summaryrefslogtreecommitdiff
path: root/README.google
blob: 21bc65f784fb6cecf429dfbf850ddb310d8d82c3 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
Remember standard operating procedure: Patches go into the FSF tree
*first*, when possible.  At the very least non-Google specific patches
should be submitted to the FSF first.  If you need to timeout on waiting
for FSF approval that's fine.  When the patch goes in upstream, then
delete its entry here and copy over the entry from the appropriate
ChangeLog(s).  The goal here is to minimize unnecessary local patches,
they are an ongoing maintenance burden.

2015-09-05  Doug Evans  <dje@google.com>

	N.B. No further README.google entries are required for google.[ch].
	You can still add them if they make sense of course.
	This one exists because of the change to Makefile.in.
	* Makefile.in: Add google.c.
	* google.c: New file.
	* google.h: New file.

2015-09-05  Doug Evans  <dje@google.com>

	Implement 1155829.
	* exec.c (elf_64_p): New function.
	(exec_file_attach): If the file format is not recognized, check if
	it's 64-bit elf and if so recommend using 64-bit gdb. Improve error
	message if binary is accidently a core file.
	* main.c (captured_main): Call warn_troublesome_environment if !quiet.
	* google.c (warn_troublesome_environment): New function.
	* symfile.c (print_no_debugging_symbols_found): New function.
	(symbol_file_add_with_addrs_or_offsets): Call it if no debugging
	symbols found.
	(reread_symbols): Ditto.
	* testsuite/lib/gdb.exp (gdb_file_cmd): Update to match
	output of print_no_debugging_symbols_found.
	Properly update gdb_file_cmd_debug_info for nodebug when
	loading new symbol table.
	* testsuite/gdb.base/sepdebug.exp: Update expected output to match
	"no debugging symbols found" warning change for 1155829.
--- README.google	2015-09-05 14:53:18.000000000 -0700
+++ README.google	2015-09-05 15:01:01.000000000 -0700
+
+2015-09-05  Doug Evans  <dje@google.com>
+
+	Fix for http://b/2630476
+	* remote.c (readchar): Use minimum timeout of 10 minutes in noack mode.
+	(putpkt_binary): Don't resend QStartNoAckMode command packets.
+
+	* remote.c (readchar): Don't lengthen timeout value of 0,
+	return immediately means return immediately.
--- README.google	2015-09-05 15:13:05.000000000 -0700
+++ README.google	2015-09-05 15:16:13.000000000 -0700
+
+2015-09-05  Doug Evans  <dje@google.com>
+
+	Temporary hack to make these tests pass with Fission.
+	To be replaced with something that can be checked in upstream
+	when support for .dwp files is completed.
+	* gdb.python/py-prettyprint.exp: Do all C testing before any C++
+	testing.
+
+	* testsuite/gdb.mi/mi-var-invalidate.exp: Reorder compilations so
+	that var-cmd.dwo isn't clobbered.
+
+	* testsuite/boards/fission.exp: Add blaze support.
--- README.google	2015-09-05 15:16:13.000000000 -0700
+++ README.google	2015-09-05 15:19:05.000000000 -0700
+
+2015-09-05  Doug Evans  <dje@google.com>
+
+	* dwarf2read.c (use_index_sections): New static global.
+	(read_index_from_section): Check it.
+	(_initialize_dwarf2_read): New option "use-index-sections".
--- README.google	2015-09-05 15:21:34.000000000 -0700
+++ README.google	2015-09-05 15:24:02.000000000 -0700
+
+2015-09-05  Doug Evans  <dje@google.com>
+
+	Temp fix for http://sourceware.org/bugzilla/show_bug.cgi?id=14704
+	* bfd/elf.c (bfd_section_from_shdr, case SHT_REL,SHT_RELA): Don't mark
+	the target section as having relocs if the reloc section is empty.
--- README.google	2015-09-05 15:24:02.000000000 -0700
+++ README.google	2015-09-05 16:13:49.000000000 -0700
+
+2015-09-05  Doug Evans  <dje@google.com>
+
+	Quick hack to workaround bfd memory wastage, bug 8033013.
+	Upstream: http://sourceware.org/bugzilla/show_bug.cgi?id=14108
+	* bfd/bfd-in.h (bfd_init_14108, bfd_release_14108): Declare.
+	* bfd/bfd-in2.h: Regenerate.
+	* bfd/bfd.c (struct bfd): New members use_14108, memory_14108.
+	* bfd/elfcode.h (elf_slurp_symbol_table): If user requested, allocate
+	space for symbols in memory_14108.
+	* bfd/opncls.c (_bfd_delete_bfd): Free memory_14108.
