RE: [PATCH] vhost: Unbreak SMMU and virtio-iommu on dev-iotlb support

> From: Peter Xu > Sent: Friday, February 5, 2021 11:31 PM > > > > > > > > > >> or virtio-iommu > > >> since dev-iotlb (or PCIe ATS) > > > > > > > > > We may need to add this in the future. > > added Jean-Philippe in CC > > So that's the part I'm unsure about.. Since everybody is cced so maybe g

Re: Increased execution time with TCI in latest git master (was: Re: [PULL 00/46] tcg patch queue)

Am 07.02.21 um 04:45 schrieb Richard Henderson: On 2/6/21 11:38 AM, Stefan Weil wrote: I am still searching what caused this detoriation. My first suspect was thread local storage, but that wasn't it. Do you have any idea? No, but since it's 1/3 of a complete patch set, I don't care to investi

[PATCH v3] travis-ci: Disable C++ optional objects on AArch64 container

Travis-CI seems to have enforced memory limit on containers, and the 'GCC check-tcg' job started to fail on AArch64 [*]: [2041/3679] Compiling C++ object libcommon.fa.p/disas_nanomips.cpp.o FAILED: libcommon.fa.p/disas_nanomips.cpp.o {standard input}: Assembler messages: {standard input}:5

[PATCH] target/i386: expose more MSRs to GDB

This patch adds 7 more model-specific registers to be usable while remote debugging in gdb. Accessing these registers can for example be useful for tracing Linux Systemcalls. Signed-off-by: Dominik Glöß --- Adding registers to GDB like this works fine for now. Should there arise the need to add

A issue about qemu for rbd attach

Hi qemu developer, I have a question that does qume support cname as host to attach rbd device? When I want to do that, it return a error: /# virsh attach-device virtlet-228fa0ac-d53a-tess-node-c7nww disk_vde.yaml error: Failed to attach device from disk_vde.yaml error: internal error: unable to ex

Re: [PATCH] vhost: Unbreak SMMU and virtio-iommu on dev-iotlb support

Hi, Kevin, On Sun, Feb 07, 2021 at 09:04:55AM +, Tian, Kevin wrote: > > From: Peter Xu > > Sent: Friday, February 5, 2021 11:31 PM > > > > > > > > > > > > > >> or virtio-iommu > > > >> since dev-iotlb (or PCIe ATS) > > > > > > > > > > > > We may need to add this in the future. > > > added Jea

[PATCH 00/26] ppc: qemu: Convert qemu-ppce500 to driver model

At present when building qemu-ppce500 the following warnings are seen: = WARNING == This board does not use CONFIG_DM. CONFIG_DM will be compulsory starting with the v2020.01 release. Failure to update may result in board removal. UPD include/generated

Re: Help with Windows XP in qemu-system-i386

On Fri, Feb 05, 2021 at 04:08:26PM -0500, Programmingkid wrote: > > > > On Feb 5, 2021, at 3:49 PM, Michael S. Tsirkin wrote: > > > > On Fri, Feb 05, 2021 at 03:25:00PM -0500, Programmingkid wrote: > >> Hi, I'm noticing that my Windows XP Service Pack 3 VM is causing > >> qemu-system-i386 to e

Re: [PATCH 2/2] hw/ssi: xilinx_spips: Implement basic QSPI DMA support

Hi Peter, On Sat, Feb 6, 2021 at 11:28 PM Peter Maydell wrote: > > On Sat, 6 Feb 2021 at 14:38, Bin Meng wrote: > > > > From: Xuzhou Cheng > > > > ZynqMP QSPI supports SPI transfer using DMA mode, but currently this > > is unimplemented. When QSPI is programmed to use DMA mode, QEMU will > > cr

[PATCH v4 0/6] colo: Introduce resource agent and test suite/CI

Hello Everyone, So here is v4. Regards, Lukas Straub Changes: v4: -use new yank api that finally has been merged -cleanup the test a bit by using numbers instead of "hosta" and "hostb" -resource-agent: Don't set master-score to 0 on invalid configuration v3: -resource-agent: Don't determine

[PATCH v4 1/6] avocado_qemu: Introduce pick_qemu_util to pick qemu utility binaries

This introduces a generic function to pick qemu utility binaries from the build dir, system or via test parameter. Signed-off-by: Lukas Straub --- tests/acceptance/avocado_qemu/__init__.py | 15 +++ 1 file changed, 15 insertions(+) diff --git a/tests/acceptance/avocado_qemu/__init__

[PATCH v4 5/6] configure,Makefile: Install colo resource-agent

Optionally install the resouce-agent so it gets picked up by pacemaker. Signed-off-by: Lukas Straub --- configure | 7 +++ meson.build | 5 + meson_options.txt | 2 ++ 3 files changed, 14 insertions(+) diff --git a/configure b/configure index a34f91171d..54fc7e533f 100755

