* rtems_servers/ftpd.c, rtems_servers/ftpd.h: Major enhancements
as listed below:
- use pool of pre-created threads to handle sessions instead of
creating/deleting threads on the fly
- LIST output is now similar to what "/bin/ls -al" would output,
thus FTP clients such Netscape are happy with it.
- LIST NAME now works (both for files and directories)
- added support for NLST, CDUP, and MDTM FTP commands to make
more FTP clients happy
- keep track of CWD for every session separately
- ability to specify root directory name for FTPD in configuration
table. FTPD will then create illusion for FTP clients that this
is actually root directory.
- ignore options sent in commands, thus LIST -al FILE works and
doesn't try to list "-al" directory.
- buffers are allocated on stack instead of heap where possible to
eliminate malloc/free calls (avoid possible heap fragmentation
troubles).
- drop using of task notepad to pass parameters - use function
arguments instead
- use snprintf() instead of sprintf() as the latter is unsafe
- use of PF_INET in socket() instead of AF_INET
Here are ftp clients I've tried new FTPD with (all of them
running on Debian GNU/Linux 2.2):
Lftp 2.1.10
NcFTP 2.4.3
Netscape 4.75
ftp
mc 4.5.49
* libc/linkaddr.c: Initialized variable to remove warning.
* modem/ppp.c, modem/ppp_tty.c: Made numerous variable declarations
conditional on PPP_COMPRESS and PPP_FILTER. Commented out variables
that were not used because the code using them was commented out.
Removed totally unused variables.
* modem/pppcompress.c: Added parentheses to avoid warnings.
* pppd/pppmain.c: Removed numerous warnings.
* wrapup/Makefile.am: Added modem subdir.
* configure.in, Makefile.am: Added modem subdir.
* net/Makefile.am: Added if_pppvar.h, pppcompress.h.
* pppd/Makefile.am: Added pppmain.c (which needs work).
* pppd/chat.c, pppd/fsm.c, pppd/fsm.h, pppd/ipxcp.c, pppd/main.c,
pppd/ppp_tty.c, pppd/upap.c: Changes from Thomas Doerfler
<Thomas.Doerfler@imd-systems.de> and cosmetic changes by me.
Actually main.c and ppp_tty.c should be scratched. The modem
subdir has the real ppp_tty.c and the real pppd main is in pppmain.c.
* wrapup/Makefile.am: Added modem subdir.
* configure.in, Makefile.am: Added modem subdir.
* net/Makefile.am: Added if_pppvar.h, pppcompress.h.
* pppd/Makefile.am: Added pppmain.c (which needs work).
* pppd/chat.c, pppd/fsm.c, pppd/fsm.h, pppd/ipxcp.c, pppd/main.c,
pppd/ppp_tty.c, pppd/upap.c: Changes from Thomas Doerfler
<Thomas.Doerfler@imd-systems.de> and cosmetic changes by me.
Actually main.c and ppp_tty.c should be scratched. The modem
subdir has the real ppp_tty.c and the real pppd main is in pppmain.c.
* POSIX include files merged into newlib. This resulted in
some definitions moving to other files and thus some secondary
effects in RTEMS source code.
* machine/types.h: Added _CLOCKID_T_ and _TIMER_T_ to be in sync
with newlib's <machine/types.h>.
* rtems/rtems_bsdnet_internal.h: newlib now includes definition
of struct itimerval in <sys/time.h>.
* lib/tftpDriver.c: add comments to handlers struct function pointers.
* rtems/rtems_glue.c: move pointer arithmetic to be _after_
pointer has been checked against NULL.
* machine/types.h, pppd/pppd.h, rtems/rtems_bsdnet_internal.h,
rtems_webserver/webmain.c: machine/types.h should not have
included rtems.h. It is now including precisely the
least amount of low level, yet portable .h files to get
the basic RTEMS types defined. This rippled into other
files since rtems_bsdnet_internal.h used machine/types.h to include
rtems.h.
