Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [tcf-dev] Source-level RunControl and value-adds
  • From: Eugene Tarassov <eugenet@xxxxxxxxxx>
  • Date: Wed, 15 Jan 2020 00:22:17 +0000
  • Accept-language: en-US
  • Arc-authentication-results: i=1; mx.microsoft.com 1; spf=pass smtp.mailfrom=xilinx.com; dmarc=pass action=none header.from=xilinx.com; dkim=pass header.d=xilinx.com; arc=none
  • Arc-message-signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector9901; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=6dCeIst10TrbTmRHKEX1EafSzrTjQ+6XsaSTm6mp5fg=; b=XjtSVdgIrKz8cNDYTrV8vUXJr4Fte6ZB+A2qLqcybc+dilLdcI/ru1aV6AAFVRXa8DOYIZ1qB2uqzVCE0dN4BrMXfG6er/oCHT5gz0Xlq8WxQb8Teed+sFvL779WqeNFD4eGjUkyUHzOHBIKYPTu1OeRh4mRJrovhf2fDE17szFwjx6P6e6JogSlEAj49m3JF0bxVJnccXiegIBAKznMWjSOjagoftTQ58+S0OXB/JJjaSlQMnMIqs7ckTXsfqKTjQTAL2Oc9uqqh7gU125EGDTTnGIfPTf/H+2DthsP+2HQzfNNRU6jQeil7cHybzrYqF7nBX9TfftCsu19qRXdXg==
  • Arc-seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=mIoiNd9022XWHRLZkQFrysN1jmp0f8EpFzjq10hPLPVGccxP/u9Jawf9Emxyb6WhJLFnL5iJIJ0l9Ki1SSFN85snqusROedhvN3oFItzolY9Bk+JU0anGEbRgCid5WJBZbjPaQ+truBOjTJfECRtUB+T+2LqmGSqDfVXVwp1+//qWO5OBUVXRoeJFAlRzMpNV42AOSnhNKNTAx7pgqznmrck8qu3KkTstkFOOsyXHkcWJPc+n40UpciXC4bVwCFiP+iZV2xCQlfMZWjUUHm21/RrN9p7m9X4KcTr4d29gM0Xn2WHJMgKDcVGBpIcU42fwJJi5k0BV+C5p+pH70pwzw==
  • Delivered-to: tcf-dev@xxxxxxxxxxx
  • List-archive: <https://www.eclipse.org/mailman/private/tcf-dev>
  • List-help: <mailto:tcf-dev-request@eclipse.org?subject=help>
  • List-subscribe: <https://www.eclipse.org/mailman/listinfo/tcf-dev>, <mailto:tcf-dev-request@eclipse.org?subject=subscribe>
  • List-unsubscribe: <https://www.eclipse.org/mailman/options/tcf-dev>, <mailto:tcf-dev-request@eclipse.org?subject=unsubscribe>
  • Thread-index: AdXLHutvZg7xQEETTCiGQYVzJCG80gAEuyGQ
  • Thread-topic: Source-level RunControl and value-adds
Hi Mircea,

I don't know.
It should be possible.
I think splitting the debugger into 2 halves, like this, is not a good idea.

> Think of A as a "gdbserver" ...

Not a good analogy.
Gdbserver is roughly equivalent of TCF context.h implementation.
While the rest of TCF Agent, including all services, is equivalent to gdb.
Your split is not like gdb/gdbserver split.

It would make sense if you put only context.h implementation on the target to save target memory.
In fact, you could use gdbserial protocol for context.h implementation.
We did that in our product to talk to QEMU, and it worked fine.

It would make sense if you put all services, except Symbols and Line Numbers, on the target for maximum performance.
This is how original VxWorks support was done, and it worked fine.

I, personally, don't believe into the value-add story, it is not my idea of a good architecture, and I don't know how to make it to work well.

Regards,
Eugene



-----Original Message-----
From: tcf-dev-bounces@xxxxxxxxxxx <tcf-dev-bounces@xxxxxxxxxxx> On Behalf Of Gherzan, Mircea
Sent: Tuesday, January 14, 2020 1:11 PM
To: tcf-dev@xxxxxxxxxxx
Cc: Mustafa, Muhammad <muhammad.mustafa@xxxxxxxxx>; Soare, Cristian <cristian.soare@xxxxxxxxx>
Subject: [tcf-dev] Source-level RunControl and value-adds

EXTERNAL EMAIL

Hi Eugene,

I've got a question about whether/how source-level run control can work over a value-add. Let's assume:

        - TCF Agent "A" with just the "basic services" enabled (Memory, Registers, RunControl, Breakpoints)
        - TCF Agent "B" (a value-add) which higher-level services (StackTrace, LineNumbers, Expressions, Symbols)

Think of A as a "gdbserver" without any debug info support (just assembly-level, "bare metal" operations), and B as the "gdb" connected to it (with full debug info capabilities). B uses the context-proxy logic and the cache mechanism. The value-add services work fine as they deal with halted contexts. But is there a chance to expose a "debug-info enabled RunControl" service in B? In other words, could a client (e.g. Eclipse) connected to the B-on-top-of-A stack perform RC primitives like step-over-line and step-out?

Thanks in advance,
- Mircea

_______________________________________________
tcf-dev mailing list
tcf-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://www.eclipse.org/mailman/listinfo/tcf-dev

Back to the top