From: Michael Snyder <msnyder@cygnus.com>
To: Jim Blandy <jimb@cygnus.com>
Cc: gdb-patches@sources.redhat.com
Subject: Re: RFA: tolerate unavailable struct return values
Date: Thu, 29 Nov 2001 14:19:00 -0000 [thread overview]
Message-ID: <3C06B4CA.11E3E3CE@cygnus.com> (raw)
In-Reply-To: <20011129220913.2D72A5E9D8@zwingli.cygnus.com>
Jim Blandy wrote:
>
> On some architectures, it's impossible for GDB to find structs
> returned by value. These shouldn't be failures. Should they be
> passes?
I kinda like the idea of their being xfails. The functionality
is not there, it's just not a surprise.
>
> 2001-11-29 Jim Blandy <jimb@redhat.com>
>
> If GDB says it can't find the struct the function returned, treat
> that as an expected failure.
> * gdb.base/structs.exp (call_struct_func): New function.
> (do_function_calls): Use call_struct_func to call the functions
> returning structs.
>
> Index: gdb/testsuite/gdb.base/structs.exp
> ===================================================================
> RCS file: /cvs/cvsfiles/devo/gdb/testsuite/gdb.base/structs.exp,v
> retrieving revision 1.17
> diff -c -r1.17 structs.exp
> *** gdb/testsuite/gdb.base/structs.exp 2001/08/09 18:02:02 1.17
> --- gdb/testsuite/gdb.base/structs.exp 2001/11/29 17:46:25
> ***************
> *** 54,59 ****
> --- 54,85 ----
> continue
> }
>
> +
> + # Call FUNC with no arguments, and expect to see the regexp RESULT in
> + # the output. If we get back the error message "Function return value
> + # unknown", call that an expected failure; on some architectures, it's
> + # impossible to find structs returned by value reliably.
> + proc call_struct_func { func result } {
> + global gdb_prompt
> +
> + set command "p ${func}()"
> + send_gdb "${command}\n"
> + gdb_expect {
> + -re "$result\[\r\n\]+$gdb_prompt $" {
> + pass "$command"
> + }
> + -re "Function return value unknown.\[\r\n\]+$gdb_prompt $" {
> + xfail "$command"
> + }
> + -re "$gdb_prompt $" {
> + fail "$command"
> + }
> + timeout {
> + fail "$command (timeout)"
> + }
> + }
> + }
> +
> # FIXME: Before calling this proc, we should probably verify that
> # we can call inferior functions and get a valid integral value
> # returned.
> ***************
> *** 67,85 ****
> global gdb_prompt
>
> # First, call the "fun" functions and examine the value they return.
> ! gdb_test "p fun1()" " = {a = 49 '1'}"
> ! gdb_test "p fun2()" " = {a = 97 'a', b = 98 'b'}"
> ! gdb_test "p fun3()" " = {a = 65 'A', b = 66 'B', c = 67 'C'}"
> ! gdb_test "p fun4()" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4'}"
> ! gdb_test "p fun5()" " = {a = 97 'a', b = 98 'b', c = 99 'c', d = 100 'd', e = 101 'e'}"
> ! gdb_test "p fun6()" " = {a = 65 'A', b = 66 'B', c = 67 'C', d = 68 'D', e = 69 'E', f = 70 'F'}"
> ! gdb_test "p fun7()" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4', e = 53 '5', f = 54 '6', g = 55 '7'}"
> ! gdb_test "p fun8()" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4', e = 53 '5', f = 54 '6', g = 55 '7', h = 56 '8'}"
> ! gdb_test "p fun9()" " = {a = 97 'a', b = 98 'b', c = 99 'c', d = 100 'd', e = 101 'e', f = 102 'f', g = 103 'g', h = 104 'h', i = 105 'i'}"
> ! gdb_test "p fun10()" " = {a = 65 'A', b = 66 'B', c = 67 'C', d = 68 'D', e = 69 'E', f = 70 'F', g = 71 'G', h = 72 'H', i = 73 'I', j = 74 'J'}"
> ! gdb_test "p fun11()" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4', e = 53 '5', f = 54 '6', g = 55 '7', h = 56 '8', i = 57 '9', j = 65 'A', k = 66 'B'}"
> ! gdb_test "p fun12()" " = {a = 65 'A', b = 66 'B', c = 67 'C', d = 68 'D', e = 69 'E', f = 70 'F', g = 71 'G', h = 72 'H', i = 73 'I', j = 74 'J', k = 75 'K', l = 76 'L'}"
> ! gdb_test "p fun16()" " = {a = 97 'a', b = 98 'b', c = 99 'c', d = 100 'd', e = 101 'e', f = 102 'f', g = 103 'g', h = 104 'h', i = 105 'i', j = 106 'j', k = 107 'k', l = 108 'l', m = 109 'm', n = 110 'n', o = 111 'o', p = 112 'p'}"
>
> # Now call the Fun functions to set the L* variables. This
> # tests that gdb properly passes structures to functions.
