Thomas Huth [Sat, 30 Jun 2018 14:58:01 +0000 (16:58 +0200)]
m68k: Add serial controller to the NeXTcube machine
The NeXTcube uses a normal 8530 serial controller, so we can simply use
our normal "escc" device here.
While we're at it, also add a boot-serial-test for the next-cube machine,
now that the serial output works.
Thomas Huth [Sat, 30 Jun 2018 06:45:25 +0000 (08:45 +0200)]
m68k: Add NeXTcube machine
It is still quite incomplete (no SCSI, no floppy emulation, no network,
etc.), but the firmware already shows up the debug monitor prompt in the
framebuffer display, so at least the very basics are already working.
This code has been taken from Bryce Lanham's GSoC 2011 NeXT branch at
Thomas Huth [Sat, 30 Jun 2018 05:50:23 +0000 (07:50 +0200)]
m68k: Add NeXTcube keyboard device
It is likely still quite incomplete (e.g. mouse and interrupts are not
implemented yet), but it is good enough for keyboard input at the firmware
monitor.
This code has been taken from Bryce Lanham's GSoC 2011 NeXT branch at
Thomas Huth [Sat, 30 Jun 2018 05:19:42 +0000 (07:19 +0200)]
m68k: Add NeXTcube framebuffer device emulation
The NeXTcube uses a linear framebuffer with 4 greyscale colors and
a fixed resolution of 1120 * 832.
This code has been taken from Bryce Lanham's GSoC 2011 NeXT branch at
Peter Maydell [Thu, 5 Sep 2019 16:09:13 +0000 (17:09 +0100)]
Merge remote-tracking branch 'remotes/huth-gitlab/tags/pull-request-2019-09-05-v2' into staging
- Make the core libqtest library independent from global_qtest
- Clean up docs from hard-coded qemu-system-* names
- Install libattr-dev and libcap-dev in gitlab-ci to test virtio-9p
* remotes/huth-gitlab/tags/pull-request-2019-09-05-v2:
gitlab-ci.yml: Install libattr-devel and libcap-devel to test virtio-9p
qemu-doc: Do not hard-code the name of the QEMU binary
tests/vm: Take the J=x setting into account for the vm-boot-ssh targets, too
tests/libqtest: Use libqtest-single.h in tests that require global_qtest
tests/libqtest: Move global_test wrapper function into a separate header
tests: Remove unnecessary global_qtest references
tests/libqos: Replace clock_step with qtest_clock_step in virtio code
tests/libqos/e1000e: Make e1000e libqos functions independent from global_qtest
tests/migration: Do not use functions anymore that rely on global_qtest
Peter Maydell [Thu, 5 Sep 2019 15:33:39 +0000 (16:33 +0100)]
Merge remote-tracking branch 'remotes/berrange/tags/docs-pull-request' into staging
docs: add docs about use of automatic cleanup functions
This is ostensibly about adding docs for the g_autofree/g_autoptr
macros. As part of doing that, however, the existing HACKING doc
is merged into the CODING_STYLE doc and the text is converted to
rst with a table of contents.
# gpg: Signature made Thu 05 Sep 2019 14:43:44 BST
# gpg: using RSA key DAF3A6FDB26B62912D0E8E3FBE86EBB415104FDF
# gpg: Good signature from "Daniel P. Berrange <[email protected]>" [full]
# gpg: aka "Daniel P. Berrange <[email protected]>" [full]
# Primary key fingerprint: DAF3 A6FD B26B 6291 2D0E 8E3F BE86 EBB4 1510 4FDF
* remotes/berrange/tags/docs-pull-request:
docs: split the CODING_STYLE doc into distinct groups
docs: document use of automatic cleanup functions in glib
docs: merge HACKING.rst contents into CODING_STYLE.rst
docs: convert README, CODING_STYLE and HACKING to RST syntax
Thomas Huth [Thu, 5 Sep 2019 10:36:50 +0000 (12:36 +0200)]
gitlab-ci.yml: Install libattr-devel and libcap-devel to test virtio-9p
So far the gitlab-ci was not testing virtio-9p yet, since we did not
install libattr-devel and libcap-devel in any of the pipelines. Do
it now to get some more test coverage.
There are only two remaining uses of gen_bx_im. In each case, we
know the destination mode -- not changing in the case of gen_jmp
or changing in the case of trans_BLX_i. Use this to simplify the
surrounding code.
For trans_BLX_i, use gen_jmp for the actual branch. For gen_jmp,
use gen_set_pc_im to set up the single-step.
We have been using store_reg and not store_reg_for_load when writing
back a loaded value into the base register. At first glance this is
incorrect when base == pc, however that case is UNPREDICTABLE.
Pass the T5 encoding of SUBS PC, LR, #IMM through the normal SUBS path
to make it clear exactly what's happening -- we hit ALUExceptionReturn
along that path.
The m-profile and a-profile decodings overlap. Only return false
for the case of wrong profile; handle UNDEFINED for permission failure
directly. This ensures that we don't accidentally pass an insn that
applies to the wrong profile.
Convert the modified immediate form of the data processing insns.
For A32, we can finally remove any code that was intertwined with
the register and register-shifted-register forms.
target/arm: Convert Data Processing (reg-shifted-reg)
Convert the register shifted by register form of the data
processing insns. For A32, we cannot yet remove any code
because the legacy decoder intertwines the immediate form.
