Commit Graph

1013 Commits

Author SHA1 Message Date
Ruslan Ermilov
2f42caa5de Unbreak this as well.
At the extra bonus of fixing the contents of the .depend file.

Not really my day.
2002-04-12 15:49:30 +00:00
Peter Wemm
31c41f1b1e Bandaid for a buffer overrun in the module searching code. When breaking
up the module_path string, we would walk one past the end of the buffer.
This hurting ia64 originally, but it was probably also happening on i386
occasionally as well.  The effects were usually harmless, it would add
bogus "binary" search directories to the places it actually looked for
files.
2002-04-11 10:00:44 +00:00
Peter Wemm
2ab188f529 Finally fix loader completely for IA64. efifs_stat() wasn't setting
the S_IFREG bit for regular files.  This caused the path search code to
skip it when it finally did find the kernel (after the common/module.c
buffer overrun bug was fixed)
2002-04-11 09:50:11 +00:00
Pierre Beyssac
2b9592079f Code cleanup, no functional change.
Patch adapted from PR.

PR:		i386/36015
Submitted by:	thomas@cuivre.fr.eu.org
MFC after:	1 week
2002-04-11 09:21:10 +00:00
Daniel C. Sobral
231b8f199f Upgrade FICL to 3.02. Forgot this one, sorry.
PR:		36308
Submitted by:	dcs
2002-04-09 20:59:34 +00:00
Daniel C. Sobral
be88b71603 Upgrade to FICL version 3.02. Anything wrong is my fault, everything right is
due Jon Mini.

PR:		36308
Submitted by:	Jon Mini <mini@haikugeek.com>
MFC after:	4 weeks
2002-04-09 17:45:28 +00:00
Jake Burkholder
bb75d97e13 Fix another unsigned long used to index the symbol table which should be
Elf_Hashelt.
2002-04-09 00:05:46 +00:00
Peter Wemm
c21ab46789 We must not let install(1) strip loader.efi when installing it, or the
resulting binary will be damaged and no longer work.
2002-04-07 04:27:50 +00:00
Peter Wemm
1e7e1ea54d Set BINDIR 2002-04-07 04:26:42 +00:00
Peter Wemm
f3693129ce Add loader bootforth infrastructure and install it. 2002-04-07 04:25:45 +00:00
Peter Wemm
01c0b884af Make it a bit closer to the EFI build and tie up some loose ends. 2002-04-06 04:33:53 +00:00
Peter Wemm
e91ab65ef6 Try and tidy up some very loose ends with paths to various libraries etc. 2002-04-06 04:29:36 +00:00
Peter Wemm
402e50591a Do not add the ficl/alpha subdir to the ia64 include path. Try ficl/ia64
instead.
2002-04-06 04:16:58 +00:00
Peter Wemm
5459842c5e Do not assume that ${OBJCOPY} variable exists. It was hidden by the
fact that the 'ia64-make' wrapper explicitly set it.
2002-04-06 04:11:46 +00:00
Peter Wemm
93b4f334b6 Use the correct elf hash table entry type. This matches a similar fix
in the kernel side of things some time ago.  The hash table entries are
always 32 bits wide, even on 64 bit machines.
2002-04-06 04:09:42 +00:00
Peter Wemm
a6d81d83a2 Fix self hosted loader building. When you use a self configured gcc, it
detects and uses the gas section merge support.  As a result, a whole bunch
of new sections arrive, including .rodata.str1.8, which was not included
in our custom ldscript.ia64.  The result was a loader binary that EFI
rejected.

While here, collect the loader shell commands linker set and include it
in the data area rather than having its own section.

/boot/loader.efi was the last holdout for having a 100% self built ia64
system.
2002-04-06 03:39:22 +00:00
Peter Wemm
65a0653eb7 Use a relative path to libstand.. /usr/src/lib/libstand may not exist
or may have the wrong header files.
2002-04-04 06:04:38 +00:00
Thomas Moestl
bc3c153560 Do not use semi-fixed TLB slots for the 4M kernel pages any more. 2002-04-02 17:10:15 +00:00
Thomas Moestl
2c94cd1ba0 1.) Rename locore.s to locore.S (by repocopy), to be able to remove
special-case make rule
2.) Cleanups, remove superfluous expicit rules, add -nostdlib to LDFLAGS,
    remove -X and -g, remove -g from CFLAGS
3.) Add BINDIR
4.) Build install the loader help file, add an empty help.sparc64
5.) Change the default configuration to only support booting from disk
6.) Get libofw.a from a path relative ${.OBJDIR}, not ${.CURDIR}

