[musl] Segmentation fault musl 1.2.4

[musl] Segmentation fault musl 1.2.4

[Cody Wetzel]

[-- Attachment #1: Type: text/plain, Size: 411 bytes --]

Hello,

I am running into a segmentation fault on musl 1.2.4 on my QNAP NAS.  QNAP
updated the pagesize to 32k and I believe musl isn't handling it
gracefully.  musl 1.2.3 does not have this problem.

https://www.qnap.com/da-dk/how-to/faq/article/why-do-the-installed-third-party-containers-not-run-successfully-on-specific-32-bit-arm-devices

Please CC me.

-- 
Cody Wetzel
codyawetzel@gmail.com
(402)490-9242

[-- Attachment #2: Type: text/html, Size: 893 bytes --]

Re: [musl] Segmentation fault musl 1.2.4

[Rich Felker]

On Wed, Dec 20, 2023 at 03:53:38PM -0600, Cody Wetzel wrote:
> Hello,
> 
> I am running into a segmentation fault on musl 1.2.4 on my QNAP NAS.  QNAP
> updated the pagesize to 32k and I believe musl isn't handling it
> gracefully.  musl 1.2.3 does not have this problem.
> 
> https://www.qnap.com/da-dk/how-to/faq/article/why-do-the-installed-third-party-containers-not-run-successfully-on-specific-32-bit-arm-devices
> 
> Please CC me.

Without any indication of where the segfault was, it's impossible to
know whether it has anything to do with the change you cited or even
with musl. There is no reason oversized pages shouldn't function with
musl. They will lead to significantly worse memory usage and possibly
worse performance, but of course should not crash. Using 4k pages is
highly recommended if you have the option.

If you think the crash is caused by or related to a bug in musl,
please report what application is crashing and let us know if you can
use a debugger to get some basic information on the location of the
crash.

Rich

Re: [musl] Segmentation fault musl 1.2.4

[Rich Felker]

On Wed, Dec 20, 2023 at 10:38:01PM -0500, Rich Felker wrote:
> On Wed, Dec 20, 2023 at 03:53:38PM -0600, Cody Wetzel wrote:
> > Hello,
> > 
> > I am running into a segmentation fault on musl 1.2.4 on my QNAP NAS.  QNAP
> > updated the pagesize to 32k and I believe musl isn't handling it
> > gracefully.  musl 1.2.3 does not have this problem.
> > 
> > https://www.qnap.com/da-dk/how-to/faq/article/why-do-the-installed-third-party-containers-not-run-successfully-on-specific-32-bit-arm-devices
> > 
> > Please CC me.

I did CC but your mail provider is refusing the mail. Can you please
complain to them?

Rich

Re: [musl] Segmentation fault musl 1.2.4

[Markus Wichmann]

Am Wed, Dec 20, 2023 at 03:53:38PM -0600 schrieb Cody Wetzel:
> Hello,
>
> I am running into a segmentation fault on musl 1.2.4 on my QNAP NAS.  QNAP
> updated the pagesize to 32k and I believe musl isn't handling it
> gracefully.  musl 1.2.3 does not have this problem.
>
> https://www.qnap.com/da-dk/how-to/faq/article/why-do-the-installed-third-party-containers-not-run-successfully-on-specific-32-bit-arm-devices
>
> Please CC me.
>
> --
> Cody Wetzel
> codyawetzel@gmail.com
> (402)490-9242

Gonna try my luck since Rich apparently had none: Please procure some
data from the crash, such as a core dump or a backtrace. Without any
indication of what is going on, there is no telling whether the issue
you are experiencing even is in musl, let alone where it is.

BTW, the linked website is a hoot and a half. "We made this change to
improve your user experience, but it may cause your applications to have
less memory and also crash. Your only recourse is to 'Verify
compatibility', i.e. sink or swim."

Ciao,
Markus

Re: [musl] Segmentation fault musl 1.2.4

[Cody Wetzel]

[-- Attachment #1: Type: text/plain, Size: 1677 bytes --]

Markus,

No kidding.  I've been dealing with issues running Home Assistant for quite
some time thanks to this page size change.  Since I'm not the most well
versed in Linux I will need to look into how to get logs.  I do have a
Github issue you can look at for more details in the meantime.

https://github.com/home-assistant/core/issues/86589

Note that I think there is an issue in rustc also.

Thanks!

Cody

On Thu, Dec 21, 2023 at 1:57 AM Markus Wichmann <nullplan@gmx.net> wrote:

> Am Wed, Dec 20, 2023 at 03:53:38PM -0600 schrieb Cody Wetzel:
> > Hello,
> >
> > I am running into a segmentation fault on musl 1.2.4 on my QNAP NAS.
> QNAP
> > updated the pagesize to 32k and I believe musl isn't handling it
> > gracefully.  musl 1.2.3 does not have this problem.
> >
> >
> https://www.qnap.com/da-dk/how-to/faq/article/why-do-the-installed-third-party-containers-not-run-successfully-on-specific-32-bit-arm-devices
> >
> > Please CC me.
> >
> > --
> > Cody Wetzel
> > codyawetzel@gmail.com
> > (402)490-9242
>
> Gonna try my luck since Rich apparently had none: Please procure some
> data from the crash, such as a core dump or a backtrace. Without any
> indication of what is going on, there is no telling whether the issue
> you are experiencing even is in musl, let alone where it is.
>
> BTW, the linked website is a hoot and a half. "We made this change to
> improve your user experience, but it may cause your applications to have
> less memory and also crash. Your only recourse is to 'Verify
> compatibility', i.e. sink or swim."
>
> Ciao,
> Markus
>


-- 
Cody Wetzel
codyawetzel@gmail.com
(402)490-9242

[-- Attachment #2: Type: text/html, Size: 2681 bytes --]

Re: [musl] Segmentation fault musl 1.2.4

[Natanael Copa]

On Thu, 21 Dec 2023 12:39:35 -0600
Cody Wetzel <codyawetzel@gmail.com> wrote:

> Markus,
> 
> No kidding.  I've been dealing with issues running Home Assistant for quite
> some time thanks to this page size change.  Since I'm not the most well
> versed in Linux I will need to look into how to get logs.  I do have a
> Github issue you can look at for more details in the meantime.
> 
> https://github.com/home-assistant/core/issues/86589

Those issues will keep popping up til we fix them, and to fix them we
will need to know exactly what to fix, a backtrace may help with that.
Alternatively we'd need a simple way to reproduce the issue.

I found this following the above issue:
https://gitlab.alpinelinux.org/alpine/aports/-/issues/15167
This issue looks more like a problem in busybox than musl.

-nc


> Note that I think there is an issue in rustc also.
> 
> Thanks!
> 
> Cody
> 
> On Thu, Dec 21, 2023 at 1:57*AM Markus Wichmann <nullplan@gmx.net>
> wrote:
> 
> > Am Wed, Dec 20, 2023 at 03:53:38PM -0600 schrieb Cody Wetzel:  
> > > Hello,
> > >
> > > I am running into a segmentation fault on musl 1.2.4 on my QNAP
> > > NAS.  
> > QNAP  
> > > updated the pagesize to 32k and I believe musl isn't handling it
> > > gracefully.  musl 1.2.3 does not have this problem.
> > >
> > >  
> > https://www.qnap.com/da-dk/how-to/faq/article/why-do-the-installed-third-party-containers-not-run-successfully-on-specific-32-bit-arm-devices
> >  
> > >
> > > Please CC me.
> > >
> > > --
> > > Cody Wetzel
> > > codyawetzel@gmail.com
> > > (402)490-9242  
> >
> > Gonna try my luck since Rich apparently had none: Please procure
> > some data from the crash, such as a core dump or a backtrace.
> > Without any indication of what is going on, there is no telling
> > whether the issue you are experiencing even is in musl, let alone
> > where it is.
> >
> > BTW, the linked website is a hoot and a half. "We made this change
> > to improve your user experience, but it may cause your applications
> > to have less memory and also crash. Your only recourse is to 'Verify
> > compatibility', i.e. sink or swim."
> >
> > Ciao,
> > Markus
> >  
> 
> 

Re: [musl] Segmentation fault musl 1.2.4

[Cody Wetzel]

[-- Attachment #1: Type: text/plain, Size: 2836 bytes --]

I am able to run:

# apk upgrade busybox --repository=
https://dl-cdn.alpinelinux.org/alpine/v3.18/main --repository=https://dl-cd
n.alpinelinux.org/alpine/v3.18/community

and not create a segmentation fault on Alpine 3.17

I will be on Christmas vacation for the next week but when I get back I can
try to get additional details.

Thanks!

