From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: Date: Wed, 8 Oct 2008 20:23:08 +0900 From: underspecified To: "Fans of the OS Plan 9 from Bell Labs" <9fans@9fans.net> In-Reply-To: <20081008083600.GN10037@hermes.my.domain> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_47270_26530528.1223464988429" References: <20081008083600.GN10037@hermes.my.domain> Subject: Re: [9fans] p9p acme freezes on devdraw read in MacOS X Topicbox-Message-UUID: 18cf1a3a-ead4-11e9-9d60-3106f5b1d025 ------=_Part_47270_26530528.1223464988429 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Greetings, I have been using a scroll wheel in p9p's acme on OS-X for some time without any problems. This raises a few questions about your setup: 1) Are you sure you are running the new, native gui version? Your hg info suggests this is the case, but I want to make sure you recompiled the binaries between now and then. 2) What kind of mouse is is? 3) Does the scroll wheel behave normally in other apps on OS-X? 4) Are you using any non-standard mouse drivers? 5) If you are on a laptop, does scrolling with the touchpad cause this error? --underspecified On Wed, Oct 8, 2008 at 5:36 PM, Christian Kellermann < Christian.Kellermann@nefkom.net> wrote: > Dear List, > > I am experiencing a strange behaviour of p9p acme: When I scroll > with the mouse wheel acme just hangs. The console prints acme: bad > rpc tag 2 (no one waiting on that tag) > > attaching with gdb says: > > 0x9526a1b5 in read$UNIX2003 () > (gdb) bt > #0 0x9526a1b5 in read$UNIX2003 () > #1 0x000089e7 in threadmain () > #2 0x0001aeac in threadmainstart (v=0x0) at thread.c:699 > #3 0x0001a808 in threadstart (y=552960, x=0) at thread.c:95 > > and detaching makes acme crash completely with > > settag/send tag 3: Broken pipe > settag/send tag 3: Broken pipe > Oct 8 10:27:23.908 write error: Broken pipe > settag/send tag 3: Broken pipe > > Has any one else seen this? It feels like a race condition since I > cannot reproduce it reliably other than opening a window with some > content and scrolling around with the mouse wheel. I have tried to > pull from the repo but it seems that > > changeset: 2915:dd72218f97d9 > tag: tip > user: Russ Cox > date: Thu Aug 14 10:29:29 2008 -0400 > summary: rc: fix local variables in functions > > is the latest tip. > > What can I do to help debugging this? It is really an annoying bug > I must say. > > Thanks for your insights, > > Christian > > -- > You may use my gpg key for replies: > pub 1024D/47F79788 2005/02/02 Christian Kellermann (C-Keen) > ------=_Part_47270_26530528.1223464988429 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline
Greetings,

I have been using a scroll wheel in p9p's acme on OS-X for some time without any problems.
This raises a few questions about your setup:

1) Are you sure you are running the new, native gui version? Your hg info suggests this is the case, but I want to make sure you recompiled the binaries between now and then.
2) What kind of mouse is is?
3) Does the scroll wheel behave normally in other apps on OS-X?
4) Are you using any non-standard mouse drivers?
5) If you are on a laptop, does scrolling with the touchpad cause this error?

--underspecified

On Wed, Oct 8, 2008 at 5:36 PM, Christian Kellermann <Christian.Kellermann@nefkom.net> wrote:
Dear List,

I am experiencing a strange behaviour of p9p acme: When I scroll
with the mouse wheel acme just hangs. The console prints acme: bad
rpc tag 2 (no one waiting on that tag)

attaching with gdb says:

0x9526a1b5 in read$UNIX2003 ()
(gdb) bt
#0  0x9526a1b5 in read$UNIX2003 ()
#1  0x000089e7 in threadmain ()
#2  0x0001aeac in threadmainstart (v=0x0) at thread.c:699
#3  0x0001a808 in threadstart (y=552960, x=0) at thread.c:95

and detaching makes acme crash completely with

settag/send tag 3: Broken pipe
settag/send tag 3: Broken pipe
Oct  8 10:27:23.908 write error: Broken pipe
settag/send tag 3: Broken pipe

Has any one else seen this? It feels like a race condition since I
cannot reproduce it reliably other than opening a window with some
content and scrolling around with the mouse wheel. I have tried to
pull from the repo but it seems that

changeset:   2915:dd72218f97d9
tag:         tip
user:        Russ Cox <rsc@swtch.com>
date:        Thu Aug 14 10:29:29 2008 -0400
summary:     rc: fix local variables in functions

is the latest tip.

What can I do to help debugging this? It is really an annoying bug
I must say.

Thanks for your insights,

Christian

--
You may use my gpg key for replies:
pub  1024D/47F79788 2005/02/02 Christian Kellermann (C-Keen)

------=_Part_47270_26530528.1223464988429--