[Thread Prev][Thread Next][Index]

Re: [ferret_users] Mac memory limit problem?



Hi Ryo,

When Ferret switched to double precision for memory the "word" size did switch to 8 bytes.  But the 2G limit on memory should only be for for 32-bit operating systems. 

Karl


On Wed, Aug 6, 2014 at 8:43 AM, Andy Jacobson <andy.jacobson@xxxxxxxx> wrote:
Hi Ryo,

I can't confirm; my Mac doesn't manifest this problem:


        NOAA/PMEL TMAP
        FERRET v6.82
        Darwin 9.8.0 - 08/06/12
         6-Aug-14 09:40

yes? set memory/size=269
 Cached data cleared from memory
yes? set memory/size=5000
 Cached data cleared from memory
Internal overflow expressing 5000.0 Mwords as words (705032704)
Unable to allocate the requested 5000.0 Mwords of memory.
Memory remaining at 269.0 Mwords.
yes?


-Andy

On Tue 05 Aug 2014, at 21:49 , Ryo Furue <furue@xxxxxxxxxx> wrote:

> Ferret users (on Mac in particular),
>
> It seems that memory that Ferret can use is limited to about 1 GiB
> (giga bytes) and I thought this can be a bug.
>
> yes? set memory/size=269
> ferret_v6.82(33692) malloc: *** mmap(size=2152001536) failed (error code=12)
> *** error: can't allocate region
> *** set a breakpoint in malloc_error_break to debug
> Unable to allocate 269.0 Mwords of memory.
> Restoring previous memory of 268.0 Mwords.
>
> I can't allocate 269 Mwords, which is about 1.00 GiB if one "word" is
> 4 bytes.  Because of this problem, I can't read a variable which is a
> little over 1 GiB.  Ferret says "request exceeds memory setting: 292
> Mwords were requested."
>
> I thought 2 GiB is the actual maximum.  The error message contains
> some hints that corroborate:
>
> 1) size=2152001536 is 2.00 GiB (if it's in bytes).
>
> 2) error code = 12 is "Cannot allocate memory" (from /usr/include/sys/errno.h).
>
> From this, I guess that at some point in the Ferret code, one "word"
> is interpreted as 8 bytes.
>
> Ferret v6.82 for Mac.
>
> Best regards,
>
> Ryo
> ----
> P.S.  I use this convention: "M" == 10^6, "Mi" = 2^20, "G" == 10^9,
> "Gi" == 2^30.

--
Andy Jacobson
andy.jacobson@xxxxxxxx

NOAA Earth System Research Lab
Global Monitoring Division
325 Broadway R/GMD1
Boulder, Colorado 80305

303/578-2237










--
Karl M. Smith, Ph.D.
JISAO Univ. Wash. and TMAP/PMEL NOAA
"The contents of this message are mine personally and do
not necessarily reflect any position of the Government
or the National Oceanic and Atmospheric Administration."

[Thread Prev][Thread Next][Index]
Contact Us
Dept of Commerce / NOAA / OAR / PMEL / Ferret

Privacy Policy | Disclaimer | Accessibility Statement