From: Yao Qi <qiyaoltc@gmail.com>
To: Pedro Alves <palves@redhat.com>
Cc: John Baldwin <jhb@freebsd.org>, gdb-patches@sourceware.org
Subject: Re: [PATCH v2] Fix PR gdb/18653: gdb disturbs inferior's inherited signal dispositions
Date: Mon, 22 Aug 2016 14:28:00 -0000 [thread overview]
Message-ID: <864m6cdgju.fsf@gmail.com> (raw)
In-Reply-To: <86d1ddee-9841-d332-27d5-15c8c185063e@redhat.com> (Pedro Alves's message of "Fri, 5 Aug 2016 11:56:56 +0100")
Pedro Alves <palves@redhat.com> writes:
> diff --git a/gdb/testsuite/gdb.base/signals-state-child.exp b/gdb/testsuite/gdb.base/signals-state-child.exp
> new file mode 100644
> index 0000000..f5fdcb2
> +
> +set gdb_txt [standard_output_file gdb.txt]
> +set standalone_txt [standard_output_file standalone.txt]
> +remote_exec host "rm -f $gdb_txt"
> +remote_exec host "rm -f $standalone_txt"
> +
> +set options [list debug "additional_flags=-DOUTPUT_TXT=\"$gdb_txt\""]
> +if {[build_executable $testfile.exp $testfile $srcfile $options]} {
> + untested $testfile.exp
> + return -1
> +}
$gdb_txt and $standalone_txt is the file location on build, but we pass
them to the test file, which will be run on target. It won't work if
build != target.
> +
> +set options [list debug "additional_flags=-DOUTPUT_TXT=\"$standalone_txt\""]
> +if {[build_executable $testfile.exp $testfile-standalone $srcfile $options]} {
> + untested $testfile.exp
> + return -1
> +}
> +
> +# Run the program directly, and dump its initial signal actions and
> +# mask in "standalone.txt".
> +
> +# Use remote_spawn instead of remote_exec, like how we spawn gdb.
> +# This is in order to take the same code code paths in dejagnu
> +# compared to when running the program through gdb. E.g., because
> +# local_exec uses -ignore SIGHUP, while remote_spawn does not, if we
> +# used remote_exec, the test program would start with SIGHUP ignored
> +# when run standalone, but not when run through gdb.
> +set res [remote_spawn host "$binfile-standalone"]
s/host/target/
> +if { $res < 0 || $res == "" } {
> + untested "spawning $binfile-standalone failed"
> + return 1
> +} else {
> + pass "collect standalone signals state"
> +}
> +remote_close host
How about the patch below? It follows the way we fix the problem of
this kind in mi-traceframe-changed.exp and tfile.exp.
How about the patch below? It should go to master and 7.12 branch.
--
Yao (齐尧)
From 6be8c11c9dc2f733449c1a988dee5f8354f5dfc6 Mon Sep 17 00:00:00 2001
From: Yao Qi <yao.qi@linaro.org>
Date: Fri, 19 Aug 2016 14:49:31 +0100
Subject: [PATCH] Fix signals-state-child.exp in remote testing
Remote testing isn't considered in signals-state-child.exp, so the it
fails like
shell diff -s /scratch/yao/gdb/build-git/aarch64-linux-gnu/gdb/testsuite/outputs/gdb.base/signals-state-child/standalone.txt /scratch/yao/gdb/build-git/aarch64-linux-gnu/gdb/testsuite/outputs/gdb.base/signals-state-child/gdb.txt^M
diff: /scratch/yao/gdb/build-git/aarch64-linux-gnu/gdb/testsuite/outputs/gdb.base/signals-state-child/standalone.txt: No such file or directory^M
(gdb) FAIL: gdb.base/signals-state-child.exp: signals states are identical
This patch is to fix it.
gdb/testsuite:
2016-08-22 Yao Qi <yao.qi@linaro.org>
* gdb.base/signals-state-child.exp: Set variables gdb_txt and
standalone_txt. Delete gdb_txt and standalone_txt on host
and target. Spawn the binary on target. Copy files from
target to host.
diff --git a/gdb/testsuite/gdb.base/signals-state-child.exp b/gdb/testsuite/gdb.base/signals-state-child.exp
index f5fdcb2..cd6a9a5 100644
--- a/gdb/testsuite/gdb.base/signals-state-child.exp
+++ b/gdb/testsuite/gdb.base/signals-state-child.exp
@@ -29,10 +29,20 @@
standard_testfile
-set gdb_txt [standard_output_file gdb.txt]
-set standalone_txt [standard_output_file standalone.txt]
-remote_exec host "rm -f $gdb_txt"
-remote_exec host "rm -f $standalone_txt"
+if {![is_remote host] && ![is_remote target]} {
+ set gdb_txt [standard_output_file gdb.txt]
+ set standalone_txt [standard_output_file standalone.txt]
+ set purely_local 1
+} else {
+ set gdb_txt gdb.txt
+ set standalone_txt standalone.txt
+ set purely_local 0
+}
+
+remote_file host delete $gdb_txt
+remote_file host delete $standalone_txt
+remote_file target delete $gdb_txt
+remote_file target delete $standalone_txt
set options [list debug "additional_flags=-DOUTPUT_TXT=\"$gdb_txt\""]
if {[build_executable $testfile.exp $testfile $srcfile $options]} {
@@ -55,14 +65,14 @@ if {[build_executable $testfile.exp $testfile-standalone $srcfile $options]} {
# local_exec uses -ignore SIGHUP, while remote_spawn does not, if we
# used remote_exec, the test program would start with SIGHUP ignored
# when run standalone, but not when run through gdb.
-set res [remote_spawn host "$binfile-standalone"]
+set res [remote_spawn target "$binfile-standalone"]
if { $res < 0 || $res == "" } {
untested "spawning $binfile-standalone failed"
return 1
} else {
pass "collect standalone signals state"
}
-remote_close host
+remote_close target
# Now run the program through gdb, and dump its initial signal actions
# and mask in "gdb.txt".
@@ -76,6 +86,12 @@ if { ! [ runto_main ] } then {
gdb_continue_to_end "collect signals state under gdb"
+if {!$purely_local} {
+ # Copy file from target to host through build.
+ remote_download host [remote_upload target gdb.txt] gdb.txt
+ remote_download host [remote_upload target standalone.txt] standalone.txt
+}
+
# Diff the .txt files. They should be identical.
gdb_test "shell diff -s $standalone_txt $gdb_txt" \
"Files .* are identical.*" \
next prev parent reply other threads:[~2016-08-22 14:28 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-08-04 17:37 [PATCH] " Pedro Alves
2016-08-05 0:08 ` John Baldwin
2016-08-05 1:04 ` Pedro Alves
2016-08-05 10:57 ` [PATCH v2] " Pedro Alves
2016-08-08 17:28 ` John Baldwin
2016-08-12 8:21 ` Yao Qi
2016-08-12 9:08 ` Pedro Alves
2016-08-22 14:28 ` Yao Qi [this message]
2016-08-22 14:34 ` Pedro Alves
2016-08-23 13:20 ` Yao Qi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=864m6cdgju.fsf@gmail.com \
--to=qiyaoltc@gmail.com \
--cc=gdb-patches@sourceware.org \
--cc=jhb@freebsd.org \
--cc=palves@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox