Stefan Weil [Sat, 30 Apr 2011 20:40:07 +0000 (22:40 +0200)]
eepro100: Pad received short frames
QEMU sends frames smaller than 60 bytes to ethernet nics.
Such frames are rejected by real NICs and their emulations.
To avoid this behaviour, other NIC emulations pad received
frames. This patch enables this workaround for eepro100, too.
All related code is marked with CONFIG_PAD_RECEIVED_FRAMES,
so we can drop this in case QEMU's networking code is
ever changed.
Stefan Weil [Sat, 30 Apr 2011 20:40:04 +0000 (22:40 +0200)]
eepro100: Avoid duplicate debug messages
When DEBUG_EEPRO100 was enabled, unsupported writes were logged twice.
Now logging in eepro100_write1 and eepro100_write2 is similar to the
logging in eepro100_write4 (which already was correct).
Cirrus VGA (at least) calls register memory region
with the same values again and again. The
registration in vhost-net slows this a lot,
optimize by checking that the same data is already registered.
use the new api to reduce the number of these (expensive)
system calls.
Note: using this API, we should be able to
get rid of vga_dirty_log_xxx APIs. Using them doesn't
affect the performance though because we detects
the log_dirty flag set and ignores the call.
The previous patch didn't change the behavior when load,
it resulted in ugly code. This patch cleans it up.
With this patch, pic irq lines are manipulated when loaded.
It is expected that it won't change the behaviour because
the interrupts are level: at the moment e.g. pci devices already
reassert interrupts on load.
Test:
- rung linux as guest and use flooding ping (ping -f) to host
in order to trigger interrupts for e1000 emulated.
- savevm/loadvm and see guest kept running after loadvm.
To be honest, I'm not sure that ping -f caused enough interrupts
because Linux e1000 driver supports NAPI.
TODO: test more OSes, stress test with save/load, live-migration
optimize irq routing in piix_pic.c which has been a TODO.
So far piix3 tracks each pirq level and checks whether a given pic pins is
asserted by seeing if each pirq is mapped into the pic pin.
This is independent on irq routing, but data path is on slow path.
Given that irq routing is rarely changed and asserting pic pins is on
data path, the path that asserts pic pins should be optimized and
chainging irq routing should be on slow path.
The new behavior with this patch series is to use bitmap which is addressed
by pirq and pic pins with a given irq routing.
When pirq is asserted, the bitmap is set and see if the pic pins is
asserted by checking the bitmaps.
When irq routing is changed, rebuild the bitmap and re-assert pic pins.
test:
- create VM with 4 e1000 nics in different pci slots
(i.e. fn=0 for each e1000)
Thus those e1000's INTA are connected to each PIRQ[A-D].
- run linux as guest and saw each devices triggers interrupt
by seeing /proc/interrupts. And then confirmed that each PIRQ[A-D]
surely asserted interrupts.
Because irq 10 and 11 are shared by 4 e1000's, it only one NIC is activated
with ifconfig ethN up/down when counting interrupts.
Introduce accessor function to know INTx levels.
It will be used later by q35.
Although piix_pci tracks the intx line levels, it can be eliminated
by this helper function.
Commit c81131db15dd1844d0db1d51f3cd7a105cfd2cf3
detects old guests by comparing virtio and
PCI status. It attempts to do this on load,
as well, but load_config callback in a binding
is invoked too early and so the virtio status
isn't set yet.
We could add yet another callback to the
binding, to invoke after load, but it
seems easier to reuse the existing vmstate
callback.
Gleb Natapov [Tue, 15 Mar 2011 11:56:04 +0000 (13:56 +0200)]
report that QEMU process was killed by a signal
Currently when rogue script kills QEMU process (using TERM/INT/HUP
signal) it looks indistinguishable from system shutdown. Lets report
that QEMU was killed and leave some clues about the killer identity.
