTH-1:0] cmd_table_tid[2**CMD_ADDR_WIDTH-1:0];
+        reg [S_DEST_WIDTH-1:0] cmd_table_tdest[2**CMD_ADDR_WIDTH-1:0];
         reg [USER_WIDTH-1:0] cmd_table_tuser[2**CMD_ADDR_WIDTH-1:0];
 
         reg [CMD_ADDR_WIDTH+1-1:0] cmd_table_start_ptr_reg = 0;
@@ -552,8 +556,8 @@ generate
         reg [ADDR_WIDTH-1:0] cmd_table_start_len;
         reg [CL_M_COUNT-1:0] cmd_table_start_select;
         reg [KEEP_WIDTH-1:0] cmd_table_start_tkeep;
-        reg [ID_WIDTH-1:0] cmd_table_start_tid;
-        reg [DEST_WIDTH-1:0] cmd_table_start_tdest;
+        reg [S_ID_WIDTH-1:0] cmd_table_start_tid;
+        reg [S_DEST_WIDTH-1:0] cmd_table_start_tdest;
         reg [USER_WIDTH-1:0] cmd_table_start_tuser;
         reg cmd_table_start_en;
         reg [CMD_ADDR_WIDTH+1-1:0] cmd_table_read_ptr_reg = 0;
@@ -563,14 +567,14 @@ generate
         reg [CMD_ADDR_WIDTH+1-1:0] cmd_table_finish_ptr_reg = 0;
         reg cmd_table_finish_en;
 
-        reg [RAM_ADDR_WIDTH-1:0] cmd_addr_reg = {RAM_ADDR_WIDTH{1'b0}}, cmd_addr_next;
-        reg [ADDR_WIDTH-1:0]     cmd_len_reg = {ADDR_WIDTH{1'b0}}, cmd_len_next;
-        reg [CMD_ADDR_WIDTH-1:0] cmd_id_reg = {CMD_ADDR_WIDTH{1'b0}}, cmd_id_next;
-        reg [KEEP_WIDTH-1:0]     cmd_tkeep_reg = {KEEP_WIDTH{1'b0}}, cmd_tkeep_next;
-        reg [ID_WIDTH-1:0]       cmd_tid_reg = {ID_WIDTH{1'b0}}, cmd_tid_next;
-        reg [DEST_WIDTH-1:0]     cmd_tdest_reg = {DEST_WIDTH{1'b0}}, cmd_tdest_next;
-        reg [USER_WIDTH-1:0]     cmd_tuser_reg = {USER_WIDTH{1'b0}}, cmd_tuser_next;
-        reg [M_COUNT-1:0]        cmd_valid_reg = 0, cmd_valid_next;
+        reg [RAM_ADDR_WIDTH-1:0]  cmd_addr_reg = {RAM_ADDR_WIDTH{1'b0}}, cmd_addr_next;
+        reg [ADDR_WIDTH-1:0]      cmd_len_reg = {ADDR_WIDTH{1'b0}}, cmd_len_next;
+        reg [CMD_ADDR_WIDTH-1:0]  cmd_id_reg = {CMD_ADDR_WIDTH{1'b0}}, cmd_id_next;
+        reg [KEEP_WIDTH-1:0]      cmd_tkeep_reg = {KEEP_WIDTH{1'b0}}, cmd_tkeep_next;
+        reg [S_ID_WIDTH-1:0]      cmd_tid_reg = {S_ID_WIDTH{1'b0}}, cmd_tid_next;
+        reg [S_DEST_WIDTH-1:0]    cmd_tdest_reg = {S_DEST_WIDTH{1'b0}}, cmd_tdest_next;
+        reg [USER_WIDTH-1:0]      cmd_tuser_reg = {USER_WIDTH{1'b0}}, cmd_tuser_next;
+        reg [M_COUNT-1:0]         cmd_valid_reg = 0, cmd_valid_next;
 
         reg cmd_status_ready_reg = 1'b0, cmd_status_ready_next;
 
@@ -590,8 +594,8 @@ generate
         assign int_cmd_len[m*ADDR_WIDTH +: ADDR_WIDTH] = cmd_len_reg;
         assign int_cmd_id[m*CMD_ADDR_WIDTH +: CMD_ADDR_WIDTH] = cmd_id_reg;
         assign int_cmd_tkeep[m*KEEP_WIDTH +: KEEP_WIDTH] = cmd_tkeep_reg;
-        assign int_cmd_tid[m*ID_WIDTH +: ID_WIDTH] = cmd_tid_reg;
-        assign int_cmd_tdest[m*DEST_WIDTH +: DEST_WIDTH] = cmd_tdest_reg;
+        assign int_cmd_tid[m*S_ID_WIDTH +: S_ID_WIDTH] = cmd_tid_reg;
+        assign int_cmd_tdest[m*S_DEST_WIDTH +: S_DEST_WIDTH] = cmd_tdest_reg;
         assign int_cmd_tuser[m*USER_WIDTH +: USER_WIDTH] = cmd_tuser_reg;
         assign int_cmd_valid[m*M_COUNT +: M_COUNT] = cmd_valid_reg;
 
@@ -825,8 +829,8 @@ generate
         wire [ADDR_WIDTH-1:0]     cmd_len_mux   = int_cmd_len[grant_encoded*ADDR_WIDTH +: ADDR_WIDTH];
         wire [CMD_ADDR_WIDTH-1:0] cmd_id_mux    = int_cmd_id[grant_encoded*CMD_ADDR_WIDTH +: CMD_ADDR_WIDTH];
         wire [KEEP_WIDTH-1:0]     cmd_tkeep_mux = int_cmd_tkeep[grant_encoded*KEEP_WIDTH +: KEEP_WIDTH];
-        wire [ID_WIDTH-1:0]       cmd_tid_mux   = int_cmd_tid[grant_encoded*ID_WIDTH +: ID_WIDTH];
-        wire [DEST_WIDTH-1:0]     cmd_tdest_mux = int_cmd_tdest[grant_encoded*DEST_WIDTH +: DEST_WIDTH];
+        wire [M_ID_WIDTH-1:0]     cmd_tid_mux   = int_cmd_tid[grant_encoded*S_ID_WIDTH +: S_ID_WIDTH];
+        wire [M_DEST_WIDTH-1:0]   cmd_tdest_mux = int_cmd_tdest[grant_encoded*S_DEST_WIDTH +: S_DEST_WIDTH];
         wire [USER_WIDTH-1:0]     cmd_tuser_mux = int_cmd_tuser[grant_encoded*USER_WIDTH +: USER_WIDTH];
         wire                      cmd_valid_mux = int_cmd_valid[grant_encoded*M_COUNT+n] && grant_valid;
         wire                      cmd_ready_mux;
@@ -846,18 +850,18 @@ generate
         reg [CMD_ADDR_WIDTH-1:0] id_reg = 0, id_next;
 
         reg [KEEP_WIDTH-1:0] last_cycle_tkeep_reg = {KEEP_WIDTH{1'b0}}, last_cycle_tkeep_next;
-        reg [ID_WIDTH-1:0] tid_reg = {ID_WIDTH{1'b0}}, tid_next;
-        reg [DEST_WIDTH-1:0] tdest_reg = {DEST_WIDTH{1'b0}}, tdest_next;
+        reg [M_ID_WIDTH-1:0] tid_reg = {M_ID_WIDTH{1'b0}}, tid_next;
+        reg [M_DEST_WIDTH-1:0] tdest_reg = {M_DEST_WIDTH{1'b0}}, tdest_next;
         reg [USER_WIDTH-1:0] tuser_reg = {USER_WIDTH{1'b0}}, tuser_next;
 
