Andrew Haley found a bug on GDB running on ARM when using
--enable-64-bit-bfd. Basically the issue happens when dealing with "bl"
instructions: GDB does branch destination calculation and (wrongly)
sign-extends the PC. Here is a piece of his original message explaining
the problem:
> next_pc = arm_get_next_pc (frame, get_frame_pc (frame));
>
> /* The Linux kernel offers some user-mode helpers in a high page. We can
> not read this page (as of 2.6.23), and even if we could then we couldn't
> set breakpoints in it, and even if we could then the atomic operations
> would fail when interrupted. They are all called as functions and return
> to the address in LR, so step to there instead. */
> if (next_pc > 0xffff0000)
> next_pc = get_frame_register_unsigned (frame, ARM_LR_REGNUM);
>
> arm_insert_single_step_breakpoint (gdbarch, aspace, next_pc);
>
> Unfortunately, branch destination addresses are SIGN EXTENDED to 64
> bits. So,
>
> (top-gdb) p/x next_pc
> $14 = 0xffffffffb6df2864
>
> Which triggers the next_pc = get_frame_register_unsigned(), and we
> cannot step into any branches because the destination PC is wrong.
Anyway, the fix is simple and Andrew himself provided it for us. It
took a while for me to figure out how to trigger the bug (in order to
write a testcase for it), but I finally made it.
The attached patch fixes the problem (by casting to `unsigned long'
instead of just `long'), and also includes a testcase to reproduce the
issue.
gdb/ChangeLog:
2013-04-22 Andrew Haley <aph@redhat.com>
* arm-tdep.c (BranchDest): Cast result as "unsigned long",
instead of "long".
Yao Qi [Sat, 20 Apr 2013 07:38:39 +0000 (07:38 +0000)]
gdb/
* ctf.c (ctf_fetch_registers): Change the type of 'regs' from
'char *' to 'gdb_byte *'. Cast the return value of
'bt_ctf_get_char_array' to 'gdb_byte *'.
Pedro Alves [Fri, 19 Apr 2013 17:48:10 +0000 (17:48 +0000)]
Make GDB -Wpointer-sign clean on MinGW too.
This is the remaining issue impeding GDB to build with "-Wpointer-sign
-Werror" with Fedora 17's i686-w64-mingw32 cross toolchain.
../../src/gdb/ser-tcp.c: In function 'net_read_prim':
../../src/gdb/ser-tcp.c:341:3: error: pointer targets in passing argument 2 of 'recv' differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/serial.h:23:0,
from ../../src/gdb/ser-tcp.c:21:
/usr/i686-w64-mingw32/sys-root/mingw/include/winsock2.h:983:34: note: expected 'char *' but argument is of type 'unsigned char *'
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* ser-tcp.c (net_read_prim): Cast second argument to recv to
'void *'.
Pedro Alves [Fri, 19 Apr 2013 17:44:18 +0000 (17:44 +0000)]
-Wpointer-sign: monitor.c.
This fixes -Wpointer-sign issues in monitor.c.
Tested by building on x86_64 Fedora 17 w/ --enable-targets=all.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* monitor.c (monitor_write_memory, monitor_write_memory_bytes):
Change type of 'myaddr' parameter to gdb_byte pointer.
(monitor_write_memory_longlongs): Likewise. Cast 'myaddr' pointer
to 'long long' pointer instead of to 'unsigned long long'.
(monitor_write_memory_block, monitor_read_memory_single)
(monitor_read_memory): Change type of 'myaddr' parameter to
gdb_byte pointer.
Pedro Alves [Fri, 19 Apr 2013 15:37:31 +0000 (15:37 +0000)]
-Wpointer-sign: record.c.
../../src/gdb/record.c: In function ‘set_record_insn_history_size’:
../../src/gdb/record.c:670:5: error: pointer targets in passing argument 2 of ‘validate_history_size’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/record.c:646:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
../../src/gdb/record.c: In function ‘set_record_call_history_size’:
../../src/gdb/record.c:682:5: error: pointer targets in passing argument 2 of ‘validate_history_size’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/record.c:646:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
This fixes it in the obvious way.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* record.c (validate_history_size): Make parameter 'setting'
unsigned.
Pedro Alves [Fri, 19 Apr 2013 15:36:52 +0000 (15:36 +0000)]
-Wpointer-sign: ctf.c.
ctf_save_write's second parameter is gdb_byte *, and all these
arguments are 'char *'. Since this function is ultimately just
writing host bytes to a local file with fwrite, an alternative would
be to change ctf_save_write to take a 'void *' instead of 'gdb_byte
*', thus removing the need for any cast (we have more calls with casts
than without).
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* ctf.c (ctf_write_uploaded_tsv, ctf_write_uploaded_tp): Add casts
to 'gdb_byte *'.
Pedro Alves [Fri, 19 Apr 2013 15:36:24 +0000 (15:36 +0000)]
-Wpointer-sign: cp-valprint.c.
Fix:
../../src/gdb/cp-valprint.c: In function ‘cp_print_class_member’:
../../src/gdb/cp-valprint.c:793:3: error: pointer targets in passing argument 2 of ‘cp_find_class_member’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/cp-valprint.c:721:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
'fieldno' is used throughout as 'int', so just follow the trend.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* cp-valprint.c (cp_print_class_member): Change type of 'fieldno'
local to int.
Pedro Alves [Fri, 19 Apr 2013 15:35:21 +0000 (15:35 +0000)]
-Wpointer-sign: dwarf2-frame.c: Pass unsigned variable to safe_read_uleb128.
The 'bytes_read' change should be obvious. As for the other hunk,
we're passing the address of the signed 'offset' to safe_read_uleb128,
which expects unsigned. Fix it by passing the address of the unsigned
'utmp' instead, like already done on other spots in the file.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* dwarf2-frame.c (execute_cfa_program): Make 'bytes_read' local
unsigned. Pass 'tmp' to safe_read_uleb128 instead of the signed
'offset', and adjust.
Pedro Alves [Fri, 19 Apr 2013 15:34:46 +0000 (15:34 +0000)]
-Wpointer-sign: dwarf2read.c.
This fixes the remaining issues necessary to make the DWARF reader
-Wpointer-sign clean. The 'filename' bit should be obvious.
'constant_pool' holds the contents of an obstack, which are 'char *'.
gdb_byte would work too, but it'd need more casts elsewhere, so I just
chose the minimal approach. Any way would be fine with me.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* dwarf2read.c (dwarf2_get_dwz_file): Add cast to const char *.
(read_index_from_section): Add cast to 'char *'.
Pedro Alves [Fri, 19 Apr 2013 15:33:58 +0000 (15:33 +0000)]
-Wpointer-sign: xcoffread.c.
../../src/gdb/xcoffread.c: In function ‘xcoff_initial_scan’:
../../src/gdb/xcoffread.c:2982:17: error: pointer targets in assignment differ in signedness [-Werror=pointer-sign]
'debugsec' is a 'bfd_byte *', holding the result of a
bfd_get_full_section_contents. 'info->debugsec' holds the same
contents throughout the whole xcoff read, and everywhere it's used to
read symbol names. Simply adding a cast feels appropriate.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* xcoffread.c (xcoff_initial_scan): Add cast to 'char *'.
Pedro Alves [Fri, 19 Apr 2013 15:33:29 +0000 (15:33 +0000)]
-Wpointer-sign: coff-pe-read.c: treat strings in PE/COFF data as char *.
A couple places take a pointer to the middle of some raw section
buffer and treat them as strings. Add casts to char * as appropriate,
fixing -Wpointer-sign warnings.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* coff-pe-read.c (read_pe_exported_syms): Handle strings as char.
Pedro Alves [Fri, 19 Apr 2013 15:32:56 +0000 (15:32 +0000)]
-Wpointer-sign: bookmarks.
Bookmarks are opaque to the core code -- by design, the target is free
to use any sort of blob as bookmark identifier.
The record target chooses to use strings for bookmarks. This adds
casts following that direction, fixing -Wpointer-sign warnings.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* record-full.c (record_full_get_bookmark): Change local 'ret'
type to char * and add cast to gdb_byte *.
(record_full_goto_bookmark): Handle 'bookmark' argument as a
string.
* reverse.c (goto_bookmark_command): Add casts to gdb_byte *.
Pedro Alves [Fri, 19 Apr 2013 15:29:09 +0000 (15:29 +0000)]
-Wpointer-sign: python/.
This fixes -Wpointer-sign warnings in the python/ code in the manner
that seems most appropriate to me.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* python/py-inferior.c (infpy_write_memory): Add cast to gdb_byte
* python/py-prettyprint.c (print_string_repr): Change type of
'output' local to char *. Add cast to gdb_byte * in
LA_PRINT_STRING call.
(print_children): Change type of 'output' local to char *.
* python/py-value.c (valpy_string): Add cast to const char * in
PyUnicode_Decode call.
Pedro Alves [Fri, 19 Apr 2013 15:28:42 +0000 (15:28 +0000)]
-Wpointer-sign: remote-mips.c.