Cody

On Thu, Dec 21, 2023 at 3:25 PM Natanael Copa <ncopa@alpinelinux.org> wrote:

> On Thu, 21 Dec 2023 12:39:35 -0600
> Cody Wetzel <codyawetzel@gmail.com> wrote:
>
> > Markus,
> >
> > No kidding.  I've been dealing with issues running Home Assistant for
> quite
> > some time thanks to this page size change.  Since I'm not the most well
> > versed in Linux I will need to look into how to get logs.  I do have a
> > Github issue you can look at for more details in the meantime.
> >
> > https://github.com/home-assistant/core/issues/86589
>
> Those issues will keep popping up til we fix them, and to fix them we
> will need to know exactly what to fix, a backtrace may help with that.
> Alternatively we'd need a simple way to reproduce the issue.
>
> I found this following the above issue:
> https://gitlab.alpinelinux.org/alpine/aports/-/issues/15167
> This issue looks more like a problem in busybox than musl.
>
> -nc
>
>
> > Note that I think there is an issue in rustc also.
> >
> > Thanks!
> >
> > Cody
> >
> > On Thu, Dec 21, 2023 at 1:57*AM Markus Wichmann <nullplan@gmx.net>
> > wrote:
> >
> > > Am Wed, Dec 20, 2023 at 03:53:38PM -0600 schrieb Cody Wetzel:
> > > > Hello,
> > > >
> > > > I am running into a segmentation fault on musl 1.2.4 on my QNAP
> > > > NAS.
> > > QNAP
> > > > updated the pagesize to 32k and I believe musl isn't handling it
> > > > gracefully.  musl 1.2.3 does not have this problem.
> > > >
> > > >
> > >
> https://www.qnap.com/da-dk/how-to/faq/article/why-do-the-installed-third-party-containers-not-run-successfully-on-specific-32-bit-arm-devices
> > >
> > > >
> > > > Please CC me.
> > > >
> > > > --
> > > > Cody Wetzel
> > > > codyawetzel@gmail.com
> > > > (402)490-9242
> > >
> > > Gonna try my luck since Rich apparently had none: Please procure
> > > some data from the crash, such as a core dump or a backtrace.
> > > Without any indication of what is going on, there is no telling
> > > whether the issue you are experiencing even is in musl, let alone
> > > where it is.
> > >
> > > BTW, the linked website is a hoot and a half. "We made this change
> > > to improve your user experience, but it may cause your applications
> > > to have less memory and also crash. Your only recourse is to 'Verify
> > > compatibility', i.e. sink or swim."
> > >
> > > Ciao,
> > > Markus
> > >
> >
> >
>
>

-- 
Cody Wetzel
codyawetzel@gmail.com
(402)490-9242

[-- Attachment #2: Type: text/html, Size: 4654 bytes --]

Re: [musl] Segmentation fault musl 1.2.4

[Jeffrey Walton]

On Wed, Dec 20, 2023 at 4:54 PM Cody Wetzel <codyawetzel@gmail.com> wrote:
>
> I am running into a segmentation fault on musl 1.2.4 on my QNAP NAS.  QNAP updated the pagesize to 32k and I believe musl isn't handling it gracefully.  musl 1.2.3 does not have this problem.
>
> https://www.qnap.com/da-dk/how-to/faq/article/why-do-the-installed-third-party-containers-not-run-successfully-on-specific-32-bit-arm-devices

Related, Apple M1's use a 32k page size. It is baked into the MMU. 32k
memory pages has caused a fair amount of trouble for Asashi Linux,
especially for some userland programs.
See<https://github.com/AsahiLinux/docs/wiki/Broken-Software>.

Jeff

Re: [musl] Segmentation fault musl 1.2.4

[Jeffrey Walton]

On Thu, Dec 21, 2023 at 9:20 PM Jeffrey Walton <noloader@gmail.com> wrote:
>
> On Wed, Dec 20, 2023 at 4:54 PM Cody Wetzel <codyawetzel@gmail.com> wrote:
> >
> > I am running into a segmentation fault on musl 1.2.4 on my QNAP NAS.  QNAP updated the pagesize to 32k and I believe musl isn't handling it gracefully.  musl 1.2.3 does not have this problem.
> >
> > https://www.qnap.com/da-dk/how-to/faq/article/why-do-the-installed-third-party-containers-not-run-successfully-on-specific-32-bit-arm-devices
>
> Related, Apple M1's use a 32k page size. It is baked into the MMU. 32k
> memory pages has caused a fair amount of trouble for Asashi Linux,
> especially for some userland programs.
> See<https://github.com/AsahiLinux/docs/wiki/Broken-Software>.

My bad, s/32k/16k.

Jeff

Re: [musl] Segmentation fault musl 1.2.4

[Cody Wetzel]

[-- Attachment #1: Type: text/plain, Size: 4508 bytes --]

Hello musl team,

I tried getting a core dump but I'm not sure if I'm doing something wrong...

/ # cat /proc/sys/kernel/core_pattern/tmp/core-%e-%s-%u-%g-%p-%t/ #
apk upgrade busybox
--repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/main
--repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/communityfetch
https://dl-cdn.alpinelinux.org/alpine/v3.18/community/armv7/APKINDEX.tar.gzfetch
https://dl-cdn.alpinelinux.org/alpine/v3.18/main/armv7/APKINDEX.tar.gzfetch
https://dl-cdn.alpinelinux.org/alpine/v3.17/main/armv7/APKINDEX.tar.gzfetch
https://dl-cdn.alpinelinux.org/alpine/v3.17/community/armv7/APKINDEX.tar.gz(1/3)
Upgrading busybox (1.35.0-r29 -> 1.36.1-r5)Executing
busybox-1.36.1-r5.post-upgrade(2/3) Upgrading busybox-binsh
(1.35.0-r29 -> 1.36.1-r5)(3/3) Upgrading ssl_client (1.35.0-r29 ->
1.36.1-r5)Executing busybox-1.36.1-r5.triggerOK: 5 MiB in 15 packages/
# apk upgrade musl
--repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/main
--repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/community(1/1)
Upgrading musl (1.2.3-r5 -> 1.2.4-r2)OK: 5 MiB in 15 packages/ # apk
upgrade rust --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/main
--repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/communitySegmentation
fault/ # apk -hSegmentation fault/ # ls /tmp/core*Segmentation fault/
# cd /tmp/tmp # lsSegmentation fault
>
>
Thanks!

Cody

On Thu, Dec 21, 2023 at 3:57 PM Cody Wetzel <codyawetzel@gmail.com> wrote:

> I am able to run:
>
> # apk upgrade busybox --repository=
> https://dl-cdn.alpinelinux.org/alpine/v3.18/main --repository=
> https://dl-cd
> n.alpinelinux.org/alpine/v3.18/community
>
> and not create a segmentation fault on Alpine 3.17
>
> I will be on Christmas vacation for the next week but when I get back I
> can try to get additional details.
>
> Thanks!
>
> Cody
>
> On Thu, Dec 21, 2023 at 3:25 PM Natanael Copa <ncopa@alpinelinux.org>
> wrote:
>
>> On Thu, 21 Dec 2023 12:39:35 -0600
>> Cody Wetzel <codyawetzel@gmail.com> wrote:
>>
>> > Markus,
>> >
>> > No kidding.  I've been dealing with issues running Home Assistant for
>> quite
>> > some time thanks to this page size change.  Since I'm not the most well
>> > versed in Linux I will need to look into how to get logs.  I do have a
>> > Github issue you can look at for more details in the meantime.
>> >
>> > https://github.com/home-assistant/core/issues/86589
>>
>> Those issues will keep popping up til we fix them, and to fix them we
>> will need to know exactly what to fix, a backtrace may help with that.
>> Alternatively we'd need a simple way to reproduce the issue.
>>
>> I found this following the above issue:
>> https://gitlab.alpinelinux.org/alpine/aports/-/issues/15167
>> This issue looks more like a problem in busybox than musl.
>>
>> -nc
>>
>>
>> > Note that I think there is an issue in rustc also.
>> >
>> > Thanks!
>> >
>> > Cody
>> >
>> > On Thu, Dec 21, 2023 at 1:57*AM Markus Wichmann <nullplan@gmx.net>
>> > wrote:
>> >
>> > > Am Wed, Dec 20, 2023 at 03:53:38PM -0600 schrieb Cody Wetzel:
>> > > > Hello,
>> > > >
>> > > > I am running into a segmentation fault on musl 1.2.4 on my QNAP
>> > > > NAS.
>> > > QNAP
>> > > > updated the pagesize to 32k and I believe musl isn't handling it
>> > > > gracefully.  musl 1.2.3 does not have this problem.
>> > > >
>> > > >
>> > >
>> https://www.qnap.com/da-dk/how-to/faq/article/why-do-the-installed-third-party-containers-not-run-successfully-on-specific-32-bit-arm-devices
>> > >
>> > > >
>> > > > Please CC me.
>> > > >
>> > > > --
>> > > > Cody Wetzel
>> > > > codyawetzel@gmail.com
>> > > > (402)490-9242
>> > >
>> > > Gonna try my luck since Rich apparently had none: Please procure
>> > > some data from the crash, such as a core dump or a backtrace.
>> > > Without any indication of what is going on, there is no telling
>> > > whether the issue you are experiencing even is in musl, let alone
>> > > where it is.
>> > >
>> > > BTW, the linked website is a hoot and a half. "We made this change
>> > > to improve your user experience, but it may cause your applications
>> > > to have less memory and also crash. Your only recourse is to 'Verify
>> > > compatibility', i.e. sink or swim."
>> > >
>> > > Ciao,
>> > > Markus
>> > >
>> >
>> >
>>
>>
>
> --
> Cody Wetzel
> codyawetzel@gmail.com
> (402)490-9242
>


-- 
Cody Wetzel
codyawetzel@gmail.com
(402)490-9242

[-- Attachment #2: Type: text/html, Size: 9559 bytes --]

Re: [musl] Segmentation fault musl 1.2.4

[Leah Neukirchen]

Cody Wetzel <codyawetzel@gmail.com> writes:

> Hello musl team,
>
> I tried getting a core dump but I'm not sure if I'm doing something wrong...

You need to run

% ulimit -c unlimited

to enable coredumps for the session.

-- 
Leah Neukirchen  <leah@vuxu.org>  https://leahneukirchen.org/

Re: [musl] Segmentation fault musl 1.2.4

[Cody Wetzel]

[-- Attachment #1: Type: text/plain, Size: 751 bytes --]

I thought I had tried that command and just to confirm I tried it again.  I
receive no core dump however.  To further clarify, even if a core dump did
occur using all basic commands result in segfault after upgrading musl.  I
can't use ls or mv for example.

Thanks!

