public inbox for gdb@sourceware.org
 help / color / mirror / Atom feed
From: Torbjorn SVENSSON <torbjorn.svensson@st.com>
To: 庄秋彬 <mark.zhuang@spacemit.com>, "Tom Tromey" <tom@tromey.com>
Cc: gdb <gdb@sourceware.org>
Subject: RE: why gdb try to read address 0x0 when I set $pc,how to suppress this behavior?
Date: Thu, 27 Jul 2023 17:36:45 +0000	[thread overview]
Message-ID: <AM6PR10MB219793057F9B1278773129718101A@AM6PR10MB2197.EURPRD10.PROD.OUTLOOK.COM> (raw)
In-Reply-To: <16c8bb4f69de36253945d0fb7891fb770dd49697.2fb804b7.2dde.4cea.a08c.33c09a7a4597@feishu.cn>

Hello,

I think the interesting frames are:

#12 0x00000000006c0de8 in riscv_insn::fetch_instruction (gdbarch=gdbarch@entry=0xf83870, addr=0, len=len@entry=0x7fffffffd9e0)
    at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/riscv-tdep.c:1664
#13 0x00000000006c0ee8 in riscv_insn::decode (this=0x7fffffffd9e0, gdbarch=0xf83870, pc=<optimized out>) at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/riscv-tdep.c:1693
#14 0x00000000006c187c in riscv_scan_prologue (gdbarch=0xf83870, start_pc=<optimized out>, end_pc=100, cache=0xe9bd20) at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/riscv-tdep.c:1873

Before you do anything else, I would suggest that you rebuild riscv-gdb with debug symbols and without optimization ("-O0 -ggdb" for clang/gcc) as it will make debugging much easier. 

Then I would either run it in a native gdb, as Tom suggested, or instrument the code with some additional printf statements to show the content of some important variables.
Particular interesting is the content of "next_pc", "cur_pc", "start_pc" and "end_pc" in the for-loop in riscv_scan_prologue.

Happy hacking! :)

