[PATCH] [gdb/testsuite] Fix linespec ambiguity in gdb.base/longjmp.exp

[PATCH] [gdb/testsuite] Fix linespec ambiguity in gdb.base/longjmp.exp

[Tom de Vries]

PR testsuite/30103 reports the following failure on aarch64-linux
(ubuntu 22.04):
...
(gdb) PASS: gdb.base/longjmp.exp: with_probes=0: pattern 1: next to longjmp
next
warning: Breakpoint address adjusted from 0x83dc305fef755015 to \
  0xffdc305fef755015.
Warning:
Cannot insert breakpoint 0.
Cannot access memory at address 0xffdc305fef755015

__libc_siglongjmp (env=0xaaaaaaab1018 <env>, val=1) at ./setjmp/longjmp.c:30
30	}
(gdb) KFAIL: gdb.base/longjmp.exp: with_probes=0: pattern 1: gdb/26967 \
  (PRMS: next over longjmp)
delete breakpoints
Delete all breakpoints? (y or n) y
(gdb) info breakpoints
No breakpoints or watchpoints.
(gdb) break 63
No line 63 in the current file.
Make breakpoint pending on future shared library load? (y or [n]) n
(gdb) FAIL: gdb.base/longjmp.exp: with_probes=0: pattern 2: setup: breakpoint \
  at pattern start (got interactive prompt)
...

The test-case intends to set the breakpoint on line number 63 in
gdb.base/longjmp.c.

It tries to do so by specifying "break 63", which specifies a line in the
"current source file".

Due to the KFAIL PR, gdb stopped in __libc_siglongjmp, and because of presence
of debug info, the "current source file" becomes glibc's ./setjmp/longjmp.c.

Consequently, setting the breakpoint fails.

Fix this by adding a $subdir/$srcfile: prefix to the breakpoint linespecs.

I've managed to reproduce the FAIL on x86_64/-m32, by installing the
glibc-32bit-debuginfo package.  This allowed me to confirm the "current source
file" that is used:
...
(gdb) KFAIL: gdb.base/longjmp.exp: with_probes=0: pattern 1: gdb/26967 \
  (PRMS: next over longjmp)
info source^M
Current source file is ../setjmp/longjmp.c^M
...

Tested on x86_64-linux, target boards unix/{-m64,-m32}.

Reported-By: Luis Machado <luis.machado@arm.com>

PR testsuite/30103
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=30103
---
 gdb/testsuite/gdb.base/longjmp.exp | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/gdb/testsuite/gdb.base/longjmp.exp b/gdb/testsuite/gdb.base/longjmp.exp
index 96d9c1c8059..f74891aa7ca 100644
--- a/gdb/testsuite/gdb.base/longjmp.exp
+++ b/gdb/testsuite/gdb.base/longjmp.exp
@@ -82,6 +82,9 @@ proc do_test { with_probes } {
 	gdb_assert { !$have_longjmp_probe }
     }
 
+    # When using these line numbers in break linespecs, prefix each of these
+    # with "$subdir/$srcfile:" to avoid referring to a glibc file when stopped
+    # in __libc_siglongjmp or similar.
     set bp_miss_step_1 [gdb_get_line_number "miss_step_1"]
     set bp_miss_step_2 [gdb_get_line_number "miss_step_2"]
 
@@ -104,13 +107,13 @@ proc do_test { with_probes } {
 	with_test_prefix setup {
 	    delete_breakpoints
 
-	    gdb_test "break $bp_start_test_1" \
+	    gdb_test "break $::subdir/$::srcfile:$bp_start_test_1" \
 		"Breakpoint.*at.* file .*$::srcfile, line.*$bp_start_test_1.*" \
 		"breakpoint at pattern start"
 	    gdb_test "continue" "patt1.*" "continue to breakpoint at pattern start"
 
 	    # set safe-net break
-	    gdb_test "break $bp_miss_step_1" \
+	    gdb_test "break $::subdir/$::srcfile:$bp_miss_step_1" \
 		"Breakpoint.*at.* file .*$::srcfile, line.*$bp_miss_step_1.*" \
 		"breakpoint at safety net"
 	}
@@ -151,13 +154,13 @@ proc do_test { with_probes } {
 	with_test_prefix setup {
 	    delete_breakpoints
 
