Merge pull request #45 from davidgiven/dtrg-fixups

Add hi16[], ha16[], lo16[] support to the PowerPC assembler

mach/powerpc/as/mach1.c

@@ -3,3 +3,7 @@
  * $State$
  */
 
+#include <stdbool.h>
+
+extern quad emit_hi(struct expr_t* expr, bool is_signed);
+extern quad emit_lo(struct expr_t* expr);

mach/powerpc/as/mach2.c

@@ -84,6 +84,9 @@
 %token <y_word> OP_LA
 %token <y_word> OP_LI32
 
+%token <y_word> OP_POWERPC_FIXUP
+%token <y_word> OP_HI OP_HA OP_LO
+
 /* Other token types */
 
 %type <y_word> c

mach/powerpc/as/mach3.c

@@ -102,6 +102,9 @@
 0,     OP_LA,                 0,                                       "la",
 0,     OP_LA,                 0,                                       "li",
 0,     OP_RS_RA_RA_C,         31<<26 | 444<<1,                         "mr",
+0,     OP_HI,                 0,                                       "hi16",
+0,     OP_HA,                0,                                        "ha16",
+0,     OP_LO,                 0,                                       "lo16",
 
 /* Branch processor instructions (page 20) */
 

mach/powerpc/as/mach4.c

@@ -75,6 +75,9 @@ e16
 			serror("16-bit value out of range");
 		$$ = (uint16_t) $1;
 	}
+	| OP_HI ASC_LPAR expr ASC_RPAR           { $$ = emit_hi(&$3, false); }
+	| OP_HA ASC_LPAR expr ASC_RPAR           { $$ = emit_hi(&$3, true); }
+	| OP_LO ASC_LPAR expr ASC_RPAR           { $$ = emit_lo(&$3); }
 	;
 
 u8
@@ -248,3 +251,14 @@ spr_num
 		$$ = ($1 >> 5) | (($1 & 0x1f) << 5);
 	}
 	;
+
+powerpcfixup
+	: expr
+	{
+		quad type = $1.typ & S_TYP;
+		quad val = $1.val;
+		if (type == S_ABS)
+			serror(".powerpcfixup is useless on absolute values");
+		newrelo($1.typ, RELOPPC | FIXUPFLAGS);
+	}
+	;

mach/powerpc/as/mach5.c

@@ -1,5 +1,38 @@
-/*
+
- * $Source$
+quad emit_hi(struct expr_t* expr, bool is_signed)
- * $State$
+{
+    /* If this is a symbol reference, discard the symbol and keep only the
+        * offset part. */
+    quad type = expr->typ & S_TYP;
+    quad val = expr->val;
+    uint16_t hi = val >> 16;
+    uint16_t lo = val & 0xffff;
+
+    if (type != S_ABS)
+		newrelo(expr->typ, RELOPPC | FIXUPFLAGS);
+
+    /* If the low half of this relocation is going to be a memory operation,
+     * then it'll be treated as a signed value. That means that values greater
+     * than 0x7fff will cause the high word to have 1 subtracted from it; so
+     * we apply an adjustment here.
      */
 
+    if (is_signed && (lo > 0x7fff))
+        hi++;
+
+    return hi;
+}
+
+quad emit_lo(struct expr_t* expr)
+{
+    quad type = expr->typ & S_TYP;
+    quad val = expr->val;
+
+    /* If the assembler stored a symbol for relocation later, we need to
+     * abandon it (because the relocation was generated by emit_ha). */
+
+    if (type != S_ABS)
+        relonami = 0;
+
+    return val & 0xffff;
+}

mach/powerpc/ncg/table

@@ -169,11 +169,15 @@ TOKENS
 /* Used only in instruction descriptions (to generate the correct syntax). */
 
 	GPRINDIRECT        = { GPR reg; INT off; }    4    off "(" reg ")".
+	GPRINDIRECT_OFFSET_LO = { GPR reg; ADDR adr; } 4   "lo16[" adr "](" reg ")".
 	CONST              = { INT val; }             4    val.
 