-        reg [DATA_WIDTH-1:0] out_axis_tdata_reg = {DATA_WIDTH{1'b0}}, out_axis_tdata_next;
-        reg [KEEP_WIDTH-1:0] out_axis_tkeep_reg = {KEEP_WIDTH{1'b0}}, out_axis_tkeep_next;
-        reg                  out_axis_tvalid_reg = 1'b0, out_axis_tvalid_next;
-        wire                 out_axis_tready;
-        reg                  out_axis_tlast_reg = 1'b0, out_axis_tlast_next;
-        reg [ID_WIDTH-1:0]   out_axis_tid_reg   = {ID_WIDTH{1'b0}}, out_axis_tid_next;
-        reg [DEST_WIDTH-1:0] out_axis_tdest_reg = {DEST_WIDTH{1'b0}}, out_axis_tdest_next;
-        reg [USER_WIDTH-1:0] out_axis_tuser_reg = {USER_WIDTH{1'b0}}, out_axis_tuser_next;
+        reg [DATA_WIDTH-1:0]    out_axis_tdata_reg = {DATA_WIDTH{1'b0}}, out_axis_tdata_next;
+        reg [KEEP_WIDTH-1:0]    out_axis_tkeep_reg = {KEEP_WIDTH{1'b0}}, out_axis_tkeep_next;
+        reg                     out_axis_tvalid_reg = 1'b0, out_axis_tvalid_next;
+        wire                    out_axis_tready;
+        reg                     out_axis_tlast_reg = 1'b0, out_axis_tlast_next;
+        reg [M_ID_WIDTH-1:0]    out_axis_tid_reg   = {M_ID_WIDTH{1'b0}}, out_axis_tid_next;
+        reg [M_DEST_WIDTH-1:0]  out_axis_tdest_reg = {M_DEST_WIDTH{1'b0}}, out_axis_tdest_next;
+        reg [USER_WIDTH-1:0]    out_axis_tuser_reg = {USER_WIDTH{1'b0}}, out_axis_tuser_next;
 
         reg  [RAM_ADDR_WIDTH-1:0] ram_rd_addr_reg = {RAM_ADDR_WIDTH{1'b0}}, ram_rd_addr_next;
         reg                       ram_rd_en_reg = 1'b0, ram_rd_en_next;
@@ -878,8 +882,8 @@ generate
         reg [DATA_WIDTH-1:0] out_fifo_tdata[31:0];
         reg [KEEP_WIDTH-1:0] out_fifo_tkeep[31:0];
         reg out_fifo_tlast[31:0];
-        reg [ID_WIDTH-1:0] out_fifo_tid[31:0];
-        reg [DEST_WIDTH-1:0] out_fifo_tdest[31:0];
+        reg [M_ID_WIDTH-1:0] out_fifo_tid[31:0];
+        reg [M_DEST_WIDTH-1:0] out_fifo_tdest[31:0];
         reg [USER_WIDTH-1:0] out_fifo_tuser[31:0];
 
         reg [5:0] out_fifo_data_wr_ptr_reg = 0;
@@ -888,8 +892,8 @@ generate
         reg [5:0] out_fifo_ctrl_wr_ptr_reg = 0;
         reg [KEEP_WIDTH-1:0] out_fifo_ctrl_wr_tkeep;
         reg out_fifo_ctrl_wr_tlast;
-        reg [ID_WIDTH-1:0] out_fifo_ctrl_wr_tid;
-        reg [DEST_WIDTH-1:0] out_fifo_ctrl_wr_tdest;
+        reg [M_ID_WIDTH-1:0] out_fifo_ctrl_wr_tid;
+        reg [M_DEST_WIDTH-1:0] out_fifo_ctrl_wr_tdest;
         reg [USER_WIDTH-1:0] out_fifo_ctrl_wr_tuser;
         reg out_fifo_ctrl_wr_en;
         reg [5:0] out_fifo_rd_ptr_reg = 0;
@@ -1071,9 +1075,9 @@ generate
             .M_KEEP_ENABLE(M_KEEP_ENABLE),
             .M_KEEP_WIDTH(M_KEEP_WIDTH),
             .ID_ENABLE(ID_ENABLE),
-            .ID_WIDTH(ID_WIDTH),
+            .ID_WIDTH(M_ID_WIDTH),
             .DEST_ENABLE(1),
-            .DEST_WIDTH(DEST_WIDTH),
+            .DEST_WIDTH(M_DEST_WIDTH),
             .USER_ENABLE(USER_ENABLE),
             .USER_WIDTH(USER_WIDTH)
         )
@@ -1095,8 +1099,8 @@ generate
             .m_axis_tvalid(m_axis_tvalid[n]),
             .m_axis_tready(m_axis_tready[n]),
             .m_axis_tlast(m_axis_tlast[n]),
-            .m_axis_tid(m_axis_tid[ID_WIDTH*n +: ID_WIDTH]),
-            .m_axis_tdest(m_axis_tdest[DEST_WIDTH*n +: DEST_WIDTH]),
+            .m_axis_tid(m_axis_tid[M_ID_WIDTH*n +: M_ID_WIDTH]),
+            .m_axis_tdest(m_axis_tdest[M_DEST_WIDTH*n +: M_DEST_WIDTH]),
             .m_axis_tuser(m_axis_tuser[USER_WIDTH*n +: USER_WIDTH])
         );
     end // m_ifaces
diff --git a/rtl/axis_ram_switch_wrap.py b/rtl/axis_ram_switch_wrap.py
index 6e07113f41f42ac8d07f954232ec31655324f6d7..4d8e25b29ac5c85c4b5328dbc791ab7e188f73da 100755
--- a/rtl/axis_ram_switch_wrap.py
+++ b/rtl/axis_ram_switch_wrap.py
@@ -100,11 +100,15 @@ module {{name}} #
     parameter M_KEEP_WIDTH = (M_DATA_WIDTH/8),
     // Propagate tid signal
     parameter ID_ENABLE = 0,
-    // tid signal width
-    parameter ID_WIDTH = 8,
-    // tdest signal width
+    // input tid signal width
+    parameter S_ID_WIDTH = 8,
+    // output tid signal width
+    parameter M_ID_WIDTH = S_ID_WIDTH+{{cm}},
+    // output tdest signal width
+    parameter M_DEST_WIDTH = 1,
+    // input tdest signal width
     // must be wide enough to uniquely address outputs
-    parameter DEST_WIDTH = {{cn}},
+    parameter S_DEST_WIDTH = M_DEST_WIDTH+{{cn}},
     // Propagate tuser signal
     parameter USER_ENABLE = 1,
     // tuser signal width
@@ -143,27 +147,27 @@ module {{name}} #
      * AXI Stream inputs
      */
 {%- for p in range(m) %}
-    input  wire [S_DATA_WIDTH-1:0] s{{'%02d'%p}}_axis_tdata,
-    input  wire [S_KEEP_WIDTH-1:0] s{{'%02d'%p}}_axis_tkeep,
-    input  wire                    s{{'%02d'%p}}_axis_tvalid,
-    output wire                    s{{'%02d'%p}}_axis_tready,
-    input  wire                    s{{'%02d'%p}}_axis_tlast,
-    input  wire [ID_WIDTH-1:0]     s{{'%02d'%p}}_axis_tid,
-    input  wire [DEST_WIDTH-1:0]   s{{'%02d'%p}}_axis_tdest,
-    input  wire [USER_WIDTH-1:0]   s{{'%02d'%p}}_axis_tuser,
+    input  wire [S_DATA_WIDTH-1:0]  s{{'%02d'%p}}_axis_tdata,
+    input  wire [S_KEEP_WIDTH-1:0]  s{{'%02d'%p}}_axis_tkeep,
+    input  wire                     s{{'%02d'%p}}_axis_tvalid,
+    output wire                     s{{'%02d'%p}}_axis_tready,
+    input  wire                     s{{'%02d'%p}}_axis_tlast,
+    input  wire [S_ID_WIDTH-1:0]    s{{'%02d'%p}}_axis_tid,
+    input  wire [S_DEST_WIDTH-1:0]  s{{'%02d'%p}}_axis_tdest,
+    input  wire [USER_WIDTH-1:0]    s{{'%02d'%p}}_axis_tuser,
 {% endfor %}
     /*
      * AXI Stream outputs
      */
 {%- for p in range(n) %}
-    output wire [M_DATA_WIDTH-1:0] m{{'%02d'%p}}_axis_tdata,
-    output wire [M_KEEP_WIDTH-1:0] m{{'%02d'%p}}_axis_tkeep,
-    output wire                    m{{'%02d'%p}}_axis_tvalid,
-    input  wire                    m{{'%02d'%p}}_axis_tready,
-    output wire                    m{{'%02d'%p}}_axis_tlast,
-    output wire [ID_WIDTH-1:0]     m{{'%02d'%p}}_axis_tid,
-    output wire [DEST_WIDTH-1:0]   m{{'%02d'%p}}_axis_tdest,
-    output wire [USER_WIDTH-1:0]   m{{'%02d'%p}}_axis_tuser,
+    output wire [M_DATA_WIDTH-1:0]  m{{'%02d'%p}}_axis_tdata,
+    output wire [M_KEEP_WIDTH-1:0]  m{{'%02d'%p}}_axis_tkeep,
+    output wire                     m{{'%02d'%p}}_axis_tvalid,
+    input  wire                     m{{'%02d'%p}}_axis_tready,
+    output wire                     m{{'%02d'%p}}_axis_tlast,
+    output wire [M_ID_WIDTH-1:0]    m{{'%02d'%p}}_axis_tid,
+    output wire [M_DEST_WIDTH-1:0]  m{{'%02d'%p}}_axis_tdest,
+    output wire [USER_WIDTH-1:0]    m{{'%02d'%p}}_axis_tuser,
 {% endfor %}
     /*
      * Status
@@ -174,7 +178,7 @@ module {{name}} #
 );
 
 // parameter sizing helpers
-function [DEST_WIDTH-1:0] w_dw(input [DEST_WIDTH-1:0] val);
+function [S_DEST_WIDTH-1:0] w_dw(input [S_DEST_WIDTH-1:0] val);
     w_dw = val;
 endfunction
 
@@ -195,8 +199,10 @@ axis_ram_switch #(
     .M_KEEP_ENABLE(M_KEEP_ENABLE),
     .M_KEEP_WIDTH(M_KEEP_WIDTH),
     .ID_ENABLE(ID_ENABLE),
-    .ID_WIDTH(ID_WIDTH),
-    .DEST_WIDTH(DEST_WIDTH),
+    .S_ID_WIDTH(S_ID_WIDTH),
+    .M_ID_WIDTH(M_ID_WIDTH),
+    .S_DEST_WIDTH(S_DEST_WIDTH),
+    .M_DEST_WIDTH(M_DEST_WIDTH),
     .USER_ENABLE(USER_ENABLE),
     .USER_WIDTH(USER_WIDTH),
     .USER_BAD_FRAME_VALUE(USER_BAD_FRAME_VALUE),
diff --git a/rtl/axis_switch.v b/rtl/axis_switch.v
index 4688d639ffa79d21348acedccb180d49ad5c2bbe..f65368f72eaf2b13f8558ea6bd97f28f39480ad4 100644
--- a/rtl/axis_switch.v
+++ b/rtl/axis_switch.v
@@ -45,22 +45,26 @@ module axis_switch #
     parameter KEEP_WIDTH = (DATA_WIDTH/8),
     // Propagate tid signal
     parameter ID_ENABLE = 0,
-    // tid signal width
-    parameter ID_WIDTH = 8,
-    // tdest signal width
+    // input tid signal width
+    parameter S_ID_WIDTH = 8,
+    // output tid signal width
+    parameter M_ID_WIDTH = S_ID_WIDTH+$clog2(S_COUNT),
+    // output tdest signal width
+    parameter M_DEST_WIDTH = 1,
+    // input tdest signal width
     // must be wide enough to uniquely address outputs
-    parameter DEST_WIDTH = $clog2(M_COUNT),
+    parameter S_DEST_WIDTH = M_DEST_WIDTH+$clog2(M_COUNT),
     // Propagate tuser signal
     parameter USER_ENABLE = 1,
     // tuser signal width
     parameter USER_WIDTH = 1,
     // Output interface routing base tdest selection
-    // Concatenate M_COUNT DEST_WIDTH sized constants
+    // Concatenate M_COUNT S_DEST_WIDTH sized constants
     // Port selected if M_BASE <= tdest <= M_TOP
     // set to zero for default routing with tdest MSBs as port index
     parameter M_BASE = 0,
     // Output interface routing top tdest selection
-    // Concatenate M_COUNT DEST_WIDTH sized constants
+    // Concatenate M_COUNT S_DEST_WIDTH sized constants
     // Port selected if M_BASE <= tdest <= M_TOP
     // set to zero to inherit from M_BASE
     parameter M_TOP = 0,
@@ -79,32 +83,32 @@ module axis_switch #
     parameter ARB_LSB_HIGH_PRIORITY = 1
 )
 (
-    input  wire                          clk,
-    input  wire                          rst,
+    input  wire                             clk,
+    input  wire                             rst,
 
     /*
      * AXI Stream inputs
      */
-    input  wire [S_COUNT*DATA_WIDTH-1:0] s_axis_tdata,
-    input  wire [S_COUNT*KEEP_WIDTH-1:0] s_axis_tkeep,
-    input  wire [S_COUNT-1:0]            s_axis_tvalid,
-    output wire [S_COUNT-1:0]            s_axis_tready,
-    input  wire [S_COUNT-1:0]            s_axis_tlast,
-    input  wire [S_COUNT*ID_WIDTH-1:0]   s_axis_tid,
-    input  wire [S_COUNT*DEST_WIDTH-1:0] s_axis_tdest,
-    input  wire [S_COUNT*USER_WIDTH-1:0] s_axis_tuser,
+    input  wire [S_COUNT*DATA_WIDTH-1:0]    s_axis_tdata,
+    input  wire [S_COUNT*KEEP_WIDTH-1:0]    s_axis_tkeep,
+    input  wire [S_COUNT-1:0]               s_axis_tvalid,
+    output wire [S_COUNT-1:0]               s_axis_tready,
+    input  wire [S_COUNT-1:0]               s_axis_tlast,
+    input  wire [S_COUNT*S_ID_WIDTH-1:0]    s_axis_tid,
+    input  wire [S_COUNT*S_DEST_WIDTH-1:0]  s_axis_tdest,
+    input  wire [S_COUNT*USER_WIDTH-1:0]    s_axis_tuser,
 
     /*
      * AXI Stream outputs
      */
-    output wire [M_COUNT*DATA_WIDTH-1:0] m_axis_tdata,
-    output wire [M_COUNT*KEEP_WIDTH-1:0] m_axis_tkeep,
-    output wire [M_COUNT-1:0]            m_axis_tvalid,
-    input  wire [M_COUNT-1:0]            m_axis_tready,
-    output wire [M_COUNT-1:0]            m_axis_tlast,
-    output wire [M_COUNT*ID_WIDTH-1:0]   m_axis_tid,
-    output wire [M_COUNT*DEST_WIDTH-1:0] m_axis_tdest,
-    output wire [M_COUNT*USER_WIDTH-1:0] m_axis_tuser
+    output wire [M_COUNT*DATA_WIDTH-1:0]    m_axis_tdata,
+    output wire [M_COUNT*KEEP_WIDTH-1:0]    m_axis_tkeep,
+    output wire [M_COUNT-1:0]               m_axis_tvalid,
+    input  wire [M_COUNT-1:0]               m_axis_tready,
+    output wire [M_COUNT-1:0]               m_axis_tlast,
+    output wire [M_COUNT*M_ID_WIDTH-1:0]    m_axis_tid,
+    output wire [M_COUNT*M_DEST_WIDTH-1:0]  m_axis_tdest,
+    output wire [M_COUNT*USER_WIDTH-1:0]    m_axis_tuser
 );
 
 parameter CL_S_COUNT = $clog2(S_COUNT);
@@ -114,8 +118,8 @@ integer i, j;
 
 // check configuration
 initial begin
-    if (DEST_WIDTH < CL_M_COUNT) begin
-        $error("Error: DEST_WIDTH too small for port count (instance %m)");
+    if (S_DEST_WIDTH < CL_M_COUNT) begin
+        $error("Error: S_DEST_WIDTH too small for port count (instance %m)");
         $finish;
     end
 
@@ -123,21 +127,21 @@ initial begin
         // M_BASE is zero, route with tdest as port index
         $display("Addressing configuration for axis_switch instance %m");
         for (i = 0; i < M_COUNT; i = i + 1) begin
-            $display("%d: %08x-%08x (connect mask %b)", i, i << (DEST_WIDTH-CL_M_COUNT), ((i+1) << (DEST_WIDTH-CL_M_COUNT))-1, M_CONNECT[i*S_COUNT +: S_COUNT]);
+            $display("%d: %08x-%08x (connect mask %b)", i, i << (S_DEST_WIDTH-CL_M_COUNT), ((i+1) << (S_DEST_WIDTH-CL_M_COUNT))-1, M_CONNECT[i*S_COUNT +: S_COUNT]);
         end
 
     end else if (M_TOP == 0) begin
         // M_TOP is zero, assume equal to M_BASE
         $display("Addressing configuration for axis_switch instance %m");
         for (i = 0; i < M_COUNT; i = i + 1) begin
-            $display("%d: %08x (connect mask %b)", i, M_BASE[i*DEST_WIDTH +: DEST_WIDTH], M_CONNECT[i*S_COUNT +: S_COUNT]);
+            $display("%d: %08x (connect mask %b)", i, M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH], M_CONNECT[i*S_COUNT +: S_COUNT]);
         end
 
