Some BIOSes (observed with an AMI BIOS on a SunFire X2200) seem to
reset the BIOS drive counter at 40:75 after a failed boot attempt.
This causes problems when attempting a Windows direct-to-iSCSI
installation: bootmgr.exe calls INT 13,0800 and gets told that there
are no hard disks, so never bothers to read the MBR in order to obtain
the boot disk signature.  The Windows iSCSI initiator will detect the
iBFT and connect to the target, and everything will appear to work
except for the error message "This computer's hardware may not support
booting to this disk.  Ensure that the disk's controller is enabled in
the computer's BIOS menu."

Fix by checking the BIOS drive counter on every INT 13 call, and
updating it whenever necessary.

INT 13,15 should return the number of sectors, not the number of
cylinders.

The case of an unsupported SAN protocol will currently not result in
any error message.  Fix by printing the error message at the top level
using strerror(), rather than using hard-coded error messages in the
error paths.

This disgustingly ugly hack just keeps getting worse.

IPoIB has a 20-byte link-layer address, of which only eight bytes
represent anything relating to a "hardware address".

The PXE and EFI SNP APIs expect the permanent address to be the same
size as the link-layer address, so fill in the "permanent address"
field with the initial link layer address (as generated by
register_netdev() based upon the real hardware address).

The hardware address is an intrinsic property of the hardware, while
the link-layer address can be changed at runtime.  This separation is
exposed via APIs such as PXE and EFI, but is currently elided by gPXE.

Expose the hardware and link-layer addresses as separate properties
within a net device.  Drivers should now fill in hw_addr, which will
be used to initialise ll_addr at the time of calling
register_netdev().

The SRP Boot Firmware Table serves a similar role to the iSCSI and AoE
Boot Firmware Tables; it provides information required by the loaded
OS in order to establish a connection back to the SRP boot device.

SRP is the SCSI RDMA Protocol.  It allows for a method of SAN booting
whereby the target is responsible for reading and writing data using
Remote DMA directly to the initiator's memory.  The software initiator
merely sends and receives SCSI commands; it never has to touch the
actual data.

Tested-by: Joshua Oreman <oremanj@xenon.get-linux.org>

IPoIB has a link-layer broadcast address that varies according to the
partition key.  We currently go through several contortions to pretend
that the link-layer address is a fixed constant; by making the
broadcast address a property of the network device rather than the
link-layer protocol it will be possible to simplify IPoIB's broadcast
handling.

PXENV_STOP_UNDI should return PXENV_STATUS_KEEP_UNDI if the UNDI
cannot be safely unloaded (e.g. due to interrupt vectors that could
not be unhooked).

Merge the pxe_set_netdev()+pxe_[un]hook_int1a() pattern into a single
pxe_[de]activate() call.

pxe_init_structures() fills in the fields of the !PXE and PXENV+
structures that aren't known until gPXE starts up.  Once gPXE is
started, these values will never change.

Make pxe_init_structures() an initialisation function so that PXE
users don't have to worry about calling it.

It is possible that the UNDI ISR may be triggered before netdev_tx()
returns control to pxenv_undi_transmit().  This means that
pxenv_undi_isr() may see a zero undi_tx_count, and so not check for TX
completions.  This is not a significant problem, since it will check
for TX completions on the next call to pxenv_undi_isr() anyway; it
just means that the NBP will see a spurious IRQ that was apparently
caused by nothing.

Fix by updating the undi_tx_count before calling netdev_tx(), so that
pxenv_undi_isr() can decrement it and report the TX completion.

Symantec Ghost requires working multicast support.  gPXE configures
all (sufficiently supported) network adapters into "receive all
multicasts" mode, which means that PXENV_UNDI_SET_MCAST_ADDRESS is
actually a no-op, but the current implementation returns
PXENV_STATUS_UNSUPPORTED instead.

Fix by making PXENV_UNDI_SET_MCAST_ADDRESS return success.  For good
measure, also implement PXENV_UNDI_GET_MCAST_ADDRESS, since the
relevant functionality is now exposed by the net device core.

Note that this will silently fail if the gPXE driver for the NIC being
used fails to configure the NIC in "receive all multicasts" mode.

The PXE debugging messages have remained pretty much unaltered since
Etherboot 5.4, and are now difficult to read in comparison to most of
the rest of gPXE.

Bring the pxe_undi debug messages up to normal gPXE standards.

