Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Sandra Loosemore <sandra@codesourcery.com>
To: Doug Evans <dje@google.com>
Cc: Gary Benson <gbenson@redhat.com>, <gdb-patches@sourceware.org>,
	Pedro Alves <palves@redhat.com>, <Paul_Koning@Dell.com>,
	Jan Kratochvil	<jan.kratochvil@redhat.com>,
	Joel Brobecker <brobecker@adacore.com>
Subject: Re: [PATCH 4/5] Use TARGET_FILENAME_PREFIX as the system root in some cases
Date: Tue, 28 Jul 2015 20:02:00 -0000	[thread overview]
Message-ID: <55B7DF58.40008@codesourcery.com> (raw)
In-Reply-To: <001a11381844e0dffe051bf2e570@google.com>

On 07/28/2015 11:35 AM, Doug Evans wrote:
>
> One thing that comes to mind is that there's no indication/warning
> here of the potential massive responsiveness hit people may take
> if they turn this feature on, plus an explanation of what's going
> on, or how they can do things differently to avoid it.

Yes.  The delay would be more tolerable if users were given some clue 
what GDB is doing.

> If, after doing:
>
> (gdb) target remote :9999
>
> the user was first prompted with something like:
>
> "Warning: I have no way to find files with debug info locally,
> and auto-target-prefix is set to "on",
> so I will try to fetch these files from the target.
> This may take time.  If you want to avoid having me try to transfer
> files from the target, you can do the following:
> blah blah blah
> Are you sure you want to continue?"
> [suitably cleaned up, I didn't want to spend any time wordsmithing that]
>
> then that may be sufficient.  What do others think?

The problem with forcing the user to answer question here is that it'll 
screw up a lot of GDB startup scripts.  Maybe stuff like launching GDB 
from Eclipse, too.

My preference would be to make GDB give a message when transferring 
files from the target, like

Reading /full/path/to/libc-2.21.so (12236020 bytes) from target.
Use "set sysroot" to set a local pathname for this file instead.

and make the read operation interruptible so that if it's taking too 
long, the user can stop the transfer and do the "set sysroot" thing as 
suggested.

-Sandra


  reply	other threads:[~2015-07-28 20:02 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-28 17:35 Doug Evans
2015-07-28 20:02 ` Sandra Loosemore [this message]
2015-07-29 10:30   ` Gary Benson
2015-08-04 16:09 ` Joel Brobecker
  -- strict thread matches above, loose matches on Subject: below --
2015-07-28 15:36 [PATCH 0/5] Change how "target:" gets into filenames Gary Benson
2015-07-28 15:36 ` [PATCH 4/5] Use TARGET_FILENAME_PREFIX as the system root in some cases Gary Benson
2015-07-28 15:40   ` Eli Zaretskii

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=55B7DF58.40008@codesourcery.com \
    --to=sandra@codesourcery.com \
    --cc=Paul_Koning@Dell.com \
    --cc=brobecker@adacore.com \
    --cc=dje@google.com \
    --cc=gbenson@redhat.com \
    --cc=gdb-patches@sourceware.org \
    --cc=jan.kratochvil@redhat.com \
    --cc=palves@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox