* [PATCH/RFA] Generic OS ABI handling for PowerPC
@ 2002-05-28 12:41 Jason R Thorpe
2002-05-28 12:52 ` Kevin Buettner
0 siblings, 1 reply; 11+ messages in thread
From: Jason R Thorpe @ 2002-05-28 12:41 UTC (permalink / raw)
To: gdb-patches
[-- Attachment #1: Type: text/plain, Size: 812 bytes --]
This makes the PowerPC/RS6000 target use the generic OS ABI framework,
and also moves some SVR4 ABI stuff out of ppc-linux-tdep.c into
rs6000-tdep.c.
OK to commit?
* Makefile.in (ppc_tdep_h): Define.
(ppc-linux-nat.o, ppc-linux-tdep.o, rs6000-tdep.o): Use $(ppc_tdep_h).
* ppc-tdep.h: Use generic OS ABI framework.
* ppc-linux-tdep.c (_initialize_ppc_linux_tdep,
ppc_linux_init_abi): New functions.
(ppc_sysv_abi_broken_use_struct_convention,
ppc_sysv_abi_use_struct_convention,
ppc_sysv_abi_push_arguments): Move to...
* rs6000-tdep.c: ...here.
(process_note_abi_tag_sections, get_elfosabi): Remove.
(rs6000_gdbarch_init): Use generic OS ABI framework.
(rs6000_dump_tdep): New function.
(_initialize_rs6000_tdep): Use gdbarch_register.
--
-- Jason R. Thorpe <thorpej@wasabisystems.com>
[-- Attachment #2: ppc-patch --]
[-- Type: text/plain, Size: 29098 bytes --]
Index: Makefile.in
===================================================================
RCS file: /cvs/src/src/gdb/Makefile.in,v
retrieving revision 1.200
diff -u -r1.200 Makefile.in
--- Makefile.in 24 May 2002 00:12:16 -0000 1.200
+++ Makefile.in 28 May 2002 19:15:46 -0000
@@ -646,6 +646,7 @@
monitor_h = monitor.h
objfiles_h = objfiles.h
parser_defs_h = parser-defs.h $(doublest_h)
+ppc_tdep_h = ppc-tdep.h osabi.h
regcache_h = regcache.h
remote_h = remote.h
remote_utils_h = remote-utils.h $(target_h)
@@ -1899,10 +1900,10 @@
$(objfiles_h) $(gdb_stabs_h) $(serial_h) ocd.h $(regcache_h)
ppc-linux-nat.o: ppc-linux-nat.c $(defs_h) $(gdbcore_h) $(frame_h) \
- $(inferior_h) $(target_h) $(regcache_h) ppc-tdep.h
+ $(inferior_h) $(target_h) $(regcache_h) $(ppc_tdep_h)
ppc-linux-tdep.o: ppc-linux-tdep.c $(defs_h) $(gdbcore_h) $(inferior_h) \
- $(target_h) ppc-tdep.h $(regcache_h) $(value_h)
+ $(target_h) $(ppc_tdep_h) $(regcache_h) $(value_h)
ppcbug-rom.o: ppcbug-rom.c $(monitor_h) $(bfd_h) $(gdb_wait_h) $(defs_h) \
$(gdbcmd_h) $(inferior_h) $(target_h) $(serial_h) $(terminal_h) \
@@ -2038,7 +2039,7 @@
$(gdb_stabs_h) $(regcache_h) $(arch_utils_h)
rs6000-tdep.o: rs6000-tdep.c $(defs_h) $(gdbcore_h) $(inferior_h) \
- $(target_h) ppc-tdep.h $(regcache_h) $(value_h) $(parser_defs_h)
+ $(target_h) $(ppc_tdep_h) $(regcache_h) $(value_h) $(parser_defs_h)
s390-tdep.o: s390-tdep.c $(defs_h) $(arch_utils_h) $(frame_h) $(inferior_h) \
$(symtab_h) $(target_h) $(gdbcore_h) $(gdbcmd_h) $(symfile_h) \
Index: ppc-linux-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/ppc-linux-tdep.c,v
retrieving revision 1.17
diff -u -r1.17 ppc-linux-tdep.c
--- ppc-linux-tdep.c 29 Apr 2002 15:10:06 -0000 1.17
+++ ppc-linux-tdep.c 28 May 2002 19:15:48 -0000
@@ -411,293 +411,6 @@
return rs6000_frame_chain (thisframe);
}
-/* FIXME: Move the following to rs6000-tdep.c (or some other file where
- it may be used generically by ports which use either the SysV ABI or
- the EABI */
-
-/* Until November 2001, gcc was not complying to the SYSV ABI for
- returning structures less than or equal to 8 bytes in size. It was
- returning everything in memory. When this was corrected, it wasn't
- fixed for native platforms. */
-int
-ppc_sysv_abi_broken_use_struct_convention (int gcc_p, struct type *value_type)
-{
- if (TYPE_LENGTH (value_type) == 16
- && TYPE_VECTOR (value_type))
- return 0;
-
- return generic_use_struct_convention (gcc_p, value_type);
-}
-
-/* Structures 8 bytes or less long are returned in the r3 & r4
- registers, according to the SYSV ABI. */
-int
-ppc_sysv_abi_use_struct_convention (int gcc_p, struct type *value_type)
-{
- if (TYPE_LENGTH (value_type) == 16
- && TYPE_VECTOR (value_type))
- return 0;
-
- return (TYPE_LENGTH (value_type) > 8);
-}
-
-/* round2 rounds x up to the nearest multiple of s assuming that s is a
- power of 2 */
-
-#undef round2
-#define round2(x,s) ((((long) (x) - 1) & ~(long)((s)-1)) + (s))
-
-/* Pass the arguments in either registers, or in the stack. Using the
- ppc sysv ABI, the first eight words of the argument list (that might
- be less than eight parameters if some parameters occupy more than one
- word) are passed in r3..r10 registers. float and double parameters are
- passed in fpr's, in addition to that. Rest of the parameters if any
- are passed in user stack.
-
- If the function is returning a structure, then the return address is passed
- in r3, then the first 7 words of the parametes can be passed in registers,
- starting from r4. */
-
-CORE_ADDR
-ppc_sysv_abi_push_arguments (int nargs, struct value **args, CORE_ADDR sp,
- int struct_return, CORE_ADDR struct_addr)
-{
- int argno;
- /* Next available general register for non-float, non-vector arguments. */
- int greg;
- /* Next available floating point register for float arguments. */
- int freg;
- /* Next available vector register for vector arguments. */
- int vreg;
- int argstkspace;
- int structstkspace;
- int argoffset;
- int structoffset;
- struct value *arg;
- struct type *type;
- int len;
- char old_sp_buf[4];
- CORE_ADDR saved_sp;
-
- greg = struct_return ? 4 : 3;
- freg = 1;
- vreg = 2;
- argstkspace = 0;
- structstkspace = 0;
-
- /* Figure out how much new stack space is required for arguments
- which don't fit in registers. Unlike the PowerOpen ABI, the
- SysV ABI doesn't reserve any extra space for parameters which
- are put in registers. */
- for (argno = 0; argno < nargs; argno++)
- {
- arg = args[argno];
- type = check_typedef (VALUE_TYPE (arg));
- len = TYPE_LENGTH (type);
-
- if (TYPE_CODE (type) == TYPE_CODE_FLT)
- {
- if (freg <= 8)
- freg++;
- else
- {
- /* SysV ABI converts floats to doubles when placed in
- memory and requires 8 byte alignment */
- if (argstkspace & 0x4)
- argstkspace += 4;
- argstkspace += 8;
- }
- }
- else if (TYPE_CODE (type) == TYPE_CODE_INT && len == 8) /* long long */
- {
- if (greg > 9)
- {
- greg = 11;
- if (argstkspace & 0x4)
- argstkspace += 4;
- argstkspace += 8;
- }
- else
- {
- if ((greg & 1) == 0)
- greg++;
- greg += 2;
- }
- }
- else if (!TYPE_VECTOR (type))
- {
- if (len > 4
- || TYPE_CODE (type) == TYPE_CODE_STRUCT
- || TYPE_CODE (type) == TYPE_CODE_UNION)
- {
- /* Rounding to the nearest multiple of 8 may not be necessary,
- but it is safe. Particularly since we don't know the
- field types of the structure */
- structstkspace += round2 (len, 8);
- }
- if (greg <= 10)
- greg++;
- else
- argstkspace += 4;
- }
- else
- {
- if (len == 16
- && TYPE_CODE (type) == TYPE_CODE_ARRAY
- && TYPE_VECTOR (type))
- {
- if (vreg <= 13)
- vreg++;
- else
- {
- /* Vector arguments must be aligned to 16 bytes on
- the stack. */
- argstkspace += round2 (argstkspace, 16);
- argstkspace += 16;
- }
- }
- }
- }
-
- /* Get current SP location */
- saved_sp = read_sp ();
-
- sp -= argstkspace + structstkspace;
-
- /* Allocate space for backchain and callee's saved lr */
- sp -= 8;
-
- /* Make sure that we maintain 16 byte alignment */
- sp &= ~0x0f;
-
- /* Update %sp before proceeding any further */
- write_register (SP_REGNUM, sp);
-
- /* write the backchain */
- store_address (old_sp_buf, 4, saved_sp);
- write_memory (sp, old_sp_buf, 4);
-
- argoffset = 8;
- structoffset = argoffset + argstkspace;
- freg = 1;
- greg = 3;
- vreg = 2;
- /* Fill in r3 with the return structure, if any */
- if (struct_return)
- {
- char val_buf[4];
- store_address (val_buf, 4, struct_addr);
- memcpy (®isters[REGISTER_BYTE (greg)], val_buf, 4);
- greg++;
- }
- /* Now fill in the registers and stack... */
- for (argno = 0; argno < nargs; argno++)
- {
- arg = args[argno];
- type = check_typedef (VALUE_TYPE (arg));
- len = TYPE_LENGTH (type);
-
- if (TYPE_CODE (type) == TYPE_CODE_FLT)
- {
- if (freg <= 8)
- {
- if (len > 8)
- printf_unfiltered (
- "Fatal Error: a floating point parameter #%d with a size > 8 is found!\n", argno);
- memcpy (®isters[REGISTER_BYTE (FP0_REGNUM + freg)],
- VALUE_CONTENTS (arg), len);
- freg++;
- }
- else
- {
- /* SysV ABI converts floats to doubles when placed in
- memory and requires 8 byte alignment */
- /* FIXME: Convert floats to doubles */
- if (argoffset & 0x4)
- argoffset += 4;
- write_memory (sp + argoffset, (char *) VALUE_CONTENTS (arg), len);
- argoffset += 8;
- }
- }
- else if (TYPE_CODE (type) == TYPE_CODE_INT && len == 8) /* long long */
- {
- if (greg > 9)
- {
- greg = 11;
- if (argoffset & 0x4)
- argoffset += 4;
- write_memory (sp + argoffset, (char *) VALUE_CONTENTS (arg), len);
- argoffset += 8;
- }
- else
- {
- if ((greg & 1) == 0)
- greg++;
-
- memcpy (®isters[REGISTER_BYTE (greg)],
- VALUE_CONTENTS (arg), 4);
- memcpy (®isters[REGISTER_BYTE (greg + 1)],
- VALUE_CONTENTS (arg) + 4, 4);
- greg += 2;
- }
- }
- else if (!TYPE_VECTOR (type))
- {
- char val_buf[4];
- if (len > 4
- || TYPE_CODE (type) == TYPE_CODE_STRUCT
- || TYPE_CODE (type) == TYPE_CODE_UNION)
- {
- write_memory (sp + structoffset, VALUE_CONTENTS (arg), len);
- store_address (val_buf, 4, sp + structoffset);
- structoffset += round2 (len, 8);
- }
- else
- {
- memset (val_buf, 0, 4);
- memcpy (val_buf, VALUE_CONTENTS (arg), len);
- }
- if (greg <= 10)
- {
- memcpy (®isters[REGISTER_BYTE (greg)], val_buf, 4);
- greg++;
- }
- else
- {
- write_memory (sp + argoffset, val_buf, 4);
- argoffset += 4;
- }
- }
- else
- {
- if (len == 16
- && TYPE_CODE (type) == TYPE_CODE_ARRAY
- && TYPE_VECTOR (type))
- {
- struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch);
- char *v_val_buf = alloca (16);
- memset (v_val_buf, 0, 16);
- memcpy (v_val_buf, VALUE_CONTENTS (arg), len);
- if (vreg <= 13)
- {
- memcpy (®isters[REGISTER_BYTE (tdep->ppc_vr0_regnum
- + vreg)],
- v_val_buf, 16);
- vreg++;
- }
- else
- {
