test: simulate the register file and the table engine, and test the tables

The switch is reached through four register functions and the SFR_DATA
registers, so a flat register file behind them is enough to run the table
code on the host. A few addresses are live: writing TBL_CTRL with EXECUTE
performs the VLAN or L2 operation at once and clears the bit, STAT_GET
fetches a counter, L2_TBL_FLUSH_CTRL drops entries. Nothing ever reports
busy, so a polling loop runs once and the tests say nothing about what
happens when hardware stalls; that question was settled elsewhere.

The header states the VLAN and L2 entry layouts on its own, so a test can
hold what rtl837x_port.c writes against a second reading of the same bits
rather than against itself. The first suite does that for the VLAN entry,
the PVID register two ports share, the static multicast and management
entries, the per-port flush and the trunk registers. Two stub headers stand
in for the ones a firmware build generates, so the harness needs no SDCC
build first.

test/hw_mock.h

@@ -0,0 +1,59 @@
+/*
+ * hw_mock.h - a simulated RTL837x register file and table engine.
+ *
+ * The firmware reaches the ASIC only through reg_read/reg_read_m/reg_write/
+ * reg_write_m and the SFR_DATA_* registers, so mocking that edge is enough to
+ * run the table code on the host. Every register is a 32-bit word in a flat
+ * file; a few addresses are live: writing TBL_CTRL with EXECUTE performs the
+ * table operation at once and clears the bit, STAT_GET fetches a counter,
+ * L2_TBL_FLUSH_CTRL drops entries. Nothing here ever reports busy, so a
+ * polling loop in the firmware runs exactly once.
+ *
+ * The VLAN and L2 entry layouts mirror what rtl837x_port.c writes and what
+ * httpd/page_impl.c reads; the decoders below are an independent statement
+ * of that layout, so a test can compare the firmware's two ends against it.
+ */
+#ifndef TEST_HW_MOCK_H
+#define TEST_HW_MOCK_H
+
+#include <stdint.h>
+#include <stdbool.h>
+
+void     hw_reset(void);                       /* every register 0, tables empty */
+void     hw_reg_set(uint16_t addr, uint32_t v);
+uint32_t hw_reg_get(uint16_t addr);
+
+extern uint8_t       sfr_data[4];               /* the firmware's register scratch, owned here */
+extern unsigned long hw_reads;                 /* reg_read + reg_read_m calls since hw_reset */
+extern unsigned long hw_writes;                /* reg_write + reg_write_m calls since hw_reset */
+
+/* VLAN table: the 32-bit word the firmware wrote for that VID, 0 = never written */
+uint32_t hw_vlan_word(uint16_t vid);
+
+/* L2 table: entries as the three data words the firmware exchanges with the
+ * engine. Word B: bit 29 valid, bits 31:30 the low two bits of the port (or
+ * of the member mask), bits 27:16 VID, bits 15:0 the first two MAC octets.
+ * Word A: the last four MAC octets. Word C: bit 16 static, bits 1:0 the high
+ * two bits of the port; for a multicast MAC bits 7:0 are the member mask
+ * shifted right by two. */
+#define HW_L2_MAX 64
+struct hw_l2 {
+	bool     present;
+	uint16_t idx;
+	uint32_t a, b, c;   /* DATA_IN_A/B/C on write, L2_DATA_OUT_A/B/C on read */
+};
+
+/* Pointers returned below stay valid until the next table operation. */
+int             hw_l2_count(void);
+const struct hw_l2 *hw_l2_at(int n);           /* n-th present entry in idx order */
+const struct hw_l2 *hw_l2_find(const uint8_t mac[6], uint16_t vid);
+/* Put an entry the way the switch would hold it: a learned unicast entry on
+ * one port, or a static one. mc entries use pmask instead of a port. */
+void hw_l2_put(uint16_t idx, const uint8_t mac[6], uint16_t vid, uint16_t port_or_pmask, bool is_static);
+void hw_l2_decode(const struct hw_l2 *e, uint8_t mac[6], uint16_t *vid, uint16_t *port_or_pmask,
+		  bool *valid, bool *is_static, bool *mc);
+
+/* MIB counter the next STAT_GET for (port, counter) returns */
+void hw_counter_set(uint8_t port, uint8_t counter, uint64_t value);
+
+#endif /* TEST_HW_MOCK_H */