* [PATCH V2 0/5] gdb/testsuite: Miscelleanous fixes
@ 2023-04-23 19:31 Sandra Loosemore
2023-04-23 19:31 ` [PATCH V2 1/5] gdb/testsuite: Fix style.exp failures on targets without argc/argv support Sandra Loosemore
` (3 more replies)
0 siblings, 4 replies; 5+ messages in thread
From: Sandra Loosemore @ 2023-04-23 19:31 UTC (permalink / raw)
To: gdb-patches; +Cc: tom
Reposting this series, which now incorporates minor tweaks suggested by
Tom Tromey to parts 2, 3, and 4. Parts 1 and 5 are unchanged.
OK for trunk?
-Sandra
Sandra Loosemore (5):
gdb/testsuite: Fix style.exp failures on targets without argc/argv
support
gdb/testsuite: Fix style.exp failures on targets without libopcodes
styling
gdb/testsuite: Adjust some testcases to allow Windows pathnames
gdb/testsuite: Disable some tests that are broken on remote Windows
host
gdb/testsuite: Make hook-stop.exp ignore termination message from GDB
stub
gdb/testsuite/gdb.base/bad-file.exp | 3 ++
gdb/testsuite/gdb.base/early-init-file.exp | 3 ++
.../gdb.base/empty-host-env-vars.exp | 3 ++
gdb/testsuite/gdb.base/hook-stop.exp | 2 +-
.../maint-expand-symbols-header-file.exp | 4 +-
gdb/testsuite/gdb.base/setshow.exp | 17 +++++--
gdb/testsuite/gdb.base/style.exp | 49 ++++++++++++++++---
gdb/testsuite/gdb.dwarf2/dw2-filename.exp | 2 +-
gdb/testsuite/gdb.mi/mi-info-sources.exp | 32 ++++++------
9 files changed, 86 insertions(+), 29 deletions(-)
--
2.31.1
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH V2 1/5] gdb/testsuite: Fix style.exp failures on targets without argc/argv support
2023-04-23 19:31 [PATCH V2 0/5] gdb/testsuite: Miscelleanous fixes Sandra Loosemore
@ 2023-04-23 19:31 ` Sandra Loosemore
2023-04-23 19:31 ` [PATCH V2 2/5] gdb/testsuite: Fix style.exp failures on targets without libopcodes styling Sandra Loosemore
` (2 subsequent siblings)
3 siblings, 0 replies; 5+ messages in thread
From: Sandra Loosemore @ 2023-04-23 19:31 UTC (permalink / raw)
To: gdb-patches; +Cc: tom
Some embedded targets don't have full support for argc/argv. argv
may print as "0x0" or as an address with a symbol name following.
This causes problems for the regexps in the style.exp line-wrapping
tests that assume it always prints as an ordinary address in backtrace
output.
This patch generalizes the regexps to handle these additional forms
and reworks some of the line-wrapping tests to account for the argv
address string being shorter or longer than a regular address.
---
gdb/testsuite/gdb.base/style.exp | 23 ++++++++++++++++++-----
1 file changed, 18 insertions(+), 5 deletions(-)
diff --git a/gdb/testsuite/gdb.base/style.exp b/gdb/testsuite/gdb.base/style.exp
index 0370550d251..60f909e2402 100644
--- a/gdb/testsuite/gdb.base/style.exp
+++ b/gdb/testsuite/gdb.base/style.exp
@@ -92,7 +92,7 @@ proc run_style_tests { } {
set argv ""
gdb_test_multiple "frame" "frame without styling" {
- -re -wrap "main \\(argc=.*, (argv=$hex)\\).*style\\.c:\[0-9\].*" {
+ -re -wrap "main \\(argc=.*, (argv=$hex.*)\\).*style\\.c:\[0-9\].*" {
set argv $expect_out(1,string)
pass $gdb_test_name
}
@@ -105,9 +105,13 @@ proc run_style_tests { } {
set file_expr "$base_file_expr:\[0-9\]+"
set arg_expr [limited_style "arg." variable]
+ # On some embedded targets that don't fully support argc/argv,
+ # argv may print as "0x0" or as an address with a symbol name
+ # following, so use a regexp general enough to match that and
+ # do not make assumptions about how long the address string is.