Cody

On Wed, Jan 3, 2024 at 11:26 AM Leah Neukirchen <leah@vuxu.org> wrote:

> Cody Wetzel <codyawetzel@gmail.com> writes:
>
> > Hello musl team,
> >
> > I tried getting a core dump but I'm not sure if I'm doing something
> wrong...
>
> You need to run
>
> % ulimit -c unlimited
>
> to enable coredumps for the session.
>
> --
> Leah Neukirchen  <leah@vuxu.org>  https://leahneukirchen.org/
>


-- 
Cody Wetzel
codyawetzel@gmail.com
(402)490-9242

[-- Attachment #2: Type: text/html, Size: 1495 bytes --]

Re: [musl] Segmentation fault musl 1.2.4

[Szabolcs Nagy]

* Cody Wetzel <codyawetzel@gmail.com> [2024-01-03 11:20:29 -0600]:
> Hello musl team,
> 
> I tried getting a core dump but I'm not sure if I'm doing something wrong...
> 
> / # cat /proc/sys/kernel/core_pattern/tmp/core-%e-%s-%u-%g-%p-%t/ #
> apk upgrade busybox
> --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/main
> --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/communityfetch
> https://dl-cdn.alpinelinux.org/alpine/v3.18/community/armv7/APKINDEX.tar.gzfetch
> https://dl-cdn.alpinelinux.org/alpine/v3.18/main/armv7/APKINDEX.tar.gzfetch
> https://dl-cdn.alpinelinux.org/alpine/v3.17/main/armv7/APKINDEX.tar.gzfetch
> https://dl-cdn.alpinelinux.org/alpine/v3.17/community/armv7/APKINDEX.tar.gz(1/3)
> Upgrading busybox (1.35.0-r29 -> 1.36.1-r5)Executing
> busybox-1.36.1-r5.post-upgrade(2/3) Upgrading busybox-binsh
> (1.35.0-r29 -> 1.36.1-r5)(3/3) Upgrading ssl_client (1.35.0-r29 ->
> 1.36.1-r5)Executing busybox-1.36.1-r5.triggerOK: 5 MiB in 15 packages/
> # apk upgrade musl
> --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/main
> --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/community(1/1)
> Upgrading musl (1.2.3-r5 -> 1.2.4-r2)OK: 5 MiB in 15 packages/ # apk
> upgrade rust --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/main
> --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/communitySegmentation
> fault/ # apk -hSegmentation fault/ # ls /tmp/core*Segmentation fault/
> # cd /tmp/tmp # lsSegmentation fault

i'd

# cp /lib/ld-musl-armhf.so.1 /tmp
# apk add gdb musl-dbg apk-tools-static

then upgrade musl using apk.static, then debug via

# /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args ls -l /tmp

or analyze a core dump, you can always install more debug tools
via apk.static and run commands using the old ld-musl-armhf.so.1

in gdb, you want to do

bt
disas $pc-40,+80
info reg
info proc map

as a starting point and post the results.

strace output can be useful too as well as readelf -aW of ld.so
depending on what is going on.

Re: [musl] Segmentation fault musl 1.2.4

[Cody Wetzel]

[-- Attachment #1: Type: text/plain, Size: 5319 bytes --]

So maybe I'm not understanding how gdb works or is used. I'm getting no
meaningful output even though these commands should result in a
segmentation fault.

> / # gdb
> Segmentation fault
> / # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args ls -l /tmp
> GNU gdb (GDB) 12.1
> Copyright (C) 2022 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.
> Type "show copying" and "show warranty" for details.
> This GDB was configured as "armv7-alpine-linux-musleabihf".
> Type "show configuration" for configuration details.
> For bug reporting instructions, please see:
> <https://www.gnu.org/software/gdb/bugs/>.
> Find the GDB manual and other documentation resources online at:
> <http://www.gnu.org/software/gdb/documentation/>.
> For help, type "help".
> Type "apropos word" to search for commands related to "word"...
> Reading symbols from ls...
> (No debugging symbols found in ls)
> (gdb)
> quit
> / # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args cd /tmp
> GNU gdb (GDB) 12.1
> Copyright (C) 2022 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.
> Type "show copying" and "show warranty" for details.
> This GDB was configured as "armv7-alpine-linux-musleabihf".
> Type "show configuration" for configuration details.
> For bug reporting instructions, please see:
> <https://www.gnu.org/software/gdb/bugs/>.
> Find the GDB manual and other documentation resources online at:
> <http://www.gnu.org/software/gdb/documentation/>.
> For help, type "help".
> Type "apropos word" to search for commands related to "word"...
> cd: No such file or directory.
> (gdb)
> quit
> / # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args apk -h /tmp
> GNU gdb (GDB) 12.1
> Copyright (C) 2022 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.
> Type "show copying" and "show warranty" for details.
> This GDB was configured as "armv7-alpine-linux-musleabihf".
> Type "show configuration" for configuration details.
> For bug reporting instructions, please see:
> <https://www.gnu.org/software/gdb/bugs/>.
> Find the GDB manual and other documentation resources online at:
> <http://www.gnu.org/software/gdb/documentation/>.
> For help, type "help".
> Type "apropos word" to search for commands related to "word"...
> Reading symbols from apk...
> (No debugging symbols found in apk)
> (gdb) bt
> No stack.
> (gdb) disas $pc-40,+80
> No registers.
> (gdb) info reg
> The program has no registers now.


On Thu, Jan 4, 2024 at 8:48 AM Szabolcs Nagy <nsz@port70.net> wrote:

> * Cody Wetzel <codyawetzel@gmail.com> [2024-01-03 11:20:29 -0600]:
> > Hello musl team,
> >
> > I tried getting a core dump but I'm not sure if I'm doing something
> wrong...
> >
> > / # cat /proc/sys/kernel/core_pattern/tmp/core-%e-%s-%u-%g-%p-%t/ #
> > apk upgrade busybox
> > --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/main
> > --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/communityfetch
> >
> https://dl-cdn.alpinelinux.org/alpine/v3.18/community/armv7/APKINDEX.tar.gzfetch
> >
> https://dl-cdn.alpinelinux.org/alpine/v3.18/main/armv7/APKINDEX.tar.gzfetch
> >
> https://dl-cdn.alpinelinux.org/alpine/v3.17/main/armv7/APKINDEX.tar.gzfetch
> >
> https://dl-cdn.alpinelinux.org/alpine/v3.17/community/armv7/APKINDEX.tar.gz(1/3)
> > Upgrading busybox (1.35.0-r29 -> 1.36.1-r5)Executing
> > busybox-1.36.1-r5.post-upgrade(2/3) Upgrading busybox-binsh
> > (1.35.0-r29 -> 1.36.1-r5)(3/3) Upgrading ssl_client (1.35.0-r29 ->
> > 1.36.1-r5)Executing busybox-1.36.1-r5.triggerOK: 5 MiB in 15 packages/
> > # apk upgrade musl
> > --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/main
> > --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/community(1/1)
> > Upgrading musl (1.2.3-r5 -> 1.2.4-r2)OK: 5 MiB in 15 packages/ # apk
> > upgrade rust --repository=
> https://dl-cdn.alpinelinux.org/alpine/v3.18/main
> > --repository=
> https://dl-cdn.alpinelinux.org/alpine/v3.18/communitySegmentation
> > fault/ # apk -hSegmentation fault/ # ls /tmp/core*Segmentation fault/
> > # cd /tmp/tmp # lsSegmentation fault
>
> i'd
>
> # cp /lib/ld-musl-armhf.so.1 /tmp
> # apk add gdb musl-dbg apk-tools-static
>
> then upgrade musl using apk.static, then debug via
>
> # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args ls -l /tmp
>
> or analyze a core dump, you can always install more debug tools
> via apk.static and run commands using the old ld-musl-armhf.so.1
>
> in gdb, you want to do
>
> bt
> disas $pc-40,+80
> info reg
> info proc map
>
> as a starting point and post the results.
>
> strace output can be useful too as well as readelf -aW of ld.so
> depending on what is going on.
>


-- 
Cody Wetzel
codyawetzel@gmail.com
(402)490-9242

[-- Attachment #2: Type: text/html, Size: 13019 bytes --]

Re: [musl] Segmentation fault musl 1.2.4

[Szabolcs Nagy]

* Cody Wetzel <codyawetzel@gmail.com> [2024-01-10 09:59:18 -0600]:
> So maybe I'm not understanding how gdb works or is used.

you have to run the process..

you never issued the gdb 'run' command.

