Bug 31514 - [gdb/testsuite] FAIL: gdb.ada/tagged-lookup.exp: only one CU expanded
Summary: [gdb/testsuite] FAIL: gdb.ada/tagged-lookup.exp: only one CU expanded
Status: RESOLVED FIXED
Alias: None
Product: gdb
Classification: Unclassified
Component: testsuite (show other bugs)
Version: HEAD
: P2 normal
Target Milestone: 15.1
Assignee: Not yet assigned to anyone
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2024-03-20 10:24 UTC by Tom de Vries
Modified: 2024-03-25 14:29 UTC (History)
0 users

See Also:
Host:
Target:
Build:
Last reconfirmed:

Attachments
passing gdb.log (on openSUSE Tumbleweed) (2.37 KB, text/x-log)
2024-03-20 15:31 UTC, Tom de Vries 		Details
failing gdb.log (on openSUSE Leap 15.4) (1.70 KB, text/x-log)
2024-03-20 16:38 UTC, Tom de Vries 		Details
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Tom de Vries 2024-03-20 10:24:33 UTC 
On openSUSE Leap 15.4, using gcc 7.5.0, I'm running into:
...
FAIL: gdb.ada/tagged-lookup.exp: only one CU expanded
...

In more detail:
...
(gdb) set debug symtab-create 1^M
(gdb) PASS: gdb.ada/tagged-lookup.exp: set debug symtab-create 1
print *the_local_var^M
[symtab-create] start_subfile: name = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb, name_for_id = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb^M
[symtab-create] start_subfile: name = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb, name_for_id = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb^M
[symtab-create] start_subfile: found existing symtab with name_for_id /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb^M
[symtab-create] start_subfile: name = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb, name_for_id = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb^M
[symtab-create] start_subfile: found existing symtab with name_for_id /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb^M
[symtab-create] start_subfile: name = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.ads, name_for_id = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.ads^M
[symtab-create] start_subfile: name = a-tags.adb, name_for_id = /home/abuild/rpmbuild/BUILD/gcc-7.5.0+r278197/obj-x86_64-suse-linux/gcc/ada/rts/a-tags.adb^M
[symtab-create] start_subfile: name = s-htable.adb, name_for_id = /home/abuild/rpmbuild/BUILD/gcc-7.5.0+r278197/obj-x86_64-suse-linux/gcc/ada/rts/s-htable.adb^M
[symtab-create] start_subfile: name = a-tags.adb, name_for_id = /home/abuild/rpmbuild/BUILD/gcc-7.5.0+r278197/obj-x86_64-suse-linux/gcc/ada/rts/a-tags.adb^M
[symtab-create] start_subfile: found existing symtab with name_for_id /home/abuild/rpmbuild/BUILD/gcc-7.5.0+r278197/obj-x86_64-suse-linux/gcc/ada/rts/a-tags.adb^M
  ...
[symtab-create] start_subfile: name = s-valllu.ads, name_for_id = /home/abuild/rpmbuild/BUILD/gcc-7.5.0+r278197/obj-x86_64-suse-linux/gcc/ada/rts/s-valllu.ads^M
$1 = (n => 2)^M
(gdb) FAIL: gdb.ada/tagged-lookup.exp: only one CU expanded
...
Comment 1 Tom de Vries 2024-03-20 15:31:41 UTC 
Created attachment 15420 [details]
passing gdb.log (on openSUSE Tumbleweed)
Comment 2 Tom de Vries 2024-03-20 15:57:24 UTC 
(In reply to Tom de Vries from comment #1)
> Created attachment 15420 [details]
> passing gdb.log (on openSUSE Tumbleweed)

Grepping reveals pck2.adb only:
...
$ egrep "start_subfile: name = .*(pck|pck2)\.adb" ~/gdb.log
[symtab-create] start_subfile: name = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb, name_for_id = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb
[symtab-create] start_subfile: name = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb, name_for_id = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb
[symtab-create] start_subfile: name = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb, name_for_id = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb
[symtab-create] start_subfile: name = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb, name_for_id = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb
...
Comment 3 Tom de Vries 2024-03-20 16:37:13 UTC 
(In reply to Tom de Vries from comment #2)
> (In reply to Tom de Vries from comment #1)
> > Created attachment 15420 [details]
> > passing gdb.log (on openSUSE Tumbleweed)
> 
> Grepping reveals pck2.adb only:

And in the failing case, no matches at all:
...
$ egrep "start_subfile: name = .*(pck|pck2)\.adb" gdb.log
$
...
Comment 4 Tom de Vries 2024-03-20 16:38:14 UTC 
Created attachment 15421 [details]
failing gdb.log (on openSUSE Leap 15.4)
Comment 5 Tom de Vries 2024-03-21 07:34:11 UTC 
Installed gcc-13 on Leap 15.4.

Passes with gcc-13, fails with gcc-12, 11, 10, 9, 8 and 7.

With gcc-13, we have pck2.adb, pck2.ads, b~foo.adb and b~foo.ads:
...
$ grep "start_subfile: name = .*tagged-lookup" gdb.log | sort -u
[symtab-create] start_subfile: name = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb, name_for_id = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb
[symtab-create] start_subfile: name = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb, name_for_id = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb
[symtab-create] start_subfile: name = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.ads, name_for_id = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.ads
[symtab-create] start_subfile: name = /data/vries/gdb/src/gdb/testsuite/gdb.ada/tagged-lookup/pck2.ads, name_for_id = /data/vries/gdb/src/gdb/testsuite/gdb.ada/tagged-lookup/pck2.ads
...