gdb_test "frame" \
[multi_line \
- "#0\\s+$main_expr\\s+\\($arg_expr=$decimal,\\s+$arg_expr=$hex\\)\\s+at\\s+$file_expr" \
+ "#0\\s+$main_expr\\s+\\($arg_expr=$decimal,\\s+$arg_expr=$hex.*\\)\\s+at\\s+$file_expr" \
"\[0-9\]+\\s+.*return.* break here .*"]
gdb_test "info breakpoints" "$main_expr at $file_expr.*"
@@ -134,16 +138,21 @@ proc run_style_tests { } {
# the line listing; this is why the words from the source
# code are spelled out in the final result line of the
# test.
+ set re0_styled \
+ [multi_line \
+ "#0\\s+$main_expr\\s+\\($arg_expr=$decimal,\\s+$arg_expr=$hex\\)" \
+ "\\s+at\\s+$file_expr" \
+ "\[0-9\]+\\s+.*return.* break here .*"]
set re1_styled \
[multi_line \
"#0\\s+$main_expr\\s+\\($arg_expr=$decimal,\\s+" \
- "\\s+$arg_expr=$hex\\)" \
+ "\\s+$arg_expr=$hex.*\\)" \
"\\s+at\\s+$file_expr" \
"\[0-9\]+\\s+.*return.* break here .*"]
set re2_styled \
[multi_line \
"#0\\s+$main_expr\\s+\\($arg_expr=.*" \
- "\\s+$arg_expr=$hex\\)\\s+at\\s+$file_expr" \
+ "\\s+$arg_expr=$hex.*\\)\\s+at\\s+$file_expr" \
"\[0-9\]+\\s+.*return.* break here .*"]
# The length of the line containing argv containing:
@@ -152,7 +161,11 @@ proc run_style_tests { } {
# - closing parenthesis
set line_len [expr 4 + $argv_len + 1]
- if { $line_len > $width } {
+ if { $argv == "argv=0x0" && $width >= 27 } {
+ # Embedded target with no argv support.
+ # #0 main (argc=0, argv=0x0)
+ set re_styled $re0_styled
+ } elseif { $line_len > $width } {
# At on the next line.
set re_styled $re1_styled
} else {
--
2.31.1
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH V2 2/5] gdb/testsuite: Fix style.exp failures on targets without libopcodes styling
2023-04-23 19:31 [PATCH V2 0/5] gdb/testsuite: Miscelleanous fixes Sandra Loosemore
2023-04-23 19:31 ` [PATCH V2 1/5] gdb/testsuite: Fix style.exp failures on targets without argc/argv support Sandra Loosemore
@ 2023-04-23 19:31 ` Sandra Loosemore
2023-04-23 19:31 ` [PATCH V2 3/5] gdb/testsuite: Adjust some testcases to allow Windows pathnames Sandra Loosemore
2023-04-24 17:16 ` [PATCH V2 0/5] gdb/testsuite: Miscelleanous fixes Tom Tromey
3 siblings, 0 replies; 5+ messages in thread
From: Sandra Loosemore @ 2023-04-23 19:31 UTC (permalink / raw)
To: gdb-patches; +Cc: tom
Do not expect disassembler output to be styled by libopcodes if GDB
reports the target does not support that.
---
gdb/testsuite/gdb.base/style.exp | 26 +++++++++++++++++++++++++-
1 file changed, 25 insertions(+), 1 deletion(-)
diff --git a/gdb/testsuite/gdb.base/style.exp b/gdb/testsuite/gdb.base/style.exp
index 60f909e2402..94c9dc46d73 100644
--- a/gdb/testsuite/gdb.base/style.exp
+++ b/gdb/testsuite/gdb.base/style.exp
@@ -24,6 +24,26 @@ if {[build_executable "failed to build" $testfile $srcfile {debug macros}]} {
return -1
}
+# Check whether libopcodes styling is supported for this target.
