Rtems-4.10.2 and RPC on PowerPC
Kate Feng
feng at bnl.gov
Thu Oct 18 15:20:41 UTC 2012
I applied PR 1786, and PR 1797 to RTEMS4-10.0.
https://www.rtems.org/bugzilla/show_bug.cgi?id=1786
https://www.rtems.org/bugzilla/show_bug.cgi?id=1797
In addition, for RTEMS4-10.0, one needs to apply the patch below for the
cpukit/libfs/src/nfsclient/src/nfs.c
********************************************************************************
diff -a -u nfs.c.orig nfs.c
--- nfs.c.orig 2010-07-01 11:18:06.000000000 -0400
+++ nfs.c 2012-04-26 12:41:25.383414776 -0400
@@ -128,7 +128,6 @@
/* we use a dynamically assigned major number */
#define NFS_MAJOR (nfsGlob.nfs_major)
-
/* NOTE: RTEMS (ss-20020301) uses a 'short st_ino' type :-( but the
* NFS fileid is 32 bit. [Later versions of RTEMS have fixed this;
* nfsInit() issues a warning if you run a version with 'short st_ino'.]
@@ -1574,7 +1573,7 @@
*/
assert( node->args.name );
- *(const char**)arg = pathname + (node->args.name - p);
+ *(arg->c) = pathname + (node->args.name - p);
#if 0
/* restore the directory node */
@@ -2616,6 +2615,9 @@
if (count > NFS_MAXDATA)
count = NFS_MAXDATA;
+ if (count > nfsStBlksize)
+ count = nfsStBlksize;
+
SERP_ARGS(node).readarg.offset = iop->offset;
SERP_ARGS(node).readarg.count = count;
SERP_ARGS(node).readarg.totalcount = UINT32_C(0xdeadbeef);
@@ -2720,6 +2722,8 @@
if (count > NFS_MAXDATA)
count = NFS_MAXDATA;
+ if (count > nfsStBlksize)
+ count = nfsStBlksize;
SERP_ARGS(node).writearg.beginoffset = UINT32_C(0xdeadbeef);
if ( LIBIO_FLAGS_APPEND & iop->flags ) {
@@ -3247,8 +3251,8 @@
return 0;
}
Yes, RTEMS4.10.1 and RTEMS4.10.2 still have bugs.
Hope this helps,
Kate Feng
On 10/18/2012 02:43 AM, rwas wrote:
> Anyone had luck getting RPC apps to run under rtems-4.10.2?
>
>
> I have code that ran fine under rtems-4.9 (powerpc) but crashes (the
> POSIX thread) under rtems-4.10.2 when rpc functions (in this case
> pmap_unset())
> are called.
>
> And yes I've called "/rtems_rpc_start_portmapper()"/.
>
> Thanks.
>
> Robert W.
>
>
> bsp: mvme5500
> os: suse12.1
> rtems: 4.10.2
> _______________________________________________
> rtems-users mailing list
> rtems-users at rtems.org
> http://www.rtems.org/mailman/listinfo/rtems-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/users/attachments/20121018/7b7b5f13/attachment-0001.html>
More information about the users
mailing list