Hi,
I have gotten shared memory to work on this machine but my shared memory
block size was small, about 700 bytes to test out my code. Now I was ready
to increase it and I get a Segmenation Fault when I /sbin/insmod my rt
process. Here is the information I have and my questions after searching
the archives for many hours and trying things !!!
First, I am running rtlinux 1.1 on kernel 2.0.36. I have approx. 128
MBytes of Memory
My /proc/cpuinfo is:
processor: 0
cpu: 586
model: Mobile Pentium MMX
vendor_id: GenuineIntel
stepping: 1
fdiv_bug: no
hlt_bug: no
f00f_bug: yes
fpu: yes
fpu_exception: yes
cpuid: yes
wp: yes
flags: fpu vme de pse tsc msr mce cx8 mmx
bogomips: 532.48
My /proc/meminfo is:
total used free shared
buffers cached
mem: 1308822144 84975616 45846528 6950912
68554752 4681728
swap: 13156766 0 13156766
memTotal: 127756 kB
memFree: 44772 kB
memShared: 6788 kB
Buffers: 66948 kB
cached: 4572 kB
SwapTotal: 128484 kB
SwapFree: 128484 kB
Which seems reasonable to me because I am using "append=mem127m" on my ~
128MByte machine. And the real memory information shows I have less than
128MBytes.
My shared memory structure has ONLY 4512 BYTES !!! (I do want to increase
this, but when I ran into trouble I decreased it to see what was going on).
AFTER I insmod my real-time process I get Messages on the monitor (General
Protection and a Dump) and my /proc/meminfo changes --- Specifically the
shared memory BUGS ME !!
My /proc/meminfo is AFTER IS:
total used free shared
buffers cached
mem: 1308822144 86937600 43884544 1078468
68554752 4829184
swap: 13156766 0 13156766
memTotal: 127756 kB
memFree: 42856 kB
memShared: 10532 kB
Buffers: 66948 kB
cached: 4716 kB
SwapTotal: 128484 kB
SwapFree: 128484 kB
I can also see that my ISR is being entered at the specifed rate when I run
my application, but memory has obviously been messed up and I get
segmenation faults after that. I do know that my rt process is sound
because this will run with a small shared memory block, I have tried up to
712 bytes.
I have all the general protection information if it is necessary I can pass
that all along.
------------------------------------------------------
I read in the archives something from Randal Carpenter --
Hey I wonder if he has that AMD K6 revision A that had the > 32M bug...it
caused seg faults. If so send it back and get a K6 rev B. I had an A andB
and can say definetly A was broken and B was fixed. It shows asstepping 1
and stepping 2 (A and B respectively) in the /proc/cpuinfo later Randal
COULD THIS APPLY TO ME ??? AND IF SO WHAT DO I DO ABOUT IT ???
--------------------------------------
Thanks for any help ---
Janet Estabridis
Electrical Engineer
NAWC Code 473E00D
China Lake, CA 93555
[EMAIL PROTECTED]
(760) 939-2896 FAX (760) 939 -3075
--- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
----
For more information on Real-Time Linux see:
http://www.rtlinux.org/~rtlinux/
