On 18.03.2024 18:46, Nicholas Piggin wrote:
record makes an initial snapshot when the machine is created, to enable
reverse-debugging. Often the issue being debugged appears near the end of
the trace, so it is important for performance to keep snapshots close to
the end.
This implements a
Reviewed-by: Pavel Dovgalyuk
On 18.03.2024 18:46, Nicholas Piggin wrote:
The regular qemu_bh_schedule() calls result in non-deterministic
execution of the bh in record-replay mode, which causes replay failure.
Signed-off-by: Nicholas Piggin
---
hw/net/virtio-net.c | 11 ++-
1 file
On 12.03.2024 14:05, Nicholas Piggin wrote:
On Tue Mar 12, 2024 at 7:09 PM AEST, Pavel Dovgalyuk wrote:
This won't work, as needed. Announce timer can't be enabled, because
it is set in post_load function. Therefore announce callbacks break
the replay, when virtio-net is used with snapshots
This won't work, as needed. Announce timer can't be enabled, because
it is set in post_load function. Therefore announce callbacks break
the replay, when virtio-net is used with snapshots.
On 11.03.2024 20:40, Nicholas Piggin wrote:
Using virtual time for announce ensures that guest visible
On 11.03.2024 20:40, Nicholas Piggin wrote:
record makes an initial snapshot when the machine is created, to enable
reverse-debugging. Often the issue being debugged appears near the end of
the trace, so it is important for performance to keep snapshots close to
the end.
This implements a
Reviewed-by: Pavel Dovgalyuk
On 01.03.2024 15:06, Markus Armbruster wrote:
Improve
Record/replay feature is not supported for '-rtc base=localtime'
Record/replay feature is not supported for 'smp'
Record/replay feature is not supported for '-snapshot'
to
Record/replay
On 26.02.2024 10:36, Nicholas Piggin wrote:
On Fri Aug 18, 2023 at 2:36 PM AEST, Pavel Dovgalyuk wrote:
On 14.08.2023 19:31, Nicholas Piggin wrote:
record makes an initial snapshot when the machine is created, to enable
reverse-debugging. Often the issue being debugged appears near the end
Reviewed-by: Pavel Dovgalyuk
On 20.02.2024 11:52, Manos Pitsidianakis wrote:
Correct typos automatically found with the `typos` tool
<https://crates.io/crates/typos>
Signed-off-by: Manos Pitsidianakis
---
tests/avocado/acpi-bits/bits-tests/smbios.py2 | 2 +-
tests/avocado/mem-addr
Reviewed-by: Pavel Dovgalyuk
On 25.01.2024 19:08, Nicholas Piggin wrote:
This runs replay-dump.py after recording a trace, and fails the test if
the script fails.
replay-dump.py is modified to exit with non-zero if an error is
encountered while parsing, to support this.
Signed-off
Reviewed-by: Pavel Dovgalyuk
On 25.01.2024 19:08, Nicholas Piggin wrote:
When replaying a trace, it is possible to go from shutdown to
running with a reverse-debugging step. This can be useful if the
problem being debugged triggers a reset or shutdown.
Signed-off-by: Nicholas Piggin
Reviewed-by: Pavel Dovgalyuk
On 24.01.2024 13:16, Philippe Mathieu-Daudé wrote:
Move this x86-specific code out of the generic accel/tcg/.
Signed-off-by: Philippe Mathieu-Daudé
---
target/i386/tcg/helper-tcg.h| 1 +
accel/tcg/cpu-exec.c| 9 -
target/i386
Reviewed-by: Pavel Dovgalyuk
On 24.01.2024 13:16, Philippe Mathieu-Daudé wrote:
In order to make accel/tcg/ target agnostic,
introduce the need_replay_interrupt() handler.
Signed-off-by: Philippe Mathieu-Daudé
---
include/hw/core/tcg-cpu-ops.h | 5 +
accel/tcg/cpu-exec.c | 5
most of the failures in replay_kernel.py
Is there an effect for console QEMU only?
I've tested this patch on Windows7 boot scenario and replay speed has
not changed.
Fixes: https://gitlab.com/qemu-project/qemu/-/issues/2013
Signed-off-by: Alex Bennée
Cc: Pavel Dovgalyuk
---
include
@@ def test_or1k_sim(self):
file_path = self.fetch_asset(tar_url, asset_hash=tar_hash)
self.do_test_advcal_2018(file_path, 'vmlinux')
-@skip("nios2 emulation is buggy under record/replay")
def test_nios2_10m50(self):
"""
:avocado: tags=arch:nios2
Acked-by: Pavel Dovgalyuk
, because RR may be started from the snapshot,
which references the middle of replayed scenario.
Pavel Dovgalyuk
/qemu/-/issues/2010
Signed-off-by: Alex Bennée
Cc: Pavel Dovgalyuk
---
chardev/char.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/chardev/char.c b/chardev/char.c
index 996a024c7a..6e5b4d7345 100644
--- a/chardev/char.c
+++ b/chardev/char.c
@@ -171,7 +171,8 @@ int
deletions(-)
Reviewed-by: Pavel Dovgalyuk
overlapping ids. */
uint64_t block_request_id;
-/*! Asynchronous event id read from the log */
uint64_t read_event_id;
} ReplayState;
extern ReplayState replay_state;
Reviewed-by: Pavel Dovgalyuk
clock */
-uint64_t host_clock_last;
/*! Asynchronous event id read from the log */
uint64_t read_event_id;
} ReplayState;
Reviewed-by: Pavel Dovgalyuk
):
+global total_insns
ins_diff = read_dword(dumpfile)
-print_event(eid, name, "0x%x" % (ins_diff))
+total_insns += ins_diff
+print_event(eid, name, "+ %d -> %d" % (ins_diff, total_insns))
return True
def decode_char_write(eid, name, dumpfile):
Reviewed-by: Pavel Dovgalyuk
1 file changed, 16 insertions(+)
Acked-by: Pavel Dovgalyuk
('successfully finished the replay')
elapsed = time.time() - start_time
logger.info('elapsed time %.2f sec' % elapsed)
return elapsed
Reviewed-by: Pavel Dovgalyuk
On 11.11.2023 17:27, Akihiko Odaki wrote:
On 2023/11/10 16:35, Pavel Dovgalyuk wrote:
You need to bump REPLAY_VERSION in replay/replay.c, because your patch
changes the replay log format.
Otherwise, for replay part:
Acked-by: Pavel Dovgalyuk
I'll drop this change then. It's just a cleanup
You need to bump REPLAY_VERSION in replay/replay.c, because your patch
changes the replay log format.
Otherwise, for replay part:
Acked-by: Pavel Dovgalyuk
On 30.10.2023 08:12, Akihiko Odaki wrote:
There is no defined flag now.
Signed-off-by: Akihiko Odaki
---
include/net/filter.h
From: Pavel Dovgalyuk
The set of patches include the following modifications:
- fix for allowing record/replay with virtio-net
- fix for loading non-replay snapshots
Pavel Dovgalyuk (3):
replay: improve determinism of virtio-net
virtio-net: added replay blocker for guest_announce
replay
From: Pavel Dovgalyuk
virtio-net device uses bottom halves for callbacks.
These callbacks should be deterministic, because they affect VM state.
This patch replaces BH invocations with corresponding replay functions,
making them deterministic in record/replay mode.
Signed-off-by: Pavel
From: Pavel Dovgalyuk
Snapshots created in regular icount execution mode can't be loaded
in recording mode, because icount value advances only by 32-bit value.
This patch initializes replay icount initial value after loading
the snapshot.
Cc: Pizarro Solar Rafael Ulises Luzius
Signed-off
From: Pavel Dovgalyuk
This patch blocks record/replay when guest_announce is enabled,
because this flag breaks loading the snapshots in deterministic
execution mode.
Signed-off-by: Pavel Dovgalyuk
---
hw/net/virtio-net.c | 4
1 file changed, 4 insertions(+)
diff --git a/hw/net/virtio
On 11.08.2023 04:47, Nicholas Piggin wrote:
RR CPU switching is driven by timers and events so it is deterministic
like everything else. Record a CPU switch event and use that to drive
the CPU switch on replay.
Signed-off-by: Nicholas Piggin
---
This is still in RFC phase because so far I've
On 14.08.2023 19:31, Nicholas Piggin wrote:
record makes an initial snapshot when the machine is created, to enable
reverse-debugging. Often the issue being debugged appears near the end of
the trace, so it is important for performance to keep snapshots close to
the end.