- write_memory (sp + argoffset, v_val_buf, 16);
- argoffset += 16;
- }
- }
- }
- }
-
- target_store_registers (-1);
- return sp;
-}
-
/* ppc_linux_memory_remove_breakpoints attempts to remove a breakpoint
in much the same fashion as memory_remove_breakpoint in mem-break.c,
but is careful not to write back the previous contents if the code
@@ -885,4 +598,45 @@
}
return lmp;
+}
+
+static void
+ppc_linux_init_abi (struct gdbarch_info info,
+ struct gdbarch *gdbarch)
+{
+ struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch);
+
+ /* Until November 2001, gcc was not complying to the SYSV ABI for
+ returning structures less than or equal to 8 bytes in size. It was
+ returning everything in memory. When this was corrected, it wasn't
+ fixed for native platforms. */
+ set_gdbarch_use_struct_convention (gdbarch,
+ ppc_sysv_abi_broken_use_struct_convention);
+
+ if (tdep->wordsize == 4)
+ {
+ /* Note: kevinb/2002-04-12: See note in rs6000_gdbarch_init regarding
+ *_push_arguments(). The same remarks hold for the methods below. */
+ set_gdbarch_frameless_function_invocation (gdbarch,
+ ppc_linux_frameless_function_invocation);
+ set_gdbarch_frame_chain (gdbarch, ppc_linux_frame_chain);
+ set_gdbarch_frame_saved_pc (gdbarch, ppc_linux_frame_saved_pc);
+
+ set_gdbarch_frame_init_saved_regs (gdbarch,
+ ppc_linux_frame_init_saved_regs);
+ set_gdbarch_init_extra_frame_info (gdbarch,
+ ppc_linux_init_extra_frame_info);
+
+ set_gdbarch_memory_remove_breakpoint (gdbarch,
+ ppc_linux_memory_remove_breakpoint);
+ set_solib_svr4_fetch_link_map_offsets
+ (gdbarch, ppc_linux_svr4_fetch_link_map_offsets);
+ }
+}
+
+void
+_initialize_ppc_linux_tdep (void)
+{
+ gdbarch_register_osabi (bfd_arch_powerpc, GDB_OSABI_LINUX,
+ ppc_linux_init_abi);
}
Index: ppc-tdep.h
===================================================================
RCS file: /cvs/src/src/gdb/ppc-tdep.h,v
retrieving revision 1.10
diff -u -r1.10 ppc-tdep.h
--- ppc-tdep.h 29 Apr 2002 15:10:06 -0000 1.10
+++ ppc-tdep.h 28 May 2002 19:15:48 -0000
@@ -22,6 +22,8 @@
#ifndef PPC_TDEP_H
#define PPC_TDEP_H
+#include "osabi.h"
+
struct frame_info;
struct value;
@@ -52,7 +54,7 @@
struct gdbarch_tdep
{
int wordsize; /* size in bytes of fixed-point word */
- int osabi; /* OS / ABI from ELF header */
+ enum gdb_osabi osabi; /* OS / ABI from ELF header */
int *regoff; /* byte offsets in register arrays */
const struct reg *regs; /* from current variant */
int ppc_gp0_regnum; /* GPR register 0 */
Index: rs6000-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/rs6000-tdep.c,v
retrieving revision 1.65
diff -u -r1.65 rs6000-tdep.c
--- rs6000-tdep.c 20 May 2002 17:45:43 -0000 1.65
+++ rs6000-tdep.c 28 May 2002 19:15:52 -0000
@@ -1128,6 +1128,263 @@
return sp;
}
+/* round2 rounds x up to the nearest multiple of s assuming that s is a
+ power of 2 */
+
+#undef round2
+#define round2(x,s) ((((long) (x) - 1) & ~(long)((s)-1)) + (s))
+
+/* Pass the arguments in either registers, or in the stack. Using the
+ ppc sysv ABI, the first eight words of the argument list (that might
+ be less than eight parameters if some parameters occupy more than one
+ word) are passed in r3..r10 registers. float and double parameters are
+ passed in fpr's, in addition to that. Rest of the parameters if any
+ are passed in user stack.
+
+ If the function is returning a structure, then the return address is passed
+ in r3, then the first 7 words of the parametes can be passed in registers,
+ starting from r4. */
+
+CORE_ADDR
+ppc_sysv_abi_push_arguments (int nargs, struct value **args, CORE_ADDR sp,
+ int struct_return, CORE_ADDR struct_addr)
+{
+ int argno;
+ /* Next available general register for non-float, non-vector arguments. */
+ int greg;
+ /* Next available floating point register for float arguments. */
+ int freg;
+ /* Next available vector register for vector arguments. */
+ int vreg;
+ int argstkspace;
+ int structstkspace;
+ int argoffset;
+ int structoffset;
+ struct value *arg;
+ struct type *type;
+ int len;
+ char old_sp_buf[4];
+ CORE_ADDR saved_sp;
+
+ greg = struct_return ? 4 : 3;
+ freg = 1;
+ vreg = 2;
+ argstkspace = 0;
+ structstkspace = 0;
+
+ /* Figure out how much new stack space is required for arguments
+ which don't fit in registers. Unlike the PowerOpen ABI, the
+ SysV ABI doesn't reserve any extra space for parameters which
+ are put in registers. */
+ for (argno = 0; argno < nargs; argno++)
+ {
+ arg = args[argno];
+ type = check_typedef (VALUE_TYPE (arg));
+ len = TYPE_LENGTH (type);
+
+ if (TYPE_CODE (type) == TYPE_CODE_FLT)
+ {
+ if (freg <= 8)
+ freg++;
+ else
+ {
+ /* SysV ABI converts floats to doubles when placed in
+ memory and requires 8 byte alignment */
+ if (argstkspace & 0x4)
+ argstkspace += 4;
+ argstkspace += 8;
+ }
+ }
+ else if (TYPE_CODE (type) == TYPE_CODE_INT && len == 8) /* long long */
+ {
+ if (greg > 9)
+ {
+ greg = 11;
+ if (argstkspace & 0x4)
+ argstkspace += 4;
+ argstkspace += 8;
+ }
+ else
+ {
+ if ((greg & 1) == 0)
+ greg++;
+ greg += 2;
+ }
+ }
+ else if (!TYPE_VECTOR (type))
+ {
+ if (len > 4
+ || TYPE_CODE (type) == TYPE_CODE_STRUCT
+ || TYPE_CODE (type) == TYPE_CODE_UNION)
+ {
+ /* Rounding to the nearest multiple of 8 may not be necessary,
+ but it is safe. Particularly since we don't know the
+ field types of the structure */
+ structstkspace += round2 (len, 8);
+ }
+ if (greg <= 10)
+ greg++;
+ else
+ argstkspace += 4;
+ }
+ else
+ {
+ if (len == 16
+ && TYPE_CODE (type) == TYPE_CODE_ARRAY
+ && TYPE_VECTOR (type))
+ {
+ if (vreg <= 13)
+ vreg++;
+ else
+ {
+ /* Vector arguments must be aligned to 16 bytes on
+ the stack. */
+ argstkspace += round2 (argstkspace, 16);
+ argstkspace += 16;
+ }
+ }
+ }
+ }
+
+ /* Get current SP location */
+ saved_sp = read_sp ();
+
+ sp -= argstkspace + structstkspace;
+
+ /* Allocate space for backchain and callee's saved lr */
+ sp -= 8;
+
+ /* Make sure that we maintain 16 byte alignment */
+ sp &= ~0x0f;
+
+ /* Update %sp before proceeding any further */
+ write_register (SP_REGNUM, sp);
+
+ /* write the backchain */
+ store_address (old_sp_buf, 4, saved_sp);
+ write_memory (sp, old_sp_buf, 4);
+
+ argoffset = 8;
+ structoffset = argoffset + argstkspace;
+ freg = 1;
+ greg = 3;
+ vreg = 2;
+ /* Fill in r3 with the return structure, if any */
+ if (struct_return)
+ {
+ char val_buf[4];
+ store_address (val_buf, 4, struct_addr);
+ memcpy (®isters[REGISTER_BYTE (greg)], val_buf, 4);
+ greg++;
+ }
+ /* Now fill in the registers and stack... */
+ for (argno = 0; argno < nargs; argno++)
+ {
+ arg = args[argno];
+ type = check_typedef (VALUE_TYPE (arg));
+ len = TYPE_LENGTH (type);
+
+ if (TYPE_CODE (type) == TYPE_CODE_FLT)
+ {
+ if (freg <= 8)
+ {
+ if (len > 8)
+ printf_unfiltered (
+ "Fatal Error: a floating point parameter #%d with a size > 8 is found!\n", argno);
+ memcpy (®isters[REGISTER_BYTE (FP0_REGNUM + freg)],
+ VALUE_CONTENTS (arg), len);
+ freg++;
+ }
+ else
+ {
+ /* SysV ABI converts floats to doubles when placed in
+ memory and requires 8 byte alignment */
+ /* FIXME: Convert floats to doubles */
+ if (argoffset & 0x4)
+ argoffset += 4;
+ write_memory (sp + argoffset, (char *) VALUE_CONTENTS (arg), len);
+ argoffset += 8;
+ }
+ }
+ else if (TYPE_CODE (type) == TYPE_CODE_INT && len == 8) /* long long */
+ {
+ if (greg > 9)
+ {
+ greg = 11;
+ if (argoffset & 0x4)
+ argoffset += 4;
+ write_memory (sp + argoffset, (char *) VALUE_CONTENTS (arg), len);
+ argoffset += 8;
+ }
+ else
+ {
+ if ((greg & 1) == 0)
+ greg++;
+
+ memcpy (®isters[REGISTER_BYTE (greg)],
+ VALUE_CONTENTS (arg), 4);
+ memcpy (®isters[REGISTER_BYTE (greg + 1)],
+ VALUE_CONTENTS (arg) + 4, 4);
+ greg += 2;
+ }
+ }
+ else if (!TYPE_VECTOR (type))
+ {
+ char val_buf[4];
+ if (len > 4
+ || TYPE_CODE (type) == TYPE_CODE_STRUCT
+ || TYPE_CODE (type) == TYPE_CODE_UNION)
+ {
+ write_memory (sp + structoffset, VALUE_CONTENTS (arg), len);
+ store_address (val_buf, 4, sp + structoffset);
+ structoffset += round2 (len, 8);
+ }
+ else
+ {
+ memset (val_buf, 0, 4);
+ memcpy (val_buf, VALUE_CONTENTS (arg), len);
+ }
+ if (greg <= 10)
+ {
+ memcpy (®isters[REGISTER_BYTE (greg)], val_buf, 4);
+ greg++;
+ }
+ else
+ {
+ write_memory (sp + argoffset, val_buf, 4);
+ argoffset += 4;
+ }
+ }
+ else
+ {
+ if (len == 16
+ && TYPE_CODE (type) == TYPE_CODE_ARRAY
+ && TYPE_VECTOR (type))
+ {
+ struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch);
+ char *v_val_buf = alloca (16);
+ memset (v_val_buf, 0, 16);
+ memcpy (v_val_buf, VALUE_CONTENTS (arg), len);
+ if (vreg <= 13)
+ {
+ memcpy (®isters[REGISTER_BYTE (tdep->ppc_vr0_regnum
+ + vreg)],
+ v_val_buf, 16);
+ vreg++;
+ }
+ else
+ {
+ write_memory (sp + argoffset, v_val_buf, 16);
+ argoffset += 16;
+ }
+ }
+ }
+ }
+
+ target_store_registers (-1);
+ return sp;
+}
+
/* Function: ppc_push_return_address (pc, sp)
Set up the return address for the inferior function call. */
@@ -1195,6 +1452,32 @@
static CORE_ADDR rs6000_struct_return_address;
+/* Until November 2001, gcc was not complying to the SYSV ABI for
+ returning structures less than or equal to 8 bytes in size. It was
+ returning everything in memory. When this was corrected, it wasn't
+ fixed for native platforms. */
+int
+ppc_sysv_abi_broken_use_struct_convention (int gcc_p, struct type *value_type)
+{
+ if (TYPE_LENGTH (value_type) == 16
+ && TYPE_VECTOR (value_type))
+ return 0;
+
+ return generic_use_struct_convention (gcc_p, value_type);
+}
+
+/* Structures 8 bytes or less long are returned in the r3 & r4
+ registers, according to the SYSV ABI. */
+int
+ppc_sysv_abi_use_struct_convention (int gcc_p, struct type *value_type)
+{
+ if (TYPE_LENGTH (value_type) == 16
+ && TYPE_VECTOR (value_type))
+ return 0;
+
+ return (TYPE_LENGTH (value_type) > 8);
+}
+
/* Return whether handle_inferior_event() should proceed through code
starting at PC in function NAME when stepping.