> --- 93,111 ----
> global gdb_prompt
>
> # First, call the "fun" functions and examine the value they return.
> ! call_struct_func "fun1" " = {a = 49 '1'}"
> ! call_struct_func "fun2" " = {a = 97 'a', b = 98 'b'}"
> ! call_struct_func "fun3" " = {a = 65 'A', b = 66 'B', c = 67 'C'}"
> ! call_struct_func "fun4" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4'}"
> ! call_struct_func "fun5" " = {a = 97 'a', b = 98 'b', c = 99 'c', d = 100 'd', e = 101 'e'}"
> ! call_struct_func "fun6" " = {a = 65 'A', b = 66 'B', c = 67 'C', d = 68 'D', e = 69 'E', f = 70 'F'}"
> ! call_struct_func "fun7" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4', e = 53 '5', f = 54 '6', g = 55 '7'}"
> ! call_struct_func "fun8" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4', e = 53 '5', f = 54 '6', g = 55 '7', h = 56 '8'}"
> ! call_struct_func "fun9" " = {a = 97 'a', b = 98 'b', c = 99 'c', d = 100 'd', e = 101 'e', f = 102 'f', g = 103 'g', h = 104 'h', i = 105 'i'}"
> ! call_struct_func "fun10" " = {a = 65 'A', b = 66 'B', c = 67 'C', d = 68 'D', e = 69 'E', f = 70 'F', g = 71 'G', h = 72 'H', i = 73 'I', j = 74 'J'}"
> ! call_struct_func "fun11" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4', e = 53 '5', f = 54 '6', g = 55 '7', h = 56 '8', i = 57 '9', j = 65 'A', k = 66 'B'}"
> ! call_struct_func "fun12" " = {a = 65 'A', b = 66 'B', c = 67 'C', d = 68 'D', e = 69 'E', f = 70 'F', g = 71 'G', h = 72 'H', i = 73 'I', j = 74 'J', k = 75 'K', l = 76 'L'}"
> ! call_struct_func "fun16" " = {a = 97 'a', b = 98 'b', c = 99 'c', d = 100 'd', e = 101 'e', f = 102 'f', g = 103 'g', h = 104 'h', i = 105 'i', j = 106 'j', k = 107 'k', l = 108 'l', m = 109 'm', n = 110 'n', o = 111 'o', p = 112 'p'}"
>
> # Now call the Fun functions to set the L* variables. This
> # tests that gdb properly passes structures to functions.
WARNING: multiple messages have this Message-ID
From: Michael Snyder <msnyder@cygnus.com>
To: Jim Blandy <jimb@cygnus.com>
Cc: gdb-patches@sources.redhat.com
Subject: Re: RFA: tolerate unavailable struct return values
Date: Mon, 19 Nov 2001 22:30:00 -0000 [thread overview]
Message-ID: <3C06B4CA.11E3E3CE@cygnus.com> (raw)
Message-ID: <20011119223000.0IL7-fQJQRfN-QHZp-551D_2Vt0SRIyiYOrxVpWqqwY@z> (raw)
In-Reply-To: <20011129220913.2D72A5E9D8@zwingli.cygnus.com>
Jim Blandy wrote:
>
> On some architectures, it's impossible for GDB to find structs
> returned by value. These shouldn't be failures. Should they be
> passes?