         for (i = 0; i < M_COUNT; i = i + 1) begin
             for (j = i+1; j < M_COUNT; j = j + 1) begin
-                if (M_BASE[i*DEST_WIDTH +: DEST_WIDTH] == M_BASE[j*DEST_WIDTH +: DEST_WIDTH]) begin
-                    $display("%d: %08x", i, M_BASE[i*DEST_WIDTH +: DEST_WIDTH]);
-                    $display("%d: %08x", j, M_BASE[j*DEST_WIDTH +: DEST_WIDTH]);
+                if (M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH] == M_BASE[j*S_DEST_WIDTH +: S_DEST_WIDTH]) begin
+                    $display("%d: %08x", i, M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH]);
+                    $display("%d: %08x", j, M_BASE[j*S_DEST_WIDTH +: S_DEST_WIDTH]);
                     $error("Error: ranges overlap (instance %m)");
                     $finish;
                 end
@@ -146,11 +150,11 @@ initial begin
     end else begin
         $display("Addressing configuration for axis_switch instance %m");
         for (i = 0; i < M_COUNT; i = i + 1) begin
-            $display("%d: %08x-%08x (connect mask %b)", i, M_BASE[i*DEST_WIDTH +: DEST_WIDTH], M_TOP[i*DEST_WIDTH +: DEST_WIDTH], M_CONNECT[i*S_COUNT +: S_COUNT]);
+            $display("%d: %08x-%08x (connect mask %b)", i, M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH], M_TOP[i*S_DEST_WIDTH +: S_DEST_WIDTH], M_CONNECT[i*S_COUNT +: S_COUNT]);
         end
 
         for (i = 0; i < M_COUNT; i = i + 1) begin
-            if (M_BASE[i*DEST_WIDTH +: DEST_WIDTH] > M_TOP[i*DEST_WIDTH +: DEST_WIDTH]) begin
+            if (M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH] > M_TOP[i*S_DEST_WIDTH +: S_DEST_WIDTH]) begin
                 $error("Error: invalid range (instance %m)");
                 $finish;
             end
@@ -158,9 +162,9 @@ initial begin
 
         for (i = 0; i < M_COUNT; i = i + 1) begin
             for (j = i+1; j < M_COUNT; j = j + 1) begin
-                if (M_BASE[i*DEST_WIDTH +: DEST_WIDTH] <= M_TOP[j*DEST_WIDTH +: DEST_WIDTH] && M_BASE[j*DEST_WIDTH +: DEST_WIDTH] <= M_TOP[i*DEST_WIDTH +: DEST_WIDTH]) begin
-                    $display("%d: %08x-%08x", i, M_BASE[i*DEST_WIDTH +: DEST_WIDTH], M_TOP[i*DEST_WIDTH +: DEST_WIDTH]);
-                    $display("%d: %08x-%08x", j, M_BASE[j*DEST_WIDTH +: DEST_WIDTH], M_TOP[j*DEST_WIDTH +: DEST_WIDTH]);
+                if (M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH] <= M_TOP[j*S_DEST_WIDTH +: S_DEST_WIDTH] && M_BASE[j*S_DEST_WIDTH +: S_DEST_WIDTH] <= M_TOP[i*S_DEST_WIDTH +: S_DEST_WIDTH]) begin
+                    $display("%d: %08x-%08x", i, M_BASE[i*S_DEST_WIDTH +: S_DEST_WIDTH], M_TOP[i*S_DEST_WIDTH +: S_DEST_WIDTH]);
+                    $display("%d: %08x-%08x", j, M_BASE[j*S_DEST_WIDTH +: S_DEST_WIDTH], M_TOP[j*S_DEST_WIDTH +: S_DEST_WIDTH]);
                     $error("Error: ranges overlap (instance %m)");
                     $finish;
                 end
@@ -169,17 +173,17 @@ initial begin
     end
 end
 
-wire [S_COUNT*DATA_WIDTH-1:0] int_s_axis_tdata;
-wire [S_COUNT*KEEP_WIDTH-1:0] int_s_axis_tkeep;
-wire [S_COUNT-1:0]            int_s_axis_tvalid;
-wire [S_COUNT-1:0]            int_s_axis_tready;
-wire [S_COUNT-1:0]            int_s_axis_tlast;
-wire [S_COUNT*ID_WIDTH-1:0]   int_s_axis_tid;
-wire [S_COUNT*DEST_WIDTH-1:0] int_s_axis_tdest;
-wire [S_COUNT*USER_WIDTH-1:0] int_s_axis_tuser;
+wire [S_COUNT*DATA_WIDTH-1:0]    int_s_axis_tdata;
+wire [S_COUNT*KEEP_WIDTH-1:0]    int_s_axis_tkeep;
+wire [S_COUNT-1:0]               int_s_axis_tvalid;
+wire [S_COUNT-1:0]               int_s_axis_tready;
+wire [S_COUNT-1:0]               int_s_axis_tlast;
+wire [S_COUNT*S_ID_WIDTH-1:0]    int_s_axis_tid;
+wire [S_COUNT*S_DEST_WIDTH-1:0]  int_s_axis_tdest;
+wire [S_COUNT*USER_WIDTH-1:0]    int_s_axis_tuser;
 
-wire [S_COUNT*M_COUNT-1:0]    int_axis_tvalid;
-wire [M_COUNT*S_COUNT-1:0]    int_axis_tready;
+wire [S_COUNT*M_COUNT-1:0]       int_axis_tvalid;
+wire [M_COUNT*S_COUNT-1:0]       int_axis_tready;
 
 generate
 
@@ -212,7 +216,7 @@ generate
                             drop_next = 1'b0;
                         end else begin
                             // M_BASE is zero, route with $clog2(M_COUNT) MSBs of tdest as port index
-                            if (int_s_axis_tdest[m*DEST_WIDTH+(DEST_WIDTH-CL_M_COUNT) +: CL_M_COUNT] == k && (M_CONNECT & (1 << (m+k*S_COUNT)))) begin
+                            if (int_s_axis_tdest[m*S_DEST_WIDTH+(S_DEST_WIDTH-CL_M_COUNT) +: CL_M_COUNT] == k && (M_CONNECT & (1 << (m+k*S_COUNT)))) begin
                                 select_next = k;
                                 select_valid_next = 1'b1;
                                 drop_next = 1'b0;
@@ -220,13 +224,13 @@ generate
                         end
                     end else if (M_TOP == 0) begin
                         // M_TOP is zero, assume equal to M_BASE
-                        if (int_s_axis_tdest[m*DEST_WIDTH +: DEST_WIDTH] == M_BASE[k*DEST_WIDTH +: DEST_WIDTH] && (M_CONNECT & (1 << (m+k*S_COUNT)))) begin
+                        if (int_s_axis_tdest[m*S_DEST_WIDTH +: S_DEST_WIDTH] == M_BASE[k*S_DEST_WIDTH +: S_DEST_WIDTH] && (M_CONNECT & (1 << (m+k*S_COUNT)))) begin
                             select_next = k;
                             select_valid_next = 1'b1;
                             drop_next = 1'b0;
                         end
                     end else begin
-                        if (int_s_axis_tdest[m*DEST_WIDTH +: DEST_WIDTH] >= M_BASE[k*DEST_WIDTH +: DEST_WIDTH] && int_s_axis_tdest[m*DEST_WIDTH +: DEST_WIDTH] <= M_TOP[k*DEST_WIDTH +: DEST_WIDTH] && (M_CONNECT & (1 << (m+k*S_COUNT)))) begin
+                        if (int_s_axis_tdest[m*S_DEST_WIDTH +: S_DEST_WIDTH] >= M_BASE[k*S_DEST_WIDTH +: S_DEST_WIDTH] && int_s_axis_tdest[m*S_DEST_WIDTH +: S_DEST_WIDTH] <= M_TOP[k*S_DEST_WIDTH +: S_DEST_WIDTH] && (M_CONNECT & (1 << (m+k*S_COUNT)))) begin
                             select_next = k;
                             select_valid_next = 1'b1;
                             drop_next = 1'b0;
@@ -258,9 +262,9 @@ generate
             .KEEP_WIDTH(KEEP_WIDTH),
             .LAST_ENABLE(1),
             .ID_ENABLE(ID_ENABLE),
-            .ID_WIDTH(ID_WIDTH),
+            .ID_WIDTH(S_ID_WIDTH),
             .DEST_ENABLE(1),
-            .DEST_WIDTH(DEST_WIDTH),
+            .DEST_WIDTH(S_DEST_WIDTH),
             .USER_ENABLE(USER_ENABLE),
             .USER_WIDTH(USER_WIDTH),
             .REG_TYPE(S_REG_TYPE)
@@ -274,8 +278,8 @@ generate
             .s_axis_tvalid(s_axis_tvalid[m]),
             .s_axis_tready(s_axis_tready[m]),
             .s_axis_tlast(s_axis_tlast[m]),
-            .s_axis_tid(s_axis_tid[m*ID_WIDTH +: ID_WIDTH]),
-            .s_axis_tdest(s_axis_tdest[m*DEST_WIDTH +: DEST_WIDTH]),
+            .s_axis_tid(s_axis_tid[m*S_ID_WIDTH +: S_ID_WIDTH]),
+            .s_axis_tdest(s_axis_tdest[m*S_DEST_WIDTH +: S_DEST_WIDTH]),
             .s_axis_tuser(s_axis_tuser[m*USER_WIDTH +: USER_WIDTH]),
             // AXI output
             .m_axis_tdata(int_s_axis_tdata[m*DATA_WIDTH +: DATA_WIDTH]),
@@ -283,8 +287,8 @@ generate
             .m_axis_tvalid(int_s_axis_tvalid[m]),
             .m_axis_tready(int_s_axis_tready[m]),
             .m_axis_tlast(int_s_axis_tlast[m]),
-            .m_axis_tid(int_s_axis_tid[m*ID_WIDTH +: ID_WIDTH]),
-            .m_axis_tdest(int_s_axis_tdest[m*DEST_WIDTH +: DEST_WIDTH]),
+            .m_axis_tid(int_s_axis_tid[m*S_ID_WIDTH +: S_ID_WIDTH]),
+            .m_axis_tdest(int_s_axis_tdest[m*S_DEST_WIDTH +: S_DEST_WIDTH]),
             .m_axis_tuser(int_s_axis_tuser[m*USER_WIDTH +: USER_WIDTH])
         );
     end // s_ifaces
@@ -316,20 +320,20 @@ generate
         );
 