remote-mips.c has a bunch of -Wpointer-sign warnings:
../../src/gdb/remote-mips.c: In function ‘mips_receive_packet’:
../../src/gdb/remote-mips.c:1128:7: error: pointer targets in passing argument 2 of ‘mips_cksum’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/remote-mips.c:830:1: note: expected ‘const unsigned char *’ but argument is of type ‘char *’
../../src/gdb/remote-mips.c:1135:7: error: pointer targets in passing argument 2 of ‘mips_cksum’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/remote-mips.c:830:1: note: expected ‘const unsigned char *’ but argument is of type ‘char *’
../../src/gdb/remote-mips.c: In function ‘mips_load_srec’:
../../src/gdb/remote-mips.c:2830:12: error: pointer targets in passing argument 4 of ‘mips_make_srec’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/remote-mips.c:134:12: note: expected ‘unsigned char *’ but argument is of type ‘char *’
../../src/gdb/remote-mips.c: In function ‘pmon_zeroset’:
../../src/gdb/remote-mips.c:3030:3: error: pointer targets in passing argument 4 of ‘pmon_makeb64’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/remote-mips.c:2977:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
../../src/gdb/remote-mips.c: In function ‘pmon_make_fastrec’:
../../src/gdb/remote-mips.c:3124:6: error: pointer targets in passing argument 3 of ‘pmon_zeroset’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/remote-mips.c:3025:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
../../src/gdb/remote-mips.c:3126:4: error: pointer targets in passing argument 4 of ‘pmon_makeb64’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/remote-mips.c:2977:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
../../src/gdb/remote-mips.c:3148:3: error: pointer targets in passing argument 3 of ‘pmon_zeroset’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/remote-mips.c:3025:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
../../src/gdb/remote-mips.c:3153:3: error: pointer targets in passing argument 3 of ‘pmon_zeroset’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/remote-mips.c:3025:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
../../src/gdb/remote-mips.c:3154:8: error: pointer targets in passing argument 4 of ‘pmon_makeb64’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/remote-mips.c:2977:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
../../src/gdb/remote-mips.c: In function ‘pmon_load_fast’:
../../src/gdb/remote-mips.c:3423:2: error: pointer targets in passing argument 4 of ‘pmon_makeb64’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/remote-mips.c:2977:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
../../src/gdb/remote-mips.c:3457:4: error: pointer targets in passing argument 3 of ‘pmon_checkset’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/remote-mips.c:3051:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
../../src/gdb/remote-mips.c:3484:8: error: pointer targets in passing argument 3 of ‘pmon_zeroset’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/remote-mips.c:3025:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
../../src/gdb/remote-mips.c:3489:3: error: pointer targets in passing argument 3 of ‘pmon_checkset’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/remote-mips.c:3051:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
The mips packet payload is ASCII, so it makes sense for
mips_send_packet and mips_receive_packet to expose 'char *'-based
interfaces, as currently they do. But, mips packets have a binary
header, so if you look at e.g., mips_receive_packet's implementation,
you'll see "unsigned char" buffers in use. I find it the most natural
to make the payload pointer passed to mips_cksum 'char *' too.
The other changes are straightforward adjustments -- a checksum is
naturally unsigned, and there's one point where we're reading a bfd section.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* remote-mips.c (mips_cksum): Rename 'data' parameter to 'datastr'
and change its type to 'const char *'. Adjust.
(mips_send_packet): Add cast to 'char *', and remove cast to
'unsigned char *'.
(mips_receive_packet): Remove cast to 'unsigned char *'.
(mips_load_srec): Use bfd_byte.
(pmon_makeb64, pmon_zeroset): Make 'chksum' parameter unsigned.
(pmon_checkset): Make 'value' parameter unsigned.
Pedro Alves [Fri, 19 Apr 2013 15:27:06 +0000 (15:27 +0000)]
gdb_byte for binary buffer, char for string: remote.c, tracepoint.c.
While the RSP is largely ASCII based (hence the packet buffer type is
char *), at places we pass around 8-bit binary packets in that buffer.
Functions like hex2bin or remote_escape_output conceptually are
handling binary buffers, so I left them as working with gdb_byte, and
added casts where necessary. Whether these are host bytes or target
bytes is blurry at present, so this is largely a matter of taste.
Switching some of these functions to take "char *" or "void *" would
be equally good.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* remote.c (remote_write_bytes_aux, compare_sections_command)
(remote_read_qxfer)
(remote_search_memory, remote_hostio_pwrite, remote_hostio_pread)
(remote_hostio_readlink, remote_bfd_iovec_pread)
(remote_set_trace_notes): Use gdb_byte when RSP buffer is used as
binary buffer, and char when buffer is used as string.
* tracepoint.c (encode_source_string, tfile_write_uploaded_tp)
(trace_save, tfile_open, traceframe_walk_blocks)
(tfile_fetch_registers): Likewise.
Pedro Alves [Fri, 19 Apr 2013 15:26:17 +0000 (15:26 +0000)]
serial_write: change prototype to take a void-pointer buffer.
While remote.c works with "char *" buffers most of the time, other
remote targets have binary-ish-er protocols, and choose to use
"unsigned char" throughout, like e.g., remote-mips.c or
remote-m32r-sdi.c. That results in -Wpointer-sign warnings in those
targets, unless we add casts in calls to serial_write. Since
serial_write is only concerned about sending raw host bytes out, and
serial_ops->write_prim already works with "void *"/"size_t", a similar
interface to the "write" or "send" system calls, I find it natural to
change serial_write's prototype accordingly, avoiding the need for
casts.
Tested on x86_64 Fedora 17, and also by building x86_64-mingw32
and DJGPP/go32 -hosted gdbs.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* ser-base.c (ser_base_write): Change prototype -- take 'void *'
buffer and size_t size. Adjust.
* ser-base.h (ser_base_write): Adjust.
* ser-go32.c (cnts): Change type to size_t.
(dos_write): Change prototype -- take 'void *'
buffer and size_t size. Adjust.
(dos_info): Print elements of 'cnts' as unsigned long.
* serial.c (serial_write): Likewise.
* serial.h (serial_write): Adjust.
(struct serial_ops) <write>: Change prototype -- take 'void *'
buffer and size_t size. Adjust.
Pedro Alves [Fri, 19 Apr 2013 15:24:25 +0000 (15:24 +0000)]
-Wpointer-sign: alpha-tdep.c.
../../src/gdb/alpha-tdep.c: In function ‘alpha_extract_return_value’:
../../src/gdb/alpha-tdep.c:520:4: error: pointer targets in passing argument 3 of ‘regcache_cooked_read_signed’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/alpha-tdep.c:36:0:
We use regcache_cooked_read_unsigned everywhere else too.
