2013-05-06 22:04:46 +08:00
|
|
|
# This testcase is part of GDB, the GNU debugger.
|
|
|
|
#
|
2024-01-12 23:30:44 +08:00
|
|
|
# Copyright 2013-2024 Free Software Foundation, Inc.
|
2013-05-06 22:04:46 +08:00
|
|
|
#
|
|
|
|
# Contributed by Intel Corp. <markus.t.metzger@intel.com>
|
|
|
|
#
|
|
|
|
# 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/>.
|
|
|
|
|
2023-01-09 01:44:15 +08:00
|
|
|
require allow_btrace_tests
|
2013-05-06 22:04:46 +08:00
|
|
|
|
2015-02-03 23:42:41 +08:00
|
|
|
# This test is stepping on instruction level. To guarantee that we always
|
|
|
|
# get the same execution trace, we use an assembly source file.
|
|
|
|
#
|
|
|
|
# We use different assembly sources based on the target architecture.
|
|
|
|
#
|
|
|
|
# Luckily, they are similar enough that a single test script can handle
|
|
|
|
# both.
|
2016-07-22 19:41:45 +08:00
|
|
|
if [info exists COMPILE] {
|
|
|
|
# make check RUNTESTFLAGS="gdb.btrace/stepi.exp COMPILE=1"
|
|
|
|
standard_testfile record_goto.c
|
|
|
|
lappend opts debug
|
|
|
|
} elseif {[istarget "i?86-*-*"] || [istarget "x86_64-*-*"]} {
|
|
|
|
if {[is_amd64_regs_target]} {
|
|
|
|
standard_testfile x86_64-record_goto.S
|
|
|
|
} else {
|
|
|
|
standard_testfile i686-record_goto.S
|
|
|
|
}
|
2015-02-03 23:42:41 +08:00
|
|
|
} else {
|
2016-12-09 20:16:22 +08:00
|
|
|
unsupported "target architecture not supported"
|
2016-07-22 19:41:45 +08:00
|
|
|
return -1
|
2015-02-03 23:42:41 +08:00
|
|
|
}
|
|
|
|
|
2020-12-17 16:34:54 +08:00
|
|
|
if [prepare_for_testing "failed to prepare" $testfile $srcfile {}] {
|
2013-05-06 22:04:46 +08:00
|
|
|
return -1
|
|
|
|
}
|
|
|
|
|
|
|
|
if ![runto_main] {
|
|
|
|
return -1
|
|
|
|
}
|
|
|
|
|
2016-12-09 20:16:22 +08:00
|
|
|
global gdb_prompt
|
|
|
|
|
2013-05-06 22:04:46 +08:00
|
|
|
proc check_replay_at { insn } {
|
2015-03-10 16:51:27 +08:00
|
|
|
gdb_test "info record" [multi_line \
|
2013-05-06 22:04:46 +08:00
|
|
|
"Active record target: record-btrace" \
|
2015-02-03 23:42:41 +08:00
|
|
|
".*" \
|
record-btrace: indicate gaps
Indicate gaps in the trace due to decode errors. Internally, a gap is
represented as a btrace function segment without instructions and with a
non-zero format-specific error code.
Show the gap when traversing the instruction or function call history.
Also indicate gaps in "info record".
It looks like this:
(gdb) info record
Active record target: record-btrace
Recording format: Branch Trace Store.
Buffer size: 64KB.
Recorded 32 instructions in 5 functions (1 gaps) for thread 1 (process 7182).
(gdb) record function-call-history /cli
1 fib inst 1,9 at src/fib.c:9,14
2 fib inst 10,20 at src/fib.c:6,14
3 [decode error (1): instruction overflow]
4 fib inst 21,28 at src/fib.c:11,14
5 fib inst 29,33 at src/fib.c:6,9
(gdb) record instruction-history 20,22
20 0x000000000040062f <fib+47>: sub $0x1,%rax
[decode error (1): instruction overflow]
21 0x0000000000400613 <fib+19>: add $0x1,%rax
22 0x0000000000400617 <fib+23>: mov %rax,0x200a3a(%rip)
(gdb)
Gaps are ignored during reverse execution and replay.
2015-02-09 Markus Metzger <markus.t.metzger@intel.com>
* btrace.c (ftrace_find_call): Skip gaps.
(ftrace_new_function): Initialize level.
(ftrace_new_call, ftrace_new_tailcall, ftrace_new_return)
(ftrace_new_switch): Update
level computation.
(ftrace_new_gap): New.