Convert the register shifted by immediate form of the data
processing insns. For A32, we cannot yet remove any code
because the legacy decoder intertwines the reg-shifted-reg
and immediate forms.
target/arm: Use store_reg_from_load in thumb2 code
This function already includes the test for an interworking write
to PC from a load. Change the T32 LDM implementation to match the
A32 LDM implementation.
For LDM, the reordering of the tests does not change valid
behaviour because the only case that differs is has rn == 15,
which is UNPREDICTABLE.
Thomas Huth [Tue, 30 Jul 2019 15:08:26 +0000 (17:08 +0200)]
qemu-doc: Do not hard-code the name of the QEMU binary
In our documentation, we use a mix of "$QEMU", "qemu-system-i386" and
"qemu-system-x86_64" when we give examples to the users how to run
QEMU. Some more consistency would be good here. Also some distributions
use different names for the QEMU binary (e.g. "qemu-kvm" in RHEL), so
providing more flexibility here would also be good. Thus let's define
some variables for the names of the QEMU command and use those in the
documentation instead: @value{qemu_system} for generic examples, and
@value{qemu_system_x86} for examples that only work with the x86
binaries.
Thomas Huth [Wed, 24 Jul 2019 09:58:48 +0000 (11:58 +0200)]
tests/vm: Take the J=x setting into account for the vm-boot-ssh targets, too
For testing whether the VMs can deal with multiple CPUs correctly,
it is useful to be able to use the "J=<cpus>" setting for the
vm-boot-ssh targets, too.
Thomas Huth [Tue, 3 Sep 2019 05:50:26 +0000 (07:50 +0200)]
tests/libqtest: Use libqtest-single.h in tests that require global_qtest
Tests that require global_qtest or the related wrapper functions now
use the libqtest-single.h header that is dedicated for everything
related to global_qtest. The core libqtest.c and libqtest.h files are
now completely indepedent from global_qtest, so that the core library
is now not depending on a global state anymore.
Thomas Huth [Tue, 3 Sep 2019 05:39:37 +0000 (07:39 +0200)]
tests/libqtest: Move global_test wrapper function into a separate header
We want libqtest.h to become completely independent from global_qtest
(so that the wrapper functions are not used by accident anymore). As
a first step, move the wrapper functions into a separate header file.
The new header is only included from libqtest.h for now, so that there
is no difference to the users of libqtest.h yet. In the next patch, we
will switch this, so that the users of the global_qtest-related
functions will be using libqtest-single.h directly and libqtest.h
becomes completely independent of this.
Thomas Huth [Tue, 20 Aug 2019 19:23:56 +0000 (21:23 +0200)]
tests: Remove unnecessary global_qtest references
We are going to remove global_qtest from the main libqtest library
soon, so tests that do not urgently need global_qtest anymore
should be cleaned from the unnecessary references.
Thomas Huth [Tue, 3 Sep 2019 06:18:46 +0000 (08:18 +0200)]
tests/libqos: Replace clock_step with qtest_clock_step in virtio code
Library functions should not rely on functions that require global_qtest
(since they might get used in tests that deal with multiple states).
Commit 1999a70a05ad603d ("Make generic virtio code independent from
global_qtest") already tried to clean the libqos virtio code, but I
missed to replace the clock_step() function. Thus change it now to
qtest_clock_step() instead.
The logic of the qvirtio_wait_config_isr() function is now pushed
to the virtio-mmio.c and virtio-pci.c files instead, since we can
get the QTestState here easily.
Thomas Huth [Tue, 27 Aug 2019 11:17:47 +0000 (13:17 +0200)]
tests/libqos/e1000e: Make e1000e libqos functions independent from global_qtest
libqos library functions should never depend on functions (like memread(),
memwrite() or clock_step()) that require global_qtest to be set, since
library functions might get used in qtests that track multiple states, too.
Thus let's replace the global_qtest-related functions with their independent
counterparts.
Thomas Huth [Tue, 20 Aug 2019 14:33:38 +0000 (16:33 +0200)]
tests/migration: Do not use functions anymore that rely on global_qtest
The migration tests deal with multiple test states, so we really should
not use functions here that rely on the single global_qtest variable.
Switch from qtest_start() to qtest_init() to make sure that global_qtest
is not set anymore. This also revealed a regression in the migrate()
function: It has once been converted to use the qtest_qmp() function,
but commit b5bbd3f315d686bd511 ("Clean up string interpolation into QMP,
part 2") accidentally reverted it back to qmp().
* remotes/rth/tags/pull-or1k-20190904:
target/openrisc: Update cpu "any" to v1.3
target/openrisc: Implement l.adrp
target/openrisc: Implement move to/from FPCSR
target/openrisc: Implement unordered fp comparisons
target/openrisc: Add support for ORFPX64A32
target/openrisc: Check CPUCFG_OF32S for float insns
target/openrisc: Fix lf.ftoi.s
target/openrisc: Add VR2 and AVR special processor registers
target/openrisc: Move VR, UPR, DMMCFGR, IMMCFGR to cpu init
target/openrisc: Make VR and PPC read-only
target/openrisc: Cache R0 in DisasContext
target/openrisc: Replace cpu register array with a function
target/openrisc: Add DisasContext parameter to check_r0_write
This is hardware support for double-precision floating-point using
pairs of 32-bit registers. Fix latent bugs in the heretofore unused
helper_itofd and helper_ftoid. Include the bit for cpu "any".
Change the default cpu for linux-user to "any".
These SPRs are read-only. The writes can simply be ignored,
as we already do for other read-only (or missing) registers.
There is no reason to mask the value in env->vr.
projects / qemu.git / log