Now that q800 VIA1 and VIA2 are completely separate devices there is no need to
add a specific device prefix to ensure that the IRQ lines remain separate.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20210830102447.10806-11-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Remove the mac_via device and wire up both q800 VIA1 and VIA2 directly for the
m68k q800 machine.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20210830102447.10806-10-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
The ADB is accessed using clock and data pins on q800 VIA1 port B and so can be
moved to MOS6522Q800VIA1State.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20210830102447.10806-6-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Commit db1ffc32dd ("qemu/bitops.h: add bitrev8 implementation") introduced
a bitrev8() function to reverse the bit ordering required for storing the
MAC address in the q800 PROM.
This function is not required since QEMU implements its own revbit8()
function which does exactly the same thing. Remove the extraneous
bitrev8() function and switch its only caller in hw/m68k/q800.c to
use revbit8() instead.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20210725110557.3007-1-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
The checksum used by MacOS to validate the PROM content is an exclusive-OR
rather than a sum over the corresponding bytes. In addition the MAC address
must be stored in bit-reversed format as indicated in comments in Linux's
macsonic.c.
With the PROM contents fixed MacOS starts to probe the device registers
when AppleTalk is enabled in the Control Panel.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Tested-by: Finn Thain <fthain@linux-m68k.org>
Message-Id: <20210625065401.30170-8-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
This is in preparation for each board to have its own separate bit storage
format and checksum for storing the MAC address.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Tested-by: Finn Thain <fthain@linux-m68k.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20210625065401.30170-5-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Stop including exec/address-spaces.h in files that don't need it.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20210416171314.2074665-5-thuth@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Stop including cpu.h in files that don't need it.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20210416171314.2074665-4-thuth@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Stop including sysemu/sysemu.h in files that don't need it.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20210416171314.2074665-2-thuth@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Many files include hw/irq.h without needing it. Remove the superfluous
include statements.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20210327050236.2232347-1-thuth@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
The include/hw/hw.h header only has a prototype for hw_error(),
so it does not make sense to include this in files that do not
use this function.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20210326151848.2217216-1-thuth@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
The machine is based on Goldfish interfaces defined by Google
for Android simulator. It uses Goldfish-rtc (timer and RTC),
Goldfish-pic (PIC) and Goldfish-tty (for serial port and early tty).
The machine is created with 128 virtio-mmio bus, and they can
be used to use serial console, GPU, disk, NIC, HID, ...
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20210312214145.2936082-6-laurent@vivier.eu>
The existing ESP QOM type currently represents a sysbus device with an embedded
ESP state. Rename the type to SYSBUS_ESP accordingly.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20210304221103.6369-3-mark.cave-ayland@ilande.co.uk>
Copy bootinfo.h and bootinfo-mac.h from arch/m68k/include/uapi/asm/
to include/standard-headers/asm-m68k/
Imported from linux v5.9 but didn't change since v4.14 (header update)
and since v4.10 (content update).
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20201220112615.933036-2-laurent@vivier.eu>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Add the vmstate for the new NeXTPC devic; this is in theory
a migration compatibility break, but this machine doesn't have
working migration currently anyway.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20210115201206.17347-11-peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
The fields scsi_irq, scsi_dma, scsi_reset and fd_irq in
NeXTState are all unused, except in commented out
"this should do something like this" code. Remove the
unused fields. As and when the functionality that might
use them is added, we can put in the correct kind of
wiring (which might or might not need to be a qemu_irq,
but which in any case will need to be in the NeXTPC
device, not in NeXTState).
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20210115201206.17347-10-peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Move the rtc into the NeXTPC struct. Since this is the last
use of the 'backdoor' NextState pointer we can now remove that.
Probably the RTC should be its own device at some point: in hardware
there is a separate MCS1850 RTC chip connected to the Peripheral
Controller via a 1-bit serial interface. That goes beyond the remit
of the current refactoring, though.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20210115201206.17347-9-peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Make the next_irq function be GPIO inputs to the NEXT_PC
device, rather than a freestanding set of qemu_irq lines.
This fixes a minor Coverity issue where it correctly points
out the trivial memory leak of the memory allocated in the
call to qemu_allocate_irqs().
Fixes: CID 1421962
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20210115201206.17347-8-peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
All the code which accesses int_status and int_mask is now doing
so via the NeXTPC->NeXTState indirection, so we can move these
fields into the NeXTPC struct where they belong.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20210115201206.17347-7-peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Make the next_irq function take a NeXTPC* as its opaque rather than
the M68kCPU*. This will make it simpler to turn the next_irq
function into a gpio input line of the NeXTPC device in the next
commit.