+set have_libopcodes_styling 1
+proc test_for_libopcodes_styling {} {
+ global binfile
+ global have_libopcodes_styling
+ clean_restart ${binfile}
+
+ gdb_test_multiple "maint set libopcodes-styling enabled on" "" {
+ -re "^maint set libopcodes-styling enabled on\r\n" {
+ exp_continue
+ }
+ -re "Use of libopcodes styling not supported on architecture \[^\r\n\]+\r\n" {
+ set have_libopcodes_styling 0
+ exp_continue
+ }
+ -re "^$::gdb_prompt $" {
+ }
+ }
+}
+
# The tests in this file are run multiple times with GDB's styles
# disabled one at a time. This variable is the style that is
# currently disabled.
@@ -68,6 +88,7 @@ proc clean_restart_and_disable { prefix args } {
proc run_style_tests { } {
global testfile srcfile hex binfile
global currently_disabled_style decimal hex
+ global have_libopcodes_styling
save_vars { env(TERM) } {
# We need an ANSI-capable terminal to get the output.
@@ -195,7 +216,7 @@ proc run_style_tests { } {
# symbol will be styled. However, if pygments is not being
# used then we can know how the symbol name will be styled.
set main [limited_style main function]
- if { $::python_disassembly_styling } {
+ if { $::python_disassembly_styling || !$have_libopcodes_styling } {
set func "some_called_function"
} else {
set func [limited_style some_called_function function]
@@ -495,6 +516,9 @@ if {[allow_python_tests] && [gdb_py_module_available "pygments"]} {
set python_disassembly_styling false
}
+# Check to see if libopcodes styling is available for this target
+test_for_libopcodes_styling
+
# Run tests with all styles in their default state.
with_test_prefix "all styles enabled" {
run_style_tests
--
2.31.1
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH V2 3/5] gdb/testsuite: Adjust some testcases to allow Windows pathnames
2023-04-23 19:31 [PATCH V2 0/5] gdb/testsuite: Miscelleanous fixes Sandra Loosemore
2023-04-23 19:31 ` [PATCH V2 1/5] gdb/testsuite: Fix style.exp failures on targets without argc/argv support Sandra Loosemore
2023-04-23 19:31 ` [PATCH V2 2/5] gdb/testsuite: Fix style.exp failures on targets without libopcodes styling Sandra Loosemore
@ 2023-04-23 19:31 ` Sandra Loosemore
2023-04-24 17:16 ` [PATCH V2 0/5] gdb/testsuite: Miscelleanous fixes Tom Tromey
3 siblings, 0 replies; 5+ messages in thread
From: Sandra Loosemore @ 2023-04-23 19:31 UTC (permalink / raw)
To: gdb-patches; +Cc: tom
This patch fixes some testcases that formerly had patterns with
hardwired "/" pathname separators in them, which broke when testing on
(remote) Windows host.
---
.../maint-expand-symbols-header-file.exp | 4 +--
gdb/testsuite/gdb.base/setshow.exp | 17 ++++++++--
gdb/testsuite/gdb.dwarf2/dw2-filename.exp | 2 +-
gdb/testsuite/gdb.mi/mi-info-sources.exp | 32 +++++++++----------
4 files changed, 33 insertions(+), 22 deletions(-)
diff --git a/gdb/testsuite/gdb.base/maint-expand-symbols-header-file.exp b/gdb/testsuite/gdb.base/maint-expand-symbols-header-file.exp
index d38d2ab7398..43b5aabea3c 100644
--- a/gdb/testsuite/gdb.base/maint-expand-symbols-header-file.exp
+++ b/gdb/testsuite/gdb.base/maint-expand-symbols-header-file.exp
@@ -40,13 +40,13 @@ gdb_test_no_output "maint info symtabs" $test
gdb_test_no_output "maint expand-symtabs maint-expand-symbols-header-file.h"
# Check that the include symtab was in fact expanded.
-set file_re "\[^\r\n\]*/maint-expand-symbols-header-file.h"
+set file_re "\[^\r\n\]*maint-expand-symbols-header-file.h"
gdb_test "maint info symtabs" \
"\r\n\t{ symtab $file_re \\(\\(struct symtab \\*\\) $hex\\)\r\n.*" \
"check header file psymtab expansion"
# Check that the symtab the include symtab was referring to was expanded.
-set file_re "\[^\r\n\]*/maint-expand-symbols-header-file.c"
+set file_re "\[^\r\n\]*maint-expand-symbols-header-file.c"
gdb_test "maint info symtabs" \
"\r\n\t{ symtab $file_re \\(\\(struct symtab \\*\\) $hex\\)\r\n.*" \
"check source file psymtab expansion"
diff --git a/gdb/testsuite/gdb.base/setshow.exp b/gdb/testsuite/gdb.base/setshow.exp
index 86821ca1db0..684e2b6c3a4 100644
--- a/gdb/testsuite/gdb.base/setshow.exp
+++ b/gdb/testsuite/gdb.base/setshow.exp
@@ -280,10 +280,21 @@ proc_with_prefix test_setshow_history {} {
#get home directory path
set HOME ""
+ set STRINGHOME ""
set test "show environment HOME"
gdb_test_multiple $test $test {
-re "\nHOME = (\[^\r\n\]*)\[\r\n\]+$::gdb_prompt $" {
set HOME $expect_out(1,string)
+ if { [ishost *-*-mingw*] } {
+ # STRINGHOME is how HOME prints with C string escapes.
