DMA functions defined. Most of the changes are based on the "bochs sync"
version of plex86. Here is the list of changes:
* register/unregister functions for DMA channels added. The DMA controller
can use the DMA read/write handlers of registered devices directly.
* "hardwired" code in dma.cc removed
* all DMA related code in devices.cc and iodev.h removed
* DMA related code in pc_system.* removed except HRQ handling
* macros for DMA functions defined in bochs.h
* floppy and SB16 code modified to use the changes described above
The new function register_pci_handlers() is similar to the register functions
for i/o addresses. A PCI device can register the read/write handlers for it's
private PCI configuration space. The i/o mapped registers of the host bridge
control the access to the configuration registers of each PCI device. You can
select the bus, device, function and register address with the confAddr
register. The confData register is a window to the configuration space of the
selected device.
- reset sets the values of the i/o mapped registers to 0
- changed some BX_INFO messages to BX_DEBUG
This means that the REALTIME_PIT now only works on machines
that are capable of sustaining 150000 IPS, in aggregate. This
seems a reasonable requirement, as even my 200Mhz PPC with 16MB
of memory can sustain that.
This is now probably ready for primetime, so I need to get it
added to the configure options.
This seems to work, but I don't think it works on slow machines (like mine.)
At least, I'm getting fails with the dlxlinux image, but they're pretty
sporadic. This code shouldn't cause any fails.
floppy_bootsig_check: disabled=[0|1]. It is not disabled by default.
Sorry for any inconvenience it could cause
- changed the default (if cmos reg is 0) behavior for the floppy boot signature
check, so the new bios is compatible with older bochs binaries (useful if we
want to use the new bios when we release 1.4.1) :
1 : disabled
0 : enabled
- added info at launch time about the boot device and the floppy boot signature
check
- added a "LBA capable" flag in the ata controller
- fixed opened netdevice name (/dev/net/tun...)
- fixed test on iface name (tun instead of tap)
- cosmetic changes
- added example in .bochsrc
Much of these fixes have been contributed by Frederic Bothamy
- The emulated vga card issues a dimension_update() using the real
screen dimensions and the real font height.
- The gui selects a font with the requested height or it recalculates
the screen height if the font height is not available.
- The text_update() function uses a new variable 'ncols' instead of the
fixed value of 80 for the number of text columns.
I have tested the changes with X11/Linux, SDL, wxGTK and win32. The changes
in the other guis are not tested yet.
cursor size issue a screen update request
- reading an invalid CRTC register returns 0 / writing is ignored
- bit 7 of CRTC address register is always 0
- debug messages for CRTC registers updated / panics removed
- use the number of visible columns in text mode when calculating the screen
width and the cursor position
. ports 0x0400 0x0401 and 0xfff0 : rombios
. ports 0x0500 0x0501 and 0x0502 : vgabios
The rombios log output was previously handled by the unmmapped device
- fixed handling of rotate_on_autoeoi for master PIC in in IAC()
- removed old IRQ handling functions (trigger_irq / untrigger_irq)
- replaced BX_INFO in case of bx_dbg.pic by BX_DEBUG calls
the SMP interrupt handling is to blame. So I'm putting the 1.46 changes
back into the code. Now keyboard.cc is exactly the same as revision 1.51
again. A copy of the changes is now in patches/patch.eks-port64 so that
they can be easily examined by SMP debugging people.
boot of linux 2.2.17 to freeze forever. I don't understand what's being
done in this rev, so I'll ask Dave to check over it and see he can find
anything suspicious.
called <linux/netlink.h> the ethertap module will be defined. If other
OSes turn up that also have ethertap, we can change the test.
- in eth_tap.cc, I replaced the calls to GPL code from maconlinux with
my own stuff.
There is a problem with the existing function rx_timer_handler. It can
retrieve multiple packets from winpcap but returns when it finds one with
the source mac address address instead of ignoring it and processing any
remaining packets.
Replacing the function with the one in the attached file rectifies this.
With this I can browse the network fairly well from the guest NT4 and indeed
see the guest NT4 from other machines on the network.
Mike Lerwill <Mike@ml-solutions.co.uk>
- at the moment we only use hFile so the other pieces of the ADAPTER struct
don't affect how this code turns out. However, it would be safer to include
the winpcap header file instead of hardcoding the structures, which are bound
to change over time.
We should really be using #defines or enums to give these constants
a proper name! Thanks to Peter Tattam <peter@jazz-1.trumpet.com.au>
for the bug report.