2013-04-19 Pedro Alves <palves@redhat.com>
* alpha-tdep.c (alpha_extract_return_value): Use
regcache_cooked_read_unsigned to read 'v0'.
Pedro Alves [Fri, 19 Apr 2013 15:23:52 +0000 (15:23 +0000)]
-Wpointer-sign: xtensa-tdep.c.
../../src/gdb/xtensa-tdep.c:2914:4: error: pointer targets in passing argument 7 of ‘xtensa_operand_get_field’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/xtensa-tdep.c:53:0:
../../src/gdb/../include/xtensa-isa.h:487:1: note: expected ‘unsigned int *’ but argument is of type ‘int *’
../../src/gdb/xtensa-tdep.c:2916:4: error: pointer targets in passing argument 4 of ‘xtensa_operand_decode’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/xtensa-tdep.c:53:0:
../../src/gdb/../include/xtensa-isa.h:507:1: note: expected ‘unsigned int *’ but argument is of type ‘int *’
../../src/gdb/xtensa-tdep.c:2918:4: error: pointer targets in passing argument 7 of ‘xtensa_operand_get_field’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/xtensa-tdep.c:53:0:
../../src/gdb/../include/xtensa-isa.h:487:1: note: expected ‘unsigned int *’ but argument is of type ‘int *’
../../src/gdb/xtensa-tdep.c:2920:4: error: pointer targets in passing argument 4 of ‘xtensa_operand_decode’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/xtensa-tdep.c:53:0:
../../src/gdb/../include/xtensa-isa.h:507:1: note: expected ‘unsigned int *’ but argument is of type ‘int *’
../../src/gdb/xtensa-tdep.c:2922:4: error: pointer targets in passing argument 7 of ‘xtensa_operand_get_field’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/xtensa-tdep.c:53:0:
../../src/gdb/../include/xtensa-isa.h:487:1: note: expected ‘unsigned int *’ but argument is of type ‘int *’
../../src/gdb/xtensa-tdep.c:2924:4: error: pointer targets in passing argument 4 of ‘xtensa_operand_decode’ differ in signedness [-Werror=pointer-sign]
In file included from ../../src/gdb/xtensa-tdep.c:53:0:
../../src/gdb/../include/xtensa-isa.h:507:1: note: expected ‘unsigned int *’ but argument is of type ‘int *’
Those bfd functions that decode instructions output uint32_t values.
Hence this fix:
2013-04-19 Pedro Alves <palves@redhat.com>
* xtensa-tdep.c (execute_l32e, execute_s32e): Change type of
parameters 'at', 'as' and 'offset' to uint32_t.
Pedro Alves [Fri, 19 Apr 2013 15:23:18 +0000 (15:23 +0000)]
-Wpointer-sign: aarch64-tdep.c.
../../src/gdb/aarch64-tdep.c: In function ‘aarch64_analyze_prologue’:
../../src/gdb/aarch64-tdep.c:713:7: error: pointer targets in passing argument 3 of ‘decode_cb’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/aarch64-tdep.c:386:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
../../src/gdb/aarch64-tdep.c:747:7: error: pointer targets in passing argument 3 of ‘decode_stur’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/aarch64-tdep.c:597:1: note: expected ‘int *’ but argument is of type ‘unsigned int *’
'is64' is just used as a boolean; signed/unsigned doesn't really matter.
2013-04-19 Pedro Alves <palves@redhat.com>
* aarch64-tdep.c (aarch64_analyze_prologue): Change type of local
'is64' to signed 'int'.
Pedro Alves [Fri, 19 Apr 2013 15:22:47 +0000 (15:22 +0000)]
-Wpointer-sign: s390-tdep.c.
-Wpointer-sign reveals a bunch of:
../../src/gdb/s390-tdep.c:1342:7: error: pointer targets in passing argument 4 of ‘is_rx’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/s390-tdep.c:1038:1: note: expected ‘unsigned int *’ but argument is of type ‘int *’
../../src/gdb/s390-tdep.c:1343:9: error: pointer targets in passing argument 5 of ‘is_rxy’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/s390-tdep.c:1055:1: note: expected ‘unsigned int *’ but argument is of type ‘int *’
../../src/gdb/s390-tdep.c:1344:9: error: pointer targets in passing argument 5 of ‘is_rxy’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/s390-tdep.c:1055:1: note: expected ‘unsigned int *’ but argument is of type ‘int *’
...
../../src/gdb/s390-tdep.c:1363:7: error: pointer targets in passing argument 5 of ‘is_rs’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/s390-tdep.c:966:1: note: expected ‘unsigned int *’ but argument is of type ‘int *’
../../src/gdb/s390-tdep.c:1364:9: error: pointer targets in passing argument 6 of ‘is_rsy’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/s390-tdep.c:983:1: note: expected ‘unsigned int *’ but argument is of type ‘int *’
../../src/gdb/s390-tdep.c:1365:9: error: pointer targets in passing argument 6 of ‘is_rsy’ differ in signedness [-Werror=pointer-sign]
../../src/gdb/s390-tdep.c:983:1: note: expected ‘unsigned int *’ but argument is of type ‘int *’
...
I don't know much about s390, but from reading the code I believe the
right fix is to treat d2 as signed.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* s390-tdep.c (is_rs, is_rsy, is_rx, is_rxy): Change type of 'd2'
parameter to int *.
