| Age | Commit message (Collapse) | Author |
|---|
|
|
|
this code was if(0) for a long time due to wrong parentesis,
fixed parentesis cause print spam on some machines making them
unusage (kenji okomoto). removing the check alltogether.
|
|
|
|
|
|
|
|
|
|
this is a reimplementation of cpu and import utilities in rc using a tlsclient
and tlssrv as the encryption and authentication layers. there is only one new
service, which after authentication and encryption setup accepts an arbitrary
rc script over the network and executes it with the standard filedescriptors
redirected to the conversaion (this is *after* authentication and in the
context of the authorized user).
the new rcpu program has a few improvements over cpu(1):
- doesnt mangle program arguments
- the remote process will get the clients standard file descriptors, so error
and output are separated and you can consume the clients input from the
remote side :-)
- forwards error status of remote process
theres no backwards mode for rimport, but a new program called rexport
for the same purpose.
all these services use exportfs without the bolted on initial handshake,
so the hope is to clean up exportfs in the future and remove all the ugly
crap in there.
|
|
path)
|
|
when we got a bad root tree, we should speak the right language
and return a Rerror response instead of crapping ascii into
the 9p conversation.
|
|
AuthInfo structure to avoid leaking secrets
|
|
|
|
theres a bootstrap problem:
when /bin/init is run, it processes /lib/namespace where we might want to
mount or bind resources to /n or /mnt. but mntgen was run later in
cpurc/termrc so these mounts would be ignored.
we already have mntgen in bootfs, so we can provide these mountpoints early.
i keep the termrc/cpurc mntgens where they are, but ignore the error
prints. this way old kernels will continue to work.
|
|
|
|
|
|
When a window is moved or reshaped, that implicitely tops
the window and makes it current. The conseqence of this
is that we always have to redraw the window as if it where
a current window in any case. This was handled for Reshaped
windows, but not when the window was just moved. We now
handle both cases the exact same way, getting rid of the
Moved wctl message.
|
|
|
|
|
|
|
|
tlsSecRSAs(), cleanup
|
|
|
|
|
|
similar to common lisp grey-streams or golang's io.Reader/io.Writer. Now 95% bug-free.
|
|
-I$HOME/source/harvey/ape/include -mno-red-zone -ffreestanding -fno-builtin -nostdlib -trigraphs -D_SUSV2_SOURCE -D_POSIX_SOURCE -D_LIMITS_EXTENSION -D_BSD_SOURCE -D_BSD_EXTENSION -DHAVE_SOCK_OPTS -DHARVEY -DPlan9" LDFLAGS="-static $HOME/source/harvey/ape/amd64/lib/crt1.o $HOME/source/harvey/ape/amd64/lib/crti.o $HOME/source/harvey/ape/amd64/lib/crtn.o -L$HOME/source/harvey/ape/amd64/lib -L$HOME/source/harvey/amd64/lib" LIBS="-lbsd -lap -lc" LIBM="" --prefix=$HOME/source/harvey/ape/ports *--host=x86_64-linux-gnu --build=x86_64-harvey *--enable-shared=no
|
|
when the 9p end is closed, post a note to our private note
group to shoot down http procs. this results in quicker
release of namespace resources.
|
|
|
|
|
|
oversight, closekey is called again even if addkey failed, which
will have Key.priv nil.
stack trace:
memset(data=0x0,n=0x1f8)+0x58 /sys/src/libc/arm/memset.s:40
p9skclosekey(k=0x55358)+0x18 /sys/src/cmd/auth/factotum/p9sk1.c:493
closekey(k=0x55358)+0x7c /sys/src/cmd/auth/factotum/util.c:247
ctlwrite(a=0x552f8,atzero=0x0)+0x320 /sys/src/cmd/auth/factotum/rpc.c:503
fswrite(r=0x53b78)+0x1b0 /sys/src/cmd/auth/factotum/fs.c:574
swrite(r=0x53b78,srv=0x423e8)+0x134 /sys/src/lib9p/srv.c:559
srvwork()+0x2a4 /sys/src/lib9p/srv.c:746
srv(srv=0x423e8)+0xcc /sys/src/lib9p/srv.c:825
postproc()+0x3c /sys/src/lib9p/post.c:120
rforker(flag=0x1,fn=0x39f98,arg=0x423e8)+0x34 /sys/src/lib9p/rfork.c:16
_postmountsrv(s=0x423e8,name=0x0,mtpt=0x46f70,flag=0x1)+0xb8 /sys/src/lib9p/post.c:27
postmountsrv(name=0x0,mtpt=0x46f70,flag=0x1)+0x20 /sys/src/lib9p/rfork.c:32
main(argv=0x1fffff9c,argc=0x0)+0x23c /sys/src/cmd/auth/factotum/fs.c:157
_main+0x28 /sys/src/libc/arm/main9.s:19
acid: /sys/src/cmd/auth/factotum/p9sk1.c:493
|
|
|
|
|
|
|
|
|
|
in ecverify
|
|
|
|
|
|
|
|
make kbdproc() and mouseproc() share fd table with the main proc
and not explicitely close the file descriptors. so /dev/mouse gets
closed *after* /dev/draw/new to avoid the white window refresh issue.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
apparently, this causes some quadcore ramnode vm to hang on boot,
even tho all cores successfully started up and are operational.
i suspect some side effect from timersinit()... this would also
mean *notsc= would break it (syncclock() would continue)...
its unclear.
i'm reverting this for now until the problem is better understood.
|
|
|
|
|
|
|
|
|
|
thunderbirdsarego
when testing in qemu, launching each ap became slower and slower
because all the ap's where spinning in syncclock() waiting for
cpu0 to update its mach0->tscticks, which happens only much later
after all cpu's have been started up.
now we wait for each cpu to do its timer callibration and
manually update our tscticks while we wait and each cpu will
not spin but halt while waiting for active.thunderbirdsarego.
this reduces the system load and noise for timer callibration
and makes the mp startup linear with regard to the number of
cores.
|
|
|
|
|