[PATCH v4 2/6] boot_linux.py: Use pick_qemu_util

Replace duplicate code with pick_qemu_util. Signed-off-by: Lukas Straub --- tests/acceptance/boot_linux.py | 11 ++- 1 file changed, 2 insertions(+), 9 deletions(-) diff --git a/tests/acceptance/boot_linux.py b/tests/acceptance/boot_linux.py index 1da4a53d6a..38029f8c70 100644 --- a/tes

[PATCH v4 4/6] colo: Introduce high-level test suite

Add high-level test relying on the colo resource-agent to test all failover cases while checking guest network connectivity. Signed-off-by: Lukas Straub --- scripts/colo-resource-agent/crm_master | 44 ++ scripts/colo-resource-agent/crm_resource | 12 + tests/acceptance/colo.py

[PATCH v4 3/6] colo: Introduce resource agent

Introduce a resource agent which can be used to manage qemu COLO in a pacemaker cluster. Signed-off-by: Lukas Straub --- scripts/colo-resource-agent/colo | 1527 ++ 1 file changed, 1527 insertions(+) create mode 100755 scripts/colo-resource-agent/colo diff --git a/s

[PATCH v4 6/6] MAINTAINERS: Add myself as maintainer for COLO resource agent

Signed-off-by: Lukas Straub --- MAINTAINERS | 6 ++ 1 file changed, 6 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 8d8b0bf966..d04567aa4d 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2773,6 +2773,12 @@ F: net/colo* F: net/filter-rewriter.c F: net/filter-mirror.c +COLO reso

Re: [PATCH v2 63/93] tcg/tci: Use ffi for calls

Am 04.02.21 um 02:44 schrieb Richard Henderson: This requires adjusting where arguments are stored. Place them on the stack at left-aligned positions. Adjust the stack frame to be at entirely positive offsets. Signed-off-by: Richard Henderson --- [...] diff --git a/tcg/tci.c b/tcg/tci.c inde

Interested in contributing to QEMU

Hello all, I am Niteesh, a junior student(3rd year) pursuing Electronics and Communication engineering. I was also a GSoC student for RTEMS last year. My main area of interest is low-level development (OS, Emulators, Hardware design, etc). I wanted to start contributing from last year itself but

Re: [PATCH] migration: Drop unused VMSTATE_FLOAT64 support

On 10/22/20 2:08 PM, Peter Maydell wrote: > Commit ef96e3ae9698d6 in January 2019 removed the last user of the > VMSTATE_FLOAT64* macros. These were used by targets which defined > their floating point register file as an array of 'float64'. Similar candidate: VMSTATE_CPUDOUBLE_ARRAY() > We used

Re: [PATCH v2 63/93] tcg/tci: Use ffi for calls

On 2/7/21 8:25 AM, Stefan Weil wrote: >> +#include "qemu-common.h" >> +#include "tcg/tcg.h"   /* MAX_OPC_PARAM_IARGS */ >> +#include "exec/cpu_ldst.h" >> +#include "tcg/tcg-op.h" >> +#include "qemu/compiler.h" >> +#include >> + > > > ffi.h is not found on macOS with Homebrew. > > This c

Re: [PATCH v4 2/5] acpi: Permit OEM ID and OEM table ID fields to be changed

"Michael S. Tsirkin" writes: > > > I queued this but there's a lot of code duplication with this. > Further, the use of g_strdup adds unnecessary dynamic memory > management where it's not needed. > I'd prefer > - a new struct AcpiBuildOem including the correct strings > - use sizeof of field

[Bug 1914117] Re: Short files returned via FTP on Qemu with various architectures and OSes

The more I look at this, the more I think it may be a macOS bug underneath. I've tested OpenBSD as a guest on a Debian AWS instance running 4.2.1 - all is fine. I've tested OpenBSD as a guest on a FreeBSD AWS instance running whatever is in ports and all is fine. Also others are having trouble:

Re: Increased execution time with TCI in latest git master (was: Re: [PULL 00/46] tcg patch queue)

On 2/7/21 2:50 AM, Stefan Weil wrote: > Your latest code from the rth7680/tci-next branch is twice as fast as my code > with BIOS boot and qemu-x86_64 on sparc64. That's great. > > With that code I don't get any BIOS output at all when running qemu-i386. > That's not so good. > > Did I test the c

Re: [PATCH v4 2/5] acpi: Permit OEM ID and OEM table ID fields to be changed

On Sun, Feb 07, 2021 at 08:23:33PM +0200, Marian Postevca wrote: > "Michael S. Tsirkin" writes: > > > > > > > I queued this but there's a lot of code duplication with this. > > Further, the use of g_strdup adds unnecessary dynamic memory > > management where it's not needed. > > I'd prefer > > -

