This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: Systemtap on ARM arch results summary
- From: Mark Wielaard <mjw at redhat dot com>
- To: William Cohen <wcohen at redhat dot com>
- Cc: David Long <dave dot long at linaro dot org>, Naresh Kamboju <naresh dot kamboju at linaro dot org>, systemtap at sourceware dot org
- Date: Wed, 18 Jun 2014 09:26:28 +0200
- Subject: Re: Systemtap on ARM arch results summary
- Authentication-results: sourceware.org; auth=none
- References: <CA+G9fYtV4apghaFPDQcD7-xbvG_Ua1mEUi=jcxVYfOQdwzAPXA at mail dot gmail dot com> <5399C1BE dot 1050409 at redhat dot com> <5399C4D6 dot 3070502 at linaro dot org> <5399CA64 dot 8050809 at redhat dot com> <539B3F2C dot 8000805 at linaro dot org> <539F3FAD dot 3050802 at redhat dot com> <539F420E dot 4030205 at linaro dot org> <53A04B96 dot 7010008 at redhat dot com> <53A04E9D dot 6020500 at linaro dot org> <53A0FA70 dot 4080002 at redhat dot com>
On Tue, Jun 17, 2014 at 10:33:20PM -0400, William Cohen wrote:
> Does GDB deal with this properly? It would be worthwhile to see
> whether GDB addresses ARM/Thumb code breakpoints. It is going to
> have the same issue.
I don't know how GDB deals with this, but I have been dealing with
some ARM/Thumb issues in elfutils [1]. If you have access to a symbol
table then you can lookup inside which STT_FUNC symbol the address
falls. The zero bit of the st_value will indicate whether the function
starts in ARM or THUMB code. There are also special $a and $t STT_FUNC
symbols in the symbol table that indicate the start of ARM or THUMB
code [2].
Cheers,
Mark
[1]
https://lists.fedorahosted.org/pipermail/elfutils-devel/2014-June/004075.html
[2]
http://infocenter.arm.com/help/topic/com.arm.doc.espc0003/ARMELF.pdf