Signed-off-by: Michael Brown <mcb30@etherboot.org>

The Symantec UNDI DOS driver fails when run on top of gPXE because we
return our interface type as "gPXE" rather than one of the predefined
NDIS interface type strings.

Fix by returning the standard "DIX+802.3" string; this isn't
necessarily always accurate, but it's highly unlikely that anything
trying to use the UNDI API would understand our IPoIB link-layer
pseudo-header anyway.

The Intel DOS UNDI driver fails when run on top of gPXE because we do
not fill in the ServiceFlags field in PXENV_UNDI_GET_IFACE_INFO.

Fix by filling in the ServiceFlags field with reasonable values
indicating our approximate feature capabilities.

The 3Com DOS UNDI driver fails when run on top of gPXE for two
reasons: firstly because PXENV_UNDI_SET_PACKET_FILTER is unsupported,
and secondly because gPXE enters the NBP without enabling interrupts
on the NIC, and the 3Com driver never calls PXENV_UNDI_OPEN.

Fix by always returning success from PXENV_UNDI_SET_PACKET_FILTER
(which is no worse than the current situation, since we already ignore
the receive packet filter in PXENV_UNDI_OPEN), and by forcibly
enabling interrupts on the NIC within PXENV_UNDI_TRANSMIT.  The latter
is something of a hack, but avoids the need to implement a complete
base-code ISR that we would otherwise need if we were to enter the NBP
with interrupts enabled.

In order to construct outgoing link-layer frames or parse incoming
ones properly, some protocols (such as 802.11) need more state than is
available in the existing variables passed to the link-layer protocol
handlers. To remedy this, add struct net_device *netdev as the first
argument to each of these functions, so that more information can be
fetched from the link layer-private part of the network device.

Updated all three call sites (netdevice.c, efi_snp.c, pxe_undi.c) and
both implementations (ethernet.c, ipoib.c) of ll_protocol to use the
new argument.

Signed-off-by: Michael Brown <mcb30@etherboot.org>

Signed-off-by: Michael Brown <mcb30@etherboot.org>

Add FILE_LICENCE declarations to almost all files that make up the
various standard builds of gPXE.

Using "lret $2" to return from an interrupt causes interrupts to be
disabled in the calling program, since the INT instruction will have
disabled interrupts.  Instead, patch CF on the stack and use iret to
return.

Interestingly, the original PC BIOS had this bug in at least one
place.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Michael Brown <mcb30@etherboot.org>

When handling a redirection event, we need to close the existing
connection before opening the new connection.

This is not strictly an entry requirement, but it does mean that a PXE
NBP returning without setting %ax will appear to have returned
success.

Avoid passing credentials in the iBFT that were available but not
required for login.  This works around a problem in the Microsoft
iSCSI initiator, which will refuse to initiate sessions if the CHAP
password is fewer than 12 characters, even if the target ends up not
asking for CHAP authentication.

The PXE 1.x spec specifies that on NBP entry or on return from INT
1Ah AX=5650h, EDX shall point to the physical address of the PXENV+
structure.  The PXE 2.x spec drops this requirement, simply stating
that EDX is clobbered.  Given the principle "be conservative in what
you send, liberal in what you accept", however, we should implement
this anyway.

COMBOOT images use INTs to issue API calls; these end up making calls
into gPXE from real mode, and so temporarily change the real-mode
stack pointer.  When our COMBOOT code uses a longjmp() to implement
the various "exit COMBOOT image" API calls, this leaves the real-mode
stack pointer stuck with its temporary value, which causes problems if
we eventually try to exit out of gPXE back to the BIOS.

Fix by adding rmsetjmp() and rmlongjmp() calls (analogous to
sigsetjmp()/siglongjmp()); these save and restore the additional state
needed for real-mode calls to function correctly.

When chaining COMBOOT images, the old images now get freed correctly.

Multi-level menus via COMBOOT rely on the COMBOOT program being able
to exit and invoke a new COMBOOT program (the next menu).  This works,
but rapidly (within about five iterations) runs out of space in gPXE's
internal stack, since each new image is executed in a new function
context.

Fix by allowing tail recursion between images; an image can now
specify a replacement image for itself, and image_exec() will perform
the necessary tail recursion.

Try to avoid future problems caused by implicit section flags and/or
type information by instituting a policy that all .section
declarations must explicitly state the flags and type.