Here are the patches (without the serial name fix). Let me know if you want
me to change that. Oh, FYI, my changes also open up the com1 terminal in raw
mode, which makes the emulation much more accurate. You'll also notice that
I added infrastructure for com2->4 in the option parsing. I didn't add it to
the serial code, as I think a bunch of things need to be untangled before
that can work.
This is basically the opposite of the slowdown timer. Instead
of trying to keep the PIT ticks in sync with bochs time, we
keep them in sync with REAL time. This is bad because it creates
unreproducible fails, but it's good if you want to run bochs at
maximum speed on your machine. However, bochs will take all of
the available resources from the machine also.
DO NOT use this with the slowdown timer. Results would be
unpredictable.
- new functions raise_irq() and lower_irq()
- all trigger_irq() / untrigger_irq() calls are replaced by the new functions
- REMARK: timer IRQ handling is not correct but it works
- TODO: IOAPIC IRQ handling needs to be changed
- interrupt reasons modem status change and receiver line status change added
- number of data bits is considered now
- REMARK: Windows 95 driver still makes trouble
- TODO: real serial communication, FIFO, second serial port
- length of 'configure' floppy command fixed
- busy flag is set until the result of a read/write command is complete
- read/write access to unsupported address causes a BX_ERROR, not a BX_PANIC
- commented BX_INFO statements removed
* floppy command 'format track' implemented
* read and write operations with MT=0 are working now
(function 'increment_sector()' updated)
* result code of floppy command 'get status' fixed
* flag FS_MS_DIO is not set while the 'configure' floppy command is pending
* diskette controller data register returns last result if no new data
is available
* reset will be activated when the reset bit is changed to normal operation
* reset sets the error bits in status register 0
* write access to port 0x3f4 will cause a BX_ERROR now
* unsupported and invalid floppy commands are setting the error status bit
'invalid command' - BX_PANIC not necessary
* flag FS_MS_DIO is not set while a floppy command is pending
* floppy command 'specify': cause a BX_ERROR when non-DMA mode is selected
* floppy command 'sense interrupt status' returns an error is no interrupt
is pending
* floppy command 'read ID' sets the 'busy' flag and returns no data if the
motor is not on
* removed SIMX86 section (not defined in bochs)
* define variable 'sTemp' for win32 only
- fopen with "wb" so that win32 doesn't translate CRLF
- add initmode setting
- fix bug, fill in behavior in write handler
- now it works for him with DOS, Linux, and Windows guests
probably been treated as ifdef instead, but a sun compiler doesn't like
them. Anyway, they were being used around a check for irq_num > 15.
This bounds check seems ok to do all the time, so I just removed the
#if BX_DEBUG lines and corresponding #endifs.
interface menus. Parallel port #1 is implemented, and I left stubs for
parallel port #2 in case we want to ever add it. If the parallel port
is enabled, the init method of parallel.cc does an fopen() on the output
file. If disabled, or if the fopen fails, the file handler remains
NULL and no characters are printed. There is no attempt to enable/disable
the operation of the parallel port, only the output to a file.
[ #468340 ] pic:slave: OCW3 not implemented
The service_master_pic() method supported special mask mode but
service_slave_pic() did not. I added the code to service_slave_pic(). I
have no clear way to test that this is actually working right. If I can put
a gdb breakpoint in the pic.cc code and then step through and watch it work,
I'll be more confident.
I compiled Bochs on Linux and installed a linux
in it, but when I ping a machine on my LAN, I get
packet loss. Sometimes as much as 70% is lost.
So I read ne2k.cc, Linux 8390 driver and 8390 chip
specification. I find that 8390 command register START
bit is misused in ne2k.cc. According to the chip
specification, even if START=0, the chip does not stop
working.
device and disk file for a while. Even though its version of
read_toc is minimal, in fact I would say broken, it lets people use
an ISO disk file as a cdrom.
- in this revision, I wrote the "unix equivalent" of the win32 code, including
the broken version of read_toc. Now win32 and unix should act very similar
when they encounter an ISO disk image.
- one important improvement is in read_toc, I have added "*length=1" for both
win32 and unix, since otherwise the function returns random junk for the
length of the TOC. I also tried "*length=0" and that created the "lost
interrupt" behavior that psyon has been trying to get rid of...I changed it
back to *length=1 of course and left a note to him in that bug report.
which were generated with gcc -MM to the end of each Makefile.in
so that make understands which files depend on which. Basically,
everything depends on bochs.h, which depends on everything, which
is not ideal.
of just panicing. In particular, if the logical sector is out of bounds
or the disk image cannot be read/written at the desired offset, we now
abort the ATA command and return an error code. Many of the old BX_PANIC
messages are turned to BX_ERROR, so they will still appear in the
log, but now the device model will try to communicate this fact to
the OS instead of simply giving up.