+	(_bfd_free_cached_info): Ditto.
+	(bfd_init_14108, bfd_release_14108): New functions.
+	(bfd_alloc_aux, bfd_release_all_aux): New functions.
+	* gdb/elfread.c (elf_symfile_read): Free memory for normal symtab.
+	* gdb/solib.c (gdb_bfd_lookup_symbol_from_symtab): Ditto.
--- README.google	2015-09-05 16:13:49.000000000 -0700
+++ README.google	2015-09-05 16:18:05.000000000 -0700
+
+2015-09-05  Doug Evans  <dje@google.com>
+
+	Workaround http://sourceware.org/bugzilla/show_bug.cgi?id=15412
+	* typeprint.c (restore_default_ptype_raw): New function.
+	(symbol_info_type_print, symbol_info_typedef_print): New functions.
+	* typeprint.h (symbol_info_type_print): Declare.
+	(symbol_info_typedef_print): Declare.
+	* symtab.c: #include "typeprint.h".
+	(print_symbol_info): Call them.
--- README.google	2015-09-05 16:18:05.000000000 -0700
+++ README.google	2015-09-05 16:20:56.000000000 -0700
+
+2015-09-05  Doug Evans  <dje@google.com>
+
+	* defs.h (begin_uninterruptible_section): Declare.
+	(defer_quit): Declare.
+	* utils.c (defer_quit): New global.
+	(end_uninterruptible_cleanup): New function.
+	(begin_uninterruptible_cleanup): New function.
+	* extension.c (check_quit_flag): Check it. [revised from original]
+	* dwarf2read.c (dw2_instantiate_symtab): Mark symtab expansion as
+	uninterruptible.
+	* psymtab.c (psymtab_to_symtab): Ditto.
+	* python/python.c (check_quit_flag): Ditto.
--- README.google	2015-09-05 16:20:56.000000000 -0700
+++ README.google	2015-09-05 16:22:58.000000000 -0700
+
+2015-09-05  Doug Evans  <dje@google.com>
+
+	Ref# 2180247
+	* infcmd.c (attach_command_post_wait): Pass zero to
+	post_create_inferior, same as run_command_1.
--- README.google	2015-09-05 16:24:43.000000000 -0700
+++ README.google	2015-09-05 16:33:32.000000000 -0700
+
+2015-09-05  Doug Evans  <google.com>
+
+	Ref# 13734692, submitted upstream
+	PR gdb/14236
+	* NEWS: Mention change in default behaviour of "interrupt" command.
+	* infcmd.c (struct interrupt_command_continuation_args): New struct.
+	(interrupt_command_continuation_free_args): New function.
+	(wait_thread_stopped): New function.
+	(interrupt_command_post_wait): New function.
+	(interrupt_command_continuation): New function.
+	(add_interrupt_continuation): New function.
+	(interrupt_command): Rewrite.
+
+	doc/
+	* gdb.texinfo (Background Execution): Document new option "&" for
+	"interrupt" command.
+
+	testsuite/
+	* gdb.base/async-shell.exp: Add "&" to "interrupt" command.
+	* gdb.base/dprintf-non-stop.exp: Ditto.
+	* gdb.base/interrupt-noterm.exp: Ditto.
+	* gdb.threads/interrupt-command.c: New file.
+	* gdb.threads/interrupt-command.exp: New file.
+
+	* gdbthread.h (any_running): Declare.
+	* thread.c (any_running): New function.
--- README.google	2015-09-05 16:33:32.000000000 -0700
+++ README.google	2015-09-05 17:01:28.000000000 -0700
+
+2015-09-05  Doug Evans  <google.com>
+
+	Ref# 12994080
+	Add support for DW_AT_str_offsets_base.
+	This patch is GOOGLE LOCAL for now.
+	Not sure when to submit it upstream.
+
+	* dwarf2read.c (struct dwarf2_per_objfile): New member str_offsets.
+	(dwarf2_elf_names): Add .debug_str_offsets.
+	(struct dwarf2_cu): New members str_offsets_base,
+	have_str_offsets_base.
+	(struct attribute): Reduce member form to 14 bits.  New member
+	string_is_str_index.
+	(DW_STRING_IS_STR_INDEX): New macro.
+	(dwarf2_locate_sections): Handle .debug_str_offsets.
+	(read_cutu_die_from_dwo): Initialize DW_STRING_IS_STR_INDEX for
+	comp_dir attribute.  Call get_comp_dir_attr.
+	(lookup_dwo_unit): Replace arg this_cu with reader.  All callers
+	updated.
+	(init_cutu_and_read_dies): Fetch value of DW_AT_str_offsets_base.