AFAIK, extraction is optional to get vlans working. The driver
requests rx detagging but should not assume that it was done. Under
Linux, the mac layer will catch the vlan ethertype. I only added this
part for completeness (to emulate the hardware more truthfully...) Signed-off-by: Blue Swirl <[email protected]>
Peter Maydell [Wed, 16 Mar 2011 15:21:31 +0000 (15:21 +0000)]
tcg/arm: Support host code being compiled for Thumb
Although the TCG generated code is always in ARM mode, it is possible
that the host code was compiled by gcc in Thumb mode (this is often the
default for Linux distributions targeting ARM v7 only). Handle this
by using BLX imm when doing a call from ARM into Thumb mode.
Since BLX imm is not a conditionalisable instruction, we make
tcg_out_call() no longer take a condition code; we were only ever
using it with COND_AL anyway.
Jes Sorensen [Wed, 9 Mar 2011 13:31:06 +0000 (14:31 +0100)]
qmp-commands.hx: Clean up mess of client_migrate_info
client_migrate_info was put into qmp-commands.hx in the middle of
migrate_set_speed, between the command and it's description. In
addition client_migrate_info put the description before the command
itself, which is the wrong order.
Jes Sorensen [Wed, 16 Mar 2011 12:33:32 +0000 (13:33 +0100)]
Introduce -display none
New option -display none. This option differs from -nographic by not
trying to take control of stdio etc. but instead behaves as if a
graphics display is enabled, except that it doesn't show one.
Peter Maydell [Mon, 14 Mar 2011 15:37:13 +0000 (15:37 +0000)]
target-arm: use make_float32() to make constant floats for VRSQRTS
The preferred way to create a constant floating point value is to use
make_float32() rather than doing a runtime int32_to_float32().
Convert the code in the VRSQRTS helper to work this way.
Peter Maydell [Mon, 14 Mar 2011 15:37:12 +0000 (15:37 +0000)]
target-arm: Fix VRECPS edge cases handling
Correct the handling of edge cases for the VRECPS instruction:
* this is a Neon instruction so uses the "standard FPSCR value"
* (zero, inf) is a special case which returns 2.0
Peter Maydell [Fri, 11 Mar 2011 10:09:58 +0000 (10:09 +0000)]
target-arm: Set Q bit for overflow in SMUAD and SMLAD
SMUAD and SMLAD are supposed to set the Q bit if the addition of
the two 16x16 multiply products and optional accumulator overflows
considered as a signed value. However we were only doing this check
for the addition of the accumulator, not when adding the products,
with the effect that we were mishandling the edge case where
both inputs are 0x80008000.
Peter Maydell [Thu, 10 Mar 2011 18:51:49 +0000 (18:51 +0000)]
target-arm: Fix GE bits for v6media signed modulo arithmetic
Fix the signed modulo arithmetic helpers for the v6media
instructions (SADD8, SSUB8, SADD16, SSUB16, SASX, SSAX) to set
the GE bits correctly (based on the result of the add or subtract
before it is truncated to 16 bits, not after).
Peter Maydell [Thu, 10 Mar 2011 16:48:49 +0000 (16:48 +0000)]
target-arm: Fix UNDEF cases in Thumb load/store
Decode of Thumb load/store was merging together the cases of 'bit 11==0'
(reg+reg LSL imm) and 'bit 11==1' (reg+imm). This happens to work for
valid instruction patterns but meant that we would not UNDEF for the
cases the architecture mandates that we must. Make the decode actually
look at bit 11 as well as [10..8] so that we UNDEF in the right places.
This change also removes what was a spurious unreachable 'case 8',
and correctly frees TCG temporaries on the illegal-insn codepaths.
Peter Maydell [Fri, 25 Feb 2011 10:27:40 +0000 (10:27 +0000)]
linux-user: Fix unlock_user() call in return from poll()
Correct the broken attempt to calculate the third argument
to unlock_user() in the code path which unlocked the pollfd
array on return from poll() and ppoll() emulation. (This
only caused a problem if unlock_user() wasn't a no-op, eg
if DEBUG_REMAP is defined.)
