You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@nuttx.apache.org by GitBox <gi...@apache.org> on 2020/11/22 06:27:58 UTC
[GitHub] [incubator-nuttx] btashton opened a new issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
btashton opened a new issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366
When I try to dump the threads with `nxthread_bt_all` nothing is shown and then it crashes the target. I am using this with STM32H7 and the ARM Embedded 2020-q2-update toolchain:
```
GNU gdb (GNU Arm Embedded Toolchain 9-2020-q2-update) 8.3.1.20191211-git
```
```
nx_start () at init/nx_start.c:832
832 up_idle();
(gdb) bt
#0 nx_start () at init/nx_start.c:832
#1 0x080003ce in __start () at chip/stm32_start.c:426
(gdb) nxthread
target examined
_target_arch.name=armv7e-m
$_target_has_fpu : 0
$_target_has_smp : 0
saved current_tcb (pid=0)
Missing argument 0 in user function.
(gdb) nxthread_all_bt
* 0 Thread 0x24000504 (Name: Idle Task, State: Running, Priority: 0) 0x80025c6 in nx_start()
#0 nx_start () at init/nx_start.c:832
#1 0x080003ce in __start () at chip/stm32_start.c:426
1 Thread 0x38000490 (Name: hpwork, State: Waiting,Signal, Priority: 224) 0x0 in No()
#0 0x00000000 in ?? ()
2 Thread 0x38001130 (Name: lpwork, State: Waiting,Signal, Priority: 100) 0x0 in No()
#0 0x00000000 in ?? ()
3 Thread 0x38001dd0 (Name: init, State: Waiting,Semaphore, Priority: 100) 0x0 in No()
#0 0x00000000 in ?? ()
4 Thread 0x38002b60 (Name: usbhost, State: Waiting,Semaphore, Priority: 100) 0x0 in No()
#0 0x00000000 in ?? ()
5 Thread 0x38003420 (Name: Telnet daemon, State: Waiting,Semaphore, Priority: 100) 0x0 in No()
#0 0x00000000 in ?? ()
```
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] btashton commented on issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
btashton commented on issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366#issuecomment-731705585
@masayuki2009 any idea what is going on here? The `Missing argument 0 in user function.` does not look good.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] masayuki2009 commented on issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
masayuki2009 commented on issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366#issuecomment-731716675
> @masayuki2009 any idea what is going on here? The `Missing argument 0 in user function.` does not look good.
What happens if you call ```info_nxthread``` before calling ```nxthread``` ?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] masayuki2009 commented on issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
masayuki2009 commented on issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366#issuecomment-731854500
>There are 22 different definitions of this {}_fpuconfig symbol so I am not quite sure what the change should be here.
@btashton
I came up with a new idea to check if the target uses FPU.
It just checks the size of tcb.xcp.regs[].
I will try this idea later.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] btashton commented on issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
btashton commented on issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366#issuecomment-731717548
Still looks like it is not grabbing the stackframes correctly:
```c
(gdb) c
Continuing.
^C
Program received signal SIGINT, Interrupt.