For this to work we have to pass the CPU to the NeXTPC device via a
link property, in the same way we do in q800.c (and for the same
reason).
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20210115201206.17347-6-peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Move the registers handled by the scr_ops struct into the NeXTPC
device.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20210115201206.17347-5-peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Move the registers handled by the mmio_ops struct into the NeXTPC
device. This allows us to also move the scr1 and scr2 data fields.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20210115201206.17347-4-peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Currently the next-cube board code open-codes a lot of handling of
interrupts and some miscellaneous registers. Move this into a proper
QOM device.
In the real hardware this functionality seems to be the
responsibility of the Peripheral Controller (PC) chip, so name the
device that.
There are several different things that will need to be moved into
this device:
* the mmio_iops register set
* the scr_ops register set
* the next_irq IRQ handling
To ease review, we structure the change as a sequence of commits: in
this first commit we create the skeleton of the NeXTPC device with no
content, but with a backdoor pointer to the NeXTState machine's state
struct so we can move parts of the code and still have refactored and
non-refactored code using the same struct data fields. Further
commits will move functionality into the new device piece by piece.
At the end we will be able to remove the backdoor pointer because all
the data fields will be in the NeXTPC struct and not the NeXTState
struct.
We'll add the VMState for the new device at the end of all that; this
is in theory a migration compatibility break but this machine does
not currently support migration at all anyway.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20210115201206.17347-3-peter.maydell@linaro.org>
[huth: Add a comment in front of struct NeXTPC]
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
The next_irq() function is global, but isn't actually used anywhere
outside next-cube.c. Make it static.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20210115201206.17347-2-peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
The handling of the GLUE (General Logic Unit) device is
currently open-coded. Make this into a proper QOM device.
This minor piece of modernisation gets rid of the free
floating qemu_irq array 'pic', which Coverity points out
is technically leaked when we exit the machine init function.
(The replacement glue device is not leaked because it gets
added to the sysbus, so it's accessible via that.)
Fixes: Coverity CID 1421883
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Laurent vivier <laurent@vivier.eu>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20201106235109.7066-3-peter.maydell@linaro.org>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
The q800 board code connects both of the IRQ outputs of the ESCC
to the same pic[3] qemu_irq. Connecting two qemu_irqs outputs directly
to the same input is not valid as it produces subtly wrong behaviour
(for instance if both the IRQ lines are high, and then one goes
low, the PIC input will see this as a high-to-low transition
even though the second IRQ line should still be holding it high).
This kind of wiring needs an explicitly created OR gate; add one.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Message-Id: <20201106235109.7066-2-peter.maydell@linaro.org>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Use the machine properties instead.
Cc: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Coverity points out that the realize function for the TYPE_MCF5206_MBAR
device leaks the IRQ array it allocates with qemu_allocate_irqs().
Keep a pointer to it in the device state struct to avoid the leak.
(Since it needs to stay around for the life of the simulation there
is no need to actually free it, and the leak was harmless.)
Fixes: Coverity CID 1432412
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Thomas Huth <huth@tuxfamily.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20201120172314.14725-1-peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Make the type checking macro name consistent with the TYPE_*
constant.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Hervé Poussineau <hpoussin@reactos.org>
Message-Id: <20200902224311.1321159-40-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Some typedefs and macros are defined after the type check macros.
This makes it difficult to automatically replace their
definitions with OBJECT_DECLARE_TYPE.
Patch generated using:
$ ./scripts/codeconverter/converter.py -i \
--pattern=QOMStructTypedefSplit $(git grep -l '' -- '*.[ch]')
which will split "typdef struct { ... } TypedefName"
declarations.
Followed by:
$ ./scripts/codeconverter/converter.py -i --pattern=MoveSymbols \
$(git grep -l '' -- '*.[ch]')
which will:
- move the typedefs and #defines above the type check macros
- add missing #include "qom/object.h" lines if necessary
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20200831210740.126168-9-ehabkost@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Message-Id: <20200831210740.126168-10-ehabkost@redhat.com>
Message-Id: <20200831210740.126168-11-ehabkost@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
The mcf5206 system integration module should be a proper device.
Let's finally QOMify it.
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Message-Id: <20200819065201.4045-1-huth@tuxfamily.org>
Each architecture's sourceset is placed in an hw_arch dictionary, and picked up
from there when building the per-emulator static_library.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The object_property_set_FOO() setters take property name and value in
an unusual order:
void object_property_set_FOO(Object *obj, FOO_TYPE value,
const char *name, Error **errp)
Having to pass value before name feels grating. Swap them.