Gleb Natapov [Mon, 31 Jan 2011 13:11:01 +0000 (15:11 +0200)]
fix linuxboot.bin and multiboot.bin to not hijack int19
Currently linuxboot.bin and multiboot.bin option roms override int19
vector to intercept boot process. No sane option rom should do that.
Provide bev entry instead that will be called by BIOS if option rom
is selected for booting.
Anthony Liguori [Tue, 15 Mar 2011 16:22:12 +0000 (11:22 -0500)]
seabios: update to latest git
- cc97564 Add config option to permit running option roms with bad checksums.
- 424f217 Add config option to disable MTRR initialization.
- fb2f10d Prep version for next release.
- f18b09b Update version to 0.6.2
- 20fcf9b lets pretend that RTC can be used to wakeup from S4
- 1efb10b Expand user configurable parameters in Kconfig.
- 3dba4c2 Support non-local build directory - allow "make OUT=abc/" to work.
- 42d9771 The vgabios build must depend on autoconf.h.
- 32aa9f3 fix parallel compilation of SeaBIOS
- 713be89 Start using Kconfig to configure SeaBIOS settings.
- b623e7c Change kconfig to emit disabled symbols in autoconf.h.
- 0da7bfd Build changes for Linux kconfig code to work in seabios dir structure.
- a4c5daf Initial commit of Kconfig build tool.
- 4d0c592 Minor boot fixes.
- 812478e Minor comment / code layout improvement to romlayout.S.
- 6fc7cf1 Fix to prevent infinite loop in build_pci_path().
- f5154e2 support T13 EDD3.0 spec
- d08eb9c Add ability to track PCI paths and add to build_pci_path().
- 89a1efd Move the CBFS payload setup to later in the boot.
- 9e881a3 Extract space trimming code from ATA and use in USB and bootorder code.
- 3c241ed seabios: acpi: add _RMV control method for PCI devices
- c2002a1 Add support for finding the boot priority of USB drives.
- a5f2b91 Extend 'usb_pipe' to track the controller and ports of each device.
- 8bf5503 Add functions for boot device path parsing.
- c4bd3b9 Use bootprio_find_named_rom() for ramdisk and cbfs payload priorities.
- ca2bc1c Remove drive->desc field.
- f13a180 Minor reorganization of some of the boot_xxx code in boot.c.
- 3da2c1c Move IPL.fw_bootorder to static variables in boot.c.
- 7bb1584 Move IPL.bev to static variables in boot.c
- baaadb6 Move IPL.checkfloppysig to a global (CheckFloppySig) in boot.c.
- a0842f8 Remove Drives global struct in favor of independent global variables.
- ecbcf77 Don't access drive_g->desc from boot_cdrom().
- 551caa2 Simplify keyboard reading code in the interactive boot menu.
- 697e63c Call setup_translation() from map_hd_drive().
- 3c5e0e1 Rename add_ordered_drive() to add_drive() and use in map_hd_drive().
- 031ef55 Add stubs to permit devices to specify their boot priority.
- 72eee3e Simplify boot ordering by building an inclusive boot list.
- ce24be5 Populate drive_g->desc prior to calling add_bcv_internal().
- 74fd942 pciinit: fix off-by-one
- 311f887 Minor build fixes.
- bca3a87 Track the source of each optionrom deployed.
- c6629e0 Support qemu based romfile wrappers called out of order.
- d1a1746 Breakup boot_setup() bootorder code into its own function.
- f9b0930 Add romfile_loadfile() helper function.
- b9a7591 Read bootorder file into memory.
- 2e109a6 Add strchr() function.
- 1703ea2 Add romfile_name() function.
- b674152 Add BAID for ahci cdrom.
- f77e179 Provide full EDD 3.0 info for virtio disk
- 67863be Enhance layoutrom.py to work around a binutils quirk.
