IDE 7.3 Disassembly Window STILL Not Fixed

[lpcware]

Content originally posted in LPCWare by MikeSimmonds on Sat Aug 16 12:10:25 MST 2014
The disassembly window still has the issues that I reported as far back as version 5.

Namely

(1) Use the scroll bar arrow to scroll back one click. All the address are now odd (which is illegal for both
Cortex-M3 and ARM)
You can't get it back even until a code jump causes the window to be re-filled (I.e. step etc. do not fix this).

(2) Show Opcodes on the right-click context menu (in the address column) will add/remove a few pixels between
the address and the rest of the line, but does not show the opcode bytes.

If it matters (it shouldn't) I'm debugging on a Cortex-M3 MCU.

Mike

[lpcware]

Content originally posted in LPCWare by MikeSimmonds on Tue Aug 19 12:33:29 MST 2014
Firstly, I want to apologise for the pejorative adjectives that I used.
They are caused by high levels of frustration.

I DO like LPCXpresso and I DO appreciate the efforts of the support team.
The version 7 debugger is better than 6 in many ways.

It is Eclipse in general that I have problems with. I stand by my assertion that is has a shoddy, fragmented
User Interface and poor documentation of anything bar what is obvious anyway.
For instance wat does "Use Step Filters" do?, why would you want it? I never found any documentation
on that. It might be a red herring, or something useful, who knows?

Eclipse is vast, writen in a language I am not familiar with, hides thing many levels down.
Noboby not involved with Eclipse development has time or aptitude to poke about in its depths
to see if there is a useful setting here or there.

I sometimes get the impression that, because my questions are not 'mainstream', they are not
always given equal priority with more newbie issues. I hope that this is wrong.

Feedback is the issue. I do not expect instant fixes to anything, but a timely reply either saying
"you have misunderstood or made a mistake", or we are "aware of the issue" would go a long way
to ease the frustration.

I can even live with "Yes we know about that but we are not going to (or cannot) fix it".

I am mainly concerned with the debugger and some issue are outstanding from version 5;
but I am not always anoyed enough to post about them. E.g breakpoints in asm code often actually stop
at the instruction after where the breakpoint is put.

If you take anything from this post, it is that feedback is vital to keep your users happy.

Respectfully, Mike

[lpcware]

Content originally posted in LPCWare by lpcxpresso-support on Tue Aug 19 10:09:34 MST 2014
The last minor update to that FAQ was made last month yes, but the FAQ has been live on LPCware for about a year now.

The LPCXpresso specific functionality for opcodes was introduced a long time ago and still exists due to issues in the general Eclipse equivalent, which are currently being worked on - and the two should "merge" over coming releases.

And on your issue of  addresses and the scroll bar, this is still to be fully looked in to. But I was trying to be help you as soon as possible by providing a pointer to information on your opcodes issue that was already publicly available.

Finally, I am sorry from your various posts that you obviously do not like using LPCXpresso IDE / Eclipse.

We know that there are some areas we could improve further. But we work incredibly hard at continuing to drive LPCXpresso forward.  And for the large majority of our extensive user base it provides very good value for money, both in the Free and Pro Editions. And we get a lot of feedback that emphasises how much our efforts really are appreciated.

We take all the constructive comments that we receive back from users, on the forums and via other sources as input for continuing to try to improve the products. But there are limits as to what we can do and when.

Finally, as I am sure you are aware there are lots of alternatives to LPCXpresso, some of which use Eclipse and/or GCC, some of which don't. NXP provide support packages for Keil and IAR, in addition to the LPCXpresso IDE. And there are other vendors that we work with too (Rowley, Atollic, etc, etc). And then there is the completely open source route as well. There is nothing to lock you in to the LPCXpresso IDE if it does not suit your working practices.

Regards,
LPCXpresso Support

[lpcware]

Content originally posted in LPCWare by MikeSimmonds on Tue Aug 19 05:19:38 MST 2014
OK -- FAQ entry enabled the display of the opcodes. Thanks for that.  ...

BUT this FAQ was only submitted on the 24th of last month!

The contex menu in the disassembly view window (Show Opcodes) DOES NOTHING!
[Therefore, either it should not be there at all (because it give false hope), or it should actually work.]

So, given what seems to be the sensible place to set/un-set the opcode display,
how is anyone supposed to know that (a) this does not work and (b) instead have to
find a buried dialog.

The UI design of Eclipse is utter rubbish and the (lack of) documentation is worse!

Does the LPCXpresso team have any control at all on the Eclipse part of the IDE
or are they falling down on the job.

And you still have not addressesd the problem of scrolling backwards in the disassembly giving odd addresses!

[lpcware]

Content originally posted in LPCWare by lpcxpresso-support on Mon Aug 18 14:23:56 MST 2014
To enable opcode display, please use the mechanism described at:

http://www.lpcware.com/content/faq/lpcxpresso/disassembly-view

Regards,
LPCXpresso Support