[v6] Enable tracing of pseudo-registers on ARM
Commit Message
In this v6:
* use https://sourceware.org/ml/gdb-patches/2016-02/msg00786.html to map
registers to remote registers. (This is already in master)
* Fix test changelog
* Test is already renamed, update patch.
-
This patch implements the ax_pseudo_register_push_stack and
ax_pseudo_register_collect gdbarch functions so that a pseudo-register can
be traced.
No regressions, tested on ubuntu 14.04 ARMv7 and x86.
With gdbserver-{native,extended} / { -marm -mthumb }
gdb/ChangeLog:
* arm-tdep.c (arm_pseudo_register_to_register): New function.
(arm_ax_pseudo_register_collect): New function.
(arm_ax_pseudo_register_push_stack): New function.
(arm_gdbarch_init): Set
gdbarch_ax_pseudo_register_{collect,push_stack} functions.
gdb/testsuite/ChangeLog:
* gdb.trace/tracefile-pseudo-reg.c: Include arm_neon.h if on arm.
(main): Add a register variable and a tracepoint label.
* gdb.trace/tracefile-pseudo-reg.exp): Add arm pseudo register
tracing test with s5 pseudo register.
---
gdb/arm-tdep.c | 68 ++++++++++++++++++++++++
gdb/testsuite/gdb.trace/tracefile-pseudo-reg.c | 12 +++++
gdb/testsuite/gdb.trace/tracefile-pseudo-reg.exp | 35 +++++++++---
3 files changed, 108 insertions(+), 7 deletions(-)
Comments
On 02/25/2016 03:32 PM, Antoine Tremblay wrote:
> In this v6:
> * use https://sourceware.org/ml/gdb-patches/2016-02/msg00786.html to map
> registers to remote registers. (This is already in master)
> * Fix test changelog
> * Test is already renamed, update patch.
> -
> This patch implements the ax_pseudo_register_push_stack and
> ax_pseudo_register_collect gdbarch functions so that a pseudo-register can
> be traced.
>
FAOD, I'm happy with this version, if Yao is happy.
> dummy (void)
> @@ -37,6 +41,7 @@ main (void)
> {
> /* Strictly speaking, it should be ymm15 (xmm15 is 128-bit), but gcc older
> than 4.9 doesn't recognize "ymm15" as a valid register name. */
> +#if (defined __x86_64__)
The comment should be within the #if.
Thanks,
Pedro Alves
Pedro Alves writes:
> On 02/25/2016 03:32 PM, Antoine Tremblay wrote:
>> In this v6:
>> * use https://sourceware.org/ml/gdb-patches/2016-02/msg00786.html to map
>> registers to remote registers. (This is already in master)
>> * Fix test changelog
>> * Test is already renamed, update patch.
>> -
>> This patch implements the ax_pseudo_register_push_stack and
>> ax_pseudo_register_collect gdbarch functions so that a pseudo-register can
>> be traced.
>>
>
> FAOD, I'm happy with this version, if Yao is happy.
>
OK. I'll wait for Yao's review.
>> dummy (void)
>> @@ -37,6 +41,7 @@ main (void)
>> {
>> /* Strictly speaking, it should be ymm15 (xmm15 is 128-bit), but gcc older
>> than 4.9 doesn't recognize "ymm15" as a valid register name. */
>> +#if (defined __x86_64__)
>
> The comment should be within the #if.
>
Fixed, thanks.
Antoine
Antoine Tremblay <antoine.tremblay@ericsson.com> writes:
> @@ -20,7 +20,11 @@
> * registers on x86_64.
> */
>
The comments above should be updated as well.
> +#if (defined __x86_64__)
> #include <immintrin.h>
> +#elif (defined __arm__ || defined __thumb2__ || defined __thumb__)
__arm__ is defined even in thumb mode, so only "defined __arm__" is enough.
> +#include <arm_neon.h>
Why do you include arm_neon.h? I don't see anything NEON specific is used.
> +#endif
>
> void
> dummy (void)
> @@ -37,6 +41,7 @@ main (void)
> {
> /* Strictly speaking, it should be ymm15 (xmm15 is 128-bit), but gcc older
> than 4.9 doesn't recognize "ymm15" as a valid register name. */
> +#if (defined __x86_64__)
> register __v8si a asm("xmm15") = {
> 0x12340001,
> 0x12340002,
> @@ -48,6 +53,13 @@ main (void)
> 0x12340008,
> };
> asm volatile ("traceme: call dummy" : : "x" (a));
> +#elif (defined __arm__ || defined __thumb2__ || defined __thumb__)
Only "defined __arm__" is needed.