* rtems/rtems_syscall.c: Changed from O_NONBLOCK to internal
RTEMS_LIBIO_FLAGS_NO_DELAY to avoid O_NONBLOCK/O_NDELAY confusion
and to work with the converted flags.
initialization. This adds an interface which makes it easier to
control the BSD stack from user code. The BSD stack initialise uses
it. It is a sort of `function' interface for an ifconfig
command.
I also added support for attaching and removing interfaces. With hot
swap PCI comming online support for hot swap PCI will be an important
factor in "state of art" RTOS's. This is also part of a general move on
my part to allow RTEMS to be configured at runtime by calls rather than
table driven at initialisation.
that contains the automake files for libnetworking plus a couple of
minor fixes. [Now only one unused/unsupported Makefile.in remains
(./c/src/lib/libbsp/hppa1.1/pxfl/Makefile.in).]
To apply:
patch -p1 < rtems-rc-20000118-7.diff
/bin/sh rtems-rc-20000118-7.rm
/bin/sh rtems-rc-20000118-7.add
./bootstrap
Notes:
* I have tested this one by building all BSPs for m68k, powerpc, sh and
unix with toolchains built since last weekend.
* I did not touch libnetworking's directory layout.
Janovetz <janovetz@tempest.ece.uiuc.edu> to return a status from
network initialization rather than panic'ing. It changes a bunch
of rtems_panics to printfs and returns a status from
rtems_bsdnet_initialize_network().
report by Nick.SIMON@syntegra.bt.co.uk:
TFTP uses UDP and UDP packets are prime targets for getting dropped when
the network gets busy. I want the number of retries quite large in my
application. I see that PACKET_REPLY_MILLISECONDS is, in fact, not
being used.
system table from when the rtems_filesystem_operations_table structure changed
and the initializer in libnetworking/lib/tftpDriver.c did not get
updated.
Tomasz only tested this on the mpc823.
The official site for the original source for this PPP implementation is:
ftp://cs.anu.edu.au/pub/software/ppp
NOTE: As of 11/30/1999, the current version of this source is 2.3.10.
EPICS needs a synchronized time-of-day clock. This patch is the changes
needed to get NTP server information from a BOOTP server.
This patch also adds NTP server information to the network configuration
structure, too.
<valette@crf.canon.fr> to add a port of the GoAhead web server
(httpd) to the RTEMS build tree. They have successfully used
this BSP on i386/pc386 and PowerPC/mcp750.
Mark and Joel spoke with Nick Berliner <nickb@goahead.com> on
26 Oct 1999 about this port and got verbal approval to include
it in RTEMS distributions.
report from Philip A. Prindeville <philipp@zembu.com>:
I was working on a device driver for a certain ethernet chipset that
occassionally wraps in its buffer, and causes a resulting mbuf chain
with only a few dozen bytes in the first mbuf of the chain.
I wouldn't have thought this would be a problem, until I ran some
stress tests that flooded the ethernet receiver with packets and
started to get panics here:
250
251 if (m->m_pkthdr.len < sizeof(struct ip))
252 goto tooshort;
253
254 #ifdef DIAGNOSTIC
255 if (m->m_len < sizeof(struct ip))
256 panic("ipintr mbuf too short");
257 #endif
258
259 if (m->m_len < sizeof (struct ip) &&
260 (m = m_pullup(m, sizeof (struct ip))) == 0) {
261 ipstat.ips_toosmall++;
262 return;
263 }
264 ip = mtod(m, struct ip *);
and the panic was at line 256. But if I #undef'd DIAGNOSTICS,
then the m_pullup() at line 260 does the right thing and the packet
ends up being processed just fine.
So I started wondering, (a) why was the test checking for
something that apparently wasn't a fatal condition but rather
one that is subsequently recovered from a couple of lines later
and (b) why panic as a diagnostic "aid" from a recoverable
condition rather than just (say) log a message to the console?
All of this seems overly severe for no reason that is readily
apparent to me.
now prototypes the malloc family in stdlib.h. This causes conflicts
with the way the network stack overrides the definitions of malloc.
As best I (being Joel) can tell, commenting stdlib.h out keeps the
files compiling and referencing the desired malloc/free but results
in more warnings.