This implements a
Acked-by: Pavel Dovgalyuk
On 14.08.2023 19:31, Nicholas Piggin wrote:
When replaying a trace, it is possible to go from shutdown to
running with a reverse-debugging step. This can be useful if the
problem being debugged triggers a reset or shutdown.
Signed-off-by: Nicholas Piggin
On 14.08.2023 19:31, Nicholas Piggin wrote:
This runs replay-dump.py after recording a trace, and fails the test if
the script fails.
replay-dump.py is modified to exit with non-zero if an error is
encountered while parsing.
I would like to have separate test for replay-dump, because
On 14.08.2023 19:31, Nicholas Piggin wrote:
This thing seems to have fallen by the wayside. This gets it working with
the current format, although does not quite implement all events.
Signed-off-by: Nicholas Piggin
The code looks ok, therefore
Rewieved-by: Pavel Dovgalyuk
However
From: Pavel Dovgalyuk
virtio-net device uses bottom halves for callbacks.
These callbacks should be deterministic, because they affect VM state.
This patch replaces BH invocations with corresponding replay functions,
making them deterministic in record/replay mode.
Signed-off-by: Pavel
From: Pavel Dovgalyuk
This patch fixes the race condition in waiting for shutdown
of the replay linux test.
Signed-off-by: Pavel Dovgalyuk
Suggested-by: John Snow
---
tests/avocado/replay_linux.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/avocado
From: Pavel Dovgalyuk
The set of patches include the following modifications:
- fix for allowing record/replay with virtio-net
- fix of the record/replay test
Pavel Dovgalyuk (3):
replay: improve determinism of virtio-net
virtio-net: added replay blocker for guest_announce
tests/avocado
From: Pavel Dovgalyuk
This patch blocks record/replay when guest_announce is enabled,
because this flag breaks loading the snapshots in deterministic
execution mode.
Signed-off-by: Pavel Dovgalyuk
---
hw/net/virtio-net.c | 4
1 file changed, 4 insertions(+)
diff --git a/hw/net/virtio
On 08.08.2023 06:09, Nicholas Piggin wrote:
On Sun Aug 6, 2023 at 9:46 PM AEST, Nicholas Piggin wrote:
On Fri Aug 4, 2023 at 6:50 PM AEST, Pavel Dovgalyuk wrote:
BTW, there is a function qemu_register_reset_nosnapshotload that can be
used in similar cases.
Can you just use it without changing
with. When loading a snapshot for record-replay, the machine
is reset again, and that tries to consume the random event record
again, crashing due to inconsistent record
Fix this by saving the seed to populate the device tree with, and
skipping the rng on snapshot load.
Cc: Pavel Dovgalyuk
Signed-off
Reviewed-by: Pavel Dovgalyuk
On 26.07.2023 21:35, Nicholas Piggin wrote:
These machines run reverse-debugging well enough to pass basic tests.
Wire them up.
Cc: Pavel Dovgalyuk
Signed-off-by: Nicholas Piggin
---
tests/avocado/reverse_debugging.py | 29 +
1
Reviewed-by: Pavel Dovgalyuk
On 26.07.2023 21:35, Nicholas Piggin wrote:
The reverse-debugging test creates a trace, then replays it and:
1. Steps the first 10 instructions and records their addresses.
2. Steps backward and verifies their addresses match.
3. Runs to (near) the end
Acked-by: Pavel Dovgalyuk
On 26.07.2023 21:35, Nicholas Piggin wrote:
This the ppc64 record-replay test is able to replay the full kernel boot
so try enabling it.
Cc: Pavel Dovgalyuk
Signed-off-by: Nicholas Piggin
---
tests/avocado/replay_kernel.py | 3 +--
1 file changed, 1 insertion
Acked-by: Pavel Dovgalyuk
On 26.07.2023 21:35, Nicholas Piggin wrote:
spapr_machine_reset gets a random number to populate the device-tree
rng seed with. When loading a snapshot for record-replay, the machine
is reset again, and that tries to consume the random event record
again, crashing due
From: Pavel Dovgalyuk
Signed-off-by: Pavel Dovgalyuk
Suggested-by: John Snow
---
tests/avocado/replay_linux.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/avocado/replay_linux.py b/tests/avocado/replay_linux.py
index a76dd507fc..270ccc1eae 100644
--- a/tests
On 10.07.2023 20:14, Philippe Mathieu-Daudé wrote:
Hi,
Per tests/avocado/replay_kernel.py:
def test_aarch64_virt(self):
...
console_pattern = 'VFS: Cannot open root device'
the test is succeeding, but Avocado reports an error.