I kinda like the idea of their being xfails. The functionality
is not there, it's just not a surprise.
>
> 2001-11-29 Jim Blandy <jimb@redhat.com>
>
> If GDB says it can't find the struct the function returned, treat
> that as an expected failure.
> * gdb.base/structs.exp (call_struct_func): New function.
> (do_function_calls): Use call_struct_func to call the functions
> returning structs.
>
> Index: gdb/testsuite/gdb.base/structs.exp
> ===================================================================
> RCS file: /cvs/cvsfiles/devo/gdb/testsuite/gdb.base/structs.exp,v
> retrieving revision 1.17
> diff -c -r1.17 structs.exp
> *** gdb/testsuite/gdb.base/structs.exp 2001/08/09 18:02:02 1.17
> --- gdb/testsuite/gdb.base/structs.exp 2001/11/29 17:46:25
> ***************
> *** 54,59 ****
> --- 54,85 ----
> continue
> }
>
> +
> + # Call FUNC with no arguments, and expect to see the regexp RESULT in
> + # the output. If we get back the error message "Function return value
> + # unknown", call that an expected failure; on some architectures, it's
> + # impossible to find structs returned by value reliably.
> + proc call_struct_func { func result } {
> + global gdb_prompt
> +
> + set command "p ${func}()"
> + send_gdb "${command}\n"
> + gdb_expect {
> + -re "$result\[\r\n\]+$gdb_prompt $" {
> + pass "$command"
> + }
> + -re "Function return value unknown.\[\r\n\]+$gdb_prompt $" {
> + xfail "$command"
> + }
> + -re "$gdb_prompt $" {
> + fail "$command"
> + }
> + timeout {
> + fail "$command (timeout)"
> + }
> + }
> + }
> +
> # FIXME: Before calling this proc, we should probably verify that
> # we can call inferior functions and get a valid integral value
> # returned.
> ***************
> *** 67,85 ****
> global gdb_prompt
>
> # First, call the "fun" functions and examine the value they return.
> ! gdb_test "p fun1()" " = {a = 49 '1'}"
> ! gdb_test "p fun2()" " = {a = 97 'a', b = 98 'b'}"
> ! gdb_test "p fun3()" " = {a = 65 'A', b = 66 'B', c = 67 'C'}"
> ! gdb_test "p fun4()" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4'}"
> ! gdb_test "p fun5()" " = {a = 97 'a', b = 98 'b', c = 99 'c', d = 100 'd', e = 101 'e'}"
> ! gdb_test "p fun6()" " = {a = 65 'A', b = 66 'B', c = 67 'C', d = 68 'D', e = 69 'E', f = 70 'F'}"
> ! gdb_test "p fun7()" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4', e = 53 '5', f = 54 '6', g = 55 '7'}"
> ! gdb_test "p fun8()" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4', e = 53 '5', f = 54 '6', g = 55 '7', h = 56 '8'}"
> ! gdb_test "p fun9()" " = {a = 97 'a', b = 98 'b', c = 99 'c', d = 100 'd', e = 101 'e', f = 102 'f', g = 103 'g', h = 104 'h', i = 105 'i'}"
> ! gdb_test "p fun10()" " = {a = 65 'A', b = 66 'B', c = 67 'C', d = 68 'D', e = 69 'E', f = 70 'F', g = 71 'G', h = 72 'H', i = 73 'I', j = 74 'J'}"
> ! gdb_test "p fun11()" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4', e = 53 '5', f = 54 '6', g = 55 '7', h = 56 '8', i = 57 '9', j = 65 'A', k = 66 'B'}"
> ! gdb_test "p fun12()" " = {a = 65 'A', b = 66 'B', c = 67 'C', d = 68 'D', e = 69 'E', f = 70 'F', g = 71 'G', h = 72 'H', i = 73 'I', j = 74 'J', k = 75 'K', l = 76 'L'}"
> ! gdb_test "p fun16()" " = {a = 97 'a', b = 98 'b', c = 99 'c', d = 100 'd', e = 101 'e', f = 102 'f', g = 103 'g', h = 104 'h', i = 105 'i', j = 106 'j', k = 107 'k', l = 108 'l', m = 109 'm', n = 110 'n', o = 111 'o', p = 112 'p'}"
>
> # Now call the Fun functions to set the L* variables. This
> # tests that gdb properly passes structures to functions.
