gdb: fix frame passed to gdbarch_value_to_register in value_assign
Checks
Commit Message
Commit 78f2fd84e83 ("gdb: remove VALUE_REGNUM, add value::regnum")
introduced an unexpected change in value_assign. It replaced
`get_prev_frame_always (next_frame)` with `next_frame`in the call to
gdbarch_value_to_register.
This is the result of a merge error, since I previously had a patch to
change gdbarch_value_to_register to take the next frame, and later
decided to drop it. Revert that change.
Add a test based on the DWARF assembler to expose the problem and test
the fix. I also tested on ppc64le to make sure the problem reported in
PR 31231 was fixed.
Change-Id: Ib8b851287ac27a4b2e386f7b680cf65865e6aee6
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=31231
---
.../assign-variable-value-to-register.exp | 86 +++++++++++++++++++
gdb/valops.c | 3 +-
2 files changed, 88 insertions(+), 1 deletion(-)
create mode 100644 gdb/testsuite/gdb.dwarf2/assign-variable-value-to-register.exp
base-commit: fa87f8e1958b8937047e3d5fcfd8c265745710ae
Comments
On 1/11/24 17:51, Simon Marchi wrote:
> Commit 78f2fd84e83 ("gdb: remove VALUE_REGNUM, add value::regnum")
> introduced an unexpected change in value_assign. It replaced
> `get_prev_frame_always (next_frame)` with `next_frame`in the call to
> gdbarch_value_to_register.
>
> This is the result of a merge error, since I previously had a patch to
> change gdbarch_value_to_register to take the next frame, and later
> decided to drop it. Revert that change.
>
> Add a test based on the DWARF assembler to expose the problem and test
> the fix. I also tested on ppc64le to make sure the problem reported in
> PR 31231 was fixed.
>
I applied the patch on x86_64, ran the test-case, observed the internal
error, rebuild gdb, and observed it pass.
Only one nit, I get:
...
DUPLICATE: gdb.dwarf2/assign-variable-value-to-register.exp: p foo
...
Tested-by: Tom de Vries <tdevries@suse.de>
Thanks,
- Tom
> Change-Id: Ib8b851287ac27a4b2e386f7b680cf65865e6aee6
> Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=31231
> ---
> .../assign-variable-value-to-register.exp | 86 +++++++++++++++++++
> gdb/valops.c | 3 +-
> 2 files changed, 88 insertions(+), 1 deletion(-)
> create mode 100644 gdb/testsuite/gdb.dwarf2/assign-variable-value-to-register.exp
>
> diff --git a/gdb/testsuite/gdb.dwarf2/assign-variable-value-to-register.exp b/gdb/testsuite/gdb.dwarf2/assign-variable-value-to-register.exp
> new file mode 100644
> index 000000000000..770f665ca959
> --- /dev/null
> +++ b/gdb/testsuite/gdb.dwarf2/assign-variable-value-to-register.exp
> @@ -0,0 +1,86 @@
> +# Copyright 2024 Free Software Foundation, Inc.
> +
> +# This program is free software; you can redistribute it and/or modify
> +# it under the terms of the GNU General Public License as published by
> +# the Free Software Foundation; either version 3 of the License, or
> +# (at your option) any later version.
> +#
> +# This program is distributed in the hope that it will be useful,
> +# but WITHOUT ANY WARRANTY; without even the implied warranty of
> +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> +# GNU General Public License for more details.
> +#
> +# You should have received a copy of the GNU General Public License
> +# along with this program. If not, see <http://www.gnu.org/licenses/>.
> +
> +# Test writing to a variable that is in a register that causes
> +# gdbarch_value_to_register to be called. This was introduced to test
> +# PR 31231 (https://sourceware.org/bugzilla/show_bug.cgi?id=31231).
> +
> +# This test uses a hard-coded x86-64 DWARF register number, so restrict it to
> +# x86-64.
> +if { ![is_x86_64_m64_target] } {
> + unsupported "unsupported architecture"
> +}
> +
> +load_lib dwarf.exp
> +
> +standard_testfile main.c -dw.S
> +set dwarf_file [standard_output_file $srcfile2]
> +
> +set st0_dwarf_regnum 33
> +
> +Dwarf::assemble $dwarf_file {
> + get_func_info main
> +
> + cu {} {
> + DW_TAG_compile_unit {} {
> + declare_labels int_label float_label
> +
> + int_label: DW_TAG_base_type {
> + { DW_AT_name int }
> + { DW_AT_byte_size 4 DW_FORM_udata }
> + { DW_AT_encoding @DW_ATE_signed }
> + }
> +
> + float_label: DW_TAG_base_type {
> + { DW_AT_name float }
> + { DW_AT_byte_size 4 DW_FORM_udata }
> + { DW_AT_encoding @DW_ATE_float }
> + }
> +
> + DW_TAG_subprogram {
> + { DW_AT_name main }
> + { DW_AT_low_pc $main_start DW_FORM_addr }
> + { DW_AT_high_pc $main_end DW_FORM_addr }
> + { DW_AT_type :$int_label }
> + } {
> + DW_TAG_variable {
> + { DW_AT_name foo }
> + { DW_AT_type :$float_label }
> + { DW_AT_location {
> + DW_OP_regx $::st0_dwarf_regnum
> + } SPECIAL_expr }
> + }
> + }
> + }
> + }
> +}
> +
> +if { [prepare_for_testing "failed to prepare" $testfile \
> + [list $srcfile $dwarf_file] {nodebug}] } {
> + return
> +}
> +
> +if { ![runto_main] } {
> + return
> +}
> +
> +# st0 is expected to be initialized to 0.