Re: [PATCH] migration: Drop unused VMSTATE_FLOAT64 support

On Sun, 7 Feb 2021 at 17:10, Philippe Mathieu-Daudé wrote: > > On 10/22/20 2:08 PM, Peter Maydell wrote: > > Commit ef96e3ae9698d6 in January 2019 removed the last user of the > > VMSTATE_FLOAT64* macros. These were used by targets which defined > > their floating point register file as an array o

Re: [PATCH v2 63/93] tcg/tci: Use ffi for calls

On Sun, 7 Feb 2021 at 17:41, Richard Henderson wrote: > > On 2/7/21 8:25 AM, Stefan Weil wrote: > >> +#include "qemu-common.h" > >> +#include "tcg/tcg.h" /* MAX_OPC_PARAM_IARGS */ > >> +#include "exec/cpu_ldst.h" > >> +#include "tcg/tcg-op.h" > >> +#include "qemu/compiler.h" > >> +#inclu

Re: [PATCH v2 63/93] tcg/tci: Use ffi for calls

On 2/7/21 11:52 AM, Peter Maydell wrote: > On Sun, 7 Feb 2021 at 17:41, Richard Henderson > wrote: >> >> On 2/7/21 8:25 AM, Stefan Weil wrote: +#include "qemu-common.h" +#include "tcg/tcg.h" /* MAX_OPC_PARAM_IARGS */ +#include "exec/cpu_ldst.h" +#include "tcg/tcg-op.h

Re: [PATCH 0/2] utils/fifo8: minor updates

On 28/01/2021 22:17, Mark Cave-Ayland wrote: This patchset contains a couple of minor updates to QEMU's Fifo8 implementation conceived whilst working on the next revision of the ESP series. Patch 1 has already been reviewed on-list whilst patch 2 adds a new VMSTATE_FIFO8_TEST macro which is req

Re: [PATCH v2 63/93] tcg/tci: Use ffi for calls

On 07.02.21 21:12, Richard Henderson wrote: > On 2/7/21 11:52 AM, Peter Maydell wrote: >> On Sun, 7 Feb 2021 at 17:41, Richard Henderson >> wrote: >>> >>> On 2/7/21 8:25 AM, Stefan Weil wrote: > +#include "qemu-common.h" > +#include "tcg/tcg.h" /* MAX_OPC_PARAM_IARGS */ > +#i

[PATCH RFC v2 2/8] hw/block/nvme: remove block accounting for write zeroes

From: Klaus Jensen A Write Zeroes commands should not be counted in either the 'Data Units Written' or in 'Host Write Commands' SMART/Health Information Log page. Signed-off-by: Klaus Jensen --- hw/block/nvme.c | 1 - 1 file changed, 1 deletion(-) diff --git a/hw/block/nvme.c b/hw/block/nvme.

[PATCH RFC v2 4/8] hw/block/nvme: try to deal with the iov/qsg duality

From: Klaus Jensen Introduce NvmeSg and try to deal with that pesky qsg/iov duality that haunts all the memory-related functions. Signed-off-by: Klaus Jensen --- hw/block/nvme.h | 8 ++- hw/block/nvme.c | 171 2 files changed, 90 insertions(+)

[PATCH RFC v2 1/8] hw/block/nvme: remove redundant len member in compare context

From: Klaus Jensen The 'len' member of the nvme_compare_ctx struct is redundant since the same information is available in the 'iov' member. Signed-off-by: Klaus Jensen --- hw/block/nvme.c | 10 -- 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/hw/block/nvme.c b/hw/block

[PATCH RFC v2 3/8] hw/block/nvme: fix strerror printing

From: Klaus Jensen Fix missing sign inversion. Signed-off-by: Klaus Jensen --- hw/block/nvme.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/hw/block/nvme.c b/hw/block/nvme.c index e4a01cf9edc5..29902038d618 100644 --- a/hw/block/nvme.c +++ b/hw/block/nvme.c @@ -1150,7 +1

[PATCH RFC v2 6/8] hw/block/nvme: refactor nvme_dma

From: Klaus Jensen The nvme_dma function doesn't just do DMA (QEMUSGList-based) memory transfers; it also handles QEMUIOVector copies. Introduce the NvmeTxDirection enum and rename to nvme_tx. Remove mapping of PRPs/SGLs from nvme_tx and instead assert that they have been mapped previously. This

[PATCH RFC v2 0/8] hw/block/nvme: metadata and end-to-end data protection support

From: Klaus Jensen This is RFC v2 of a series that adds support for metadata and end-to-end data protection. First, on the subject of metadata, in v1, support was restricted to extended logical blocks, which was pretty trivial to implement, but required special initialization and broke DULBE. In

[PATCH RFC v2 7/8] hw/block/nvme: add metadata support

From: Klaus Jensen Add support for metadata in the form of extended logical blocks as well as a separate buffer of data. The new `ms` nvme-ns device parameter specifies the size of metadata per logical block in bytes. The `mset` nvme-ns device parameter controls whether metadata is transfered as