Most of this change was achieved using

    perl -pi \
      -e 's/".text"$/".text", "ax", \@progbits/ ; ' \
      -e 's/".text16"$/".text16", "ax", \@progbits/ ; ' \
      -e 's/".text16.null"$/".text16.null", "ax", \@progbits/ ; ' \
      -e 's/".text16.data"$/".text16.data", "aw", \@progbits/ ; ' \
      -e 's/".data"$/".data", "aw", \@progbits/ ; ' \
      -e 's/".data16"$/".data16", "aw", \@progbits/ ; ' \
      -e 's/".bss"$/".bss", "aw", \@nobits/ ; ' \
      -e 's/".bss16"$/".bss16", "aw", \@nobits/ ; ' \
      -e 's/".prefix"$/".prefix", "ax", \@progbits/ ; ' \
      -e 's/".prefix.lib"$/".prefix.lib", "awx", \@progbits/ ; ' \
      -e 's/".prefix.data"$/".prefix.data", "aw", \@progbits/ ; ' \
      -e 's/".weak"$/".weak", "a", \@nobits/ ; ' \
      `git grep -l '\.section'`

If it happens that _textdata_memsz ends up being an exact multiple of
4kB, then this will cause the .textdata section (after relocation) to
start on a page boundary.  This means that the hidden memory region
(which is rounded down to the nearest page boundary) will start
exactly at virtual address 0, i.e. UNULL.  This means that
init_eheap() will erroneously assume that it has failed to allocate a
an external heap, since it typically ends up choosing the area that
lies immediately below .textdata, which in this case will be the
region with top==UNULL.

A subsequent error is that memtop_urealloc() passes through the error
return status -ENOMEM to the caller, which (rightly) assumes that the
result represents a valid userptr_t address.

Fixed by using alternative tests for heap non-existence, and by
returning UNULL in case of an error from init_eheap().

The DHCP client code now implements only the mechanism of the DHCP and
PXE Boot Server protocols.  Boot Server Discovery can be initiated
manually using the "pxebs" command.  The menuing code is separated out
into a user-level function on a par with boot_root_path(), and is
entered in preference to a normal filename boot if the DHCP vendor
class is "PXEClient" and the PXE boot menu option exists.

pxe_tftp.c assumes that the first seek on its data-transfer interface
represents the block size.  Apart from being an ugly hack, this will
also screw up file size calculation for files smaller than one block.

The proper solution would be to extend the data-transfer interface to
support the reporting of stat()-like data.  This is not going to
happen until the cost of adding interface methods is reduced (a fix I
have planned since June 2008).

In the meantime, abuse the xfer_window() method to return the block
size, since it is not being used for anything else and is vaguely
justifiable.

Astonishingly, having returned the incorrect TFTP blocksize via
PXENV_TFTP_OPEN for almost a year seems not to have affected any of
the test cases run during that time; this bug was found only when
someone tried running the heavily-patched version of pxegrub found in
OpenSolaris.

Currently the only supported platform for x86_64 is EFI.

Building an EFI64 gPXE requires a version of gcc that supports
__attribute__((ms_abi)).  This currently means a development build of
gcc; the feature should be present when gcc 4.4 is released.

In the meantime; you can grab a suitable gcc tree from

  git://git.etherboot.org/scm/people/mcb30/gcc/.git

EFI provides a copy of the SMBIOS table accessible via the EFI system
table, which we should use instead of manually scanning through the
F000:0000 segment.

Remove the assortment of miscellaneous hacks to guess the "network
boot device", and replace them each with a call to last_opened_netdev().

It still isn't guaranteed correct, but it won't be any worse than
before, and it will at least be consistent.

Signed-off-by: Laurent Vivier <Laurent.Vivier@bull.net>

This brings us in to line with Linux definitions, and also simplifies
adding x86_64 support since both platforms have 2-byte shorts, 4-byte
ints and 8-byte long longs.

__cdecl is a misleading name, since it currently encapsulates both
cdecl and regparm(0) attributes.  Rename to __asmcall.

The initial PXE implementation in Etherboot had the goal of being
architecture-agnostic, but this goal has not been realised.

Code paths that automatically allocate memory from the FBMS at 40:13
should also free it, if possible.

Freeing this memory will not be possible if either

  1. The FBMS has been modified since our allocation, or

  2. We have not been able to unhook one or more BIOS interrupt vectors.

We have EFI APIs for CPU I/O, PCI I/O, timers, console I/O, user
access and user memory allocation.

EFI executables are created using the vanilla GNU toolchain, with the
EXE header handcrafted in assembly and relocations generated by a
custom efilink utility.