@@ -2402,88 +2685,7 @@
return NULL;
}
-
-
-
\f
-static void
-process_note_abi_tag_sections (bfd *abfd, asection *sect, void *obj)
-{
- int *os_ident_ptr = obj;
- const char *name;
- unsigned int sectsize;
-
- name = bfd_get_section_name (abfd, sect);
- sectsize = bfd_section_size (abfd, sect);
- if (strcmp (name, ".note.ABI-tag") == 0 && sectsize > 0)
- {
- unsigned int name_length, data_length, note_type;
- char *note = alloca (sectsize);
-
- bfd_get_section_contents (abfd, sect, note,
- (file_ptr) 0, (bfd_size_type) sectsize);
-
- name_length = bfd_h_get_32 (abfd, note);
- data_length = bfd_h_get_32 (abfd, note + 4);
- note_type = bfd_h_get_32 (abfd, note + 8);
-
- if (name_length == 4 && data_length == 16 && note_type == 1
- && strcmp (note + 12, "GNU") == 0)
- {
- int os_number = bfd_h_get_32 (abfd, note + 16);
-
- /* The case numbers are from abi-tags in glibc */
- switch (os_number)
- {
- case 0 :
- *os_ident_ptr = ELFOSABI_LINUX;
- break;
- case 1 :
- *os_ident_ptr = ELFOSABI_HURD;
- break;
- case 2 :
- *os_ident_ptr = ELFOSABI_SOLARIS;
- break;
- default :
- internal_error (__FILE__, __LINE__,
- "process_note_abi_sections: unknown OS number %d",
- os_number);
- break;
- }
- }
- }
-}
-
-/* Return one of the ELFOSABI_ constants for BFDs representing ELF
- executables. If it's not an ELF executable or if the OS/ABI couldn't
- be determined, simply return -1. */
-
-static int
-get_elfosabi (bfd *abfd)
-{
- int elfosabi = -1;
-
- if (abfd != NULL && bfd_get_flavour (abfd) == bfd_target_elf_flavour)
- {
- elfosabi = elf_elfheader (abfd)->e_ident[EI_OSABI];
-
- /* When elfosabi is 0 (ELFOSABI_NONE), this is supposed to indicate
- that we're on a SYSV system. However, GNU/Linux uses a note section
- to record OS/ABI info, but leaves e_ident[EI_OSABI] zero. So we
- have to check the note sections too. */
- if (elfosabi == 0)
- {
- bfd_map_over_sections (abfd,
- process_note_abi_tag_sections,
- &elfosabi);
- }
- }
-
- return elfosabi;
-}
-
-\f
-
/* Initialize the current architecture based on INFO. If possible, re-use an
architecture from ARCHES, which is a list of architectures already created
during this debugging session.
@@ -2502,7 +2704,8 @@
enum bfd_architecture arch;
unsigned long mach;
bfd abfd;
- int osabi, sysv_abi;
+ int sysv_abi;
+ enum gdb_osabi osabi = GDB_OSABI_UNKNOWN;
gdbarch_print_insn_ftype *print_insn;
from_xcoff_exec = info.abfd && info.abfd->format == bfd_object &&
@@ -2513,7 +2716,8 @@
sysv_abi = info.abfd && bfd_get_flavour (info.abfd) == bfd_target_elf_flavour;
- osabi = get_elfosabi (info.abfd);
+ if (info.abfd)
+ osabi = gdbarch_lookup_osabi (info.abfd);
/* Check word size. If INFO is from a binary file, infer it from
that, else choose a likely default. */
@@ -2726,57 +2930,23 @@
/* Not sure on this. FIXMEmgo */
set_gdbarch_frame_args_skip (gdbarch, 8);
- /* Until November 2001, gcc was not complying to the SYSV ABI for
- returning structures less than or equal to 8 bytes in size. It was
- returning everything in memory. When this was corrected, it wasn't
- fixed for native platforms. */
if (sysv_abi)
- {
- if (osabi == ELFOSABI_LINUX
- || osabi == ELFOSABI_NETBSD
- || osabi == ELFOSABI_FREEBSD)
- set_gdbarch_use_struct_convention (gdbarch,
- ppc_sysv_abi_broken_use_struct_convention);
- else
- set_gdbarch_use_struct_convention (gdbarch,
- ppc_sysv_abi_use_struct_convention);
- }
+ set_gdbarch_use_struct_convention (gdbarch,
+ ppc_sysv_abi_use_struct_convention);
else
- {
- set_gdbarch_use_struct_convention (gdbarch,
- generic_use_struct_convention);
- }
+ set_gdbarch_use_struct_convention (gdbarch,
+ generic_use_struct_convention);
set_gdbarch_frame_chain_valid (gdbarch, file_frame_chain_valid);
- /* Note: kevinb/2002-04-12: See note above regarding *_push_arguments().
- The same remarks hold for the methods below. */
- if (osabi == ELFOSABI_LINUX && wordsize == 4)
- {
- set_gdbarch_frameless_function_invocation (gdbarch,
- ppc_linux_frameless_function_invocation);
- set_gdbarch_frame_chain (gdbarch, ppc_linux_frame_chain);
- set_gdbarch_frame_saved_pc (gdbarch, ppc_linux_frame_saved_pc);
-
- set_gdbarch_frame_init_saved_regs (gdbarch,
- ppc_linux_frame_init_saved_regs);
- set_gdbarch_init_extra_frame_info (gdbarch,
- ppc_linux_init_extra_frame_info);
-
- set_gdbarch_memory_remove_breakpoint (gdbarch,
- ppc_linux_memory_remove_breakpoint);
- set_solib_svr4_fetch_link_map_offsets
- (gdbarch, ppc_linux_svr4_fetch_link_map_offsets);
- }
- else
- {
- set_gdbarch_frameless_function_invocation (gdbarch,
- rs6000_frameless_function_invocation);
- set_gdbarch_frame_chain (gdbarch, rs6000_frame_chain);
- set_gdbarch_frame_saved_pc (gdbarch, rs6000_frame_saved_pc);
- set_gdbarch_frame_init_saved_regs (gdbarch, rs6000_frame_init_saved_regs);
- set_gdbarch_init_extra_frame_info (gdbarch, rs6000_init_extra_frame_info);
- }
+ set_gdbarch_frameless_function_invocation (gdbarch,
+ rs6000_frameless_function_invocation);
+ set_gdbarch_frame_chain (gdbarch, rs6000_frame_chain);
+ set_gdbarch_frame_saved_pc (gdbarch, rs6000_frame_saved_pc);
+
+ set_gdbarch_frame_init_saved_regs (gdbarch, rs6000_frame_init_saved_regs);
+ set_gdbarch_init_extra_frame_info (gdbarch, rs6000_init_extra_frame_info);
+
if (!sysv_abi)
{
/* Handle RS/6000 function pointers (which are really function
@@ -2792,9 +2962,24 @@
now that the C compiler delays popping them. */
set_gdbarch_frame_num_args (gdbarch, frame_num_args_unknown);
+ /* Hook in ABI-specific overrides, if they have been registered. */
+ gdbarch_init_osabi (info, gdbarch, osabi);
+
return gdbarch;
}
+static void
+rs6000_dump_tdep (struct gdbarch *current_gdbarch, struct ui_file *file)
+{
+ struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch);
+
+ if (tdep == NULL)
+ return;
+
+ fprintf_unfiltered (file, "rs6000_dump_tdep: OS ABI = %s\n",
+ gdbarch_osabi_name (tdep->osabi));
+}
+
static struct cmd_list_element *info_powerpc_cmdlist = NULL;
static void
@@ -2808,8 +2993,8 @@
void
_initialize_rs6000_tdep (void)
{
- register_gdbarch_init (bfd_arch_rs6000, rs6000_gdbarch_init);
- register_gdbarch_init (bfd_arch_powerpc, rs6000_gdbarch_init);
+ gdbarch_register (bfd_arch_rs6000, rs6000_gdbarch_init, rs6000_dump_tdep);
+ gdbarch_register (bfd_arch_powerpc, rs6000_gdbarch_init, rs6000_dump_tdep);
/* Add root prefix command for "info powerpc" commands */
add_prefix_cmd ("powerpc", class_info, rs6000_info_powerpc_command,
@@ -2819,5 +3004,4 @@
add_cmd ("altivec", class_info, rs6000_altivec_registers_info,
"Display the contents of the AltiVec registers.",
&info_powerpc_cmdlist);
-
}
Index: config/powerpc/nbsd.mh
===================================================================
RCS file: /cvs/src/src/gdb/config/powerpc/nbsd.mh,v
retrieving revision 1.10
diff -u -r1.10 nbsd.mh
--- config/powerpc/nbsd.mh 28 May 2002 18:43:08 -0000 1.10
+++ config/powerpc/nbsd.mh 28 May 2002 19:15:53 -0000
@@ -1,3 +1,3 @@
# Host: PowerPC, running NetBSD
-NATDEPFILES= fork-child.o infptrace.o inftarg.o ppcnbsd-nat.o solib-legacy.o
+NATDEPFILES= fork-child.o infptrace.o inftarg.o ppcnbsd-nat.o
NAT_FILE= nm-nbsd.h
Index: config/powerpc/nbsd.mt
===================================================================
RCS file: /cvs/src/src/gdb/config/powerpc/nbsd.mt,v
retrieving revision 1.10
diff -u -r1.10 nbsd.mt
--- config/powerpc/nbsd.mt 28 May 2002 18:43:08 -0000 1.10
+++ config/powerpc/nbsd.mt 28 May 2002 19:15:53 -0000
@@ -1,5 +1,5 @@
# Target: PowerPC, running NetBSD
-TDEPFILES= rs6000-tdep.o ppcnbsd-tdep.o nbsd-tdep.o ppc-linux-tdep.o corelow.o \
+TDEPFILES= rs6000-tdep.o ppcnbsd-tdep.o nbsd-tdep.o corelow.o \
solib.o solib-svr4.o
TM_FILE= tm-nbsd.h