> --- 93,111 ----
> global gdb_prompt
>
> # First, call the "fun" functions and examine the value they return.
> ! call_struct_func "fun1" " = {a = 49 '1'}"
> ! call_struct_func "fun2" " = {a = 97 'a', b = 98 'b'}"
> ! call_struct_func "fun3" " = {a = 65 'A', b = 66 'B', c = 67 'C'}"
> ! call_struct_func "fun4" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4'}"
> ! call_struct_func "fun5" " = {a = 97 'a', b = 98 'b', c = 99 'c', d = 100 'd', e = 101 'e'}"
> ! call_struct_func "fun6" " = {a = 65 'A', b = 66 'B', c = 67 'C', d = 68 'D', e = 69 'E', f = 70 'F'}"
> ! call_struct_func "fun7" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4', e = 53 '5', f = 54 '6', g = 55 '7'}"
> ! call_struct_func "fun8" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4', e = 53 '5', f = 54 '6', g = 55 '7', h = 56 '8'}"
> ! call_struct_func "fun9" " = {a = 97 'a', b = 98 'b', c = 99 'c', d = 100 'd', e = 101 'e', f = 102 'f', g = 103 'g', h = 104 'h', i = 105 'i'}"
> ! call_struct_func "fun10" " = {a = 65 'A', b = 66 'B', c = 67 'C', d = 68 'D', e = 69 'E', f = 70 'F', g = 71 'G', h = 72 'H', i = 73 'I', j = 74 'J'}"
> ! call_struct_func "fun11" " = {a = 49 '1', b = 50 '2', c = 51 '3', d = 52 '4', e = 53 '5', f = 54 '6', g = 55 '7', h = 56 '8', i = 57 '9', j = 65 'A', k = 66 'B'}"
> ! call_struct_func "fun12" " = {a = 65 'A', b = 66 'B', c = 67 'C', d = 68 'D', e = 69 'E', f = 70 'F', g = 71 'G', h = 72 'H', i = 73 'I', j = 74 'J', k = 75 'K', l = 76 'L'}"
> ! call_struct_func "fun16" " = {a = 97 'a', b = 98 'b', c = 99 'c', d = 100 'd', e = 101 'e', f = 102 'f', g = 103 'g', h = 104 'h', i = 105 'i', j = 106 'j', k = 107 'k', l = 108 'l', m = 109 'm', n = 110 'n', o = 111 'o', p = 112 'p'}"
>
> # Now call the Fun functions to set the L* variables. This
> # tests that gdb properly passes structures to functions.
next prev parent reply other threads:[~2001-11-29 14:19 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2001-11-19 20:48 Jim Blandy
2001-11-29 14:08 ` Jim Blandy
2001-11-29 14:19 ` Michael Snyder [this message]
2001-11-19 22:30 ` Michael Snyder
2001-11-29 14:37 ` Daniel Jacobowitz
2001-11-20 7:19 ` Daniel Jacobowitz
2001-11-30 12:48 ` Jim Blandy
2001-11-23 13:51 ` Jim Blandy
2001-11-30 13:33 ` Daniel Jacobowitz
2001-11-24 10:23 ` Daniel Jacobowitz
2001-11-30 13:51 ` Michael Snyder
2001-11-24 23:01 ` Michael Snyder
2001-12-12 11:27 ` Elena Zannoni
2001-12-17 15:09 ` Jim Blandy
2001-12-18 9:35 ` Elena Zannoni
[not found] ` <87667t2loj.fsf@creche.redhat.com>
2001-11-29 22:41 ` Jim Blandy
2001-11-21 8:07 ` Jim Blandy
2001-11-29 22:46 Jim Blandy
2001-11-21 13:10 ` Jim Blandy
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=3C06B4CA.11E3E3CE@cygnus.com \
--to=msnyder@cygnus.com \
--cc=gdb-patches@sources.redhat.com \
--cc=jimb@cygnus.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