> I'm getting no
> meaningful output even though these commands should result in a
> segmentation fault.
> 
> > / # gdb
> > Segmentation fault
> > / # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args ls -l /tmp
> > GNU gdb (GDB) 12.1
> > Copyright (C) 2022 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.
> > Type "show copying" and "show warranty" for details.
> > This GDB was configured as "armv7-alpine-linux-musleabihf".
> > Type "show configuration" for configuration details.
> > For bug reporting instructions, please see:
> > <https://www.gnu.org/software/gdb/bugs/>.
> > Find the GDB manual and other documentation resources online at:
> > <http://www.gnu.org/software/gdb/documentation/>.
> > For help, type "help".
> > Type "apropos word" to search for commands related to "word"...
> > Reading symbols from ls...
> > (No debugging symbols found in ls)
> > (gdb)
> > quit
> > / # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args cd /tmp
> > GNU gdb (GDB) 12.1
> > Copyright (C) 2022 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.
> > Type "show copying" and "show warranty" for details.
> > This GDB was configured as "armv7-alpine-linux-musleabihf".
> > Type "show configuration" for configuration details.
> > For bug reporting instructions, please see:
> > <https://www.gnu.org/software/gdb/bugs/>.
> > Find the GDB manual and other documentation resources online at:
> > <http://www.gnu.org/software/gdb/documentation/>.
> > For help, type "help".
> > Type "apropos word" to search for commands related to "word"...
> > cd: No such file or directory.
> > (gdb)
> > quit
> > / # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args apk -h /tmp
> > GNU gdb (GDB) 12.1
> > Copyright (C) 2022 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.
> > Type "show copying" and "show warranty" for details.
> > This GDB was configured as "armv7-alpine-linux-musleabihf".
> > Type "show configuration" for configuration details.
> > For bug reporting instructions, please see:
> > <https://www.gnu.org/software/gdb/bugs/>.
> > Find the GDB manual and other documentation resources online at:
> > <http://www.gnu.org/software/gdb/documentation/>.
> > For help, type "help".
> > Type "apropos word" to search for commands related to "word"...
> > Reading symbols from apk...
> > (No debugging symbols found in apk)
> > (gdb) bt
> > No stack.
> > (gdb) disas $pc-40,+80
> > No registers.
> > (gdb) info reg
> > The program has no registers now.
> 
> 
> On Thu, Jan 4, 2024 at 8:48 AM Szabolcs Nagy <nsz@port70.net> wrote:
> 
> > * Cody Wetzel <codyawetzel@gmail.com> [2024-01-03 11:20:29 -0600]:
> > > Hello musl team,
> > >
> > > I tried getting a core dump but I'm not sure if I'm doing something
> > wrong...
> > >
> > > / # cat /proc/sys/kernel/core_pattern/tmp/core-%e-%s-%u-%g-%p-%t/ #
> > > apk upgrade busybox
> > > --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/main
> > > --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/communityfetch
> > >
> > https://dl-cdn.alpinelinux.org/alpine/v3.18/community/armv7/APKINDEX.tar.gzfetch
> > >
> > https://dl-cdn.alpinelinux.org/alpine/v3.18/main/armv7/APKINDEX.tar.gzfetch
> > >
> > https://dl-cdn.alpinelinux.org/alpine/v3.17/main/armv7/APKINDEX.tar.gzfetch
> > >
> > https://dl-cdn.alpinelinux.org/alpine/v3.17/community/armv7/APKINDEX.tar.gz(1/3)
> > > Upgrading busybox (1.35.0-r29 -> 1.36.1-r5)Executing
> > > busybox-1.36.1-r5.post-upgrade(2/3) Upgrading busybox-binsh
> > > (1.35.0-r29 -> 1.36.1-r5)(3/3) Upgrading ssl_client (1.35.0-r29 ->
> > > 1.36.1-r5)Executing busybox-1.36.1-r5.triggerOK: 5 MiB in 15 packages/
> > > # apk upgrade musl
> > > --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/main
> > > --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/community(1/1)
> > > Upgrading musl (1.2.3-r5 -> 1.2.4-r2)OK: 5 MiB in 15 packages/ # apk
> > > upgrade rust --repository=
> > https://dl-cdn.alpinelinux.org/alpine/v3.18/main
> > > --repository=
> > https://dl-cdn.alpinelinux.org/alpine/v3.18/communitySegmentation
> > > fault/ # apk -hSegmentation fault/ # ls /tmp/core*Segmentation fault/
> > > # cd /tmp/tmp # lsSegmentation fault
> >
> > i'd
> >
> > # cp /lib/ld-musl-armhf.so.1 /tmp
> > # apk add gdb musl-dbg apk-tools-static
> >
> > then upgrade musl using apk.static, then debug via
> >
> > # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args ls -l /tmp
> >
> > or analyze a core dump, you can always install more debug tools
> > via apk.static and run commands using the old ld-musl-armhf.so.1
> >
> > in gdb, you want to do
> >
> > bt
> > disas $pc-40,+80
> > info reg
> > info proc map
> >
> > as a starting point and post the results.
> >
> > strace output can be useful too as well as readelf -aW of ld.so
> > depending on what is going on.
> >
> 
> 
> -- 
> Cody Wetzel
> codyawetzel@gmail.com
> (402)490-9242

Re: [musl] Segmentation fault musl 1.2.4

[Cody Wetzel]

[-- Attachment #1: Type: text/plain, Size: 7543 bytes --]

Sorry, I'm definitely not that well versed in linux.  I received the
following...

/ # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args ls -l /tmp
> GNU gdb (GDB) 12.1
> Copyright (C) 2022 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.
> Type "show copying" and "show warranty" for details.
> This GDB was configured as "armv7-alpine-linux-musleabihf".
> Type "show configuration" for configuration details.
> For bug reporting instructions, please see:
> <https://www.gnu.org/software/gdb/bugs/>.
> Find the GDB manual and other documentation resources online at:
>     <http://www.gnu.org/software/gdb/documentation/>.
>
> For help, type "help".
> Type "apropos word" to search for commands related to "word"...
> Reading symbols from ls...
> (No debugging symbols found in ls)
> (gdb) run
> Starting program: /bin/ls -l /tmp
> warning: Error disabling address space randomization: Operation not
> permitted
> warning: Could not trace the inferior process.
> warning: ptrace: Operation not permitted
> warning: Error restoring address space randomization: Operation not
> permitted
> During startup program exited with code 127.
>

Thanks!

On Thu, Jan 11, 2024 at 11:03 AM Szabolcs Nagy <nsz@port70.net> wrote:

> * Cody Wetzel <codyawetzel@gmail.com> [2024-01-10 09:59:18 -0600]:
> > So maybe I'm not understanding how gdb works or is used.
>
> you have to run the process..
>
> you never issued the gdb 'run' command.
>
> > I'm getting no
> > meaningful output even though these commands should result in a
> > segmentation fault.
> >
> > > / # gdb
> > > Segmentation fault
> > > / # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args ls -l /tmp
> > > GNU gdb (GDB) 12.1
> > > Copyright (C) 2022 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.
> > > Type "show copying" and "show warranty" for details.
> > > This GDB was configured as "armv7-alpine-linux-musleabihf".
> > > Type "show configuration" for configuration details.
> > > For bug reporting instructions, please see:
> > > <https://www.gnu.org/software/gdb/bugs/>.
> > > Find the GDB manual and other documentation resources online at:
> > > <http://www.gnu.org/software/gdb/documentation/>.
> > > For help, type "help".
> > > Type "apropos word" to search for commands related to "word"...
> > > Reading symbols from ls...
> > > (No debugging symbols found in ls)
> > > (gdb)
> > > quit
> > > / # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args cd /tmp
> > > GNU gdb (GDB) 12.1
> > > Copyright (C) 2022 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.
> > > Type "show copying" and "show warranty" for details.
> > > This GDB was configured as "armv7-alpine-linux-musleabihf".
> > > Type "show configuration" for configuration details.
> > > For bug reporting instructions, please see:
> > > <https://www.gnu.org/software/gdb/bugs/>.
> > > Find the GDB manual and other documentation resources online at:
> > > <http://www.gnu.org/software/gdb/documentation/>.
> > > For help, type "help".
> > > Type "apropos word" to search for commands related to "word"...
> > > cd: No such file or directory.
> > > (gdb)
> > > quit
> > > / # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args apk -h /tmp
> > > GNU gdb (GDB) 12.1
> > > Copyright (C) 2022 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.
> > > Type "show copying" and "show warranty" for details.
> > > This GDB was configured as "armv7-alpine-linux-musleabihf".
> > > Type "show configuration" for configuration details.
> > > For bug reporting instructions, please see:
> > > <https://www.gnu.org/software/gdb/bugs/>.
> > > Find the GDB manual and other documentation resources online at:
> > > <http://www.gnu.org/software/gdb/documentation/>.
> > > For help, type "help".
> > > Type "apropos word" to search for commands related to "word"...
> > > Reading symbols from apk...
> > > (No debugging symbols found in apk)
> > > (gdb) bt
> > > No stack.
> > > (gdb) disas $pc-40,+80
> > > No registers.
> > > (gdb) info reg
> > > The program has no registers now.
> >
> >
> > On Thu, Jan 4, 2024 at 8:48 AM Szabolcs Nagy <nsz@port70.net> wrote:
> >
> > > * Cody Wetzel <codyawetzel@gmail.com> [2024-01-03 11:20:29 -0600]:
> > > > Hello musl team,
> > > >
> > > > I tried getting a core dump but I'm not sure if I'm doing something
> > > wrong...
> > > >
> > > > / # cat /proc/sys/kernel/core_pattern/tmp/core-%e-%s-%u-%g-%p-%t/ #
> > > > apk upgrade busybox
> > > > --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/main
> > > > --repository=
> https://dl-cdn.alpinelinux.org/alpine/v3.18/communityfetch
> > > >
> > >
> https://dl-cdn.alpinelinux.org/alpine/v3.18/community/armv7/APKINDEX.tar.gzfetch
> > > >
> > >
> https://dl-cdn.alpinelinux.org/alpine/v3.18/main/armv7/APKINDEX.tar.gzfetch
> > > >
> > >
> https://dl-cdn.alpinelinux.org/alpine/v3.17/main/armv7/APKINDEX.tar.gzfetch
> > > >
> > >
> https://dl-cdn.alpinelinux.org/alpine/v3.17/community/armv7/APKINDEX.tar.gz(1/3)
> > > > Upgrading busybox (1.35.0-r29 -> 1.36.1-r5)Executing
> > > > busybox-1.36.1-r5.post-upgrade(2/3) Upgrading busybox-binsh
> > > > (1.35.0-r29 -> 1.36.1-r5)(3/3) Upgrading ssl_client (1.35.0-r29 ->
> > > > 1.36.1-r5)Executing busybox-1.36.1-r5.triggerOK: 5 MiB in 15
> packages/
> > > > # apk upgrade musl
> > > > --repository=https://dl-cdn.alpinelinux.org/alpine/v3.18/main
> > > > --repository=
> https://dl-cdn.alpinelinux.org/alpine/v3.18/community(1/1)
> > > > Upgrading musl (1.2.3-r5 -> 1.2.4-r2)OK: 5 MiB in 15 packages/ # apk
> > > > upgrade rust --repository=
> > > https://dl-cdn.alpinelinux.org/alpine/v3.18/main
> > > > --repository=
> > > https://dl-cdn.alpinelinux.org/alpine/v3.18/communitySegmentation
> > > > fault/ # apk -hSegmentation fault/ # ls /tmp/core*Segmentation fault/
> > > > # cd /tmp/tmp # lsSegmentation fault
> > >
> > > i'd
> > >
> > > # cp /lib/ld-musl-armhf.so.1 /tmp
> > > # apk add gdb musl-dbg apk-tools-static
> > >
> > > then upgrade musl using apk.static, then debug via
> > >
> > > # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args ls -l /tmp
> > >
> > > or analyze a core dump, you can always install more debug tools
> > > via apk.static and run commands using the old ld-musl-armhf.so.1
> > >
> > > in gdb, you want to do
> > >
> > > bt
> > > disas $pc-40,+80
> > > info reg
> > > info proc map
> > >
> > > as a starting point and post the results.
> > >
> > > strace output can be useful too as well as readelf -aW of ld.so
> > > depending on what is going on.
> > >
> >
> >
> > --
> > Cody Wetzel
> > codyawetzel@gmail.com
> > (402)490-9242
>