Same for object_property_set(), object_property_get(), and
object_property_parse().
Convert callers with this Coccinelle script:
@@
identifier fun = {
object_property_get, object_property_parse, object_property_set_str,
object_property_set_link, object_property_set_bool,
object_property_set_int, object_property_set_uint, object_property_set,
object_property_set_qobject
};
expression obj, v, name, errp;
@@
- fun(obj, v, name, errp)
+ fun(obj, name, v, errp)
Chokes on hw/arm/musicpal.c's lcd_refresh() with the unhelpful error
message "no position information". Convert that one manually.
Fails to convert hw/arm/armsse.c, because Coccinelle gets confused by
ARMSSE being used both as typedef and function-like macro there.
Convert manually.
Fails to convert hw/rx/rx-gdbsim.c, because Coccinelle gets confused
by RXCPU being used both as typedef and function-like macro there.
Convert manually. The other files using RXCPU that way don't need
conversion.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20200707160613.848843-27-armbru@redhat.com>
[Straightforwad conflict with commit 2336172d9b "audio: set default
value for pcspk.iobase property" resolved]
hw_error() dumps the CPU state and exits QEMU. This is ok during initial
code development (to see where the guest code is currently executing),
but it is certainly not the desired behavior that we want to present to
normal users, and it can also cause trouble when e.g. fuzzing devices.
Thus let's replace these hw_error()s by qemu_log_mask()s instead.
Message-Id: <20200611055807.15921-1-huth@tuxfamily.org>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
qdev_prop_set_drive() can fail. None of the other qdev_prop_set_FOO()
can; they abort on error.
To clean up this inconsistency, rename qdev_prop_set_drive() to
qdev_prop_set_drive_err(), and create a qdev_prop_set_drive() that
aborts on error.
Coccinelle script to update callers:
@ depends on !(file in "hw/core/qdev-properties-system.c")@
expression dev, name, value;
symbol error_abort;
@@
- qdev_prop_set_drive(dev, name, value, &error_abort);
+ qdev_prop_set_drive(dev, name, value);
@@
expression dev, name, value, errp;
@@
- qdev_prop_set_drive(dev, name, value, errp);
+ qdev_prop_set_drive_err(dev, name, value, errp);
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200622094227.1271650-14-armbru@redhat.com>
This is the transformation explained in the commit before previous.
Takes care of just one pattern that needs conversion. More to come in
this series.
Coccinelle script:
@ depends on !(file in "hw/arm/highbank.c")@
expression bus, type_name, dev, expr;
@@
- dev = qdev_create(bus, type_name);
+ dev = qdev_new(type_name);
... when != dev = expr
- qdev_init_nofail(dev);
+ qdev_realize_and_unref(dev, bus, &error_fatal);
@@
expression bus, type_name, dev, expr;
identifier DOWN;
@@
- dev = DOWN(qdev_create(bus, type_name));
+ dev = DOWN(qdev_new(type_name));
... when != dev = expr
- qdev_init_nofail(DEVICE(dev));
+ qdev_realize_and_unref(DEVICE(dev), bus, &error_fatal);
@@
expression bus, type_name, expr;
identifier dev;
@@
- DeviceState *dev = qdev_create(bus, type_name);
+ DeviceState *dev = qdev_new(type_name);
... when != dev = expr
- qdev_init_nofail(dev);
+ qdev_realize_and_unref(dev, bus, &error_fatal);
@@
expression bus, type_name, dev, expr, errp;
symbol true;
@@
- dev = qdev_create(bus, type_name);
+ dev = qdev_new(type_name);
... when != dev = expr
- object_property_set_bool(OBJECT(dev), true, "realized", errp);
+ qdev_realize_and_unref(dev, bus, errp);
@@
expression bus, type_name, expr, errp;
identifier dev;
symbol true;
@@
- DeviceState *dev = qdev_create(bus, type_name);
+ DeviceState *dev = qdev_new(type_name);
... when != dev = expr
- object_property_set_bool(OBJECT(dev), true, "realized", errp);
+ qdev_realize_and_unref(dev, bus, errp);
The first rule exempts hw/arm/highbank.c, because it matches along two
control flow paths there, with different @type_name. Covered by the
next commit's manual conversions.
Missing #include "qapi/error.h" added manually.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200610053247.1583243-10-armbru@redhat.com>
[Conflicts in hw/misc/empty_slot.c and hw/sparc/leon3.c resolved]
hw_error() calls exit(). This a bit overkill when we can log
the accesses as unimplemented or guest error.
