tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Caldarale, Charles R" <Chuck.Caldar...@unisys.com>
Subject RE: Tomcat dies suddenly
Date Thu, 04 Feb 2010 23:19:52 GMT
> From: Christopher Schultz [mailto:chris@christopherschultz.net]
> Subject: Re: Tomcat dies suddenly
> 
> 1. 32-bit Linux didn't support 64GiB of system memory, which it does

I didn't make it clear: this is not a software issue, it's the way the x86 architecture works
for addressing memory over 4 GB.  (Like the man said, "Who would ever need more than 64k?")

> 2. Processes used direct-memory addressing, which they don't

It's not a problem for user code, but rather for the privileged OS code that has to manage
the memory space.

> 3. All kernel structures must reside in <4GiB memory space (maybe
> true?)

No, it's generally not the Linux kernel structures, it's the ones required by the hardware
- paging and I/O stuff, IIRC.  (It's been awhile since I've tried to do anything interesting
on a 32-bit platform.)

> 4. The kernel was too stupid to move it's structures when necessary

There are some that the standard Linux kernel won't move even on 64-bit systems (which causes
us a great deal of pain during dynamic reconfiguration).

> 6. Carl was using 32-bit Linux, which he isn't :(

Correct, which made the whole point moot, so I'm not sure why Dan even brought it up.

> AFAIK, 64-bit Linux has a wide-open memory addressing scheme. Maybe it
> considers everything under 17 billion GiB to be "low memory", now :)

No, the hardware restrictions don't exist in 64-bit mode.

 - Chuck


THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE PROPRIETARY MATERIAL and is thus
for use only by the intended recipient. If you received this in error, please contact the
sender and delete the e-mail and its attachments from all computers.

Mime
View raw message