public inbox for gdb-prs@sourceware.org
help / color / mirror / Atom feed
From: "vries at gcc dot gnu.org" <sourceware-bugzilla@sourceware.org>
To: gdb-prs@sourceware.org
Subject: [Bug testsuite/30250] New: [gdb/testsuite] remotedir review
Date: Mon, 20 Mar 2023 12:21:34 +0000	[thread overview]
Message-ID: <bug-30250-4717@http.sourceware.org/bugzilla/> (raw)

https://sourceware.org/bugzilla/show_bug.cgi?id=30250

            Bug ID: 30250
           Summary: [gdb/testsuite] remotedir review
           Product: gdb
           Version: HEAD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: testsuite
          Assignee: unassigned at sourceware dot org
          Reporter: vries at gcc dot gnu.org
  Target Milestone: ---

Board variable remotedir is documented as intended for remote target:
...
name       example             description
remotedir  /tmp/runtest.[pid]  Directory on the remote target in which
                               executables are downloaded and executed
...
I didn't spot anything target-specific though in the implementation, so I
wonder if it can also be used for remote host.

Anyway, first looking at the implementation:
....
    if { [board_info $dest exists remotedir] } {
        set remotedir [board_info $dest remotedir]
        set status [remote_exec $dest mkdir "-p $remotedir"]
        if { [lindex $status 0] != 0 } {
            perror "Couldn't create remote directory $remotedir on $dest"
            return ""
        }
        set destfile $remotedir/$destfile
    }
...
it looks odd to me to hardcode "/" as path separator, but perhaps that's
considered ok because it's not enabled by default.  I see something similar for
nfsdir.

Then, the implementation doesn't look complete.

That is, remote_exec and remote_download both take remotedir into account, but
not remote_upload.

So, say we have remotedir=/tmp, and we have an exec ./a.out on build/host, and 
the exec generates ./bla.txt.

A remote_download of a.out to target gives us /tmp/a.out.

Doing a remote_exec of /tmp/a.out gives us /tmp/bla.txt.

Then doing a remote_upload from host of ./bla.txt doesn't work, we need to use
./tmp/bla.txt.

I don't see any reason why remote_upload cannot do the same as remote_download.

Finally, target_compile uses this style remote_upload for host, so if we want
remote host to work with remotedir, we need to fix this.

I'd start out with a local override of remote_upload, then try to upstream the
change.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

             reply	other threads:[~2023-03-20 12:21 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-20 12:21 vries at gcc dot gnu.org [this message]
2023-03-20 16:06 ` [Bug testsuite/30250] " cvs-commit at gcc dot gnu.org

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=bug-30250-4717@http.sourceware.org/bugzilla/ \
    --to=sourceware-bugzilla@sourceware.org \
    --cc=gdb-prs@sourceware.org \
    /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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).