aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2023-06-28block: Fix AioContext locking in bdrv_open_inherit()Kevin Wolf
bdrv_open_inherit() calls several functions for which it needs to hold the AioContext lock, but currently doesn't. This includes calls in bdrv_append_temp_snapshot(), for which bdrv_open_inherit() is the only caller. Fix the locking in these places. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Message-ID: <20230605085711.21261-8-kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2023-06-28block: Fix AioContext locking in bdrv_reopen_parse_file_or_backing()Kevin Wolf
bdrv_set_file_or_backing_noperm() requires the caller to hold the AioContext lock for the child node, but we hold the one for the parent node in bdrv_reopen_parse_file_or_backing(). Take the other one temporarily. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Message-ID: <20230605085711.21261-7-kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2023-06-28block: Fix AioContext locking in bdrv_attach_child_common()Kevin Wolf
The function can move the child node to a different AioContext. In this case, it also must take the AioContext lock for the new context before calling functions that require the caller to hold the AioContext for the child node. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Message-ID: <20230605085711.21261-6-kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2023-06-28block: Fix AioContext locking in bdrv_open_child()Kevin Wolf
bdrv_attach_child() requires that the caller holds the AioContext lock for the new child node. Take it in bdrv_open_child() and document that the caller must not hold any AioContext apart from the main AioContext. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Message-ID: <20230605085711.21261-5-kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2023-06-28test-block-iothread: Lock AioContext for blk_insert_bs()Kevin Wolf
blk_insert_bs() requires that callers hold the AioContext lock for the node that should be inserted. Take it. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Message-ID: <20230605085711.21261-4-kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2023-06-28qdev-properties-system: Lock AioContext for blk_insert_bs()Kevin Wolf
blk_insert_bs() requires that callers hold the AioContext lock for the node that should be inserted. Take it. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Message-ID: <20230605085711.21261-3-kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2023-06-28iotests: Test active commit with iothread and background I/OKevin Wolf
This is a better regression test for the bugs hidden by commit 80fc5d26 ('graph-lock: Disable locking for now'). With that commit reverted, it hangs instantaneously and reliably for me. It is important to have a reliable test like this, because the following commits will set out to fix the actual root cause of the deadlocks and then finally revert commit 80fc5d26, which was only a stopgap solution. Signed-off-by: Kevin Wolf <kwolf@redhat.com> Message-ID: <20230605085711.21261-2-kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2023-06-28Merge tag 'ui-pull-request' of https://gitlab.com/marcandre.lureau/qemu into ↵Richard Henderson
staging wef # -----BEGIN PGP SIGNATURE----- # # iQJQBAABCAA6FiEEh6m9kz+HxgbSdvYt2ujhCXWWnOUFAmSa+6McHG1hcmNhbmRy # ZS5sdXJlYXVAcmVkaGF0LmNvbQAKCRDa6OEJdZac5YxfD/46nwpTCTWWdKdTeo5p # OUdrF0rfHFqu4FN3OWHbfRxZCO/AdHL+UZ52owV4+5bJJI5uOnXwYKpkrwKYBrFd # H8Bll7NitzA41rw4AQa0GeaQYCPJ99OOfnhbRI5Aep2NG2DfX5PK4RWnfqYw8LD1 # TiHtRv2lWnX9EyMjnEh93C+n17OfquP5Ew3ozZNQJ0+SiJ3CvsUn6hEqxOA8OdyX # lj6l00CASQA2BxW+zjXjJKvRakCV4gfdvrL9eMf4eu0UopzET7ombBJGPnYVsrDU # /4R7b0JgGM4iOpXFxK4Ng6myP28vPdOEJAU/OJLH+oMRz1caohS+0Ijl2KviUCex # SGpb9plxqI7fI2QQt+1CxAlXADSW7oV1zV0/tLkKl/n5+MF3HJ/5qR3tefLhYu1p # 2LpfbPMKGQ9V3+5Z/UvWx6GQYP1iBRm5THPLn+HSDMSqLmt6yp5cOTwP3KTx1Zlc # JfpBtekT2Cgs54nnCcfnXa6/EPo4uR7cMFzrgXdSacPz/GssMVa1c2mNUYkgYEYU # PeyDWZG2Rt/70y+CFDPBpKWEQVICnf7Ha43oj4BtGTqqUFeuZClMTTtZ5poSg3ir # FcRNJ5zSWg2KhHIQ9TQKxIAwrxxVBY0AiQleNRyDzx+YGAuBBadO6i5eCqqpGgOa # QRVBsP33Pg/QD1JdxN9GSSEh0w== # =cR6x # -----END PGP SIGNATURE----- # gpg: Signature made Tue 27 Jun 2023 05:09:23 PM CEST # gpg: using RSA key 87A9BD933F87C606D276F62DDAE8E10975969CE5 # gpg: issuer "marcandre.lureau@redhat.com" # gpg: Good signature from "Marc-André Lureau <marcandre.lureau@redhat.com>" [full] # gpg: aka "Marc-André Lureau <marcandre.lureau@gmail.com>" [full] * tag 'ui-pull-request' of https://gitlab.com/marcandre.lureau/qemu: (33 commits) ui/dbus: use shared D3D11 Texture2D when possible virtio-gpu-virgl: use D3D11_SHARE_TEXTURE when available ui: add optional d3d texture pointer to scanout texture ui/egl: query ANGLE d3d device virtio-gpu-virgl: teach it to get the QEMU EGL display ui/dbus: add some GL traces ui/dbus: add GL support on win32 ui: add egl_fb_read_rect() ui/egl: default to GLES on windows ui: add egl-headless support on win32 ui/dbus: use shared memory when possible on win32 virtio-gpu/win32: allocate shareable 2d resources/images console/win32: allocate shareable display surface ui/dbus: introduce "Interfaces" properties tests: make dbus-display-test work on win32 qtest: add qtest_pid() ui/dbus: win32 support scripts: add a XML preprocessor script ui/dbus: compile without gio/gunixfdlist.h ui/egl: fix make_context_current() callback return value ... Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-06-27ui/dbus: use shared D3D11 Texture2D when possibleMarc-André Lureau
When the client implements "org.qemu.Display1.Listener.Win32.D3d11" and we are running on ANGLE/win32, share the scanout texture with the peer process, and draw with ScanoutTexture2d/UpdateTexture2d methods. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-22-marcandre.lureau@redhat.com>
2023-06-27virtio-gpu-virgl: use D3D11_SHARE_TEXTURE when availableMarc-André Lureau
Enable D3D texture sharing when possible, and pass it to the texture display callbacks. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-21-marcandre.lureau@redhat.com>
2023-06-27ui: add optional d3d texture pointer to scanout textureMarc-André Lureau
The following patch will get the underlying D3D11 Texture2D from the virgl renderer scanout. Pass it along to the texture scanout callbacks as a priliminary step, to simplify review. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-20-marcandre.lureau@redhat.com>
2023-06-27ui/egl: query ANGLE d3d deviceMarc-André Lureau
Check if ANGLE is being used with D3D backend. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-19-marcandre.lureau@redhat.com>
2023-06-27virtio-gpu-virgl: teach it to get the QEMU EGL displayMarc-André Lureau
virgl offers a few features that require to have access to the underlying EGLDisplay. This is the case for the D3D texture sharing support. The API callback is merged for virgl 1.0: https://gitlab.freedesktop.org/virgl/virglrenderer/-/merge_requests/1113 Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-18-marcandre.lureau@redhat.com>
2023-06-27ui/dbus: add some GL tracesMarc-André Lureau
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-17-marcandre.lureau@redhat.com>
2023-06-27ui/dbus: add GL support on win32Marc-André Lureau
Enable usage of dbus,gl= on win32. At this point, the scanout texture is read on the DisplaySurface memory, and the client is then updated with the "2D" API (with shared memory if possible). Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-16-marcandre.lureau@redhat.com>
2023-06-27ui: add egl_fb_read_rect()Marc-André Lureau
Similar to egl_fb_read(), same limitations, but with extra arguments to read a subset of the framebuffer. Used in following commits. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-15-marcandre.lureau@redhat.com>
2023-06-27ui/egl: default to GLES on windowsMarc-André Lureau
Windows GL drivers are notoriously not very good. Otoh, ANGLE provides rock solid GLES implementation on top of direct3d. We should recommend it and default to ES when using EGL (users can easily override this if necessary) Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-14-marcandre.lureau@redhat.com>