Pedro Alves [Fri, 19 Apr 2013 15:22:14 +0000 (15:22 +0000)]
ppc-linux-tdep.c: Wrong signness for buffer holding instructions.
There seems to be no reason for this buffer to be signed. We pass it
around to functions expecting it to be unsigned (which triggers
-Wpointer-sign warnings).
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* ppc-linux-tdep.c (ppc_skip_trampoline_code): Change local
'insnbuf' buffer type to unsigned int[].
Pedro Alves [Fri, 19 Apr 2013 15:21:24 +0000 (15:21 +0000)]
mips-tdep.c: Wrong signness for local holding PC register.
Addresses on MIPS are signed, and we're indeed using
regcache_cooked_read_signed to read the PC, but, we're passing it the
address of an unsigned variable, which triggers a -Wpointer-sign
warning. I've chosen to change the variable's type. I believe this
will end up being the same (though I can't test it).
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* mips-tdep.c (mips_read_pc): Change local 'pc' type to LONGEST.
Pedro Alves [Fri, 19 Apr 2013 15:15:21 +0000 (15:15 +0000)]
alpha-tdep.c/mips-tdep.c: "set heuristic-fence-post" is signed/zinteger.
These commands are currently var_zinteger, hence their control
variable is signed.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* alpha-tdep.c (heuristic_fence_post): Change type to int.
(alpha_heuristic_proc_start): Adjust to check -1 instead of
UINT_MAX.
* mips-tdep.c (heuristic_fence_post): Change type to int.
(heuristic_proc_start): Adjust to check -1 instead of UINT_MAX.
Pedro Alves [Fri, 19 Apr 2013 15:14:23 +0000 (15:14 +0000)]
cris-tdep.c: Use unsigned variable for unsigned command.
"set cris-version" is an unsigned command:
/* CRIS-specific user-commands. */
add_setshow_uinteger_cmd ("cris-version", class_support,
&usr_cmd_cris_version,
_("Set the current CRIS version."),
_("Show the current CRIS version."),
_("\
Set to 10 for CRISv10 or 32 for CRISv32 if autodetection fails.\n\
Defaults to 10. "),
Tested by building on x86_64 Fedora 17.
2013-04-19 Pedro Alves <palves@redhat.com>
* cris-tdep.c (usr_cmd_cris_version): Make unsigned.
(struct gdbarch_tdep) <cris_version>: Make unsigned.
(cris_spec_reg_applicable, cris_gdbarch_init): Adjust locals.
Pedro Alves [Fri, 19 Apr 2013 15:10:53 +0000 (15:10 +0000)]
-Wpointer-sign: gdb_byte -> char.
This is sort of the opposite of the previous patch. Places that
manipulate strings or interfaces that return strings are changed to
use char* instead of gdb_byte*.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* avr-tdep.c (avr_io_reg_read_command): New local 'bufstr'. Use
it to get a string view of the byte buffer.
* i386-cygwin-tdep.c (core_process_module_section): Change local 'buf'
type to gdb_byte *. Adjust.
* linux-tdep.c (linux_info_proc, linux_find_memory_regions_full):
Change local to char *.
* solib-darwin.c (find_program_interpreter): Change return type to
char *. Adjust.
(darwin_solib_get_all_image_info_addr_at_init): Adjust.
* solib-dsbt.c (enable_break2): Change local 'buf' to char *.
* solib-frv.c (enable_break2): Change local 'buf' to char *.
* solib-spu.c (spu_current_sos): Add gdb_byte * cast.
* solib-svr4.c (find_program_interpreter): Change return type to
char *. Adjust.
(enable_break): Change local 'interp_name' to char *.
* spu-multiarch.c (spu_xfer_partial): Add cast to 'char *'.
* spu-tdep.c (spu_pseudo_register_read_spu): Add cast to 'char *'.
(spu_pseudo_register_write_spu): Use char for string buffer.
Adjust.
(info_spu_event_command, info_spu_signal_command): Add casts to
'char *'.
Pedro Alves [Fri, 19 Apr 2013 15:09:46 +0000 (15:09 +0000)]
-Wpointer-sign: char -> gdb_byte.
-Wpointer-sign catches all these cases across the codebase that should
be using gdb_byte for raw target bytes. I think these are all
obvious, hence I've collapsed into a single patch.
gdb/
2013-04-19 Pedro Alves <palves@redhat.com>
* aarch64-tdep.c (aarch64_default_breakpoint): Change type to
gdb_byte[].
(aarch64_breakpoint_from_pc): Change return type to gdb_byte *.
* ada-lang.c (ada_value_assign): Use gdb_byte.
* alphanbsd-tdep.c (sigtramp_retcode): Change type to gdb_byte[].
(alphanbsd_sigtramp_offset): Use gdb_byte.
* arm-linux-tdep.c (arm_linux_arm_le_breakpoint)
(arm_linux_arm_be_breakpoint, eabi_linux_arm_le_breakpoint)
(eabi_linux_arm_be_breakpoint, arm_linux_thumb_be_breakpoint)
(arm_linux_thumb_le_breakpoint, arm_linux_thumb2_be_breakpoint)
(arm_linux_thumb2_le_breakpoint): Change type to gdb_byte[].
* arm-tdep.c (arm_stub_unwind_sniffer)
(arm_displaced_init_closure): Use gdb_byte.