-	    gdb_test "break $bp_start_test_2" \
+	    gdb_test "break $::subdir/$::srcfile:$bp_start_test_2" \
 		"Breakpoint.*at.* file .*$::srcfile, line.*$bp_start_test_2.*" \
 		"breakpoint at pattern start"
 	    gdb_test "continue" "patt2.*" "continue to breakpoint at pattern start"
 
 	    # set safe-net break
-	    gdb_test "break $bp_miss_step_2" \
+	    gdb_test "break $::subdir/$::srcfile:$bp_miss_step_2" \
 		"Breakpoint.*at.* file .*$::srcfile, line.*$bp_miss_step_2.*" \
 		"breakpoint at safety net"
 	}
@@ -198,7 +201,7 @@ proc do_test { with_probes } {
 	with_test_prefix setup {
 	    delete_breakpoints
 
-	    gdb_test "break $bp_start_test_3" \
+	    gdb_test "break $::subdir/$::srcfile:$bp_start_test_3" \
 		"Breakpoint.*at.* file .*$::srcfile, line.*$bp_start_test_3.*" \
 		"breakpoint at pattern start"
 	    gdb_test "continue" "patt3.*" "continue to breakpoint at pattern start"

base-commit: fe8cdc8ec145a166414fc375cf2cb65d9a8085a1
-- 
2.35.3

Re: [PATCH] [gdb/testsuite] Fix linespec ambiguity in gdb.base/longjmp.exp

[Tom Tromey]

>>>>> "Tom" == Tom de Vries via Gdb-patches <gdb-patches@sourceware.org> writes:

Tom> Due to the KFAIL PR, gdb stopped in __libc_siglongjmp, and because of presence
Tom> of debug info, the "current source file" becomes glibc's ./setjmp/longjmp.c.

Tom> Consequently, setting the breakpoint fails.

Tom> Fix this by adding a $subdir/$srcfile: prefix to the breakpoint linespecs.

Makes sense to me.

Reviewed-By: Tom Tromey <tom@tromey.com>

Tom

Re: [PATCH] [gdb/testsuite] Fix linespec ambiguity in gdb.base/longjmp.exp

[Luis Machado]