         // mux
-        wire [DATA_WIDTH-1:0] s_axis_tdata_mux   = int_s_axis_tdata[grant_encoded*DATA_WIDTH +: DATA_WIDTH];
-        wire [KEEP_WIDTH-1:0] s_axis_tkeep_mux   = int_s_axis_tkeep[grant_encoded*KEEP_WIDTH +: KEEP_WIDTH];
-        wire                  s_axis_tvalid_mux  = int_axis_tvalid[grant_encoded*M_COUNT+n] && grant_valid;
-        wire                  s_axis_tready_mux;
-        wire                  s_axis_tlast_mux   = int_s_axis_tlast[grant_encoded];
-        wire [ID_WIDTH-1:0]   s_axis_tid_mux     = int_s_axis_tid[grant_encoded*ID_WIDTH +: ID_WIDTH];
-        wire [DEST_WIDTH-1:0] s_axis_tdest_mux   = int_s_axis_tdest[grant_encoded*DEST_WIDTH +: DEST_WIDTH];
-        wire [USER_WIDTH-1:0] s_axis_tuser_mux   = int_s_axis_tuser[grant_encoded*USER_WIDTH +: USER_WIDTH];
+        wire [DATA_WIDTH-1:0]    m_axis_tdata_mux   = int_s_axis_tdata[grant_encoded*DATA_WIDTH +: DATA_WIDTH];
+        wire [KEEP_WIDTH-1:0]    m_axis_tkeep_mux   = int_s_axis_tkeep[grant_encoded*KEEP_WIDTH +: KEEP_WIDTH];
+        wire                     m_axis_tvalid_mux  = int_axis_tvalid[grant_encoded*M_COUNT+n] && grant_valid;
+        wire                     m_axis_tready_mux;
+        wire                     m_axis_tlast_mux   = int_s_axis_tlast[grant_encoded];
+        wire [M_ID_WIDTH-1:0]    m_axis_tid_mux     = int_s_axis_tid[grant_encoded*S_ID_WIDTH +: S_ID_WIDTH];
+        wire [M_DEST_WIDTH-1:0]  m_axis_tdest_mux   = int_s_axis_tdest[grant_encoded*S_DEST_WIDTH +: S_DEST_WIDTH];
+        wire [USER_WIDTH-1:0]    m_axis_tuser_mux   = int_s_axis_tuser[grant_encoded*USER_WIDTH +: USER_WIDTH];
 
-        assign int_axis_tready[n*S_COUNT +: S_COUNT] = (grant_valid && s_axis_tready_mux) << grant_encoded;
+        assign int_axis_tready[n*S_COUNT +: S_COUNT] = (grant_valid && m_axis_tready_mux) << grant_encoded;
 
         for (m = 0; m < S_COUNT; m = m + 1) begin
             assign request[m] = int_axis_tvalid[m*M_COUNT+n] && !grant[m];
-            assign acknowledge[m] = grant[m] && int_axis_tvalid[m*M_COUNT+n] && s_axis_tlast_mux && s_axis_tready_mux;
+            assign acknowledge[m] = grant[m] && int_axis_tvalid[m*M_COUNT+n] && m_axis_tlast_mux && m_axis_tready_mux;
         end
 