I still need to do add some commands from older specs that are obsolete
(and not listed) in ATAPI-6.
- commands that aren't in the spec will still panic.
- fill in names from spec on various commands
- add command aborted for 0x08 device reset on disks (it's only for ATAPI)
- add command aborted for 0xe1 idle immediate
This is a patch from Volker Ruppert <Volker.Ruppert@t-online.de>, who
comments: "The fdisk command reports an unusable second harddisk if the cdrom
is enabled. This patch helps, but I don't know if it is the right way."
> I have inspected the header of output file and several sample MIDI files. I
> have found two different bytes in the header. After I have changed this
> bytes Winamp could play the output file, but it showed a track time of
> 0:00. The Windows Media Player still doesn't like the file.
posted to bochs-developers on Wed, 29 Aug 2001 00:08:45 +0100
David Haslam wrote:
> I have been looking at the keyboard problem with Minix, which for
> those that haven't tried Minix results in every key press giving the
> response: ^@
>
> I am aware of the comments in the changelog that suggest removing 2
> keyboard ACKs in iodev/keyboard.cc, but this is a bit of a hack,
> (which is presumably why it was never incorporated).
>
> The problem seems to be that the Minix keyboard driver doesn't obey
> the rules, and Bochs doesn't model the 8042 accurately. When issuing
> commands to set the LEDs, Minix polls the 8042 output data register
> waiting for an ACK even though the OBF flag isn't set.
>
> Bochs returns zero under these circumstances, which seems to
> trigger obscure behaviour that messes up the Minix internal
> keyboard queue. I don't fully understand why Minix breaks, but I
> think the fact that the ACK generates an interrupt, and Minix treats
> it as a scan code also has something to do with it.
>
> In any case, I believe, the fix is for Bochs to return the output
> buffer contents, regardless of whether the OBF flag (called outb in
> Bochs) is set.
>
> I expect the real hardware allows the register to be read at anytime,
> and with this fix we are modelling the behaviour of the hardware more
> accurately.
>
> I have tested it with Minix 2.0.0. Also DOS 6.22 still works after this fix.
> Does anyone think this will break anything else?
>
> This diff is against the latest CVS of iodev/keyboard.cc, version 1.29
> The last part of the diff is a minor fix to an unrelated debug print.
Fixed a "feature" in pc_system.cc with setting timers to small values
that can cause bochs to hang.
Significantly improved the performance of the new PIT.
It's probably ready to become the default now.
Added a preliminary implementation of the slowdown timer
that Bryce and I had talked about.
Also added a hack to keep the OpenBSD timer problem from filling the log.
The new PIT seems to work, but until some
enhancements are made to the way the timers
and devices.cc work, it'll be slower than
the old one.
The original code for determining the capacity of a disk only worked for
ATAPI drives, leaving us poor SCSI users in the cold. The code uses the
standard Linux CD-ROM driver routines, so it should work on any supported
drive. It's basically just a copy of Keith Jones FreeBSD code.
appeared in the guest OS. Full description:
> After much grovelling through the 8390 docs, I think this is the
> correct answer to the odd-length packet problem I was having with
> the ne2k driver under Linux.
>
> According to the datasheet, the 8390 always accesses its buffer
> memory in word-size chunks if the WTS bit of the DCR is set. So
> it will always send a word to the host bus interface if WTS==1.
> It's up to the host bus interface to deliver the the number of
> requested bytes to the host. So disallowing a byte read when the
> WTS bit is set is wrong (IMO) as the bus interface may allow it,
> as the NE2000 appears to.
>
> The patch to ne2k.h bumps the receive buffer memory size to 32K.
> This fixes the "out-of-bounds chipmem read" errors I was getting.
>
> Can someone with an NE2K datasheet verify these changes? They
> jibe with the Linux ne.c driver, anyway.
for Linux!!! I tested this using host OS kernel 2.2.14, and was able
to use telnet, ftp, irc, lynx, etc. Because it is a packet filter
solution, you aren't able to talk to the host machine, only to other
machines on the network. The patch itself is in
patches/patch.ethlinux-splite.
X servers that I've seen, however on other X servers it makes all
key mappings into absolute junk. We need to continue to work on this
patch to support all X servers and all key maps.