-- 
Cody Wetzel
codyawetzel@gmail.com
(402)490-9242

[-- Attachment #2: Type: text/html, Size: 11977 bytes --]

Re: [musl] Segmentation fault musl 1.2.4

[Szabolcs Nagy]

* Cody Wetzel <codyawetzel@gmail.com> [2024-01-11 12:19:16 -0600]:
> Sorry, I'm definitely not that well versed in linux.  I received the
> following...
> 
> / # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args ls -l /tmp
> > GNU gdb (GDB) 12.1
> > Copyright (C) 2022 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.
> > Type "show copying" and "show warranty" for details.
> > This GDB was configured as "armv7-alpine-linux-musleabihf".
> > Type "show configuration" for configuration details.
> > For bug reporting instructions, please see:
> > <https://www.gnu.org/software/gdb/bugs/>.
> > Find the GDB manual and other documentation resources online at:
> >     <http://www.gnu.org/software/gdb/documentation/>.
> >
> > For help, type "help".
> > Type "apropos word" to search for commands related to "word"...
> > Reading symbols from ls...
> > (No debugging symbols found in ls)
> > (gdb) run
> > Starting program: /bin/ls -l /tmp
> > warning: Error disabling address space randomization: Operation not
> > permitted
> > warning: Could not trace the inferior process.
> > warning: ptrace: Operation not permitted
> > warning: Error restoring address space randomization: Operation not
> > permitted
> > During startup program exited with code 127.

use a container that's not restricted e.g.

docker run --cap-add=SYS_PTRACE --security-opt seccomp=unconfined

(i found this with 2min web search)

Re: [musl] Segmentation fault musl 1.2.4

[Cody Wetzel]

[-- Attachment #1: Type: text/plain, Size: 3022 bytes --]

I did actually find that stackoverflow post and tried spinning up a new
container but the results don't provide any useful information...

/ # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args ls -l /tmp
> GNU gdb (GDB) 12.1
> Copyright (C) 2022 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.
> Type "show copying" and "show warranty" for details.
> This GDB was configured as "armv7-alpine-linux-musleabihf".
> Type "show configuration" for configuration details.
> For bug reporting instructions, please see:
> <https://www.gnu.org/software/gdb/bugs/>.
> Find the GDB manual and other documentation resources online at:
>     <http://www.gnu.org/software/gdb/documentation/>.
>
> For help, type "help".
> Type "apropos word" to search for commands related to "word"...
> Reading symbols from ls...
> (No debugging symbols found in ls)
> (gdb) run
> Starting program: /bin/ls -l /tmp
> During startup program terminated with signal SIGSEGV, Segmentation fault.
>

On Fri, Jan 12, 2024 at 12:57 PM Szabolcs Nagy <nsz@port70.net> wrote:

> * Cody Wetzel <codyawetzel@gmail.com> [2024-01-11 12:19:16 -0600]:
> > Sorry, I'm definitely not that well versed in linux.  I received the
> > following...
> >
> > / # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args ls -l /tmp
> > > GNU gdb (GDB) 12.1
> > > Copyright (C) 2022 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.
> > > Type "show copying" and "show warranty" for details.
> > > This GDB was configured as "armv7-alpine-linux-musleabihf".
> > > Type "show configuration" for configuration details.
> > > For bug reporting instructions, please see:
> > > <https://www.gnu.org/software/gdb/bugs/>.
> > > Find the GDB manual and other documentation resources online at:
> > >     <http://www.gnu.org/software/gdb/documentation/>.
> > >
> > > For help, type "help".
> > > Type "apropos word" to search for commands related to "word"...
> > > Reading symbols from ls...
> > > (No debugging symbols found in ls)
> > > (gdb) run
> > > Starting program: /bin/ls -l /tmp
> > > warning: Error disabling address space randomization: Operation not
> > > permitted
> > > warning: Could not trace the inferior process.
> > > warning: ptrace: Operation not permitted
> > > warning: Error restoring address space randomization: Operation not
> > > permitted
> > > During startup program exited with code 127.
>
> use a container that's not restricted e.g.
>
> docker run --cap-add=SYS_PTRACE --security-opt seccomp=unconfined
>
> (i found this with 2min web search)
>


-- 
Cody Wetzel
codyawetzel@gmail.com
(402)490-9242

[-- Attachment #2: Type: text/html, Size: 4545 bytes --]

Re: [musl] Segmentation fault musl 1.2.4

[Szabolcs Nagy]

* Cody Wetzel <codyawetzel@gmail.com> [2024-01-12 17:10:24 -0600]:
> I did actually find that stackoverflow post and tried spinning up a new
> container but the results don't provide any useful information...
> 
> / # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args ls -l /tmp
> > GNU gdb (GDB) 12.1
> > Copyright (C) 2022 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.
> > Type "show copying" and "show warranty" for details.
> > This GDB was configured as "armv7-alpine-linux-musleabihf".
> > Type "show configuration" for configuration details.
> > For bug reporting instructions, please see:
> > <https://www.gnu.org/software/gdb/bugs/>.
> > Find the GDB manual and other documentation resources online at:
> >     <http://www.gnu.org/software/gdb/documentation/>.
> >
> > For help, type "help".
> > Type "apropos word" to search for commands related to "word"...
> > Reading symbols from ls...
> > (No debugging symbols found in ls)
> > (gdb) run
> > Starting program: /bin/ls -l /tmp
> > During startup program terminated with signal SIGSEGV, Segmentation fault.
> >

so i looked at alpine armv7 binaries and they seem to be built with
4k page size.

since the elf files are not possible to map with 32k pages, it is
expected that they crash on exec in the kernel and thus you get no
backtrace.

i'm not sure why gdb works, unless it's an old gdb binary that's built
with large pagesize support.

an easy way to verify is to compare proram headers of old and new
binaries like

readelf -lW /tmp/ld-musl-armhf.so.1

Re: [musl] Segmentation fault musl 1.2.4

[Cody Wetzel]

[-- Attachment #1: Type: text/plain, Size: 4206 bytes --]

Here is the output for the old
....
>
> / # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW /tmp/ld-musl-armhf.so.1
>
> Elf file type is DYN (Shared object file)
> Entry point 0x359cd
> There are 6 program headers, starting at offset 52
>
> Program Headers:
>   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
>   EXIDX          0x07acec 0x0007acec 0x0007acec 0x00008 0x00008 R   0x4
>   LOAD           0x000000 0x00000000 0x00000000 0x7acf4 0x7acf4 R E 0x10000
>   LOAD           0x07fd6c 0x0008fd6c 0x0008fd6c 0x0054a 0x02258 RW  0x10000
>   DYNAMIC        0x07febc 0x0008febc 0x0008febc 0x000c0 0x000c0 RW  0x4
>   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW  0x10
>   GNU_RELRO      0x07fd6c 0x0008fd6c 0x0008fd6c 0x00294 0x00294 R   0x1
>
>  Section to Segment mapping:
>   Segment Sections...
>    00     .ARM.exidx
>    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> .rodata .ARM.exidx
>    02     .data.rel.ro .dynamic .got .data .bss
>    03     .dynamic
>    04
>    05     .data.rel.ro .dynamic .got
>

And the new...

/ # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW /lib/ld-musl-armhf.so.1
>
> Elf file type is DYN (Shared object file)
> Entry point 0x362f1
> There are 6 program headers, starting at offset 52
>
> Program Headers:
>   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
>   EXIDX          0x07b81c 0x0007b81c 0x0007b81c 0x00008 0x00008 R   0x4
>   LOAD           0x000000 0x00000000 0x00000000 0x7b824 0x7b824 R E 0x1000
>   LOAD           0x07bd74 0x0007cd74 0x0007cd74 0x0054a 0x0225c RW  0x1000
>   DYNAMIC        0x07bebc 0x0007cebc 0x0007cebc 0x000c0 0x000c0 RW  0x4
>   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW  0x10
>   GNU_RELRO      0x07bd74 0x0007cd74 0x0007cd74 0x0028c 0x0028c R   0x1
>
>  Section to Segment mapping:
>   Segment Sections...
>    00     .ARM.exidx
>    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> .rodata .ARM.exidx
>    02     .data.rel.ro .dynamic .got .data .bss
>    03     .dynamic
>    04
>    05     .data.rel.ro .dynamic .got


I hope that helps.

Thanks!

