[SPI] re-add high-speed mode (#730)

CHANGELOG.md

@@ -32,6 +32,7 @@ mimpid = 0x01040312 -> Version 01.04.03.12 -> v1.4.3.12
 
 | Date (*dd.mm.yyyy*) | Version | Comment |
 |:-------------------:|:-------:|:--------|
+| 18.11.2023 | 1.9.1.1 | (re-)add SPI high-speed mode, :bug: fix bug in SPI shift register - introduced in v1.9.0.9; [730](https://github.com/stnolting/neorv32/pull/730) |
 | 14.11.2023 | [**:rocket:1.9.1**](https://github.com/stnolting/neorv32/releases/tag/v1.9.1) | **New release** |
 | 11.11.2023 | 1.9.0.9 | :test_tube: add full hardware reset for **all** flip flops in CPU/processor; [#724](https://github.com/stnolting/neorv32/pull/724) |
 | 09.11.2023 | 1.9.0.8 | minor rtl code cleanups; [#723](https://github.com/stnolting/neorv32/pull/723) |

docs/datasheet/soc_spi.adoc

@@ -68,8 +68,7 @@ image::SPI_timing_diagram2.wikimedia.png[]
 
 The SPI clock frequency (`spi_clk_o`) is programmed by the 3-bit `SPI_CTRL_PRSCx` clock prescaler for a coarse clock selection
 and a 4-bit clock divider `SPI_CTRL_CDIVx` for a fine clock configuration.
-
-The following clock prescalers (_SPI_CTRL_PRSCx_) are available:
+The following clock prescalers (`SPI_CTRL_PRSCx`) are available:
 
 .SPI prescaler configuration
 [cols="<4,^1,^1,^1,^1,^1,^1,^1,^1"]
@@ -79,7 +78,7 @@ The following clock prescalers (_SPI_CTRL_PRSCx_) are available:
 | Resulting `clock_prescaler` |       2 |       4 |       8 |      64 |     128 |    1024 |    2048 |    4096
 |=======================
 
-Based on the programmen clock configuration, the actual SPI clock frequency f~SPI~ is derived
+Based on the programmed clock configuration, the actual SPI clock frequency f~SPI~ is derived
 from the processor's main clock f~main~ according to the following equation:
 
 _**f~SPI~**_ = _f~main~[Hz]_ / (2 * `clock_prescaler` * (1 + `SPI_CTRL_CDIVx`))
@@ -88,6 +87,17 @@ Hence, the maximum SPI clock is f~main~ / 4 and the lowest SPI clock is f~main~
 symmetric having a duty cycle of 50%.
 
 
+**High-Speed Mode**
+
+The SPI provides a high-speed mode to further boost the maximum SPI clock frequency. When enabled via the control
+register's `SPI_CTRL_HIGHSPEED` bit the clock prescaler configuration (`SPI_CTRL_PRSCx` bits) is overridden setting it
+to a minimal factor of 1. However, the clock speed can still be fine-tuned using the `SPI_CTRL_CDIVx` bits.
+
+_**f~SPI~**_ = _f~main~[Hz]_ / (2 * 1 * (1 + `SPI_CTRL_CDIVx`))
+
+Hence, the maximum SPI clock when in high-speed mode is f~main~ / 2.
+
+
 **SPI Interrupt**
 