Does the emulator exit when error "Cannot
e500 has the same problem, I think, according to this issue:
https://gitlab.com/qemu-project/qemu/-/issues/1634
Btw, ARM virt platform rebuilds fdt only at initialization phase, not
when reset.
Isn't this behavior correct? Shouldn't PPC platforms do the similar thing?
On 23.06.2023 15:57,
Reviewed-by: Pavel Dovgalyuk
On 23.06.2023 15:57, Nicholas Piggin wrote:
Timebase save uses a random number for a legacy vmstate field, which
makes rr snapshot loading unbalanced. The easiest way to deal with this
is just to skip the rng if record-replay is active.
Signed-off-by: Nicholas
On 23.06.2023 15:57, Nicholas Piggin wrote:
pseries can run reverse-debugging well enough to pass basic tests.
There is strangeness with reverse-continue possibly relating to a bp
being set on the first instruction or on a snapshot, that causes
the PC to be reported on the first instruction
Acked-by: Pavel Dovgalyuk
On 23.06.2023 15:57, Nicholas Piggin wrote:
ppc only migrates reserve_addr, so the destination machine can get a
valid reservation with an incorrect reservation value of 0. Prior to
commit 392d328abe753 ("target/ppc: Ensure stcx size matches larx"),
this co
replay/replay-input.c part:
Reviewed-by: Pavel Dovgalyuk
On 18.02.2023 19:22, Sergio Lopez wrote:
Add a virtio-multitouch device to the family of devices emulated by
virtio-input implementing the Multi-touch protocol as descripted here:
https://www.kernel.org/doc/html/latest/input/multi-touch
in xmega controllers,
that could be added later. For the currently supported MCUs this
feature is disabled and RAMPZ is not used for RAM access.
Signed-off-by: Pavel Dovgalyuk
Reviewed-by: Richard Henderson
---
target/avr/cpu.h |2 ++
target/avr/translate.c | 63
of deposit tcg instructions.
Signed-off-by: Pavel Dovgalyuk
Reviewed-by: Richard Henderson
Reviewed-by: Michael Rolnik
---
target/avr/translate.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/target/avr/translate.c b/target/avr/translate.c
index 2bed56f135..552f739b3d
Interrupt bit vector has 64 bits, but interrupt vector is found with ctz32
function. This patch replaces it with ctz64.
Signed-off-by: Pavel Dovgalyuk
Reviewed-by: Philippe Mathieu-Daudé
Reviewed-by: Richard Henderson
---
target/avr/helper.c |4 ++--
1 file changed, 2 insertions(+), 2
This set of patches includes multiple changes for AVR target.
v2 changes:
- fixed instruction translation in icount mode
---
Pavel Dovgalyuk (5):
target/avr: fix long address calculation
target/avr: implement small RAM/large RAM feature
target/avr: fix avr features processing
-by: Pavel Dovgalyuk
Reviewed-by: Richard Henderson
---
target/avr/translate.c | 30 ++
1 file changed, 30 insertions(+)
diff --git a/target/avr/translate.c b/target/avr/translate.c
index 40b15d116e..ee137dfe54 100644
--- a/target/avr/translate.c
+++ b/target/avr
Bit vector for features has 64 bits. This patch fixes bit shifts in
avr_feature and set_avr_feature functions to be 64-bit too.
Signed-off-by: Pavel Dovgalyuk
Reviewed-by: Michael Rolnik
Reviewed-by: Philippe Mathieu-Daudé
Reviewed-by: Richard Henderson
---
target/avr/cpu.h |4 ++--
1
of deposit tcg instructions.
Signed-off-by: Pavel Dovgalyuk
Reviewed-by: Richard Henderson
Reviewed-by: Michael Rolnik
---
target/avr/translate.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/target/avr/translate.c b/target/avr/translate.c
index 2bed56f135..552f739b3d
in xmega controllers,
that could be added later. For the currently supported MCUs this
feature is disabled and RAMPZ is not used for RAM access.
Signed-off-by: Pavel Dovgalyuk
---
target/avr/cpu.h |2 ++
target/avr/translate.c | 63 ++--
2 files
Bit vector for features has 64 bits. This patch fixes bit shifts in
avr_feature and set_avr_feature functions to be 64-bit too.