Cody

On Mon, Jan 15, 2024 at 4:30 PM Szabolcs Nagy <nsz@port70.net> wrote:

> * Cody Wetzel <codyawetzel@gmail.com> [2024-01-12 17:10:24 -0600]:
> > I did actually find that stackoverflow post and tried spinning up a new
> > container but the results don't provide any useful information...
> >
> > / # /tmp/ld-musl-armhf.so.1 /usr/bin/gdb --args ls -l /tmp
> > > GNU gdb (GDB) 12.1
> > > Copyright (C) 2022 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.
> > > Type "show copying" and "show warranty" for details.
> > > This GDB was configured as "armv7-alpine-linux-musleabihf".
> > > Type "show configuration" for configuration details.
> > > For bug reporting instructions, please see:
> > > <https://www.gnu.org/software/gdb/bugs/>.
> > > Find the GDB manual and other documentation resources online at:
> > >     <http://www.gnu.org/software/gdb/documentation/>.
> > >
> > > For help, type "help".
> > > Type "apropos word" to search for commands related to "word"...
> > > Reading symbols from ls...
> > > (No debugging symbols found in ls)
> > > (gdb) run
> > > Starting program: /bin/ls -l /tmp
> > > During startup program terminated with signal SIGSEGV, Segmentation
> fault.
> > >
>
> so i looked at alpine armv7 binaries and they seem to be built with
> 4k page size.
>
> since the elf files are not possible to map with 32k pages, it is
> expected that they crash on exec in the kernel and thus you get no
> backtrace.
>
> i'm not sure why gdb works, unless it's an old gdb binary that's built
> with large pagesize support.
>
> an easy way to verify is to compare proram headers of old and new
> binaries like
>
> readelf -lW /tmp/ld-musl-armhf.so.1
>


-- 
Cody Wetzel
codyawetzel@gmail.com
(402)490-9242

[-- Attachment #2: Type: text/html, Size: 5979 bytes --]

Re: [musl] Segmentation fault musl 1.2.4

[Szabolcs Nagy]

* Cody Wetzel <codyawetzel@gmail.com> [2024-01-16 09:21:05 -0600]:
> Here is the output for the old
> ....
> >
> > / # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW /tmp/ld-musl-armhf.so.1
> >
> > Elf file type is DYN (Shared object file)
> > Entry point 0x359cd
> > There are 6 program headers, starting at offset 52
> >
> > Program Headers:
> >   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
> >   EXIDX          0x07acec 0x0007acec 0x0007acec 0x00008 0x00008 R   0x4
> >   LOAD           0x000000 0x00000000 0x00000000 0x7acf4 0x7acf4 R E 0x10000
> >   LOAD           0x07fd6c 0x0008fd6c 0x0008fd6c 0x0054a 0x02258 RW  0x10000

this load segment is 64k aligned.

> >   DYNAMIC        0x07febc 0x0008febc 0x0008febc 0x000c0 0x000c0 RW  0x4
> >   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW  0x10
> >   GNU_RELRO      0x07fd6c 0x0008fd6c 0x0008fd6c 0x00294 0x00294 R   0x1
> >
> >  Section to Segment mapping:
> >   Segment Sections...
> >    00     .ARM.exidx
> >    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> > .rodata .ARM.exidx
> >    02     .data.rel.ro .dynamic .got .data .bss
> >    03     .dynamic
> >    04
> >    05     .data.rel.ro .dynamic .got
> >
> 
> And the new...
> 
> / # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW /lib/ld-musl-armhf.so.1
> >
> > Elf file type is DYN (Shared object file)
> > Entry point 0x362f1
> > There are 6 program headers, starting at offset 52
> >
> > Program Headers:
> >   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
> >   EXIDX          0x07b81c 0x0007b81c 0x0007b81c 0x00008 0x00008 R   0x4
> >   LOAD           0x000000 0x00000000 0x00000000 0x7b824 0x7b824 R E 0x1000
> >   LOAD           0x07bd74 0x0007cd74 0x0007cd74 0x0054a 0x0225c RW  0x1000

this load segment is 4k aligned and offset vs addr is not congruent
modulo 64k, or 32k, so won't work on systems with such page size.

> >   DYNAMIC        0x07bebc 0x0007cebc 0x0007cebc 0x000c0 0x000c0 RW  0x4
> >   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW  0x10
> >   GNU_RELRO      0x07bd74 0x0007cd74 0x0007cd74 0x0028c 0x0028c R   0x1
> >
> >  Section to Segment mapping:
> >   Segment Sections...
> >    00     .ARM.exidx
> >    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> > .rodata .ARM.exidx
> >    02     .data.rel.ro .dynamic .got .data .bss
> >    03     .dynamic
> >    04
> >    05     .data.rel.ro .dynamic .got
> 
> 
> I hope that helps.

yes, this is a linking issue, not musl libc.

alpine linux links binaries for 4k pagesize only.

arm linkers were updated at some point to create binaries supporting
up to 64k pagesize.  i suspect some ppl ran into issues in practice
and decided the larger binaries are not worth it, if they dont work
reliably and forced 4k page size at link time.

you have to raise an issue with alpine linux, if you think 32k
oage size is useful and reliably supportable.

Re: [musl] Segmentation fault musl 1.2.4

[Cody Wetzel]

[-- Attachment #1: Type: text/plain, Size: 3419 bytes --]

Thank you so much for pointing me in the right direction!

Cody

On Tue, Jan 16, 2024 at 12:29 PM Szabolcs Nagy <nsz@port70.net> wrote:

> * Cody Wetzel <codyawetzel@gmail.com> [2024-01-16 09:21:05 -0600]:
> > Here is the output for the old
> > ....
> > >
> > > / # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW
> /tmp/ld-musl-armhf.so.1
> > >
> > > Elf file type is DYN (Shared object file)
> > > Entry point 0x359cd
> > > There are 6 program headers, starting at offset 52
> > >
> > > Program Headers:
> > >   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg
> Align
> > >   EXIDX          0x07acec 0x0007acec 0x0007acec 0x00008 0x00008 R   0x4
> > >   LOAD           0x000000 0x00000000 0x00000000 0x7acf4 0x7acf4 R E
> 0x10000
> > >   LOAD           0x07fd6c 0x0008fd6c 0x0008fd6c 0x0054a 0x02258 RW
> 0x10000
>
> this load segment is 64k aligned.
>
> > >   DYNAMIC        0x07febc 0x0008febc 0x0008febc 0x000c0 0x000c0 RW  0x4
> > >   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW
> 0x10
> > >   GNU_RELRO      0x07fd6c 0x0008fd6c 0x0008fd6c 0x00294 0x00294 R   0x1
> > >
> > >  Section to Segment mapping:
> > >   Segment Sections...
> > >    00     .ARM.exidx
> > >    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> > > .rodata .ARM.exidx
> > >    02     .data.rel.ro .dynamic .got .data .bss
> > >    03     .dynamic
> > >    04
> > >    05     .data.rel.ro .dynamic .got
> > >
> >
> > And the new...
> >
> > / # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW /lib/ld-musl-armhf.so.1
> > >
> > > Elf file type is DYN (Shared object file)
> > > Entry point 0x362f1
> > > There are 6 program headers, starting at offset 52
> > >
> > > Program Headers:
> > >   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg
> Align
> > >   EXIDX          0x07b81c 0x0007b81c 0x0007b81c 0x00008 0x00008 R   0x4
> > >   LOAD           0x000000 0x00000000 0x00000000 0x7b824 0x7b824 R E
> 0x1000
> > >   LOAD           0x07bd74 0x0007cd74 0x0007cd74 0x0054a 0x0225c RW
> 0x1000
>
> this load segment is 4k aligned and offset vs addr is not congruent
> modulo 64k, or 32k, so won't work on systems with such page size.
>
> > >   DYNAMIC        0x07bebc 0x0007cebc 0x0007cebc 0x000c0 0x000c0 RW  0x4
> > >   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW
> 0x10
> > >   GNU_RELRO      0x07bd74 0x0007cd74 0x0007cd74 0x0028c 0x0028c R   0x1
> > >
> > >  Section to Segment mapping:
> > >   Segment Sections...
> > >    00     .ARM.exidx
> > >    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> > > .rodata .ARM.exidx
> > >    02     .data.rel.ro .dynamic .got .data .bss
> > >    03     .dynamic
> > >    04
> > >    05     .data.rel.ro .dynamic .got
> >
> >
> > I hope that helps.
>
> yes, this is a linking issue, not musl libc.
>
> alpine linux links binaries for 4k pagesize only.
>
> arm linkers were updated at some point to create binaries supporting
> up to 64k pagesize.  i suspect some ppl ran into issues in practice
> and decided the larger binaries are not worth it, if they dont work
> reliably and forced 4k page size at link time.
>
> you have to raise an issue with alpine linux, if you think 32k
> oage size is useful and reliably supportable.
>
>

-- 
Cody Wetzel
codyawetzel@gmail.com
(402)490-9242

[-- Attachment #2: Type: text/html, Size: 4874 bytes --]

Re: [musl] Segmentation fault musl 1.2.4

[Rich Felker]

On Tue, Jan 16, 2024 at 07:29:18PM +0100, Szabolcs Nagy wrote:
> * Cody Wetzel <codyawetzel@gmail.com> [2024-01-16 09:21:05 -0600]:
> > Here is the output for the old
> > ....
> > >
> > > / # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW /tmp/ld-musl-armhf.so.1
> > >
> > > Elf file type is DYN (Shared object file)
> > > Entry point 0x359cd
> > > There are 6 program headers, starting at offset 52
> > >
> > > Program Headers:
> > >   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
> > >   EXIDX          0x07acec 0x0007acec 0x0007acec 0x00008 0x00008 R   0x4
> > >   LOAD           0x000000 0x00000000 0x00000000 0x7acf4 0x7acf4 R E 0x10000
> > >   LOAD           0x07fd6c 0x0008fd6c 0x0008fd6c 0x0054a 0x02258 RW  0x10000
> 
> this load segment is 64k aligned.
> 
> > >   DYNAMIC        0x07febc 0x0008febc 0x0008febc 0x000c0 0x000c0 RW  0x4
> > >   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW  0x10
> > >   GNU_RELRO      0x07fd6c 0x0008fd6c 0x0008fd6c 0x00294 0x00294 R   0x1
> > >
> > >  Section to Segment mapping:
> > >   Segment Sections...
> > >    00     .ARM.exidx
> > >    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> > > .rodata .ARM.exidx
> > >    02     .data.rel.ro .dynamic .got .data .bss
> > >    03     .dynamic
> > >    04
> > >    05     .data.rel.ro .dynamic .got
> > >
> > 
> > And the new...
> > 
> > / # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW /lib/ld-musl-armhf.so.1
> > >
> > > Elf file type is DYN (Shared object file)
> > > Entry point 0x362f1
> > > There are 6 program headers, starting at offset 52
> > >
> > > Program Headers:
> > >   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
> > >   EXIDX          0x07b81c 0x0007b81c 0x0007b81c 0x00008 0x00008 R   0x4
> > >   LOAD           0x000000 0x00000000 0x00000000 0x7b824 0x7b824 R E 0x1000
> > >   LOAD           0x07bd74 0x0007cd74 0x0007cd74 0x0054a 0x0225c RW  0x1000
> 
> this load segment is 4k aligned and offset vs addr is not congruent
> modulo 64k, or 32k, so won't work on systems with such page size.
> 
> > >   DYNAMIC        0x07bebc 0x0007cebc 0x0007cebc 0x000c0 0x000c0 RW  0x4
> > >   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW  0x10
> > >   GNU_RELRO      0x07bd74 0x0007cd74 0x0007cd74 0x0028c 0x0028c R   0x1
> > >
> > >  Section to Segment mapping:
> > >   Segment Sections...
> > >    00     .ARM.exidx
> > >    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> > > .rodata .ARM.exidx
> > >    02     .data.rel.ro .dynamic .got .data .bss
> > >    03     .dynamic
> > >    04
> > >    05     .data.rel.ro .dynamic .got
> > 
> > 
> > I hope that helps.
> 
> yes, this is a linking issue, not musl libc.
> 
> alpine linux links binaries for 4k pagesize only.
> 
> arm linkers were updated at some point to create binaries supporting
> up to 64k pagesize.  i suspect some ppl ran into issues in practice
> and decided the larger binaries are not worth it, if they dont work
> reliably and forced 4k page size at link time.
> 
> you have to raise an issue with alpine linux, if you think 32k
> oage size is useful and reliably supportable.

Are they using -Wl,-z,separate-code? That incurs a large
binary-size-on-disk penalty when supporting oversized pages, and IIRC
something was done to make the linker default to not supporting
oversized pages when that's used. It might be the reason, if arm
linking is normally expected to use a larger max pagesize.

Rich

Re: [musl] Segmentation fault musl 1.2.4

[Szabolcs Nagy]

* Rich Felker <dalias@libc.org> [2024-01-16 15:45:52 -0500]:

> On Tue, Jan 16, 2024 at 07:29:18PM +0100, Szabolcs Nagy wrote:
> > * Cody Wetzel <codyawetzel@gmail.com> [2024-01-16 09:21:05 -0600]:
> > > Here is the output for the old
> > > ....
> > > >
> > > > / # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW /tmp/ld-musl-armhf.so.1
> > > >
> > > > Elf file type is DYN (Shared object file)
> > > > Entry point 0x359cd
> > > > There are 6 program headers, starting at offset 52
> > > >
> > > > Program Headers:
> > > >   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
> > > >   EXIDX          0x07acec 0x0007acec 0x0007acec 0x00008 0x00008 R   0x4
> > > >   LOAD           0x000000 0x00000000 0x00000000 0x7acf4 0x7acf4 R E 0x10000
> > > >   LOAD           0x07fd6c 0x0008fd6c 0x0008fd6c 0x0054a 0x02258 RW  0x10000
> > 
> > this load segment is 64k aligned.
> > 
> > > >   DYNAMIC        0x07febc 0x0008febc 0x0008febc 0x000c0 0x000c0 RW  0x4
> > > >   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW  0x10
> > > >   GNU_RELRO      0x07fd6c 0x0008fd6c 0x0008fd6c 0x00294 0x00294 R   0x1
> > > >
> > > >  Section to Segment mapping:
> > > >   Segment Sections...
> > > >    00     .ARM.exidx
> > > >    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> > > > .rodata .ARM.exidx
> > > >    02     .data.rel.ro .dynamic .got .data .bss
> > > >    03     .dynamic
> > > >    04
> > > >    05     .data.rel.ro .dynamic .got
> > > >
> > > 
> > > And the new...
> > > 
> > > / # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW /lib/ld-musl-armhf.so.1
> > > >
> > > > Elf file type is DYN (Shared object file)
> > > > Entry point 0x362f1
> > > > There are 6 program headers, starting at offset 52
> > > >
> > > > Program Headers:
> > > >   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
> > > >   EXIDX          0x07b81c 0x0007b81c 0x0007b81c 0x00008 0x00008 R   0x4
> > > >   LOAD           0x000000 0x00000000 0x00000000 0x7b824 0x7b824 R E 0x1000
> > > >   LOAD           0x07bd74 0x0007cd74 0x0007cd74 0x0054a 0x0225c RW  0x1000
> > 
> > this load segment is 4k aligned and offset vs addr is not congruent
> > modulo 64k, or 32k, so won't work on systems with such page size.
> > 
> > > >   DYNAMIC        0x07bebc 0x0007cebc 0x0007cebc 0x000c0 0x000c0 RW  0x4
> > > >   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW  0x10
> > > >   GNU_RELRO      0x07bd74 0x0007cd74 0x0007cd74 0x0028c 0x0028c R   0x1
> > > >
> > > >  Section to Segment mapping:
> > > >   Segment Sections...
> > > >    00     .ARM.exidx
> > > >    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> > > > .rodata .ARM.exidx
> > > >    02     .data.rel.ro .dynamic .got .data .bss
> > > >    03     .dynamic
> > > >    04
> > > >    05     .data.rel.ro .dynamic .got
> > > 
> > > 
> > > I hope that helps.
> > 
> > yes, this is a linking issue, not musl libc.
> > 
> > alpine linux links binaries for 4k pagesize only.
> > 
> > arm linkers were updated at some point to create binaries supporting
> > up to 64k pagesize.  i suspect some ppl ran into issues in practice
> > and decided the larger binaries are not worth it, if they dont work
> > reliably and forced 4k page size at link time.
> > 
> > you have to raise an issue with alpine linux, if you think 32k
> > oage size is useful and reliably supportable.
> 
> Are they using -Wl,-z,separate-code? That incurs a large
> binary-size-on-disk penalty when supporting oversized pages, and IIRC
> something was done to make the linker default to not supporting
> oversized pages when that's used. It might be the reason, if arm
> linking is normally expected to use a larger max pagesize.

i looked at this now, turns out they just changed the
pagesize back to 4k (i missed this change):

https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=1a26a53a0dee39106ba58fcb15496c5f13074652

Re: [musl] Segmentation fault musl 1.2.4

[Rich Felker]

On Tue, Jan 30, 2024 at 11:43:38AM +0100, Szabolcs Nagy wrote:
> * Rich Felker <dalias@libc.org> [2024-01-16 15:45:52 -0500]:
> 
> > On Tue, Jan 16, 2024 at 07:29:18PM +0100, Szabolcs Nagy wrote:
> > > * Cody Wetzel <codyawetzel@gmail.com> [2024-01-16 09:21:05 -0600]:
> > > > Here is the output for the old
> > > > ....
> > > > >
> > > > > / # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW /tmp/ld-musl-armhf.so.1
> > > > >
> > > > > Elf file type is DYN (Shared object file)
> > > > > Entry point 0x359cd
> > > > > There are 6 program headers, starting at offset 52
> > > > >
> > > > > Program Headers:
> > > > >   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
> > > > >   EXIDX          0x07acec 0x0007acec 0x0007acec 0x00008 0x00008 R   0x4
> > > > >   LOAD           0x000000 0x00000000 0x00000000 0x7acf4 0x7acf4 R E 0x10000
> > > > >   LOAD           0x07fd6c 0x0008fd6c 0x0008fd6c 0x0054a 0x02258 RW  0x10000
> > > 
> > > this load segment is 64k aligned.
> > > 
> > > > >   DYNAMIC        0x07febc 0x0008febc 0x0008febc 0x000c0 0x000c0 RW  0x4
> > > > >   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW  0x10
> > > > >   GNU_RELRO      0x07fd6c 0x0008fd6c 0x0008fd6c 0x00294 0x00294 R   0x1
> > > > >
> > > > >  Section to Segment mapping:
> > > > >   Segment Sections...
> > > > >    00     .ARM.exidx
> > > > >    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> > > > > .rodata .ARM.exidx
> > > > >    02     .data.rel.ro .dynamic .got .data .bss
> > > > >    03     .dynamic
> > > > >    04
> > > > >    05     .data.rel.ro .dynamic .got
> > > > >
> > > > 
> > > > And the new...
> > > > 
> > > > / # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW /lib/ld-musl-armhf.so.1
> > > > >
> > > > > Elf file type is DYN (Shared object file)
> > > > > Entry point 0x362f1
> > > > > There are 6 program headers, starting at offset 52
> > > > >
> > > > > Program Headers:
> > > > >   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
> > > > >   EXIDX          0x07b81c 0x0007b81c 0x0007b81c 0x00008 0x00008 R   0x4
> > > > >   LOAD           0x000000 0x00000000 0x00000000 0x7b824 0x7b824 R E 0x1000
> > > > >   LOAD           0x07bd74 0x0007cd74 0x0007cd74 0x0054a 0x0225c RW  0x1000
> > > 
> > > this load segment is 4k aligned and offset vs addr is not congruent
> > > modulo 64k, or 32k, so won't work on systems with such page size.
> > > 
> > > > >   DYNAMIC        0x07bebc 0x0007cebc 0x0007cebc 0x000c0 0x000c0 RW  0x4
> > > > >   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW  0x10
> > > > >   GNU_RELRO      0x07bd74 0x0007cd74 0x0007cd74 0x0028c 0x0028c R   0x1
> > > > >
> > > > >  Section to Segment mapping:
> > > > >   Segment Sections...
> > > > >    00     .ARM.exidx
> > > > >    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> > > > > .rodata .ARM.exidx
> > > > >    02     .data.rel.ro .dynamic .got .data .bss
> > > > >    03     .dynamic
> > > > >    04
> > > > >    05     .data.rel.ro .dynamic .got
> > > > 
> > > > 
> > > > I hope that helps.
> > > 
> > > yes, this is a linking issue, not musl libc.
> > > 
> > > alpine linux links binaries for 4k pagesize only.
> > > 
> > > arm linkers were updated at some point to create binaries supporting
> > > up to 64k pagesize.  i suspect some ppl ran into issues in practice
> > > and decided the larger binaries are not worth it, if they dont work
> > > reliably and forced 4k page size at link time.
> > > 
> > > you have to raise an issue with alpine linux, if you think 32k
> > > oage size is useful and reliably supportable.
> > 
> > Are they using -Wl,-z,separate-code? That incurs a large
> > binary-size-on-disk penalty when supporting oversized pages, and IIRC
> > something was done to make the linker default to not supporting
> > oversized pages when that's used. It might be the reason, if arm
> > linking is normally expected to use a larger max pagesize.
> 
> i looked at this now, turns out they just changed the
> pagesize back to 4k (i missed this change):
> 
> https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=1a26a53a0dee39106ba58fcb15496c5f13074652

