Eliminate is_remote check in gdb.base/scope.exp

This commit makes --target_board=native-gdbserver (and in principle
all other is_remote boards) pass all the same gdb.base/scope.exp tests
as native testing.

I first wrote the gdb.base/scope.exp change described in the ChangeLog
below and in the new comments in the patch, knowing that gdb_file_cmd
was the right thing to use here.

However, that revealed that the native-extended-gdbserver board should
be overriding gdb_file_cmd+gdb_reload instead of gdb_load, as is
hinted at by the comments on top of the default implementations in
testsuite/lib/gdb.exp, because otherwise a gdb_run_cmd after
gdb_file_cmd misses setting "set remote exec-file".  However, if we do
that and remove gdb_load, then we regress gdb.base/dbx.exp, so for now
keep the gdb_load override as well.

gdb/testsuite/ChangeLog:
2017-10-13  Pedro Alves  <palves@redhat.com>

	* gdb.base/scope.exp: Use build_executable + clean_restart +
	gdb_file_cmd instead of prepare_for_testing and no longer skip
	"before run" tests on is_remote target boards.  Update comments.
	* boards/native-extended-gdbserver.exp
	(extended_gdbserver_load_last_file): New, factored out from ...
	(gdb_load): ... this.  Move further below and add comment.
	(extended_gdbserver_gdb_file_cmd, gdb_file_cmd, gdb_reload): New.

diff --git a/gdb/testsuite/ChangeLog b/gdb/testsuite/ChangeLog
index 06c62f4d6b122a1ed14e0101bcd87af83bfca340..38add6f50b21769407e9401eba8bf9f8cbe99f82 100644 (file)
--- a/gdb/testsuite/ChangeLog
+++ b/gdb/testsuite/ChangeLog
@@ -1,3 +1,13 @@
+2017-10-13  Pedro Alves  <palves@redhat.com>
+
+       * gdb.base/scope.exp: Use build_executable + clean_restart +
+       gdb_file_cmd instead of prepare_for_testing and no longer skip
+       "before run" tests on is_remote target boards.  Update comments.
+       * boards/native-extended-gdbserver.exp
+       (extended_gdbserver_load_last_file): New, factored out from ...
+       (gdb_load): ... this.  Move further below and add comment.
+       (extended_gdbserver_gdb_file_cmd, gdb_file_cmd, gdb_reload): New.
+
 2017-10-12  Pedro Alves  <palves@redhat.com>
 
        * gdb.base/attach.exp: Remove references to gdb64.

diff --git a/gdb/testsuite/boards/native-extended-gdbserver.exp b/gdb/testsuite/boards/native-extended-gdbserver.exp
index 4ea210f0953acc94b387116da151f647d43fc5e5..564ffd2229baed145da9c89c7ad1c0dfe48a7338 100644 (file)
--- a/gdb/testsuite/boards/native-extended-gdbserver.exp
+++ b/gdb/testsuite/boards/native-extended-gdbserver.exp
@@ -80,17 +80,12 @@ proc mi_gdb_start { args } {
     return 0
 }
 
-# Overriden in order to set the remote exec-file whenever a file is
-# loaded to gdb.
-#
-proc gdb_load { arg } {
+# Helper that runs "set remote exec-file" with the last loaded file.
+
+proc extended_gdbserver_load_last_file {} {
     global gdb_prompt
     global last_loaded_file
 
-    if { $arg != "" } {
-       if [gdb_file_cmd $arg] then { return -1 }
-    }
-
     send_gdb "set remote exec-file $last_loaded_file\n"
     gdb_expect {
        -re "$gdb_prompt $" {}
@@ -103,6 +98,37 @@ proc gdb_load { arg } {
     return 0
 }
 
+# Overriden in order to set the remote exec-file whenever a file is
+# loaded to gdb.
+#
+if { [info procs extended_gdbserver_gdb_file_cmd] == "" } {
+    rename gdb_file_cmd extended_gdbserver_gdb_file_cmd
+}
+proc gdb_file_cmd { arg } {
+    if [extended_gdbserver_gdb_file_cmd $arg] {
+       return -1
+    }
+    return [extended_gdbserver_load_last_file]
+}
+
+proc gdb_reload { } {
+    return [extended_gdbserver_load_last_file]
+}
+
+# With the two procedure overrides above, it shouldn't be necessary to
+# override this one too.  However, not doing so regresses
+# gdb.base/dbx.exp.  See comments above gdb.base/dbx.exp:gdb_file_cmd.
+# Once testing of the "symbol-file"/"exec-file" commands is moved out
+# to a separate non-dbx testcase, we should be able to remove this.
+proc gdb_load { arg } {
+    if { $arg != "" } {
+       if [gdb_file_cmd $arg] then { return -1 }
+    }
+
+    return [extended_gdbserver_load_last_file]
+}
+
+
 # Likewise, for MI.
 #
 if { [info procs extended_gdbserver_mi_gdb_load] == "" } {

diff --git a/gdb/testsuite/gdb.base/scope.exp b/gdb/testsuite/gdb.base/scope.exp
index 48d8245776faf3556eedb86cc4c8268702b235b9..f26775d91f4958bdae92db042cfe29c967a27b24 100644 (file)
--- a/gdb/testsuite/gdb.base/scope.exp
+++ b/gdb/testsuite/gdb.base/scope.exp
@@ -18,7 +18,7 @@
 
 standard_testfile scope0.c scope1.c
 
-if {[prepare_for_testing "failed to prepare" ${testfile} \
+if {[build_executable "failed to build" ${testfile} \
         [list $srcfile $srcfile2] {debug}]} {
     return -1
 }
@@ -497,6 +497,13 @@ proc test_at_localscopes {} {
     gdb_stop_suppressing_tests
 }
 
+# Note, we don't pass a binfile to clean_restart because we need to
+# avoid gdb_load here.  With remote targets, gdb_load connects to the
+# remote target, and we specificaly want the following tests to run
+# before the program is started.
+clean_restart
+gdb_file_cmd $binfile
+
 # Test that variables in various segments print out correctly before
 # the program is run.  
 
@@ -505,21 +512,15 @@ setup_xfail "rs6000-*-*"
 
 gdb_test "print 'scope0.c'::filelocal_ro" "= 201"
 
-# gdb currently cannot access bss memory on some targets if the inferior
-# is not running.
-#
-# For PA boards using monitor/remote-pa.c, the bss test is going to 
-# randomly fail.  We've already put remote-pa on the target stack,
-# so we actually read memory from the board.  Problem is crt0.o
-# is responsible for clearing bss and that hasnt' happened yet.
+# Check that gdb can access bss memory if the inferior is not running.
 #
-# This is a problem for all non-native targets. -- manson
-if [is_remote target] {
-    unsupported "print 'scope0.c'::filelocal_bss before run"
-} else {
-    gdb_test "print 'scope0.c'::filelocal_bss" "= 0" \
-       "print 'scope0.c'::filelocal_bss before run"
-}
+# Note the care above about avoiding gdb_load.  Otherwise, for
+# embedded stub-like boards, this test would randomly fail.  If we'd
+# already put target remote on the target stack, we'd read memory from
+# the board instead of the program's binary, and we would have
+# connected before crt0.o had had a chance to clear bss.
+gdb_test "print 'scope0.c'::filelocal_bss" "= 0" \
+    "print 'scope0.c'::filelocal_bss before run"
 
 gdb_test "print 'scope0.c'::filelocal" "= 1" \
     "print 'scope0.c'::filelocal before run"