ETRAX-FS: Add support for DMA channel resets, needed for recent linux kernels.
* Correct numeric value for the RST state.
* Add emulation for reseting a DMA channel.
* Add a few sanity checks.
* Make it compile with debug enabled.
This patch enhances the pcnet NIC emulation with better loopback mode
support, including CRC generation for looped-back packets in "raw" mode.
The patch has practically no impact on the normal RX and TX path.
Successfully tested against an ancient proprietary pcnet driver that
does a lot of hardware checks on boot-up and now works fine over qemu as
well.
This adds trivial support for the R2D-PLUS FPGA, mostly just for the
versioning information that the kernel uses for IRL mappings, in addition
to handling the heartbeat and poweroff writes.
Trivial patch adding CPU listing and the ability to do per-subtype
CVR/PVR/PRR values. Presently SH7750R and SH7751R definitions are
provided, as these are the ones in present use in-tree.
The CVR value for SH7751R is intentionally restricted so the kernel
boots, though this will want to be switched to the proper CVR value
once system emulation has sufficiently stabilized.
This also makes it trivial to abstract subtype specific registers like
MMU_PTEA and to set up feature bits in line with the kernel probing for
things like conditionalizing FPU/DSP context.
Trivial fix for a corner case: system_shutdown on isapc machines causes
qemu to segfaults due to accessing the uninitialized pm_state. Issue a
system shutdown instead.
blueswir1 [Fri, 29 Aug 2008 20:50:21 +0000 (20:50 +0000)]
Fix Sparc64 boot on i386 host:
- move do_interrupt() back to op_helper.c
- move non-helper prototypes from helper.h to exec.h
- move some prototypes from cpu.h to exec.h
- do not export either set_cwp() or cpu_set_cwp() from op_helper.c,
but instead provide inline functions
aliguori [Sat, 23 Aug 2008 23:27:37 +0000 (23:27 +0000)]
VNC: Support for ExtendedKeyEvent client message
This patch adds support for the ExtendedKeyEvent client message. This message
allows a client to send raw scan codes directly to the server. If the client
and server are using the same keymap, then it's unnecessary to use the '-k'
option with QEMU when this extension is supported.
This is extension is currently only implemented by gtk-vnc based clients
(gvncviewer, virt-manager, vinagre, etc.).
aurel32 [Sat, 23 Aug 2008 17:22:19 +0000 (17:22 +0000)]
MIPS: don't free TCG temporary variable twice
In gen_dmtc0 function, TCG temporary variable t0 is freed at the end of
the function. Variable is freed again in the gen_dmtc0 caller.
I removed the free in gen_dmtc0, to do like in gen_dmfc0, gen_mfc0,
gen_mtc0.
aurel32 [Fri, 22 Aug 2008 09:23:06 +0000 (09:23 +0000)]
uhci: improved TD matching, working ISOC transfers
While trying to make VX-3000 camera work on XP under KVM I realized that
we do not necessarily have to find original TD address. All we care about
is the token which identifies the transfer rather well (direction, endpoint,
size, etc).
This is especially important for the isochronous transfers because otherwise
they are being canceled left and right and we do not make much progress.
With this patch all devices that used bulk transfers that I've tried so
far continue to work just as well. And now my USB web cammera (isoc transfers)
is working well tool. It's not as smooth as native Windows but it's pretty
darn smooth.
The cool thing is that new USB code (both usb-uhci and usb-linux) is totaly
generic and does not need any special logic for ISOC.
aurel32 [Fri, 22 Aug 2008 08:58:00 +0000 (08:58 +0000)]
[sh4] code translation bug fix
When a TLB miss occurs while pre-decrement store instruction such as
"mov.l Rm, @-Rn" is executed, re-execution of such instruction cause
status confusion. Because pre Rn decrement is executed before TLB miss,
re-execution decrements Rn again.
In other words, in a translated instruction array, register status
should not modified before memory access instruction.
aurel32 [Fri, 22 Aug 2008 08:57:52 +0000 (08:57 +0000)]
[sh4] MMU bug fix
Some bugs on SH4 MMU are fixed.
- When a TLB entry is overwritten or invalidated, tlb_flush_page() should be
invoked to invalidate old entry.
- When a ASID is changed, tlb_flush() should be invoke to invalidate entries
which have old ASID.
- The check for shared bit in TLB entry causes multiple TLB hit exception.
As SH3's MMU, shared bit is ignored.
- ASID is used when MMUCR's SV bit or SR's MD bit is zero.
No need to check both bits are zero.
aurel32 [Fri, 22 Aug 2008 08:57:43 +0000 (08:57 +0000)]
[sh4] memory mapped TLB entries
SH4 MMU's memory mapped TLB feature is implemented.
SH-Linux seems to write to memory mapped TLB to invalidate a TLB entry,
but does not to read it. So only memory write feature is implemented.
Work on memory read feature is left.
aurel32 [Fri, 22 Aug 2008 08:57:35 +0000 (08:57 +0000)]
[sh4] delay slot bug fix
Two bugs about delay slot handlings are fixed.
- After an exception occurred in delay slot, the branch instruction
before delay slot should be executed again. To judge such re-execution
is necessery or not, delay slot status is kept in SH4 CPU data structure.
- When a branch instruction is placed at the end of memory segment,
the delay slot is placed at the start of next memory segment.
It means delay slot comes to the start of a translation block.