> +gdb_test "p foo" " = 0"
> +
> +# Set foo. This used to cause an internal error.
> +gdb_test_no_output "set foo = 1234"
> +
> +# Confirm the value.
> +gdb_test "p foo" " = 1234"
> diff --git a/gdb/valops.c b/gdb/valops.c
> index 16cdf1f45530..147793400d90 100644
> --- a/gdb/valops.c
> +++ b/gdb/valops.c
> @@ -1244,7 +1244,8 @@ value_assign (struct value *toval, struct value *fromval)
> /* If TOVAL is a special machine register requiring
> conversion of program values to a special raw
> format. */
> - gdbarch_value_to_register (gdbarch, next_frame,
> + gdbarch_value_to_register (gdbarch,
> + get_prev_frame_always (next_frame),
> toval->regnum (), type,
> fromval->contents ().data ());
> }
>
> base-commit: fa87f8e1958b8937047e3d5fcfd8c265745710ae
On 1/11/24 12:24, Tom de Vries wrote:
> On 1/11/24 17:51, Simon Marchi wrote:
>> Commit 78f2fd84e83 ("gdb: remove VALUE_REGNUM, add value::regnum")
>> introduced an unexpected change in value_assign. It replaced
>> `get_prev_frame_always (next_frame)` with `next_frame`in the call to
>> gdbarch_value_to_register.
>>
>> This is the result of a merge error, since I previously had a patch to
>> change gdbarch_value_to_register to take the next frame, and later
>> decided to drop it. Revert that change.
>>
>> Add a test based on the DWARF assembler to expose the problem and test
>> the fix. I also tested on ppc64le to make sure the problem reported in
>> PR 31231 was fixed.
>>
>
> I applied the patch on x86_64, ran the test-case, observed the internal error, rebuild gdb, and observed it pass.
>
> Only one nit, I get:
> ...
> DUPLICATE: gdb.dwarf2/assign-variable-value-to-register.exp: p foo
Ah thanks, missed it.
> ...
>
> Tested-by: Tom de Vries <tdevries@suse.de>
Thanks, will push with the duplicate test name fixed.
Simon
new file mode 100644
@@ -0,0 +1,86 @@
+# Copyright 2024 Free Software Foundation, Inc.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+# Test writing to a variable that is in a register that causes
+# gdbarch_value_to_register to be called. This was introduced to test
+# PR 31231 (https://sourceware.org/bugzilla/show_bug.cgi?id=31231).
+
+# This test uses a hard-coded x86-64 DWARF register number, so restrict it to
+# x86-64.
+if { ![is_x86_64_m64_target] } {
+ unsupported "unsupported architecture"
+}
+
+load_lib dwarf.exp
+
+standard_testfile main.c -dw.S
+set dwarf_file [standard_output_file $srcfile2]
+
+set st0_dwarf_regnum 33
+
+Dwarf::assemble $dwarf_file {
+ get_func_info main
+
+ cu {} {
+ DW_TAG_compile_unit {} {
+ declare_labels int_label float_label
+
+ int_label: DW_TAG_base_type {
+ { DW_AT_name int }
+ { DW_AT_byte_size 4 DW_FORM_udata }
+ { DW_AT_encoding @DW_ATE_signed }
+ }
+
+ float_label: DW_TAG_base_type {
+ { DW_AT_name float }
+ { DW_AT_byte_size 4 DW_FORM_udata }
+ { DW_AT_encoding @DW_ATE_float }
+ }
+
+ DW_TAG_subprogram {
+ { DW_AT_name main }
+ { DW_AT_low_pc $main_start DW_FORM_addr }
+ { DW_AT_high_pc $main_end DW_FORM_addr }
+ { DW_AT_type :$int_label }
+ } {
+ DW_TAG_variable {
+ { DW_AT_name foo }
+ { DW_AT_type :$float_label }
+ { DW_AT_location {
+ DW_OP_regx $::st0_dwarf_regnum
+ } SPECIAL_expr }
+ }
+ }
+ }
+ }
+}
+
+if { [prepare_for_testing "failed to prepare" $testfile \
+ [list $srcfile $dwarf_file] {nodebug}] } {
+ return
+}
+
+if { ![runto_main] } {
+ return
+}
+
+# st0 is expected to be initialized to 0.
+gdb_test "p foo" " = 0"
+
+# Set foo. This used to cause an internal error.
+gdb_test_no_output "set foo = 1234"
+
+# Confirm the value.
+gdb_test "p foo" " = 1234"
@@ -1244,7 +1244,8 @@ value_assign (struct value *toval, struct value *fromval)
/* If TOVAL is a special machine register requiring
conversion of program values to a special raw
format. */
- gdbarch_value_to_register (gdbarch, next_frame,
+ gdbarch_value_to_register (gdbarch,
+ get_prev_frame_always (next_frame),
toval->regnum (), type,
fromval->contents ().data ());
}