         // M side register
@@ -339,9 +343,9 @@ generate
             .KEEP_WIDTH(KEEP_WIDTH),
             .LAST_ENABLE(1),
             .ID_ENABLE(ID_ENABLE),
-            .ID_WIDTH(ID_WIDTH),
+            .ID_WIDTH(M_ID_WIDTH),
             .DEST_ENABLE(1),
-            .DEST_WIDTH(DEST_WIDTH),
+            .DEST_WIDTH(M_DEST_WIDTH),
             .USER_ENABLE(USER_ENABLE),
             .USER_WIDTH(USER_WIDTH),
             .REG_TYPE(M_REG_TYPE)
@@ -350,22 +354,22 @@ generate
             .clk(clk),
             .rst(rst),
             // AXI input
-            .s_axis_tdata(s_axis_tdata_mux),
-            .s_axis_tkeep(s_axis_tkeep_mux),
-            .s_axis_tvalid(s_axis_tvalid_mux),
-            .s_axis_tready(s_axis_tready_mux),
-            .s_axis_tlast(s_axis_tlast_mux),
-            .s_axis_tid(s_axis_tid_mux),
-            .s_axis_tdest(s_axis_tdest_mux),
-            .s_axis_tuser(s_axis_tuser_mux),
+            .s_axis_tdata(m_axis_tdata_mux),
+            .s_axis_tkeep(m_axis_tkeep_mux),
+            .s_axis_tvalid(m_axis_tvalid_mux),
+            .s_axis_tready(m_axis_tready_mux),
+            .s_axis_tlast(m_axis_tlast_mux),
+            .s_axis_tid(m_axis_tid_mux),
+            .s_axis_tdest(m_axis_tdest_mux),
+            .s_axis_tuser(m_axis_tuser_mux),
             // AXI output
             .m_axis_tdata(m_axis_tdata[n*DATA_WIDTH +: DATA_WIDTH]),
             .m_axis_tkeep(m_axis_tkeep[n*KEEP_WIDTH +: KEEP_WIDTH]),
             .m_axis_tvalid(m_axis_tvalid[n]),
             .m_axis_tready(m_axis_tready[n]),
             .m_axis_tlast(m_axis_tlast[n]),
-            .m_axis_tid(m_axis_tid[n*ID_WIDTH +: ID_WIDTH]),
-            .m_axis_tdest(m_axis_tdest[n*DEST_WIDTH +: DEST_WIDTH]),
+            .m_axis_tid(m_axis_tid[n*M_ID_WIDTH +: M_ID_WIDTH]),
+            .m_axis_tdest(m_axis_tdest[n*M_DEST_WIDTH +: M_DEST_WIDTH]),
             .m_axis_tuser(m_axis_tuser[n*USER_WIDTH +: USER_WIDTH])
         );
     end // m_ifaces
diff --git a/rtl/axis_switch_wrap.py b/rtl/axis_switch_wrap.py
index d7d3bbfc6620c625acae55b8f50f2b7594a8130c..87efd5494ba2b7a48731adbe7a3d316bd13581fc 100755
--- a/rtl/axis_switch_wrap.py
+++ b/rtl/axis_switch_wrap.py
@@ -84,11 +84,15 @@ module {{name}} #
     parameter KEEP_WIDTH = (DATA_WIDTH/8),
     // Propagate tid signal
     parameter ID_ENABLE = 0,
-    // tid signal width
-    parameter ID_WIDTH = 8,
-    // tdest signal width
+    // input tid signal width
+    parameter S_ID_WIDTH = 8,
+    // output tid signal width
+    parameter M_ID_WIDTH = S_ID_WIDTH+{{cm}},
+    // output tdest signal width
+    parameter M_DEST_WIDTH = 1,
+    // input tdest signal width
     // must be wide enough to uniquely address outputs
-    parameter DEST_WIDTH = {{cn}},
+    parameter S_DEST_WIDTH = M_DEST_WIDTH+{{cn}},
     // Propagate tuser signal
     parameter USER_ENABLE = 1,
     // tuser signal width
@@ -122,32 +126,32 @@ module {{name}} #
      * AXI Stream inputs
      */
 {%- for p in range(m) %}
-    input  wire [DATA_WIDTH-1:0] s{{'%02d'%p}}_axis_tdata,
-    input  wire [KEEP_WIDTH-1:0] s{{'%02d'%p}}_axis_tkeep,
-    input  wire                  s{{'%02d'%p}}_axis_tvalid,
-    output wire                  s{{'%02d'%p}}_axis_tready,
-    input  wire                  s{{'%02d'%p}}_axis_tlast,
-    input  wire [ID_WIDTH-1:0]   s{{'%02d'%p}}_axis_tid,
-    input  wire [DEST_WIDTH-1:0] s{{'%02d'%p}}_axis_tdest,
-    input  wire [USER_WIDTH-1:0] s{{'%02d'%p}}_axis_tuser,
+    input  wire [DATA_WIDTH-1:0]    s{{'%02d'%p}}_axis_tdata,
+    input  wire [KEEP_WIDTH-1:0]    s{{'%02d'%p}}_axis_tkeep,
+    input  wire                     s{{'%02d'%p}}_axis_tvalid,
+    output wire                     s{{'%02d'%p}}_axis_tready,
+    input  wire                     s{{'%02d'%p}}_axis_tlast,
+    input  wire [S_ID_WIDTH-1:0]    s{{'%02d'%p}}_axis_tid,
+    input  wire [S_DEST_WIDTH-1:0]  s{{'%02d'%p}}_axis_tdest,
+    input  wire [USER_WIDTH-1:0]    s{{'%02d'%p}}_axis_tuser,
 {% endfor %}
     /*
      * AXI Stream outputs
      */
 {%- for p in range(n) %}
-    output wire [DATA_WIDTH-1:0] m{{'%02d'%p}}_axis_tdata,
-    output wire [KEEP_WIDTH-1:0] m{{'%02d'%p}}_axis_tkeep,
-    output wire                  m{{'%02d'%p}}_axis_tvalid,
-    input  wire                  m{{'%02d'%p}}_axis_tready,
-    output wire                  m{{'%02d'%p}}_axis_tlast,
-    output wire [ID_WIDTH-1:0]   m{{'%02d'%p}}_axis_tid,
-    output wire [DEST_WIDTH-1:0] m{{'%02d'%p}}_axis_tdest,
-    output wire [USER_WIDTH-1:0] m{{'%02d'%p}}_axis_tuser{% if not loop.last %},{% endif %}
+    output wire [DATA_WIDTH-1:0]    m{{'%02d'%p}}_axis_tdata,
+    output wire [KEEP_WIDTH-1:0]    m{{'%02d'%p}}_axis_tkeep,
+    output wire                     m{{'%02d'%p}}_axis_tvalid,
+    input  wire                     m{{'%02d'%p}}_axis_tready,
+    output wire                     m{{'%02d'%p}}_axis_tlast,
+    output wire [M_ID_WIDTH-1:0]    m{{'%02d'%p}}_axis_tid,
+    output wire [M_DEST_WIDTH-1:0]  m{{'%02d'%p}}_axis_tdest,
+    output wire [USER_WIDTH-1:0]    m{{'%02d'%p}}_axis_tuser{% if not loop.last %},{% endif %}
 {% endfor -%}
 );
 
 // parameter sizing helpers
-function [DEST_WIDTH-1:0] w_dw(input [DEST_WIDTH-1:0] val);
+function [S_DEST_WIDTH-1:0] w_dw(input [S_DEST_WIDTH-1:0] val);
     w_dw = val;
 endfunction
 
@@ -162,8 +166,10 @@ axis_switch #(
     .KEEP_ENABLE(KEEP_ENABLE),
     .KEEP_WIDTH(KEEP_WIDTH),
     .ID_ENABLE(ID_ENABLE),
-    .ID_WIDTH(ID_WIDTH),
-    .DEST_WIDTH(DEST_WIDTH),
+    .S_ID_WIDTH(S_ID_WIDTH),
+    .M_ID_WIDTH(M_ID_WIDTH),
+    .S_DEST_WIDTH(S_DEST_WIDTH),
+    .M_DEST_WIDTH(M_DEST_WIDTH),
     .USER_ENABLE(USER_ENABLE),
     .USER_WIDTH(USER_WIDTH),
     .M_BASE({ {% for p in range(n-1,-1,-1) %}w_dw(M{{'%02d'%p}}_BASE){% if not loop.last %}, {% endif %}{% endfor %} }),
diff --git a/tb/axis_arb_mux/Makefile b/tb/axis_arb_mux/Makefile
index 3e3dc20a6feaa63e22faac2d8740ff5124568abf..7d91e58974f2ade0cb0003028043466b67e21daa 100644
--- a/tb/axis_arb_mux/Makefile
+++ b/tb/axis_arb_mux/Makefile
@@ -42,7 +42,8 @@ export PARAM_DATA_WIDTH ?= 8
 export PARAM_KEEP_ENABLE ?= $(shell expr $(PARAM_DATA_WIDTH) \> 8 )
 export PARAM_KEEP_WIDTH ?= $(shell expr $(PARAM_DATA_WIDTH) / 8 )
 export PARAM_ID_ENABLE ?= 1
-export PARAM_ID_WIDTH ?= 8
+export PARAM_S_ID_WIDTH ?= 8
+export PARAM_M_ID_WIDTH ?= $(shell python -c "print($(PARAM_S_ID_WIDTH) + ($(PARAM_PORTS)-1).bit_length())")
 export PARAM_DEST_ENABLE ?= 1
 export PARAM_DEST_WIDTH ?= 8
 export PARAM_USER_ENABLE ?= 1
@@ -58,7 +59,8 @@ ifeq ($(SIM), icarus)
 	COMPILE_ARGS += -P $(TOPLEVEL).KEEP_ENABLE=$(PARAM_KEEP_ENABLE)
 	COMPILE_ARGS += -P $(TOPLEVEL).KEEP_WIDTH=$(PARAM_KEEP_WIDTH)
 	COMPILE_ARGS += -P $(TOPLEVEL).ID_ENABLE=$(PARAM_ID_ENABLE)
-	COMPILE_ARGS += -P $(TOPLEVEL).ID_WIDTH=$(PARAM_ID_WIDTH)
+	COMPILE_ARGS += -P $(TOPLEVEL).M_ID_WIDTH=$(PARAM_M_ID_WIDTH)
+	COMPILE_ARGS += -P $(TOPLEVEL).S_ID_WIDTH=$(PARAM_S_ID_WIDTH)
 	COMPILE_ARGS += -P $(TOPLEVEL).DEST_ENABLE=$(PARAM_DEST_ENABLE)
 	COMPILE_ARGS += -P $(TOPLEVEL).DEST_WIDTH=$(PARAM_DEST_WIDTH)
 	COMPILE_ARGS += -P $(TOPLEVEL).USER_ENABLE=$(PARAM_USER_ENABLE)
@@ -78,7 +80,8 @@ else ifeq ($(SIM), verilator)
 	COMPILE_ARGS += -GKEEP_ENABLE=$(PARAM_KEEP_ENABLE)
 	COMPILE_ARGS += -GKEEP_WIDTH=$(PARAM_KEEP_WIDTH)
 	COMPILE_ARGS += -GID_ENABLE=$(PARAM_ID_ENABLE)
-	COMPILE_ARGS += -GID_WIDTH=$(PARAM_ID_WIDTH)
+	COMPILE_ARGS += -GM_ID_WIDTH=$(PARAM_M_ID_WIDTH)
+	COMPILE_ARGS += -GS_ID_WIDTH=$(PARAM_S_ID_WIDTH)
 	COMPILE_ARGS += -GDEST_ENABLE=$(PARAM_DEST_ENABLE)
 	COMPILE_ARGS += -GDEST_WIDTH=$(PARAM_DEST_WIDTH)
 	COMPILE_ARGS += -GUSER_ENABLE=$(PARAM_USER_ENABLE)
diff --git a/tb/axis_arb_mux/test_axis_arb_mux.py b/tb/axis_arb_mux/test_axis_arb_mux.py
index b67a410f794c1403e72848152f3401216b1344c3..d69b0b156094f73d578de87b9f8fcfe48d7cef54 100644
--- a/tb/axis_arb_mux/test_axis_arb_mux.py
+++ b/tb/axis_arb_mux/test_axis_arb_mux.py
@@ -316,7 +316,8 @@ def test_axis_arb_mux(request, ports, data_width, round_robin):
     parameters['KEEP_ENABLE'] = int(parameters['DATA_WIDTH'] > 8)
     parameters['KEEP_WIDTH'] = parameters['DATA_WIDTH'] // 8
     parameters['ID_ENABLE'] = 1
-    parameters['ID_WIDTH'] = 8
+    parameters['S_ID_WIDTH'] = 8
+    parameters['M_ID_WIDTH'] = parameters['S_ID_WIDTH'] + (ports-1).bit_length()
     parameters['DEST_ENABLE'] = 1
     parameters['DEST_WIDTH'] = 8
     parameters['USER_ENABLE'] = 1
diff --git a/tb/axis_demux/Makefile b/tb/axis_demux/Makefile
index 8f529ef7781097596096d0da8ce9a3df5c39c52f..a7e2a1dd332eab31a15ed88ac063a9cf0945cfcb 100644
--- a/tb/axis_demux/Makefile
+++ b/tb/axis_demux/Makefile
@@ -42,7 +42,8 @@ export PARAM_KEEP_WIDTH ?= $(shell expr $(PARAM_DATA_WIDTH) / 8 )
 export PARAM_ID_ENABLE ?= 1
 export PARAM_ID_WIDTH ?= 8
 export PARAM_DEST_ENABLE ?= 1
-export PARAM_DEST_WIDTH ?= 8
+export PARAM_M_DEST_WIDTH ?= 8
+export PARAM_S_DEST_WIDTH ?= $(shell python -c "print($(PARAM_M_DEST_WIDTH) + ($(PARAM_PORTS)-1).bit_length())")
 export PARAM_USER_ENABLE ?= 1
 export PARAM_USER_WIDTH ?= 1
 