2023-06-27ui: add egl-headless support on win32Marc-André Lureau
Make GBM optional for EGL code, and enable the build for win32. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-13-marcandre.lureau@redhat.com>
2023-06-27ui/dbus: use shared memory when possible on win32Marc-André Lureau
When the display surface has an associated HANDLE, we can duplicate it to the client process and let it map the memory to avoid expensive copies. Introduce two new win32-specific methods ScanoutMap and UpdateMap. The first is used to inform the listener about the a shared map availability, and the second for display updates. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-12-marcandre.lureau@redhat.com>
2023-06-27virtio-gpu/win32: allocate shareable 2d resources/imagesMarc-André Lureau
Allocate pixman bits for scanouts with qemu_win32_map_alloc() so we can set a shareable handle on the associated display surface. Note: when bits are provided to pixman_image_create_bits(), you must also give the rowstride (the argument is ignored when bits is NULL) Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-11-marcandre.lureau@redhat.com>
2023-06-27console/win32: allocate shareable display surfaceMarc-André Lureau
Introduce qemu_win32_map_alloc() and qemu_win32_map_free() to allocate shared memory mapping. The handle can be used to share the mapping with another process. Teach qemu_create_displaysurface() to allocate shared memory. Following patches will introduce other places for shared memory allocation. Other patches for -display dbus will share the memory when possible with the client, to avoid expensive memory copy between the processes. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-10-marcandre.lureau@redhat.com>
2023-06-27ui/dbus: introduce "Interfaces" propertiesMarc-André Lureau
This property is similar to ``org.freedesktop.DBus.Interfaces`` property on the bus interface: it's an array of strings listing the extra interfaces and capabilities available, in a convenient way. Most interfaces are implicit, as they are required. For ``org/qemu/Display1_$id``, we can list the Keyboard And Mouse interfaces. Those could be optional. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-9-marcandre.lureau@redhat.com>
2023-06-27tests: make dbus-display-test work on win32Marc-André Lureau
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Acked-by: Thomas Huth <thuth@redhat.com> Message-Id: <20230606115658.677673-8-marcandre.lureau@redhat.com>
2023-06-27qtest: add qtest_pid()Marc-André Lureau
Used in the following test on win32, to share sockets with the QEMU process. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Reviewed-by: Thomas Huth <thuth@redhat.com> Message-Id: <20230606115658.677673-7-marcandre.lureau@redhat.com>
2023-06-27ui/dbus: win32 supportMarc-André Lureau
D-Bus doesn't support fd-passing on Windows (AF_UNIX doesn't have SCM_RIGHTS yet, but there are other means to share objects. I have proposed various solutions upstream, but none seem fitting enough atm). To make the "-display dbus" work on Windows, implement an alternative D-Bus interface where all the 'h' (FDs) arguments are replaced with 'ay' (WSASocketW data), and sockets are passed to the other end via WSADuplicateSocket(). Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-6-marcandre.lureau@redhat.com>
2023-06-27scripts: add a XML preprocessor scriptMarc-André Lureau
gdbus-codegen doesn't support conditions or pre-processing. Rather than duplicating D-Bus interfaces for win32 adaptation, let's have a preprocess step, so we can have platform-specific interfaces. The python script is based on https://github.com/peitaosu/XML-Preprocessor, with bug fixes, some testing and replacing lxml dependency with the built-in xml module. This preprocessing syntax style is not very common, but is similar to the one provided by WiX (https://wixtoolset.org/docs/v3/overview/preprocessor/) or wixl, that we adopted in QEMU for packaging the guest agent. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-5-marcandre.lureau@redhat.com>