> The Linux 2.4.5 CD-ROM driver sends a READ_DISC_INFO command which caused
> an "unrecognized ATAPI command" panic. Looks like READ_DISC_INFO is only
> recognized by CD-R and CD-RW drives, so I ignore it for now. (I don't
> know if ASC_INV_FIELD_IN_CMD_PACKET is the right code, but it shouldn't
> matter to Linux anyway.)
been converted into parameters temporarily have the letter "O" appended
to their name. I don't want to keep it this way, but it has helped
in the conversion process because the compiler refuses to compile the
old uses of the name. Before I started using the "O" trick, there were
many bugs like this: if (bx_options.diskc.present) {...}
This was legal with the new parameters, but it was testing whether the
parameter structure had been created, instead of testing the value of
the present parameter. Renaming present to Opresent turns this into
a compile error, which points out the incorrect use of the param.
- the "--disable-control-panel" no longer works, I'm afraid. I can no
longer support this and continue progress.
declared as bx_param_c * types in the bx_options structure. They are
initialized in main.cc (bx_init_options) with default values.
Access to parameters of this type should always be like this:
bx_options.mouse_enabled->get ();
bx_options.mouse_enabled->set (newval);
Eventually I will be transferring all options to this format.
a read-only disk image. For systems such as DOS that actually use the
BIOS services, it was also necessary to add code in int13_diskette_function
to recognize a write-protected error and return the correct error
status code (AH=3, Carry Set).
now floppy.cc no longer crashes if you try to open a write-protected
disk or read-only disk image. Instead, it tries a second time to
open the image read-only and only panics if this also fails. If the
image is opened read-only, a readonly flag is set
(bx_floppy.s.media[drive].read_only). If you try to write the floppy
when this flag is set, the write silently fails except for some messages
into the log. Instead of failing silently we should learn what the
floppy controller would really do in this situation and emulate it.
BX_SUPPORT_APIC were used. To follow the pattern used by other
names like this, I changed them all to BX_SUPPORT_APIC.
Thanks to Tom Lindström for chasing this down!
an fpos_t, use ftell which returns an int. Without the patch,
Bochs gets an fpos_t and assumes it is an integer type, but
on some systems (like linux with newer glibc libraries) this
assumption is wrong. Malte Cornils <malte@cornils.net> first
reported this bug, and he warned me that ftell may not be portable
to some platforms which treat binary and ascii streams differently.
I haven't found any alternative yet.
- now the HALT macro in rombios.c writes to panic port but does not actually
execute a "hlt" instruction. This allows the .bochsrc to control
whether the BIOS panic is fatal or not.
To see the commit logs for this use either cvsweb or
cvs update -r BRANCH-io-cleanup and then 'cvs log' the various files.
In general this provides a generic interface for logging.
logfunctions:: is a class that is inherited by some classes, and also
. allocated as a standalone global called 'genlog'. All logging uses
. one of the ::info(), ::error(), ::ldebug(), ::panic() methods of this
. class through 'BX_INFO(), BX_ERROR(), BX_DEBUG(), BX_PANIC()' macros
. respectively.
.
. An example usage:
. BX_INFO(("Hello, World!\n"));
iofunctions:: is a class that is allocated once by default, and assigned
as the iofunction of each logfunctions instance. It is this class that
maintains the file descriptor and other output related code, at this
point using vfprintf(). At some future point, someone may choose to
write a gui 'console' for bochs to which messages would be redirected
simply by assigning a different iofunction class to the various logfunctions
objects.
More cleanup is coming, but this works for now. If you want to see alot
of debugging output, in main.cc, change onoff[LOGLEV_DEBUG]=0 to =1.
Comments, bugs, flames, to me: todd@fries.net
- made the code that fills 0x1b-0x23 for diskc conditional on diskc being
present; this was probably not necessary.
- added some code (still commented out) that will help in supporting a second
IDE interface.
Basicly what it does is, it help VGA16 find the syncing, because of the
technique it uses, bochs did not help it figure out the timing diffrences
with its one size fits all timing.
The 82c54 model (pit.cc) implements timer modes 0, 2, and 3 in its handler
functions, without caring which timer number is involved. However, the
I/O write code that sets the mode is inconsistent.
Timer 0 can be set to modes 0,2,3 only.
Timer 1 can be set to mode 2 only.
Timer 2 can be set to mode 2,3 only.
From a quick reading of an 8254 datasheet, I can't see any reason to
restrict which timer can be in which mode, so I think it's correct to
allow ALL timers to go into ALL modes that are implemented.
