[Bug 50869] New: Regressions caused by new start.exe behaviour in Wine 6.5 (stdout redirect, killing inferior, wine start /unix)
https://bugs.winehq.org/show_bug.cgi?id=50869 Bug ID: 50869 Summary: Regressions caused by new start.exe behaviour in Wine 6.5 (stdout redirect, killing inferior, wine start /unix) Product: Wine Version: 6.5 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: programs Assignee: wine-bugs(a)winehq.org Reporter: mail(a)robbertvanderhelm.nl Distribution: --- I noticed a few regressions and new weird behaviours after updating to Wine 6.5 with regards to launching applications now that everything gets launched through start.exe: 1. When launching Winelib applications all STDOUT output gets lost. I tested this with a simple C++ file compiled with wineg++ that tries to print some text using both `printf()` and `std::cout`. Printing to STDERR using `std::cerr` still works fine. 2. When spawning a Wine child process from another application, you can no longer kill that child process by raising SIGINT or SIGKILL for the pid of the child. In Wine 6.5 this now kills the start.exe process, but it leaves the inferior process running in the background. To reproduce this, run `wine notepad&` in `fish` (bash seems to kill the child process by itself), use `ps -aux | grep notepad` or `jobs` to find the pid of the process just named `notepad`, and then run `kill -SIGKINT <pid>` (or with SIGKILL). Even though the process launched using `wine notepad&` has exited, Notepad is still running in the background. 3. (duplicate of bug 50867, but I still wanted to mention it here for completeness' sake) Launching binaries with Wine from the GUI by clicking on them (using the bundled wine.desktop) no longer works. This uses `wine start /unix`, which was already reported to no longer work as of Wine 6.5 in the other issue because `ShellExecuteEx` cannot find the file. 4. And lastly, as discovered by @stefand, `wine https://google.com` now launches the regular system browser. This is probably not really a bug as much as a consequence of the start.exe change, but I just wanted to mention it. Please let me know if I should split this up into separate tickets. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 Stefan Dösinger <stefan(a)codeweavers.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |stefan(a)codeweavers.com --- Comment #1 from Stefan Dösinger <stefan(a)codeweavers.com> --- 50e33a213d7382906594e565b5813a41d87117fe is probably the culprit here. Can you confirm that it works OK before this commit? -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 --- Comment #2 from Robbert van der Helm <mail(a)robbertvanderhelm.nl> --- I compiled from the parent of that commit (aef821cc), and issues 1, 2 and 3 still there. Is there another potentially relevant commit I should test? I can do a bisect, but that's probably doing to take a while. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 --- Comment #3 from Alexandre Julliard <julliard(a)winehq.org> --- (In reply to Robbert van der Helm from comment #0)
Please let me know if I should split this up into separate tickets.
Yes please. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 Robbert van der Helm <mail(a)robbertvanderhelm.nl> changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|Regressions caused by new |Killing Wine process in |start.exe behaviour in Wine |Wine 6.5 doesn't terminate |6.5 (stdout redirect, |the application because of |killing inferior, wine |the new use of start.exe |start /unix) | -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 --- Comment #4 from Robbert van der Helm <mail(a)robbertvanderhelm.nl> --- I've created a separate issue for (1) (bug 50879), and I've added a comment to bug 50867 for (3). (4) is probably not a bug but more of an unintional consequence of this change, right? Or would `wine 'https://google.com'` cause `iexplore` or something to pop up? Anyways, this is (2) repeated once more (since I can't change the description of this bug): When spawning a Wine child process from another application, you can no longer kill that child process by raising SIGINT or SIGKILL for the pid of the child process you spawned using (v)fork and exec. In Wine 6.5 this now kills the start.exe process, but it leaves the actual inferior process running in the background. To reproduce this, run `wine notepad&` in `fish` (bash seems to kill the child process by itself), use `ps -aux | grep notepad` or `jobs` to find the pid of the process just named `notepad`, and then run `kill -SIGKINT <pid>` (or with SIGKILL). Even though the process launched using `wine notepad&` has exited, Notepad is still running in the background. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 --- Comment #5 from Alexandre Julliard <julliard(a)winehq.org> --- (In reply to Robbert van der Helm from comment #4)
I've created a separate issue for (1) (bug 50879), and I've added a comment to bug 50867 for (3). (4) is probably not a bug but more of an unintional consequence of this change, right? Or would `wine 'https://google.com'` cause `iexplore` or something to pop up?
Yes, that's intentional. It may not be all that useful with a URL, but it lets you run a .msi file directly for instance. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 Eric Pouech <eric.pouech(a)orange.fr> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |eric.pouech(a)orange.fr --- Comment #6 from Eric Pouech <eric.pouech(a)orange.fr> --- Created attachment 71260 --> https://bugs.winehq.org/attachment.cgi?id=71260 tentative patch -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 --- Comment #7 from Eric Pouech <eric.pouech(a)orange.fr> --- for issue #2 in comment 1: tested with: - wine notepad& - kill <pid of start.exe> - both start.exe and notepad are terminated A+ -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 --- Comment #8 from Robbert van der Helm <mail(a)robbertvanderhelm.nl> --- I can confirm that this patch also resolves the issue in the situation where I initially encountered it, namely launching and subsequently trying to kill Winelib processes (https://github.com/robbert-vdh/yabridge/blob/b7a13652e910c1956677d296be69575...). I have not yet done any further testing beyond that though. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 Eric Pouech <eric.pouech(a)orange.fr> changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #71260|0 |1 is obsolete| | --- Comment #9 from Eric Pouech <eric.pouech(a)orange.fr> --- Created attachment 71267 --> https://bugs.winehq.org/attachment.cgi?id=71267 tentative patch (v2) thanks Robert for testing V2 of the tentative patch - Only including child of start.exe in the job, not the grandchildren -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 --- Comment #10 from Gijs Vermeulen <gijsvrm(a)gmail.com> --- Hi Robbert, Could you confirm that all the described issues disappeared after <https://source.winehq.org/git/wine.git/commit/a1abb524fee5b033e761d700b87353f14b009680>? -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 Robbert van der Helm <mail(a)robbertvanderhelm.nl> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution|--- |FIXED --- Comment #11 from Robbert van der Helm <mail(a)robbertvanderhelm.nl> --- Yes, all of these issues have been fixed! -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 Gijs Vermeulen <gijsvrm(a)gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Fixed by SHA1| |a1abb524fee5b033e761d700b87 | |353f14b009680 -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 Alexandre Julliard <julliard(a)winehq.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED --- Comment #12 from Alexandre Julliard <julliard(a)winehq.org> --- Closing bugs fixed in 7.5. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 Michael Stefaniuc <mstefani(a)winehq.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Target Milestone|--- |7.0.x -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
https://bugs.winehq.org/show_bug.cgi?id=50869 Michael Stefaniuc <mstefani(a)winehq.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Target Milestone|7.0.x |--- --- Comment #13 from Michael Stefaniuc <mstefani(a)winehq.org> --- Removing the 7.0.x milestone from bug fixes included in 7.0.1. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
participants (1)
-
WineHQ Bugzilla