 /* Primitives */
 
 	LABEL              = { ADDR adr; }            4    adr.
+	LABEL_OFFSET_HI    = { ADDR adr; }            4    "hi16[" adr "]".
+	LABEL_OFFSET_HA    = { ADDR adr; }            4    "ha16[" adr "]".
+	LABEL_OFFSET_LO    = { ADDR adr; }            4    "lo16[" adr "]".
 	LOCAL              = { INT off; }             4.
 
 /* Allows us to use regvar() to refer to registers */
@@ -282,7 +286,7 @@ INSTRUCTIONS
   add             GPR:wo, GPR:ro, GPR:ro.
   addX "add."     GPR:wo, GPR:ro, GPR:ro.
   addi            GPR:wo, GPR:ro, CONST:ro.
-  addis           GPR:wo, GPR:ro, CONST:ro.
+  addis           GPR:wo, GPR:ro, CONST+LABEL_OFFSET_HI+LABEL_OFFSET_HA:ro.
   and             GPR:wo, GPR:ro, GPR:ro.
   andc            GPR:wo, GPR:ro, GPR:ro.
   andiX  "andi."  GPR:wo:cc, GPR:ro, CONST:ro.
@@ -329,10 +333,9 @@ INSTRUCTIONS
   lhz             GPR:wo, GPRINDIRECT:ro cost(4, 3).
   lhzx            GPR:wo, GPR:ro, GPR:ro cost(4, 3).
   li32            GPR:wo, CONST:ro cost(8, 2).
-  li32            GPR:wo, LABEL:ro cost(8, 2).
   lwzu            GPR:wo, GPRINDIRECT:ro cost(4, 3).
   lwzx            GPR:wo, GPR:ro, GPR:ro cost(4, 3).
-  lwz             GPR:wo, GPRINDIRECT:ro cost(4, 3).
+  lwz             GPR:wo, GPRINDIRECT+GPRINDIRECT_OFFSET_LO:ro cost(4, 3).
   nand            GPR:wo, GPR:ro, GPR:ro.
   neg             GPR:wo, GPR:ro.
   nor             GPR:wo, GPR:ro, GPR:ro.
@@ -342,7 +345,7 @@ INSTRUCTIONS
   mtspr           SPR:wo, GPR:ro cost(4, 2).
   or              GPR:wo, GPR:ro, GPR:ro.
   orc             GPR:wo, GPR:ro, GPR:ro.
-  ori             GPR:wo, GPR:ro, CONST:ro.
+  ori             GPR:wo, GPR:ro, CONST+LABEL_OFFSET_LO:ro.
   oris            GPR:wo, GPR:ro, CONST:ro.
   orX "or."       GPR:wo:cc, GPR:ro, GPR:ro.
   rlwinm          GPR:wo, GPR:ro, CONST:ro, CONST:ro, CONST:ro.
@@ -405,7 +408,8 @@ MOVES
 	from LABEL to GPR
 		gen
 			COMMENT("move LABEL->GPR")
-			li32 %2, {LABEL, %1.adr}
+			addis %2, R0, {LABEL_OFFSET_HI, %1.adr}
+			ori %2, %2, {LABEL_OFFSET_LO, %1.adr}
 
 /* Sign extension */
 
@@ -1142,6 +1146,12 @@ PATTERNS
 			yields {IND_RC_H, %1.reg, %1.off}
 
 	pat loi $1==INT32                  /* Load word indirect */
+		with LABEL
+			uses REG
+			gen
+				addis %a, R0, {LABEL_OFFSET_HA, %1.adr}
+				lwz %a, {GPRINDIRECT_OFFSET_LO, %a, %1.adr}
+			yields %a
 		with GPR
 			yields {IND_RC_W, %1, 0}
 		with SUM_RC

man/powerpc_as.6

@@ -0,0 +1,33 @@
+.TH POWERPC_AS 1
+.ad
+.SH NAME
+powerpc_as \- assembler for PowerPC
+
+.SH SYNOPSIS
+as [options] argument ...
+
+.SH DESCRIPTION
+This assembler is made with the general framework
+described in \fIuni_ass\fP(6).
+
+.SH SYNTAX
+Most 32-bit integer and floating point instructions are supported, but not many
+short form instructions. Instructions which take 16-bit operands can additionally
+use the following special functions:
+
+.IP hi16[value], ha16[value]
+Returns the high half of the value of the expression; if the value is not absolute,
+also generates the appropriate fixup. Use of either of these \fImust\fR be followed,
+in the next instruction, by the corresponding use of \fBlo16[]\fR. Use \fBhi16[]\fR
+if the low half is going to interpret its payload as an unsigned value, and
+\fBha16[]\fR if it will be interpreted as a signed value (so that the high half can
+be adjusted to match).
+
+.IP lo16[]
+Returns the low half of the value of the expression. No fixup is generated. Use of
+\fBlo16[]\fR must come in the instruction immediately after a use of \fBhi16[]\fR or
+\fBha16[]\fR.
+
+.SH "SEE ALSO"
+uni_ass(6),
+ack(1)

util/led/archive.c

@@ -6,8 +6,10 @@
 static char rcsid[] = "$Id$";
 #endif
 
-#include <stdlib.h>
 #include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdbool.h>
 #include "arch.h"
 #include "out.h"
 #include "ranlib.h"

util/led/const.h

@@ -4,8 +4,6 @@
  */
 /* $Id$ */
 
-typedef int		bool;
-
 #define FALSE		0
 #define TRUE		1
 

util/led/debug.h

@@ -17,3 +17,5 @@ extern int DEB;
 
 extern int Verbose;
 #define verbose(s, a1, a2, a3, a4)	(Verbose && do_verbose(s, a1, a2, a3, a4))
+
+extern void fatal(char* format, ...);

util/led/error.c

@@ -6,9 +6,11 @@
 static char rcsid[] = "$Id$";
 #endif
 
-#include <stdarg.h>
-#include <stdlib.h>
 #include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdbool.h>
+#include <stdarg.h>
 #include <out.h>
 #include "const.h"
 

util/led/extract.c

@@ -6,8 +6,10 @@
 static char rcsid[] = "$Id$";
 #endif
 
-#include <stdlib.h>
 #include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdbool.h>
 #include "out.h"
 #include "const.h"
 #include "debug.h"

util/led/finish.c

@@ -6,6 +6,10 @@
 static char rcsid[] = "$Id$";
 #endif
 
+#include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdbool.h>
 #include <out.h>
 #include "const.h"
 #include "defs.h"

util/led/main.c

@@ -11,7 +11,9 @@ static char rcsid[] = "$Id$";
  */
 
 #include <stdio.h>
+#include <stdlib.h>
 #include <stdint.h>
+#include <stdbool.h>
 #include <out.h>
 #include "const.h"
 #include "debug.h"

util/led/memory.c

@@ -22,6 +22,9 @@ static char rcsid[] = "$Id$";
  */
 
 #include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdbool.h>
 #include <out.h>
 #include "const.h"
 #include "assert.h"

util/led/output.c

@@ -6,6 +6,10 @@
 static char rcsid[] = "$Id$";
 #endif
 
+#include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdbool.h>
 #include <out.h>
 #include "const.h"
 #include "memory.h"

util/led/read.c

@@ -6,6 +6,11 @@
 static char rcsid[] = "$Id$";
 #endif
 
+#include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdbool.h>
+
 int	infile;	/* The current input file. */
 
 rd_fatal()

util/led/relocate.c

@@ -9,6 +9,7 @@ static char rcsid[] = "$Id$";
 #include <stdlib.h>
 #include <stdio.h>
 #include <stdint.h>
+#include <stdbool.h>
 #include <assert.h>
 #include "out.h"
 #include "const.h"
@@ -104,6 +105,24 @@ static uint32_t get_vc4_valu(char* addr)
 	assert(0 && "unrecognised VC4 instruction");
 }
 