(arm_default_arm_le_breakpoint, arm_default_arm_be_breakpoint)
(arm_default_thumb_le_breakpoint)
(arm_default_thumb_be_breakpoint): Change type to gdb_byte[].
* arm-tdep.h (struct gdbarch_tdep) <arm_breakpoint,
thumb_breakpoint, thumb2_breakpoint>: Change type to gdb_byte *.
* arm-wince-tdep.c (arm_wince_le_breakpoint)
(arm_wince_thumb_le_breakpoint): Change type to gdb_byte[].
* armnbsd-tdep.c (arm_nbsd_arm_le_breakpoint)
(arm_nbsd_arm_be_breakpoint, arm_nbsd_thumb_le_breakpoint)
(arm_nbsd_thumb_be_breakpoint): Change type to gdb_byte[].
* armobsd-tdep.c (arm_obsd_thumb_le_breakpoint)
(arm_obsd_thumb_be_breakpoint): Change type to gdb_byte[].
* cris-tdep.c (push_stack_item, cris_push_dummy_call)
(cris_store_return_value, cris_extract_return_value): Use
gdb_byte.
(constraint): Change type of parameter to char * from signed
char*. Use gdb_byte.
* dwarf2loc.c (read_pieced_value, write_pieced_value): Change type
of local buffer to gdb_byte *.
* dwarf2read.c (read_index_from_section): Use gdb_byte.
(create_dwp_hash_table): Change type of locals to gdb_byte *.
(add_address_entry): Change type of local buffer to gdb_byte[].
* frv-tdep.c (frv_adjust_breakpoint_address, find_func_descr)
(frv_push_dummy_call): Use gdb_byte.
* hppa-hpux-tdep.c (hppa_hpux_push_dummy_code)
(hppa_hpux_supply_ss_fpblock, hppa_hpux_supply_ss_wide)
(hppa_hpux_supply_save_state): Use gdb_byte.
* hppa-tdep.c (hppa32_push_dummy_call)
(hppa64_convert_code_addr_to_fptr): Use gdb_byte.
* ia64-tdep.c (extract_bit_field, replace_bit_field)
(slotN_contents, replace_slotN_contents): Change type of parameter
to gdb_byte *.
(fetch_instruction, ia64_pseudo_register_write)
(ia64_register_to_value, ia64_value_to_register)
(ia64_extract_return_value, ia64_store_return_value)
(ia64_push_dummy_call): Use gdb_byte.
* m32c-tdep.c (m32c_return_value): Remove cast.
* m68hc11-tdep.c (m68hc11_pseudo_register_write)
(m68hc11_push_dummy_call, m68hc11_store_return_value): Use
gdb_byte.
* mipsnbsd-tdep.c (mipsnbsd_get_longjmp_target): Use gdb_byte.
* mn10300-tdep.c (mn10300_store_return_value)
(mn10300_breakpoint_from_pc, mn10300_push_dummy_call): Use
gdb_byte.
* moxie-tdep.c (moxie_process_readu): Use gdb_byte.
(moxie_process_record): Remove casts.
* ppc-ravenscar-thread.c (supply_register_at_address)
(ppc_ravenscar_generic_store_registers): Use gdb_byte.
* ravenscar-thread.c (get_running_thread_id): Use gdb_byte.
* remote-m32r-sdi.c (m32r_fetch_register): Use gdb_byte.
* remote-mips.c (mips_xfer_memory): Use gdb_byte.
* remote.c (compare_sections_command): Use gdb_byte.
* score-tdep.c (score7_free_memblock): Change type of parameter to
gdb_byte *.
* sh-tdep.c (sh_justify_value_in_reg): Change return type to
gdb_byte *. Use gdb_byte.
(sh_push_dummy_call_fpu): Use gdb_byte.
(sh_extract_return_value_nofpu, sh_extract_return_value_fpu)
(sh_store_return_value_nofpu, sh_store_return_value_fpu)
(sh_register_convert_to_virtual, sh_register_convert_to_raw):
Change parameter type to 'gdb_byte *'. Use gdb_byte.
(sh_pseudo_register_read, sh_pseudo_register_write): Use gdb_byte.
* sh64-tdep.c (sh64_push_dummy_call): Use gdb_byte.
(sh64_store_return_value, sh64_register_convert_to_virtual):
Change parameter type to 'gdb_byte *'. Use gdb_byte.
(sh64_pseudo_register_write): Use gdb_byte.
* solib-darwin.c (darwin_current_sos): Add casts to 'gdb_byte *'.
* solib-irix.c (fetch_lm_info): Likewise. Use gdb_byte for byte
buffer.
(irix_current_sos): Use gdb_byte.
* solib-som.c (som_current_sos): Use gdb_byte.
* sparc-ravenscar-thread.c (supply_register_at_address)
(sparc_ravenscar_generic_store_registers): Use gdb_byte.
* spu-multiarch.c (spu_xfer_partial): Add cast to 'char *'.
* spu-tdep.c (spu_get_overlay_table): Use gdb_byte.
* tic6x-tdep.c (tic6x_breakpoint_from_pc): Change return type to
'gdb_byte *'.
* tic6x-tdep.h (struct gdbarch_tdep) <breakpoint>: Change type to
'gdb_byte *'.
* tracepoint.c (tfile_fetch_registers): Use gdb_byte.
