public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Yao Qi <yao@codesourcery.com>
To: Eli Zaretskii <eliz@gnu.org>
Cc: <gdb-patches@sourceware.org>
Subject: Re: [PATCH 2/3] Unbuffer stdout and stderr in cygwin
Date: Thu, 01 Aug 2013 08:06:00 -0000	[thread overview]
Message-ID: <51FA16C6.6010006@codesourcery.com> (raw)
In-Reply-To: <83siyxxtif.fsf@gnu.org>

On 07/29/2013 11:42 PM, Eli Zaretskii wrote:
> I would suggest to call the function using_cygwin_pty or some such.
> "is_in_cygwin" is IMO too ambiguous.
> 

OK, fixed in patch 1/3.

>> >+      setvbuf (stdout, NULL, _IONBF, BUFSIZ);
>> >+      setvbuf (stderr, NULL, _IONBF, BUFSIZ);
> How about using line buffering instead on both streams?  Or at least
> leave stdout line-buffered?  Did you try that, and if so, did that
> have the same problems that triggered these patches?

I tried line-buffering for stdout, but get some regressions in python
related testing,

FAIL: gdb.python/py-explore-cc.exp: explore int_ptr_ref (timeout)
FAIL: gdb.python/py-explore-cc.exp: explore b (timeout)
FAIL: gdb.python/py-explore-cc.exp: explore B (timeout)

FAIL: gdb.python/py-explore.exp: explore ss_ptr (timeout)
FAIL: gdb.python/py-explore.exp: explore darray_ref (timeout)
FAIL: gdb.python/py-explore.exp: explore su (timeout)
FAIL: gdb.python/py-explore.exp: explore cs (timeout)
FAIL: gdb.python/py-explore.exp: explore cu (timeout)

however, I don't triage these fails yet.
> 
> You see, the way your patch works, using GDB from a Cygwin shell
> window will always do the above, even when not running the test
> suite.  Users might be unhappy that their standard output suddenly
> becomes much less efficient.

Yes, I get your point.  GDB would be less efficient on output.  How
about this patch? still unbuffering stdout and stderr, due to these
regression?

-- 
Yao (齐尧)

gdb:

2013-08-01  Joseph Myers  <joseph@codesourcery.com>
	    Yao Qi  <yao@codesourcery.com>

	* main.c (captured_main) [__MINGW32__]: Set stdout
	and stderr unbuffered if GDB is using Cygwin pty.
---
 gdb/main.c |   12 ++++++++++++
 1 files changed, 12 insertions(+), 0 deletions(-)

diff --git a/gdb/main.c b/gdb/main.c
index 677f587..3a2fee6 100644
--- a/gdb/main.c
+++ b/gdb/main.c
@@ -375,6 +375,18 @@ captured_main (void *data)
   saved_command_line[0] = '\0';
   instream = stdin;
 
+#ifdef __MINGW32__
+  if (using_cygwin_pty ())
+    {
+      /* A Cygwin session may not look like a terminal to the Windows
+	 runtime; ensure stdout and stderr is unbuffered.  Note that
+	 setvbuf may be used after the file is opened but before any
+	 other operation is performed.  */
+      setvbuf (stdout, NULL, _IONBF, BUFSIZ);
+      setvbuf (stderr, NULL, _IONBF, BUFSIZ);
+    }
+#endif
+
   gdb_stdout = stdio_fileopen (stdout);
   gdb_stderr = stdio_fileopen (stderr);
   gdb_stdlog = gdb_stderr;	/* for moment */
-- 
1.7.7.6

  reply	other threads:[~2013-08-01  8:06 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-07-29  8:46 [PATCH 0/3 V3] Test mingw32 GDB " Yao Qi
2013-07-29  8:46 ` [PATCH 1/3] Detect GDB is " Yao Qi
2013-07-29 15:38   ` Eli Zaretskii
2013-07-30  9:27     ` Yao Qi
2013-07-30 15:33       ` Eli Zaretskii
2013-08-01  7:52         ` Yao Qi
2013-08-01 16:33           ` Eli Zaretskii
2013-08-02  2:51             ` Yao Qi
2013-08-02  6:10               ` Eli Zaretskii
2013-08-03  4:55           ` Christopher Faylor
2013-08-04  8:45             ` Yao Qi
2013-08-05  4:41               ` Christopher Faylor
2013-08-05  6:23                 ` Yao Qi
2013-08-06  2:08                   ` Christopher Faylor
2013-08-06  3:05                     ` Yao Qi
2013-08-08  5:11                       ` Christopher Faylor
2013-08-08  7:24                         ` Yao Qi
2013-08-15 17:40                           ` Christopher Faylor
2013-08-15 18:58                             ` Tom Tromey
2013-08-15 19:14                               ` Eli Zaretskii
2013-08-16  0:06                                 ` Yao Qi
2013-08-16  2:01                                   ` Tom Tromey
2013-08-16  1:07                             ` Yao Qi
2013-08-16 16:37                               ` Christopher Faylor
2013-08-08  7:28                         ` Pierre Muller
2013-08-13  8:12                           ` Yao Qi
2013-08-13  8:23                             ` Pierre Muller
2013-07-29  8:46 ` [PATCH 2/3] Unbuffer stdout and stderr " Yao Qi
2013-07-29 15:42   ` Eli Zaretskii
2013-08-01  8:06     ` Yao Qi [this message]
2013-08-01 16:36       ` Eli Zaretskii
2013-08-02  0:40         ` Yao Qi
2013-07-29  8:46 ` [PATCH 3/3] Set stdin/stdout/stderr to binary mode " Yao Qi
2013-07-29 15:44   ` Eli Zaretskii
2013-08-01  8:10     ` Yao Qi
2013-08-01 16:37       ` Eli Zaretskii
2013-07-29 14:03 ` [PATCH 0/3 V3] Test mingw32 GDB " Pierre Muller
2013-07-30  6:03   ` Yao Qi
2013-07-29 18:03 ` Tom Tromey
2013-07-29 18:43   ` Eli Zaretskii
2013-08-13  9:35 [PATCH 0/3 V4] " Yao Qi
2013-08-13  9:35 ` [PATCH 2/3] Unbuffer stdout and stderr " Yao Qi

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=51FA16C6.6010006@codesourcery.com \
    --to=yao@codesourcery.com \
    --cc=eliz@gnu.org \
    --cc=gdb-patches@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).