 The SPI module provides a set of programmable interrupt conditions based on the level of the RX/TX FIFO. The different
@@ -107,14 +117,15 @@ Furthermore, an active SPI interrupt has to be explicitly cleared again by writi
 [options="header",grid="all"]
 |=======================
 | Address | Name [C] | Bit(s), Name [C] | R/W | Function
-.18+<| `0xfffff800` .18+<| `CTRL` <|`0`     `SPI_CTRL_EN`                           ^| r/w <| SPI module enable
+.19+<| `0xfffff800` .19+<| `CTRL` <|`0`     `SPI_CTRL_EN`                           ^| r/w <| SPI module enable
                                   <|`1`     `SPI_CTRL_CPHA`                         ^| r/w <| clock phase
                                   <|`2`     `SPI_CTRL_CPOL`                         ^| r/w <| clock polarity
                                   <|`5:3`   `SPI_CTRL_CS_SEL2 : SPI_CTRL_CS_SEL0`   ^| r/w <| Direct chip-select 0..7
                                   <|`6`     `SPI_CTRL_CS_EN`                        ^| r/w <| Direct chip-select enable: setting `spi_csn_o(SPI_CTRL_CS_SEL)` low when set
                                   <|`9:7`   `SPI_CTRL_PRSC2 : SPI_CTRL_PRSC0`       ^| r/w <| 3-bit clock prescaler select
-                                  <|`13:10` `SPI_CTRL_CDIV2 : SPI_CTRL_CDIV0`       ^| r/w <| 4-bit clock divider
-                                  <|`15:14`  _reserved_                             ^| r/- <| reserved, read as zero
+                                  <|`13:10` `SPI_CTRL_CDIV2 : SPI_CTRL_CDIV0`       ^| r/w <| 4-bit clock divider for fine-tuning
+                                  <|`14`    `SPI_CTRL_HIGHSPEED`                    ^| r/w <| high-speed mode enable (overriding `SPI_CTRL_PRSC`)
+                                  <|`15`     _reserved_                             ^| r/- <| reserved, read as zero
                                   <|`16`    `SPI_CTRL_RX_AVAIL`                     ^| r/- <| RX FIFO data available (RX FIFO not empty)
                                   <|`17`    `SPI_CTRL_TX_EMPTY`                     ^| r/- <| TX FIFO empty
                                   <|`18`    `SPI_CTRL_TX_NHALF`                     ^| r/- <| TX FIFO _not_ at least half full

rtl/core/neorv32_package.vhd

@@ -59,7 +59,7 @@ package neorv32_package is
 
   -- Architecture Constants -----------------------------------------------------------------
   -- -------------------------------------------------------------------------------------------
-  constant hw_version_c : std_ulogic_vector(31 downto 0) := x"01090100"; -- hardware version
+  constant hw_version_c : std_ulogic_vector(31 downto 0) := x"01090101"; -- hardware version
   constant archid_c     : natural := 19; -- official RISC-V architecture ID
   constant XLEN         : natural := 32; -- native data path width, do not change!
 

rtl/core/neorv32_spi.vhd

@@ -75,6 +75,7 @@ architecture neorv32_spi_rtl of neorv32_spi is
   constant ctrl_cdiv1_c        : natural := 11; -- r/w: clock divider bit 1
   constant ctrl_cdiv2_c        : natural := 12; -- r/w: clock divider bit 2
   constant ctrl_cdiv3_c        : natural := 13; -- r/w: clock divider bit 3
+  constant ctrl_highspeed_c    : natural := 14; -- r/w: high-speed mode
   --
   constant ctrl_rx_avail_c     : natural := 16; -- r/-: rx fifo data available (fifo not empty)
   constant ctrl_tx_empty_c     : natural := 17; -- r/-: tx fifo empty
@@ -99,6 +100,7 @@ architecture neorv32_spi_rtl of neorv32_spi is
     cs_en        : std_ulogic;
     prsc         : std_ulogic_vector(2 downto 0);
     cdiv         : std_ulogic_vector(3 downto 0);
+    highspeed    : std_ulogic;
     irq_rx_avail : std_ulogic;
     irq_tx_empty : std_ulogic;
     irq_tx_nhalf : std_ulogic;
@@ -116,7 +118,7 @@ architecture neorv32_spi_rtl of neorv32_spi is
     sreg     : std_ulogic_vector(7 downto 0);
     bitcnt   : std_ulogic_vector(3 downto 0);
     sdi_sync : std_ulogic;
-    sck : std_ulogic;
+    sck      : std_ulogic;
     done     : std_ulogic;
   end record;
   signal rtx_engine : rtx_engine_t;
@@ -151,6 +153,7 @@ begin
       ctrl.cs_en        <= '0';
       ctrl.prsc         <= (others => '0');
       ctrl.cdiv         <= (others => '0');
+      ctrl.highspeed    <= '0';
       ctrl.irq_rx_avail <= '0';
       ctrl.irq_tx_empty <= '0';
       ctrl.irq_tx_nhalf <= '0';
@@ -171,6 +174,7 @@ begin
             ctrl.cs_en        <= bus_req_i.data(ctrl_cs_en_c);
             ctrl.prsc         <= bus_req_i.data(ctrl_prsc2_c downto ctrl_prsc0_c);
             ctrl.cdiv         <= bus_req_i.data(ctrl_cdiv3_c downto ctrl_cdiv0_c);
+            ctrl.highspeed    <= bus_req_i.data(ctrl_highspeed_c);
             ctrl.irq_rx_avail <= bus_req_i.data(ctrl_irq_rx_avail_c);
             ctrl.irq_tx_empty <= bus_req_i.data(ctrl_irq_tx_empty_c);
             ctrl.irq_tx_nhalf <= bus_req_i.data(ctrl_irq_tx_nhalf_c);
@@ -186,6 +190,7 @@ begin
             bus_rsp_o.data(ctrl_cs_en_c)                         <= ctrl.cs_en;
             bus_rsp_o.data(ctrl_prsc2_c downto ctrl_prsc0_c)     <= ctrl.prsc;
             bus_rsp_o.data(ctrl_cdiv3_c downto ctrl_cdiv0_c)     <= ctrl.cdiv;
+            bus_rsp_o.data(ctrl_highspeed_c)                     <= ctrl.highspeed;
             --
             bus_rsp_o.data(ctrl_rx_avail_c)     <= rx_fifo.avail;
             bus_rsp_o.data(ctrl_tx_empty_c)     <= not tx_fifo.avail;
@@ -208,11 +213,15 @@ begin
   end process bus_access;
 
   -- direct chip-select (low-active) --
-  chip_select: process(ctrl)
+  chip_select: process(rstn_i, clk_i)
   begin
-    spi_csn_o <= (others => '1'); -- default: all disabled
-    if (ctrl.cs_en = '1') and (ctrl.enable = '1') then
-      spi_csn_o(to_integer(unsigned(ctrl.cs_sel))) <= '0';
+    if (rstn_i = '0') then
+      spi_csn_o <= (others => '1');
+    elsif rising_edge(clk_i) then
+      spi_csn_o <= (others => '1'); -- default: all disabled
+      if (ctrl.cs_en = '1') and (ctrl.enable = '1') then
+        spi_csn_o(to_integer(unsigned(ctrl.cs_sel))) <= '0';
+      end if;
     end if;
   end process chip_select;
 
@@ -317,8 +326,8 @@ begin
         -- ------------------------------------------------------------
           rtx_engine.sck    <= ctrl.cpol;
           rtx_engine.bitcnt <= (others => '0');
-          rtx_engine.sreg   <= tx_fifo.rdata;
           if (tx_fifo.avail = '1') then -- trigger new transmission
+            rtx_engine.sreg              <= tx_fifo.rdata;
             rtx_engine.state(1 downto 0) <= "01";
           end if;
 
@@ -371,25 +380,23 @@ begin
   spi_clk_o <= rtx_engine.sck;
 
 
-  -- clock generator --
+  -- SPI Clock Generator --------------------------------------------------------------------
+  -- -------------------------------------------------------------------------------------------
   clock_generator: process(rstn_i, clk_i)
   begin
     if (rstn_i = '0') then
       spi_clk_en <= '0';
       cdiv_cnt   <= (others => '0');
     elsif rising_edge(clk_i) then
+      spi_clk_en <= '0'; -- default
       if (ctrl.enable = '0') then -- reset/disabled
-        spi_clk_en <= '0';
-        cdiv_cnt   <= (others => '0');
-      else
-        spi_clk_en <= '0'; -- default
-        if (clkgen_i(to_integer(unsigned(ctrl.prsc))) = '1') then -- pre-scaled clock
-          if (cdiv_cnt = ctrl.cdiv) then -- clock divider for fine-tuning
-            spi_clk_en <= '1';
-            cdiv_cnt   <= (others => '0');
-          else
-            cdiv_cnt <= std_ulogic_vector(unsigned(cdiv_cnt) + 1);
-          end if;
+        cdiv_cnt <= (others => '0');
+      elsif (clkgen_i(to_integer(unsigned(ctrl.prsc))) = '1') or (ctrl.highspeed = '1') then -- pre-scaled clock
+        if (cdiv_cnt = ctrl.cdiv) then -- clock divider for fine-tuning
+          spi_clk_en <= '1';
+          cdiv_cnt   <= (others => '0');
+        else
+          cdiv_cnt <= std_ulogic_vector(unsigned(cdiv_cnt) + 1);
         end if;
       end if;
     end if;

