xref: /openbmc/qemu/docs/system/gdb.rst (revision 436c831a)
1.. _GDB usage:
2
3GDB usage
4---------
5
6QEMU supports working with gdb via gdb's remote-connection facility
7(the "gdbstub"). This allows you to debug guest code in the same
8way that you might with a low-level debug facility like JTAG
9on real hardware. You can stop and start the virtual machine,
10examine state like registers and memory, and set breakpoints and
11watchpoints.
12
13In order to use gdb, launch QEMU with the ``-s`` and ``-S`` options.
14The ``-s`` option will make QEMU listen for an incoming connection
15from gdb on TCP port 1234, and ``-S`` will make QEMU not start the
16guest until you tell it to from gdb. (If you want to specify which
17TCP port to use or to use something other than TCP for the gdbstub
18connection, use the ``-gdb dev`` option instead of ``-s``. See
19`Using unix sockets`_ for an example.)
20
21.. parsed-literal::
22
23   |qemu_system| -s -S -kernel bzImage -hda rootdisk.img -append "root=/dev/hda"
24
25QEMU will launch but will silently wait for gdb to connect.
26
27Then launch gdb on the 'vmlinux' executable::
28
29   > gdb vmlinux
30
31In gdb, connect to QEMU::
32
33   (gdb) target remote localhost:1234
34
35Then you can use gdb normally. For example, type 'c' to launch the
36kernel::
37
38   (gdb) c
39
40Here are some useful tips in order to use gdb on system code:
41
421. Use ``info reg`` to display all the CPU registers.
43
442. Use ``x/10i $eip`` to display the code at the PC position.
45
463. Use ``set architecture i8086`` to dump 16 bit code. Then use
47   ``x/10i $cs*16+$eip`` to dump the code at the PC position.
48
49Debugging multicore machines
50============================
51
52GDB's abstraction for debugging targets with multiple possible
53parallel flows of execution is a two layer one: it supports multiple
54"inferiors", each of which can have multiple "threads". When the QEMU
55machine has more than one CPU, QEMU exposes each CPU cluster as a
56separate "inferior", where each CPU within the cluster is a separate
57"thread". Most QEMU machine types have identical CPUs, so there is a
58single cluster which has all the CPUs in it.  A few machine types are
59heterogenous and have multiple clusters: for example the ``sifive_u``
60machine has a cluster with one E51 core and a second cluster with four
61U54 cores. Here the E51 is the only thread in the first inferior, and
62the U54 cores are all threads in the second inferior.
63
64When you connect gdb to the gdbstub, it will automatically
65connect to the first inferior; you can display the CPUs in this
66cluster using the gdb ``info thread`` command, and switch between
67them using gdb's usual thread-management commands.
68
69For multi-cluster machines, unfortunately gdb does not by default
70handle multiple inferiors, and so you have to explicitly connect
71to them. First, you must connect with the ``extended-remote``
72protocol, not ``remote``::
73
74    (gdb) target extended-remote localhost:1234
75
76Once connected, gdb will have a single inferior, for the
77first cluster. You need to create inferiors for the other
78clusters and attach to them, like this::
79
80  (gdb) add-inferior
81  Added inferior 2
82  (gdb) inferior 2
83  [Switching to inferior 2 [<null>] (<noexec>)]
84  (gdb) attach 2
85  Attaching to process 2
86  warning: No executable has been specified and target does not support
87  determining executable automatically.  Try using the "file" command.
88  0x00000000 in ?? ()
89
90Once you've done this, ``info threads`` will show CPUs in
91all the clusters you have attached to::
92
93  (gdb) info threads
94    Id   Target Id         Frame
95    1.1  Thread 1.1 (cortex-m33-arm-cpu cpu [running]) 0x00000000 in ?? ()
96  * 2.1  Thread 2.2 (cortex-m33-arm-cpu cpu [halted ]) 0x00000000 in ?? ()
97
98You probably also want to set gdb to ``schedule-multiple`` mode,
99so that when you tell gdb to ``continue`` it resumes all CPUs,
100not just those in the cluster you are currently working on::
101
102  (gdb) set schedule-multiple on
103
104Using unix sockets
105==================
106
107An alternate method for connecting gdb to the QEMU gdbstub is to use
108a unix socket (if supported by your operating system). This is useful when
109running several tests in parallel, or if you do not have a known free TCP
110port (e.g. when running automated tests).
111
112First create a chardev with the appropriate options, then
113instruct the gdbserver to use that device:
114
115.. parsed-literal::
116
117   |qemu_system| -chardev socket,path=/tmp/gdb-socket,server=on,wait=off,id=gdb0 -gdb chardev:gdb0 -S ...
118
119Start gdb as before, but this time connect using the path to
120the socket::
121
122   (gdb) target remote /tmp/gdb-socket
123
124Note that to use a unix socket for the connection you will need
125gdb version 9.0 or newer.
126
127Advanced debugging options
128==========================
129
130Changing single-stepping behaviour
131^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
132
133The default single stepping behavior is step with the IRQs and timer
134service routines off. It is set this way because when gdb executes a
135single step it expects to advance beyond the current instruction. With
136the IRQs and timer service routines on, a single step might jump into
137the one of the interrupt or exception vectors instead of executing the
138current instruction. This means you may hit the same breakpoint a number
139of times before executing the instruction gdb wants to have executed.
140Because there are rare circumstances where you want to single step into
141an interrupt vector the behavior can be controlled from GDB. There are
142three commands you can query and set the single step behavior:
143
144``maintenance packet qqemu.sstepbits``
145   This will display the MASK bits used to control the single stepping
146   IE:
147
148   ::
149
150      (gdb) maintenance packet qqemu.sstepbits
151      sending: "qqemu.sstepbits"
152      received: "ENABLE=1,NOIRQ=2,NOTIMER=4"
153
154``maintenance packet qqemu.sstep``
155   This will display the current value of the mask used when single
156   stepping IE:
157
158   ::
159
160      (gdb) maintenance packet qqemu.sstep
161      sending: "qqemu.sstep"
162      received: "0x7"
163
164``maintenance packet Qqemu.sstep=HEX_VALUE``
165   This will change the single step mask, so if wanted to enable IRQs on
166   the single step, but not timers, you would use:
167
168   ::
169
170      (gdb) maintenance packet Qqemu.sstep=0x5
171      sending: "qemu.sstep=0x5"
172      received: "OK"
173
174Examining physical memory
175^^^^^^^^^^^^^^^^^^^^^^^^^
176
177Another feature that QEMU gdbstub provides is to toggle the memory GDB
178works with, by default GDB will show the current process memory respecting
179the virtual address translation.
180
181If you want to examine/change the physical memory you can set the gdbstub
182to work with the physical memory rather with the virtual one.
183
184The memory mode can be checked by sending the following command:
185
186``maintenance packet qqemu.PhyMemMode``
187    This will return either 0 or 1, 1 indicates you are currently in the
188    physical memory mode.
189
190``maintenance packet Qqemu.PhyMemMode:1``
191    This will change the memory mode to physical memory.
192
193``maintenance packet Qqemu.PhyMemMode:0``
194    This will change it back to normal memory mode.
195