xfs
[Top] [All Lists]

Re: bug? sv_wait() and gcc3

To: linux-xfs@xxxxxxxxxxx
Subject: Re: bug? sv_wait() and gcc3
From: Kelledin <kelledin+XFS@xxxxxxxxxxxxxxxxxxx>
Date: Mon, 12 May 2003 17:48:41 -0500
In-reply-to: <200305101230.28794.kelledin+XFS@xxxxxxxxxxxxxxxxxxx>
References: <200305101230.28794.kelledin+XFS@xxxxxxxxxxxxxxxxxxx>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: KMail/1.5.1
Further details:

I investigated sv_wait, and it seems that function/macro is 
intended to deliberately schedule a non-interruptible deadlock 
for as long as the platform will allow (MAX_SCHEDULE_TIMEOUT).  
That's a pretty long time in jiffies--and on platforms where 
long ints are 64 bits (like Alpha), it's an outrageously long 
time.

My questions now:
1) Apparently the XFS driver code is deadlocking deliberately.  
Why?

2) Why does the forced deadlock have to last for such a long 
time?  Could we at least clamp it down to some smaller value?  
Voluntarily forcing a billion-or-so-year deadlock period just 
seems unreasonable.

-- 
Kelledin
"If a server crashes in a server farm and no one pings it, does 
it still cost four figures to fix?"


<Prev in Thread] Current Thread [Next in Thread>