The shortcomings addressed by this patch:
- the semihosting trace messages disapeared when the GDB session was started
- the semihosting exit code was not passed back to the host
- the semihosting command line was not passed properly, because it generated a
very large string, including the image fu
Signed-off-by: Liviu Ionescu
---
gdbstub.c | 12 +++
hw/arm/armv7m.c | 27 +++-
hw/arm/stellaris.c | 14 +
include/hw/arm/arm.h| 8 ---
include/hw/boards.h | 1 +
include/sysemu/sysemu.h | 7 +++
qemu-options.hx
On 18 Nov 2014, at 21:46, Liviu Ionescu wrote:
> once you integrate them...
thank you for integrating them.
so can I base my next patches on them?
> I'll work on another patch, to fix passing the semihosting command line down
> to the armv7m code, since currently this triggers exceptions due
On 19 November 2014 11:37, Liviu Ionescu wrote:
>
> On 18 Nov 2014, at 21:46, Liviu Ionescu wrote:
>
>> once you integrate them...
>
> thank you for integrating them.
>
> so can I base my next patches on them?
You can, yes. (Bear in mind that my target-arm.next
branch is a rebasing branch, so yo
On 19 Nov 2014, at 14:08, Peter Maydell wrote:
> ... (Bear in mind that my target-arm.next
> branch is a rebasing branch, so you can't git merge
> or pull it; you need to fetch it and then rebase
> your patchstack on it.)
I currently have two more branches in my git, for each of the patches app
On 19 November 2014 13:06, Liviu Ionescu wrote:
> I already explained, but will do it again.
Sorry if I missed that; there's a lot of mail on qemu-devel...
> first, we are talking about two completely different use cases.
>
> on the A profile you have a kernel, and when you start it, you 'append
On 19 Nov 2014, at 15:20, Peter Maydell wrote:
> Yes, we should do this. (We have to make cmdline= optional anyway,
> to support old and previously working command lines.)
ok, so I'll add the optional 'cmdline=' to the newly added
'-semihosting-config'. since this was not used before, it will
Hi,
with the latest submitted patches, the functionality I expect for
qemu-system-arm is complete. (I have some more cosmetic suggestions, to be
discussed later).
to test the functionality, you can download an unit test application from:
https://dl.dropboxusercontent.com/u/78151643/gcm
to test the patch, you can download an unit test application from:
https://dl.dropboxusercontent.com/u/78151643/gcm.elf
there are two use cases:
1) standalone (usually unit tests integrated in a continuous integration system)
$ ./qemu-system-arm -machine lm3s6965evb -image gcm.elf -se
On 14 Nov 2014, at 02:25, Liviu Ionescu wrote:
> (gdb) system_reset
in fact this is
(gdb) mon system_reset
Liviu
On 13 November 2014 23:55, Liviu Ionescu wrote:
> The shortcomings addressed by this patch:
> - the semihosting trace messages disapeared when the GDB session was started
> - the semihosting exit code was not passed back to the host
> - the semihosting command line was not passed properly, because
On 17 November 2014 12:32, Peter Maydell wrote:
> This (the extra command line option to specify where semihosting
> should go) is definitely a feature we should add. I think it's
> possible to make use of the QemuOpts infrastructure to support
> -semihosting # current option name with existing
On 17 Nov 2014, at 14:32, Peter Maydell wrote:
> that it's combining six different features and bug fixes into a
> single commit. Could you separate them out into their own patches?
sure. in practical terms, this requires separate branches and each be applied
to master, right?
my experience w
On 17 November 2014 14:32, Liviu Ionescu wrote:
> On 17 Nov 2014, at 14:32, Peter Maydell wrote:
>
>> that it's combining six different features and bug fixes into a
>> single commit. Could you separate them out into their own patches?
>
> sure. in practical terms, this requires separate branches
On 17 Nov 2014, at 14:35, Peter Maydell wrote:
> Compare the handling of 'rtc' and similar options: you get
> to have an "implied option name" so "-foo" is treated like
> "-foo enable=true",
I already considered this in a previous tentative version, but the
implementation doesn't work like thi
On 17 Nov 2014, at 14:32, Peter Maydell wrote:
> You can start out by separating out one or two and submitting
> those.
just did that, I submitted two patches, one to add the -semihosting-config and
one to fix the exit code.
if you have further comments, please let me know.
once you integrat
16 matches
Mail list logo