(ftrace_update_function): Create new function after gap.
(btrace_compute_ftrace_bts): Create gap on error.
(btrace_stitch_bts): Update parameters. Clear trace if it
becomes empty.
(btrace_stitch_trace): Update parameters. Update callers.
(btrace_clear): Reset the number of gaps.
(btrace_insn_get): Return NULL if the iterator points to a gap.
(btrace_insn_number): Return zero if the iterator points to a gap.
(btrace_insn_end): Allow gaps at the end.
(btrace_insn_next, btrace_insn_prev, btrace_insn_cmp): Handle gaps.
(btrace_find_insn_by_number): Assert that the found iterator does
not point to a gap.
(btrace_call_next, btrace_call_prev): Assert that the last function
is not a gap.
* btrace.h (btrace_bts_error): New.
(btrace_function): Update comment.
(btrace_function) <insn, insn_offset, number>: Update comment.
(btrace_function) <errcode>: New.
(btrace_thread_info) <ngaps>: New.
(btrace_thread_info) <replay>: Update comment.
(btrace_insn_get): Update comment.
* record-btrace.c (btrace_ui_out_decode_error): New.
(record_btrace_info): Print number of gaps.
(btrace_insn_history, btrace_call_history): Call
btrace_ui_out_decode_error for gaps.
(record_btrace_step_thread, record_btrace_start_replaying): Skip gaps.
testsuite/
* gdb.btrace/buffer-size.exp: Update "info record" output.
* gdb.btrace/delta.exp: Update "info record" output.
* gdb.btrace/enable.exp: Update "info record" output.
* gdb.btrace/finish.exp: Update "info record" output.
* gdb.btrace/instruction_history.exp: Update "info record" output.
* gdb.btrace/next.exp: Update "info record" output.
* gdb.btrace/nexti.exp: Update "info record" output.
* gdb.btrace/step.exp: Update "info record" output.
* gdb.btrace/stepi.exp: Update "info record" output.
* gdb.btrace/nohist.exp: Update "info record" output.
2014-01-30 16:51:10 +08:00
|
|
|
"Recorded 40 instructions in 16 functions \\\(0 gaps\\\) for .*" \
|
2013-05-06 22:04:46 +08:00
|
|
|
"Replay in progress\. At instruction $insn\." \
|
2020-12-04 17:01:29 +08:00
|
|
|
] "check replay at $insn"
|
2013-05-06 22:04:46 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
# trace the call to the test function
|
2020-12-04 17:01:29 +08:00
|
|
|
with_test_prefix "record" {
|
|
|
|
gdb_test_no_output "record btrace"
|
|
|
|
gdb_test "next" ".*" "next.1"
|
|
|
|
}
|
2013-05-06 22:04:46 +08:00
|
|
|
|
|
|
|
# we start with stepping to make sure that the trace is fetched automatically
|
2020-12-04 17:01:29 +08:00
|
|
|
with_test_prefix "fetch" {
|
|
|
|
gdb_test "reverse-stepi" ".*fun4\.5.*" "reverse-stepi.1"
|
|
|
|
gdb_test "reverse-stepi" ".*fun4\.5.*" "reverse-stepi.2"
|
2013-05-06 22:04:46 +08:00
|
|
|
|
2020-12-04 17:01:29 +08:00
|
|
|
# let's check where we are in the trace
|
|
|
|
check_replay_at 39
|
|
|
|
}
|
2013-05-06 22:04:46 +08:00
|
|
|
|
|
|
|
# let's step forward and check again
|
2020-12-04 17:01:29 +08:00
|
|
|
with_test_prefix "stepi" {
|
|
|
|
gdb_test "stepi" ".*fun4\.5.*"
|
|
|
|
check_replay_at 40
|
|
|
|
}
|
2013-05-06 22:04:46 +08:00
|
|
|
|
|
|
|
# with the next step, we stop replaying
|
2020-12-04 17:01:29 +08:00
|
|
|
with_test_prefix "end" {
|
|
|
|
gdb_test "stepi" ".*main\.3.*"
|
|
|
|
gdb_test "info record" [multi_line \
|
|
|
|
"Active record target: record-btrace" \
|
|
|
|
".*" \
|
|
|
|
"Recorded 40 instructions in 16 functions \\\(0 gaps\\\) for \[^\\\r\\\n\]*" \
|
|
|
|
]
|
|
|
|
}
|
2013-05-06 22:04:46 +08:00
|
|
|
|
2015-02-03 23:42:41 +08:00
|
|
|
# let's try nexti
|
2020-12-04 17:01:29 +08:00
|
|
|
with_test_prefix "reverse-nexti.1" {
|
|
|
|
gdb_test "reverse-nexti" ".*main\.2.*"
|
|
|
|
check_replay_at 1
|
|
|
|
}
|
2015-02-03 23:42:41 +08:00
|
|
|
|
|
|
|
# we can't reverse-nexti any further
|
2020-12-04 17:01:29 +08:00
|
|
|
with_test_prefix "reverse-nexti.2" {
|
|
|
|
gdb_test "reverse-nexti" \
|
Change message when reaching end of reverse history.