> + register uint32_t a asm("s5") = {
> + 0x2
> + };
I'd like to write an inline asm to set s5 a value and the value can be shown as
an integer so that the test is more reliable (current test tests float
"2.80259693e-45").
> + asm volatile ("traceme: bl dummy" : : "x" (a));
> +#endif
> +
> end ();
> return 0;
> }
> diff --git a/gdb/testsuite/gdb.trace/tracefile-pseudo-reg.exp b/gdb/testsuite/gdb.trace/tracefile-pseudo-reg.exp
> index 4c52c64..12a2740 100644
> --- a/gdb/testsuite/gdb.trace/tracefile-pseudo-reg.exp
> +++ b/gdb/testsuite/gdb.trace/tracefile-pseudo-reg.exp
> @@ -12,8 +12,8 @@
> # You should have received a copy of the GNU General Public License
> # along with this program. If not, see <http://www.gnu.org/licenses/>.
>
> -if { ! [is_amd64_regs_target] } {
> - verbose "Skipping tfile AVX test (target is not x86_64)."
> +if { ! [is_amd64_regs_target] && ! [istarget "arm*-*-*"] } {
> + verbose "Skipping tracefile pseudo register tests, target is not supported."
> return
> }
>
> @@ -21,8 +21,14 @@ load_lib "trace-support.exp"
>
> standard_testfile
>
> +if { [is_amd64_regs_target] } {
> + set add_flags "-mavx"
> +} elseif { [istarget "arm*-*-*"] } {
> + set add_flags "-mfpu=neon"
Don't have to pass -mfpu=neon, because the case is also valid for vfp.
> +}
> +
> if {[prepare_for_testing $testfile.exp $testfile $srcfile \
> - [list debug additional_flags=-mavx]]} {
> + [list debug additional_flags=$add_flags]]} {
> return -1
> }
>
> @@ -36,20 +42,31 @@ if ![gdb_target_supports_trace] {
> return -1
> }
>
> -gdb_test_multiple "print \$ymm15" "check for AVX support" {
> +if { [is_amd64_regs_target] } {
> + set reg "\$ymm15"
> + set reg_message "check for AVX support"
> +} elseif { [istarget "arm*-*-*"] } {
> + set reg "\$s5"
> + set reg_message "check for Neon support"
$s5 exists on the processors which support NEON or VFP, so the
$reg_message isn't accurate. We can change reg_message to "check
register $reg".
Yao Qi writes:
> Antoine Tremblay <antoine.tremblay@ericsson.com> writes:
>
>> @@ -20,7 +20,11 @@
>> * registers on x86_64.
>> */
>>
>
> The comments above should be updated as well.
Done.
>
>> +#if (defined __x86_64__)
>> #include <immintrin.h>
>> +#elif (defined __arm__ || defined __thumb2__ || defined __thumb__)
>
> __arm__ is defined even in thumb mode, so only "defined __arm__" is enough.
>
>> +#include <arm_neon.h>
>
> Why do you include arm_neon.h? I don't see anything NEON specific is used.
>
>> +#endif
Indeed I was playing with neon types before, forgot to remove it. Fixed.
>>
>> void
>> dummy (void)
>> @@ -37,6 +41,7 @@ main (void)
>> {
>> /* Strictly speaking, it should be ymm15 (xmm15 is 128-bit), but gcc older
>> than 4.9 doesn't recognize "ymm15" as a valid register name. */
>> +#if (defined __x86_64__)
>> register __v8si a asm("xmm15") = {
>> 0x12340001,
>> 0x12340002,
>> @@ -48,6 +53,13 @@ main (void)
>> 0x12340008,
>> };
>> asm volatile ("traceme: call dummy" : : "x" (a));
>> +#elif (defined __arm__ || defined __thumb2__ || defined __thumb__)
>
> Only "defined __arm__" is needed.
>
Done.
>> + register uint32_t a asm("s5") = {
>> + 0x2
>> + };
>
> I'd like to write an inline asm to set s5 a value and the value can be shown as
> an integer so that the test is more reliable (current test tests float
> "2.80259693e-45").
>
Done.