- 4c90a20 Create separate IPL entry for each CD/DVD
- 0e6f636 ahci: set controller id
- 1e924bb ahci: fix off-by-one in port count
- 7eb0222 ahci: enable io/mem/dma
- c19fc71 ahci: set dma feature flag
- 80c2b6e Check if capability enabled in XXX_cmd_data functions.
- f3fe3aa Require a "_cfuncXX_" symbol prefix for inter-mode c function references.
- d52fdf6 add ahci support
- 54fa8ec ata: make helpers available
- 0a80608 util: add memset_fl()
- f1f18eb pci: add helper functions for mmio bar access from real mode.
- af9629b Enhance call32() to pass a parameter to called function.
- 4057f98 Don't pass return address to transition(32,16,16big) on stack.
- e2623fc pciinit: use pci_region functions.
- 58a38e8 pci: introduce pci_region to manage pci io/memory/prefmemory regions.
- 6d5a217 mark irq9 active high in DSDT
- 9433098 Report meaningful error if pyserial not present in tools/readserial.py
- 94dc9c4 Show size of non-relocatable runtime code during build.
- 4d96edc Cleanup - it's no longer necessary to manually reset global variables.
- a899945 Allow rom to grow to 256K.
- 4a446d7 Fix typo preventing relocated space from being used for option roms.
- 0f67397 fix virtio-blk failure after reboot
- d4bded4 Add a debug method to AML code.
- 2f96800 Warn that ohci bulk is not supported.
- e826465 Enable optionroms to use freed space due to CONFIG_RELOCATE_INIT.
- 7f55fd3 Move the 32bit init code to high memory at runtime.
- 025cabd Move init code from _start() to post().
- 12fa24a Add memalign_tmp() helper function.
- 244caf8 Try to hard-reboot on rerun of post even on emulators.
- 5bd01de Don't do shadow copying of optionroms when CONFIG_OPTIONROMS_DEPLOYED.
- adaf373 Try to hard-reboot processor on rerun of post under coreboot.
- 402fd9c Enhance build to emit relocation information.
- d1b4f96 Separate out init code from the rest of the 32bit flat code.
- 1a4885e Modify tools/layoutrom.py to use classes instead of tuples.
- 6c2e781 Use str.startswith() in python scripts.
- d9c916e Prep version for next release.
Andreas Färber [Mon, 7 Mar 2011 00:34:05 +0000 (01:34 +0100)]
softfloat: Resolve type mismatches between declaration and implementation
The original SoftFloat 2.0b library avoided the use of custom integer types
in its public headers. This requires the definitions of int{8,16,32,64} to
match the assumptions in the declarations. This breaks on BeOS R5 and Haiku/x86,
where int32 is defined in {be,os}/support/SupportDefs.h in terms of a long
rather than an int. Spotted by Michael Lotz.
Since QEMU already breaks this distinction by defining those types just above,
do use them for consistency and to allow #ifndef'ing them out as done for
[u]int16 on AIX.
Andreas Färber [Mon, 7 Mar 2011 00:34:04 +0000 (01:34 +0100)]
softfloat: Prepend QEMU-style header with derivation notice
The SoftFloat license requires "prominent notice that the work
is derivative". Having added features like improved 16-bit support
for arm already, add such a notice to the sources.
softfloat-native.[ch] are not under the SoftFloat license
and thus are not changed.
The PCI/PCI-X Family of Gigabit Ethernet Controllers Software
Developer’s Manual states the following about the POPTS field:
Provides a number of options which control the handling of this
packet. This field is ignored except on the first data descriptor of
a packet.
The current implementation always loads the field and its checksum
offload flags. This patch uses only the first descriptor's POPTS field
in order to comply with the specification.
When Solaris sends multi-descriptor packets it fills in POPTS for the
first descriptor only. Therefore this patch is necessary in order to
perform checksum offload correctly for multi-descriptor packets.