[PATCH RFC v2 5/8] hw/block/nvme: remove the req dependency in map functions

From: Klaus Jensen The PRP and SGL mapping functions does not have any particular need for the entire NvmeRequest as a parameter. Clean it up. Signed-off-by: Klaus Jensen --- hw/block/nvme.c | 61 ++- hw/block/trace-events | 4 +-- 2 files changed

[PATCH RFC v2 8/8] hw/block/nvme: end-to-end data protection

From: Klaus Jensen Add support for namespaces formatted with protection information. The type of end-to-end data protection (i.e. Type 1, Type 2 or Type 3) is selected with the `pi` nvme-ns device parameter. If the number of metadata bytes is larger than 8, the `pil` nvme-ns device parameter may

Re: Increased execution time with TCI in latest git master (was: Re: [PULL 00/46] tcg patch queue)

On 07.02.21 19:37, Richard Henderson wrote: > On 2/7/21 2:50 AM, Stefan Weil wrote: >> Your latest code from the rth7680/tci-next branch is twice as fast as my code >> with BIOS boot and qemu-x86_64 on sparc64. That's great. >> >> With that code I don't get any BIOS output at all when running qemu-

[PULL 0/2] qemu-sparc queue 20210207

The following changes since commit 5b19cb63d9dfda41b412373b8c9fe14641bcab60: Merge remote-tracking branch 'remotes/rth-gitlab/tags/pull-tcg-20210205' into staging (2021-02-05 22:59:12 +) are available in the Git repository at: git://github.com/mcayland/qemu.git tags/qemu-spar

[PULL 2/2] utils/fifo8: add VMSTATE_FIFO8_TEST macro

Rewrite the existing VMSTATE_FIFO8 macro to use VMSTATE_FIFO8_TEST as per the standard pattern in include/migration/vmstate.h. Signed-off-by: Mark Cave-Ayland Reviewed-by: Peter Maydell Message-Id: <20210128221728.14887-3-mark.cave-ayl...@ilande.co.uk> --- include/qemu/fifo8.h | 16 ++--

[PULL 1/2] utils/fifo8: change fatal errors from abort() to assert()

Developer errors are better represented with assert() rather than abort(). Also improve the strictness of the checks by using range checks within the assert() rather than converting the existing equality checks to inequality checks. Signed-off-by: Mark Cave-Ayland Reviewed-by: Claudio Fontana Re

[RFC PATCH 0/6] exec: Remove "tcg/tcg.h" from "exec/cpu_ldst.h"

Hi, I wondered why changing something in "tcg/tcg.h" would trigger rebuilding the whole tree and figured the inclusion in "exec/cpu_ldst.h". By making tlb_addr_write() static to accel/tcg/cputlb.c we can remove the "tcg/tcg.h" inclusion and reduce the number of objects to rebuild. I added tlb_as

[PATCH 6/6] exec/cpu_ldst: Move tlb* declarations to "exec/exec-all.h"

Keep MMU functions in "exec/cpu_ldst.h", and move TLB functions to "exec/exec-all.h". As tlb_addr_write() is only called in accel/tcg/cputlb.c, make move it there as a static function. Doing so we removed the "tcg/tcg.h" dependency on "exec/cpu_ldst.h". Signed-off-by: Philippe Mathieu-Daudé ---

[PATCH 2/6] target/m68k: Include missing "tcg/tcg.h" header

Commit 14f944063af ("target-m68k: add cas/cas2 ops") introduced use of typedef/prototypes declared in "tcg/tcg.h" without including it. This was not a problem because "tcg/tcg.h" is pulled in by "exec/cpu_ldst.h". To be able to remove this header there, we first need to include it here in op_helper

[RFC PATCH 1/6] target: Replace tcg_debug_assert() by assert()

Since commit 262a69f4282 ("osdep.h: Prohibit disabling assert() in supported builds") we can not build QEMU with assert() disabled. tcg_debug_assert() does nothing until QEMU is configured with --enable-debug-tcg. Since there is no obvious logic whether to use tcg_debug_assert() or assert() for f

[PATCH 3/6] target/mips: Include missing "tcg/tcg.h" header

Commit 83be6b54123 ("Fix MSA instructions LD. on big endian host") introduced use of typedef/prototypes declared in "tcg/tcg.h" without including it. This was not a problem because "tcg/tcg.h" is pulled in by "exec/cpu_ldst.h". To be able to remove this header there, we first need to include it her

[PATCH 4/6] accel/tcg: Include missing "tcg/tcg.h" header

Commit 3468b59e18b ("tcg: enable multiple TCG contexts in softmmu") introduced use of typedef/prototypes declared in "tcg/tcg.h" without including it. This was not a problem because "tcg/tcg.h" is pulled in by "exec/cpu_ldst.h". To be able to remove this header there, we first need to include it he