2023-06-27ui/dbus: compile without gio/gunixfdlist.hMarc-André Lureau
D-Bus on windows doesn't support fd-passing. Let's isolate the fdlist-related code as a first step, before adding Windows support, using another mechanism. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-4-marcandre.lureau@redhat.com>
2023-06-27ui/egl: fix make_context_current() callback return valueMarc-André Lureau
eglMakeCurrent() returns 1/EGL_TRUE on success. This is not what the callback expects, where 0 indicates success. While at it, print the EGL error to ease debugging. As with virgl_renderer_callbacks, the return value is now checked since version >= 4: https://gitlab.freedesktop.org/virgl/virglrenderer/-/commit/7f09e6bf0c6ceea6727bd0049781256a28cab0e5 Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-3-marcandre.lureau@redhat.com>
2023-06-27ui/egl: export qemu_egl_get_error_string()Marc-André Lureau
It will be used from other units. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230606115658.677673-2-marcandre.lureau@redhat.com>
2023-06-27ui/gtk: making dmabuf NULL when it's released.Dongwon Kim
Set vc->gfx.guest_fb.dmabuf to NULL to prevent any further access to it after the dmabuf is released. v2: move declaration of vc inside ifdef Cc: Gerd Hoffmann <kraxel@redhat.com> Cc: Marc-André Lureau <marcandre.lureau@redhat.com> Cc: Vivek Kasireddy <vivek.kasireddy@intel.com> Signed-off-by: Dongwon Kim <dongwon.kim@intel.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-ID: <20230627005316.5627-1-dongwon.kim@intel.com>
2023-06-27virtio-gpu: OUT_OF_MEMORY if failing to create udmabufDongwon Kim
Respond with VIRTIO_GPU_RESP_ERR_OUT_OF_MEMORY if it fails to create an udmabuf for the blob resource. v2: consolidated return statments and removed an unnecessary style change Cc: Gerd Hoffmann <kraxel@redhat.com> Cc: Marc-André Lureau <marcandre.lureau@redhat.com> Cc: Vivek Kasireddy <vivek.kasireddy@intel.com> Signed-off-by: Dongwon Kim <dongwon.kim@intel.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-ID: <20230627003453.5321-1-dongwon.kim@intel.com>
2023-06-27ui/gtk: set the area of the scanout texture correctlyDongwon Kim
x and y offsets and width and height of the scanout texture is not correctly configured in case guest scanout frame is dmabuf. Cc: Gerd Hoffmann <kraxel@redhat.com> Cc: Marc-André Lureau <marcandre.lureau@redhat.com> Cc: Vivek Kasireddy <vivek.kasireddy@intel.com> Signed-off-by: Dongwon Kim <dongwon.kim@intel.com> Message-ID: <20230621213150.29573-1-dongwon.kim@intel.com>
2023-06-27virtio-gpu-udmabuf: create udmabuf for blob even when iov_cnt == 1Dongwon Kim
There were often cases where a scanout blob sometimes has just 1 entry that is linked to many pages in it. So just checking whether iov_cnt is 1 is not enough for screening small, non-scanout blobs. Therefore adding iov_len check as well to make sure it creates an udmabuf only for a scanout blob, which is at least bigger than one page size. Cc: Gerd Hoffmann <kraxel@redhat.com> Cc: Marc-André Lureau <marcandre.lureau@redhat.com> Cc: Vivek Kasireddy <vivek.kasireddy@intel.com> Signed-off-by: Dongwon Kim <dongwon.kim@intel.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-ID: <20230621222704.29932-1-dongwon.kim@intel.com>
2023-06-27virtio-gpu: Make non-gl display updates work again when blob=trueVivek Kasireddy
In the case where the console does not have gl capability, and if blob is set to true, make sure that the display updates still work. Commit e86a93f55463 accidentally broke this by misplacing the return statement (in resource_flush) causing the updates to be silently ignored. Fixes: e86a93f55463 ("virtio-gpu: splitting one extended mode guest fb into n-scanouts") Cc: Gerd Hoffmann <kraxel@redhat.com> Cc: Marc-André Lureau <marcandre.lureau@redhat.com> Cc: Dongwon Kim <dongwon.kim@intel.com> Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-ID: <20230623060454.3749910-1-vivek.kasireddy@intel.com>
2023-06-27ui/dbus: Expose a touch device interfaceBilal Elmoussaoui
So that clients making use of the DBus backend could send touch events through the new org.qemu.Display1.Touch interface Signed-off-by: Bilal Elmoussaoui <belmouss@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230619095337.9899-3-belmouss@redhat.com>
2023-06-27ui/touch: Move event handling to a common helperBilal Elmoussaoui
To share code between the GTK and DBus UI bakcends see the next commit for details Signed-off-by: Bilal Elmoussaoui <belmouss@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230619095337.9899-2-belmouss@redhat.com>
2023-06-27chardev/char-win-stdio: Support VT sequences on Windows 11 hostZhang Huasen
If the monitor or the serial port use STDIO as backend on Windows 11 host, e.g. -nographic options is used, the monitor or the guest Linux do not response to arrow keys. When Windows creates a console, ENABLE_VIRTUAL_PROCESS_INPUT is disabled by default. Arrow keys cannot be retrieved by ReadFile or ReadConsoleInput functions. Add ENABLE_VIRTUAL_PROCESS_INPUT to the flag which is passed to SetConsoleMode, when opening stdio console. Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1674 Signed-off-by: Zhang Huasen <huasenzhang@foxmail.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <tencent_8DA57B405D427A560FD40F8FB0C0B1ADDE09@qq.com>
2023-06-27virtio-gpu: Optimize 2D resource data transferKeqian Zhu
The following points sometimes can reduce much data to copy: 1. When width matches, we can transfer data with one call of iov_to_buf(). 2. Only the required height need to transfer, not whole image. Signed-off-by: Keqian Zhu <zhukeqian1@huawei.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230612021358.25068-1-zhukeqian1@huawei.com>
2023-06-27ui/sdl2: OpenGL window contextAntonio Caggiano
When OpenGL is enabled, create only the OpenGL context, ignoring the SDL renderer as it is unused anyway. Signed-off-by: Antonio Caggiano <quic_acaggian@quicinc.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230612091959.2983-1-quic_acaggian@quicinc.com>
2023-06-27egl: no need to lookup EGL functions manuallyMarc-André Lureau
libepoxy handles loading the function pointer and dispatching the call, so you don't have to worry about it. Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <20230515132455.1025608-1-marcandre.lureau@redhat.com>
2023-06-27ui: return NULL when getting cursor without a consoleMarc-André Lureau
VNC may try to get the current cursor even when there are no consoles and crashes. Simple reproducer is qemu with -nodefaults. Fixes: (again) https://gitlab.com/qemu-project/qemu/-/issues/1548 Fixes: commit 385ac97f8 ("ui: keep current cursor with QemuConsole") Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Message-Id: <20230428154807.2143652-1-marcandre.lureau@redhat.com>
2023-06-27icount: don't adjust virtual time backwards after warpNicholas Piggin
The icount-based QEMU_CLOCK_VIRTUAL runs ahead of the RT clock at times. When warping, it is possible it is still ahead at the end of the warp, which causes icount adaptive mode to adjust it backward. This can result in the machine observing time going backwards. Prevent this by clamping adaptive adjustment to 0 at minimum. Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Message-ID: <20230627061406.241847-1-npiggin@gmail.com> Cc: qemu-stable@nongnu.org Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2023-06-26Merge tag 'pull-tcg-20230626' of https://gitlab.com/rth7680/qemu into stagingRichard Henderson
accel/tcg: Replace target_ulong in some APIs accel/tcg: Remove CONFIG_PROFILER accel/tcg: Store some tlb flags in CPUTLBEntryFull tcg: Issue memory barriers as required for the guest memory model tcg: Fix temporary variable in tcg_gen_gvec_andcs # -----BEGIN PGP SIGNATURE----- # # iQFRBAABCgA7FiEEekgeeIaLTbaoWgXAZN846K9+IV8FAmSZsPgdHHJpY2hhcmQu # aGVuZGVyc29uQGxpbmFyby5vcmcACgkQZN846K9+IV+kWAf+ODI9qRvVbb4/uYv8 # k7wMhCxX9kk5bRVr+QcqDn9RekAdsyOKSdkAAv4NeRFqHs3ukxhMxu0N2aiVXGDw # WtpsV73FrivAXaCxRj0aaYCsX8qFUQM4eWORZX2+V4AO0BtMHx1loK3bUQwdBTqN # jgkpn8BYeFdfUJjvvEj9XeSJ7s0n/p7esaf6VKajef/PbrcgYAeHg72tb5Vv5LTI # oxhU4icpaq/FT+SolnGzh4nRV7yqji9qFJ2INb0Uanx/WxCMD6CQJ0rDw55UouvH # t7zGDn8FKDZJGQGxAbUav3evqWcBlkG5VzuhQli3P1+WbGF9jV0KI1nelOuafCKI # 0enECg== # =XvZb # -----END PGP SIGNATURE----- # gpg: Signature made Mon 26 Jun 2023 05:38:32 PM CEST # gpg: using RSA key 7A481E78868B4DB6A85A05C064DF38E8AF7E215F # gpg: issuer "richard.henderson@linaro.org" # gpg: Good signature from "Richard Henderson <richard.henderson@linaro.org>" [ultimate] * tag 'pull-tcg-20230626' of https://gitlab.com/rth7680/qemu: (22 commits) accel/tcg: Renumber TLB_DISCARD_WRITE accel/tcg: Move TLB_WATCHPOINT to TLB_SLOW_FLAGS_MASK accel/tcg: Store some tlb flags in CPUTLBEntryFull accel/tcg: Remove check_tcg_memory_orders_compatible tcg: Add host memory barriers to cpu_ldst.h interfaces tcg: Do not elide memory barriers for !CF_PARALLEL in system mode target/microblaze: Define TCG_GUEST_DEFAULT_MO tcg: Fix temporary variable in tcg_gen_gvec_andcs accel/tcg: remove CONFIG_PROFILER tests/plugin: Remove duplicate insn log from libinsn.so softfloat: use QEMU_FLATTEN to avoid mistaken isra inlining cpu: Replace target_ulong with hwaddr in tb_invalidate_phys_addr() accel/tcg: Replace target_ulong with vaddr in translator_*() accel/tcg: Replace target_ulong with vaddr in *_mmu_lookup() accel: Replace target_ulong with vaddr in probe_*() accel/tcg: Widen pc to vaddr in CPUJumpCache accel/tcg/cpu-exec.c: Widen pc to vaddr accel/tcg/cputlb.c: Widen addr in MMULookupPageData accel/tcg/cputlb.c: Widen CPUTLBEntry access functions target: Widen pc/cs_base in cpu_get_tb_cpu_state ... Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-06-26accel/tcg: Renumber TLB_DISCARD_WRITERichard Henderson
Move to fill a hole in the set of bits. Reduce the total number of tlb bits by 1. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-06-26accel/tcg: Move TLB_WATCHPOINT to TLB_SLOW_FLAGS_MASKRichard Henderson
This frees up one bit of the primary tlb flags without impacting the TLB_NOTDIRTY logic. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-06-26accel/tcg: Store some tlb flags in CPUTLBEntryFullRichard Henderson
We have run out of bits we can use within the CPUTLBEntry comparators, as TLB_FLAGS_MASK cannot overlap alignment. Store slow_flags[] in CPUTLBEntryFull, and merge with the flags from the comparator. A new TLB_FORCE_SLOW bit is set within the comparator as an indication that the slow path must be used. Move TLB_BSWAP to TLB_SLOW_FLAGS_MASK. Since we are out of bits, we cannot create a new bit without moving an old one. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-06-26accel/tcg: Remove check_tcg_memory_orders_compatibleRichard Henderson
We now issue host memory barriers to match the guest memory order. Continue to disable MTTCG only if the guest has not been ported. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-06-26tcg: Add host memory barriers to cpu_ldst.h interfacesRichard Henderson
Bring the helpers into line with the rest of tcg in respecting guest memory ordering. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-06-26tcg: Do not elide memory barriers for !CF_PARALLEL in system modeRichard Henderson
The virtio devices require proper memory ordering between the vcpus and the iothreads. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2023-06-26target/microblaze: Define TCG_GUEST_DEFAULT_MORichard Henderson
The microblaze architecture does not reorder instructions. While there is an MBAR wait-for-data-access instruction, this concerns synchronizing with DMA. This should have been defined when enabling MTTCG. Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Edgar E. Iglesias <edgar@zeroasic.com> Fixes: d449561b130 ("configure: microblaze: Enable mttcg") Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
generated by cgit v1.2.3 (git 2.26.2) at 2025-03-17 18:10:41 +0000