Aurelien Jarno [Mon, 21 Mar 2011 20:28:38 +0000 (21:28 +0100)]
Merge branch 'for-anthony' of git://github.com/bonzini/qemu
* 'for-anthony' of git://github.com/bonzini/qemu:
remove qemu_get_clock
add a generic scaling mechanism for timers
change all other clock references to use nanosecond resolution accessors
change all rt_clock references to use millisecond resolution accessors
add more helper functions with explicit milli/nanosecond resolution
Aurelien Jarno [Mon, 21 Mar 2011 20:28:19 +0000 (21:28 +0100)]
Merge branch 'for-anthony' of git://repo.or.cz/qemu/kevin
* 'for-anthony' of git://repo.or.cz/qemu/kevin:
Add qcow2 documentation
hw/xen_disk: aio_inflight not released in handling ioreq when nr_segments==0
Improve error handling in do_snapshot_blkdev()
Fix ATA SMART and CHECK POWER MODE
Don't allow multiwrites against a block device without underlying medium
tools: Use real async.c instead of stubs
Add error message for loading snapshot without VM state
block/qcow: Don't ignore immediate read/write and other failures
block/vdi: Don't ignore immediate read/write failures
Aurelien Jarno [Mon, 7 Mar 2011 06:17:49 +0000 (07:17 +0100)]
rbd: don't link with -lcrypto
rbd support tries to both link with -lrados and -lcrypto. While the
first one is of course necessary, the second is not necessary (only
librados ifself needs to link with libcrypto).
This fixes a licensing issue: qemu as a whole is GPL v2, and thus can't
be linked with OpenSSL without an exception in the license, which seems
difficult to get given the number of persons involved.
Peter Maydell [Mon, 7 Mar 2011 11:10:31 +0000 (11:10 +0000)]
hw/arm_sysctl.c: Add the Versatile Express system registers
Add support for the Versatile Express SYS_CFG registers, which provide
a generic means of reading or writing configuration information from
various parts of the board. We only implement shutdown and reset.
Also make the RESETCTL register RAZ/WI on Versatile Express rather
than reset the board. Other system registers are generally the same
as Versatile and Realview.
This includes a VMState version number bump for arm_sysctl,
since we have new register state to preserve. It also adds
sys_mci to the VMState while we're bumping the version number
(an accidental omission from commit b50ff6f5).
Amit Shah [Tue, 15 Mar 2011 08:43:09 +0000 (14:13 +0530)]
virtio-console: Keep chardev open for other users after hot-unplug
After a hot-unplug operation, the previous behaviour was to close the
chardev. That meant the chardev couldn't be re-used. Also, since
chardev hot-plug isn't possible so far, this means virtio-console
hot-plug isn't feasible as well.
With this change, the chardev is kept around. A new virtio-console
channel can then be hot-plugged with the same chardev and things will
continue to work.
Amit Shah [Thu, 3 Mar 2011 07:59:45 +0000 (13:29 +0530)]
virtio-serial: Don't clear ->have_data() pointer after unplug
After a port unplug operation, the port->info->have_data() pointer was
set to NULL. The problem is, the ->info struct is shared by all ports,
effectively disabling writes to other ports.
Amit Shah [Thu, 3 Feb 2011 07:35:07 +0000 (13:05 +0530)]
virtio-serial: Disallow generic ports at id 0
Port 0 is reserved for virtconsole devices for backward compatibility
with the old -virtioconsole (from qemu 0.12) device type.
libvirt prior to commit 8e28c5d40200b4c5d483bd585d237b9d870372e5 used
port 0 for generic ports. libvirt will no longer do that, but disallow
instantiating generic ports at id 0 from qemu as well.
Amit Shah [Thu, 3 Feb 2011 05:52:32 +0000 (11:22 +0530)]
virtio-serial: Use a struct to pass config information from proxy
Instead of using a single variable to pass to the virtio_serial_init
function, use a struct so that expanding the number of variables to be
passed on later is easier.