In such occasion, DELAY_SLOT_CLAREME flag is used to transmit status
between translation blocks. When an exception occurs on this kind of
delay slot, DELAY_SLOT_CLEARME flag cause a status confusion in exception
handling. DELAY_SLOT_CLEARME flag should be cleared on exceptions.
aurel32 [Fri, 22 Aug 2008 08:57:09 +0000 (08:57 +0000)]
Parallel Port Direction Fix
The direction bit in the control register should not be directly
set using PPWCONTROL. The kernel gives the following debug message.
parport0 (ppdev0): use data_reverse for this!
More over setting the data pins to forward mode does not work,
perhaps a bug in the Linux PP driver. The right way to do this is
to use PPDATADIR to set the direction. The patch checks if the
user is toggling the direction bit, and invokes PPDATADIR to
do the job.
aliguori [Thu, 21 Aug 2008 22:40:00 +0000 (22:40 +0000)]
Ignore IDE command if issued while IDE is busy (Gleb Natapov)
Feature, Sector Count, LBA Low/Mid/High and Device registers should be
written only when both BSY and DRQ are cleared to zero.
Command register shall only be written when BSY and DRQ are set to zero
for all commands except DEVICE RESET.
Data Port register shall be accessed for host PIO data transfer only when
DRQ is set to one.
aliguori [Thu, 21 Aug 2008 20:08:55 +0000 (20:08 +0000)]
add DisplayState->idle (Samuel Thibault)
Add idle field to DisplayState struct, so drivers can figure
the display is idle and take advantage of that.
The xen framebuffer driver will use this to communicate the
idle state to the guest, so it knows it can stop doing updates
to a virtual display which is invisible anyway.
aliguori [Thu, 21 Aug 2008 20:08:03 +0000 (20:08 +0000)]
Handle terminating signals (Gerd Hoffmann)
This patch makes qemu handle signals better. It sets the request_shutdown
flag, making the main_loop exit and qemu taking the usual exit route, with
atexit handlers being called and so on, instead of qemu just being killed
by the signal.
To avoid calling vm_start() from the signal handler main_loop() got an
additional check so qemu_system_shutdown_request() works even when the
vm is in stopped state.
aliguori [Thu, 21 Aug 2008 19:33:09 +0000 (19:33 +0000)]
uhci: fixes for save/load-vm (Max Krasnyansky)
For some reason we were not registering save/load-vm handler for piix3
flavor of UHCI and hence save/load was broken.
Async transactions need to be canceled when we save the VM because there
is no way we can save/restore all that state. Since we do not mess the
original TD/QH the driver will simply resubmit the transfers.
Tested with Windows XP-SP2 running under QEMU/KQEMU.
aliguori [Thu, 21 Aug 2008 19:31:10 +0000 (19:31 +0000)]
husb: rewrite Linux host USB layer, fully async operation (Max Krasnyansky)
This is a follow up to the async UHCI patch. Both BULK and ISOC transactions
are now fully asynchrounous. I left CONTROL synchronous for now, ideally
we want it to be async too and it should not be that hard to do now.
This patch obviously requires UHCI patch. The combo has been tested with
various devices. See the UHCI patch description for list of the devices.
Most of the testing was done with the KVM flavor of QEMU.
This is esentially a re-write of the QEMU UHCI layer. My initial goal
was to support fully async operation with multiple outstanding async
transactions. Along the way I realized that I can greatly simplify
and cleanup the overall logic. There was a lot of duplicate and confusing
code in the UHCI data structure parsing and other places.
We were actually violating UHCI spec in handling async ISOC transaction
(host controller is not supposed to write into the frame pointer).
The reason I wanted to support fully async operation is because current
synchronous version is unusable with most devices exported from host
(via usb-linux.c). Transactions take a long time and the whole VM becomes
slow as hell.
Current async support is very rudimentory and for the most part
non-functional. Single transaction at a time is simply not enough. I have
a device for which XP driver submits both IN and OUT packets at the same
time. IN packet always times out unless OUT packet makes it to the device.
Hence we must be able to process both in order for that device to work.
The new code is backwards compatible and was first tested agains original
synchronous usb-linux.c and builtin usb devices like tablet which is also
synchronous. Rewrite of the usb-linux.c is coming up next.
Async support was tested against various XP versions (ie XP, SP2, SP3) and
a bunch of different USB devices: serial port controllers, mice, keyboard,
JTAG dongles (from Xilinx and Altera).
ISOC support was only lighly tested and needs more work. It's not any worse
than current code though.
UHCI parser changes are probably somewhat hard to review without the
understanding of the UHCI spec.
The async design should be fairly easy to follow. Basically we have a list
of async objects for each pending transfer. Async objects are tagged with
the original TD (transfer descriptor) address and token. We now support
unlimited number of outstanding isoc and one outstanding bulk/intr/ctrl
transfer per QH (queue head). UHCI spec does not have a clear protocol for
the cancelation of the trasfer requests. Driver can yank out TDs on any
frame boundary. In oder to handle that I added somewhat fancy TD validation
logic logic to avoid unnecessary cancelations.
aliguori [Thu, 21 Aug 2008 19:29:38 +0000 (19:29 +0000)]
usb: generic packet handler cleanup and documentation (Max Krasnyansky)
A bit better documentation of the USB device API, namely
return codes.
Rewrite of usb_generic_handle_packet() to make it more
reable and easier to follow.
projects / qemu.git / log