In a record session, when we move backward, GDB switches from normal
execution to simulation. Moving forward again, the emulation continues
until the end of the reverse history. When the end is reached, the
execution stops, and a warning message is shown. This message has been
modified to indicate that the forward emulation has reached the end, but
the execution can continue as normal, and the recording will also continue.
Before this patch, the warning message shown in that case was the same as
in the reverse case. This meant that when the end of history was reached in
either backward or forward emulation, the same message was displayed:
"No more reverse-execution history."
This message has changed for these two cases. Backward emulation:
"Reached end of recorded history; stopping.
Backward execution from here not possible."
Forward emulation:
"Reached end of recorded history; stopping.
Following forward execution will be added to history."
The reason for this change is that the initial message was deceiving, for
the forward case, making the user believe that forward debugging could not
continue.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=31224
Reviewed-By: Markus T. Metzger <markus.t.metzger@intel.com> (btrace)
Approved-By: Guinevere Larsen <blarsen@redhat.com>
2024-08-26 21:33:17 +08:00
|
|
|
"Reached end of recorded history; stopping\.\r\n.*main\.2.*" \
|
2020-12-04 17:01:29 +08:00
|
|
|
"reverse-nexti.2"
|
|
|
|
check_replay_at 1
|
|
|
|
}
|
2015-02-03 23:42:41 +08:00
|
|
|
|
|
|
|
# but we can step back again
|
2020-12-04 17:01:29 +08:00
|
|
|
with_test_prefix "nexti" {
|
|
|
|
gdb_test "nexti" ".*main\.3.*"
|
|
|
|
gdb_test "info record" [multi_line \
|
|
|
|
"Active record target: record-btrace" \
|
|
|
|
".*" \
|
|
|
|
"Recorded 40 instructions in 16 functions \\\(0 gaps\\\) for \[^\\\r\\\n\]*" \
|
|
|
|
]
|
|
|
|
}
|
2015-02-03 23:42:41 +08:00
|
|
|
|
2013-05-06 22:04:46 +08:00
|
|
|
# let's step from a goto position somewhere in the middle
|
2020-12-04 17:01:29 +08:00
|
|
|
with_test_prefix "goto" {
|
|
|
|
gdb_test "record goto 22" ".*fun3\.2.*"
|
|
|
|
with_test_prefix "goto 22" { check_replay_at 22 }
|
2013-05-06 22:04:46 +08:00
|
|
|
|
2020-12-04 17:01:29 +08:00
|
|
|
gdb_test "stepi" ".*fun1\.1.*" "stepi.3"
|
|
|
|
with_test_prefix "stepi to 23" { check_replay_at 23 }
|
2013-05-06 22:04:46 +08:00
|
|
|
|
2020-12-04 17:01:29 +08:00
|
|
|
gdb_test "reverse-stepi" ".*fun3\.2.*" "reverse-stepi.3"
|
|
|
|
with_test_prefix "reverse-stepi to 22" { check_replay_at 22 }
|
2015-02-03 23:42:41 +08:00
|
|
|
|
2020-12-04 17:01:29 +08:00
|
|
|
gdb_test "nexti" ".*fun3\.3.*"
|
|
|
|
with_test_prefix "nexti to 27" { check_replay_at 27 }
|
2015-02-03 23:42:41 +08:00
|
|
|
|
2020-12-04 17:01:29 +08:00
|
|
|
gdb_test "reverse-nexti" ".*fun3\.2.*" "reverse-nexti.3"
|
|
|
|
with_test_prefix "reverse-nexti to 22" { check_replay_at 22 }
|
|
|
|
}
|
2013-05-06 22:04:46 +08:00
|
|
|
|
|
|
|
# let's try to step off the left end
|
2020-12-04 17:01:29 +08:00
|
|
|
with_test_prefix "goto begin" {
|
|
|
|
gdb_test "record goto begin" ".*main\.2.*"
|
|
|
|
check_replay_at 1
|
|
|
|
|
|
|
|
with_test_prefix "reverse-stepi" {
|
|
|
|
gdb_test "reverse-stepi" \
|
Change message when reaching end of reverse history.