Signed-off-by: Pavel Dovgalyuk
Reviewed-by: Michael Rolnik
---
target/avr/cpu.h |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/target/avr
This set of patches includes multiple changes for AVR target.
v2 changes:
- fixed instruction translation in icount mode
---
Pavel Dovgalyuk (5):
target/avr: fix long address calculation
target/avr: implement small RAM/large RAM feature
target/avr: fix avr features processing
On 19.01.2023 07:40, Hyeonggon Yoo wrote:
On Wed, Jan 18, 2023 at 12:39:16PM +0300, Pavel Dovgalyuk wrote:
Sometimes replay (or reverse debugging) have problems due to incomplete or
incorrect virtual device save/load implementation.
Can you try removing -cpu from your command line?
Or you can
Interrupt bit vector has 64 bits, but interrupt vector is found with ctz32
function. This patch replaces it with ctz64.
Signed-off-by: Pavel Dovgalyuk
---
target/avr/helper.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/target/avr/helper.c b/target/avr/helper.c
-by: Pavel Dovgalyuk
---
target/avr/translate.c | 30 ++
1 file changed, 30 insertions(+)
diff --git a/target/avr/translate.c b/target/avr/translate.c
index 40b15d116e..ee137dfe54 100644
--- a/target/avr/translate.c
+++ b/target/avr/translate.c
@@ -1406,6 +1406,10
, Jan 18, 2023 at 10:12:48AM +0300, Pavel Dovgalyuk wrote:
As replay works well, the reverse debugging should be ok too.
But for "going back" it needs a VM snapshot that can be used for reload.
Snapshots are saved on qcow2 images connected to QEMU.
Therefore you need to add an empty qco
As replay works well, the reverse debugging should be ok too.
But for "going back" it needs a VM snapshot that can be used for reload.
Snapshots are saved on qcow2 images connected to QEMU.
Therefore you need to add an empty qcow2 to your command line with the
following option: -drive
Reviewed-by: Pavel Dovgalyuk
On 19.12.2022 20:08, Philippe Mathieu-Daudé wrote:
From: Philippe Mathieu-Daude
replay API is used deeply within TCG common code (common to user
and system emulation). Unfortunately "sysemu/replay.h" requires
some QAPI headers for few system-specific de
Reviewed-by: Pavel Dovgalyuk
On 29.11.2022 04:05, Richard Henderson wrote:
Fixes the build with gcc 13:
replay/replay-time.c:34:6: error: conflicting types for \
'replay_read_next_clock' due to enum/integer mismatch; \
have 'void(ReplayClockKind)' [-Werror=enum-int-mismatch]
34
in xmega controllers,
that could be added later. For the currently supported MCUs this
feature is disabled and RAMPZ is not used for RAM access.
Signed-off-by: Pavel Dovgalyuk
---
target/avr/cpu.h |2 ++
target/avr/translate.c | 63 ++--
2 files
of deposit tcg instructions.
Signed-off-by: Pavel Dovgalyuk
Reviewed-by: Richard Henderson
Reviewed-by: Michael Rolnik
---
target/avr/translate.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/target/avr/translate.c b/target/avr/translate.c
index e65b6008c0..c9a0a39c2d
Interrupt bit vector has 64 bits, but interrupt vector is found with ctz32
function. This patch replaces it with ctz64.
Signed-off-by: Pavel Dovgalyuk
---
target/avr/helper.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/target/avr/helper.c b/target/avr/helper.c
Bit vector for features has 64 bits. This patch fixes bit shifts in
avr_feature and set_avr_feature functions to be 64-bit too.
Signed-off-by: Pavel Dovgalyuk
---
target/avr/cpu.h |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/target/avr/cpu.h b/target/avr/cpu.h
---
Pavel Dovgalyuk (4):
target/avr: fix long address calculation
target/avr: implement small RAM/large RAM feature
target/avr: fix avr features processing
target/avr: fix interrupt processing
target/avr/cpu.h | 6 ++--
target/avr/helper.c| 4 +--
target/avr
of deposit tcg instructions.
Signed-off-by: Pavel Dovgalyuk
---
target/avr/translate.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/target/avr/translate.c b/target/avr/translate.c
index e65b6008c0..c9a0a39c2d 100644
--- a/target/avr/translate.c
+++ b/target/avr
rom other make tasks.