Regards,
Torbjörn

> -----Original Message-----
> From: Gdb <gdb-bounces+torbjorn.svensson=st.com@sourceware.org> On
> Behalf Of ???
> Sent: den 27 juli 2023 17:51
> To: Tom Tromey <tom@tromey.com>
> Cc: gdb <gdb@sourceware.org>
> Subject: Re: why gdb try to read address 0x0 when I set $pc,how to
> suppress this behavior?
> 
> Hi Tom,
> 
> Thank you for your reply, I'm trying to debug it.
> 
> I use qemu-riscv64 to run a program, then use riscv-gdb to debug it, riscv-gdb
> show that it send package "$m0,2#fb" to qemu, the full log is:
> (riscv-gdb) target remote localhost:1235
> Remote debugging using localhost:1235
> warning: No executable has been specified and target does not support
> determining executable automatically.  Try using the "file" command.
> 0x0000000000010528 in ?? ()
> (riscv-gdb) set debug remote 1
> (riscv-gdb) set $pc=0x10528
> [remote] Sending packet: $m0,2#fb
> [remote] Received Ack
> [remote] Packet received: E14
> [remote] Sending packet: $g#67
> [remote] Received Ack
> [remote] Packet received:
> 00000000000000000000000000000000900280004000000000000000000000
> 00000000000000000000000000000000000000000000000000000000000000
> 00000000000000000000000000000000000000000000000000000000000000
> 00000000000000000000000000000000000000000000000000000000000000
> 00000000000000000000000000000000000000000000000000000000000000
> 00000000000000000000000000000000000000000000000000000000000000
> 00000000000000000000000000000000000000000000000000000000000000
> 00000000000000000000000000000000000000000000000000000000000000
> 0000000000000000 [16 bytes omitted]
> [remote] Sending packet: $m10528,4#cd
> [remote] Received Ack
> [remote] Packet received: ef00e002
> [remote] Sending packet: $m10524,4#c9
> [remote] Received Ack
> [remote] Packet received: f5b70000
> [remote] Sending packet: $m0,2#fb
> [remote] Received Ack
> [remote] Packet received: E14
> 
> 
> Then I use gdb(x86-gdb) to debug this riscv-gdb, the package "$m0,2#fb"
> comes from some code about riscv frame, full log is:
> 
> Thread 1 "riscv64-unknown" hit Breakpoint 2, remote_target::putpkt_binary
> (this=0xf399d0, buf=<optimized out>, cnt=<optimized out>)
>     at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/remote.c:9602
> 9602        /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/remote.c:
> No such file or directory.
> (gdb) bt
> #0  remote_target::putpkt_binary (this=0xf399d0, buf=<optimized out>,
> cnt=<optimized out>) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/remote.c:9602
> #1  0x00000000006b7d38 in remote_target::putpkt (buf=0xfaeac0 "m0,2",
> this=0xf399d0) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/remote.c:9516
> #2  remote_target::putpkt (buf=..., this=0xf399d0) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/remote.c:907
> #3  remote_target::remote_read_bytes_1 (this=this@entry=0xf399d0,
> memaddr=<optimized out>, memaddr@entry=0,
> myaddr=myaddr@entry=0x7fffffffd968 "\204\036f",
> len_units=len_units@entry=2,
>     unit_size=unit_size@entry=1,
> xfered_len_units=xfered_len_units@entry=0x7fffffffd8f8) at
> /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/remote.c:9190
> #4  0x00000000006b7f91 in remote_target::remote_read_bytes
> (this=0xf399d0, memaddr=0, myaddr=0x7fffffffd968 "\204\036f", len=2,
> unit_size=1, xfered_len=0x7fffffffd8f8)
>     at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/remote.c:9315
> #5  0x000000000072cbcf in raw_memory_xfer_partial
> (ops=ops@entry=0xf399d0, readbuf=readbuf@entry=0x7fffffffd968
> "\204\036f", writebuf=writebuf@entry=0x0, memaddr=memaddr@entry=0,
> len=2,
>     xfered_len=xfered_len@entry=0x7fffffffd8f8) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/target.c:1514
> #6  0x000000000072cd1f in memory_xfer_partial_1
> (ops=ops@entry=0xf399d0,
> object=object@entry=TARGET_OBJECT_MEMORY,
> readbuf=readbuf@entry=0x7fffffffd968 "\204\036f",
> writebuf=writebuf@entry=0x0,
>     memaddr=memaddr@entry=0, len=len@entry=2,
> xfered_len=0x7fffffffd8f8) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/target.c:1645
> #7  0x000000000072cf9a in memory_xfer_partial (xfered_len=0x7fffffffd8f8,
> len=2, memaddr=0, writebuf=0x0, readbuf=0x7fffffffd968 "\204\036f",
> object=TARGET_OBJECT_MEMORY, ops=0xf399d0)
>     at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/target.c:1674
> #8  target_xfer_partial (ops=0xf399d0, object=TARGET_OBJECT_MEMORY,
> annex=0x0, readbuf=0x7fffffffd968 "\204\036f", writebuf=0x0, offset=0,
> len=<optimized out>, xfered_len=0x7fffffffd8f8)
>     at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/target.c:1731
> #9  0x000000000072d3a9 in target_read_partial (xfered_len=0x7fffffffd8f8,
> len=2, offset=0, buf=0x7fffffffd968 "\204\036f", annex=0x0,
> object=TARGET_OBJECT_MEMORY, ops=0xf399d0)
>     at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/target.c:2024
> #10 target_read (ops=0xf399d0,
> object=object@entry=TARGET_OBJECT_MEMORY,
> annex=annex@entry=0x0, buf=buf@entry=0x7fffffffd968 "\204\036f",
> offset=offset@entry=0, len=len@entry=2)
>     at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/target.c:2024
> #11 0x000000000072d416 in target_read_memory
> (memaddr=memaddr@entry=0, myaddr=myaddr@entry=0x7fffffffd968
> "\204\036f", len=len@entry=2)
>     at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/target.h:1394
> #12 0x00000000006c0de8 in riscv_insn::fetch_instruction
> (gdbarch=gdbarch@entry=0xf83870, addr=0, len=len@entry=0x7fffffffd9e0)
>     at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/riscv-
> tdep.c:1664
> #13 0x00000000006c0ee8 in riscv_insn::decode (this=0x7fffffffd9e0,
> gdbarch=0xf83870, pc=<optimized out>) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/riscv-tdep.c:1693
> #14 0x00000000006c187c in riscv_scan_prologue (gdbarch=0xf83870,
> start_pc=<optimized out>, end_pc=100, cache=0xe9bd20) at
> /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/riscv-tdep.c:1873