Submitted by:	jake (1 - 5), obrien (6)
2002-04-02 17:08:37 +00:00
Thomas Moestl
ad68ab89e2 Add support for booting from CD-ROM. Make it possible to enable UFS
support using make arguments.
2002-04-01 23:28:35 +00:00
Thomas Moestl
163f47b1ec 1.) Produce a boot1 disklabel template of the format disklabel(8) expects.
2.) Clean up and change over to using bsd.prog.mk

Submitted by:	jake (2)
2002-04-01 22:57:51 +00:00
Marcel Moolenaar
16ed9fd6bb Only install the help file if we can find it. Use ${BASE}.help
in both the condition and for the install. We expect to find
the help file in ${.OBJDIR}.
2002-03-31 20:48:13 +00:00
Marcel Moolenaar
cd2c8e3ef6 Pass the address of the bootinfo block to the kernel in register
r8. Keep it at the hardwired address for now. Bump the version.
2002-03-30 23:52:34 +00:00
Marcel Moolenaar
c459265de7 Pass the physical address of the bootinfo block to the kernel in
register r8. We continue to write the bootinfo block at the same
hardwired address, because the kernel still expects it there.
It is expected that future kernels use register r8 to get to the
bootinfo block and don't depend on the hardwired address anymore.

Bump the loader version once again due to the interface change.
2002-03-30 23:00:05 +00:00
Yoshihiro Takahashi
bd30f72967 MFi386: revision 1.55 2002-03-30 11:18:30 +00:00
Marcel Moolenaar
a918e110c4 Add a quick and dirty way to determine where we're loaded from. We
only care if it's network or not at this time. If we're loaded from
the network, we set currdev (=loaddev) so that the kernel is loaded
from the network as well. In all other cases we initialize to disk.
This makes netbooting more convenient and can easily be enhanced to
do more elaborate checking.
2002-03-30 07:32:08 +00:00
Marcel Moolenaar
4e357dbc58 The EFI loader has been improved a lot since it was first added.
Most significantly (from an interfacing point of view) is the
support for the FPSWA pointer passing. Even though that was added
4 months ago, it's probably not a bad idea to bump the version
number to reflect this.
2002-03-30 04:54:54 +00:00
Marcel Moolenaar
7608f7cb80 Fix the initialization of the protocol:
o  Query the state field of the protocol mode to determine whether
   we need to start and/or initialize the protocol. When we're
   loaded across the network, the protocol has already been started
   and is already initialized. When no networking has happened yet,
   we have to start and initialize the protocol ourselves.
o  After initialization, we have to set the receive filters. Not
   doing this results in a deaf interface. We set the unicast and
   broadcast filters. Multicast may not be supported. This specific
   change fixes the problem we had that we could not netboot if
   the loader was started from the EFI shell.
o  To help future debugging, add a function that dumps the current
   mode of the interface. It's conditional on EFINET_DEBUG.
o  To help in runtime problems, emit a diagnostic message when we
   could not initialize the protocol properly.
2002-03-30 04:50:52 +00:00
Marcel Moolenaar
c61a2c84f0 Don't blindly dereference f->f_devdata as if it's always a pointer to
an efi_devdesc structure. When we're netbooting, f->f_devdata holds
the address of the network socket variable. Dereferencing this caused
some very unpredictable behaviour, including proper functioning.
So, as a sanity check, we first make sure f->f_dev points to our
own devsw. If not, the open will fail before we use f->f_devdata.

This solves the netboot hangs I invariably got whenever I used the
latest toolchain to compile the EFI loader.
2002-03-30 01:36:03 +00:00
Marcel Moolenaar
9423456018 o Make efinet_put a blocking call by waiting for the protocol
layer to signal transmission of the packet. This resolves the
   problem I'm seeing that an immediate call to net->Receive
   after calling net->Transmit returns EFI_DEVICE_ERROR. This
   condition seems to be sufficiently persistent that BOOTP and
   RARP fail.
o  While here, unify all functions to have 'nif' defined. Some
   have it as arguments. The others now have them as locals. We
   now always get the protocol interface by using the 'nif' var.