With gcc-12, we have just b~foo.adb and b~foo.ads:
...
$ grep "start_subfile: name = .*tagged-lookup" gdb.log | sort -u
[symtab-create] start_subfile: name = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb, name_for_id = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb
[symtab-create] start_subfile: name = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.ads, name_for_id = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.ads
...
Comment 6 Tom de Vries 2024-03-21 07:50:43 UTC 
With the patch reverted, we have the following.

With gcc-13, we have pck2.adb, pck2.ads, pck.adb, pck.ads, b~foo.adb, b~foo.ads:
...
$ grep "start_subfile: name = .*tagged-lookup" gdb.log | sort -u
[symtab-create] start_subfile: name = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb, name_for_id = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb
[symtab-create] start_subfile: name = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck.adb, name_for_id = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck.adb
[symtab-create] start_subfile: name = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb, name_for_id = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb
[symtab-create] start_subfile: name = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.ads, name_for_id = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.ads
[symtab-create] start_subfile: name = /data/vries/gdb/src/gdb/testsuite/gdb.ada/tagged-lookup/pck2.ads, name_for_id = /data/vries/gdb/src/gdb/testsuite/gdb.ada/tagged-lookup/pck2.ads
[symtab-create] start_subfile: name = /data/vries/gdb/src/gdb/testsuite/gdb.ada/tagged-lookup/pck.ads, name_for_id = /data/vries/gdb/src/gdb/testsuite/gdb.ada/tagged-lookup/pck.ads
...

With gcc-12, the same:
...
$ grep "start_subfile: name = .*tagged-lookup" gdb.log | sort -u
[symtab-create] start_subfile: name = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb, name_for_id = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck2.adb
[symtab-create] start_subfile: name = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck.adb, name_for_id = /data/vries/gdb/binutils-gdb.git/gdb/testsuite/gdb.ada/tagged-lookup/pck.adb
[symtab-create] start_subfile: name = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb, name_for_id = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.adb
[symtab-create] start_subfile: name = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.ads, name_for_id = /data/vries/gdb/leap-15-4/build/gdb/testsuite/outputs/gdb.ada/tagged-lookup/b~foo.ads
[symtab-create] start_subfile: name = /data/vries/gdb/src/gdb/testsuite/gdb.ada/tagged-lookup/pck2.ads, name_for_id = /data/vries/gdb/src/gdb/testsuite/gdb.ada/tagged-lookup/pck2.ads
[symtab-create] start_subfile: name = /data/vries/gdb/src/gdb/testsuite/gdb.ada/tagged-lookup/pck.ads, name_for_id = /data/vries/gdb/src/gdb/testsuite/gdb.ada/tagged-lookup/pck.ads
...
Comment 7 Tom de Vries 2024-03-21 08:00:49 UTC 
Tentative patch:
...
diff --git a/gdb/testsuite/gdb.ada/tagged-lookup.exp b/gdb/testsuite/gdb.ada/tagged-lookup.exp
index 4bc088ba8d5..3803319c505 100644
--- a/gdb/testsuite/gdb.ada/tagged-lookup.exp
+++ b/gdb/testsuite/gdb.ada/tagged-lookup.exp
@@ -52,7 +52,7 @@ gdb_test_multiple "print *the_local_var" "only one CU expanded" -lbl {
 	exp_continue
     }
     -re -wrap ".* = \\\(n => $decimal\\\)" {
-	if {$found_pck + $found_pck2 == 1} {
+	if {$found_pck + $found_pck2 <= 1} {
 	    pass $gdb_test_name
 	} else {
 	    fail $gdb_test_name
...

It detects the fail with fix reverted for gcc-12 and gcc-13, though not before.

And it make the test pass for gcc-12 and gcc-13 with the fix present.
Comment 9 Sourceware Commits 2024-03-25 14:28:46 UTC 
The master branch has been updated by Tom de Vries <vries@sourceware.org>:

https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=874f4887f0361e953f7098459ae60a5f894d362a

commit 874f4887f0361e953f7098459ae60a5f894d362a
Author: Tom de Vries <tdevries@suse.de>
Date:   Mon Mar 25 15:28:35 2024 +0100

    [gdb/testsuite] Fix gdb.ada/tagged-lookup.exp with gcc <= 12
    
    With gcc 13, test-case gdb.ada/tagged-lookup.exp passes for me, but with gcc
    12, I get:
    ...
    (gdb) set debug symtab-create 1^M
    (gdb) print *the_local_var^M
      ...
    $1 = (n => 2)^M
    (gdb) FAIL: gdb.ada/tagged-lookup.exp: only one CU expanded
    ...
    
    The problem is that this fails:
    ...
        -re -wrap ".* = \\\(n => $decimal\\\)" {
            if {$found_pck + $found_pck2 == 1} {
                pass $gdb_test_name
            } else {
                fail $gdb_test_name
            }
    ...
    because $found_pck == 0 and $found_pck2 == 0.
    
    Indeed, with gcc 13 we have:
    ...
    $ grep "start_subfile: name = .*/tagged-lookup/" gdb.log | sed 's%.*/%%'
    b~foo.adb
    b~foo.adb
    b~foo.adb
    b~foo.ads
    pck2.adb
    pck2.adb
    pck2.ads
    pck2.adb
    pck2.ads
    ...
    and with gcc 12:
    ...
    $ grep "start_subfile: name = .*/tagged-lookup/" gdb.log | sed 's%.*/%%'
    b~foo.adb
    b~foo.adb
    b~foo.adb
    b~foo.ads
    ...
    
    Fix this by checking for "$found_pck + $found_pck2 <= 1" instead.
    
    Tested on x86_64-linux.
    
    Approved-By: Tom Tromey <tom@tromey.com>
    
    PR testsuite/31514
    Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=31514
Comment 10 Tom de Vries 2024-03-25 14:29:14 UTC 
Fixed.