[RFC PATCH 5/6] accel/tcg: Refactor debugging tlb_assert_iotlb_entry_for_ptr_present()

Refactor debug code as tlb_assert_iotlb_entry_for_ptr_present() helper. Signed-off-by: Philippe Mathieu-Daudé --- What this code does is out of my league, but refactoring it allow keeping tlb_addr_write() local to accel/tcg/cputlb.c in the next patch. --- include/exec/exec-all.h | 9 +

Re: [PATCH 6/6] exec/cpu_ldst: Move tlb* declarations to "exec/exec-all.h"

On 2/7/21 11:57 PM, Philippe Mathieu-Daudé wrote: > Keep MMU functions in "exec/cpu_ldst.h", and move TLB functions > to "exec/exec-all.h". As tlb_addr_write() is only called in > accel/tcg/cputlb.c, make move it there as a static function. > > Doing so we removed the "tcg/tcg.h" dependency on "ex

[RFC PATCH v2 0/6] exec: Remove "tcg/tcg.h" from "exec/cpu_ldst.h"

Since v1: - Do not move tlb_vaddr_to_host() Hi, I wondered why changing something in "tcg/tcg.h" would trigger rebuilding the whole tree and figured the inclusion in "exec/cpu_ldst.h". By making tlb_addr_write() static to accel/tcg/cputlb.c we can remove the "tcg/tcg.h" inclusion and reduce the

[PATCH v2 3/6] target/mips: Include missing "tcg/tcg.h" header

Commit 83be6b54123 ("Fix MSA instructions LD. on big endian host") introduced use of typedef/prototypes declared in "tcg/tcg.h" without including it. This was not a problem because "tcg/tcg.h" is pulled in by "exec/cpu_ldst.h". To be able to remove this header there, we first need to include it her

[PATCH v2 2/6] target/m68k: Include missing "tcg/tcg.h" header

Commit 14f944063af ("target-m68k: add cas/cas2 ops") introduced use of typedef/prototypes declared in "tcg/tcg.h" without including it. This was not a problem because "tcg/tcg.h" is pulled in by "exec/cpu_ldst.h". To be able to remove this header there, we first need to include it here in op_helper

[RFC PATCH v2 1/6] target: Replace tcg_debug_assert() by assert()

Since commit 262a69f4282 ("osdep.h: Prohibit disabling assert() in supported builds") we can not build QEMU with assert() disabled. tcg_debug_assert() does nothing until QEMU is configured with --enable-debug-tcg. Since there is no obvious logic whether to use tcg_debug_assert() or assert() for f

[PATCH v2 4/6] accel/tcg: Include missing "tcg/tcg.h" header

Commit 3468b59e18b ("tcg: enable multiple TCG contexts in softmmu") introduced use of typedef/prototypes declared in "tcg/tcg.h" without including it. This was not a problem because "tcg/tcg.h" is pulled in by "exec/cpu_ldst.h". To be able to remove this header there, we first need to include it he

[RFC PATCH v2 5/6] accel/tcg: Refactor debugging tlb_assert_iotlb_entry_for_ptr_present()

Refactor debug code as tlb_assert_iotlb_entry_for_ptr_present() helper. Signed-off-by: Philippe Mathieu-Daudé --- What this code does is out of my league, but refactoring it allow keeping tlb_addr_write() local to accel/tcg/cputlb.c in the next patch. --- include/exec/exec-all.h | 9 +

[PATCH v2 6/6] exec/cpu_ldst: Move tlb* declarations to "exec/exec-all.h"

Keep MMU functions in "exec/cpu_ldst.h", and move TLB functions to "exec/exec-all.h". As tlb_addr_write() is only called in accel/tcg/cputlb.c, make move it there as a static function. Doing so we removed the "tcg/tcg.h" dependency on "exec/cpu_ldst.h". Signed-off-by: Philippe Mathieu-Daudé ---

Re: [PATCH v2] hw/block/nvme: add missing mor/mar constraint checks

On Tue, 2021-01-26 at 13:15 +0100, Klaus Jensen wrote: > From: Klaus Jensen > > Firstly, if zoned.max_active is non-zero, zoned.max_open must be less > than or equal to zoned.max_active. > > Secondly, if only zones.max_active is set, we have to explicitly set > zones.max_open or we end up with a

[PATCH] hw/block/nvme: fix Close Zone

Implicitly and Explicitly Open zones can be closed by Close Zone management function. This got broken by a recent commit and now such commands fail with Invalid Zone State Transition status. Modify nvm_zrm_close() function to make Close Zone work correctly. Signed-off-by: Dmitry Fomichev Fixes:

Re: [PATCH] scsi: mptsas: dequeue request object in case of an error (CVE-2021-3392)