Signed-off-by: Maksim Bakulin
Signed-off-by: Pavel Dovgalyuk
---
tests/Makefile.include | 20 ++--
tests/jobs.py | 42 ++
tests/requirements.txt |2 +-
3 files changed, 61 insertions(+), 3 deletions(-)
ber which one). It hangs with avocado, but does not hang when I run
it with the same command line without avocado.
It could be some replay issue (like infinite waiting for input in
main_loop_wait), but I couldn't trigger this behavior with
logging/debugging enabled.
Pavel Dovgalyuk
MU as well.
That's true. But there is one exception: LBX/LWX/LDX instruction.
These are grouped into DSP extension in QEMU, but Octeon supports them.
I've sent a patch for enabling these instructions.
Reviewed-by: Pavel Dovgalyuk
Signed-off-by: Jiaxun Yang
Acked-by: Richard Henderson
---
t
This patch changes condition and function name for enabling
indexed load instructions for Octeon vCPUs. Octeons do not
have DSP extension, but implement LBX-and-others.
Signed-off-by: Pavel Dovgalyuk
---
target/mips/tcg/translate.c | 10 +++---
1 file changed, 7 insertions(+), 3 deletions
Acked-by: Pavel Dovgalyuk
On 31.10.2022 16:25, Jiaxun Yang wrote:
As per "Cavium Networks OCTEON Plus CN50XX Hardware Reference
Manual" offset field is signed 16 bit value. However arg_BBIT.offset
is unsigned. We need to cast it as signed to do address calculation.
Signed-off-by: J
re external to the machine, they come from the outer world.
It means that this is completely new reset case and new solution should
be found for it.
Pavel Dovgalyuk
Last line of the test is missing by accident.
This patch fixes the script.
Signed-off-by: Pavel Dovgalyuk
---
tests/avocado/replay_linux.py |1 +
1 file changed, 1 insertion(+)
diff --git a/tests/avocado/replay_linux.py b/tests/avocado/replay_linux.py
index 40e4f6908e..e1f9981a34 100644
Tested-by: Pavel Dovgalyuk
On 01.08.2022 19:45, Peter Maydell wrote:
The function icount_prepare_for_run() is called with the iothread
unlocked, but it can call icount_notify_aio_contexts() which will
run qemu timer handlers. Those are supposed to be run only with
the iothread lock held, so
On 06.07.2022 23:53, Philippe Mathieu-Daudé wrote:
On 4/7/22 12:59, Pavel Dovgalyuk wrote:
ping
This is the only non-reviewed patch in the series.
I've been looking for doc/datasheets but no luck (except the Linux
kernel comments).
What kind of testing are you doing?
BTW, we found SDK
On 06.07.2022 23:53, Philippe Mathieu-Daudé wrote:
On 4/7/22 12:59, Pavel Dovgalyuk wrote:
ping
This is the only non-reviewed patch in the series.
I've been looking for doc/datasheets but no luck (except the Linux
kernel comments).
What kind of testing are you doing?
We compared
ping
This is the only non-reviewed patch in the series.
On 20.06.2022 15:05, Pavel Dovgalyuk wrote:
This patch adds Cavium Octeon 68XX vCPU which provides
Octeon-specific instructions.
Signed-off-by: Pavel Dovgalyuk
--
v3 changes:
- split the patch to instruction set introduction and new
This patch implements several Octeon-specific instructions:
- BADDU
- DMUL
- EXTS/EXTS32
- CINS/CINS32
- POP/DPOP
- SEQ/SEQI
- SNE/SNEI
Signed-off-by: Pavel Dovgalyuk
--
v3 changes:
- Fixed length field for EXTS/CINS
(bug found by Richard Henderson)
v2 changes:
- Using existing tcg
(suggested by Philippe Mathieu-Daudé)
- fixed length field for EXTS/CINS (bug found by Richard Henderson)
v2 changes:
- simplified instruction decoding and translation (suggested by Richard
Henderson)
---
Pavel Dovgalyuk (4):
target/mips: introduce decodetree structure for Cavium
This patch introduces Octeon-specific decoder and implements
check-bit-and-jump instructions.