> #15 0x00000000006c1fdc in riscv_frame_cache (this_frame=0xe9bc60,
> this_cache=<optimized out>) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/riscv-tdep.c:3340
> #16 0x00000000006c2139 in riscv_frame_this_id (this_frame=<optimized
> out>, prologue_cache=<optimized out>, this_id=0xe9bcc0)
>     at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/riscv-
> tdep.c:3392
> #17 0x00000000005bf142 in compute_frame_id (fi=0xe9bc60) at
> /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/frame.c:583
> #18 0x00000000005bf3cf in get_frame_id (fi=0xe9bc60) at /toolchain/riscv-
> gnu-toolchain/build_gcc12/../gdb/gdb/frame.c:626
> #19 0x000000000078919a in value_assign (toval=0xf80340,
> fromval=0xf18890) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/valops.c:1103
> #20 0x000000000059e5fd in expression::evaluate (this=0xf8cf40,
> expect_type=0x0, noside=EVAL_NORMAL) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/eval.c:101
> #21 0x0000000000678e98 in set_command (exp=<optimized out>,
> from_tty=<optimized out>) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/printcmd.c:1520
> #22 0x00000000004eeba8 in cmd_func (cmd=<optimized out>,
> args=<optimized out>, from_tty=<optimized out>) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/cli/cli-decode.c:2514
> #23 0x00000000007486ae in execute_command (p=<optimized out>,
> p@entry=0xe4c6f0 "set $pc=0x10528", from_tty=1) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/top.c:702
> #24 0x00000000005a45bc in command_handler (command=0xe4c6f0 "set
> $pc=0x10528") at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/event-top.c:597
> #25 0x00000000005a48bb in command_line_handler (rl=...) at
> /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/event-top.c:800
> #26 0x00000000005a4ead in gdb_rl_callback_handler (rl=0xf887d0 "set
> $pc=0x10528") at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/event-top.c:229
> #27 0x000000000080f7a0 in rl_callback_read_char () at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/readline/readline/callback.c:281
> #28 0x00000000005a39fe in gdb_rl_callback_read_char_wrapper_noexcept
> () at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/event-top.c:187
> #29 0x00000000005a4d9e in gdb_rl_callback_read_char_wrapper
> (client_data=<optimized out>) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/event-top.c:203
> #30 0x00000000005a3810 in stdin_event_handler (error=<optimized out>,
> client_data=0xe47f80) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/event-top.c:524
> #31 0x00000000008b5665 in gdb_wait_for_event (block=block@entry=1) at
> /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdbsupport/event-
> loop.cc:700
> #32 0x00000000008b573d in gdb_wait_for_event (block=1) at
> /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdbsupport/event-
> loop.cc:596
> #33 gdb_do_one_event () at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdbsupport/event-loop.cc:237
> #34 0x000000000062c765 in start_event_loop () at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/main.c:421
> #35 captured_command_loop () at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/main.c:481
> #36 0x000000000062dcd5 in captured_main
> (data=data@entry=0x7fffffffe180) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/main.c:1351
> #37 gdb_main (args=args@entry=0x7fffffffe1a0) at /toolchain/riscv-gnu-
> toolchain/build_gcc12/../gdb/gdb/main.c:1366
> #38 0x000000000042f625 in main (argc=<optimized out>, argv=<optimized
> out>) at /toolchain/riscv-gnu-toolchain/build_gcc12/../gdb/gdb/gdb.c:32
> 
> I'm not familiar with gdb, so it will take some time to figure out exactly
> what's going on.
> 
> 
> > From:"Tom Tromey"<tom@tromey.com>
> > Date:Thu, Jul 27, 2023, 00:09
> > Subject:Re: why gdb try to read address 0x0 when I set $pc,how to
> suppress this behavior?
> > To:"庄秋彬"<mark.zhuang@spacemit.com>
> > Cc:"gdb@sourceware.org"<gdb@sourceware.org>
> > >>>>> 庄秋彬  <mark.zhuang@spacemit.com> writes:
> >
> > Sorry about the delay on this.
> >
> > > I use gdb with openocd to debug a riscv board, and found that when set
> > > $pc=xxx(a non-zero value) in gdb, gdb would first read the address
> > > 0x0, which caused an error on this riscv device.  Is there any
> > > documentation or suggestions for reading the source code to solve this
> > > problem?
> >
> >
> > I don't think so.  Reading 0x0 when trying to set $pc is unexpected to
> > me at least.  I'd normally approach this by debugging gdb to try to
> > understand why it is happening.
> >
> > Tom
> 
> 
> This message and any attachment are confidential and may be privileged or
> otherwise protected from disclosure. If you are not an intended recipient of
> this message, please delete it and any attachment from your system and
> notify the sender immediately by reply e-mail. Unintended recipients should
> not use, copy, disclose or take any action based on this message or any
> information contained in this message. Emails cannot be guaranteed to be
> secure or error free as they can be intercepted, amended, lost or destroyed,
> and you should take full responsibility for security checking.
> 
> 本邮件及其任何附件具有保密性质,并可能受其他保护或不允许被披露
> 给第三方。如阁下误收到本邮件,敬请立即以回复电子邮件的方式通知
> 发件人,并将本邮件及其任何附件从阁下系统中予以删除。如阁下并非
> 本邮件写明之收件人,敬请切勿使用、复制、披露本邮件或其任何内容
> ,亦请切勿依本邮件或其任何内容而采取任何行动。电子邮件无法保证
> 是一种安全和不会出现任何差错的通信方式,可能会被拦截、修改、丢
> 失或损坏,收件人需自行负责做好安全检查。

      reply	other threads:[~2023-07-27 17:36 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-07-19  7:59 庄秋彬
2023-07-19 12:44 ` 庄秋彬
2023-07-26 16:09 ` Tom Tromey
2023-07-27 15:51   ` 庄秋彬
2023-07-27 17:36     ` Torbjorn SVENSSON [this message]

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=AM6PR10MB219793057F9B1278773129718101A@AM6PR10MB2197.EURPRD10.PROD.OUTLOOK.COM \
    --to=torbjorn.svensson@st.com \
    --cc=gdb@sourceware.org \
    --cc=mark.zhuang@spacemit.com \
    --cc=tom@tromey.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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).