This doesn't help immediately, but a major ingredient to fix this
situation would be getting the kernel to stop doing the wrong thing.
Right now, it's ignoring the fact that the ELF program header
constraints are incompatible with mmap given the oversized system
pagesize, and just incorrectly mapping the executable and trying to
run it anyway, whereby it blows up.

The right thing to do would be either to fail with ENOEXEC in this
case, or when mmap with the required offset constraint fails, falling
back to making an anonymous map and copying the whole content of the
loadable segment into that (no COW sharing). The latter is really not
all that bad for got/data/etc. mappings which you expect will be dirty
(modified) anyway.

BTW the former choice (ENOEXEC) would allow doing the latter in
userspace with a binfmt_misc loader.

Rich

[musl] Fixing ELF loader for systems with oversized pages [was: Re: [musl] Segmentation fault musl 1.2.4]

[Rich Felker]

[-- Attachment #1: Type: text/plain, Size: 5368 bytes --]

On Tue, Jan 30, 2024 at 10:37:30AM -0500, Rich Felker wrote:
> On Tue, Jan 30, 2024 at 11:43:38AM +0100, Szabolcs Nagy wrote:
> > * Rich Felker <dalias@libc.org> [2024-01-16 15:45:52 -0500]:
> > 
> > > On Tue, Jan 16, 2024 at 07:29:18PM +0100, Szabolcs Nagy wrote:
> > > > * Cody Wetzel <codyawetzel@gmail.com> [2024-01-16 09:21:05 -0600]:
> > > > > Here is the output for the old
> > > > > ....
> > > > > >
> > > > > > / # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW /tmp/ld-musl-armhf.so.1
> > > > > >
> > > > > > Elf file type is DYN (Shared object file)
> > > > > > Entry point 0x359cd
> > > > > > There are 6 program headers, starting at offset 52
> > > > > >
> > > > > > Program Headers:
> > > > > >   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
> > > > > >   EXIDX          0x07acec 0x0007acec 0x0007acec 0x00008 0x00008 R   0x4
> > > > > >   LOAD           0x000000 0x00000000 0x00000000 0x7acf4 0x7acf4 R E 0x10000
> > > > > >   LOAD           0x07fd6c 0x0008fd6c 0x0008fd6c 0x0054a 0x02258 RW  0x10000
> > > > 
> > > > this load segment is 64k aligned.
> > > > 
> > > > > >   DYNAMIC        0x07febc 0x0008febc 0x0008febc 0x000c0 0x000c0 RW  0x4
> > > > > >   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW  0x10
> > > > > >   GNU_RELRO      0x07fd6c 0x0008fd6c 0x0008fd6c 0x00294 0x00294 R   0x1
> > > > > >
> > > > > >  Section to Segment mapping:
> > > > > >   Segment Sections...
> > > > > >    00     .ARM.exidx
> > > > > >    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> > > > > > .rodata .ARM.exidx
> > > > > >    02     .data.rel.ro .dynamic .got .data .bss
> > > > > >    03     .dynamic
> > > > > >    04
> > > > > >    05     .data.rel.ro .dynamic .got
> > > > > >
> > > > > 
> > > > > And the new...
> > > > > 
> > > > > / # /tmp/ld-musl-armhf.so.1 /usr/bin/readelf -lW /lib/ld-musl-armhf.so.1
> > > > > >
> > > > > > Elf file type is DYN (Shared object file)
> > > > > > Entry point 0x362f1
> > > > > > There are 6 program headers, starting at offset 52
> > > > > >
> > > > > > Program Headers:
> > > > > >   Type           Offset   VirtAddr   PhysAddr   FileSiz MemSiz  Flg Align
> > > > > >   EXIDX          0x07b81c 0x0007b81c 0x0007b81c 0x00008 0x00008 R   0x4
> > > > > >   LOAD           0x000000 0x00000000 0x00000000 0x7b824 0x7b824 R E 0x1000
> > > > > >   LOAD           0x07bd74 0x0007cd74 0x0007cd74 0x0054a 0x0225c RW  0x1000
> > > > 
> > > > this load segment is 4k aligned and offset vs addr is not congruent
> > > > modulo 64k, or 32k, so won't work on systems with such page size.
> > > > 
> > > > > >   DYNAMIC        0x07bebc 0x0007cebc 0x0007cebc 0x000c0 0x000c0 RW  0x4
> > > > > >   GNU_STACK      0x000000 0x00000000 0x00000000 0x00000 0x00000 RW  0x10
> > > > > >   GNU_RELRO      0x07bd74 0x0007cd74 0x0007cd74 0x0028c 0x0028c R   0x1
> > > > > >
> > > > > >  Section to Segment mapping:
> > > > > >   Segment Sections...
> > > > > >    00     .ARM.exidx
> > > > > >    01     .hash .gnu.hash .dynsym .dynstr .rel.dyn .rel.plt .plt .text
> > > > > > .rodata .ARM.exidx
> > > > > >    02     .data.rel.ro .dynamic .got .data .bss
> > > > > >    03     .dynamic
> > > > > >    04
> > > > > >    05     .data.rel.ro .dynamic .got
> > > > > 
> > > > > 
> > > > > I hope that helps.
> > > > 
> > > > yes, this is a linking issue, not musl libc.
> > > > 
> > > > alpine linux links binaries for 4k pagesize only.
> > > > 
> > > > arm linkers were updated at some point to create binaries supporting
> > > > up to 64k pagesize.  i suspect some ppl ran into issues in practice
> > > > and decided the larger binaries are not worth it, if they dont work
> > > > reliably and forced 4k page size at link time.
> > > > 
> > > > you have to raise an issue with alpine linux, if you think 32k
> > > > oage size is useful and reliably supportable.
> > > 
> > > Are they using -Wl,-z,separate-code? That incurs a large
> > > binary-size-on-disk penalty when supporting oversized pages, and IIRC
> > > something was done to make the linker default to not supporting
> > > oversized pages when that's used. It might be the reason, if arm
> > > linking is normally expected to use a larger max pagesize.
> > 
> > i looked at this now, turns out they just changed the
> > pagesize back to 4k (i missed this change):
> > 
> > https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=1a26a53a0dee39106ba58fcb15496c5f13074652
> 
> This doesn't help immediately, but a major ingredient to fix this
> situation would be getting the kernel to stop doing the wrong thing.
> Right now, it's ignoring the fact that the ELF program header
> constraints are incompatible with mmap given the oversized system
> pagesize, and just incorrectly mapping the executable and trying to
> run it anyway, whereby it blows up.
> 
> The right thing to do would be either to fail with ENOEXEC in this
> case, or when mmap with the required offset constraint fails, falling
> back to making an anonymous map and copying the whole content of the
> loadable segment into that (no COW sharing). The latter is really not
> all that bad for got/data/etc. mappings which you expect will be dirty
> (modified) anyway.
> 
> BTW the former choice (ENOEXEC) would allow doing the latter in
> userspace with a binfmt_misc loader.

Completely untested draft patch showing the concept is attached.

Rich

[-- Attachment #2: fix_elf_loader_with_oversized_pages.diff --]
[-- Type: text/plain, Size: 1034 bytes --]

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index f8c7f26f1fbb..45c50f379377 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -861,6 +861,12 @@ static int load_elf_binary(struct linux_binprm *bprm)
 	if (!elf_phdata)
 		goto out;
 
+	elf_ppnt = elf_phdata;
+	for (i = 0; i < elf_ex->e_phnum; i++, elf_ppnt++) {
+		if (elf_ppnt->p_type != PT_LOAD) continue;
+		if (ELF_PAGEOFFSET(elf_ppnt->p_vaddr - elf_ppnt->p_offset))
+			goto out;
+	}
 	elf_ppnt = elf_phdata;
 	for (i = 0; i < elf_ex->e_phnum; i++, elf_ppnt++) {
 		char *elf_interpreter;
@@ -962,6 +968,13 @@ static int load_elf_binary(struct linux_binprm *bprm)
 		if (!interp_elf_phdata)
 			goto out_free_dentry;
 
+		elf_ppnt = interp_elf_phdata;
+		for (i = 0; i < elf_ex->e_phnum; i++, elf_ppnt++) {
+			if (elf_ppnt->p_type != PT_LOAD) continue;
+			if (ELF_PAGEOFFSET(elf_ppnt->p_vaddr - elf_ppnt->p_offset))
+				goto out_free_dentry;
+		}
+
 		/* Pass PT_LOPROC..PT_HIPROC headers to arch code */
 		elf_property_phdata = NULL;
 		elf_ppnt = interp_elf_phdata;