similar to: nfsroot work in progress

nfsmount doesnt work anymore for me. This patch fixes at least udp nfs mounts. tcp does still hang. client and server is 2.6. --- /dev/shm/bindresvport.c 2003-12-08 02:33:24.000000000 +0100 +++ klibc/inet/bindresvport.c 2004-01-06 21:05:02.283779072 +0100 @@ -23,7 +23,7 @@ int bindresvport(int sd, struct sockaddr if (sin == NULL) { sin = &me;

Peter, we found that nfs over udp will corrupt data under very extrem load, there is no way to fix it due to the way how UDP works. TCP will not have these problems. I also wonder why the package size is only 1k. Everyone who wishes a slow connection can pass the desired options via the kernel cmdline. Everyone else prefers probably the fast mount. The defaults should look more like that: ---

I just checked in changes to kinit, such that it is now a single 29KB executable that replaces the kernel's ipconfig and nfsroot code. The kinit code isn't very elegant, but I figured it was better to save space than to fully generalise it so that it could be extended in a sort of modular fashion. If someone wants modularity, I figure they can spend the few extra KB and run the

Hey all, bug report time. I started tracking this down, but am at a bit of a loss for time, so I figured I'd report over this list to see if anyone else has the time to fix this or at least find the error. When calling nfsmount on it's own, say "nfsmount 1.2.3.4:/home/nfs /root" it works just fine, however, when calling nfsmount_main from within kinit (via nfsroot=), things get

In case anyone's interested, I followed up in the linux-nfs mailing list: https://marc.info/?l=linux-nfs&m=156887818618861&w=2 Thanks, Alkis On 9/15/19 10:51 AM, Alkis Georgopoulos wrote: > I think I got it. > > Both nfsmount and `mount -t nfs` now default to rsize/wsize = 1 MB. > By lowering this to 32K, all issues are gone, even with the default > timeo=7. And

some files depend on nodes.h, but there is no dependency for that fact. olaf@mandarine:~/klibc-0.87> grep -Er 'include.*nodes.h' ash/ | sed 's-^[^/]\+/\([^.]*\).*-\1.o-' | xargs echo OBJ_NODES = OBJ_NODES = cd.o eval.o exec.o expand.o jobs.o main.o nodes.o options.o parser.o redir.o show.o trap.o var.o make spotless doesnt work reliable, the patch adds a few more targets. Is

2ad780978b7c0c3e7877949f098cbd06e7c73839 broke klibc build MNTPROC_MNT and MNTPROC_UMNT no longer defined. usr/kinit/nfsmount/mount.c: In function ?mount_call?: usr/kinit/nfsmount/mount.c:179: error: ?MNTPROC_MNT? undeclared (first use in this function) usr/kinit/nfsmount/mount.c:179: error: (Each undeclared identifier is reported only once usr/kinit/nfsmount/mount.c:179: error: for each function

Without this patch gcc 4.6 warns: KLIBCCC usr/kinit/nfsmount/dummypmap.o usr/kinit/nfsmount/dummypmap.c: In function 'dummy_portmap': usr/kinit/nfsmount/dummypmap.c:191:13: warning: array subscript is below array bounds [-Warray-bounds] Signed-off-by: Greg Thelen <gthelen at google.com> --- usr/kinit/nfsmount/dummypmap.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)

I can't explain why 700 msecs aren't enough to avoid timeouts in 100 Mbps networks, but my tests verify it, so I'm writing to the list to request that you increase the default timeo to at least 30, or to 600 which is the default for `mount -t nfs`. How to reproduce: 1) Cabling: server <=> 100 Mbps switch <=> client Alternatively, one can use a 1000 Mbps switch and

kinit/nfsmount.c:mount_nfs_root() should use the bootpath specified by bootp/dhcp. If the "nfsroot" option is specified then it overrides the boot server bootpath and a message indicating the override is printed. --- klibc-0.194/kinit/nfsroot.c.orig 2005-01-05 04:13:47.043897880 -0700 +++ klibc-0.194/kinit/nfsroot.c 2005-01-05 04:13:09.316633296 -0700 @@ -66,34 +66,21 @@ const int