Paolo Bonzini [Fri, 11 Mar 2011 17:27:26 +0000 (18:27 +0100)]
remove qemu_get_clock
These patches are already not doing a great service to out-of-tree
modifications to QEMU. However, at least we can warn them by getting
rid of the old confusing functions, or otherwise causing compilation
errors. This patch removes qemu_get_clock; the previous one changed
qemu_new_timer's signature.
Paolo Bonzini [Fri, 11 Mar 2011 15:33:58 +0000 (16:33 +0100)]
add a generic scaling mechanism for timers
This enables rt_clock timers to use nanosecond resolution, just by
using the _ns functions; there is really no reason to forbid that.
Migrated timers are all using vm_clock (of course; but I checked that
anyway) so the timers in the savevm files are already in nanosecond
resolution. So this patch makes no change to the migration format.
Paolo Bonzini [Fri, 11 Mar 2011 15:47:48 +0000 (16:47 +0100)]
change all other clock references to use nanosecond resolution accessors
This was done with:
sed -i 's/qemu_get_clock\>/qemu_get_clock_ns/' \
$(git grep -l 'qemu_get_clock\>' )
sed -i 's/qemu_new_timer\>/qemu_new_timer_ns/' \
$(git grep -l 'qemu_new_timer\>' )
after checking that get_clock and new_timer never occur twice
on the same line. There were no missed occurrences; however, even
if there had been, they would have been caught by the compiler.
There was exactly one false positive in qemu_run_timers:
Paolo Bonzini [Fri, 11 Mar 2011 15:47:48 +0000 (16:47 +0100)]
change all rt_clock references to use millisecond resolution accessors
This was done with:
sed -i '/get_clock\>.*rt_clock/s/get_clock\>/get_clock_ms/' \
$(git grep -l 'get_clock\>.*rt_clock' )
sed -i '/new_timer\>.*rt_clock/s/new_timer\>/new_timer_ms/' \
$(git grep -l 'new_timer\>.*rt_clock' )
after checking that get_clock and new_timer never occur twice
on the same line. There were no missed occurrences; however, even
if there had been, they would have been caught by the compiler.
Blue Swirl [Sat, 19 Mar 2011 08:43:22 +0000 (08:43 +0000)]
petalogix_ml605_mmu: remove unused variable
Remove a write-only variable, spotted by GCC 4.6.0:
/src/qemu/hw/petalogix_ml605_mmu.c: In function 'petalogix_ml605_init':
/src/qemu/hw/petalogix_ml605_mmu.c:153:11: error: variable 'serial' set but not used [-Werror=unused-but-set-variable]
Jan Kiszka [Tue, 15 Mar 2011 11:26:31 +0000 (12:26 +0100)]
Expose thread_id in info cpus
Based on patch by Glauber Costa:
To allow management applications like libvirt to apply CPU affinities to
the VCPU threads, expose their ID via info cpus. This patch provides the
pre-existing and used interface from qemu-kvm.
Jan Kiszka [Tue, 15 Mar 2011 11:26:28 +0000 (12:26 +0100)]
kvm: Align kvm_arch_handle_exit to kvm_cpu_exec changes
Make the return code of kvm_arch_handle_exit directly usable for
kvm_cpu_exec. This is straightforward for x86 and ppc, just s390
would require more work. Avoid this for now by pushing the return code
translation logic into s390's kvm_arch_handle_exit.
Jan Kiszka [Tue, 15 Mar 2011 11:26:27 +0000 (12:26 +0100)]
kvm: Rework inner loop of kvm_cpu_exec
Let kvm_cpu_exec return EXCP_* values consistently and generate those
codes already inside its inner loop. This means we will now re-enter the
kernel while ret == 0.
Update kvm_handle_internal_error accordingly, but keep
kvm_arch_handle_exit untouched, it will be converted in a separate step.
projects / qemu.git / log