When fuzzing the devices, we don't want the whole process to
exit. Replace some hw_error() calls by qemu_log_mask().
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20200526094052.1723-3-f4bug@amsat.org>
Reviewed-by: Thomas Huth <huth@tuxfamily.org>
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
All calls to m5206_mbar_read/m5206_mbar_write are used with
'offset = hwaddr & 0x3ff', so we are sure the offset fits
in 16-bit.
Suggested-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20200526094052.1723-2-f4bug@amsat.org>
Signed-off-by: Thomas Huth <huth@tuxfamily.org>
This commit was produced with the Coccinelle script
scripts/coccinelle/memory-region-housekeeping.cocci.
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Description copied from Linux kernel commit from Gustavo A. R. Silva
(see [3]):
--v-- description start --v--
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to
declare variable-length types such as these ones is a flexible
array member [1], introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler
warning in case the flexible array does not occur last in the
structure, which will help us prevent some kind of undefined
behavior bugs from being unadvertenly introduced [2] to the
Linux codebase from now on.
--^-- description end --^--
Do the similar housekeeping in the QEMU codebase (which uses
C99 since commit 7be41675f7).
All these instances of code were found with the help of the
following Coccinelle script:
@@
identifier s, m, a;
type t, T;
@@
struct s {
...
t m;
- T a[0];
+ T a[];
};
@@
identifier s, m, a;
type t, T;
@@
struct s {
...
t m;
- T a[0];
+ T a[];
} QEMU_PACKED;
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=76497732932f
[3] https://git.kernel.org/pub/scm/linux/kernel/git/gustavoars/linux.git/commit/?id=17642a2fbd2c1
Inspired-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
There's no good reason for it to be type int, change it to bool.
Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200207161948.15972-3-philmd@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
The MachineClass is already zeroed on creation.
Note: The code setting is_default=0 in hw/i386/pc_piix.c is
different (related to compat options). When adding a
new versioned machine, we want it to be the new default,
so we have to mark the previous one as not default.
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200207161948.15972-2-philmd@redhat.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
This series removes ad hoc RAM allocation API (memory_region_allocate_system_memory)
and consolidates it around hostmem backend. It allows to
* resolve conflicts between global -mem-prealloc and hostmem's "policy" option,
fixing premature allocation before binding policy is applied
* simplify complicated memory allocation routines which had to deal with 2 ways
to allocate RAM.
* reuse hostmem backends of a choice for main RAM without adding extra CLI
options to duplicate hostmem features. A recent case was -mem-shared, to
enable vhost-user on targets that don't support hostmem backends [1] (ex: s390)
* move RAM allocation from individual boards into generic machine code and
provide them with prepared MemoryRegion.
* clean up deprecated NUMA features which were tied to the old API (see patches)
- "numa: remove deprecated -mem-path fallback to anonymous RAM"
- (POSTPONED, waiting on libvirt side) "forbid '-numa node,mem' for 5.0 and newer machine types"
- (POSTPONED) "numa: remove deprecated implicit RAM distribution between nodes"
Introduce a new machine.memory-backend property and wrapper code that aliases
global -mem-path and -mem-alloc into automatically created hostmem backend
properties (provided memory-backend was not set explicitly given by user).
A bulk of trivial patches then follow to incrementally convert individual
boards to using machine.memory-backend provided MemoryRegion.
Board conversion typically involves:
* providing MachineClass::default_ram_size and MachineClass::default_ram_id
so generic code could create default backend if user didn't explicitly provide
memory-backend or -m options
* dropping memory_region_allocate_system_memory() call
* using convenience MachineState::ram MemoryRegion, which points to MemoryRegion
allocated by ram-memdev
On top of that for some boards:
* missing ram_size checks are added (typically it were boards with fixed ram size)
* ram_size fixups are replaced by checks and hard errors, forcing user to
provide correct "-m" values instead of ignoring it and continuing running.
After all boards are converted, the old API is removed and memory allocation
routines are cleaned up.
memory_region_allocate_system_memory() API is going away, so
replace it with memdev allocated MemoryRegion. The later is
initialized by generic code, so board only needs to opt in
to memdev scheme by providing
MachineClass::default_ram_id
and using MachineState::ram instead of manually initializing
RAM memory region.
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Acked-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200219160953.13771-50-imammedo@redhat.com>
Mark Cave-Ayland
David Hildenbrand
Thomas Huth
Laurent Vivier
Peter Maydell
Paolo Bonzini
Eduardo Habkost
Thomas Huth
Marc-André Lureau
Markus Armbruster
Philippe Mathieu-Daudé
Philippe Mathieu-Daudé
Igor Mammedov