On 2/10/23 10:57, Tom de Vries wrote:
> PR testsuite/30103 reports the following failure on aarch64-linux
> (ubuntu 22.04):
> ...
> (gdb) PASS: gdb.base/longjmp.exp: with_probes=0: pattern 1: next to longjmp
> next
> warning: Breakpoint address adjusted from 0x83dc305fef755015 to \
>    0xffdc305fef755015.
> Warning:
> Cannot insert breakpoint 0.
> Cannot access memory at address 0xffdc305fef755015
> 
> __libc_siglongjmp (env=0xaaaaaaab1018 <env>, val=1) at ./setjmp/longjmp.c:30
> 30	}
> (gdb) KFAIL: gdb.base/longjmp.exp: with_probes=0: pattern 1: gdb/26967 \
>    (PRMS: next over longjmp)
> delete breakpoints
> Delete all breakpoints? (y or n) y
> (gdb) info breakpoints
> No breakpoints or watchpoints.
> (gdb) break 63
> No line 63 in the current file.
> Make breakpoint pending on future shared library load? (y or [n]) n
> (gdb) FAIL: gdb.base/longjmp.exp: with_probes=0: pattern 2: setup: breakpoint \
>    at pattern start (got interactive prompt)
> ...
> 
> The test-case intends to set the breakpoint on line number 63 in
> gdb.base/longjmp.c.
> 
> It tries to do so by specifying "break 63", which specifies a line in the
> "current source file".
> 
> Due to the KFAIL PR, gdb stopped in __libc_siglongjmp, and because of presence
> of debug info, the "current source file" becomes glibc's ./setjmp/longjmp.c.
> 
> Consequently, setting the breakpoint fails.
> 
> Fix this by adding a $subdir/$srcfile: prefix to the breakpoint linespecs.
> 
> I've managed to reproduce the FAIL on x86_64/-m32, by installing the
> glibc-32bit-debuginfo package.  This allowed me to confirm the "current source
> file" that is used:
> ...
> (gdb) KFAIL: gdb.base/longjmp.exp: with_probes=0: pattern 1: gdb/26967 \
>    (PRMS: next over longjmp)
> info source^M
> Current source file is ../setjmp/longjmp.c^M
> ...
> 
> Tested on x86_64-linux, target boards unix/{-m64,-m32}.
> 
> Reported-By: Luis Machado <luis.machado@arm.com>
> 
> PR testsuite/30103
> Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=30103
> ---
>   gdb/testsuite/gdb.base/longjmp.exp | 13 ++++++++-----
>   1 file changed, 8 insertions(+), 5 deletions(-)
> 
> diff --git a/gdb/testsuite/gdb.base/longjmp.exp b/gdb/testsuite/gdb.base/longjmp.exp
> index 96d9c1c8059..f74891aa7ca 100644
> --- a/gdb/testsuite/gdb.base/longjmp.exp
> +++ b/gdb/testsuite/gdb.base/longjmp.exp
> @@ -82,6 +82,9 @@ proc do_test { with_probes } {
>   	gdb_assert { !$have_longjmp_probe }
>       }
>   
> +    # When using these line numbers in break linespecs, prefix each of these
> +    # with "$subdir/$srcfile:" to avoid referring to a glibc file when stopped
> +    # in __libc_siglongjmp or similar.
>       set bp_miss_step_1 [gdb_get_line_number "miss_step_1"]
>       set bp_miss_step_2 [gdb_get_line_number "miss_step_2"]
>   
> @@ -104,13 +107,13 @@ proc do_test { with_probes } {
>   	with_test_prefix setup {
>   	    delete_breakpoints
>   
> -	    gdb_test "break $bp_start_test_1" \
> +	    gdb_test "break $::subdir/$::srcfile:$bp_start_test_1" \
>   		"Breakpoint.*at.* file .*$::srcfile, line.*$bp_start_test_1.*" \
>   		"breakpoint at pattern start"
>   	    gdb_test "continue" "patt1.*" "continue to breakpoint at pattern start"
>   
>   	    # set safe-net break
> -	    gdb_test "break $bp_miss_step_1" \
> +	    gdb_test "break $::subdir/$::srcfile:$bp_miss_step_1" \
>   		"Breakpoint.*at.* file .*$::srcfile, line.*$bp_miss_step_1.*" \
>   		"breakpoint at safety net"
>   	}
> @@ -151,13 +154,13 @@ proc do_test { with_probes } {
>   	with_test_prefix setup {
>   	    delete_breakpoints
>   
> -	    gdb_test "break $bp_start_test_2" \
> +	    gdb_test "break $::subdir/$::srcfile:$bp_start_test_2" \
>   		"Breakpoint.*at.* file .*$::srcfile, line.*$bp_start_test_2.*" \
>   		"breakpoint at pattern start"
>   	    gdb_test "continue" "patt2.*" "continue to breakpoint at pattern start"
>   
>   	    # set safe-net break
> -	    gdb_test "break $bp_miss_step_2" \
> +	    gdb_test "break $::subdir/$::srcfile:$bp_miss_step_2" \
>   		"Breakpoint.*at.* file .*$::srcfile, line.*$bp_miss_step_2.*" \
>   		"breakpoint at safety net"
>   	}
> @@ -198,7 +201,7 @@ proc do_test { with_probes } {
>   	with_test_prefix setup {
>   	    delete_breakpoints
>   
> -	    gdb_test "break $bp_start_test_3" \
> +	    gdb_test "break $::subdir/$::srcfile:$bp_start_test_3" \
>   		"Breakpoint.*at.* file .*$::srcfile, line.*$bp_start_test_3.*" \
>   		"breakpoint at pattern start"
>   	    gdb_test "continue" "patt3.*" "continue to breakpoint at pattern start"
> 
> base-commit: fe8cdc8ec145a166414fc375cf2cb65d9a8085a1

Thanks Tom. I gave this a try and it fixes the FAIL's I was seeing.

Reviewed-By: Luis Machado <luis.machado@arm.com>
Tested-By: Luis Machado <luis.machado@arm.com>

Re: [PATCH] [gdb/testsuite] Fix linespec ambiguity in gdb.base/longjmp.exp

[Tom de Vries]

On 2/10/23 15:58, Luis Machado wrote:
> Thanks Tom. I gave this a try and it fixes the FAIL's I was seeing.
> 
> Reviewed-By: Luis Machado <luis.machado@arm.com>
> Tested-By: Luis Machado <luis.machado@arm.com>

Thanks for confirming.

Unfortunately, I was once more too trigger-happy, so the commit is 
missing your tested-by tag.

Thanks,
- Tom