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [PATCH/RFA] Generic OS ABI handling for PowerPC 2002-05-28 12:41 [PATCH/RFA] Generic OS ABI handling for PowerPC Jason R Thorpe @ 2002-05-28 12:52 ` Kevin Buettner 2002-05-28 13:46 ` Jason R Thorpe 0 siblings, 1 reply; 11+ messages in thread From: Kevin Buettner @ 2002-05-28 12:52 UTC (permalink / raw) To: Jason R Thorpe, gdb-patches On May 28, 12:26pm, Jason R Thorpe wrote: > This makes the PowerPC/RS6000 target use the generic OS ABI framework, > and also moves some SVR4 ABI stuff out of ppc-linux-tdep.c into > rs6000-tdep.c. Why did you move the SVR4 ABI stuff out of ppc-linux-tdep.c? I know there's a FIXME comment to that effect in this file, but I don't think that rs6000-tdep.c is really the right place for for the SVR4 ABI code. Kevin ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH/RFA] Generic OS ABI handling for PowerPC 2002-05-28 12:52 ` Kevin Buettner @ 2002-05-28 13:46 ` Jason R Thorpe 2002-05-28 14:12 ` Kevin Buettner 0 siblings, 1 reply; 11+ messages in thread From: Jason R Thorpe @ 2002-05-28 13:46 UTC (permalink / raw) To: Kevin Buettner; +Cc: gdb-patches On Tue, May 28, 2002 at 12:41:09PM -0700, Kevin Buettner wrote: > Why did you move the SVR4 ABI stuff out of ppc-linux-tdep.c? I know > there's a FIXME comment to that effect in this file, but I don't think > that rs6000-tdep.c is really the right place for for the SVR4 ABI > code. Well, ppc-linux-tdep.c isn't exactly the right place for it, either. I could add another ppc-tdep.c file, but it seems to me that as long as rs6000-tdep.c makes references to sysv abi stuff, it should be actually in rs6000-tdep.c. -- -- Jason R. Thorpe <thorpej@wasabisystems.com> ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH/RFA] Generic OS ABI handling for PowerPC 2002-05-28 13:46 ` Jason R Thorpe @ 2002-05-28 14:12 ` Kevin Buettner 2002-05-28 16:35 ` Jason R Thorpe 0 siblings, 1 reply; 11+ messages in thread From: Kevin Buettner @ 2002-05-28 14:12 UTC (permalink / raw) To: Jason R Thorpe, Kevin Buettner; +Cc: gdb-patches On May 28, 12:52pm, Jason R Thorpe wrote: > On Tue, May 28, 2002 at 12:41:09PM -0700, Kevin Buettner wrote: > > > Why did you move the SVR4 ABI stuff out of ppc-linux-tdep.c? I know > > there's a FIXME comment to that effect in this file, but I don't think > > that rs6000-tdep.c is really the right place for for the SVR4 ABI > > code. > > Well, ppc-linux-tdep.c isn't exactly the right place for it, either. Agreed, but I think it's a better place than rs6000-tdep.c. > I could add another ppc-tdep.c file, but it seems to me that as long as > rs6000-tdep.c makes references to sysv abi stuff, it should be actually in > rs6000-tdep.c. Umm... rs6000-tdep.c references ppc_linux_* functions too. Are you suggesting that they too should be moved to rs6000-tdep.c? The way I view it is that (for historical reasons) rs6000-tdep.c contains architecture specific code for the PowerPC and Power architectures. It also contains PowerOpen ABI specific code as well as code specific to the AIX operating system. BTW, the fact that it contains PowerOpen ABI specific code is the reason that I object to moving the SVR4 ABI specific code there. IMO, ideally, the architecture specific code ought to go in it's own file. The ABI-specific code ought to go in separate files of their own and inherit from the arch specific file. The OS-dependent code ought to go in still other files and ought to inherit from the ABI specific code. For the Power/PowerPC targets, I'd like to someday break it down as follows: ppc-tdep.c - generic Power / PowerPC architecture ppc-sysv-tdep.c - PowerPC SysV ABI ppc-eabi-tdep.c - PowerPC EABI ppc-poweropen-tdep.c - Power / PowerPC PowerOpen ABI (or whatever it is that IBM's calling it these days.) ... (maybe more ABI specific files) ppc-linux-tdep.c - Linux specific code for Power/PowerPC ppc-aix-tdep.c - AIX specific code for Power/PowerPC ... (maybe more OS specific files) The gdbarch machinery doesn't yet have an inheritance mechanism, so, unfortunately, we're left with the situation whereby (e.g.) the rs6000_gdbarch_init() function has to refer to a lot of code that it really shouldn't be referring to. It also makes arranging the code as described above somewhat more difficult. Kevin ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH/RFA] Generic OS ABI handling for PowerPC 2002-05-28 14:12 ` Kevin Buettner @ 2002-05-28 16:35 ` Jason R Thorpe 2002-05-28 16:46 ` Kevin Buettner 0 siblings, 1 reply; 11+ messages in thread From: Jason R Thorpe @ 2002-05-28 16:35 UTC (permalink / raw) To: Kevin Buettner; +Cc: gdb-patches On Tue, May 28, 2002 at 01:46:29PM -0700, Kevin Buettner wrote: > Umm... rs6000-tdep.c references ppc_linux_* functions too. Are > you suggesting that they too should be moved to rs6000-tdep.c? No it doesn't. My patch moves all references to those functions into ppc-linux-tdep.c only, and uses the GDB OS ABI stuff to hook it in. > ppc-sysv-tdep.c - PowerPC SysV ABI I'm happy to move the ppc_sysv stuff into ppc-sysv-tdep.c, then, for now, and add ppc-sysv-tdep.o to all the targets that currently use rs6000-tdep.o. Would that be an acceptable interim solution? It seems certainly better than requiring all PowerPC targets to slurp in ppc-linux-tdep.o (which contains a lot of code that is completely useless for NetBSD, for example). -- -- Jason R. Thorpe <thorpej@wasabisystems.com> ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH/RFA] Generic OS ABI handling for PowerPC 2002-05-28 16:35 ` Jason R Thorpe @ 2002-05-28 16:46 ` Kevin Buettner 2002-05-28 16:47 ` Jason R Thorpe 2002-05-28 17:59 ` Jason R Thorpe 0 siblings, 2 replies; 11+ messages in thread From: Kevin Buettner @ 2002-05-28 16:46 UTC (permalink / raw) To: Jason R Thorpe, Kevin Buettner; +Cc: gdb-patches On May 28, 3:26pm, Jason R Thorpe wrote: > On Tue, May 28, 2002 at 01:46:29PM -0700, Kevin Buettner wrote: > > > Umm... rs6000-tdep.c references ppc_linux_* functions too. Are > > you suggesting that they too should be moved to rs6000-tdep.c? > > No it doesn't. My patch moves all references to those functions into > ppc-linux-tdep.c only, and uses the GDB OS ABI stuff to hook it in. I see. Up 'til now I hadn't looked at the new machinery you added. Nicely done. > > ppc-sysv-tdep.c - PowerPC SysV ABI > > I'm happy to move the ppc_sysv stuff into ppc-sysv-tdep.c, then, for now, > and add ppc-sysv-tdep.o to all the targets that currently use rs6000-tdep.o. > > Would that be an acceptable interim solution? Yeah, I like this better. > It seems certainly better than requiring all PowerPC targets to > slurp in ppc-linux-tdep.o (which contains a lot of code that is > completely useless for NetBSD, for example). Hmm. I see where you're coming from, but be careful in how you go about it because we don't want to preclude the building of a cross debugger which can debug both NetBSD/PPC and Linux/PPC binaries (connected to a suitable target of course). Kevin ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH/RFA] Generic OS ABI handling for PowerPC 2002-05-28 16:46 ` Kevin Buettner @ 2002-05-28 16:47 ` Jason R Thorpe 2002-05-28 19:42 ` Andrew Cagney 2002-05-28 17:59 ` Jason R Thorpe 1 sibling, 1 reply; 11+ messages in thread From: Jason R Thorpe @ 2002-05-28 16:47 UTC (permalink / raw) To: Kevin Buettner; +Cc: gdb-patches On Tue, May 28, 2002 at 04:35:17PM -0700, Kevin Buettner wrote: > > Would that be an acceptable interim solution? > > Yeah, I like this better. Ok, I'll post another patch shortly. > Hmm. I see where you're coming from, but be careful in how you go > about it because we don't want to preclude the building of a cross > debugger which can debug both NetBSD/PPC and Linux/PPC binaries > (connected to a suitable target of course). Yes, I understand. But it still seems correct to cleanly separate the targets so that one day you'll be able to specify precisely which ones you want. (Even without my patch, you can't properly cross debug both Linux and NetBSD PowerPC binaries in the same gdb, because the Linux cross debug stuff is incomplete.) So, yes, one day I want to be able to say: --target=powerpc-unknown-netbsd,powerpc-unknown-linux-gnu ...or whatever, and have the right thing happen, all based on the markers in the debugged executable that the gdbarch_osabi_*() stuff can examine (the osabi stuff will even warn you if you try to debug an executable which requires OS ABI support not compiled into GDB). -- -- Jason R. Thorpe <thorpej@wasabisystems.com> ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH/RFA] Generic OS ABI handling for PowerPC 2002-05-28 16:47 ` Jason R Thorpe @ 2002-05-28 19:42 ` Andrew Cagney 0 siblings, 0 replies; 11+ messages in thread From: Andrew Cagney @ 2002-05-28 19:42 UTC (permalink / raw) To: Jason R Thorpe; +Cc: Kevin Buettner, gdb-patches > So, yes, one day I want to be able to say: > > --target=powerpc-unknown-netbsd,powerpc-unknown-linux-gnu From memory the syntax is: --target=powerpc-unknown-netbsd \ --enable-targets=powerpc-unknown-linux-gnu same as BFD. There is a very old WIP patch to add this, sounds like it it time to try it again. Andrew > ...or whatever, and have the right thing happen, all based on the > markers in the debugged executable that the gdbarch_osabi_*() stuff > can examine (the osabi stuff will even warn you if you try to debug > an executable which requires OS ABI support not compiled into GDB). ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH/RFA] Generic OS ABI handling for PowerPC 2002-05-28 16:46 ` Kevin Buettner 2002-05-28 16:47 ` Jason R Thorpe @ 2002-05-28 17:59 ` Jason R Thorpe 2002-05-29 17:37 ` Kevin Buettner 1 sibling, 1 reply; 11+ messages in thread From: Jason R Thorpe @ 2002-05-28 17:59 UTC (permalink / raw) To: Kevin Buettner; +Cc: gdb-patches [-- Attachment #1: Type: text/plain, Size: 1594 bytes --] On Tue, May 28, 2002 at 04:35:17PM -0700, Kevin Buettner wrote: > Yeah, I like this better. Ok, new patch. Moves the ppc_sysv stuff into new ppc-sysv-tdep.c, and updates the various .mt files. I also touched up some dependency lists in Makefile.in. OK to commit? * Makefile.in (ppc_tdep_h): Define. (ppc-linux-nat.o, ppc-linux-tdep.o, rs6000-tdep.o): Use $(ppc_tdep_h). (ppc-sysv-tdep.o, ppcnbsd-nat.o, ppcnbsd-tdep.o): New dependency lists. * ppc-tdep.h: Use generic OS ABI framework. * ppc-linux-tdep.c (_initialize_ppc_linux_tdep, ppc_linux_init_abi): New functions. (ppc_sysv_abi_broken_use_struct_convention, ppc_sysv_abi_use_struct_convention, ppc_sysv_abi_push_arguments): Move to... * ppc-sysv-tdep.c: ...here. * ppcnbsd-nat.c: Don't include gdbcore.h and regcache.h. * rs6000-tdep.c (process_note_abi_tag_sections, get_elfosabi): Remove. (rs6000_gdbarch_init): Use generic OS ABI framework. (rs6000_dump_tdep): New function. (_initialize_rs6000_tdep): Use gdbarch_register. * config/powerpc/linux.mt (TDEPFILES): Add ppc-sysv-tdep.o. * config/powerpc/nbsd.mh (NATDEPFILES): Remove solib-legacy.o. * config/powerpc/aix.mt (TDEPFILES): Use ppc-sysv-tdep.o instead of ppc-linux-tdep.o. * config/powerpc/nbsd.mt (TDEPFILES): Likewise. * config/powerpc/ppc-eabi.mt (TDEPFILES): Likewise. * config/powerpc/ppc-sim.mt (TDEPFILES): Likewise. * config/powerpc/ppcle-eabi.mt (TDEPFILES): Likewise. * config/powerpc/ppcle-sim.mt (TDEPFILES): Likewise. * config/powerpc/vxworks.mt (TDEPFILES): Likewise. -- -- Jason R. Thorpe <thorpej@wasabisystems.com> [-- Attachment #2: ppc-patch --] [-- Type: text/plain, Size: 35626 bytes --] Index: Makefile.in =================================================================== RCS file: /cvs/src/src/gdb/Makefile.in,v retrieving revision 1.200 diff -u -r1.200 Makefile.in --- Makefile.in 24 May 2002 00:12:16 -0000 1.200 +++ Makefile.in 29 May 2002 00:07:43 -0000 @@ -646,6 +646,7 @@ monitor_h = monitor.h objfiles_h = objfiles.h parser_defs_h = parser-defs.h $(doublest_h) +ppc_tdep_h = ppc-tdep.h osabi.h regcache_h = regcache.h remote_h = remote.h remote_utils_h = remote-utils.h $(target_h) @@ -1225,7 +1226,8 @@ nindy-share/ttyflush.c nindy-tdep.c \ ns32k-tdep.c solib-osf.c \ somread.c somsolib.c $(HPREAD_SOURCE) \ - ppc-linux-nat.c ppc-linux-tdep.c \ + ppc-sysv-tdep.o ppc-linux-nat.c ppc-linux-tdep.c \ + ppcnbsd-nat.o ppcnbsd-tdep.o \ procfs.c \ remote-adapt.c remote-array.c remote-bug.c remote-e7000.c remote-eb.c \ remote-es.c remote-hms.c remote-mips.c \ @@ -1899,10 +1901,20 @@ $(objfiles_h) $(gdb_stabs_h) $(serial_h) ocd.h $(regcache_h) ppc-linux-nat.o: ppc-linux-nat.c $(defs_h) $(gdbcore_h) $(frame_h) \ - $(inferior_h) $(target_h) $(regcache_h) ppc-tdep.h + $(inferior_h) $(target_h) $(regcache_h) $(ppc_tdep_h) ppc-linux-tdep.o: ppc-linux-tdep.c $(defs_h) $(gdbcore_h) $(inferior_h) \ - $(target_h) ppc-tdep.h $(regcache_h) $(value_h) + $(target_h) $(ppc_tdep_h) $(regcache_h) $(value_h) + +ppc-sysv-tdep.o: ppc-sysv-tdep.c $(defs_h) $(gdbcore_h) $(inferior_h) \ + $(value_h) $(regcache_h) $(ppc_tdep_h) + +ppcnbsd-nat.o: ppcnbsd-nat.c $(defs_h) $(inferior_h) $(ppc_tdep_h) \ + ppcnbsd-tdep.h + +ppcnbsd-tdep.o: ppcnbsd-tdep.c $(defs_h) $(gdbcore_h) $(regcache_h) \ + $(target_h) $(breakpoint_h) $(value_h) $(ppc_tdep_t) ppcnbsd-tdep.h \ + nbsd-tdep.h ppcbug-rom.o: ppcbug-rom.c $(monitor_h) $(bfd_h) $(gdb_wait_h) $(defs_h) \ $(gdbcmd_h) $(inferior_h) $(target_h) $(serial_h) $(terminal_h) \ @@ -2038,7 +2050,7 @@ $(gdb_stabs_h) $(regcache_h) $(arch_utils_h) rs6000-tdep.o: rs6000-tdep.c $(defs_h) $(gdbcore_h) $(inferior_h) \ - $(target_h) ppc-tdep.h $(regcache_h) $(value_h) $(parser_defs_h) + $(target_h) $(ppc_tdep_h) $(regcache_h) $(value_h) $(parser_defs_h) s390-tdep.o: s390-tdep.c $(defs_h) $(arch_utils_h) $(frame_h) $(inferior_h) \ $(symtab_h) $(target_h) $(gdbcore_h) $(gdbcmd_h) $(symfile_h) \ Index: ppc-linux-tdep.c =================================================================== RCS file: /cvs/src/src/gdb/ppc-linux-tdep.c,v retrieving revision 1.17 diff -u -r1.17 ppc-linux-tdep.c --- ppc-linux-tdep.c 29 Apr 2002 15:10:06 -0000 1.17 +++ ppc-linux-tdep.c 29 May 2002 00:07:48 -0000 @@ -411,293 +411,6 @@ return rs6000_frame_chain (thisframe); } -/* FIXME: Move the following to rs6000-tdep.c (or some other file where - it may be used generically by ports which use either the SysV ABI or - the EABI */ - -/* Until November 2001, gcc was not complying to the SYSV ABI for - returning structures less than or equal to 8 bytes in size. It was - returning everything in memory. When this was corrected, it wasn't - fixed for native platforms. */ -int -ppc_sysv_abi_broken_use_struct_convention (int gcc_p, struct type *value_type) -{ - if (TYPE_LENGTH (value_type) == 16 - && TYPE_VECTOR (value_type)) - return 0; - - return generic_use_struct_convention (gcc_p, value_type); -} - -/* Structures 8 bytes or less long are returned in the r3 & r4 - registers, according to the SYSV ABI. */ -int -ppc_sysv_abi_use_struct_convention (int gcc_p, struct type *value_type) -{ - if (TYPE_LENGTH (value_type) == 16 - && TYPE_VECTOR (value_type)) - return 0; - - return (TYPE_LENGTH (value_type) > 8); -} - -/* round2 rounds x up to the nearest multiple of s assuming that s is a - power of 2 */ - -#undef round2 -#define round2(x,s) ((((long) (x) - 1) & ~(long)((s)-1)) + (s)) - -/* Pass the arguments in either registers, or in the stack. Using the - ppc sysv ABI, the first eight words of the argument list (that might - be less than eight parameters if some parameters occupy more than one - word) are passed in r3..r10 registers. float and double parameters are - passed in fpr's, in addition to that. Rest of the parameters if any - are passed in user stack. - - If the function is returning a structure, then the return address is passed - in r3, then the first 7 words of the parametes can be passed in registers, - starting from r4. */ - -CORE_ADDR -ppc_sysv_abi_push_arguments (int nargs, struct value **args, CORE_ADDR sp, - int struct_return, CORE_ADDR struct_addr) -{ - int argno; - /* Next available general register for non-float, non-vector arguments. */ - int greg; - /* Next available floating point register for float arguments. */ - int freg; - /* Next available vector register for vector arguments. */ - int vreg; - int argstkspace; - int structstkspace; - int argoffset; - int structoffset; - struct value *arg; - struct type *type; - int len; - char old_sp_buf[4]; - CORE_ADDR saved_sp; - - greg = struct_return ? 4 : 3; - freg = 1; - vreg = 2; - argstkspace = 0; - structstkspace = 0; - - /* Figure out how much new stack space is required for arguments - which don't fit in registers. Unlike the PowerOpen ABI, the - SysV ABI doesn't reserve any extra space for parameters which - are put in registers. */ - for (argno = 0; argno < nargs; argno++) - { - arg = args[argno]; - type = check_typedef (VALUE_TYPE (arg)); - len = TYPE_LENGTH (type); - - if (TYPE_CODE (type) == TYPE_CODE_FLT) - { - if (freg <= 8) - freg++; - else - { - /* SysV ABI converts floats to doubles when placed in - memory and requires 8 byte alignment */ - if (argstkspace & 0x4) - argstkspace += 4; - argstkspace += 8; - } - } - else if (TYPE_CODE (type) == TYPE_CODE_INT && len == 8) /* long long */ - { - if (greg > 9) - { - greg = 11; - if (argstkspace & 0x4) - argstkspace += 4; - argstkspace += 8; - } - else - { - if ((greg & 1) == 0) - greg++; - greg += 2; - } - } - else if (!TYPE_VECTOR (type)) - { - if (len > 4 - || TYPE_CODE (type) == TYPE_CODE_STRUCT - || TYPE_CODE (type) == TYPE_CODE_UNION) - { - /* Rounding to the nearest multiple of 8 may not be necessary, - but it is safe. Particularly since we don't know the - field types of the structure */ - structstkspace += round2 (len, 8); - } - if (greg <= 10) - greg++; - else - argstkspace += 4; - } - else - { - if (len == 16 - && TYPE_CODE (type) == TYPE_CODE_ARRAY - && TYPE_VECTOR (type)) - { - if (vreg <= 13) - vreg++; - else - { - /* Vector arguments must be aligned to 16 bytes on - the stack. */ - argstkspace += round2 (argstkspace, 16); - argstkspace += 16; - } - } - } - } - - /* Get current SP location */ - saved_sp = read_sp (); - - sp -= argstkspace + structstkspace; - - /* Allocate space for backchain and callee's saved lr */ - sp -= 8; - - /* Make sure that we maintain 16 byte alignment */ - sp &= ~0x0f; - - /* Update %sp before proceeding any further */ - write_register (SP_REGNUM, sp); - - /* write the backchain */ - store_address (old_sp_buf, 4, saved_sp); - write_memory (sp, old_sp_buf, 4); - - argoffset = 8; - structoffset = argoffset + argstkspace; - freg = 1; - greg = 3; - vreg = 2; - /* Fill in r3 with the return structure, if any */ - if (struct_return) - { - char val_buf[4]; - store_address (val_buf, 4, struct_addr); - memcpy (®isters[REGISTER_BYTE (greg)], val_buf, 4); - greg++; - } - /* Now fill in the registers and stack... */ - for (argno = 0; argno < nargs; argno++) - { - arg = args[argno]; - type = check_typedef (VALUE_TYPE (arg)); - len = TYPE_LENGTH (type); - - if (TYPE_CODE (type) == TYPE_CODE_FLT) - { - if (freg <= 8) - { - if (len > 8) - printf_unfiltered ( - "Fatal Error: a floating point parameter #%d with a size > 8 is found!