The current status of netbooting is that even though we now reliably
have BOOTP working (again), opening a file (ie loading a kernel)
across the network causes the loader to hang. I'm working on that now.
2002-03-29 23:10:15 +00:00
Marcel Moolenaar
ede9f03a1a Fix the beforeinstall target. We install ${PROG}.help if loader.help
exists, otherwise we install it anyway. I interpret this as a very
high desire to install ${PROG}.help. Alas, ${PROG}.help doesn't exist
at the moment and neither does loader.help, so in practice this just
doesn't work, no matter how you interpret it. The compromise is to
install ${PROG}.help IFF it exists. I realize we lost creativity with
this commit, but style should have been preserved, AFAICT :-)
2002-03-29 22:53:56 +00:00
David E. O'Brien
10d23bba69 Needs a.out support built into the loader. 2002-03-28 19:09:44 +00:00
Marcel Moolenaar
127d4e90b0 o Don't include sys/cdefs.h 2002-03-28 07:07:45 +00:00
Marcel Moolenaar
95c0f344ee o Add -L${DESTDIR}${LIBDIR} on the link line for -lstand.
o  Add -j .dynstr to objcopy. This makes .efi binaries work
   when built with a 3.x based toolchain.
2002-03-28 06:58:46 +00:00
Marcel Moolenaar
b0d9287123 Duplicate the logic used elsewhere to define LIBSTAND. 2002-03-28 06:52:10 +00:00
David E. O'Brien
77dfe39221 was repocopied to ../boot1 2002-03-28 02:56:05 +00:00
David E. O'Brien
20d9715213 Apparently either gcc or ld, in their infinite wisdom, want to
put a bunch of crap before the code in .text.  Since the firmware
doesn't seem to honour the a.out entry point, we need to include
a little assmbler file which jumps to where we want to be in C.

Submitted by:	jake
2002-03-28 02:41:52 +00:00
David E. O'Brien
0ce1fc3fc3 Opps, bootblock component is no more. 2002-03-28 01:41:23 +00:00
David E. O'Brien
8e5b3e7b5b sparc64 is an ofw consumer. 2002-03-28 01:37:10 +00:00
David E. O'Brien
ef313bf935 Not all platforms have and want a.out format support. 2002-03-28 01:28:21 +00:00
David E. O'Brien
8fa3779837 Add a Makefile for sparc64 at this level. 2002-03-28 01:01:43 +00:00
Pierre Beyssac
038148d678 Add option -n to i386 boot2 to disallow boot interruption by keypress.
PR:		i386/36016
Submitted by:	Thomas Quinot <thomas@cuivre.fr.eu.org>
Reviewed by:	rnordier
MFC after:	1 week
2002-03-23 19:40:27 +00:00
Alfred Perlstein
03d1b8bc94 Remove __P. 2002-03-20 08:00:54 +00:00
Peter Wemm
efcbdfb72b Add EFI write support to loader 2002-03-19 23:05:33 +00:00
Peter Wemm
1d7914a5bd Add -ffreestanding to avoid printf/puts/putchar conversions 2002-03-19 10:51:57 +00:00
Peter Wemm
436122c232 Boot from efifs first. 2002-03-19 10:50:41 +00:00
Peter Wemm
9b6a75edb3 gcc-3.1 likes to have extra { } around the internal array initializers in
the GUID templates.
2002-03-19 10:50:09 +00:00
Maxim Sobolev
fb74e5f595 Add splitfs vfs layer into libstand, which allows loading big kernels and
modules split across several physical medias. Following is how it works:

The splitfs code, when asked to open "foo" looks for a file "foo.split"
which is a text file containing a list of filenames and media names, e.g.

	foo.aa "Kernel floppy 1"
	foo.ab "Kernel floppy 2"
	foo.ac "Kernel and modules floppy"

For each file segment, the process is:

- try to open the file
- prompt "Insert the disk labelled <whatever> and press any key..."
- try to open the file
- return error if file could not be located

RE team is free to use this feature in the upcoming 5.0-DP1.

Reviewed by:	msmith, dcs
2002-03-17 12:18:05 +00:00
Ruslan Ermilov
3a3c44ac20 mdoc(7) police: fix a typo and markup. 2002-03-15 15:12:10 +00:00
Ruslan Ermilov
9662d32c73 Implement -m and -p loader(8) "boot" command options in boot2.
(This is more useful for 4.x where boot blocks can still load
kernels, modulo the PR kern/17422.)
2002-03-13 11:03:36 +00:00