P J P 于2021年2月2日周二 下午9:23写道： > > From: Prasad J Pandit > > While processing SCSI i/o requests in mptsas_process_scsi_io_request(), > the Megaraid emulator appends new MPTSASRequest object 'req' to > the 's->pending' queue. In case of an error, this same object gets > dequeued in mptsas_free_reque

[PATCH v3 00/70] TCI fixes and cleanups

Changes since v2: * 20-something patches are now upstream. * Increase testing timeout for tci. * Gitlab testing for tci w/ 32-bit host. r~ Richard Henderson (70): gdbstub: Fix handle_query_xfer_auxv tcg: Split out tcg_raise_tb_overflow tcg: Manage splitwx in tc_ptr_to_region_tree by

[PATCH v3 01/70] gdbstub: Fix handle_query_xfer_auxv

The main problem was that we were treating a guest address as a host address with a mere cast. Use the correct interface for accessing guest memory. Do not allow offset == auxv_len, which would result in an empty packet. Fixes: 51c623b0de1 ("gdbstub: add support to Xfer:auxv:read: packet") Signe

[PATCH v3 06/70] tcg/tci: Remove tci_read_r8s

Use explicit casts for ext8s opcodes. Signed-off-by: Richard Henderson --- tcg/tci.c | 25 - 1 file changed, 4 insertions(+), 21 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index c44a4aec7b..25db479e62 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -57,13 +57,6 @@ static

[PATCH v3 07/70] tcg/tci: Remove tci_read_r16

Use explicit casts for ext16u opcodes, and allow truncation to happen with the store for st16 opcodes, and with the call for bswap16 opcodes. Signed-off-by: Richard Henderson --- tcg/tci.c | 28 +++- 1 file changed, 7 insertions(+), 21 deletions(-) diff --git a/tcg/tci.c

[PATCH v3 02/70] tcg: Split out tcg_raise_tb_overflow

Allow other places in tcg to restart with a smaller tb. Signed-off-by: Richard Henderson --- tcg/tcg.c | 9 +++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/tcg/tcg.c b/tcg/tcg.c index 63a12b197b..bbe3dcee03 100644 --- a/tcg/tcg.c +++ b/tcg/tcg.c @@ -346,6 +346,12 @@ stati

[PATCH v3 03/70] tcg: Manage splitwx in tc_ptr_to_region_tree by hand

The use in tcg_tb_lookup is given a random pc that comes from the pc of a signal handler. Do not assert that the pointer is already within the code gen buffer at all, much less the writable mirror of it. Fixes: db0c51a3803 Signed-off-by: Richard Henderson --- For TCI, this indicates a bug in ha

[PATCH v3 14/70] tcg/tci: Remove ifdefs for TCG_TARGET_HAS_ext32[us]_i64

These operations are always available under different names: INDEX_op_ext_i32_i64 and INDEX_op_extu_i32_i64, so we remove no code with the ifdef. Signed-off-by: Richard Henderson --- tcg/tci.c | 4 1 file changed, 4 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index cdfd9b7af8..1819652c

[PATCH v3 08/70] tcg/tci: Remove tci_read_r16s

Use explicit casts for ext16s opcodes. Signed-off-by: Richard Henderson --- tcg/tci.c | 26 -- 1 file changed, 4 insertions(+), 22 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 547be0c2f0..d2bfcb3c93 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -57,13 +57,6 @@ stati

[PATCH v3 04/70] tcg/tci: Merge identical cases in generation

Use CASE_32_64 and CASE_64 to reduce ifdefs and merge cases that are identical between 32-bit and 64-bit hosts. Signed-off-by: Richard Henderson --- tcg/tci/tcg-target.c.inc | 204 ++- 1 file changed, 73 insertions(+), 131 deletions(-) diff --git a/tcg/tci/tc

[PATCH v3 10/70] tcg/tci: Remove tci_read_r32s

Use explicit casts for ext32s opcodes. Signed-off-by: Richard Henderson --- tcg/tci.c | 20 ++-- 1 file changed, 2 insertions(+), 18 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 72ec63e18e..9c8395397a 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -57,13 +57,6 @@ static tcg_

[PATCH v3 15/70] tcg/tci: Merge bswap operations

This includes bswap16 and bswap32. Signed-off-by: Richard Henderson --- tcg/tci.c | 22 -- 1 file changed, 4 insertions(+), 18 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 1819652c5a..c979215332 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -652,15 +652,15 @@ uintptr_t

[PATCH v3 09/70] tcg/tci: Remove tci_read_r32

Use explicit casts for ext32u opcodes, and allow truncation to happen for other users. Signed-off-by: Richard Henderson --- tcg/tci.c | 122 -- 1 file changed, 54 insertions(+), 68 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index d2bfcb3c

[PATCH v3 05/70] tcg/tci: Remove tci_read_r8

Use explicit casts for ext8u opcodes, and allow truncation to happen with the store for st8 opcodes. Signed-off-by: Richard Henderson --- tcg/tci.c | 23 +-- 1 file changed, 5 insertions(+), 18 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index fb3c97aaf1..c44a4aec7b 1006

[PATCH v3 12/70] tcg/tci: Merge basic arithmetic operations

This includes add, sub, mul, and, or, xor. Signed-off-by: Richard Henderson --- tcg/tci.c | 83 +-- 1 file changed, 25 insertions(+), 58 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 0246e663a3..894e87e1b0 100644 --- a/tcg/tci.c +++ b/

[PATCH v3 18/70] tcg/tci: Split out tci_args_rrs

Begin splitting out functions that do pure argument decode, without actually loading values from the register set. This means that decoding need not concern itself between input and output registers. We can assert that the register number is in range during decode, so that it is safe to simply de

[PATCH v3 13/70] tcg/tci: Merge extension operations

This includes ext8s, ext8u, ext16s, ext16u. Signed-off-by: Richard Henderson --- tcg/tci.c | 44 1 file changed, 8 insertions(+), 36 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 894e87e1b0..cdfd9b7af8 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @

[PATCH v3 11/70] tcg/tci: Reduce use of tci_read_r64

In all cases restricted to 64-bit hosts, tcg_read_r is identical. We retain the 64-bit symbol for the single case of INDEX_op_qemu_st_i64. Signed-off-by: Richard Henderson --- tcg/tci.c | 93 +-- 1 file changed, 42 insertions(+), 51 deletions(

[PATCH v3 16/70] tcg/tci: Merge mov, not and neg operations

Signed-off-by: Richard Henderson --- tcg/tci.c | 29 + 1 file changed, 5 insertions(+), 24 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index c979215332..225cb698e8 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -404,7 +404,7 @@ uintptr_t QEMU_DISABLE_CFI tcg_qemu_tb_ex

[PATCH v3 20/70] tcg/tci: Split out tci_args_rrr

Signed-off-by: Richard Henderson --- tcg/tci.c | 154 -- 1 file changed, 57 insertions(+), 97 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 0bc5294e8b..1736234bfd 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -191,6 +191,14 @@ static void t

[PATCH v3 17/70] tcg/tci: Rename tci_read_r to tci_read_rval

In the next patches, we want to use tci_read_r to return the raw register number. So rename the existing function, which returns the register value, to tci_read_rval. Signed-off-by: Richard Henderson --- tcg/tci.c | 192 +++--- 1 file changed, 96

[PATCH v3 19/70] tcg/tci: Split out tci_args_rr

Signed-off-by: Richard Henderson --- tcg/tci.c | 67 +-- 1 file changed, 31 insertions(+), 36 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index be298ae39d..0bc5294e8b 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -184,6 +184,13 @@ static tcg_ta

[PATCH v3 21/70] tcg/tci: Split out tci_args_rrrc

Signed-off-by: Richard Henderson --- tcg/tci.c | 23 +-- 1 file changed, 13 insertions(+), 10 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 1736234bfd..86625061f1 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -207,6 +207,15 @@ static void tci_args_rrs(const uint8_t **tb_p

[PATCH v3 22/70] tcg/tci: Split out tci_args_l

Signed-off-by: Richard Henderson --- tcg/tci.c | 9 +++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 86625061f1..8bc9dd27b0 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -184,6 +184,11 @@ static tcg_target_ulong tci_read_label(const uint8_t **tb_ptr

[PATCH v3 29/70] tcg/tci: Split out tci_args_rrrrrr

Signed-off-by: Richard Henderson --- tcg/tci.c | 31 --- 1 file changed, 20 insertions(+), 11 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 0301ee63a7..84d77855ee 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -258,6 +258,17 @@ static void tci_args_rc(const uin

[PATCH v3 26/70] tcg/tci: Reuse tci_args_l for calls.

Signed-off-by: Richard Henderson --- tcg/tci.c | 38 +++--- 1 file changed, 19 insertions(+), 19 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 5cc05fa554..92b13829c3 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -452,30 +452,30 @@ uintptr_t QEMU_DISABLE_CFI tc

[PATCH v3 23/70] tcg/tci: Split out tci_args_rrrrrc

Signed-off-by: Richard Henderson --- tcg/tci.c | 25 +++-- 1 file changed, 19 insertions(+), 6 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 8bc9dd27b0..692b95b5c2 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -221,6 +221,19 @@ static void tci_args_rrrc(const uint8_t **tb

[PATCH v3 24/70] tcg/tci: Split out tci_args_rrcl and tci_args_rrrrcl

Signed-off-by: Richard Henderson --- tcg/tci.c | 52 1 file changed, 32 insertions(+), 20 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 692b95b5c2..1e2f78a9f9 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -212,6 +212,15 @@ static void tci_

[PATCH v3 32/70] tcg/tci: Reduce qemu_ld/st TCGMemOpIdx operand to 32-bits

We are currently using the "natural" size routine, which uses 64-bits on a 64-bit host. The TCGMemOpIdx operand has 11 bits, so we can safely reduce to 32-bits. Signed-off-by: Richard Henderson --- tcg/tci.c| 8 tcg/tci/tcg-target.c.inc | 4 ++-- 2 files changed, 6 inse

[PATCH v3 30/70] tcg/tci: Split out tci_args_rrrr

Signed-off-by: Richard Henderson --- tcg/tci.c | 16 +++- 1 file changed, 11 insertions(+), 5 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 84d77855ee..cb24295cd9 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -237,6 +237,15 @@ static void tci_args_rrrc(const uint8_t **tb_ptr, }

[PATCH v3 27/70] tcg/tci: Reuse tci_args_l for exit_tb

Do not emit a uint64_t, but a tcg_target_ulong, aka uintptr_t. This reduces the size of the constant on 32-bit hosts. The assert for label != NULL has to be removed because that is a valid value for exit_tb. Signed-off-by: Richard Henderson --- tcg/tci.c| 13 - tcg/tc

[PATCH v3 28/70] tcg/tci: Reuse tci_args_l for goto_tb

Convert to indirect jumps, as it's less complicated. Then we just have a pointer to the tb address at which the chain is stored, from which we read. Signed-off-by: Richard Henderson --- tcg/tci/tcg-target.h | 11 +++ tcg/tci.c| 8 +++- tcg/tci/tcg-target.c.inc |

[PATCH v3 34/70] tcg/tci: Hoist op_size checking into tci_args_*

This performs the size check while reading the arguments, which means that we don't have to arrange for it to be done after the operation. Which tidies all of the branches. Signed-off-by: Richard Henderson --- tcg/tci.c | 87 ++- 1 file change

[PATCH v3 40/70] tcg/tci: Move call-return regs to end of tcg_target_reg_alloc_order

As the only call-clobbered regs for TCI, these should receive the least priority. Signed-off-by: Richard Henderson --- tcg/tci/tcg-target.c.inc | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tcg/tci/tcg-target.c.inc b/tcg/tci/tcg-target.c.inc index 4dae09deda..53edc50a3b

[PATCH v3 25/70] tcg/tci: Split out tci_args_ri and tci_args_rI

Signed-off-by: Richard Henderson --- tcg/tci.c | 38 ++ 1 file changed, 22 insertions(+), 16 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index 1e2f78a9f9..5cc05fa554 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -121,16 +121,6 @@ static int32_t tci_read_s32(co

[PATCH v3 33/70] tcg/tci: Split out tci_args_{rrm,rrrm,rrrrm}

Signed-off-by: Richard Henderson --- tcg/tci.c | 147 ++ 1 file changed, 81 insertions(+), 66 deletions(-) diff --git a/tcg/tci.c b/tcg/tci.c index ddc138359b..a1846825ea 100644 --- a/tcg/tci.c +++ b/tcg/tci.c @@ -66,22 +66,18 @@ tci_write_reg(

[PATCH v3 35/70] tcg/tci: Remove tci_disas

This function is unused. It's not even the disassembler, which is print_insn_tci, located in disas/tci.c. Signed-off-by: Richard Henderson --- tcg/tci/tcg-target.h | 2 -- tcg/tci/tcg-target.c.inc | 10 -- 2 files changed, 12 deletions(-) diff --git a/tcg/tci/tcg-target.h b/tcg/tc

[PATCH v3 42/70] tcg/tci: Split out tcg_out_op_rrs

Signed-off-by: Richard Henderson --- tcg/tci/tcg-target.c.inc | 84 +++- 1 file changed, 39 insertions(+), 45 deletions(-) diff --git a/tcg/tci/tcg-target.c.inc b/tcg/tci/tcg-target.c.inc index 050d514853..707f801099 100644 --- a/tcg/tci/tcg-target.c.inc +++ b

[PATCH v3 31/70] tcg/tci: Clean up deposit operations

Use the correct set of asserts during code generation. We do not require the first input to overlap the output; the existing interpreter already supported that. Split out tci_args_rrrbb in the translator. Use the deposit32/64 functions rather than inline expansion. Signed-off-by: Richard Henderso

[PATCH v3 36/70] tcg/tci: Implement the disassembler properly

Actually print arguments as opposed to simply the opcodes and, uselessly, the argument counts. Reuse all of the helpers developed as part of the interpreter. Signed-off-by: Richard Henderson --- meson.build | 2 +- include/tcg/tcg-opc.h | 2 - disas/tci.c | 61 -

[PATCH v3 39/70] tcg/tci: Improve tcg_target_call_clobber_regs

The current setting is much too pessimistic. Indicating only the one or two registers that are actually assigned after a call should avoid unnecessary movement between the register array and the stack array. Signed-off-by: Richard Henderson --- tcg/tci/tcg-target.c.inc | 10 -- 1 file c

  1   2   3   >