stm32_phyread (phydevaddr=0, phyregaddr=1, value=0x38002866) at chip/stm32_ethernet.c:3228
3228 if ((stm32_getreg(STM32_ETH_MACMDIOAR) & ETH_MACMDIOAR_MB) == 0)
(gdb) info_nxthreads
0 Thread 0x240004a4 (Name: Idle Task, State: Ready, Priority: 0) 0x0 in No()
1 Thread 0x38000490 (Name: hpwork, State: Waiting,Signal, Priority: 224) 0x0 in No()
2 Thread 0x38001130 (Name: lpwork, State: Ready, Priority: 100) 0x0 in No()
3 Thread 0x38001dd0 (Name: init, State: Running, Priority: 100) 0x0 in No()
4 Thread 0x38002b60 (Name: usbhost, State: Ready, Priority: 100) 0x0 in No()
5 Thread 0x38003420 (Name: Telnet daemon, State: Ready, Priority: 100) 0x0 in No()
(gdb) bt
#0 stm32_phyread (phydevaddr=0, phyregaddr=1, value=0x38002866) at chip/stm32_ethernet.c:3228
#1 0x0800a264 in stm32_phyinit (priv=0x240068e0 <g_stm32ethmac>) at chip/stm32_ethernet.c:3517
#2 0x0800a80c in stm32_ethconfig (priv=0x240068e0 <g_stm32ethmac>) at chip/stm32_ethernet.c:4312
#3 0x08009c90 in stm32_ifup (dev=0x2400693c <g_stm32ethmac+92>) at chip/stm32_ethernet.c:2580
#4 0x08032034 in netdev_ifup (dev=0x2400693c <g_stm32ethmac+92>) at netdev/netdev_ioctl.c:1833
#5 0x08031bce in netdev_ifr_ioctl (psock=0x38002164, cmd=1818, req=0x3800297c) at netdev/netdev_ioctl.c:907
#6 0x08031f6e in psock_vioctl (psock=0x38002164, cmd=1818, ap=...) at netdev/netdev_ioctl.c:1681
#7 0x08032006 in netdev_vioctl (sockfd=8, cmd=1818, ap=...) at netdev/netdev_ioctl.c:1810
#8 0x0801edfa in nx_vioctl (fd=8, req=1818, ap=...) at vfs/fs_ioctl.c:151
#9 0x0801ef38 in ioctl (fd=8, req=1818) at vfs/fs_ioctl.c:259
#10 0x0800add4 in netlib_ifup (ifname=0x8038ee4 "eth0") at netlib_setifstatus.c:97
#11 0x0800acb8 in netinit_net_bringup () at netinit.c:603
#12 0x0800ad60 in netinit_configure () at netinit.c:701
#13 0x0800ad6c in netinit_bringup () at netinit.c:1045
#14 0x0800aabe in nsh_consolemain (argc=1, argv=0x38002a58) at nsh_consolemain.c:104
#15 0x0800aa6c in nsh_main (argc=1, argv=0x38002a58) at nsh_main.c:168
#16 0x08006a00 in nxtask_startup (entrypt=0x800aa0d <nsh_main>, argc=1, argv=0x38002a58) at sched/task_startup.c:165
#17 0x08002b36 in nxtask_start () at task/task_start.c:144
#18 0x00000000 in ?? ()
(gdb)
```
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] btashton commented on issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
btashton commented on issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366#issuecomment-731850217
@masayuki2009 I had to make an additional change to use `lookup_symbol` instead of `lookup_global_symbol` which returns a tuple with the first value as the symbol (or None).
`python if (type(gdb.lookup_symbol("stm32_fpuconfig")[0]) is gdb.Symbol) : \`
There are 22 different definitions of this {}_fpuconfig symbol so I am not quite sure what the change should be here.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] btashton closed issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
btashton closed issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] btashton edited a comment on issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
btashton edited a comment on issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366#issuecomment-731850217
@masayuki2009 I had to make an additional change to use `lookup_symbol` instead of `lookup_global_symbol` which returns a tuple with the first value as the symbol (or None).
`python if (type(gdb.lookup_symbol("stm32_fpuconfig")[0]) is gdb.Symbol) : \`
```
(gdb) info_nxthreads
0 Thread 0x24000358 (Name: Idle Task, State: Running, Priority: 0) 0x800238e in nx_start()
1 Thread 0x380002f0 (Name: hpwork, State: Waiting,Signal, Priority: 224) 0x80091bc in arm_switchcontext()
2 Thread 0x38000ec0 (Name: lpwork, State: Waiting,Signal, Priority: 100) 0x80091bc in arm_switchcontext()
3 Thread 0x38001a90 (Name: init, State: Waiting,Semaphore, Priority: 100) 0x80091bc in arm_switchcontext()
* 4 Thread 0x38002710 (Name: usbhost, State: Waiting,Semaphore, Priority: 100) 0x80091bc in arm_switchcontext()
```
There are 22 different definitions of this {}_fpuconfig symbol so I am not quite sure what the change should be here.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] masayuki2009 commented on issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
masayuki2009 commented on issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366#issuecomment-731719828
>Still looks like it is not grabbing the stackframes correctly:
@btashton
Which board and configuration are you using now?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] masayuki2009 edited a comment on issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
masayuki2009 edited a comment on issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366#issuecomment-731716675
> @masayuki2009 any idea what is going on here? The `Missing argument 0 in user function.` does not look good.
