#1820 closed defect (fixed)
nfs-utils-1.0.7 and gcc4
| Reported by: | Owned by: | ||
|---|---|---|---|
| Priority: | normal | Milestone: | |
| Component: | BOOK | Version: | SVN |
| Severity: | blocker | Keywords: | |
| Cc: |
Description ¶
The build of nfs-utils (http://www.linuxfromscratch.org/blfs/view/svn/server/nfs-utils.html) fails with the following error on gcc4 systems.
gcc -g -O2 -D_GNU_SOURCE -DHAVE_TCP_WRAPPER -I../../support/include -I/include -Wall -pipe -DVERSION="\"nfs-utils 1.0.7\"" -DNFS3_SUPPORTED -I../../support/rpc/include -DDEBUG -c -o svc_auth_gss.o svc_auth_gss.c svc_auth_gss.c: In function 'svcauth_gss_import_name': svc_auth_gss.c:124: warning: passing argument 3 of 'gss_import_name' discards qualifiers from pointer target type svc_auth_gss.c: In function '_svcauth_gss': svc_auth_gss.c:385: error: invalid lvalue in assignment make[2]: * [svc_auth_gss.o] Error 1 make[1]: * [all] Error 2 make: * [all] Error 2
A patch already exists in the patches repo to fix this issue. http://www.linuxfromscratch.org/patches/downloads/nfs-utils/nfs-utils-1.0.7-gcc4-1.patch
Change History (9)
by , 19 years ago
| Attachment: | nfs-utils_gcc-4.patch added |
|---|
comment:1 by , 19 years ago
I came across this when I was recently building nfs-utils. It was an easy addition, so I went ahead and made a patch for the BLFS book. Hope it's helpful.
comment:2 by , 19 years ago
| Priority: | low → normal |
|---|---|
| Severity: | minor → blocker |
Increasing Priority and Severity because until this is fixed, the instructions for nfs-utils are broken -- it will not compile.
comment:3 by , 19 years ago
| Owner: | changed from to |
|---|---|
| Status: | new → assigned |
OK. I'm going to build this once to make sure the patch does what it says it's going to do. But, I don't have access to any nfs shares. I take it that you (Jeremy) and Joe both found that nfs still works after patching?
Thanks for the book patch.
comment:4 by , 19 years ago
Hmm. Before you commit you better let me test an actual share at home. I haven't actually connected via a client yet. We use the patch in the LiveCD, the build finishes fine with the patch and is broken without it, and I know the server starts up successfully when built with the patch -- that's as far as I can say right at the moment.
Joe probably has tested an actual connection, though. If not, as I said, I'll check later tonight when I get home.
comment:5 by , 19 years ago
I've been using that patch for a while now. I use it on my nfs-server which serves packages and netboots. That system is gcc-4.0.2 and glib-2.3.6. I use the client part of it on my laptop and sparc which are both gcc-4.1.0 and glibc-2.4. I havn't found an issue with it yet.
comment:6 by , 19 years ago
I have been using the newer version (a release candidate) instead of using a patched 1.0.7 and it works just fine. There's been a whole slew of bugfixes, it seems mature and a full release is probably right around the corner.
1.0.7 is almost a year and a half old, the 1.0.8-rc4 is a few days old. Many, many bug fixes. Of course, there is no GCC issue and everything works just fine as best as I can tell.
A little bit of juggling would be required in the dependencies. I am going to give a shout out to -dev about this as you haven't closed the bug yet, and using the rc4 release may be a better way to go.
comment:7 by , 19 years ago
| Resolution: | → fixed |
|---|---|
| Status: | assigned → closed |
Fix applied in r5845. Thanks, Joe and Jeremy.
Updates the BLFS book to close this ticket