david/ipxe
Archived
1
0
Commit Graph

199 Commits

Author SHA1 Message Date
Michael Brown
476d02051f Windows Server 2003 sulks if we feed it an empty region in base memory
as a result of our memory map editing, so strip out any empty regions.
2006-06-09 12:20:03 +00:00
Michael Brown
6fd1bd0510 Added missing check-in 2006-06-09 10:19:17 +00:00
Michael Brown
17eea9f933 Fix bug with >256 cylinders.
Allow our functions to return a non-zero, non-error status (since the
INT 13 Extensions Check has to return the API version in the register
that is otherwise always used for the error code).

Report a non-zero API version from the INT 13 Extensions Check; GRUB
now uses extended reads.
2006-06-08 19:28:08 +00:00
Michael Brown
fdf62528ca It's astonishing how long really fundamental bugs can survive without
being noticed...
2006-06-08 15:06:09 +00:00
Michael Brown
191a906338 Fix some remarkably obvious mistakes in pcidirect.h 2006-06-08 14:57:04 +00:00
Michael Brown
763b7e2ef9 We now have just romprefix, rather than having isaprefix and pciprefix 2006-06-06 15:41:21 +00:00
Michael Brown
ef37f78cbe Cut out almost all the optional code paths, drastically simplifying the
flow of control through this code.

We now always add PCI and PnP headers even for ISA ROMs; there's no harm
in doing so.

UNDI loader is still missing.
2006-06-06 15:33:39 +00:00
Michael Brown
e7eca7b7cb Code segment may not be writable; create the temporary pointer to the GDT
on the stack.
2006-06-06 15:30:28 +00:00
Michael Brown
46253ba618 Calculate _rom_size for the ROM prefix 2006-06-06 15:29:39 +00:00
Michael Brown
35b04b1514 Tidy up comment 2006-05-28 14:40:04 +00:00
Michael Brown
041a04335e Fixed assembly on old versions of gas 2006-05-28 14:21:29 +00:00
Michael Brown
9f86754f90 Updated to work with libprefix. 2006-05-25 22:04:47 +00:00
Michael Brown
29340d7728 Documented the fact that the prefix can prot_call(main) without
worrying about its stack being destroyed during the Etherboot run.
2006-05-25 22:04:17 +00:00
Michael Brown
d82f71efd0 Also hide base memory region from E820 map. 2006-05-25 00:10:58 +00:00
Michael Brown
89da833c5d init_librm() and prot_call() are now real-mode far calls.
install() now calls relocate(), moves the protected-mode code to the new
location, and calls hide_etherboot().
2006-05-25 00:06:45 +00:00
Michael Brown
4d81b48504 Has been obsolete for a while. 2006-05-25 00:04:37 +00:00
Michael Brown
1966945a5d Update relocate() to work with get_memmap().
Change semantics; relocate() now just finds a suitable location; it
doesn't actually perform the relocation itself.  Code in libprefix does
the copy in flat real mode.
2006-05-25 00:04:13 +00:00
Michael Brown
f939ff7cb5 Fix printf format warning error 2006-05-25 00:02:27 +00:00
Michael Brown
3bd3c13c7b Why should pushal/popal put the registers on the stack in a logical
order?
2006-05-25 00:01:46 +00:00
Michael Brown
dd75965057 Typo. 2006-05-25 00:00:53 +00:00
Michael Brown
e4f8d6eed7 Add a "count" field to struct memory_map. 2006-05-25 00:00:30 +00:00
Michael Brown
986f6ffff1 Define a struct memory_map with a fixed number of entries, rather than
requiring each caller to decide how many entries it wants to permit.
2006-05-24 14:41:27 +00:00
Michael Brown
edcf89e2b0 Remove the automatic segment register save and restore; most users of
REAL_CODE() don't need it.
2006-05-24 14:11:36 +00:00
Michael Brown
02cf755f3f Removed REAL_EXEC(); there is no longer any code using it. 2006-05-24 14:10:15 +00:00
Michael Brown
103aff3ae2 Added potentially required "\n\t" on the end of asm_code_str 2006-05-24 14:06:28 +00:00
Michael Brown
190276a8b6 NBI is due for removal 2006-05-24 13:53:35 +00:00
Michael Brown
52125e158c Will be replaced with a block-device implementation. 2006-05-24 13:52:46 +00:00
Michael Brown
8110f9b082 Updated to REAL_CODE() 2006-05-24 13:50:40 +00:00
Michael Brown
595f161b51 Updated to REAL_CODE().
int13_boot() now does its own segment register save and restore, ready
for the removal of segment register restoration from the libkir
generic code.
2006-05-24 13:48:16 +00:00
Michael Brown
e8550035af Updated to REAL_CODE() 2006-05-24 13:28:39 +00:00
Michael Brown
c4d688c005 Updated to REAL_CODE(). 2006-05-24 13:24:22 +00:00
Michael Brown
04d6005d77 Update to REAL_CODE() 2006-05-24 12:45:50 +00:00
Michael Brown
6631f46999 Updated to REAL_CODE(). 2006-05-24 12:32:57 +00:00
Michael Brown
53935e9dc6 Set carry flag before calling INT 15,e801 and INT 15,e820.
Ignore carry flag for INT 15,88; the Ralf Brown interrupt list says
that CF is unreliable for this call.
2006-05-24 12:19:16 +00:00
Michael Brown
7a53d07027 Split out REAL_CODE() from REAL_EXEC(), preparatory to removing REAL_EXEC
completely.
2006-05-24 09:51:04 +00:00
Michael Brown
83d80d9e79 Remove references to obsoleted REAL_CALL from documentation. 2006-05-24 09:16:48 +00:00
Michael Brown
3b525385af Removed REAL_FRAGMENT and REAL_CALL, and left just REAL_EXEC (which is
the only one we actually use).  This allows REAL_EXEC fragments to
contain proper references to constraints (e.g. "%w0"), rather than having
to force the use of specific registers.