Year, this should be avoided, because ```nxthread``` needs to have an argument. (i.e. ```nxthread 0```)
By the way, what happens if you call ```info_nxthread``` before calling ```nxthread``` ?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] masayuki2009 commented on issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
masayuki2009 commented on issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366#issuecomment-731731581
@btashton
Perhaps, I found the root cause.
Actually, ```nucleo-.h743zi:otg_fs_host``` uses FPU but it was not detected by the nuttx-gdbinit script.
Currently, it just checks the symbol named ```fpuconfig``` but ```stm32_fpuconfig``` was used instead.
For the interim solution, could you try it by replacing fpuconfig with stm32_fpuconfig?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] masayuki2009 edited a comment on issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
masayuki2009 edited a comment on issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366#issuecomment-731718113
> Still looks like it is not grabbing the stackframes correctly:
Hmm, it's strange...
```
$ arm-none-eabi-gdb-py --version
GNU gdb (GNU Arm Embedded Toolchain 9-2020-q2-update) 8.3.1.20191211-git
Copyright (C) 2019 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
```
I've just tried ```stm32f4discovery:rndis```
```
(gdb) target extended-remote:3333
Remote debugging using :3333
up_idle () at chip/stm32_idle.c:208
208 END_IDLE();
(gdb) info_nxthreads
target examined
_target_arch.name=armv7e-m
$_target_has_fpu : 0
$_target_has_smp : 0
saved current_tcb (pid=0)
* 0 Thread 0x20000308 (Name: Idle Task, State: Running, Priority: 0) 0x80051c6 in up_idle()
1 Thread 0x20004a80 (Name: lpwork, State: Waiting,Signal, Priority: 100) 0x8005778 in arm_switchcontext()
2 Thread 0x200056c0 (Name: init, State: Waiting,Semaphore, Priority: 100) 0x8005778 in arm_switchcontext()
3 Thread 0x20007820 (Name: Telnet daemon, State: Waiting,Semaphore, Priority: 100) 0x8005778 in arm_switchcontext()
```
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] btashton commented on issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
btashton commented on issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366#issuecomment-731721097
I am using `nucleo-h743zi:otg_fs_host` which was just merged in, the only change I made is setting the thread name limit to 31 instead of 0. When I used `stm32f429i-disco:nsh` everything was fine.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] masayuki2009 edited a comment on issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
masayuki2009 edited a comment on issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366#issuecomment-731731581
@btashton
Perhaps, I found the root cause.
Actually, ```nucleo-.h743zi:otg_fs_host``` uses FPU but it was not detected by the nuttx-gdbinit script.
Currently, it just checks the symbol named ```fpuconfig``` but ```stm32_fpuconfig``` was used instead.
For the interim solution, could you try it by replacing ```fpuconfig``` with ```stm32_fpuconfig```?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [incubator-nuttx] masayuki2009 commented on issue #2366: debug: nuttx-gdbinit not producing backtraces on armv7e-m
Posted by GitBox <gi...@apache.org>.
masayuki2009 commented on issue #2366:
URL: https://github.com/apache/incubator-nuttx/issues/2366#issuecomment-731718113
> Still looks like it is not grabbing the stackframes correctly:
Hmm, it's strange...
```
$ arm-none-eabi-gdb --version
GNU gdb (GNU Arm Embedded Toolchain 9-2020-q2-update) 8.3.1.20191211-git
Copyright (C) 2019 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
```
I've just tried ```stm32f4discovery:rndis```
```
(gdb) target extended-remote:3333
Remote debugging using :3333
up_idle () at chip/stm32_idle.c:208
208 END_IDLE();
(gdb) info_nxthreads
target examined
_target_arch.name=armv7e-m
$_target_has_fpu : 0
$_target_has_smp : 0
saved current_tcb (pid=0)
* 0 Thread 0x20000308 (Name: Idle Task, State: Running, Priority: 0) 0x80051c6 in up_idle()
1 Thread 0x20004a80 (Name: lpwork, State: Waiting,Signal, Priority: 100) 0x8005778 in arm_switchcontext()
2 Thread 0x200056c0 (Name: init, State: Waiting,Semaphore, Priority: 100) 0x8005778 in arm_switchcontext()
3 Thread 0x20007820 (Name: Telnet daemon, State: Waiting,Semaphore, Priority: 100) 0x8005778 in arm_switchcontext()
```
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org