>> + asm volatile ("traceme: bl dummy" : : "x" (a));
>> +#endif
>> +
>> end ();
>> return 0;
>> }
>> diff --git a/gdb/testsuite/gdb.trace/tracefile-pseudo-reg.exp b/gdb/testsuite/gdb.trace/tracefile-pseudo-reg.exp
>> index 4c52c64..12a2740 100644
>> --- a/gdb/testsuite/gdb.trace/tracefile-pseudo-reg.exp
>> +++ b/gdb/testsuite/gdb.trace/tracefile-pseudo-reg.exp
>> @@ -12,8 +12,8 @@
>> # You should have received a copy of the GNU General Public License
>> # along with this program. If not, see <http://www.gnu.org/licenses/>.
>>
>> -if { ! [is_amd64_regs_target] } {
>> - verbose "Skipping tfile AVX test (target is not x86_64)."
>> +if { ! [is_amd64_regs_target] && ! [istarget "arm*-*-*"] } {
>> + verbose "Skipping tracefile pseudo register tests, target is not supported."
>> return
>> }
>>
>> @@ -21,8 +21,14 @@ load_lib "trace-support.exp"
>>
>> standard_testfile
>>
>> +if { [is_amd64_regs_target] } {
>> + set add_flags "-mavx"
>> +} elseif { [istarget "arm*-*-*"] } {
>> + set add_flags "-mfpu=neon"
>
> Don't have to pass -mfpu=neon, because the case is also valid for vfp.
>
Right, Fixed.
>> +}
>> +
>> if {[prepare_for_testing $testfile.exp $testfile $srcfile \
>> - [list debug additional_flags=-mavx]]} {
>> + [list debug additional_flags=$add_flags]]} {
>> return -1
>> }
>>
>> @@ -36,20 +42,31 @@ if ![gdb_target_supports_trace] {
>> return -1
>> }
>>
>> -gdb_test_multiple "print \$ymm15" "check for AVX support" {
>> +if { [is_amd64_regs_target] } {
>> + set reg "\$ymm15"
>> + set reg_message "check for AVX support"
>> +} elseif { [istarget "arm*-*-*"] } {
>> + set reg "\$s5"
>> + set reg_message "check for Neon support"
>
> $s5 exists on the processors which support NEON or VFP, so the
> $reg_message isn't accurate. We can change reg_message to "check
> register $reg".
Done.
A v7 patch follows.
@@ -8716,6 +8716,70 @@ arm_pseudo_write (struct gdbarch *gdbarch, struct regcache *regcache,
}
}
+/* Map the pseudo register number REG to the proper register number. */
+
+static int
+arm_pseudo_register_to_register (struct gdbarch *gdbarch, int reg)
+{
+ int double_regnum = 0;
+ int num_regs = gdbarch_num_regs (gdbarch);
+ char name_buf[4];
+
+ /* Single precision pseudo registers. s0-s31. */
+ if (reg >= num_regs && reg < num_regs + 32)
+ {
+ xsnprintf (name_buf, sizeof (name_buf), "d%d", (reg - num_regs) / 2);
+ double_regnum = user_reg_map_name_to_regnum (gdbarch, name_buf,
+ strlen (name_buf));
+ }
+ /* Quadruple precision pseudo regisers. q0-q15. */
+ else if (reg >= num_regs + 32 && reg < num_regs + 32 + 16)
+ {
+ xsnprintf (name_buf, sizeof (name_buf), "d%d", (reg - num_regs - 32) * 2);
+ double_regnum = user_reg_map_name_to_regnum (gdbarch, name_buf,
+ strlen (name_buf));
+ }
+ /* Error bad register number. */
+ else
+ return -1;
+
+ return double_regnum;
+}
+
+/* Implementation of the ax_pseudo_register_collect gdbarch function. */
+
+static int
+arm_ax_pseudo_register_collect (struct gdbarch *gdbarch,
+ struct agent_expr *ax, int reg)
+{
+ int rawnum = arm_pseudo_register_to_register (gdbarch, reg);
+
+ /* Error. */
+ if (rawnum < 0)
+ return 1;
+
+ ax_reg_mask (ax, rawnum);
+
+ return 0;
+}
+
+/* Implementation of the ax_pseudo_register_push_stack gdbarch function. */
+
+static int
+arm_ax_pseudo_register_push_stack (struct gdbarch *gdbarch,
+ struct agent_expr *ax, int reg)
+{
+ int rawnum = arm_pseudo_register_to_register (gdbarch, reg);
+
+ /* Error. */
+ if (rawnum < 0)
+ return 1;
+
+ ax_reg (ax, rawnum);
+
+ return 0;
+}
+
static struct value *
value_of_arm_user_reg (struct frame_info *frame, const void *baton)
{
@@ -9377,6 +9441,10 @@ arm_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
set_gdbarch_num_pseudo_regs (gdbarch, num_pseudos);
set_gdbarch_pseudo_register_read (gdbarch, arm_pseudo_read);
set_gdbarch_pseudo_register_write (gdbarch, arm_pseudo_write);
+ set_gdbarch_ax_pseudo_register_push_stack
+ (gdbarch, arm_ax_pseudo_register_push_stack);
+ set_gdbarch_ax_pseudo_register_collect
+ (gdbarch, arm_ax_pseudo_register_collect);
}
if (tdesc_data)
@@ -20,7 +20,11 @@
* registers on x86_64.