In a record session, when we move backward, GDB switches from normal
execution to simulation. Moving forward again, the emulation continues
until the end of the reverse history. When the end is reached, the
execution stops, and a warning message is shown. This message has been
modified to indicate that the forward emulation has reached the end, but
the execution can continue as normal, and the recording will also continue.
Before this patch, the warning message shown in that case was the same as
in the reverse case. This meant that when the end of history was reached in
either backward or forward emulation, the same message was displayed:
"No more reverse-execution history."
This message has changed for these two cases. Backward emulation:
"Reached end of recorded history; stopping.
Backward execution from here not possible."
Forward emulation:
"Reached end of recorded history; stopping.
Following forward execution will be added to history."
The reason for this change is that the initial message was deceiving, for
the forward case, making the user believe that forward debugging could not
continue.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=31224
Reviewed-By: Markus T. Metzger <markus.t.metzger@intel.com> (btrace)
Approved-By: Guinevere Larsen <blarsen@redhat.com>
2024-08-26 21:33:17 +08:00
|
|
|
"Reached end of recorded history; stopping\.\r\n.*main\.2.*" \
|
2020-12-04 17:01:29 +08:00
|
|
|
"reverse-stepi.1"
|
|
|
|
gdb_test "reverse-stepi" \
|
Change message when reaching end of reverse history.
In a record session, when we move backward, GDB switches from normal
execution to simulation. Moving forward again, the emulation continues
until the end of the reverse history. When the end is reached, the
execution stops, and a warning message is shown. This message has been
modified to indicate that the forward emulation has reached the end, but
the execution can continue as normal, and the recording will also continue.
Before this patch, the warning message shown in that case was the same as
in the reverse case. This meant that when the end of history was reached in
either backward or forward emulation, the same message was displayed:
"No more reverse-execution history."
This message has changed for these two cases. Backward emulation:
"Reached end of recorded history; stopping.
Backward execution from here not possible."
Forward emulation:
"Reached end of recorded history; stopping.
Following forward execution will be added to history."
The reason for this change is that the initial message was deceiving, for
the forward case, making the user believe that forward debugging could not
continue.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=31224
Reviewed-By: Markus T. Metzger <markus.t.metzger@intel.com> (btrace)
Approved-By: Guinevere Larsen <blarsen@redhat.com>
2024-08-26 21:33:17 +08:00
|
|
|
"Reached end of recorded history; stopping\.\r\n.*main\.2.*" \
|
2020-12-04 17:01:29 +08:00
|
|
|
"reverse-stepi.2"
|
|
|
|
check_replay_at 1
|
|
|
|
}
|
|
|
|
|
|
|
|
with_test_prefix "reverse-nexti" {
|
|
|
|
gdb_test "reverse-nexti" \
|
Change message when reaching end of reverse history.
In a record session, when we move backward, GDB switches from normal
execution to simulation. Moving forward again, the emulation continues
until the end of the reverse history. When the end is reached, the
execution stops, and a warning message is shown. This message has been
modified to indicate that the forward emulation has reached the end, but
the execution can continue as normal, and the recording will also continue.
Before this patch, the warning message shown in that case was the same as
in the reverse case. This meant that when the end of history was reached in
either backward or forward emulation, the same message was displayed:
"No more reverse-execution history."
This message has changed for these two cases. Backward emulation:
"Reached end of recorded history; stopping.
Backward execution from here not possible."
Forward emulation:
"Reached end of recorded history; stopping.
Following forward execution will be added to history."
The reason for this change is that the initial message was deceiving, for
the forward case, making the user believe that forward debugging could not
continue.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=31224
Reviewed-By: Markus T. Metzger <markus.t.metzger@intel.com> (btrace)
Approved-By: Guinevere Larsen <blarsen@redhat.com>
2024-08-26 21:33:17 +08:00
|
|
|
"Reached end of recorded history; stopping\.\r\n.*main\.2.*" \
|
2020-12-04 17:01:29 +08:00
|
|
|
"reverse-nexti.1"
|
|
|
|
gdb_test "reverse-nexti" \
|
Change message when reaching end of reverse history.