+static bool is_powerpc_memory_op(uint32_t opcode)
+{
+	/* Tests for any PowerPC memory indirection instruction where the payload
+	 * is a *signed* 16-bit value. */
+	switch ((opcode & 0xfc000000) >> 26)
+	{
+		case 34: /* lbz */
+		case 40: /* lhz */
+		case 32: /* lwz */
+		case 38: /* stb */
+		case 44: /* sth */
+		case 36: /* stw */
+			return true;
+	}
+
+	return false;
+}
+
 /* PowerPC fixups are complex as we need to patch up to the next two
  * instructions in one of several different ways, depending on what the
  * instructions area.
@@ -125,8 +144,23 @@ static uint32_t get_powerpc_valu(char* addr, uint16_t type)
 		/* addis / ori instruction pair */
 		return ((opcode1 & 0xffff) << 16) | (opcode2 & 0xffff);
 	}
+	else if (((opcode1 & 0xfc1f0000) == 0x3c000000) &&
+	         is_powerpc_memory_op(opcode2))
+	{
+		/* addis / memoryop instruction pair */
+		uint16_t hi = opcode1 & 0xffff;
+		uint16_t lo = opcode2 & 0xffff;
 
-	assert(0 && "unrecognised PowerPC instruction");
+		/* Undo the sign adjustment (see mach/powerpc/as/mach5.c). */
+
+		if (lo > 0x7fff)
+			hi--;
+
+		return ((hi << 16) | lo);
+	}
+
+	fatal("Don't know how to read from PowerPC fixup on instructions 0x%08x+0x%08x",
+		opcode1, opcode2);
 }
 
 /*
@@ -263,14 +297,32 @@ static void put_powerpc_valu(char* addr, uint32_t value, uint16_t type)
 	else if (((opcode1 & 0xfc1f0000) == 0x3c000000) &&
 	         ((opcode2 & 0xfc000000) == 0x60000000))
 	{
+		/* addis / ori instruction pair */
 		uint16_t hi = value >> 16;
 		uint16_t lo = value & 0xffff;
 
 		write4((opcode1 & 0xffff0000) | hi, addr+0, type);
 		write4((opcode2 & 0xffff0000) | lo, addr+4, type);
 	}
+	else if (((opcode1 & 0xfc1f0000) == 0x3c000000) &&
+	         is_powerpc_memory_op(opcode2))
+	{
+		/* addis / memoryop instruction pair */
+		uint16_t hi = value >> 16;
+		uint16_t lo = value & 0xffff;
+
+		/* Apply the sign adjustment (see mach/powerpc/as/mach5.c). */
+
+		if (lo > 0x7fff)
+			hi++;
+
+		write4((opcode1 & 0xffff0000) | hi, addr+0, type);
+		write4((opcode2 & 0xffff0000) | lo, addr+4, type);
+	}
+
 	else
-		assert(0 && "unrecognised PowerPC instruction");
+		fatal("Don't know how to write a PowerPC fixup to instructions 0x%08x+0x%08x",
+			opcode1, opcode2);
 }
 
 /*

util/led/save.c

@@ -10,8 +10,10 @@ static char rcsid[] = "$Id$";
  * If everything is kept in core, we must save some things for the second pass.
  */
 
-#include <stdlib.h>
 #include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdbool.h>
 #include <string.h>
 #include "arch.h"
 #include "out.h"

util/led/scan.c

@@ -6,8 +6,10 @@
 static char rcsid[] = "$Id$";
 #endif
 
-#include <stdlib.h>
 #include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdbool.h>
 #ifdef SYMDBUG
 #include <sys/types.h>
 #include <sys/stat.h>

util/led/sym.c

@@ -10,8 +10,10 @@ static char rcsid[] = "$Id$";
  * Symbol table management.
  */
 
-#include <stdlib.h>
 #include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdbool.h>
 #include "out.h"
 #include "const.h"
 #include "memory.h"

util/led/write.c

@@ -6,8 +6,10 @@
 static char rcsid[] = "$Id$";
 #endif
 
-#include <stdlib.h>
 #include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <stdbool.h>
 #include <string.h>
 #include "out.h"
 #include "const.h"