+ # Specifically, all backslashes "\" in the pathname
+ # string have to be escaped as "\\". If you have other
+ # weird characters in your HOME pathname that need
+ # escaping too, maybe you shouldn't do that. :-P
+ regsub -all {\\} $HOME {\\\\} STRINGHOME
+ } else {
+ set STRINGHOME $HOME
+ }
pass $test
}
}
@@ -294,13 +305,13 @@ proc_with_prefix test_setshow_history {} {
#test show history filename ~/foobar.baz
gdb_test "show history filename" \
- "The filename in which to record the command history is \"[string_to_regexp $HOME]/foobar.baz\"..*" \
+ "The filename in which to record the command history is \"[string_to_regexp $HOME].foobar.baz\"..*" \
"show history filename (~/foobar.baz)"
gdb_test "p \$_gdb_setting(\"history filename\")" \
- " = \"[string_to_regexp $HOME]/foobar.baz\"..*" \
+ " = \"[string_to_regexp $STRINGHOME].foobar.baz\".*" \
"_gdb_setting history filename"
gdb_test "p \$_gdb_setting_str(\"history filename\")" \
- " = \"[string_to_regexp $HOME]/foobar.baz\"..*" \
+ " = \"[string_to_regexp $STRINGHOME].foobar.baz\".*" \
"_gdb_setting_str history filename"
#get current working directory
diff --git a/gdb/testsuite/gdb.dwarf2/dw2-filename.exp b/gdb/testsuite/gdb.dwarf2/dw2-filename.exp
index 3bdac4c1cb8..c0ee3ac0d3f 100644
--- a/gdb/testsuite/gdb.dwarf2/dw2-filename.exp
+++ b/gdb/testsuite/gdb.dwarf2/dw2-filename.exp
@@ -38,4 +38,4 @@ gdb_test "interpreter-exec mi -file-list-exec-source-files" \
".*{file=\"file1\\.txt\",fullname=\".+file1\\.txt\",debug-fully-read=\"\[^\"\]+\"}.*"
# And `info sources' should return the fullname incl. the directories.
-gdb_test "info sources" {[/]file1\.txt.*}
+gdb_test "info sources" {.+file1\.txt.*}
diff --git a/gdb/testsuite/gdb.mi/mi-info-sources.exp b/gdb/testsuite/gdb.mi/mi-info-sources.exp
index f639f3cdef9..3778ccb3fbe 100644
--- a/gdb/testsuite/gdb.mi/mi-info-sources.exp
+++ b/gdb/testsuite/gdb.mi/mi-info-sources.exp
@@ -118,16 +118,16 @@ proc check_info_sources { debug_fully_read } {
[mi_list "files" \
".*" \
[mi_tuple "" \
- [mi_field "file" "\[^\"\]+/mi-info-sources-base\\.c"] \
- [mi_field "fullname" "\[^\"\]+/mi-info-sources-base\\.c"] \
+ [mi_field "file" "\[^\"\]*mi-info-sources-base\\.c"] \
+ [mi_field "fullname" "\[^\"\]+mi-info-sources-base\\.c"] \
[mi_field "debug-fully-read" "${debug_fully_read}"]] \
".*"]
set p2 \
[mi_list "files" \
".*" \
[mi_tuple "" \
- [mi_field "file" "\[^\"\]+/mi-info-sources\\.c"] \
- [mi_field "fullname" "\[^\"\]+/mi-info-sources\\.c"] \
+ [mi_field "file" "\[^\"\]*mi-info-sources\\.c"] \
+ [mi_field "fullname" "\[^\"\]+mi-info-sources\\.c"] \
[mi_field "debug-fully-read" "true"]] \
".*"]
} else {
@@ -135,16 +135,16 @@ proc check_info_sources { debug_fully_read } {
[mi_list "files" \
".*" \
[mi_tuple "" \
- [mi_field "file" "\[^\"\]+/mi-info-sources\\.c"] \
- [mi_field "fullname" "\[^\"\]+/mi-info-sources\\.c"] \
+ [mi_field "file" "\[^\"\]*mi-info-sources\\.c"] \
+ [mi_field "fullname" "\[^\"\]+mi-info-sources\\.c"] \
[mi_field "debug-fully-read" "true"]] \
".*"]
set p2 \
[mi_list "files" \
".*" \
[mi_tuple "" \
- [mi_field "file" "\[^\"\]+/mi-info-sources-base\\.c"] \
- [mi_field "fullname" "\[^\"\]+/mi-info-sources-base\\.c"] \
+ [mi_field "file" "\[^\"\]*mi-info-sources-base\\.c"] \
+ [mi_field "fullname" "\[^\"\]+mi-info-sources-base\\.c"] \
[mi_field "debug-fully-read" "${debug_fully_read}"]] \
".*"]
}