\n", argno); - memcpy (®isters[REGISTER_BYTE (FP0_REGNUM + freg)], - VALUE_CONTENTS (arg), len); - freg++; - } - else - { - /* SysV ABI converts floats to doubles when placed in - memory and requires 8 byte alignment */ - /* FIXME: Convert floats to doubles */ - if (argoffset & 0x4) - argoffset += 4; - write_memory (sp + argoffset, (char *) VALUE_CONTENTS (arg), len); - argoffset += 8; - } - } - else if (TYPE_CODE (type) == TYPE_CODE_INT && len == 8) /* long long */ - { - if (greg > 9) - { - greg = 11; - if (argoffset & 0x4) - argoffset += 4; - write_memory (sp + argoffset, (char *) VALUE_CONTENTS (arg), len); - argoffset += 8; - } - else - { - if ((greg & 1) == 0) - greg++; - - memcpy (®isters[REGISTER_BYTE (greg)], - VALUE_CONTENTS (arg), 4); - memcpy (®isters[REGISTER_BYTE (greg + 1)], - VALUE_CONTENTS (arg) + 4, 4); - greg += 2; - } - } - else if (!TYPE_VECTOR (type)) - { - char val_buf[4]; - if (len > 4 - || TYPE_CODE (type) == TYPE_CODE_STRUCT - || TYPE_CODE (type) == TYPE_CODE_UNION) - { - write_memory (sp + structoffset, VALUE_CONTENTS (arg), len); - store_address (val_buf, 4, sp + structoffset); - structoffset += round2 (len, 8); - } - else - { - memset (val_buf, 0, 4); - memcpy (val_buf, VALUE_CONTENTS (arg), len); - } - if (greg <= 10) - { - memcpy (®isters[REGISTER_BYTE (greg)], val_buf, 4); - greg++; - } - else - { - write_memory (sp + argoffset, val_buf, 4); - argoffset += 4; - } - } - else - { - if (len == 16 - && TYPE_CODE (type) == TYPE_CODE_ARRAY - && TYPE_VECTOR (type)) - { - struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch); - char *v_val_buf = alloca (16); - memset (v_val_buf, 0, 16); - memcpy (v_val_buf, VALUE_CONTENTS (arg), len); - if (vreg <= 13) - { - memcpy (®isters[REGISTER_BYTE (tdep->ppc_vr0_regnum - + vreg)], - v_val_buf, 16); - vreg++; - } - else - { - write_memory (sp + argoffset, v_val_buf, 16); - argoffset += 16; - } - } - } - } - - target_store_registers (-1); - return sp; -} - /* ppc_linux_memory_remove_breakpoints attempts to remove a breakpoint in much the same fashion as memory_remove_breakpoint in mem-break.c, but is careful not to write back the previous contents if the code @@ -885,4 +598,45 @@ } return lmp; +} + +static void +ppc_linux_init_abi (struct gdbarch_info info, + struct gdbarch *gdbarch) +{ + struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch); + + /* Until November 2001, gcc was not complying to the SYSV ABI for + returning structures less than or equal to 8 bytes in size. It was + returning everything in memory. When this was corrected, it wasn't + fixed for native platforms. */ + set_gdbarch_use_struct_convention (gdbarch, + ppc_sysv_abi_broken_use_struct_convention); + + if (tdep->wordsize == 4) + { + /* Note: kevinb/2002-04-12: See note in rs6000_gdbarch_init regarding + *_push_arguments(). The same remarks hold for the methods below. */ + set_gdbarch_frameless_function_invocation (gdbarch, + ppc_linux_frameless_function_invocation); + set_gdbarch_frame_chain (gdbarch, ppc_linux_frame_chain); + set_gdbarch_frame_saved_pc (gdbarch, ppc_linux_frame_saved_pc); + + set_gdbarch_frame_init_saved_regs (gdbarch, + ppc_linux_frame_init_saved_regs); + set_gdbarch_init_extra_frame_info (gdbarch, + ppc_linux_init_extra_frame_info); + + set_gdbarch_memory_remove_breakpoint (gdbarch, + ppc_linux_memory_remove_breakpoint); + set_solib_svr4_fetch_link_map_offsets + (gdbarch, ppc_linux_svr4_fetch_link_map_offsets); + } +} + +void +_initialize_ppc_linux_tdep (void) +{ + gdbarch_register_osabi (bfd_arch_powerpc, GDB_OSABI_LINUX, + ppc_linux_init_abi); } Index: ppc-sysv-tdep.c =================================================================== RCS file: ppc-sysv-tdep.c diff -N ppc-sysv-tdep.c --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ ppc-sysv-tdep.c 29 May 2002 00:07:49 -0000 @@ -0,0 +1,312 @@ +/* Target-dependent code for PowerPC systems using the SVR4 ABI + for GDB, the GNU debugger. + + Copyright 2000, 2001, 2002 Free Software Foundation, Inc. + + This file is part of GDB. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 59 Temple Place - Suite 330, + Boston, MA 02111-1307, USA. */ + +#include "defs.h" +#include "gdbcore.h" +#include "inferior.h" +#include "regcache.h" +#include "value.h" + +#include "ppc-tdep.h" + +/* round2 rounds x up to the nearest multiple of s assuming that s is a + power of 2 */ + +#undef round2 +#define round2(x,s) ((((long) (x) - 1) & ~(long)((s)-1)) + (s)) + +/* Pass the arguments in either registers, or in the stack. Using the + ppc sysv ABI, the first eight words of the argument list (that might + be less than eight parameters if some parameters occupy more than one + word) are passed in r3..r10 registers. float and double parameters are + passed in fpr's, in addition to that. Rest of the parameters if any + are passed in user stack. + + If the function is returning a structure, then the return address is passed + in r3, then the first 7 words of the parametes can be passed in registers, + starting from r4. */ + +CORE_ADDR +ppc_sysv_abi_push_arguments (int nargs, struct value **args, CORE_ADDR sp, + int struct_return, CORE_ADDR struct_addr) +{ + int argno; + /* Next available general register for non-float, non-vector arguments. */ + int greg; + /* Next available floating point register for float arguments. */ + int freg; + /* Next available vector register for vector arguments. */ + int vreg; + int argstkspace; + int structstkspace; + int argoffset; + int structoffset; + struct value *arg; + struct type *type; + int len; + char old_sp_buf[4]; + CORE_ADDR saved_sp; + + greg = struct_return ? 4 : 3; + freg = 1; + vreg = 2; + argstkspace = 0; + structstkspace = 0; + + /* Figure out how much new stack space is required for arguments + which don't fit in registers. Unlike the PowerOpen ABI, the + SysV ABI doesn't reserve any extra space for parameters which + are put in registers. */ + for (argno = 0; argno < nargs; argno++) + { + arg = args[argno]; + type = check_typedef (VALUE_TYPE (arg)); + len = TYPE_LENGTH (type); + + if (TYPE_CODE (type) == TYPE_CODE_FLT) + { + if (freg <= 8) + freg++; + else + { + /* SysV ABI converts floats to doubles when placed in + memory and requires 8 byte alignment */ + if (argstkspace & 0x4) + argstkspace += 4; + argstkspace += 8; + } + } + else if (TYPE_CODE (type) == TYPE_CODE_INT && len == 8) /* long long */ + { + if (greg > 9) + { + greg = 11; + if (argstkspace & 0x4) + argstkspace += 4; + argstkspace += 8; + } + else + { + if ((greg & 1) == 0) + greg++; + greg += 2; + } + } + else if (!TYPE_VECTOR (type)) + { + if (len > 4 + || TYPE_CODE (type) == TYPE_CODE_STRUCT + || TYPE_CODE (type) == TYPE_CODE_UNION) + { + /* Rounding to the nearest multiple of 8 may not be necessary, + but it is safe. Particularly since we don't know the + field types of the structure */ + structstkspace += round2 (len, 8); + } + if (greg <= 10) + greg++; + else + argstkspace += 4; + } + else + { + if (len == 16 + && TYPE_CODE (type) == TYPE_CODE_ARRAY + && TYPE_VECTOR (type)) + { + if (vreg <= 13) + vreg++; + else + { + /* Vector arguments must be aligned to 16 bytes on + the stack. */ + argstkspace += round2 (argstkspace, 16); + argstkspace += 16; + } + } + } + } + + /* Get current SP location */ + saved_sp = read_sp (); + + sp -= argstkspace + structstkspace; + + /* Allocate space for backchain and callee's saved lr */ + sp -= 8; + + /* Make sure that we maintain 16 byte alignment */ + sp &= ~0x0f; + + /* Update %sp before proceeding any further */ + write_register (SP_REGNUM, sp); + + /* write the backchain */ + store_address (old_sp_buf, 4, saved_sp); + write_memory (sp, old_sp_buf, 4); + + argoffset = 8; + structoffset = argoffset + argstkspace; + freg = 1; + greg = 3; + vreg = 2; + /* Fill in r3 with the return structure, if any */ + if (struct_return) + { + char val_buf[4]; + store_address (val_buf, 4, struct_addr); + memcpy (®isters[REGISTER_BYTE (greg)], val_buf, 4); + greg++; + } + /* Now fill in the registers and stack... */ + for (argno = 0; argno < nargs; argno++) + { + arg = args[argno]; + type = check_typedef (VALUE_TYPE (arg)); + len = TYPE_LENGTH (type); + + if (TYPE_CODE (type) == TYPE_CODE_FLT) + { + if (freg <= 8) + { + if (len > 8) + printf_unfiltered ( + "Fatal Error: a floating point parameter #%d with a size > 8 is found!