Signed-off-by: Pavel Dovgalyuk
Reviewed-by: Richard Henderson
--
v3 changes:
- Split new decodetree and BBIT decoding into two patches
(suggested by Philippe Mathieu-Daudé)
v2 changes
(suggested by Philippe Mathieu-Daudé)
- fixed length field for EXTS/CINS (bug found by Richard Henderson)
v2 changes:
- simplified instruction decoding and translation (suggested by Richard
Henderson)
---
Pavel Dovgalyuk (2):
target/mips: introduce Cavium Octeon CPU model
target
This patch adds Cavium Octeon 68XX vCPU which provides
Octeon-specific instructions.
Signed-off-by: Pavel Dovgalyuk
--
v3 changes:
- split the patch to instruction set introduction and new vCPU
(suggested by Philippe Mathieu-Daudé)
v2 changes:
- vCPU name changed to Octeon68XX (suggested
This patch adds decodetree for Cavium Octeon extension and
an instruction set extension flag for using it in CPU models.
Signed-off-by: Pavel Dovgalyuk
---
target/mips/mips-defs.h|1 +
target/mips/tcg/meson.build|2 ++
target/mips/tcg/octeon.decode |6
On 09.06.2022 18:53, Richard Henderson wrote:
On 6/9/22 01:23, Pavel Dovgalyuk wrote:
+static bool trans_BADDU(DisasContext *ctx, arg_BADDU *a)
+{
+ TCGv t0, t1;
+
+ if (a->rt == 0) {
+ /* nop */
+ return true;
+ }
I believe that we're standardizing on us
This patch introduces Octeon-specific decoder and implements
check-bit-and-jump instructions.
Signed-off-by: Pavel Dovgalyuk
--
v2 changes:
- Changed insn field description and simplified the jumps
(suggested by Richard Henderson)
---
target/mips/tcg/meson.build|2
This patch adds Cavium Octeon vCPU for providing
Octeon-specific instructions.
Signed-off-by: Pavel Dovgalyuk
--
v2 changes:
- vCPU name changed to Octeon68XX (suggested by Richard Henderson)
---
target/mips/cpu-defs.c.inc | 28
target/mips/mips-defs.h|1
by Richard
Henderson)
---
Pavel Dovgalyuk (3):
target/mips: introduce Cavium Octeon CPU model
target/mips: implement Octeon-specific BBIT instructions
target/mips: implement Octeon-specific arithmetic instructions
target/mips/tcg/meson.build| 2 +
target/mips/tcg
This patch implements several Octeon-specific instructions:
- BADDU
- DMUL
- EXTS/EXTS32
- CINS/CINS32
- POP/DPOP
- SEQ/SEQI
- SNE/SNEI
Signed-off-by: Pavel Dovgalyuk
--
v2 changes:
- Using existing tcg instructions for exts, cins, pop
(suggested by Richard Henderson)
---
target/mips
On 07.06.2022 20:06, Richard Henderson wrote:
On 6/7/22 01:59, Pavel Dovgalyuk wrote:
+# Branch on bit set or clear
+# BBIT0 110010 . .
+# BBIT032 110110 . .
+# BBIT1 111010 . .
+# BBIT132 10
This patch implements several Octeon-specific instructions:
- BADDU
- DMUL
- EXTS/EXTS32
- CINS/CINS32
- POP/DPOP
- SEQ/SEQI
- SNE/SNEI
Signed-off-by: Pavel Dovgalyuk
---
target/mips/helper.h|1
target/mips/tcg/meson.build |1
target/mips/tcg/octeon.decode
This patch introduces Octeon-specific decoder and implements
check-bit-and-jump instructions.
Signed-off-by: Pavel Dovgalyuk
---
target/mips/tcg/meson.build|2 +
target/mips/tcg/octeon.decode | 14 ++
target/mips/tcg/octeon_translate.c | 53
This patch adds generic Octeon vCPU for providing
Octeon-specific instructions.
Signed-off-by: Pavel Dovgalyuk
---
target/mips/cpu-defs.c.inc | 30 ++
target/mips/mips-defs.h|1 +
2 files changed, 31 insertions(+)
diff --git a/target/mips/cpu-defs.c.inc b
The following series includes emulation of the platform-specific MIPS extension
for Cavium Octeon CPUS:
- basic Octeon vCPU model
- custom instruction decoder for Octeon
- implementation of arithmetic and logic instructions
---
Pavel Dovgalyuk (3):
target/mips: introduce generic Cavium
1 - 100 of 2324 matches
Mail list logo