sw/lib/include/neorv32_spi.h

@@ -72,6 +72,7 @@ enum NEORV32_SPI_CTRL_enum {
   SPI_CTRL_CDIV1        = 11, /**< SPI control register(11) (r/w): Clock divider bit 1 */
   SPI_CTRL_CDIV2        = 12, /**< SPI control register(12) (r/w): Clock divider bit 2 */
   SPI_CTRL_CDIV3        = 13, /**< SPI control register(13) (r/w): Clock divider bit 3 */
+  SPI_CTRL_HIGHSPEED    = 14, /**< SPI control register(14) (r/w): High-speed mode */
 
   SPI_CTRL_RX_AVAIL     = 16, /**< SPI control register(16) (r/-): RX FIFO data available (RX FIFO not empty) */
   SPI_CTRL_TX_EMPTY     = 17, /**< SPI control register(17) (r/-): TX FIFO empty */
@@ -96,6 +97,8 @@ enum NEORV32_SPI_CTRL_enum {
 /**@{*/
 int      neorv32_spi_available(void);
 void     neorv32_spi_setup(int prsc, int cdiv, int clk_phase, int clk_polarity, uint32_t irq_mask);
+void     neorv32_spi_highspeed_enable(void);
+void     neorv32_spi_highspeed_disable(void);
 uint32_t neorv32_spi_get_clock_speed(void);
 void     neorv32_spi_disable(void);
 void     neorv32_spi_enable(void);

sw/lib/source/neorv32_spi.c

@@ -85,6 +85,24 @@ void neorv32_spi_setup(int prsc, int cdiv, int clk_phase, int clk_polarity, uint
 }
 
 
+/**********************************************************************//**
+ * Enable high-speed mode.
+ **************************************************************************/
+void neorv32_spi_highspeed_enable(void) {
+
+  NEORV32_SPI->CTRL |= ((uint32_t)(1 << SPI_CTRL_HIGHSPEED));
+}
+
+
+/**********************************************************************//**
+ * Disable high-speed mode.
+ **************************************************************************/
+void neorv32_spi_highspeed_disable(void) {
+
+  NEORV32_SPI->CTRL &= ~((uint32_t)(1 << SPI_CTRL_HIGHSPEED));
+}
+
+
 /**********************************************************************//**
  * Get configured clock speed in Hz.
  *
@@ -95,10 +113,18 @@ uint32_t neorv32_spi_get_clock_speed(void) {
   const uint16_t PRSC_LUT[8] = {2, 4, 8, 64, 128, 1024, 2048, 4096};
 
   uint32_t ctrl = NEORV32_SPI->CTRL;
-  uint32_t prsc_sel = (ctrl >> SPI_CTRL_PRSC0) & 0x7;
+  uint32_t prsc_sel  = (ctrl >> SPI_CTRL_PRSC0) & 0x7;
   uint32_t clock_div = (ctrl >> SPI_CTRL_CDIV0) & 0xf;
 
-  uint32_t tmp = 2 * PRSC_LUT[prsc_sel] * (1 + clock_div);
+  uint32_t tmp;
+
+  if (ctrl & (1 << SPI_CTRL_HIGHSPEED)) { // high-speed mode enabled?
+    tmp = 2 * 1 * (1 + clock_div);
+  }
+  else {
+    tmp = 2 * PRSC_LUT[prsc_sel] * (1 + clock_div);
+  }
+
   return NEORV32_SYSINFO->CLK / tmp;
 }