In a record session, when we move backward, GDB switches from normal
execution to simulation. Moving forward again, the emulation continues
until the end of the reverse history. When the end is reached, the
execution stops, and a warning message is shown. This message has been
modified to indicate that the forward emulation has reached the end, but
the execution can continue as normal, and the recording will also continue.
Before this patch, the warning message shown in that case was the same as
in the reverse case. This meant that when the end of history was reached in
either backward or forward emulation, the same message was displayed:
"No more reverse-execution history."
This message has changed for these two cases. Backward emulation:
"Reached end of recorded history; stopping.
Backward execution from here not possible."
Forward emulation:
"Reached end of recorded history; stopping.
Following forward execution will be added to history."
The reason for this change is that the initial message was deceiving, for
the forward case, making the user believe that forward debugging could not
continue.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=31224
Reviewed-By: Markus T. Metzger <markus.t.metzger@intel.com> (btrace)
Approved-By: Guinevere Larsen <blarsen@redhat.com>
2024-08-26 21:33:17 +08:00
|
|
|
"Reached end of recorded history; stopping\.\r\n.*main\.2.*" \
|
2020-12-04 17:01:29 +08:00
|
|
|
"reverse-nexti.2"
|
|
|
|
check_replay_at 1
|
|
|
|
}
|
|
|
|
|
|
|
|
# we can step forward, though
|
|
|
|
with_test_prefix "stepi" {
|
|
|
|
gdb_test "stepi" ".*fun4\.1.*"
|
|
|
|
check_replay_at 2
|
|
|
|
}
|
|
|
|
}
|
2013-05-06 22:04:46 +08:00
|
|
|
|
|
|
|
# let's try to step off the left end again
|
2020-12-04 17:01:29 +08:00
|
|
|
with_test_prefix "reverse-stepi" {
|
|
|
|
gdb_test "reverse-stepi" ".*main\.2.*" "reverse-stepi.1"
|
|
|
|
gdb_test "reverse-stepi" \
|
Change message when reaching end of reverse history.
In a record session, when we move backward, GDB switches from normal
execution to simulation. Moving forward again, the emulation continues
until the end of the reverse history. When the end is reached, the
execution stops, and a warning message is shown. This message has been
modified to indicate that the forward emulation has reached the end, but
the execution can continue as normal, and the recording will also continue.
Before this patch, the warning message shown in that case was the same as
in the reverse case. This meant that when the end of history was reached in
either backward or forward emulation, the same message was displayed:
"No more reverse-execution history."
This message has changed for these two cases. Backward emulation:
"Reached end of recorded history; stopping.
Backward execution from here not possible."
Forward emulation:
"Reached end of recorded history; stopping.
Following forward execution will be added to history."
The reason for this change is that the initial message was deceiving, for
the forward case, making the user believe that forward debugging could not
continue.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=31224
Reviewed-By: Markus T. Metzger <markus.t.metzger@intel.com> (btrace)
Approved-By: Guinevere Larsen <blarsen@redhat.com>
2024-08-26 21:33:17 +08:00
|
|
|
"Reached end of recorded history; stopping\.\r\n.*main\.2.*" \
|
2020-12-04 17:01:29 +08:00
|
|
|
"reverse-stepi.2"
|
|
|
|
gdb_test "reverse-stepi" \
|
Change message when reaching end of reverse history.
In a record session, when we move backward, GDB switches from normal
execution to simulation. Moving forward again, the emulation continues
until the end of the reverse history. When the end is reached, the
execution stops, and a warning message is shown. This message has been
modified to indicate that the forward emulation has reached the end, but
the execution can continue as normal, and the recording will also continue.
Before this patch, the warning message shown in that case was the same as
in the reverse case. This meant that when the end of history was reached in
either backward or forward emulation, the same message was displayed:
"No more reverse-execution history."
This message has changed for these two cases. Backward emulation:
"Reached end of recorded history; stopping.
Backward execution from here not possible."
Forward emulation:
"Reached end of recorded history; stopping.
Following forward execution will be added to history."
The reason for this change is that the initial message was deceiving, for
the forward case, making the user believe that forward debugging could not
continue.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=31224
Reviewed-By: Markus T. Metzger <markus.t.metzger@intel.com> (btrace)
Approved-By: Guinevere Larsen <blarsen@redhat.com>
2024-08-26 21:33:17 +08:00
|
|
|
"Reached end of recorded history; stopping\.\r\n.*main\.2.*" \
|
2020-12-04 17:01:29 +08:00
|
|
|
"reverse-stepi.3"
|
|
|
|
check_replay_at 1
|
|
|
|
}
|