Note that the "num_constraints" parameter is now completely obsolete, and
that we can probably reduce the syntax to something like
  __asm__ __volatile__ ( REAL_CODE ( "asm statements" )
                         : output constraints
                         : input constraints
                         : clobber );
which would look much more natural, and avoid the need to always specify
a clobber list.


Add userptr_t to libkir.h, to allow it to at least compile.
2006-05-24 09:14:44 +00:00
Michael Brown
77a65075a6 Replaced memsizes.c with smaller memmap.c, taking advantage of __data16,
and creating a memory map that's easier to work with than the E820 map.
2006-05-24 01:37:46 +00:00
Michael Brown
5e80847827 Code to install the new E820 mangler (which doesn't require copying
down to base memory, since it is now in the base-memory-resident
.text16 section).
2006-05-23 23:47:05 +00:00
Michael Brown
897313007b Add __{text,data}16_array, since there's no way I can see to make the
usual __{text,data}16 trick work with arrays.  gcc seems to accept the
__asm__ ( asmlabel ) only after the [] of the array declaration, not
before.
2006-05-23 23:42:45 +00:00
Michael Brown
5ec2b2c251 Cope with regions bigger than 4GB.
We now split e820 regions around ourselves, rather than just
truncating the e820 region.  This avoids the worst-case scenario of
losing all memory over 4GB.

It's more important to get the memory map right now that we're
expecting to still be loaded when the OS starts in several situations
(e.g. Linux with UNDI driver, any OS with iSCSI/AoE boot, etc.).
2006-05-23 23:33:37 +00:00
Michael Brown
ca2519cea3 Refactored to reduce code size; now down from 1304 to 1105 bytes.
Tidied up debug messages; the log now contains one line per INT 13
operation, looking like

 INT 13,08 (80): Get drive parameters
 INT 13,02 (80): Read: C/H/S 0/47/14 = LBA 0xb9e <-> 1084:0000 (count 106)
2006-05-20 15:33:32 +00:00
Michael Brown
d48d0fb1bb Add the concept of a "user pointer" (similar to the void __user * in
the kernel), which encapsulates the information needed to refer to an
external buffer.  Under normal operation, this can just be a void *
equivalent, but under -DKEEP_IT_REAL it would be a segoff_t equivalent.

Use this concept to avoid the need for bounce buffers in int13.c,
which reduces memory usage and opens up the possibility of using
multi-sector reads.

Extend the block-device API and the SCSI block device implementation
to support multi-sector reads.

Update iscsi.c to use user buffers.

Move the obsolete portions of realmode.h to old_realmode.h.

MS-DOS now boots an order of magnitude faster over iSCSI (~10 seconds
from power-up to C:> prompt in bochs).
2006-05-19 15:06:51 +00:00
Michael Brown
0ab92faedb Use typeof(sizeof(...)) to define a size_t. This stops gcc complaining
about format errors for %zX arguments.
2006-05-19 14:51:59 +00:00
Michael Brown
6b6096d28b Strip down i386 PCI configuration space I/O to the bare minimum. A
typical build will now include 880 bytes of PCI support code, compared to
2327 bytes in Etherboot 5.4.

(There is a slight cost of around 5 extra bytes per access to a
non-constant config space address; this should be an overall win.
Driver-specific accesses will usually be to constant addresses, for
which there is no additional cost.)
2006-05-17 01:12:11 +00:00
Michael Brown
15ee09ed10 Restructured PCI subsystem to fit the new device model.
Generic PCI code now handles 64-bit BARs correctly when setting
"membase"; drivers should need to call pci_bar_start() only if they want
to use BARs other than the first memory or I/O BAR.

Split rarely-used PCI functions out into pciextra.c.

Core PCI code is now 662 bytes (down from 1308 bytes in Etherboot 5.4).
284 bytes of this saving comes from the pci/pciextra split.

Cosmetic changes to lots of drivers (e.g. vendor_id->vendor in order to
match the names used in Linux).
2006-05-16 15:12:06 +00:00
Michael Brown
fcdab6299c Missed a reference to heap.h. 2006-05-16 15:01:26 +00:00
Michael Brown
295e11b2b0 Add INT 13,41 (extensions installation check). LILO's MBR now uses
linear calls to load the MS-DOS boot sector in my test setup.
2006-05-14 23:46:28 +00:00
Michael Brown
29fb77b0a4 Fix a couple of minor typos. 2006-05-14 18:17:29 +00:00
Michael Brown
6e42cb95bd Also capture INT 19 as a failure path, to allow for boot sectors that
don't conform to BBS (e.g. the one created by mkdosfs).
2006-05-14 14:13:55 +00:00