@@ -156,8 +156,8 @@ proc check_info_sources { debug_fully_read } {
set p [mi_list "files" \
[mi_tuple "" \
- [mi_field "file" "\[^\"\]+/mi-info-sources-base\\.c"] \
- [mi_field "fullname" "\[^\"\]+/mi-info-sources-base\\.c"] \
+ [mi_field "file" "\[^\"\]*mi-info-sources-base\\.c"] \
+ [mi_field "fullname" "\[^\"\]+mi-info-sources-base\\.c"] \
[mi_field "debug-fully-read" "${debug_fully_read}"]]]
mi_gdb_test "-file-list-exec-source-files --basename -- base" ".*\\^done,${p}" \
"-file-list-exec-source-files --basename -- base"
@@ -171,13 +171,13 @@ proc check_info_sources { debug_fully_read } {
set p [mi_list "files" \
[mi_tuple "" \
- [mi_field "filename" "\[^\"\]+/mi-info-sources(\.debug)?"] \
+ [mi_field "filename" "\[^\"\]+mi-info-sources(\.debug)?"] \
[mi_field "debug-info" "${debug_info}"] \
[mi_list "sources" \
".*" \
[mi_tuple "" \
- [mi_field "file" "\[^\"\]+/mi-info-sources\\.c"] \
- [mi_field "fullname" "\[^\"\]+/mi-info-sources\\.c"] \
+ [mi_field "file" "\[^\"\]*mi-info-sources\\.c"] \
+ [mi_field "fullname" "\[^\"\]+mi-info-sources\\.c"] \
[mi_field "debug-fully-read" "true"]] \
".*"]]]
mi_gdb_test "-file-list-exec-source-files --group-by-objfile" \
@@ -186,13 +186,13 @@ proc check_info_sources { debug_fully_read } {
set p [mi_list "files" \
[mi_tuple "" \
- [mi_field "filename" "\[^\"\]+/mi-info-sources(\.debug)?"] \
+ [mi_field "filename" "\[^\"\]+mi-info-sources(\.debug)?"] \
[mi_field "debug-info" "${debug_info}"] \
[mi_list "sources" \
".*" \
[mi_tuple "" \
- [mi_field "file" "\[^\"\]+/mi-info-sources-base\\.c"] \
- [mi_field "fullname" "\[^\"\]+/mi-info-sources-base\\.c"] \
+ [mi_field "file" "\[^\"\]*mi-info-sources-base\\.c"] \
+ [mi_field "fullname" "\[^\"\]+mi-info-sources-base\\.c"] \
[mi_field "debug-fully-read" "${debug_fully_read}"]] \
".*"]]]
mi_gdb_test "-file-list-exec-source-files --group-by-objfile" \
--
2.31.1
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH V2 0/5] gdb/testsuite: Miscelleanous fixes
2023-04-23 19:31 [PATCH V2 0/5] gdb/testsuite: Miscelleanous fixes Sandra Loosemore
` (2 preceding siblings ...)
2023-04-23 19:31 ` [PATCH V2 3/5] gdb/testsuite: Adjust some testcases to allow Windows pathnames Sandra Loosemore
@ 2023-04-24 17:16 ` Tom Tromey
3 siblings, 0 replies; 5+ messages in thread
From: Tom Tromey @ 2023-04-24 17:16 UTC (permalink / raw)
To: Sandra Loosemore; +Cc: gdb-patches, tom
>>>>> "Sandra" == Sandra Loosemore <sandra@codesourcery.com> writes:
Sandra> Reposting this series, which now incorporates minor tweaks suggested by
Sandra> Tom Tromey to parts 2, 3, and 4. Parts 1 and 5 are unchanged.
Sandra> OK for trunk?
Thank you. This looks good to me.
Reviewed-By: Tom Tromey <tom@tromey.com>
Tom
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2023-04-24 17:16 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-04-23 19:31 [PATCH V2 0/5] gdb/testsuite: Miscelleanous fixes Sandra Loosemore
2023-04-23 19:31 ` [PATCH V2 1/5] gdb/testsuite: Fix style.exp failures on targets without argc/argv support Sandra Loosemore
2023-04-23 19:31 ` [PATCH V2 2/5] gdb/testsuite: Fix style.exp failures on targets without libopcodes styling Sandra Loosemore
2023-04-23 19:31 ` [PATCH V2 3/5] gdb/testsuite: Adjust some testcases to allow Windows pathnames Sandra Loosemore
2023-04-24 17:16 ` [PATCH V2 0/5] gdb/testsuite: Miscelleanous fixes Tom Tromey
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).