* xstormy16-tdep.c (xstormy16_extract_return_value)
(xstormy16_store_return_value): Change parameter type to
'gdb_byte *'. Adjust.
(xstormy16_push_dummy_call): Use gdb_byte.
* xtensa-tdep.c (xtensa_scan_prologue, call0_ret)
(call0_analyze_prologue, execute_code): Use gdb_byte.
Pedro Alves [Fri, 19 Apr 2013 14:13:30 +0000 (14:13 +0000)]
Fix the x87 FP register printout when issuing the “info float” command.
Consider the following simple program:
.globl _start
.text
_start:
fldt val
.data
val: .byte 0x00,0x00,0x45,0x07,0x11,0x19,0x22,0xe9,0xfe,0xbf
With current GDB on x86-64 GNU/Linux hosts, after the moment the fldt
command has been executed the register st(0) looks like this,
according to the “info regs” output (TOP=7):
which is clearly wrong (just count its length). The problem is due to
the printf statement (see patch) printing a promoted integer value of
a char argument "raw[i]", and, since char is signed on x86-64
GNU/Linux, the erroneous “ffffff” are printed for the first three
bytes which turn out to be "negative". The fix is to use gdb_byte
instead which is unsigned (and is the type of value_contents(), the
type to be used for raw target bytes anyway). After the fix the value
will be printed correctly:
* dwarf2read.c (struct signatured_type): New member type.
(struct attribute): Replace member signatured_type with signature.
(DW_SIGNATURE): Replaces DW_SIGNATURE_TYPE.
(read_call_site_scope): Call follow_die_ref instead of
follow_die_ref_or_sig.
(read_structure_type): Rewrite handling of signatured types.
(read_enumeration_type): Ditto.
(read_attribute_value): Update.
(build_error_marker_type): New function.
(lookup_die_type): Add assert. Rewrite handling of signatured types.
Don't call error for bad types, just build an error marker type.
(dump_die_shallow): Update.
(follow_die_sig_1): Renamed from follow_die_sig.
Don't call error for bad types, instead return NULL.
(follow_die_sig): New function.
(get_signatured_type, get_DW_AT_signature_type): New functions.
Pedro Alves [Tue, 16 Apr 2013 18:20:40 +0000 (18:20 +0000)]
Fix remaining GDBserver issues with !HAVE_THREAD_DB_H.
The previous patches are still not sufficient to build gdbserver with
our copy of thread_db.h.
../../../src/gdb/gdbserver/thread-db.c: In function ‘find_one_thread’:
../../../src/gdb/gdbserver/thread-db.c:316:6: error: ‘struct lwp_info’ has no member named ‘th’
../../../src/gdb/gdbserver/thread-db.c: In function ‘attach_thread’:
../../../src/gdb/gdbserver/thread-db.c:341:6: error: ‘struct lwp_info’ has no member named ‘th’
../../../src/gdb/gdbserver/thread-db.c: In function ‘thread_db_get_tls_address’:
../../../src/gdb/gdbserver/thread-db.c:514:47: error: ‘struct lwp_info’ has no member named ‘th’
make: *** [thread-db.o] Error 1
First, linux-low.h is including <thread_db.h> directly instead of our
gdb_thread_db.h, although thread-db.c includes the latter. Then the
'th' field of struct lwp_info is only defined if HAVE_THREAD_DB_H is
defined, which is not true if we're using our replacement copy of
thread_db.h. We have a USE_THREAD_DB symbol defined if we're building
thread-db.c that's ideal for this, however, it's currently only
defined when compiling linux-low.c (through a Makefile rule). The
patch makes it defined when compiling any file.
gdb/gdbserver/
2013-04-16 Pedro Alves <palves@redhat.com>
* Makefile.in (INTERNAL_CFLAGS): Add @USE_THREAD_DB@.
(linux-low.o): Delete rule.
* linux-low.h: Always include "gdb_thread_db.h" instead of
conditionally including thread_db.h.
(struct lwp_info) <th>: Guard with #ifdef USE_THREAD_DB instead of
HAVE_THREAD_DB_H.
Pedro Alves [Tue, 16 Apr 2013 18:17:11 +0000 (18:17 +0000)]
Update glibc_thread_db.h from upstream.
Attempting to build gdbserver with our copy of thread_db.h yields:
In file included from ../../../src/gdb/gdbserver/../common/gdb_thread_db.h:4:0,
from ../../../src/gdb/gdbserver/thread-db.c:30:
../../../src/gdb/gdbserver/../common/glibc_thread_db.h:108:3: error: unknown type name ‘uint32_t’
In file included from ../../../src/gdb/gdbserver/../common/gdb_thread_db.h:4:0,
from ../../../src/gdb/gdbserver/thread-db.c:30:
../../../src/gdb/gdbserver/../common/glibc_thread_db.h:199:5: error: unknown type name ‘uintptr_t’
../../../src/gdb/gdbserver/../common/glibc_thread_db.h:269:3: error: unknown type name ‘intptr_t’
../../../src/gdb/gdbserver/../common/glibc_thread_db.h:270:3: error: unknown type name ‘intptr_t’
We used to have a workaround for this, but the patch to import
gnulib's stdint.h removed it:
and defs.h made to always include stdint.h. However, gdbserver
doesn't include stdint.h in its equivalent server.h.
Rather than working around the issue, I've imported a more recent
version from glibc, which itself includes <stdint.h>. Other than
copyright years and FSF snail mail address, the file hasn't been
touched since 2003 in glibc. AFAICS, our version was updated last in
2000-09-03.
A note on the apparent license change: before the previous patch, this
file's contents were part of gdb_thread_db.h, and we can see that its
license's text was changed in this patch
<http://sourceware.org/ml/gdb-patches/2009-03/msg00251.html>. That
was certainly just an easy to overlook grep/sed mistake that fell
through the cracks.
gdb/common/
2013-04-16 Pedro Alves <palves@redhat.com>
Pedro Alves [Tue, 16 Apr 2013 18:14:55 +0000 (18:14 +0000)]
Move fallback thread_db.h to a separate file.
Having this on a separate file makes it easier to import a new version
-- one can just copy over instead of having to care about preserving
the GDB-specific bits.
2013-04-16 Pedro Alves <palves@redhat.com>
* common/gdb_thread_db.h [!HAVE_THREAD_DB_H]: Factor out to ...
* common/glibc_thread_db.h: ... this new file ...
* common/gdb_thread_db.h [!HAVE_THREAD_DB_H]: ... and include it.
Changed behaviour by only defining LIBTHREAD_DB_SO if thread_db.h exists. The
definition of LIBTHREAD_DB_SO and LIBTHREAD_DB_SEARCH_PATH should be moved
outside of #ifdef HAVE_THREAD_DB_H.
This is based on a patch attached to the PR, however, it needed a
tweak, as it was it broke the HAVE_THREAD_DB_H path.
2013-04-16 Will Newton <will.newton@gmail.com>
Pedro Alves <palves@redhat.com>
PR build/11881
* common/gdb_thread_db.h (LIBTHREAD_DB_SO)
(LIBTHREAD_DB_SEARCH_PATH): Move outside of #ifdef
HAVE_THREAD_DB_H.
* dwarf2read.c (struct dwo_file): Replace member "cus" with "cu".
(struct create_dwo_cu_data): Renamed from create_dwo_info_table_data.
(create_dwo_cu_reader): Renamed from
create_dwo_debug_info_hash_table_reader.
(create_dwo_cu): Renamed from create_dwo_debug_info_hash_table.
Remove support for multiple CUs in a DWO file.
(open_and_init_dwo_file, lookup_dwo_cutu): Update.
H.J. Lu [Mon, 15 Apr 2013 21:16:18 +0000 (21:16 +0000)]
Check regular reference without non-GOT reference
non_got_ref may not be set when building shared library. We need to set
non_got_ref if there are any non-PIC relocations. But we only did this
when there were no PLT/GOT relocations. It failed when there is a PLT
relocation. This checkin moves the non_got_ref check out.
bfd/
2013-04-15 H.J. Lu <hongjiu.lu@intel.com>
PR ld/15371
* elf-ifunc.c (_bfd_elf_allocate_ifunc_dyn_relocs): Check
regular reference without non-GOT reference when building
shared library.
Tom Tromey [Mon, 15 Apr 2013 17:30:36 +0000 (17:30 +0000)]
PR c++/11990:
* c-lang.c (cplus_language_defn): Use gdb_demangle.
* c-typeprint.c (c_type_print_base): Use gdb_demangle.
* cp-support.c (mangled_name_to_comp): Use gdb_demangle.
(gdb_demangle): New function.
* cp-support.h (gdb_demangle): Declare.
* dwarf2read.c (dwarf2_physname, fixup_partial_die)
(dwarf2_name): Use gdb_demangle.
* gdbtypes.c (check_stub_method): Use gdb_demangle.
* gnu-v3-abi.c (gnuv3_rtti_type): Strip @plt and version
suffixes from name.
(gnuv3_print_method_ptr): Use gdb_demangle.
* jv-lang.c (java_demangle): Use gdb_demangle.
* jv-typeprint.c (java_type_print_base): Use gdb_demangle.
* language.c (unk_lang_demangle): Use gdb_demangle.
* symtab.c (symbol_find_demangled_name)
(demangle_for_lookup): Use gdb_demangle.
Julian Brown [Mon, 15 Apr 2013 10:30:23 +0000 (10:30 +0000)]
gas/
* expr.c (add_to_result, subtract_from_result): Make global.
* expr.h (add_to_result, subtract_from_result): Add prototypes.
* config/tc-sh.c (sh_optimize_expr): Use add_to_result,
subtract_from_result to handle extra bit of precision for .sleb128
directive operands.
gas/testsuite/
* gas/all/gas.exp (sleb128-7): Don't run for tic4x, tic54x.
* gas/all/sleb128-2.s: Reformat, use _ at start of labels, remove
cruft.
* gas/all/sleb128-3.s: Likewise.
* gas/all/sleb128-4.s: Likewise.
* gas/all/sleb128-5.s: Likewise.
* gas/all/sleb128-7.s: Likewise.
* gas/all/sleb128-2.d: Handle data sections named $DATA$.
* gas/all/sleb128-3.d: Likewise.
* gas/all/sleb128-4.d: Likewise.
* gas/all/sleb128-5.d: Likewise.
* gas/all/sleb128-7.d: Likewise.
projects / thirdparty / binutils-gdb.git / log