@@ -55,7 +56,8 @@ ifeq ($(SIM), icarus)
 	COMPILE_ARGS += -P $(TOPLEVEL).ID_ENABLE=$(PARAM_ID_ENABLE)
 	COMPILE_ARGS += -P $(TOPLEVEL).ID_WIDTH=$(PARAM_ID_WIDTH)
 	COMPILE_ARGS += -P $(TOPLEVEL).DEST_ENABLE=$(PARAM_DEST_ENABLE)
-	COMPILE_ARGS += -P $(TOPLEVEL).DEST_WIDTH=$(PARAM_DEST_WIDTH)
+	COMPILE_ARGS += -P $(TOPLEVEL).S_DEST_WIDTH=$(PARAM_S_DEST_WIDTH)
+	COMPILE_ARGS += -P $(TOPLEVEL).M_DEST_WIDTH=$(PARAM_M_DEST_WIDTH)
 	COMPILE_ARGS += -P $(TOPLEVEL).USER_ENABLE=$(PARAM_USER_ENABLE)
 	COMPILE_ARGS += -P $(TOPLEVEL).USER_WIDTH=$(PARAM_USER_WIDTH)
 
@@ -72,7 +74,8 @@ else ifeq ($(SIM), verilator)
 	COMPILE_ARGS += -GID_ENABLE=$(PARAM_ID_ENABLE)
 	COMPILE_ARGS += -GID_WIDTH=$(PARAM_ID_WIDTH)
 	COMPILE_ARGS += -GDEST_ENABLE=$(PARAM_DEST_ENABLE)
-	COMPILE_ARGS += -GDEST_WIDTH=$(PARAM_DEST_WIDTH)
+	COMPILE_ARGS += -GS_DEST_WIDTH=$(PARAM_S_DEST_WIDTH)
+	COMPILE_ARGS += -GM_DEST_WIDTH=$(PARAM_M_DEST_WIDTH)
 	COMPILE_ARGS += -GUSER_ENABLE=$(PARAM_USER_ENABLE)
 	COMPILE_ARGS += -GUSER_WIDTH=$(PARAM_USER_WIDTH)
 
diff --git a/tb/axis_demux/test_axis_demux.py b/tb/axis_demux/test_axis_demux.py
index dbd9e6098bb1ec3d486e88e73d7804169749f566..1231bf4e7f5eefc4684e8a7acc0fd6ec9cd88b3e 100644
--- a/tb/axis_demux/test_axis_demux.py
+++ b/tb/axis_demux/test_axis_demux.py
@@ -183,7 +183,8 @@ def test_axis_demux(request, ports, data_width):
     parameters['ID_ENABLE'] = 1
     parameters['ID_WIDTH'] = 8
     parameters['DEST_ENABLE'] = 1
-    parameters['DEST_WIDTH'] = 8
+    parameters['M_DEST_WIDTH'] = 8
+    parameters['S_DEST_WIDTH'] = parameters['M_DEST_WIDTH'] + (ports-1).bit_length()
     parameters['USER_ENABLE'] = 1
     parameters['USER_WIDTH'] = 1
 