+	(init_cutu_and_read_dies_no_follow): Ditto.
+	(process_psymtab_comp_unit_reader): Call get_comp_dir_attr.
+	(find_file_and_directory): Ditto.
+	(read_attribute_value): Initialize DW_STRING_IS_STR_INDEX for all
+	string forms.
+	(read_attribute_value) <DW_FORM_GNU_str_index>: Handle appearance in
+	Fission stubs.
+	(read_str_index): Replace arg reader with cu, str_section,
+	str_offsets_section, str_offsets_base.  All callers updated.
+	(read_dwo_str_index, read_stub_str_index): New functions.
+	(get_stub_string_attr, get_comp_dir_attr): New functions.
+	* symfile.h (struct dwarf2_debug_sections): New member str_offsets.
+	* xcoffread.c (dwarf2_xcoff_names): Add entry for str_offsets.
+
+	include/
+	* dwarf2.def (DW_AT_str_offsets_base): New attribute.
--- README.google	2015-09-05 17:04:41.000000000 -0700
+++ README.google	2015-09-05 18:02:29.000000000 -0700
+
+2015-09-05  Doug Evans  <google.com>
+
+	Upstream PR libc/13097, local ref# 17474967
+	* solib-svr4.c (svr4_read_so_list): Ignore linux-vdso.so.1
+	and linux-gate.so.1.
+
+	Upstream PR libc/13097, local ref# 17474967
+	* solib-svr4.c (svr4_read_so_list): Ignore linux-vdso64.so.1, for ppc.
--- README.google	2015-09-05 18:02:29.000000000 -0700
+++ README.google	2015-09-05 18:13:45.000000000 -0700
+
+2015-09-05  Doug Evans  <google.com>
+
+	* NEWS: Mention gdb.Objfile.have_debug_info.
+	* python/py-objfile.c (objfpy_get_have_debug_info): New function.
+	(objfile_getset): Add "have_debug_info".
+
+	doc/
+	* python.texi (Objfiles In Python): Document Objfile.have_debug_info.
+
+	testsuite/
+	* gdb.python/py-objfile.exp: Add tests for objfile.have_debug_info.
--- README.google	2015-09-05 18:13:45.000000000 -0700
+++ README.google	2015-09-05 18:20:10.000000000 -0700
+
+2015-09-05  Doug Evans  <google.com>
+
+	Ref: 15188412
+	* breakpoint.c (create_solib_event_breakpoint): Apply
+	gdbarch_skip_entrypoint if it's defined.
--- README.google	2015-09-05 18:23:16.000000000 -0700
+++ README.google	2015-09-05 18:26:36.000000000 -0700
+
+2015-09-05  Doug Evans  <google.com>
+
+	PR python/17936
+	* symfile.c (symbol_file_add_with_addrs): Call
+	observer_notify_new_objfile (NULL) early, before loading symbols.
+	(symbol_file_add_separate): Turn off SYMFILE_MAINLINE when loading
+	separate debug files.
+	(clear_symtab_users): Don't call observer_notify_new_objfile (NULL)
+	here if SYMFILE_MAINLINE.
--- README.google	2015-09-05 18:26:36.000000000 -0700
+++ README.google	2015-09-05 18:44:15.000000000 -0700
+
+2015-09-05  Doug Evans  <dje@google.com>
+
+	PR python/17936
+	* symfile.c (symbol_file_add_with_addrs): Don't query whether to load
+	a new symbol table for separate debug files.  Don't call
+	observer_notify_new_objfile (NULL) for separate debug files.  Turn off
+	SYMFILE_MAINLINE when calling finish_new_objfile for separate debug
+	files.
+	(symbol_file_add_separate): Don't turn off SYMFILE_MAINLINE.  Add a
+	comment explaining why.
+
+	testsuite/
+	* gdb.base/sepdebug.exp: Add test to verify two "Reading symbols ..."
+	messages are printed.
--- README.google	2015-09-05 18:44:15.000000000 -0700
+++ README.google	2015-09-05 18:49:14.000000000 -0700
-	* solib-svr4.c (svr4_read_so_list): Ignore linux-vdso64.so.1, for ppc.
+	* solib-svr4.c (svr4_read_so_list): Ignore linux-vdso64.so.1, for ppc,
+	and update for -Wc++-compat changes.
--- README.google	2015-09-05 18:49:14.000000000 -0700
+++ README.google	2015-09-05 18:55:44.000000000 -0700
+
+2015-09-05  Doug Evans  <dje@google.com>
+
+	testsuite/
+	* gdb.base/dprintf-detach.exp: Cope with local patch. Ref# 2180247