*/
+#if (defined __x86_64__)
#include <immintrin.h>
+#elif (defined __arm__ || defined __thumb2__ || defined __thumb__)
+#include <arm_neon.h>
+#endif
void
dummy (void)
@@ -37,6 +41,7 @@ main (void)
{
/* Strictly speaking, it should be ymm15 (xmm15 is 128-bit), but gcc older
than 4.9 doesn't recognize "ymm15" as a valid register name. */
+#if (defined __x86_64__)
register __v8si a asm("xmm15") = {
0x12340001,
0x12340002,
@@ -48,6 +53,13 @@ main (void)
0x12340008,
};
asm volatile ("traceme: call dummy" : : "x" (a));
+#elif (defined __arm__ || defined __thumb2__ || defined __thumb__)
+ register uint32_t a asm("s5") = {
+ 0x2
+ };
+ asm volatile ("traceme: bl dummy" : : "x" (a));
+#endif
+
end ();
return 0;
}
@@ -12,8 +12,8 @@
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
-if { ! [is_amd64_regs_target] } {
- verbose "Skipping tfile AVX test (target is not x86_64)."
+if { ! [is_amd64_regs_target] && ! [istarget "arm*-*-*"] } {
+ verbose "Skipping tracefile pseudo register tests, target is not supported."
return
}
@@ -21,8 +21,14 @@ load_lib "trace-support.exp"
standard_testfile
+if { [is_amd64_regs_target] } {
+ set add_flags "-mavx"
+} elseif { [istarget "arm*-*-*"] } {
+ set add_flags "-mfpu=neon"
+}
+
if {[prepare_for_testing $testfile.exp $testfile $srcfile \
- [list debug additional_flags=-mavx]]} {
+ [list debug additional_flags=$add_flags]]} {
return -1
}
@@ -36,20 +42,31 @@ if ![gdb_target_supports_trace] {
return -1
}
-gdb_test_multiple "print \$ymm15" "check for AVX support" {
+if { [is_amd64_regs_target] } {
+ set reg "\$ymm15"
+ set reg_message "check for AVX support"
+} elseif { [istarget "arm*-*-*"] } {
+ set reg "\$s5"
+ set reg_message "check for Neon support"
+}
+
+gdb_test_multiple "print $reg" $reg_message {
-re " = void.*$gdb_prompt $" {
- verbose "Skipping tfile AVX test (target doesn't support AVX)."
+ verbose "Skipping tracefile pseudo register tests, target is not supported."
return
}
-re " = \\{.*}.*$gdb_prompt $" {
# All is well.
}
+ -re " = 0.*$gdb_prompt $" {
+ # All is well.
+ }
}
gdb_test "trace traceme" ".*"
gdb_trace_setactions "set actions for tracepoint" "" \
- "collect \$ymm15" "^$"
+ "collect $reg" "^$"
gdb_breakpoint "end"
@@ -70,4 +87,8 @@ gdb_test "target tfile ${tracefile}.tf" "" "change to tfile target" \
gdb_test "tfind 0" "Found trace frame 0, tracepoint .*"
-gdb_test "print/x \$ymm15.v8_int32" " = \\{0x12340001, .*, 0x12340008}"
+if { [is_amd64_regs_target] } {
+ gdb_test "print/x \$ymm15.v8_int32" " = \\{0x12340001, .*, 0x12340008}"
+} elseif { [istarget "arm*-*-*"] } {
+ gdb_test "print \$s5" "2.80259693e-45"
+}