Various coding style fixes checkpatch warns about. The goal is not to be 100% checkpatch compliant, but to have more consistent coding style. As this is a trivial patch serie, will land in 24 hours in klibc git, unless of course ml review hits a bugger. Checked with size(3) that the generated kinit, fstype, ipconfig and nfsmount are the same. maximilian attems (4): [klibc] ipconfig: reduce

There is a conflict in glibc between <linux/in.h> and <netinet/in.h>: one defines protocols with an enum, the other does the same, but adds a self-define for every enumerated value. #define IPPROTO_IP IPPROTO_IP This means the two cannot be mixed, and since loads of include files refer to <netinet/in.h>, effectively <linux/in.h> cannot be used with glibc. Since

hello hpa! added on top of queue ext4 fix, that Ubuntu is carrying from cjwatson. Got missed out in previous pull requests. Btrfs recognition in fstype. the patch queue contains sync with latest dash, sh4 fix by Debian porters, dprintf usage in ipconfig, kinit and nfsmount instead of buggy DEBUG macro. ipconfig memcpy usage to avoid strict aliasing warnings. Some interesting README's got

Hi all. Proposal for directory structure in the kernel: usr/include <= _all_ include files usr/include/{net,sys,..} usr/include/arch/$(ARCH) <= arch specific include files *1 usr/klibc <= klibc files usr/klibc/arch/$(ARCH) <= arch specific config and files except include files usr/kinit <= program integrated in the kernel usr/kinit/nfsmount <= used by kinit,

Hi everybody and thanks to those persons who make klibc being possible. I am booting a lessdisks system from a server. I have got it in NFSv3 but I would like use the features of NFSv4. I am using initramfs-tools so I have in my initrd.img the utility nfsmount wich make possible booting in NFSv3. I would like know if exists some version of nfsmount or some patch or, if does not, some program (a

Hi, The following set of patches is intended mostly to improve ipconfig and nfsmount compatibility with glibc. The context: I'm working on yaird, an alternative to mkinitrd, and currently implementing nfs root file systems. This is based on ipconfig and nfsmount from klibc, with a rewritten version of kinit. Users should be able to build the package with vanilla glibc, since klibc is not

Commit-ID: c4b811a1e4647224ddc717fac59900d16d0e9d4d Gitweb: http://git.kernel.org/?p=libs/klibc/klibc.git;a=commit;h=c4b811a1e4647224ddc717fac59900d16d0e9d4d Author: Baptiste Jonglez <baptiste.jonglez at imag.fr> AuthorDate: Thu, 14 Sep 2017 09:22:21 -0700 Committer: Ben Hutchings <ben at decadent.org.uk> CommitDate: Wed, 2 Jan 2019 03:08:04 +0000 [klibc] nfsmount: support

Hi, I'm resending since I got no feedback: I assume that the message was lost... The attached patch fixes a few bugs related to the spoofing portmapper used by nfsmount. Regards, Louis -- Dr Louis Rilling Kerlabs Skype: louis.rilling Batiment Germanium Phone: (+33|0) 6 80 89 08 23 80 avenue des Buttes de Coesmes http://www.kerlabs.com/ 35700 Rennes -------------- next part

Hi Bryan and other klibc people. I have taken a stamp on the Makefiles for klibc, this is what I came up with. 1) No longer recompile on every invocation 2) Correct checking on dependencies 3) Simpler makefile syntax (almost all over the place) I compile-time tested it only. Two open issues: a) Do we realy use .a files for initramfs. I renemed that to the executable. b) I renamed

The standard mount option nowadays to specify NFS version is "nfsvers", as documented in nfs(5) on modern Linux systems. Up to now, nfsmount only supported the old "v2" or "v3" boolean options. Extend option parsing to support both "nfsvers=X" and "vers=X", with X being equal to either 2 or 3 (nfsmount does not support NFSv4 at present). If both