diff --git a/tb/axis_ram_switch/Makefile b/tb/axis_ram_switch/Makefile
index c5e6aa307f218e1f7f55d3a286ec099531f6e47e..49866a772443351933630b17d948865b1c02dfbf 100644
--- a/tb/axis_ram_switch/Makefile
+++ b/tb/axis_ram_switch/Makefile
@@ -50,8 +50,10 @@ export PARAM_M_DATA_WIDTH ?= 8
 export PARAM_M_KEEP_ENABLE ?= $(shell expr $(PARAM_M_DATA_WIDTH) \> 8 )
 export PARAM_M_KEEP_WIDTH ?= $(shell expr $(PARAM_M_DATA_WIDTH) / 8 )
 export PARAM_ID_ENABLE ?= 1
-export PARAM_ID_WIDTH ?= 16
-export PARAM_DEST_WIDTH ?= 8
+export PARAM_S_ID_WIDTH ?= 16
+export PARAM_M_ID_WIDTH ?= $(shell python -c "print($(PARAM_S_ID_WIDTH) + ($(PARAM_S_COUNT)-1).bit_length())")
+export PARAM_M_DEST_WIDTH ?= 8
+export PARAM_S_DEST_WIDTH ?= $(shell python -c "print($(PARAM_M_DEST_WIDTH) + ($(PARAM_M_COUNT)-1).bit_length())")
 export PARAM_USER_ENABLE ?= 1
 export PARAM_USER_WIDTH ?= 1
 export PARAM_USER_BAD_FRAME_VALUE ?= 1
@@ -75,8 +77,10 @@ ifeq ($(SIM), icarus)
 	COMPILE_ARGS += -P $(TOPLEVEL).M_KEEP_ENABLE=$(PARAM_M_KEEP_ENABLE)
 	COMPILE_ARGS += -P $(TOPLEVEL).M_KEEP_WIDTH=$(PARAM_M_KEEP_WIDTH)
 	COMPILE_ARGS += -P $(TOPLEVEL).ID_ENABLE=$(PARAM_ID_ENABLE)
-	COMPILE_ARGS += -P $(TOPLEVEL).ID_WIDTH=$(PARAM_ID_WIDTH)
-	COMPILE_ARGS += -P $(TOPLEVEL).DEST_WIDTH=$(PARAM_DEST_WIDTH)
+	COMPILE_ARGS += -P $(TOPLEVEL).S_ID_WIDTH=$(PARAM_S_ID_WIDTH)
+	COMPILE_ARGS += -P $(TOPLEVEL).M_ID_WIDTH=$(PARAM_M_ID_WIDTH)
+	COMPILE_ARGS += -P $(TOPLEVEL).S_DEST_WIDTH=$(PARAM_S_DEST_WIDTH)
+	COMPILE_ARGS += -P $(TOPLEVEL).M_DEST_WIDTH=$(PARAM_M_DEST_WIDTH)
 	COMPILE_ARGS += -P $(TOPLEVEL).USER_ENABLE=$(PARAM_USER_ENABLE)
 	COMPILE_ARGS += -P $(TOPLEVEL).USER_WIDTH=$(PARAM_USER_WIDTH)
 	COMPILE_ARGS += -P $(TOPLEVEL).USER_BAD_FRAME_VALUE=$(PARAM_USER_BAD_FRAME_VALUE)
@@ -104,8 +108,10 @@ else ifeq ($(SIM), verilator)
 	COMPILE_ARGS += -GM_KEEP_ENABLE=$(PARAM_M_KEEP_ENABLE)
 	COMPILE_ARGS += -GM_KEEP_WIDTH=$(PARAM_M_KEEP_WIDTH)
 	COMPILE_ARGS += -GID_ENABLE=$(PARAM_ID_ENABLE)
-	COMPILE_ARGS += -GID_WIDTH=$(PARAM_ID_WIDTH)
-	COMPILE_ARGS += -GDEST_WIDTH=$(PARAM_DEST_WIDTH)
+	COMPILE_ARGS += -GS_ID_WIDTH=$(PARAM_S_ID_WIDTH)
+	COMPILE_ARGS += -GM_ID_WIDTH=$(PARAM_M_ID_WIDTH)
+	COMPILE_ARGS += -GS_DEST_WIDTH=$(PARAM_S_DEST_WIDTH)
+	COMPILE_ARGS += -GM_DEST_WIDTH=$(PARAM_M_DEST_WIDTH)
 	COMPILE_ARGS += -GUSER_ENABLE=$(PARAM_USER_ENABLE)
 	COMPILE_ARGS += -GUSER_WIDTH=$(PARAM_USER_WIDTH)
 	COMPILE_ARGS += -GUSER_BAD_FRAME_VALUE=$(PARAM_USER_BAD_FRAME_VALUE)
diff --git a/tb/axis_ram_switch/test_axis_ram_switch.py b/tb/axis_ram_switch/test_axis_ram_switch.py
index 244a8af1c514328313762454afd67da164352d75..16380a3913d92c503b5e16e3797cbfc715957bf3 100644
--- a/tb/axis_ram_switch/test_axis_ram_switch.py
+++ b/tb/axis_ram_switch/test_axis_ram_switch.py
@@ -335,8 +335,10 @@ def test_axis_ram_switch(request, s_count, m_count, s_data_width, m_data_width):
     parameters['M_KEEP_ENABLE'] = int(parameters['M_DATA_WIDTH'] > 8)
     parameters['M_KEEP_WIDTH'] = parameters['M_DATA_WIDTH'] // 8
     parameters['ID_ENABLE'] = 1
-    parameters['ID_WIDTH'] = 16
-    parameters['DEST_WIDTH'] = 8
+    parameters['S_ID_WIDTH'] = 16
+    parameters['M_ID_WIDTH'] = parameters['S_ID_WIDTH'] + (s_count-1).bit_length()
+    parameters['M_DEST_WIDTH'] = 8
+    parameters['S_DEST_WIDTH'] = parameters['M_DEST_WIDTH'] + (m_count-1).bit_length()
     parameters['USER_ENABLE'] = 1
     parameters['USER_WIDTH'] = 1
     parameters['USER_BAD_FRAME_VALUE'] = 1
diff --git a/tb/axis_switch/Makefile b/tb/axis_switch/Makefile
index d066c53933d30672b42dcb6566b06bfc2ec1c78d..115ea325582939c4aa9bb269aad5d5964905615c 100644
--- a/tb/axis_switch/Makefile
+++ b/tb/axis_switch/Makefile
@@ -44,8 +44,10 @@ export PARAM_DATA_WIDTH ?= 8
 export PARAM_KEEP_ENABLE ?= $(shell expr $(PARAM_DATA_WIDTH) \> 8 )
 export PARAM_KEEP_WIDTH ?= $(shell expr $(PARAM_DATA_WIDTH) / 8 )
 export PARAM_ID_ENABLE ?= 1
-export PARAM_ID_WIDTH ?= 16
-export PARAM_DEST_WIDTH ?= 8
+export PARAM_S_ID_WIDTH ?= 16
+export PARAM_M_ID_WIDTH ?= $(shell python -c "print($(PARAM_S_ID_WIDTH) + ($(PARAM_S_COUNT)-1).bit_length())")
+export PARAM_M_DEST_WIDTH ?= 8
+export PARAM_S_DEST_WIDTH ?= $(shell python -c "print($(PARAM_M_DEST_WIDTH) + ($(PARAM_M_COUNT)-1).bit_length())")
 export PARAM_USER_ENABLE ?= 1
 export PARAM_USER_WIDTH ?= 1
 export PARAM_S_REG_TYPE ?= 0
@@ -60,8 +62,10 @@ ifeq ($(SIM), icarus)
 	COMPILE_ARGS += -P $(TOPLEVEL).KEEP_ENABLE=$(PARAM_KEEP_ENABLE)
 	COMPILE_ARGS += -P $(TOPLEVEL).KEEP_WIDTH=$(PARAM_KEEP_WIDTH)
 	COMPILE_ARGS += -P $(TOPLEVEL).ID_ENABLE=$(PARAM_ID_ENABLE)
-	COMPILE_ARGS += -P $(TOPLEVEL).ID_WIDTH=$(PARAM_ID_WIDTH)
-	COMPILE_ARGS += -P $(TOPLEVEL).DEST_WIDTH=$(PARAM_DEST_WIDTH)
+	COMPILE_ARGS += -P $(TOPLEVEL).S_ID_WIDTH=$(PARAM_S_ID_WIDTH)
+	COMPILE_ARGS += -P $(TOPLEVEL).M_ID_WIDTH=$(PARAM_M_ID_WIDTH)
+	COMPILE_ARGS += -P $(TOPLEVEL).S_DEST_WIDTH=$(PARAM_S_DEST_WIDTH)
+	COMPILE_ARGS += -P $(TOPLEVEL).M_DEST_WIDTH=$(PARAM_M_DEST_WIDTH)
 	COMPILE_ARGS += -P $(TOPLEVEL).USER_ENABLE=$(PARAM_USER_ENABLE)
 	COMPILE_ARGS += -P $(TOPLEVEL).USER_WIDTH=$(PARAM_USER_WIDTH)
 	COMPILE_ARGS += -P $(TOPLEVEL).S_REG_TYPE=$(PARAM_S_REG_TYPE)
@@ -80,8 +84,10 @@ else ifeq ($(SIM), verilator)
 	COMPILE_ARGS += -GKEEP_ENABLE=$(PARAM_KEEP_ENABLE)
 	COMPILE_ARGS += -GKEEP_WIDTH=$(PARAM_KEEP_WIDTH)
 	COMPILE_ARGS += -GID_ENABLE=$(PARAM_ID_ENABLE)
-	COMPILE_ARGS += -GID_WIDTH=$(PARAM_ID_WIDTH)
-	COMPILE_ARGS += -GDEST_WIDTH=$(PARAM_DEST_WIDTH)
+	COMPILE_ARGS += -GS_ID_WIDTH=$(PARAM_S_ID_WIDTH)
+	COMPILE_ARGS += -GM_ID_WIDTH=$(PARAM_M_ID_WIDTH)
+	COMPILE_ARGS += -GS_DEST_WIDTH=$(PARAM_S_DEST_WIDTH)
+	COMPILE_ARGS += -GM_DEST_WIDTH=$(PARAM_M_DEST_WIDTH)
 	COMPILE_ARGS += -GUSER_ENABLE=$(PARAM_USER_ENABLE)
 	COMPILE_ARGS += -GUSER_WIDTH=$(PARAM_USER_WIDTH)
 	COMPILE_ARGS += -GS_REG_TYPE=$(PARAM_S_REG_TYPE)
diff --git a/tb/axis_switch/test_axis_switch.py b/tb/axis_switch/test_axis_switch.py
index ef8486426d1cf9a5ff53ab6c7333b5b1fb3e372b..3097836f365778f7726d7ad19ada518ef35891bf 100644
--- a/tb/axis_switch/test_axis_switch.py
+++ b/tb/axis_switch/test_axis_switch.py
@@ -328,8 +328,10 @@ def test_axis_switch(request, s_count, m_count, data_width):
     parameters['KEEP_ENABLE'] = int(parameters['DATA_WIDTH'] > 8)
     parameters['KEEP_WIDTH'] = parameters['DATA_WIDTH'] // 8
     parameters['ID_ENABLE'] = 1
-    parameters['ID_WIDTH'] = 16
-    parameters['DEST_WIDTH'] = 8
+    parameters['S_ID_WIDTH'] = 16
+    parameters['M_ID_WIDTH'] = parameters['S_ID_WIDTH'] + (s_count-1).bit_length()
+    parameters['M_DEST_WIDTH'] = 8
+    parameters['S_DEST_WIDTH'] = parameters['M_DEST_WIDTH'] + (m_count-1).bit_length()
     parameters['USER_ENABLE'] = 1
     parameters['USER_WIDTH'] = 1
     parameters['S_REG_TYPE'] = 0