\n", argno); + memcpy (®isters[REGISTER_BYTE (FP0_REGNUM + freg)], + VALUE_CONTENTS (arg), len); + freg++; + } + else + { + /* SysV ABI converts floats to doubles when placed in + memory and requires 8 byte alignment */ + /* FIXME: Convert floats to doubles */ + if (argoffset & 0x4) + argoffset += 4; + write_memory (sp + argoffset, (char *) VALUE_CONTENTS (arg), len); + argoffset += 8; + } + } + else if (TYPE_CODE (type) == TYPE_CODE_INT && len == 8) /* long long */ + { + if (greg > 9) + { + greg = 11; + if (argoffset & 0x4) + argoffset += 4; + write_memory (sp + argoffset, (char *) VALUE_CONTENTS (arg), len); + argoffset += 8; + } + else + { + if ((greg & 1) == 0) + greg++; + + memcpy (®isters[REGISTER_BYTE (greg)], + VALUE_CONTENTS (arg), 4); + memcpy (®isters[REGISTER_BYTE (greg + 1)], + VALUE_CONTENTS (arg) + 4, 4); + greg += 2; + } + } + else if (!TYPE_VECTOR (type)) + { + char val_buf[4]; + if (len > 4 + || TYPE_CODE (type) == TYPE_CODE_STRUCT + || TYPE_CODE (type) == TYPE_CODE_UNION) + { + write_memory (sp + structoffset, VALUE_CONTENTS (arg), len); + store_address (val_buf, 4, sp + structoffset); + structoffset += round2 (len, 8); + } + else + { + memset (val_buf, 0, 4); + memcpy (val_buf, VALUE_CONTENTS (arg), len); + } + if (greg <= 10) + { + memcpy (®isters[REGISTER_BYTE (greg)], val_buf, 4); + greg++; + } + else + { + write_memory (sp + argoffset, val_buf, 4); + argoffset += 4; + } + } + else + { + if (len == 16 + && TYPE_CODE (type) == TYPE_CODE_ARRAY + && TYPE_VECTOR (type)) + { + struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch); + char *v_val_buf = alloca (16); + memset (v_val_buf, 0, 16); + memcpy (v_val_buf, VALUE_CONTENTS (arg), len); + if (vreg <= 13) + { + memcpy (®isters[REGISTER_BYTE (tdep->ppc_vr0_regnum + + vreg)], + v_val_buf, 16); + vreg++; + } + else + { + write_memory (sp + argoffset, v_val_buf, 16); + argoffset += 16; + } + } + } + } + + target_store_registers (-1); + return sp; +} + +/* Until November 2001, gcc was not complying to the SYSV ABI for + returning structures less than or equal to 8 bytes in size. It was + returning everything in memory. When this was corrected, it wasn't + fixed for native platforms. */ +int +ppc_sysv_abi_broken_use_struct_convention (int gcc_p, struct type *value_type) +{ + if (TYPE_LENGTH (value_type) == 16 + && TYPE_VECTOR (value_type)) + return 0; + + return generic_use_struct_convention (gcc_p, value_type); +} + +/* Structures 8 bytes or less long are returned in the r3 & r4 + registers, according to the SYSV ABI. */ +int +ppc_sysv_abi_use_struct_convention (int gcc_p, struct type *value_type) +{ + if (TYPE_LENGTH (value_type) == 16 + && TYPE_VECTOR (value_type)) + return 0; + + return (TYPE_LENGTH (value_type) > 8); +} Index: ppc-tdep.h =================================================================== RCS file: /cvs/src/src/gdb/ppc-tdep.h,v retrieving revision 1.10 diff -u -r1.10 ppc-tdep.h --- ppc-tdep.h 29 Apr 2002 15:10:06 -0000 1.10 +++ ppc-tdep.h 29 May 2002 00:07:49 -0000 @@ -22,6 +22,8 @@ #ifndef PPC_TDEP_H #define PPC_TDEP_H +#include "osabi.h" + struct frame_info; struct value; @@ -52,7 +54,7 @@ struct gdbarch_tdep { int wordsize; /* size in bytes of fixed-point word */ - int osabi; /* OS / ABI from ELF header */ + enum gdb_osabi osabi; /* OS / ABI from ELF header */ int *regoff; /* byte offsets in register arrays */ const struct reg *regs; /* from current variant */ int ppc_gp0_regnum; /* GPR register 0 */ Index: ppcnbsd-nat.c =================================================================== RCS file: /cvs/src/src/gdb/ppcnbsd-nat.c,v retrieving revision 1.9 diff -u -r1.9 ppcnbsd-nat.c --- ppcnbsd-nat.c 28 May 2002 18:43:07 -0000 1.9 +++ ppcnbsd-nat.c 29 May 2002 00:07:49 -0000 @@ -25,8 +25,6 @@ #include "defs.h" #include "inferior.h" -#include "gdbcore.h" -#include "regcache.h" #include "ppc-tdep.h" #include "ppcnbsd-tdep.h" Index: rs6000-tdep.c =================================================================== RCS file: /cvs/src/src/gdb/rs6000-tdep.c,v retrieving revision 1.65 diff -u -r1.65 rs6000-tdep.c --- rs6000-tdep.c 20 May 2002 17:45:43 -0000 1.65 +++ rs6000-tdep.c 29 May 2002 00:07:55 -0000 @@ -2402,88 +2402,7 @@ return NULL; } - - - -\f -static void -process_note_abi_tag_sections (bfd *abfd, asection *sect, void *obj) -{ - int *os_ident_ptr = obj; - const char *name; - unsigned int sectsize; - - name = bfd_get_section_name (abfd, sect); - sectsize = bfd_section_size (abfd, sect); - if (strcmp (name, ".note.ABI-tag") == 0 && sectsize > 0) - { - unsigned int name_length, data_length, note_type; - char *note = alloca (sectsize); - - bfd_get_section_contents (abfd, sect, note, - (file_ptr) 0, (bfd_size_type) sectsize); - - name_length = bfd_h_get_32 (abfd, note); - data_length = bfd_h_get_32 (abfd, note + 4); - note_type = bfd_h_get_32 (abfd, note + 8); - - if (name_length == 4 && data_length == 16 && note_type == 1 - && strcmp (note + 12, "GNU") == 0) - { - int os_number = bfd_h_get_32 (abfd, note + 16); - - /* The case numbers are from abi-tags in glibc */ - switch (os_number) - { - case 0 : - *os_ident_ptr = ELFOSABI_LINUX; - break; - case 1 : - *os_ident_ptr = ELFOSABI_HURD; - break; - case 2 : - *os_ident_ptr = ELFOSABI_SOLARIS; - break; - default : - internal_error (__FILE__, __LINE__, - "process_note_abi_sections: unknown OS number %d", - os_number); - break; - } - } - } -} - -/* Return one of the ELFOSABI_ constants for BFDs representing ELF - executables. If it's not an ELF executable or if the OS/ABI couldn't - be determined, simply return -1. */ - -static int -get_elfosabi (bfd *abfd) -{ - int elfosabi = -1; - - if (abfd != NULL && bfd_get_flavour (abfd) == bfd_target_elf_flavour) - { - elfosabi = elf_elfheader (abfd)->e_ident[EI_OSABI]; - - /* When elfosabi is 0 (ELFOSABI_NONE), this is supposed to indicate - that we're on a SYSV system. However, GNU/Linux uses a note section - to record OS/ABI info, but leaves e_ident[EI_OSABI] zero. So we - have to check the note sections too. */ - if (elfosabi == 0) - { - bfd_map_over_sections (abfd, - process_note_abi_tag_sections, - &elfosabi); - } - } - - return elfosabi; -} - \f - /* Initialize the current architecture based on INFO. If possible, re-use an architecture from ARCHES, which is a list of architectures already created during this debugging session. @@ -2502,7 +2421,8 @@ enum bfd_architecture arch; unsigned long mach; bfd abfd; - int osabi, sysv_abi; + int sysv_abi; + enum gdb_osabi osabi = GDB_OSABI_UNKNOWN; gdbarch_print_insn_ftype *print_insn; from_xcoff_exec = info.abfd && info.abfd->format == bfd_object && @@ -2513,7 +2433,8 @@ sysv_abi = info.abfd && bfd_get_flavour (info.abfd) == bfd_target_elf_flavour; - osabi = get_elfosabi (info.abfd); + if (info.abfd) + osabi = gdbarch_lookup_osabi (info.abfd); /* Check word size. If INFO is from a binary file, infer it from that, else choose a likely default. */ @@ -2726,57 +2647,23 @@ /* Not sure on this. FIXMEmgo */ set_gdbarch_frame_args_skip (gdbarch, 8); - /* Until November 2001, gcc was not complying to the SYSV ABI for - returning structures less than or equal to 8 bytes in size. It was - returning everything in memory. When this was corrected, it wasn't - fixed for native platforms. */ if (sysv_abi) - { - if (osabi == ELFOSABI_LINUX - || osabi == ELFOSABI_NETBSD - || osabi == ELFOSABI_FREEBSD) - set_gdbarch_use_struct_convention (gdbarch, - ppc_sysv_abi_broken_use_struct_convention); - else - set_gdbarch_use_struct_convention (gdbarch, - ppc_sysv_abi_use_struct_convention); - } + set_gdbarch_use_struct_convention (gdbarch, + ppc_sysv_abi_use_struct_convention); else - { - set_gdbarch_use_struct_convention (gdbarch, - generic_use_struct_convention); - } + set_gdbarch_use_struct_convention (gdbarch, + generic_use_struct_convention); set_gdbarch_frame_chain_valid (gdbarch, file_frame_chain_valid); - /* Note: kevinb/2002-04-12: See note above regarding *_push_arguments(). - The same remarks hold for the methods below. */ - if (osabi == ELFOSABI_LINUX && wordsize == 4) - { - set_gdbarch_frameless_function_invocation (gdbarch, - ppc_linux_frameless_function_invocation); - set_gdbarch_frame_chain (gdbarch, ppc_linux_frame_chain); - set_gdbarch_frame_saved_pc (gdbarch, ppc_linux_frame_saved_pc); - - set_gdbarch_frame_init_saved_regs (gdbarch, - ppc_linux_frame_init_saved_regs); - set_gdbarch_init_extra_frame_info (gdbarch, - ppc_linux_init_extra_frame_info); - - set_gdbarch_memory_remove_breakpoint (gdbarch, - ppc_linux_memory_remove_breakpoint); - set_solib_svr4_fetch_link_map_offsets - (gdbarch, ppc_linux_svr4_fetch_link_map_offsets); - } - else - { - set_gdbarch_frameless_function_invocation (gdbarch, - rs6000_frameless_function_invocation); - set_gdbarch_frame_chain (gdbarch, rs6000_frame_chain); - set_gdbarch_frame_saved_pc (gdbarch, rs6000_frame_saved_pc); - set_gdbarch_frame_init_saved_regs (gdbarch, rs6000_frame_init_saved_regs); - set_gdbarch_init_extra_frame_info (gdbarch, rs6000_init_extra_frame_info); - } + set_gdbarch_frameless_function_invocation (gdbarch, + rs6000_frameless_function_invocation); + set_gdbarch_frame_chain (gdbarch, rs6000_frame_chain); + set_gdbarch_frame_saved_pc (gdbarch, rs6000_frame_saved_pc); + + set_gdbarch_frame_init_saved_regs (gdbarch, rs6000_frame_init_saved_regs); + set_gdbarch_init_extra_frame_info (gdbarch, rs6000_init_extra_frame_info); + if (!sysv_abi) { /* Handle RS/6000 function pointers (which are really function @@ -2792,9 +2679,24 @@ now that the C compiler delays popping them. */ set_gdbarch_frame_num_args (gdbarch, frame_num_args_unknown); + /* Hook in ABI-specific overrides, if they have been registered. */ + gdbarch_init_osabi (info, gdbarch, osabi); + return gdbarch; } +static void +rs6000_dump_tdep (struct gdbarch *current_gdbarch, struct ui_file *file) +{ + struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch); + + if (tdep == NULL) + return; + + fprintf_unfiltered (file, "rs6000_dump_tdep: OS ABI = %s\n", + gdbarch_osabi_name (tdep->osabi)); +} + static struct cmd_list_element *info_powerpc_cmdlist = NULL; static void @@ -2808,8 +2710,8 @@ void _initialize_rs6000_tdep (void) { - register_gdbarch_init (bfd_arch_rs6000, rs6000_gdbarch_init); - register_gdbarch_init (bfd_arch_powerpc, rs6000_gdbarch_init); + gdbarch_register (bfd_arch_rs6000, rs6000_gdbarch_init, rs6000_dump_tdep); + gdbarch_register (bfd_arch_powerpc, rs6000_gdbarch_init, rs6000_dump_tdep); /* Add root prefix command for "info powerpc" commands */ add_prefix_cmd ("powerpc", class_info, rs6000_info_powerpc_command, @@ -2819,5 +2721,4 @@ add_cmd ("altivec", class_info, rs6000_altivec_registers_info, "Display the contents of the AltiVec registers.", &info_powerpc_cmdlist); - } Index: config/powerpc/aix.mt =================================================================== RCS file: /cvs/src/src/gdb/config/powerpc/aix.mt,v retrieving revision 1.5 diff -u -r1.5 aix.mt --- config/powerpc/aix.mt 24 Dec 2001 20:07:02 -0000 1.5 +++ config/powerpc/aix.mt 29 May 2002 00:07:57 -0000 @@ -1,3 +1,3 @@ # Target: PowerPC running AIX -TDEPFILES= rs6000-tdep.o xcoffread.o ppc-linux-tdep.o solib.o solib-svr4.o +TDEPFILES= rs6000-tdep.o xcoffread.o ppc-sysv-tdep.o solib.o solib-svr4.o TM_FILE= tm-ppc-aix.h Index: config/powerpc/linux.mt =================================================================== RCS file: /cvs/src/src/gdb/config/powerpc/linux.mt,v retrieving revision 1.4 diff -u -r1.4 linux.mt --- config/powerpc/linux.mt 20 Oct 2001 00:16:44 -0000 1.4 +++ config/powerpc/linux.mt 29 May 2002 00:07:57 -0000 @@ -1,5 +1,6 @@ # Target: Motorola PPC on Linux -TDEPFILES= rs6000-tdep.o ppc-linux-tdep.o solib.o solib-svr4.o solib-legacy.o +TDEPFILES= rs6000-tdep.o ppc-linux-tdep.o ppc-sysv-tdep.o solib.o \ + solib-svr4.o solib-legacy.o TM_FILE= tm-linux.h SIM_OBS = remote-sim.o Index: config/powerpc/nbsd.mh =================================================================== RCS file: /cvs/src/src/gdb/config/powerpc/nbsd.mh,v retrieving revision 1.10 diff -u -r1.10 nbsd.mh --- config/powerpc/nbsd.mh 28 May 2002 18:43:08 -0000 1.10 +++ config/powerpc/nbsd.mh 29 May 2002 00:07:57 -0000 @@ -1,3 +1,3 @@ # Host: PowerPC, running NetBSD -NATDEPFILES= fork-child.o infptrace.o inftarg.o ppcnbsd-nat.o solib-legacy.o +NATDEPFILES= fork-child.o infptrace.o inftarg.o ppcnbsd-nat.o NAT_FILE= nm-nbsd.h Index: config/powerpc/nbsd.mt =================================================================== RCS file: /cvs/src/src/gdb/config/powerpc/nbsd.mt,v retrieving revision 1.10 diff -u -r1.10 nbsd.mt --- config/powerpc/nbsd.mt 28 May 2002 18:43:08 -0000 1.10 +++ config/powerpc/nbsd.mt 29 May 2002 00:07:57 -0000 @@ -1,5 +1,5 @@ # Target: PowerPC, running NetBSD -TDEPFILES= rs6000-tdep.o ppcnbsd-tdep.o nbsd-tdep.o ppc-linux-tdep.o corelow.o \ +TDEPFILES= rs6000-tdep.o ppc-sysv-tdep.o ppcnbsd-tdep.o nbsd-tdep.o corelow.o \ solib.o solib-svr4.o TM_FILE= tm-nbsd.h Index: config/powerpc/ppc-eabi.mt =================================================================== RCS file: /cvs/src/src/gdb/config/powerpc/ppc-eabi.mt,v retrieving revision 1.4 diff -u -r1.4 ppc-eabi.mt --- config/powerpc/ppc-eabi.mt 1 Dec 2001 00:23:38 -0000 1.4 +++ config/powerpc/ppc-eabi.mt 29 May 2002 00:07:57 -0000 @@ -1,3 +1,3 @@ # Target: PowerPC running eabi -TDEPFILES= rs6000-tdep.o monitor.o dsrec.o ppcbug-rom.o dink32-rom.o ppc-bdm.o ocd.o remote-sds.o ppc-linux-tdep.o solib.o solib-svr4.o +TDEPFILES= rs6000-tdep.o monitor.o dsrec.o ppcbug-rom.o dink32-rom.o ppc-bdm.o ocd.o remote-sds.o ppc-sysv-tdep.o solib.o solib-svr4.o TM_FILE= tm-ppc-eabi.h Index: config/powerpc/ppc-sim.mt =================================================================== RCS file: /cvs/src/src/gdb/config/powerpc/ppc-sim.mt,v retrieving revision 1.4 diff -u -r1.4 ppc-sim.mt --- config/powerpc/ppc-sim.mt 1 Dec 2001 00:23:38 -0000 1.4 +++ config/powerpc/ppc-sim.mt 29 May 2002 00:07:57 -0000 @@ -1,5 +1,5 @@ # Target: PowerPC running eabi and including the simulator -TDEPFILES= rs6000-tdep.o monitor.o dsrec.o ppcbug-rom.o dink32-rom.o ppc-bdm.o ocd.o remote-sds.o ppc-linux-tdep.o solib.o solib-svr4.o +TDEPFILES= rs6000-tdep.o monitor.o dsrec.o ppcbug-rom.o dink32-rom.o ppc-bdm.o ocd.o remote-sds.o ppc-sysv-tdep.o solib.o solib-svr4.o TM_FILE= tm-ppc-eabi.h SIM_OBS = remote-sim.o Index: config/powerpc/ppcle-eabi.mt =================================================================== RCS file: /cvs/src/src/gdb/config/powerpc/ppcle-eabi.mt,v retrieving revision 1.4 diff -u -r1.4 ppcle-eabi.mt --- config/powerpc/ppcle-eabi.mt 1 Dec 2001 00:23:38 -0000 1.4 +++ config/powerpc/ppcle-eabi.mt 29 May 2002 00:07:57 -0000 @@ -1,3 +1,3 @@ # Target: PowerPC running eabi in little endian mode -TDEPFILES= rs6000-tdep.o monitor.o dsrec.o ppcbug-rom.o ppc-bdm.o ocd.o ppc-linux-tdep.o solib.o solib-svr4.o +TDEPFILES= rs6000-tdep.o monitor.o dsrec.o ppcbug-rom.o ppc-bdm.o ocd.o ppc-sysv-tdep.o solib.o solib-svr4.o TM_FILE= tm-ppcle-eabi.h Index: config/powerpc/ppcle-sim.mt =================================================================== RCS file: /cvs/src/src/gdb/config/powerpc/ppcle-sim.mt,v retrieving revision 1.4 diff -u -r1.4 ppcle-sim.mt --- config/powerpc/ppcle-sim.mt 1 Dec 2001 00:23:38 -0000 1.4 +++ config/powerpc/ppcle-sim.mt 29 May 2002 00:07:57 -0000 @@ -1,5 +1,5 @@ # Target: PowerPC running eabi in little endian mode under the simulator -TDEPFILES= rs6000-tdep.o monitor.o dsrec.o ppcbug-rom.o ppc-bdm.o ocd.o ppc-linux-tdep.o solib.o solib-svr4.o +TDEPFILES= rs6000-tdep.o monitor.o dsrec.o ppcbug-rom.o ppc-bdm.o ocd.o ppc-sysv-tdep.o solib.o solib-svr4.o TM_FILE= tm-ppcle-eabi.h SIM_OBS = remote-sim.o Index: config/powerpc/vxworks.mt =================================================================== RCS file: /cvs/src/src/gdb/config/powerpc/vxworks.mt,v retrieving revision 1.3 diff -u -r1.3 vxworks.mt --- config/powerpc/vxworks.mt 1 Dec 2001 00:23:38 -0000 1.3 +++ config/powerpc/vxworks.mt 29 May 2002 00:07:57 -0000 @@ -1,3 +1,3 @@ # Target: Powerpc running VxWorks -TDEPFILES= rs6000-tdep.o ppc-linux-tdep.o solib.o solib-svr4.o +TDEPFILES= rs6000-tdep.o ppc-sysv-tdep.o solib.o solib-svr4.o TM_FILE= tm-vxworks.h ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH/RFA] Generic OS ABI handling for PowerPC 2002-05-28 17:59 ` Jason R Thorpe @ 2002-05-29 17:37 ` Kevin Buettner 2002-05-29 19:08 ` Jason R Thorpe 0 siblings, 1 reply; 11+ messages in thread From: Kevin Buettner @ 2002-05-29 17:37 UTC (permalink / raw) To: Jason R Thorpe; +Cc: gdb-patches On May 28, 5:26pm, Jason R Thorpe wrote: > Ok, new patch. Moves the ppc_sysv stuff into new ppc-sysv-tdep.c, > and updates the various .mt files. I also touched up some dependency > lists in Makefile.in. > > OK to commit? Sure, go ahead. The only nit I have is with parts of your ChangeLog entry. Entries like this one... > * ppc-linux-tdep.c (_initialize_ppc_linux_tdep, > ppc_linux_init_abi): New functions. > (ppc_sysv_abi_broken_use_struct_convention, > ppc_sysv_abi_use_struct_convention, > ppc_sysv_abi_push_arguments): Move to... ...should be formatted as follows: * ppc-linux-tdep.c (_initialize_ppc_linux_tdep) (ppc_linux_init_abi): New functions. (ppc_sysv_abi_broken_use_struct_convention) (ppc_sysv_abi_use_struct_convention) (ppc_sysv_abi_push_arguments): Move to... (Actually, I prefer the way you did it, but I've been told that it's incorrect.) Kevin ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH/RFA] Generic OS ABI handling for PowerPC 2002-05-29 17:37 ` Kevin Buettner @ 2002-05-29 19:08 ` Jason R Thorpe 0 siblings, 0 replies; 11+ messages in thread From: Jason R Thorpe @ 2002-05-29 19:08 UTC (permalink / raw) To: Kevin Buettner; +Cc: gdb-patches On Wed, May 29, 2002 at 05:23:23PM -0700, Kevin Buettner wrote: > Sure, go ahead. Ok, done, thanks. > (Actually, I prefer the way you did it, but I've been told that it's > incorrect.) Noted. (Sigh, it's so inconsistent all over the place, whee :-) -- -- Jason R. Thorpe <thorpej@wasabisystems.com> ^ permalink raw reply [flat|nested] 11+ messages in thread
end of thread, other threads:[~2002-05-30 1:21 UTC | newest] Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2002-05-28 12:41 [PATCH/RFA] Generic OS ABI handling for PowerPC Jason R Thorpe 2002-05-28 12:52 ` Kevin Buettner 2002-05-28 13:46 ` Jason R Thorpe 2002-05-28 14:12 ` Kevin Buettner 2002-05-28 16:35 ` Jason R Thorpe 2002-05-28 16:46 ` Kevin Buettner 2002-05-28 16:47 ` Jason R Thorpe 2002-05-28 19:42 ` Andrew Cagney 2002-05-28 17:59 ` Jason R Thorpe 2002-05-29 17:37 ` Kevin Buettner 2002-05-29 19:08 ` Jason R Thorpe
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox