diff options
98 files changed, 6103 insertions, 2662 deletions
diff --git a/.travis.yml b/.travis.yml index d472fd650b..93fd0164a0 100644 --- a/.travis.yml +++ b/.travis.yml @@ -1,14 +1,13 @@ -# The current Travis default is a container based 14.04 Trust on EC2 +# The current Travis default is a VM based 16.04 Xenial on GCE # Additional builds with specific requirements for a full VM need to # be added as additional matrix: entries later on -sudo: false -dist: trusty +dist: xenial language: c -python: - - "2.6" compiler: - gcc cache: ccache + + addons: apt: packages: @@ -35,10 +34,17 @@ addons: - libssh2-1-dev - liburcu-dev - libusb-1.0-0-dev - - libvte-2.90-dev + - libvte-2.91-dev - sparse - uuid-dev - gcovr + homebrew: + packages: + - libffi + - gettext + - glib + - pixman + # The channel name "irc.oftc.net#qemu" is encrypted against qemu/qemu # to prevent IRC notifications from forks. This was created using: @@ -49,88 +55,145 @@ notifications: - secure: "F7GDRgjuOo5IUyRLqSkmDL7kvdU4UcH3Lm/W2db2JnDHTGCqgEdaYEYKciyCLZ57vOTsTsOgesN8iUT7hNHBd1KWKjZe9KDTZWppWRYVwAwQMzVeSOsbbU4tRoJ6Pp+3qhH1Z0eGYR9ZgKYAoTumDFgSAYRp4IscKS8jkoedOqM=" on_success: change on_failure: always + + env: global: - SRC_DIR="." - BUILD_DIR="." - - TEST_CMD="make check" - - MAKEFLAGS="-j3" - matrix: - - CONFIG="--disable-system" - - CONFIG="--disable-user" - - CONFIG="--enable-debug --enable-debug-tcg" - - CONFIG="--disable-linux-aio --disable-cap-ng --disable-attr --disable-brlapi --disable-uuid --disable-libusb --disable-user" - - CONFIG="--enable-modules --disable-linux-user" - - CONFIG="--with-coroutine=ucontext --disable-linux-user" - - CONFIG="--with-coroutine=sigaltstack --disable-linux-user" + - TEST_CMD="make check -j3 V=1" + + git: # we want to do this ourselves submodules: false -before_install: - - if [ "$TRAVIS_OS_NAME" == "osx" ]; then brew update ; fi - - if [ "$TRAVIS_OS_NAME" == "osx" ]; then brew install libffi gettext glib pixman ; fi - - git submodule update --init --recursive capstone dtc ui/keycodemapdb + + before_script: - mkdir -p ${BUILD_DIR} && cd ${BUILD_DIR} - ${SRC_DIR}/configure ${CONFIG} || { cat config.log && exit 1; } script: - - make ${MAKEFLAGS} && ${TEST_CMD} + - make -j3 && ${TEST_CMD} + + matrix: include: + - env: + - CONFIG="--disable-system" + + + - env: + - CONFIG="--disable-user" + + + - env: + - CONFIG="--enable-debug --enable-debug-tcg" + + + - env: + - CONFIG="--disable-linux-aio --disable-cap-ng --disable-attr --disable-brlapi --disable-uuid --disable-libusb --disable-user" + + + - env: + - CONFIG="--enable-modules --disable-linux-user" + + + - env: + - CONFIG="--with-coroutine=ucontext --disable-linux-user" + + + - env: + - CONFIG="--with-coroutine=sigaltstack --disable-linux-user" + + # Test out-of-tree builds - - env: CONFIG="--enable-debug --enable-debug-tcg" - BUILD_DIR="out-of-tree/build/dir" SRC_DIR="../../.." + - env: + - CONFIG="--enable-debug --enable-debug-tcg" + - BUILD_DIR="out-of-tree/build/dir" SRC_DIR="../../.." + + # Test with Clang for compile portability (Travis uses clang-5.0) - - env: CONFIG="--disable-system" + - env: + - CONFIG="--disable-system" compiler: clang - - env: CONFIG="--disable-user" + + + - env: + - CONFIG="--disable-user" compiler: clang + + # gprof/gcov are GCC features - - env: CONFIG="--enable-gprof --enable-gcov --disable-pie --target-list=aarch64-softmmu,arm-softmmu,i386-softmmu,mips-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu" + - env: + - CONFIG="--enable-gprof --enable-gcov --disable-pie --target-list=aarch64-softmmu,arm-softmmu,i386-softmmu,mips-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu" after_success: - ${SRC_DIR}/scripts/travis/coverage-summary.sh - compiler: gcc + + # We manually include builds which we disable "make check" for - - env: CONFIG="--enable-debug --enable-tcg-interpreter" - TEST_CMD="" - compiler: gcc + - env: + - CONFIG="--enable-debug --enable-tcg-interpreter" + - TEST_CMD="" + + # We don't need to exercise every backend with every front-end - - env: CONFIG="--enable-trace-backends=log,simple,syslog --disable-system" - TEST_CMD="" - compiler: gcc - - env: CONFIG="--enable-trace-backends=ftrace --target-list=x86_64-softmmu" - TEST_CMD="" - compiler: gcc - - env: CONFIG="--enable-trace-backends=ust --target-list=x86_64-softmmu" - TEST_CMD="" - compiler: gcc - - env: CONFIG="--disable-tcg" - TEST_CMD="" - compiler: gcc + - env: + - CONFIG="--enable-trace-backends=log,simple,syslog --disable-system" + - TEST_CMD="" + + + - env: + - CONFIG="--enable-trace-backends=ftrace --target-list=x86_64-softmmu" + - TEST_CMD="" + + + - env: + - CONFIG="--enable-trace-backends=ust --target-list=x86_64-softmmu" + - TEST_CMD="" + + + - env: + - CONFIG="--disable-tcg" + - TEST_CMD="" + + # MacOSX builds - - env: CONFIG="--target-list=aarch64-softmmu,arm-softmmu,i386-softmmu,mips-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu" + - env: + - CONFIG="--target-list=aarch64-softmmu,arm-softmmu,i386-softmmu,mips-softmmu,mips64-softmmu,ppc64-softmmu,riscv64-softmmu,s390x-softmmu,x86_64-softmmu" os: osx osx_image: xcode9.4 compiler: clang - - env: CONFIG="--target-list=i386-softmmu,ppc-softmmu,ppc64-softmmu,m68k-softmmu,x86_64-softmmu" + + + - env: + - CONFIG="--target-list=i386-softmmu,ppc-softmmu,ppc64-softmmu,m68k-softmmu,x86_64-softmmu" os: osx osx_image: xcode10 compiler: clang + + # Python builds - - env: CONFIG="--target-list=x86_64-softmmu" + - env: + - CONFIG="--target-list=x86_64-softmmu" python: - "3.0" - - env: CONFIG="--target-list=x86_64-softmmu" + + + - env: + - CONFIG="--target-list=x86_64-softmmu" python: - "3.6" + + # Acceptance (Functional) tests - - env: CONFIG="--python=/usr/bin/python3 --target-list=x86_64-softmmu" - TEST_CMD="make AVOCADO_SHOW=app check-acceptance" + - env: + - CONFIG="--python=/usr/bin/python3 --target-list=x86_64-softmmu" + - TEST_CMD="make AVOCADO_SHOW=app check-acceptance" addons: apt: packages: - python3-pip - - python3.4-venv + - python3.5-venv # Using newer GCC with sanitizers - addons: apt: @@ -164,7 +227,7 @@ matrix: - libssh2-1-dev - liburcu-dev - libusb-1.0-0-dev - - libvte-2.90-dev + - libvte-2.91-dev - sparse - uuid-dev language: generic @@ -175,11 +238,8 @@ matrix: - TEST_CMD="" before_script: - ./configure ${CONFIG} --extra-cflags="-g3 -O0 -fsanitize=thread -fuse-ld=gold" || { cat config.log && exit 1; } + + - env: - CONFIG="--disable-system --disable-docs" - - TEST_CMD="make check-tcg" - script: - - make ${MAKEFLAGS} && ${TEST_CMD} ${MAKEFLAGS} - sudo: required - dist: trusty - compiler: gcc + - TEST_CMD="make -j3 check-tcg V=1" diff --git a/MAINTAINERS b/MAINTAINERS index 87f90721b9..af339b86db 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -407,6 +407,7 @@ Guest CPU Cores (Xen): X86 M: Stefano Stabellini <sstabellini@kernel.org> M: Anthony Perard <anthony.perard@citrix.com> +M: Paul Durrant <paul.durrant@citrix.com> L: xen-devel@lists.xenproject.org S: Supported F: */xen* @@ -414,10 +415,12 @@ F: hw/9pfs/xen-9p-backend.c F: hw/char/xen_console.c F: hw/display/xenfb.c F: hw/net/xen_nic.c -F: hw/block/xen_* +F: hw/block/xen* +F: hw/block/dataplane/xen* F: hw/xen/ F: hw/xenpv/ F: hw/i386/xen/ +F: include/hw/block/dataplane/xen* F: include/hw/xen/ F: include/sysemu/xen-mapcache.h @@ -2494,6 +2497,12 @@ M: Daniel P. Berrange <berrange@redhat.com> S: Odd Fixes F: docs/devel/build-system.txt +GIT Data Mining Config +M: Alex Bennée <alex.bennee@linaro.org> +S: Odd Fixes +F: gitdm.config +F: contrib/gitdm/* + Incompatible changes R: libvir-list@redhat.com F: qemu-deprecated.texi diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index 89fd1d7f8b..6b688394e4 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -625,7 +625,6 @@ void bdrv_clear_dirty_bitmap(BdrvDirtyBitmap *bitmap, HBitmap **out) void bdrv_restore_dirty_bitmap(BdrvDirtyBitmap *bitmap, HBitmap *backup) { HBitmap *tmp = bitmap->bitmap; - assert(bdrv_dirty_bitmap_enabled(bitmap)); assert(!bdrv_dirty_bitmap_readonly(bitmap)); bitmap->bitmap = backup; hbitmap_free(tmp); diff --git a/blockdev-nbd.c b/blockdev-nbd.c index 1d170c80b8..c76d5416b9 100644 --- a/blockdev-nbd.c +++ b/blockdev-nbd.c @@ -140,7 +140,8 @@ void qmp_nbd_server_start(SocketAddressLegacy *addr, } void qmp_nbd_server_add(const char *device, bool has_name, const char *name, - bool has_writable, bool writable, Error **errp) + bool has_writable, bool writable, + bool has_bitmap, const char *bitmap, Error **errp) { BlockDriverState *bs = NULL; BlockBackend *on_eject_blk; @@ -174,14 +175,13 @@ void qmp_nbd_server_add(const char *device, bool has_name, const char *name, writable = false; } - exp = nbd_export_new(bs, 0, -1, writable ? 0 : NBD_FLAG_READ_ONLY, + exp = nbd_export_new(bs, 0, -1, name, NULL, bitmap, + writable ? 0 : NBD_FLAG_READ_ONLY, NULL, false, on_eject_blk, errp); if (!exp) { return; } - nbd_export_set_name(exp, name); - /* The list of named exports has a strong reference to this export now and * our only way of accessing it is through nbd_export_find(), so we can drop * the strong reference that is @exp. */ @@ -214,31 +214,13 @@ void qmp_nbd_server_remove(const char *name, void qmp_nbd_server_stop(Error **errp) { - nbd_export_close_all(); - - nbd_server_free(nbd_server); - nbd_server = NULL; -} - -void qmp_x_nbd_server_add_bitmap(const char *name, const char *bitmap, - bool has_bitmap_export_name, - const char *bitmap_export_name, - Error **errp) -{ - NBDExport *exp; - if (!nbd_server) { error_setg(errp, "NBD server not running"); return; } - exp = nbd_export_find(name); - if (exp == NULL) { - error_setg(errp, "Export '%s' is not found", name); - return; - } + nbd_export_close_all(); - nbd_export_bitmap(exp, bitmap, - has_bitmap_export_name ? bitmap_export_name : bitmap, - errp); + nbd_server_free(nbd_server); + nbd_server = NULL; } diff --git a/blockdev.c b/blockdev.c index 1cc893fe61..a8fa8748a9 100644 --- a/blockdev.c +++ b/blockdev.c @@ -1339,7 +1339,7 @@ struct BlkActionState { const BlkActionOps *ops; JobTxn *block_job_txn; TransactionProperties *txn_props; - QSIMPLEQ_ENTRY(BlkActionState) entry; + QTAILQ_ENTRY(BlkActionState) entry; }; /* internal snapshot private data */ @@ -1963,7 +1963,7 @@ static void block_dirty_bitmap_add_prepare(BlkActionState *common, action->has_granularity, action->granularity, action->has_persistent, action->persistent, action->has_autoload, action->autoload, - action->has_x_disabled, action->x_disabled, + action->has_disabled, action->disabled, &local_err); if (!local_err) { @@ -2048,7 +2048,7 @@ static void block_dirty_bitmap_enable_prepare(BlkActionState *common, return; } - action = common->action->u.x_block_dirty_bitmap_enable.data; + action = common->action->u.block_dirty_bitmap_enable.data; state->bitmap = block_dirty_bitmap_lookup(action->node, action->name, NULL, @@ -2089,7 +2089,7 @@ static void block_dirty_bitmap_disable_prepare(BlkActionState *common, return; } - action = common->action->u.x_block_dirty_bitmap_disable.data; + action = common->action->u.block_dirty_bitmap_disable.data; state->bitmap = block_dirty_bitmap_lookup(action->node, action->name, NULL, @@ -2119,33 +2119,28 @@ static void block_dirty_bitmap_disable_abort(BlkActionState *common) } } +static BdrvDirtyBitmap *do_block_dirty_bitmap_merge(const char *node, + const char *target, + strList *bitmaps, + HBitmap **backup, + Error **errp); + static void block_dirty_bitmap_merge_prepare(BlkActionState *common, Error **errp) { BlockDirtyBitmapMerge *action; BlockDirtyBitmapState *state = DO_UPCAST(BlockDirtyBitmapState, common, common); - BdrvDirtyBitmap *merge_source; if (action_check_completion_mode(common, errp) < 0) { return; } - action = common->action->u.x_block_dirty_bitmap_merge.data; - state->bitmap = block_dirty_bitmap_lookup(action->node, - action->dst_name, - &state->bs, - errp); - if (!state->bitmap) { - return; - } - - merge_source = bdrv_find_dirty_bitmap(state->bs, action->src_name); - if (!merge_source) { - return; - } + action = common->action->u.block_dirty_bitmap_merge.data; - bdrv_merge_dirty_bitmap(state->bitmap, merge_source, &state->backup, errp); + state->bitmap = do_block_dirty_bitmap_merge(action->node, action->target, + action->bitmaps, &state->backup, + errp); } static void abort_prepare(BlkActionState *common, Error **errp) @@ -2209,17 +2204,17 @@ static const BlkActionOps actions[] = { .commit = block_dirty_bitmap_free_backup, .abort = block_dirty_bitmap_restore, }, - [TRANSACTION_ACTION_KIND_X_BLOCK_DIRTY_BITMAP_ENABLE] = { + [TRANSACTION_ACTION_KIND_BLOCK_DIRTY_BITMAP_ENABLE] = { .instance_size = sizeof(BlockDirtyBitmapState), .prepare = block_dirty_bitmap_enable_prepare, .abort = block_dirty_bitmap_enable_abort, }, - [TRANSACTION_ACTION_KIND_X_BLOCK_DIRTY_BITMAP_DISABLE] = { + [TRANSACTION_ACTION_KIND_BLOCK_DIRTY_BITMAP_DISABLE] = { .instance_size = sizeof(BlockDirtyBitmapState), .prepare = block_dirty_bitmap_disable_prepare, .abort = block_dirty_bitmap_disable_abort, }, - [TRANSACTION_ACTION_KIND_X_BLOCK_DIRTY_BITMAP_MERGE] = { + [TRANSACTION_ACTION_KIND_BLOCK_DIRTY_BITMAP_MERGE] = { .instance_size = sizeof(BlockDirtyBitmapState), .prepare = block_dirty_bitmap_merge_prepare, .commit = block_dirty_bitmap_free_backup, @@ -2266,8 +2261,8 @@ void qmp_transaction(TransactionActionList *dev_list, BlkActionState *state, *next; Error *local_err = NULL; - QSIMPLEQ_HEAD(, BlkActionState) snap_bdrv_states; - QSIMPLEQ_INIT(&snap_bdrv_states); + QTAILQ_HEAD(, BlkActionState) snap_bdrv_states; + QTAILQ_INIT(&snap_bdrv_states); /* Does this transaction get canceled as a group on failure? * If not, we don't really need to make a JobTxn. @@ -2298,7 +2293,7 @@ void qmp_transaction(TransactionActionList *dev_list, state->action = dev_info; state->block_job_txn = block_job_txn; state->txn_props = props; - QSIMPLEQ_INSERT_TAIL(&snap_bdrv_states, state, entry); + QTAILQ_INSERT_TAIL(&snap_bdrv_states, state, entry); state->ops->prepare(state, &local_err); if (local_err) { @@ -2307,7 +2302,7 @@ void qmp_transaction(TransactionActionList *dev_list, } } - QSIMPLEQ_FOREACH(state, &snap_bdrv_states, entry) { + QTAILQ_FOREACH(state, &snap_bdrv_states, entry) { if (state->ops->commit) { state->ops->commit(state); } @@ -2318,13 +2313,13 @@ void qmp_transaction(TransactionActionList *dev_list, delete_and_fail: /* failure, and it is all-or-none; roll back all operations */ - QSIMPLEQ_FOREACH(state, &snap_bdrv_states, entry) { + QTAILQ_FOREACH_REVERSE(state, &snap_bdrv_states, entry) { if (state->ops->abort) { state->ops->abort(state); } } exit: - QSIMPLEQ_FOREACH_SAFE(state, &snap_bdrv_states, entry, next) { + QTAILQ_FOREACH_SAFE(state, &snap_bdrv_states, entry, next) { if (state->ops->clean) { state->ops->clean(state); } @@ -2935,7 +2930,7 @@ void qmp_block_dirty_bitmap_clear(const char *node, const char *name, bdrv_clear_dirty_bitmap(bitmap, NULL); } -void qmp_x_block_dirty_bitmap_enable(const char *node, const char *name, +void qmp_block_dirty_bitmap_enable(const char *node, const char *name, Error **errp) { BlockDriverState *bs; @@ -2956,7 +2951,7 @@ void qmp_x_block_dirty_bitmap_enable(const char *node, const char *name, bdrv_enable_dirty_bitmap(bitmap); } -void qmp_x_block_dirty_bitmap_disable(const char *node, const char *name, +void qmp_block_dirty_bitmap_disable(const char *node, const char *name, Error **errp) { BlockDriverState *bs; @@ -2977,24 +2972,56 @@ void qmp_x_block_dirty_bitmap_disable(const char *node, const char *name, bdrv_disable_dirty_bitmap(bitmap); } -void qmp_x_block_dirty_bitmap_merge(const char *node, const char *dst_name, - const char *src_name, Error **errp) +static BdrvDirtyBitmap *do_block_dirty_bitmap_merge(const char *node, + const char *target, + strList *bitmaps, + HBitmap **backup, + Error **errp) { BlockDriverState *bs; - BdrvDirtyBitmap *dst, *src; + BdrvDirtyBitmap *dst, *src, *anon; + strList *lst; + Error *local_err = NULL; - dst = block_dirty_bitmap_lookup(node, dst_name, &bs, errp); + dst = block_dirty_bitmap_lookup(node, target, &bs, errp); if (!dst) { - return; + return NULL; } - src = bdrv_find_dirty_bitmap(bs, src_name); - if (!src) { - error_setg(errp, "Dirty bitmap '%s' not found", src_name); - return; + anon = bdrv_create_dirty_bitmap(bs, bdrv_dirty_bitmap_granularity(dst), + NULL, errp); + if (!anon) { + return NULL; + } + + for (lst = bitmaps; lst; lst = lst->next) { + src = bdrv_find_dirty_bitmap(bs, lst->value); + if (!src) { + error_setg(errp, "Dirty bitmap '%s' not found", lst->value); + dst = NULL; + goto out; + } + + bdrv_merge_dirty_bitmap(anon, src, NULL, &local_err); + if (local_err) { + error_propagate(errp, local_err); + dst = NULL; + goto out; + } } - bdrv_merge_dirty_bitmap(dst, src, NULL, errp); + /* Merge into dst; dst is unchanged on failure. */ + bdrv_merge_dirty_bitmap(dst, anon, backup, errp); + + out: + bdrv_release_dirty_bitmap(bs, anon); + return dst; +} + +void qmp_block_dirty_bitmap_merge(const char *node, const char *target, + strList *bitmaps, Error **errp) +{ + do_block_dirty_bitmap_merge(node, target, bitmaps, NULL, errp); } BlockDirtyBitmapSha256 *qmp_x_debug_block_dirty_bitmap_sha256(const char *node, @@ -357,7 +357,6 @@ vnc_png="" xkbcommon="" xen="" xen_ctrl_version="" -xen_pv_domain_build="no" xen_pci_passthrough="" linux_aio="" cap_ng="" @@ -1119,10 +1118,6 @@ for opt do ;; --enable-xen-pci-passthrough) xen_pci_passthrough="yes" ;; - --disable-xen-pv-domain-build) xen_pv_domain_build="no" - ;; - --enable-xen-pv-domain-build) xen_pv_domain_build="yes" - ;; --disable-brlapi) brlapi="no" ;; --enable-brlapi) brlapi="yes" @@ -1685,8 +1680,6 @@ Advanced options (experts only): --tls-priority default TLS protocol/cipher priority string --enable-gprof QEMU profiling with gprof --enable-profiler profiler support - --enable-xen-pv-domain-build - xen pv domain builder --enable-debug-stack-usage track the maximum stack usage of stacks created by qemu_alloc_stack @@ -2678,12 +2671,6 @@ if test "$xen_pci_passthrough" != "no"; then fi fi -if test "$xen_pv_domain_build" = "yes" && - test "$xen" != "yes"; then - error_exit "User requested Xen PV domain builder support" \ - "which requires Xen support." -fi - ########################################## # Windows Hypervisor Platform accelerator (WHPX) check if test "$whpx" != "no" ; then @@ -6069,7 +6056,6 @@ fi echo "xen support $xen" if test "$xen" = "yes" ; then echo "xen ctrl version $xen_ctrl_version" - echo "pv dom build $xen_pv_domain_build" fi echo "brlapi support $brlapi" echo "bluez support $bluez" @@ -6539,9 +6525,6 @@ fi if test "$xen" = "yes" ; then echo "CONFIG_XEN_BACKEND=y" >> $config_host_mak echo "CONFIG_XEN_CTRL_INTERFACE_VERSION=$xen_ctrl_version" >> $config_host_mak - if test "$xen_pv_domain_build" = "yes" ; then - echo "CONFIG_XEN_PV_DOMAIN_BUILD=y" >> $config_host_mak - fi fi if test "$linux_aio" = "yes" ; then echo "CONFIG_LINUX_AIO=y" >> $config_host_mak @@ -7487,7 +7470,7 @@ alpha) esac if test "$gprof" = "yes" ; then - echo "TARGET_GPROF=yes" >> $config_target_mak + echo "TARGET_GPROF=y" >> $config_target_mak if test "$target_linux_user" = "yes" ; then cflags="-p $cflags" ldflags="-p $ldflags" diff --git a/contrib/gitdm/domain-map b/contrib/gitdm/domain-map index 8cbbcfe93d..0ab41ee27a 100644 --- a/contrib/gitdm/domain-map +++ b/contrib/gitdm/domain-map @@ -9,7 +9,9 @@ greensocs.com GreenSocs ibm.com IBM igalia.com Igalia linaro.org Linaro +nokia.com Nokia oracle.com Oracle +proxmox.com Proxmox redhat.com Red Hat siemens.com Siemens sifive.com SiFive diff --git a/contrib/gitdm/group-map-ibm b/contrib/gitdm/group-map-ibm index b66db5f4a8..22727319b3 100644 --- a/contrib/gitdm/group-map-ibm +++ b/contrib/gitdm/group-map-ibm @@ -2,5 +2,12 @@ # Some IBM contributors submit via another domain # +aik@ozlabs.ru +andrew@aj.id.au +benh@kernel.crashing.org clg@kaod.org +danielhb413@gmail.com groug@kaod.org +jcfaracco@gmail.com +joel@jms.id.au +sjitindarsingh@gmail.com diff --git a/contrib/gitdm/group-map-wavecomp b/contrib/gitdm/group-map-wavecomp index c571a52c65..2801a966b6 100644 --- a/contrib/gitdm/group-map-wavecomp +++ b/contrib/gitdm/group-map-wavecomp @@ -9,6 +9,7 @@ amarkovic@wavecomp.com arikalo@wavecomp.com dnikolic@wavecomp.com james.hogan@mips.com +leon.alrae@imgtec.com matthew.fortune@mips.com paul.burton@imgtec.com pburton@wavecomp.com diff --git a/default-configs/riscv32-softmmu.mak b/default-configs/riscv32-softmmu.mak index dbc9398284..c9c5971409 100644 --- a/default-configs/riscv32-softmmu.mak +++ b/default-configs/riscv32-softmmu.mak @@ -1,6 +1,7 @@ # Default configuration for riscv-softmmu include pci.mak +include usb.mak CONFIG_SERIAL=y CONFIG_VIRTIO_MMIO=y diff --git a/default-configs/riscv64-softmmu.mak b/default-configs/riscv64-softmmu.mak index dbc9398284..c9c5971409 100644 --- a/default-configs/riscv64-softmmu.mak +++ b/default-configs/riscv64-softmmu.mak @@ -1,6 +1,7 @@ # Default configuration for riscv-softmmu include pci.mak +include usb.mak CONFIG_SERIAL=y CONFIG_VIRTIO_MMIO=y @@ -2326,7 +2326,7 @@ void hmp_nbd_server_start(Monitor *mon, const QDict *qdict) } qmp_nbd_server_add(info->value->device, false, NULL, - true, writable, &local_err); + true, writable, false, NULL, &local_err); if (local_err != NULL) { qmp_nbd_server_stop(NULL); @@ -2347,7 +2347,8 @@ void hmp_nbd_server_add(Monitor *mon, const QDict *qdict) bool writable = qdict_get_try_bool(qdict, "writable", false); Error *local_err = NULL; - qmp_nbd_server_add(device, !!name, name, true, writable, &local_err); + qmp_nbd_server_add(device, !!name, name, true, writable, + false, NULL, &local_err); hmp_handle_error(mon, &local_err); } diff --git a/hw/9pfs/xen-9p-backend.c b/hw/9pfs/xen-9p-backend.c index 9015fe7773..25ab04d95a 100644 --- a/hw/9pfs/xen-9p-backend.c +++ b/hw/9pfs/xen-9p-backend.c @@ -12,7 +12,7 @@ #include "hw/hw.h" #include "hw/9pfs/9p.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include "hw/9pfs/xen-9pfs.h" #include "qapi/error.h" #include "qemu/config-file.h" @@ -45,7 +45,7 @@ typedef struct Xen9pfsRing { } Xen9pfsRing; typedef struct Xen9pfsDev { - struct XenDevice xendev; /* must be first */ + struct XenLegacyDevice xendev; /* must be first */ V9fsState state; char *path; char *security_model; @@ -56,7 +56,7 @@ typedef struct Xen9pfsDev { Xen9pfsRing *rings; } Xen9pfsDev; -static void xen_9pfs_disconnect(struct XenDevice *xendev); +static void xen_9pfs_disconnect(struct XenLegacyDevice *xendev); static void xen_9pfs_in_sg(Xen9pfsRing *ring, struct iovec *in_sg, @@ -243,7 +243,7 @@ static const V9fsTransport xen_9p_transport = { .push_and_notify = xen_9pfs_push_and_notify, }; -static int xen_9pfs_init(struct XenDevice *xendev) +static int xen_9pfs_init(struct XenLegacyDevice *xendev) { return 0; } @@ -305,7 +305,7 @@ static void xen_9pfs_evtchn_event(void *opaque) qemu_bh_schedule(ring->bh); } -static void xen_9pfs_disconnect(struct XenDevice *xendev) +static void xen_9pfs_disconnect(struct XenLegacyDevice *xendev) { Xen9pfsDev *xen_9pdev = container_of(xendev, Xen9pfsDev, xendev); int i; @@ -321,7 +321,7 @@ static void xen_9pfs_disconnect(struct XenDevice *xendev) } } -static int xen_9pfs_free(struct XenDevice *xendev) +static int xen_9pfs_free(struct XenLegacyDevice *xendev) { Xen9pfsDev *xen_9pdev = container_of(xendev, Xen9pfsDev, xendev); int i; @@ -354,7 +354,7 @@ static int xen_9pfs_free(struct XenDevice *xendev) return 0; } -static int xen_9pfs_connect(struct XenDevice *xendev) +static int xen_9pfs_connect(struct XenLegacyDevice *xendev) { Error *err = NULL; int i; @@ -467,7 +467,7 @@ out: return -1; } -static void xen_9pfs_alloc(struct XenDevice *xendev) +static void xen_9pfs_alloc(struct XenLegacyDevice *xendev) { xenstore_write_be_str(xendev, "versions", VERSIONS); xenstore_write_be_int(xendev, "max-rings", MAX_RINGS); diff --git a/hw/block/Makefile.objs b/hw/block/Makefile.objs index 53ce5751ae..e206b8e712 100644 --- a/hw/block/Makefile.objs +++ b/hw/block/Makefile.objs @@ -4,7 +4,7 @@ common-obj-$(CONFIG_SSI_M25P80) += m25p80.o common-obj-$(CONFIG_NAND) += nand.o common-obj-$(CONFIG_PFLASH_CFI01) += pflash_cfi01.o common-obj-$(CONFIG_PFLASH_CFI02) += pflash_cfi02.o -common-obj-$(CONFIG_XEN) += xen_disk.o +common-obj-$(CONFIG_XEN) += xen-block.o common-obj-$(CONFIG_ECC) += ecc.o common-obj-$(CONFIG_ONENAND) += onenand.o common-obj-$(CONFIG_NVME_PCI) += nvme.o diff --git a/hw/block/dataplane/Makefile.objs b/hw/block/dataplane/Makefile.objs index e786f66421..c6c68dbc00 100644 --- a/hw/block/dataplane/Makefile.objs +++ b/hw/block/dataplane/Makefile.objs @@ -1 +1,2 @@ obj-y += virtio-blk.o +obj-$(CONFIG_XEN) += xen-block.o diff --git a/hw/block/dataplane/xen-block.c b/hw/block/dataplane/xen-block.c new file mode 100644 index 0000000000..d0d8905a33 --- /dev/null +++ b/hw/block/dataplane/xen-block.c @@ -0,0 +1,827 @@ +/* + * Copyright (c) 2018 Citrix Systems Inc. + * (c) Gerd Hoffmann <kraxel@redhat.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; under version 2 of the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, see <http://www.gnu.org/licenses/>. + * + * Contributions after 2012-01-13 are licensed under the terms of the + * GNU GPL, version 2 or (at your option) any later version. + */ + +#include "qemu/osdep.h" +#include "qemu/error-report.h" +#include "qapi/error.h" +#include "hw/hw.h" +#include "hw/xen/xen_common.h" +#include "hw/block/xen_blkif.h" +#include "sysemu/block-backend.h" +#include "sysemu/iothread.h" +#include "xen-block.h" + +typedef struct XenBlockRequest { + blkif_request_t req; + int16_t status; + off_t start; + QEMUIOVector v; + void *buf; + size_t size; + int presync; + int aio_inflight; + int aio_errors; + XenBlockDataPlane *dataplane; + QLIST_ENTRY(XenBlockRequest) list; + BlockAcctCookie acct; +} XenBlockRequest; + +struct XenBlockDataPlane { + XenDevice *xendev; + XenEventChannel *event_channel; + unsigned int *ring_ref; + unsigned int nr_ring_ref; + void *sring; + int64_t file_blk; + int64_t file_size; + int protocol; + blkif_back_rings_t rings; + int more_work; + QLIST_HEAD(inflight_head, XenBlockRequest) inflight; + QLIST_HEAD(freelist_head, XenBlockRequest) freelist; + int requests_total; + int requests_inflight; + unsigned int max_requests; + BlockBackend *blk; + QEMUBH *bh; + IOThread *iothread; + AioContext *ctx; +}; + +static void reset_request(XenBlockRequest *request) +{ + memset(&request->req, 0, sizeof(request->req)); + request->status = 0; + request->start = 0; + request->size = 0; + request->presync = 0; + + request->aio_inflight = 0; + request->aio_errors = 0; + + request->dataplane = NULL; + memset(&request->list, 0, sizeof(request->list)); + memset(&request->acct, 0, sizeof(request->acct)); + + qemu_iovec_reset(&request->v); +} + +static XenBlockRequest *xen_block_start_request(XenBlockDataPlane *dataplane) +{ + XenBlockRequest *request = NULL; + + if (QLIST_EMPTY(&dataplane->freelist)) { + if (dataplane->requests_total >= dataplane->max_requests) { + goto out; + } + /* allocate new struct */ + request = g_malloc0(sizeof(*request)); + request->dataplane = dataplane; + /* + * We cannot need more pages per requests than this, and since we + * re-use requests, allocate the memory once here. It will be freed + * xen_block_dataplane_destroy() when the request list is freed. + */ + request->buf = qemu_memalign(XC_PAGE_SIZE, + BLKIF_MAX_SEGMENTS_PER_REQUEST * + XC_PAGE_SIZE); + dataplane->requests_total++; + qemu_iovec_init(&request->v, 1); + } else { + /* get one from freelist */ + request = QLIST_FIRST(&dataplane->freelist); + QLIST_REMOVE(request, list); + } + QLIST_INSERT_HEAD(&dataplane->inflight, request, list); + dataplane->requests_inflight++; + +out: + return request; +} + +static void xen_block_finish_request(XenBlockRequest *request) +{ + XenBlockDataPlane *dataplane = request->dataplane; + + QLIST_REMOVE(request, list); + dataplane->requests_inflight--; +} + +static void xen_block_release_request(XenBlockRequest *request) +{ + XenBlockDataPlane *dataplane = request->dataplane; + + QLIST_REMOVE(request, list); + reset_request(request); + request->dataplane = dataplane; + QLIST_INSERT_HEAD(&dataplane->freelist, request, list); + dataplane->requests_inflight--; +} + +/* + * translate request into iovec + start offset + * do sanity checks along the way + */ +static int xen_block_parse_request(XenBlockRequest *request) +{ + XenBlockDataPlane *dataplane = request->dataplane; + size_t len; + int i; + + switch (request->req.operation) { + case BLKIF_OP_READ: + break; + case BLKIF_OP_FLUSH_DISKCACHE: + request->presync = 1; + if (!request->req.nr_segments) { + return 0; + } + /* fall through */ + case BLKIF_OP_WRITE: + break; + case BLKIF_OP_DISCARD: + return 0; + default: + error_report("error: unknown operation (%d)", request->req.operation); + goto err; + }; + + if (request->req.operation != BLKIF_OP_READ && + blk_is_read_only(dataplane->blk)) { + error_report("error: write req for ro device"); + goto err; + } + + request->start = request->req.sector_number * dataplane->file_blk; + for (i = 0; i < request->req.nr_segments; i++) { + if (i == BLKIF_MAX_SEGMENTS_PER_REQUEST) { + error_report("error: nr_segments too big"); + goto err; + } + if (request->req.seg[i].first_sect > request->req.seg[i].last_sect) { + error_report("error: first > last sector"); + goto err; + } + if (request->req.seg[i].last_sect * dataplane->file_blk >= + XC_PAGE_SIZE) { + error_report("error: page crossing"); + goto err; + } + + len = (request->req.seg[i].last_sect - + request->req.seg[i].first_sect + 1) * dataplane->file_blk; + request->size += len; + } + if (request->start + request->size > dataplane->file_size) { + error_report("error: access beyond end of file"); + goto err; + } + return 0; + +err: + request->status = BLKIF_RSP_ERROR; + return -1; +} + +static int xen_block_copy_request(XenBlockRequest *request) +{ + XenBlockDataPlane *dataplane = request->dataplane; + XenDevice *xendev = dataplane->xendev; + XenDeviceGrantCopySegment segs[BLKIF_MAX_SEGMENTS_PER_REQUEST]; + int i, count; + int64_t file_blk = dataplane->file_blk; + bool to_domain = (request->req.operation == BLKIF_OP_READ); + void *virt = request->buf; + Error *local_err = NULL; + + if (request->req.nr_segments == 0) { + return 0; + } + + count = request->req.nr_segments; + + for (i = 0; i < count; i++) { + if (to_domain) { + segs[i].dest.foreign.ref = request->req.seg[i].gref; + segs[i].dest.foreign.offset = request->req.seg[i].first_sect * + file_blk; + segs[i].source.virt = virt; + } else { + segs[i].source.foreign.ref = request->req.seg[i].gref; + segs[i].source.foreign.offset = request->req.seg[i].first_sect * + file_blk; + segs[i].dest.virt = virt; + } + segs[i].len = (request->req.seg[i].last_sect - + request->req.seg[i].first_sect + 1) * file_blk; + virt += segs[i].len; + } + + xen_device_copy_grant_refs(xendev, to_domain, segs, count, &local_err); + + if (local_err) { + error_reportf_err(local_err, "failed to copy data: "); + + request->aio_errors++; + return -1; + } + + return 0; +} + +static int xen_block_do_aio(XenBlockRequest *request); +static int xen_block_send_response(XenBlockRequest *request); + +static void xen_block_complete_aio(void *opaque, int ret) +{ + XenBlockRequest *request = opaque; + XenBlockDataPlane *dataplane = request->dataplane; + + aio_context_acquire(dataplane->ctx); + + if (ret != 0) { + error_report("%s I/O error", + request->req.operation == BLKIF_OP_READ ? + "read" : "write"); + request->aio_errors++; + } + + request->aio_inflight--; + if (request->presync) { + request->presync = 0; + xen_block_do_aio(request); + goto done; + } + if (request->aio_inflight > 0) { + goto done; + } + + switch (request->req.operation) { + case BLKIF_OP_READ: + /* in case of failure request->aio_errors is increased */ + if (ret == 0) { + xen_block_copy_request(request); + } + break; + case BLKIF_OP_WRITE: + case BLKIF_OP_FLUSH_DISKCACHE: + if (!request->req.nr_segments) { + break; + } + break; + default: + break; + } + + request->status = request->aio_errors ? BLKIF_RSP_ERROR : BLKIF_RSP_OKAY; + xen_block_finish_request(request); + + switch (request->req.operation) { + case BLKIF_OP_WRITE: + case BLKIF_OP_FLUSH_DISKCACHE: + if (!request->req.nr_segments) { + break; + } + case BLKIF_OP_READ: + if (request->status == BLKIF_RSP_OKAY) { + block_acct_done(blk_get_stats(dataplane->blk), &request->acct); + } else { + block_acct_failed(blk_get_stats(dataplane->blk), &request->acct); + } + break; + case BLKIF_OP_DISCARD: + default: + break; + } + if (xen_block_send_response(request)) { + Error *local_err = NULL; + + xen_device_notify_event_channel(dataplane->xendev, + dataplane->event_channel, + &local_err); + if (local_err) { + error_report_err(local_err); + } + } + xen_block_release_request(request); + + qemu_bh_schedule(dataplane->bh); + +done: + aio_context_release(dataplane->ctx); +} + +static bool xen_block_split_discard(XenBlockRequest *request, + blkif_sector_t sector_number, + uint64_t nr_sectors) +{ + XenBlockDataPlane *dataplane = request->dataplane; + int64_t byte_offset; + int byte_chunk; + uint64_t byte_remaining, limit; + uint64_t sec_start = sector_number; + uint64_t sec_count = nr_sectors; + + /* Wrap around, or overflowing byte limit? */ + if (sec_start + sec_count < sec_count || + sec_start + sec_count > INT64_MAX / dataplane->file_blk) { + return false; + } + + limit = BDRV_REQUEST_MAX_SECTORS * dataplane->file_blk; + byte_offset = sec_start * dataplane->file_blk; + byte_remaining = sec_count * dataplane->file_blk; + + do { + byte_chunk = byte_remaining > limit ? limit : byte_remaining; + request->aio_inflight++; + blk_aio_pdiscard(dataplane->blk, byte_offset, byte_chunk, + xen_block_complete_aio, request); + byte_remaining -= byte_chunk; + byte_offset += byte_chunk; + } while (byte_remaining > 0); + + return true; +} + +static int xen_block_do_aio(XenBlockRequest *request) +{ + XenBlockDataPlane *dataplane = request->dataplane; + + if (request->req.nr_segments && + (request->req.operation == BLKIF_OP_WRITE || + request->req.operation == BLKIF_OP_FLUSH_DISKCACHE) && + xen_block_copy_request(request)) { + goto err; + } + + request->aio_inflight++; + if (request->presync) { + blk_aio_flush(request->dataplane->blk, xen_block_complete_aio, + request); + return 0; + } + + switch (request->req.operation) { + case BLKIF_OP_READ: + qemu_iovec_add(&request->v, request->buf, request->size); + block_acct_start(blk_get_stats(dataplane->blk), &request->acct, + request->v.size, BLOCK_ACCT_READ); + request->aio_inflight++; + blk_aio_preadv(dataplane->blk, request->start, &request->v, 0, + xen_block_complete_aio, request); + break; + case BLKIF_OP_WRITE: + case BLKIF_OP_FLUSH_DISKCACHE: + if (!request->req.nr_segments) { + break; + } + + qemu_iovec_add(&request->v, request->buf, request->size); + block_acct_start(blk_get_stats(dataplane->blk), &request->acct, + request->v.size, + request->req.operation == BLKIF_OP_WRITE ? + BLOCK_ACCT_WRITE : BLOCK_ACCT_FLUSH); + request->aio_inflight++; + blk_aio_pwritev(dataplane->blk, request->start, &request->v, 0, + xen_block_complete_aio, request); + break; + case BLKIF_OP_DISCARD: + { + struct blkif_request_discard *req = (void *)&request->req; + if (!xen_block_split_discard(request, req->sector_number, + req->nr_sectors)) { + goto err; + } + break; + } + default: + /* unknown operation (shouldn't happen -- parse catches this) */ + goto err; + } + + xen_block_complete_aio(request, 0); + + return 0; + +err: + xen_block_finish_request(request); + request->status = BLKIF_RSP_ERROR; + return -1; +} + +static int xen_block_send_response(XenBlockRequest *request) +{ + XenBlockDataPlane *dataplane = request->dataplane; + int send_notify = 0; + int have_requests = 0; + blkif_response_t *resp; + + /* Place on the response ring for the relevant domain. */ + switch (dataplane->protocol) { + case BLKIF_PROTOCOL_NATIVE: + resp = (blkif_response_t *)RING_GET_RESPONSE( + &dataplane->rings.native, + dataplane->rings.native.rsp_prod_pvt); + break; + case BLKIF_PROTOCOL_X86_32: + resp = (blkif_response_t *)RING_GET_RESPONSE( + &dataplane->rings.x86_32_part, + dataplane->rings.x86_32_part.rsp_prod_pvt); + break; + case BLKIF_PROTOCOL_X86_64: + resp = (blkif_response_t *)RING_GET_RESPONSE( + &dataplane->rings.x86_64_part, + dataplane->rings.x86_64_part.rsp_prod_pvt); + break; + default: + return 0; + } + + resp->id = request->req.id; + resp->operation = request->req.operation; + resp->status = request->status; + + dataplane->rings.common.rsp_prod_pvt++; + + RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&dataplane->rings.common, + send_notify); + if (dataplane->rings.common.rsp_prod_pvt == + dataplane->rings.common.req_cons) { + /* + * Tail check for pending requests. Allows frontend to avoid + * notifications if requests are already in flight (lower + * overheads and promotes batching). + */ + RING_FINAL_CHECK_FOR_REQUESTS(&dataplane->rings.common, + have_requests); + } else if (RING_HAS_UNCONSUMED_REQUESTS(&dataplane->rings.common)) { + have_requests = 1; + } + + if (have_requests) { + dataplane->more_work++; + } + return send_notify; +} + +static int xen_block_get_request(XenBlockDataPlane *dataplane, + XenBlockRequest *request, RING_IDX rc) +{ + switch (dataplane->protocol) { + case BLKIF_PROTOCOL_NATIVE: { + blkif_request_t *req = + RING_GET_REQUEST(&dataplane->rings.native, rc); + + memcpy(&request->req, req, sizeof(request->req)); + break; + } + case BLKIF_PROTOCOL_X86_32: { + blkif_x86_32_request_t *req = + RING_GET_REQUEST(&dataplane->rings.x86_32_part, rc); + + blkif_get_x86_32_req(&request->req, req); + break; + } + case BLKIF_PROTOCOL_X86_64: { + blkif_x86_64_request_t *req = + RING_GET_REQUEST(&dataplane->rings.x86_64_part, rc); + + blkif_get_x86_64_req(&request->req, req); + break; + } + } + /* Prevent the compiler from accessing the on-ring fields instead. */ + barrier(); + return 0; +} + +/* + * Threshold of in-flight requests above which we will start using + * blk_io_plug()/blk_io_unplug() to batch requests. + */ +#define IO_PLUG_THRESHOLD 1 + +static void xen_block_handle_requests(XenBlockDataPlane *dataplane) +{ + RING_IDX rc, rp; + XenBlockRequest *request; + int inflight_atstart = dataplane->requests_inflight; + int batched = 0; + + dataplane->more_work = 0; + + rc = dataplane->rings.common.req_cons; + rp = dataplane->rings.common.sring->req_prod; + xen_rmb(); /* Ensure we see queued requests up to 'rp'. */ + + /* + * If there was more than IO_PLUG_THRESHOLD requests in flight + * when we got here, this is an indication that there the bottleneck + * is below us, so it's worth beginning to batch up I/O requests + * rather than submitting them immediately. The maximum number + * of requests we're willing to batch is the number already in + * flight, so it can grow up to max_requests when the bottleneck + * is below us. + */ + if (inflight_atstart > IO_PLUG_THRESHOLD) { + blk_io_plug(dataplane->blk); + } + while (rc != rp) { + /* pull request from ring */ + if (RING_REQUEST_CONS_OVERFLOW(&dataplane->rings.common, rc)) { + break; + } + request = xen_block_start_request(dataplane); + if (request == NULL) { + dataplane->more_work++; + break; + } + xen_block_get_request(dataplane, request, rc); + dataplane->rings.common.req_cons = ++rc; + + /* parse them */ + if (xen_block_parse_request(request) != 0) { + switch (request->req.operation) { + case BLKIF_OP_READ: + block_acct_invalid(blk_get_stats(dataplane->blk), + BLOCK_ACCT_READ); + break; + case BLKIF_OP_WRITE: + block_acct_invalid(blk_get_stats(dataplane->blk), + BLOCK_ACCT_WRITE); + break; + case BLKIF_OP_FLUSH_DISKCACHE: + block_acct_invalid(blk_get_stats(dataplane->blk), + BLOCK_ACCT_FLUSH); + default: + break; + }; + + if (xen_block_send_response(request)) { + Error *local_err = NULL; + + xen_device_notify_event_channel(dataplane->xendev, + dataplane->event_channel, + &local_err); + if (local_err) { + error_report_err(local_err); + } + } + xen_block_release_request(request); + continue; + } + + if (inflight_atstart > IO_PLUG_THRESHOLD && + batched >= inflight_atstart) { + blk_io_unplug(dataplane->blk); + } + xen_block_do_aio(request); + if (inflight_atstart > IO_PLUG_THRESHOLD) { + if (batched >= inflight_atstart) { + blk_io_plug(dataplane->blk); + batched = 0; + } else { + batched++; + } + } + } + if (inflight_atstart > IO_PLUG_THRESHOLD) { + blk_io_unplug(dataplane->blk); + } + + if (dataplane->more_work && + dataplane->requests_inflight < dataplane->max_requests) { + qemu_bh_schedule(dataplane->bh); + } +} + +static void xen_block_dataplane_bh(void *opaque) +{ + XenBlockDataPlane *dataplane = opaque; + + aio_context_acquire(dataplane->ctx); + xen_block_handle_requests(dataplane); + aio_context_release(dataplane->ctx); +} + +static void xen_block_dataplane_event(void *opaque) +{ + XenBlockDataPlane *dataplane = opaque; + + qemu_bh_schedule(dataplane->bh); +} + +XenBlockDataPlane *xen_block_dataplane_create(XenDevice *xendev, + BlockConf *conf, + IOThread *iothread) +{ + XenBlockDataPlane *dataplane = g_new0(XenBlockDataPlane, 1); + + dataplane->xendev = xendev; + dataplane->file_blk = conf->logical_block_size; + dataplane->blk = conf->blk; + dataplane->file_size = blk_getlength(dataplane->blk); + + QLIST_INIT(&dataplane->inflight); + QLIST_INIT(&dataplane->freelist); + + if (iothread) { + dataplane->iothread = iothread; + object_ref(OBJECT(dataplane->iothread)); + dataplane->ctx = iothread_get_aio_context(dataplane->iothread); + } else { + dataplane->ctx = qemu_get_aio_context(); + } + dataplane->bh = aio_bh_new(dataplane->ctx, xen_block_dataplane_bh, + dataplane); + + return dataplane; +} + +void xen_block_dataplane_destroy(XenBlockDataPlane *dataplane) +{ + XenBlockRequest *request; + + if (!dataplane) { + return; + } + + while (!QLIST_EMPTY(&dataplane->freelist)) { + request = QLIST_FIRST(&dataplane->freelist); + QLIST_REMOVE(request, list); + qemu_iovec_destroy(&request->v); + qemu_vfree(request->buf); + g_free(request); + } + + qemu_bh_delete(dataplane->bh); + if (dataplane->iothread) { + object_unref(OBJECT(dataplane->iothread)); + } + + g_free(dataplane); +} + +void xen_block_dataplane_stop(XenBlockDataPlane *dataplane) +{ + XenDevice *xendev; + + if (!dataplane) { + return; + } + + aio_context_acquire(dataplane->ctx); + blk_set_aio_context(dataplane->blk, qemu_get_aio_context()); + aio_context_release(dataplane->ctx); + + xendev = dataplane->xendev; + + if (dataplane->event_channel) { + Error *local_err = NULL; + + xen_device_unbind_event_channel(xendev, dataplane->event_channel, + &local_err); + dataplane->event_channel = NULL; + + if (local_err) { + error_report_err(local_err); + } + } + + if (dataplane->sring) { + Error *local_err = NULL; + + xen_device_unmap_grant_refs(xendev, dataplane->sring, + dataplane->nr_ring_ref, &local_err); + dataplane->sring = NULL; + + if (local_err) { + error_report_err(local_err); + } + } + + g_free(dataplane->ring_ref); + dataplane->ring_ref = NULL; +} + +void xen_block_dataplane_start(XenBlockDataPlane *dataplane, + const unsigned int ring_ref[], + unsigned int nr_ring_ref, + unsigned int event_channel, + unsigned int protocol, + Error **errp) +{ + XenDevice *xendev = dataplane->xendev; + Error *local_err = NULL; + unsigned int ring_size; + unsigned int i; + + dataplane->nr_ring_ref = nr_ring_ref; + dataplane->ring_ref = g_new(unsigned int, nr_ring_ref); + + for (i = 0; i < nr_ring_ref; i++) { + dataplane->ring_ref[i] = ring_ref[i]; + } + + dataplane->protocol = protocol; + + ring_size = XC_PAGE_SIZE * dataplane->nr_ring_ref; + switch (dataplane->protocol) { + case BLKIF_PROTOCOL_NATIVE: + { + dataplane->max_requests = __CONST_RING_SIZE(blkif, ring_size); + break; + } + case BLKIF_PROTOCOL_X86_32: + { + dataplane->max_requests = __CONST_RING_SIZE(blkif_x86_32, ring_size); + break; + } + case BLKIF_PROTOCOL_X86_64: + { + dataplane->max_requests = __CONST_RING_SIZE(blkif_x86_64, ring_size); + break; + } + default: + error_setg(errp, "unknown protocol %u", dataplane->protocol); + return; + } + + xen_device_set_max_grant_refs(xendev, dataplane->nr_ring_ref, + &local_err); + if (local_err) { + error_propagate(errp, local_err); + goto stop; + } + + dataplane->sring = xen_device_map_grant_refs(xendev, + dataplane->ring_ref, + dataplane->nr_ring_ref, + PROT_READ | PROT_WRITE, + &local_err); + if (local_err) { + error_propagate(errp, local_err); + goto stop; + } + + switch (dataplane->protocol) { + case BLKIF_PROTOCOL_NATIVE: + { + blkif_sring_t *sring_native = dataplane->sring; + + BACK_RING_INIT(&dataplane->rings.native, sring_native, ring_size); + break; + } + case BLKIF_PROTOCOL_X86_32: + { + blkif_x86_32_sring_t *sring_x86_32 = dataplane->sring; + + BACK_RING_INIT(&dataplane->rings.x86_32_part, sring_x86_32, + ring_size); + break; + } + case BLKIF_PROTOCOL_X86_64: + { + blkif_x86_64_sring_t *sring_x86_64 = dataplane->sring; + + BACK_RING_INIT(&dataplane->rings.x86_64_part, sring_x86_64, + ring_size); + break; + } + } + + dataplane->event_channel = + xen_device_bind_event_channel(xendev, event_channel, + xen_block_dataplane_event, dataplane, + &local_err); + if (local_err) { + error_propagate(errp, local_err); + goto stop; + } + + aio_context_acquire(dataplane->ctx); + blk_set_aio_context(dataplane->blk, dataplane->ctx); + aio_context_release(dataplane->ctx); + return; + +stop: + xen_block_dataplane_stop(dataplane); +} diff --git a/hw/block/dataplane/xen-block.h b/hw/block/dataplane/xen-block.h new file mode 100644 index 0000000000..d6fa6d26dd --- /dev/null +++ b/hw/block/dataplane/xen-block.h @@ -0,0 +1,29 @@ +/* + * Copyright (c) 2018 Citrix Systems Inc. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#ifndef HW_BLOCK_DATAPLANE_XEN_BLOCK_H +#define HW_BLOCK_DATAPLANE_XEN_BLOCK_H + +#include "hw/block/block.h" +#include "hw/xen/xen-bus.h" +#include "sysemu/iothread.h" + +typedef struct XenBlockDataPlane XenBlockDataPlane; + +XenBlockDataPlane *xen_block_dataplane_create(XenDevice *xendev, + BlockConf *conf, + IOThread *iothread); +void xen_block_dataplane_destroy(XenBlockDataPlane *dataplane); +void xen_block_dataplane_start(XenBlockDataPlane *dataplane, + const unsigned int ring_ref[], + unsigned int nr_ring_ref, + unsigned int event_channel, + unsigned int protocol, + Error **errp); +void xen_block_dataplane_stop(XenBlockDataPlane *dataplane); + +#endif /* HW_BLOCK_DATAPLANE_XEN_BLOCK_H */ diff --git a/hw/block/trace-events b/hw/block/trace-events index 335c092450..55e5a5500c 100644 --- a/hw/block/trace-events +++ b/hw/block/trace-events @@ -127,3 +127,17 @@ xen_disk_init(char *name) "%s" xen_disk_connect(char *name) "%s" xen_disk_disconnect(char *name) "%s" xen_disk_free(char *name) "%s" + +# hw/block/xen-block.c +xen_block_realize(const char *type, uint32_t disk, uint32_t partition) "%s d%up%u" +xen_block_connect(const char *type, uint32_t disk, uint32_t partition) "%s d%up%u" +xen_block_disconnect(const char *type, uint32_t disk, uint32_t partition) "%s d%up%u" +xen_block_unrealize(const char *type, uint32_t disk, uint32_t partition) "%s d%up%u" +xen_disk_realize(void) "" +xen_disk_unrealize(void) "" +xen_cdrom_realize(void) "" +xen_cdrom_unrealize(void) "" +xen_block_blockdev_add(char *str) "%s" +xen_block_blockdev_del(const char *node_name) "%s" +xen_block_device_create(unsigned int number) "%u" +xen_block_device_destroy(unsigned int number) "%u" diff --git a/hw/block/xen-block.c b/hw/block/xen-block.c new file mode 100644 index 0000000000..be28b63442 --- /dev/null +++ b/hw/block/xen-block.c @@ -0,0 +1,963 @@ +/* + * Copyright (c) 2018 Citrix Systems Inc. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#include "qemu/osdep.h" +#include "qemu/cutils.h" +#include "qemu/option.h" +#include "qapi/error.h" +#include "qapi/qapi-commands-block-core.h" +#include "qapi/qapi-commands-misc.h" +#include "qapi/qapi-visit-block-core.h" +#include "qapi/qobject-input-visitor.h" +#include "qapi/visitor.h" +#include "qapi/qmp/qdict.h" +#include "qapi/qmp/qstring.h" +#include "hw/hw.h" +#include "hw/xen/xen_common.h" +#include "hw/block/xen_blkif.h" +#include "hw/xen/xen-block.h" +#include "hw/xen/xen-backend.h" +#include "sysemu/blockdev.h" +#include "sysemu/block-backend.h" +#include "sysemu/iothread.h" +#include "dataplane/xen-block.h" +#include "trace.h" + +static char *xen_block_get_name(XenDevice *xendev, Error **errp) +{ + XenBlockDevice *blockdev = XEN_BLOCK_DEVICE(xendev); + XenBlockVdev *vdev = &blockdev->props.vdev; + + return g_strdup_printf("%lu", vdev->number); +} + +static void xen_block_disconnect(XenDevice *xendev, Error **errp) +{ + XenBlockDevice *blockdev = XEN_BLOCK_DEVICE(xendev); + const char *type = object_get_typename(OBJECT(blockdev)); + XenBlockVdev *vdev = &blockdev->props.vdev; + + trace_xen_block_disconnect(type, vdev->disk, vdev->partition); + + xen_block_dataplane_stop(blockdev->dataplane); +} + +static void xen_block_connect(XenDevice *xendev, Error **errp) +{ + XenBlockDevice *blockdev = XEN_BLOCK_DEVICE(xendev); + const char *type = object_get_typename(OBJECT(blockdev)); + XenBlockVdev *vdev = &blockdev->props.vdev; + unsigned int order, nr_ring_ref, *ring_ref, event_channel, protocol; + char *str; + + trace_xen_block_connect(type, vdev->disk, vdev->partition); + + if (xen_device_frontend_scanf(xendev, "ring-page-order", "%u", + &order) != 1) { + nr_ring_ref = 1; + ring_ref = g_new(unsigned int, nr_ring_ref); + + if (xen_device_frontend_scanf(xendev, "ring-ref", "%u", + &ring_ref[0]) != 1) { + error_setg(errp, "failed to read ring-ref"); + g_free(ring_ref); + return; + } + } else if (order <= blockdev->props.max_ring_page_order) { + unsigned int i; + + nr_ring_ref = 1 << order; + ring_ref = g_new(unsigned int, nr_ring_ref); + + for (i = 0; i < nr_ring_ref; i++) { + const char *key = g_strdup_printf("ring-ref%u", i); + + if (xen_device_frontend_scanf(xendev, key, "%u", + &ring_ref[i]) != 1) { + error_setg(errp, "failed to read %s", key); + g_free((gpointer)key); + g_free(ring_ref); + return; + } + + g_free((gpointer)key); + } + } else { + error_setg(errp, "invalid ring-page-order (%d)", order); + return; + } + + if (xen_device_frontend_scanf(xendev, "event-channel", "%u", + &event_channel) != 1) { + error_setg(errp, "failed to read event-channel"); + g_free(ring_ref); + return; + } + + if (xen_device_frontend_scanf(xendev, "protocol", "%ms", + &str) != 1) { + protocol = BLKIF_PROTOCOL_NATIVE; + } else { + if (strcmp(str, XEN_IO_PROTO_ABI_X86_32) == 0) { + protocol = BLKIF_PROTOCOL_X86_32; + } else if (strcmp(str, XEN_IO_PROTO_ABI_X86_64) == 0) { + protocol = BLKIF_PROTOCOL_X86_64; + } else { + protocol = BLKIF_PROTOCOL_NATIVE; + } + + free(str); + } + + xen_block_dataplane_start(blockdev->dataplane, ring_ref, nr_ring_ref, + event_channel, protocol, errp); + + g_free(ring_ref); +} + +static void xen_block_unrealize(XenDevice *xendev, Error **errp) +{ + XenBlockDevice *blockdev = XEN_BLOCK_DEVICE(xendev); + XenBlockDeviceClass *blockdev_class = + XEN_BLOCK_DEVICE_GET_CLASS(xendev); + const char *type = object_get_typename(OBJECT(blockdev)); + XenBlockVdev *vdev = &blockdev->props.vdev; + + if (vdev->type == XEN_BLOCK_VDEV_TYPE_INVALID) { + return; + } + + trace_xen_block_unrealize(type, vdev->disk, vdev->partition); + + /* Disconnect from the frontend in case this has not already happened */ + xen_block_disconnect(xendev, NULL); + + xen_block_dataplane_destroy(blockdev->dataplane); + blockdev->dataplane = NULL; + + if (blockdev_class->unrealize) { + blockdev_class->unrealize(blockdev, errp); + } +} + +static void xen_block_realize(XenDevice *xendev, Error **errp) +{ + XenBlockDevice *blockdev = XEN_BLOCK_DEVICE(xendev); + XenBlockDeviceClass *blockdev_class = + XEN_BLOCK_DEVICE_GET_CLASS(xendev); + const char *type = object_get_typename(OBJECT(blockdev)); + XenBlockVdev *vdev = &blockdev->props.vdev; + BlockConf *conf = &blockdev->props.conf; + Error *local_err = NULL; + + if (vdev->type == XEN_BLOCK_VDEV_TYPE_INVALID) { + error_setg(errp, "vdev property not set"); + return; + } + + trace_xen_block_realize(type, vdev->disk, vdev->partition); + + if (blockdev_class->realize) { + blockdev_class->realize(blockdev, &local_err); + if (local_err) { + error_propagate(errp, local_err); + return; + } + } + + /* + * The blkif protocol does not deal with removable media, so it must + * always be present, even for CDRom devices. + */ + assert(conf->blk); + if (!blk_is_inserted(conf->blk)) { + error_setg(errp, "device needs media, but drive is empty"); + return; + } + + if (!blkconf_apply_backend_options(conf, blockdev->info & VDISK_READONLY, + false, errp)) { + return; + } + + if (!(blockdev->info & VDISK_CDROM) && + !blkconf_geometry(conf, NULL, 65535, 255, 255, errp)) { + return; + } + + blkconf_blocksizes(conf); + + if (conf->logical_block_size > conf->physical_block_size) { + error_setg( + errp, "logical_block_size > physical_block_size not supported"); + return; + } + + blk_set_guest_block_size(conf->blk, conf->logical_block_size); + + if (conf->discard_granularity > 0) { + xen_device_backend_printf(xendev, "feature-discard", "%u", 1); + } + + xen_device_backend_printf(xendev, "feature-flush-cache", "%u", 1); + xen_device_backend_printf(xendev, "max-ring-page-order", "%u", + blockdev->props.max_ring_page_order); + xen_device_backend_printf(xendev, "info", "%u", blockdev->info); + + xen_device_frontend_printf(xendev, "virtual-device", "%lu", + vdev->number); + xen_device_frontend_printf(xendev, "device-type", "%s", + blockdev->device_type); + + xen_device_backend_printf(xendev, "sector-size", "%u", + conf->logical_block_size); + xen_device_backend_printf(xendev, "sectors", "%lu", + blk_getlength(conf->blk) / + conf->logical_block_size); + + blockdev->dataplane = + xen_block_dataplane_create(xendev, conf, blockdev->props.iothread); +} + +static void xen_block_frontend_changed(XenDevice *xendev, + enum xenbus_state frontend_state, + Error **errp) +{ + enum xenbus_state backend_state = xen_device_backend_get_state(xendev); + Error *local_err = NULL; + + switch (frontend_state) { + case XenbusStateInitialised: + case XenbusStateConnected: + if (backend_state == XenbusStateConnected) { + break; + } + + xen_block_disconnect(xendev, &local_err); + if (local_err) { + error_propagate(errp, local_err); + break; + } + + xen_block_connect(xendev, &local_err); + if (local_err) { + error_propagate(errp, local_err); + break; + } + + xen_device_backend_set_state(xendev, XenbusStateConnected); + break; + + case XenbusStateClosing: + xen_device_backend_set_state(xendev, XenbusStateClosing); + break; + + case XenbusStateClosed: + xen_block_disconnect(xendev, &local_err); + if (local_err) { + error_propagate(errp, local_err); + break; + } + + xen_device_backend_set_state(xendev, XenbusStateClosed); + break; + + default: + break; + } +} + +static char *disk_to_vbd_name(unsigned int disk) +{ + char *name, *prefix = (disk >= 26) ? + disk_to_vbd_name((disk / 26) - 1) : g_strdup(""); + + name = g_strdup_printf("%s%c", prefix, 'a' + disk % 26); + g_free(prefix); + + return name; +} + +static void xen_block_get_vdev(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + DeviceState *dev = DEVICE(obj); + Property *prop = opaque; + XenBlockVdev *vdev = qdev_get_prop_ptr(dev, prop); + char *str; + + switch (vdev->type) { + case XEN_BLOCK_VDEV_TYPE_DP: + str = g_strdup_printf("d%lup%lu", vdev->disk, vdev->partition); + break; + + case XEN_BLOCK_VDEV_TYPE_XVD: + case XEN_BLOCK_VDEV_TYPE_HD: + case XEN_BLOCK_VDEV_TYPE_SD: { + char *name = disk_to_vbd_name(vdev->disk); + + str = g_strdup_printf("%s%s%lu", + (vdev->type == XEN_BLOCK_VDEV_TYPE_XVD) ? + "xvd" : + (vdev->type == XEN_BLOCK_VDEV_TYPE_HD) ? + "hd" : + "sd", + name, vdev->partition); + g_free(name); + break; + } + default: + error_setg(errp, "invalid vdev type"); + return; + } + + visit_type_str(v, name, &str, errp); + g_free(str); +} + +static unsigned int vbd_name_to_disk(const char *name, const char **endp) +{ + unsigned int disk = 0; + + while (*name != '\0') { + if (!g_ascii_isalpha(*name) || !g_ascii_islower(*name)) { + break; + } + + disk *= 26; + disk += *name++ - 'a' + 1; + } + *endp = name; + + return disk - 1; +} + +static void xen_block_set_vdev(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + DeviceState *dev = DEVICE(obj); + Property *prop = opaque; + XenBlockVdev *vdev = qdev_get_prop_ptr(dev, prop); + Error *local_err = NULL; + char *str, *p; + const char *end; + + if (dev->realized) { + qdev_prop_set_after_realize(dev, name, errp); + return; + } + + visit_type_str(v, name, &str, &local_err); + if (local_err) { + error_propagate(errp, local_err); + return; + } + + p = strchr(str, 'd'); + if (!p) { + goto invalid; + } + + *p++ = '\0'; + if (*str == '\0') { + vdev->type = XEN_BLOCK_VDEV_TYPE_DP; + } else if (strcmp(str, "xv") == 0) { + vdev->type = XEN_BLOCK_VDEV_TYPE_XVD; + } else if (strcmp(str, "h") == 0) { + vdev->type = XEN_BLOCK_VDEV_TYPE_HD; + } else if (strcmp(str, "s") == 0) { + vdev->type = XEN_BLOCK_VDEV_TYPE_SD; + } else { + goto invalid; + } + + if (vdev->type == XEN_BLOCK_VDEV_TYPE_DP) { + if (qemu_strtoul(p, &end, 10, &vdev->disk)) { + goto invalid; + } + + if (*end == 'p') { + p = (char *) ++end; + if (*end == '\0') { + goto invalid; + } + } + } else { + vdev->disk = vbd_name_to_disk(p, &end); + } + + if (*end != '\0') { + p = (char *)end; + + if (qemu_strtoul(p, &end, 10, &vdev->partition)) { + goto invalid; + } + + if (*end != '\0') { + goto invalid; + } + } else { + vdev->partition = 0; + } + + switch (vdev->type) { + case XEN_BLOCK_VDEV_TYPE_DP: + case XEN_BLOCK_VDEV_TYPE_XVD: + if (vdev->disk < (1 << 4) && vdev->partition < (1 << 4)) { + vdev->number = (202 << 8) | (vdev->disk << 4) | + vdev->partition; + } else if (vdev->disk < (1 << 20) && vdev->partition < (1 << 8)) { + vdev->number = (1 << 28) | (vdev->disk << 8) | + vdev->partition; + } else { + goto invalid; + } + break; + + case XEN_BLOCK_VDEV_TYPE_HD: + if ((vdev->disk == 0 || vdev->disk == 1) && + vdev->partition < (1 << 6)) { + vdev->number = (3 << 8) | (vdev->disk << 6) | vdev->partition; + } else if ((vdev->disk == 2 || vdev->disk == 3) && + vdev->partition < (1 << 6)) { + vdev->number = (22 << 8) | ((vdev->disk - 2) << 6) | + vdev->partition; + } else { + goto invalid; + } + break; + + case XEN_BLOCK_VDEV_TYPE_SD: + if (vdev->disk < (1 << 4) && vdev->partition < (1 << 4)) { + vdev->number = (8 << 8) | (vdev->disk << 4) | vdev->partition; + } else { + goto invalid; + } + break; + + default: + goto invalid; + } + + g_free(str); + return; + +invalid: + error_setg(errp, "invalid virtual disk specifier"); + + vdev->type = XEN_BLOCK_VDEV_TYPE_INVALID; + g_free(str); +} + +/* + * This property deals with 'vdev' names adhering to the Xen VBD naming + * scheme described in: + * + * https://xenbits.xen.org/docs/unstable/man/xen-vbd-interface.7.html + */ +const PropertyInfo xen_block_prop_vdev = { + .name = "str", + .description = "Virtual Disk specifier: d*p*/xvd*/hd*/sd*", + .get = xen_block_get_vdev, + .set = xen_block_set_vdev, +}; + +static Property xen_block_props[] = { + DEFINE_PROP("vdev", XenBlockDevice, props.vdev, + xen_block_prop_vdev, XenBlockVdev), + DEFINE_BLOCK_PROPERTIES(XenBlockDevice, props.conf), + DEFINE_PROP_UINT32("max-ring-page-order", XenBlockDevice, + props.max_ring_page_order, 4), + DEFINE_PROP_LINK("iothread", XenBlockDevice, props.iothread, + TYPE_IOTHREAD, IOThread *), + DEFINE_PROP_END_OF_LIST() +}; + +static void xen_block_class_init(ObjectClass *class, void *data) +{ + DeviceClass *dev_class = DEVICE_CLASS(class); + XenDeviceClass *xendev_class = XEN_DEVICE_CLASS(class); + + xendev_class->backend = "qdisk"; + xendev_class->device = "vbd"; + xendev_class->get_name = xen_block_get_name; + xendev_class->realize = xen_block_realize; + xendev_class->frontend_changed = xen_block_frontend_changed; + xendev_class->unrealize = xen_block_unrealize; + + dev_class->props = xen_block_props; +} + +static const TypeInfo xen_block_type_info = { + .name = TYPE_XEN_BLOCK_DEVICE, + .parent = TYPE_XEN_DEVICE, + .instance_size = sizeof(XenBlockDevice), + .abstract = true, + .class_size = sizeof(XenBlockDeviceClass), + .class_init = xen_block_class_init, +}; + +static void xen_disk_unrealize(XenBlockDevice *blockdev, Error **errp) +{ + trace_xen_disk_unrealize(); +} + +static void xen_disk_realize(XenBlockDevice *blockdev, Error **errp) +{ + BlockConf *conf = &blockdev->props.conf; + + trace_xen_disk_realize(); + + blockdev->device_type = "disk"; + + if (!conf->blk) { + error_setg(errp, "drive property not set"); + return; + } + + blockdev->info = blk_is_read_only(conf->blk) ? VDISK_READONLY : 0; +} + +static void xen_disk_class_init(ObjectClass *class, void *data) +{ + DeviceClass *dev_class = DEVICE_CLASS(class); + XenBlockDeviceClass *blockdev_class = XEN_BLOCK_DEVICE_CLASS(class); + + blockdev_class->realize = xen_disk_realize; + blockdev_class->unrealize = xen_disk_unrealize; + + dev_class->desc = "Xen Disk Device"; +} + +static const TypeInfo xen_disk_type_info = { + .name = TYPE_XEN_DISK_DEVICE, + .parent = TYPE_XEN_BLOCK_DEVICE, + .instance_size = sizeof(XenDiskDevice), + .class_init = xen_disk_class_init, +}; + +static void xen_cdrom_unrealize(XenBlockDevice *blockdev, Error **errp) +{ + trace_xen_cdrom_unrealize(); +} + +static void xen_cdrom_realize(XenBlockDevice *blockdev, Error **errp) +{ + BlockConf *conf = &blockdev->props.conf; + + trace_xen_cdrom_realize(); + + blockdev->device_type = "cdrom"; + + if (!conf->blk) { + int rc; + + /* Set up an empty drive */ + conf->blk = blk_new(0, BLK_PERM_ALL); + + rc = blk_attach_dev(conf->blk, DEVICE(blockdev)); + if (!rc) { + error_setg_errno(errp, -rc, "failed to create drive"); + return; + } + } + + blockdev->info = VDISK_READONLY | VDISK_CDROM; +} + +static void xen_cdrom_class_init(ObjectClass *class, void *data) +{ + DeviceClass *dev_class = DEVICE_CLASS(class); + XenBlockDeviceClass *blockdev_class = XEN_BLOCK_DEVICE_CLASS(class); + + blockdev_class->realize = xen_cdrom_realize; + blockdev_class->unrealize = xen_cdrom_unrealize; + + dev_class->desc = "Xen CD-ROM Device"; +} + +static const TypeInfo xen_cdrom_type_info = { + .name = TYPE_XEN_CDROM_DEVICE, + .parent = TYPE_XEN_BLOCK_DEVICE, + .instance_size = sizeof(XenCDRomDevice), + .class_init = xen_cdrom_class_init, +}; + +static void xen_block_register_types(void) +{ + type_register_static(&xen_block_type_info); + type_register_static(&xen_disk_type_info); + type_register_static(&xen_cdrom_type_info); +} + +type_init(xen_block_register_types) + +static void xen_block_blockdev_del(const char *node_name, Error **errp) +{ + trace_xen_block_blockdev_del(node_name); + + qmp_blockdev_del(node_name, errp); +} + +static char *xen_block_blockdev_add(const char *id, QDict *qdict, + Error **errp) +{ + const char *driver = qdict_get_try_str(qdict, "driver"); + BlockdevOptions *options = NULL; + Error *local_err = NULL; + char *node_name; + Visitor *v; + + if (!driver) { + error_setg(errp, "no 'driver' parameter"); + return NULL; + } + + node_name = g_strdup_printf("%s-%s", id, driver); + qdict_put_str(qdict, "node-name", node_name); + + trace_xen_block_blockdev_add(node_name); + + v = qobject_input_visitor_new(QOBJECT(qdict)); + visit_type_BlockdevOptions(v, NULL, &options, &local_err); + visit_free(v); + + if (local_err) { + error_propagate(errp, local_err); + goto fail; + } + + qmp_blockdev_add(options, &local_err); + + if (local_err) { + error_propagate(errp, local_err); + goto fail; + } + + qapi_free_BlockdevOptions(options); + + return node_name; + +fail: + if (options) { + qapi_free_BlockdevOptions(options); + } + g_free(node_name); + + return NULL; +} + +static void xen_block_drive_destroy(XenBlockDrive *drive, Error **errp) +{ + char *node_name = drive->node_name; + + if (node_name) { + Error *local_err = NULL; + + xen_block_blockdev_del(node_name, &local_err); + if (local_err) { + error_propagate(errp, local_err); + return; + } + g_free(node_name); + drive->node_name = NULL; + } + g_free(drive->id); + g_free(drive); +} + +static XenBlockDrive *xen_block_drive_create(const char *id, + const char *device_type, + QDict *opts, Error **errp) +{ + const char *params = qdict_get_try_str(opts, "params"); + const char *mode = qdict_get_try_str(opts, "mode"); + const char *direct_io_safe = qdict_get_try_str(opts, "direct-io-safe"); + const char *discard_enable = qdict_get_try_str(opts, "discard-enable"); + char *driver = NULL; + char *filename = NULL; + XenBlockDrive *drive = NULL; + Error *local_err = NULL; + QDict *file_layer; + QDict *driver_layer; + + if (params) { + char **v = g_strsplit(params, ":", 2); + + if (v[1] == NULL) { + filename = g_strdup(v[0]); + driver = g_strdup("raw"); + } else { + if (strcmp(v[0], "aio") == 0) { + driver = g_strdup("raw"); + } else if (strcmp(v[0], "vhd") == 0) { + driver = g_strdup("vpc"); + } else { + driver = g_strdup(v[0]); + } + filename = g_strdup(v[1]); + } + + g_strfreev(v); + } + + if (!filename) { + error_setg(errp, "no filename"); + goto done; + } + assert(driver); + + drive = g_new0(XenBlockDrive, 1); + drive->id = g_strdup(id); + + file_layer = qdict_new(); + + qdict_put_str(file_layer, "driver", "file"); + qdict_put_str(file_layer, "filename", filename); + + if (mode && *mode != 'w') { + qdict_put_bool(file_layer, "read-only", true); + } + + if (direct_io_safe) { + unsigned long value; + + if (!qemu_strtoul(direct_io_safe, NULL, 2, &value) && !!value) { + QDict *cache_qdict = qdict_new(); + + qdict_put_bool(cache_qdict, "direct", true); + qdict_put_obj(file_layer, "cache", QOBJECT(cache_qdict)); + + qdict_put_str(file_layer, "aio", "native"); + } + } + + if (discard_enable) { + unsigned long value; + + if (!qemu_strtoul(discard_enable, NULL, 2, &value) && !!value) { + qdict_put_str(file_layer, "discard", "unmap"); + } + } + + /* + * It is necessary to turn file locking off as an emulated device + * may have already opened the same image file. + */ + qdict_put_str(file_layer, "locking", "off"); + + driver_layer = qdict_new(); + + qdict_put_str(driver_layer, "driver", driver); + qdict_put_obj(driver_layer, "file", QOBJECT(file_layer)); + + g_assert(!drive->node_name); + drive->node_name = xen_block_blockdev_add(drive->id, driver_layer, + &local_err); + +done: + g_free(driver); + g_free(filename); + + if (local_err) { + error_propagate(errp, local_err); + xen_block_drive_destroy(drive, NULL); + return NULL; + } + + return drive; +} + +static const char *xen_block_drive_get_node_name(XenBlockDrive *drive) +{ + return drive->node_name ? drive->node_name : ""; +} + +static void xen_block_iothread_destroy(XenBlockIOThread *iothread, + Error **errp) +{ + qmp_object_del(iothread->id, errp); + + g_free(iothread->id); + g_free(iothread); +} + +static XenBlockIOThread *xen_block_iothread_create(const char *id, + Error **errp) +{ + XenBlockIOThread *iothread = g_new(XenBlockIOThread, 1); + Error *local_err = NULL; + + iothread->id = g_strdup(id); + + qmp_object_add(TYPE_IOTHREAD, id, false, NULL, &local_err); + if (local_err) { + error_propagate(errp, local_err); + + g_free(iothread->id); + g_free(iothread); + return NULL; + } + + return iothread; +} + +static void xen_block_device_create(XenBackendInstance *backend, + QDict *opts, Error **errp) +{ + XenBus *xenbus = xen_backend_get_bus(backend); + const char *name = xen_backend_get_name(backend); + unsigned long number; + const char *vdev, *device_type; + XenBlockDrive *drive = NULL; + XenBlockIOThread *iothread = NULL; + XenDevice *xendev = NULL; + Error *local_err = NULL; + const char *type; + XenBlockDevice *blockdev; + + if (qemu_strtoul(name, NULL, 10, &number)) { + error_setg(errp, "failed to parse name '%s'", name); + goto fail; + } + + trace_xen_block_device_create(number); + + vdev = qdict_get_try_str(opts, "dev"); + if (!vdev) { + error_setg(errp, "no dev parameter"); + goto fail; + } + + device_type = qdict_get_try_str(opts, "device-type"); + if (!device_type) { + error_setg(errp, "no device-type parameter"); + goto fail; + } + + if (!strcmp(device_type, "disk")) { + type = TYPE_XEN_DISK_DEVICE; + } else if (!strcmp(device_type, "cdrom")) { + type = TYPE_XEN_CDROM_DEVICE; + } else { + error_setg(errp, "invalid device-type parameter '%s'", device_type); + goto fail; + } + + drive = xen_block_drive_create(vdev, device_type, opts, &local_err); + if (!drive) { + error_propagate_prepend(errp, local_err, "failed to create drive: "); + goto fail; + } + + iothread = xen_block_iothread_create(vdev, &local_err); + if (local_err) { + error_propagate_prepend(errp, local_err, + "failed to create iothread: "); + goto fail; + } + + xendev = XEN_DEVICE(qdev_create(BUS(xenbus), type)); + blockdev = XEN_BLOCK_DEVICE(xendev); + + object_property_set_str(OBJECT(xendev), vdev, "vdev", &local_err); + if (local_err) { + error_propagate_prepend(errp, local_err, "failed to set 'vdev': "); + goto fail; + } + + object_property_set_str(OBJECT(xendev), + xen_block_drive_get_node_name(drive), "drive", + &local_err); + if (local_err) { + error_propagate_prepend(errp, local_err, "failed to set 'drive': "); + goto fail; + } + + object_property_set_str(OBJECT(xendev), iothread->id, "iothread", + &local_err); + if (local_err) { + error_propagate_prepend(errp, local_err, + "failed to set 'iothread': "); + goto fail; + } + + blockdev->iothread = iothread; + blockdev->drive = drive; + + object_property_set_bool(OBJECT(xendev), true, "realized", &local_err); + if (local_err) { + error_propagate_prepend(errp, local_err, + "realization of device %s failed: ", + type); + goto fail; + } + + xen_backend_set_device(backend, xendev); + return; + +fail: + if (xendev) { + object_unparent(OBJECT(xendev)); + } + + if (iothread) { + xen_block_iothread_destroy(iothread, NULL); + } + + if (drive) { + xen_block_drive_destroy(drive, NULL); + } +} + +static void xen_block_device_destroy(XenBackendInstance *backend, + Error **errp) +{ + XenDevice *xendev = xen_backend_get_device(backend); + XenBlockDevice *blockdev = XEN_BLOCK_DEVICE(xendev); + XenBlockVdev *vdev = &blockdev->props.vdev; + XenBlockDrive *drive = blockdev->drive; + XenBlockIOThread *iothread = blockdev->iothread; + + trace_xen_block_device_destroy(vdev->number); + + object_unparent(OBJECT(xendev)); + + if (iothread) { + Error *local_err = NULL; + + xen_block_iothread_destroy(iothread, &local_err); + if (local_err) { + error_propagate_prepend(errp, local_err, + "failed to destroy iothread: "); + return; + } + } + + if (drive) { + Error *local_err = NULL; + + xen_block_drive_destroy(drive, &local_err); + if (local_err) { + error_propagate_prepend(errp, local_err, + "failed to destroy drive: "); + } + } +} + +static const XenBackendInfo xen_block_backend_info = { + .type = "qdisk", + .create = xen_block_device_create, + .destroy = xen_block_device_destroy, +}; + +static void xen_block_register_backend(void) +{ + xen_backend_register(&xen_block_backend_info); +} + +xen_backend_init(xen_block_register_backend); diff --git a/hw/block/xen_disk.c b/hw/block/xen_disk.c deleted file mode 100644 index 2a254b99d0..0000000000 --- a/hw/block/xen_disk.c +++ /dev/null @@ -1,1011 +0,0 @@ -/* - * xen paravirt block device backend - * - * (c) Gerd Hoffmann <kraxel@redhat.com> - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; under version 2 of the License. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, see <http://www.gnu.org/licenses/>. - * - * Contributions after 2012-01-13 are licensed under the terms of the - * GNU GPL, version 2 or (at your option) any later version. - */ - -#include "qemu/osdep.h" -#include "qemu/units.h" -#include <sys/ioctl.h> -#include <sys/uio.h> - -#include "hw/hw.h" -#include "hw/xen/xen_backend.h" -#include "xen_blkif.h" -#include "sysemu/blockdev.h" -#include "sysemu/iothread.h" -#include "sysemu/block-backend.h" -#include "qapi/error.h" -#include "qapi/qmp/qdict.h" -#include "qapi/qmp/qstring.h" -#include "trace.h" - -/* ------------------------------------------------------------- */ - -#define BLOCK_SIZE 512 -#define IOCB_COUNT (BLKIF_MAX_SEGMENTS_PER_REQUEST + 2) - -struct ioreq { - blkif_request_t req; - int16_t status; - - /* parsed request */ - off_t start; - QEMUIOVector v; - void *buf; - size_t size; - int presync; - - /* aio status */ - int aio_inflight; - int aio_errors; - - struct XenBlkDev *blkdev; - QLIST_ENTRY(ioreq) list; - BlockAcctCookie acct; -}; - -#define MAX_RING_PAGE_ORDER 4 - -struct XenBlkDev { - struct XenDevice xendev; /* must be first */ - char *params; - char *mode; - char *type; - char *dev; - char *devtype; - bool directiosafe; - const char *fileproto; - const char *filename; - unsigned int ring_ref[1 << MAX_RING_PAGE_ORDER]; - unsigned int nr_ring_ref; - void *sring; - int64_t file_blk; - int64_t file_size; - int protocol; - blkif_back_rings_t rings; - int more_work; - - /* request lists */ - QLIST_HEAD(, ioreq) inflight; - QLIST_HEAD(, ioreq) finished; - QLIST_HEAD(, ioreq) freelist; - int requests_total; - int requests_inflight; - int requests_finished; - unsigned int max_requests; - - gboolean feature_discard; - - /* qemu block driver */ - DriveInfo *dinfo; - BlockBackend *blk; - QEMUBH *bh; - - IOThread *iothread; - AioContext *ctx; -}; - -/* ------------------------------------------------------------- */ - -static void ioreq_reset(struct ioreq *ioreq) -{ - memset(&ioreq->req, 0, sizeof(ioreq->req)); - ioreq->status = 0; - ioreq->start = 0; - ioreq->buf = NULL; - ioreq->size = 0; - ioreq->presync = 0; - - ioreq->aio_inflight = 0; - ioreq->aio_errors = 0; - - ioreq->blkdev = NULL; - memset(&ioreq->list, 0, sizeof(ioreq->list)); - memset(&ioreq->acct, 0, sizeof(ioreq->acct)); - - qemu_iovec_reset(&ioreq->v); -} - -static struct ioreq *ioreq_start(struct XenBlkDev *blkdev) -{ - struct ioreq *ioreq = NULL; - - if (QLIST_EMPTY(&blkdev->freelist)) { - if (blkdev->requests_total >= blkdev->max_requests) { - goto out; - } - /* allocate new struct */ - ioreq = g_malloc0(sizeof(*ioreq)); - ioreq->blkdev = blkdev; - blkdev->requests_total++; - qemu_iovec_init(&ioreq->v, 1); - } else { - /* get one from freelist */ - ioreq = QLIST_FIRST(&blkdev->freelist); - QLIST_REMOVE(ioreq, list); - } - QLIST_INSERT_HEAD(&blkdev->inflight, ioreq, list); - blkdev->requests_inflight++; - -out: - return ioreq; -} - -static void ioreq_finish(struct ioreq *ioreq) -{ - struct XenBlkDev *blkdev = ioreq->blkdev; - - QLIST_REMOVE(ioreq, list); - QLIST_INSERT_HEAD(&blkdev->finished, ioreq, list); - blkdev->requests_inflight--; - blkdev->requests_finished++; -} - -static void ioreq_release(struct ioreq *ioreq, bool finish) -{ - struct XenBlkDev *blkdev = ioreq->blkdev; - - QLIST_REMOVE(ioreq, list); - ioreq_reset(ioreq); - ioreq->blkdev = blkdev; - QLIST_INSERT_HEAD(&blkdev->freelist, ioreq, list); - if (finish) { - blkdev->requests_finished--; - } else { - blkdev->requests_inflight--; - } -} - -/* - * translate request into iovec + start offset - * do sanity checks along the way - */ -static int ioreq_parse(struct ioreq *ioreq) -{ - struct XenBlkDev *blkdev = ioreq->blkdev; - struct XenDevice *xendev = &blkdev->xendev; - size_t len; - int i; - - xen_pv_printf(xendev, 3, - "op %d, nr %d, handle %d, id %" PRId64 ", sector %" PRId64 "\n", - ioreq->req.operation, ioreq->req.nr_segments, - ioreq->req.handle, ioreq->req.id, ioreq->req.sector_number); - switch (ioreq->req.operation) { - case BLKIF_OP_READ: - break; - case BLKIF_OP_FLUSH_DISKCACHE: - ioreq->presync = 1; - if (!ioreq->req.nr_segments) { - return 0; - } - /* fall through */ - case BLKIF_OP_WRITE: - break; - case BLKIF_OP_DISCARD: - return 0; - default: - xen_pv_printf(xendev, 0, "error: unknown operation (%d)\n", - ioreq->req.operation); - goto err; - }; - - if (ioreq->req.operation != BLKIF_OP_READ && blkdev->mode[0] != 'w') { - xen_pv_printf(xendev, 0, "error: write req for ro device\n"); - goto err; - } - - ioreq->start = ioreq->req.sector_number * blkdev->file_blk; - for (i = 0; i < ioreq->req.nr_segments; i++) { - if (i == BLKIF_MAX_SEGMENTS_PER_REQUEST) { - xen_pv_printf(xendev, 0, "error: nr_segments too big\n"); - goto err; - } - if (ioreq->req.seg[i].first_sect > ioreq->req.seg[i].last_sect) { - xen_pv_printf(xendev, 0, "error: first > last sector\n"); - goto err; - } - if (ioreq->req.seg[i].last_sect * BLOCK_SIZE >= XC_PAGE_SIZE) { - xen_pv_printf(xendev, 0, "error: page crossing\n"); - goto err; - } - - len = (ioreq->req.seg[i].last_sect - ioreq->req.seg[i].first_sect + 1) * blkdev->file_blk; - ioreq->size += len; - } - if (ioreq->start + ioreq->size > blkdev->file_size) { - xen_pv_printf(xendev, 0, "error: access beyond end of file\n"); - goto err; - } - return 0; - -err: - ioreq->status = BLKIF_RSP_ERROR; - return -1; -} - -static int ioreq_grant_copy(struct ioreq *ioreq) -{ - struct XenBlkDev *blkdev = ioreq->blkdev; - struct XenDevice *xendev = &blkdev->xendev; - XenGrantCopySegment segs[BLKIF_MAX_SEGMENTS_PER_REQUEST]; - int i, count, rc; - int64_t file_blk = blkdev->file_blk; - bool to_domain = (ioreq->req.operation == BLKIF_OP_READ); - void *virt = ioreq->buf; - - if (ioreq->req.nr_segments == 0) { - return 0; - } - - count = ioreq->req.nr_segments; - - for (i = 0; i < count; i++) { - if (to_domain) { - segs[i].dest.foreign.ref = ioreq->req.seg[i].gref; - segs[i].dest.foreign.offset = ioreq->req.seg[i].first_sect * file_blk; - segs[i].source.virt = virt; - } else { - segs[i].source.foreign.ref = ioreq->req.seg[i].gref; - segs[i].source.foreign.offset = ioreq->req.seg[i].first_sect * file_blk; - segs[i].dest.virt = virt; - } - segs[i].len = (ioreq->req.seg[i].last_sect - - ioreq->req.seg[i].first_sect + 1) * file_blk; - virt += segs[i].len; - } - - rc = xen_be_copy_grant_refs(xendev, to_domain, segs, count); - - if (rc) { - xen_pv_printf(xendev, 0, - "failed to copy data %d\n", rc); - ioreq->aio_errors++; - return -1; - } - - return rc; -} - -static int ioreq_runio_qemu_aio(struct ioreq *ioreq); - -static void qemu_aio_complete(void *opaque, int ret) -{ - struct ioreq *ioreq = opaque; - struct XenBlkDev *blkdev = ioreq->blkdev; - struct XenDevice *xendev = &blkdev->xendev; - - aio_context_acquire(blkdev->ctx); - - if (ret != 0) { - xen_pv_printf(xendev, 0, "%s I/O error\n", - ioreq->req.operation == BLKIF_OP_READ ? "read" : "write"); - ioreq->aio_errors++; - } - - ioreq->aio_inflight--; - if (ioreq->presync) { - ioreq->presync = 0; - ioreq_runio_qemu_aio(ioreq); - goto done; - } - if (ioreq->aio_inflight > 0) { - goto done; - } - - switch (ioreq->req.operation) { - case BLKIF_OP_READ: - /* in case of failure ioreq->aio_errors is increased */ - if (ret == 0) { - ioreq_grant_copy(ioreq); - } - qemu_vfree(ioreq->buf); - break; - case BLKIF_OP_WRITE: - case BLKIF_OP_FLUSH_DISKCACHE: - if (!ioreq->req.nr_segments) { - break; - } - qemu_vfree(ioreq->buf); - break; - default: - break; - } - - ioreq->status = ioreq->aio_errors ? BLKIF_RSP_ERROR : BLKIF_RSP_OKAY; - ioreq_finish(ioreq); - - switch (ioreq->req.operation) { - case BLKIF_OP_WRITE: - case BLKIF_OP_FLUSH_DISKCACHE: - if (!ioreq->req.nr_segments) { - break; - } - case BLKIF_OP_READ: - if (ioreq->status == BLKIF_RSP_OKAY) { - block_acct_done(blk_get_stats(blkdev->blk), &ioreq->acct); - } else { - block_acct_failed(blk_get_stats(blkdev->blk), &ioreq->acct); - } - break; - case BLKIF_OP_DISCARD: - default: - break; - } - qemu_bh_schedule(blkdev->bh); - -done: - aio_context_release(blkdev->ctx); -} - -static bool blk_split_discard(struct ioreq *ioreq, blkif_sector_t sector_number, - uint64_t nr_sectors) -{ - struct XenBlkDev *blkdev = ioreq->blkdev; - int64_t byte_offset; - int byte_chunk; - uint64_t byte_remaining, limit; - uint64_t sec_start = sector_number; - uint64_t sec_count = nr_sectors; - - /* Wrap around, or overflowing byte limit? */ - if (sec_start + sec_count < sec_count || - sec_start + sec_count > INT64_MAX >> BDRV_SECTOR_BITS) { - return false; - } - - limit = BDRV_REQUEST_MAX_SECTORS << BDRV_SECTOR_BITS; - byte_offset = sec_start << BDRV_SECTOR_BITS; - byte_remaining = sec_count << BDRV_SECTOR_BITS; - - do { - byte_chunk = byte_remaining > limit ? limit : byte_remaining; - ioreq->aio_inflight++; - blk_aio_pdiscard(blkdev->blk, byte_offset, byte_chunk, - qemu_aio_complete, ioreq); - byte_remaining -= byte_chunk; - byte_offset += byte_chunk; - } while (byte_remaining > 0); - - return true; -} - -static int ioreq_runio_qemu_aio(struct ioreq *ioreq) -{ - struct XenBlkDev *blkdev = ioreq->blkdev; - - ioreq->buf = qemu_memalign(XC_PAGE_SIZE, ioreq->size); - if (ioreq->req.nr_segments && - (ioreq->req.operation == BLKIF_OP_WRITE || - ioreq->req.operation == BLKIF_OP_FLUSH_DISKCACHE) && - ioreq_grant_copy(ioreq)) { - qemu_vfree(ioreq->buf); - goto err; - } - - ioreq->aio_inflight++; - if (ioreq->presync) { - blk_aio_flush(ioreq->blkdev->blk, qemu_aio_complete, ioreq); - return 0; - } - - switch (ioreq->req.operation) { - case BLKIF_OP_READ: - qemu_iovec_add(&ioreq->v, ioreq->buf, ioreq->size); - block_acct_start(blk_get_stats(blkdev->blk), &ioreq->acct, - ioreq->v.size, BLOCK_ACCT_READ); - ioreq->aio_inflight++; - blk_aio_preadv(blkdev->blk, ioreq->start, &ioreq->v, 0, - qemu_aio_complete, ioreq); - break; - case BLKIF_OP_WRITE: - case BLKIF_OP_FLUSH_DISKCACHE: - if (!ioreq->req.nr_segments) { - break; - } - - qemu_iovec_add(&ioreq->v, ioreq->buf, ioreq->size); - block_acct_start(blk_get_stats(blkdev->blk), &ioreq->acct, - ioreq->v.size, - ioreq->req.operation == BLKIF_OP_WRITE ? - BLOCK_ACCT_WRITE : BLOCK_ACCT_FLUSH); - ioreq->aio_inflight++; - blk_aio_pwritev(blkdev->blk, ioreq->start, &ioreq->v, 0, - qemu_aio_complete, ioreq); - break; - case BLKIF_OP_DISCARD: - { - struct blkif_request_discard *req = (void *)&ioreq->req; - if (!blk_split_discard(ioreq, req->sector_number, req->nr_sectors)) { - goto err; - } - break; - } - default: - /* unknown operation (shouldn't happen -- parse catches this) */ - goto err; - } - - qemu_aio_complete(ioreq, 0); - - return 0; - -err: - ioreq_finish(ioreq); - ioreq->status = BLKIF_RSP_ERROR; - return -1; -} - -static int blk_send_response_one(struct ioreq *ioreq) -{ - struct XenBlkDev *blkdev = ioreq->blkdev; - int send_notify = 0; - int have_requests = 0; - blkif_response_t *resp; - - /* Place on the response ring for the relevant domain. */ - switch (blkdev->protocol) { - case BLKIF_PROTOCOL_NATIVE: - resp = (blkif_response_t *) RING_GET_RESPONSE(&blkdev->rings.native, - blkdev->rings.native.rsp_prod_pvt); - break; - case BLKIF_PROTOCOL_X86_32: - resp = (blkif_response_t *) RING_GET_RESPONSE(&blkdev->rings.x86_32_part, - blkdev->rings.x86_32_part.rsp_prod_pvt); - break; - case BLKIF_PROTOCOL_X86_64: - resp = (blkif_response_t *) RING_GET_RESPONSE(&blkdev->rings.x86_64_part, - blkdev->rings.x86_64_part.rsp_prod_pvt); - break; - default: - return 0; - } - - resp->id = ioreq->req.id; - resp->operation = ioreq->req.operation; - resp->status = ioreq->status; - - blkdev->rings.common.rsp_prod_pvt++; - - RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&blkdev->rings.common, send_notify); - if (blkdev->rings.common.rsp_prod_pvt == blkdev->rings.common.req_cons) { - /* - * Tail check for pending requests. Allows frontend to avoid - * notifications if requests are already in flight (lower - * overheads and promotes batching). - */ - RING_FINAL_CHECK_FOR_REQUESTS(&blkdev->rings.common, have_requests); - } else if (RING_HAS_UNCONSUMED_REQUESTS(&blkdev->rings.common)) { - have_requests = 1; - } - - if (have_requests) { - blkdev->more_work++; - } - return send_notify; -} - -/* walk finished list, send outstanding responses, free requests */ -static void blk_send_response_all(struct XenBlkDev *blkdev) -{ - struct ioreq *ioreq; - int send_notify = 0; - - while (!QLIST_EMPTY(&blkdev->finished)) { - ioreq = QLIST_FIRST(&blkdev->finished); - send_notify += blk_send_response_one(ioreq); - ioreq_release(ioreq, true); - } - if (send_notify) { - xen_pv_send_notify(&blkdev->xendev); - } -} - -static int blk_get_request(struct XenBlkDev *blkdev, struct ioreq *ioreq, RING_IDX rc) -{ - switch (blkdev->protocol) { - case BLKIF_PROTOCOL_NATIVE: - memcpy(&ioreq->req, RING_GET_REQUEST(&blkdev->rings.native, rc), - sizeof(ioreq->req)); - break; - case BLKIF_PROTOCOL_X86_32: - blkif_get_x86_32_req(&ioreq->req, - RING_GET_REQUEST(&blkdev->rings.x86_32_part, rc)); - break; - case BLKIF_PROTOCOL_X86_64: - blkif_get_x86_64_req(&ioreq->req, - RING_GET_REQUEST(&blkdev->rings.x86_64_part, rc)); - break; - } - /* Prevent the compiler from accessing the on-ring fields instead. */ - barrier(); - return 0; -} - -static void blk_handle_requests(struct XenBlkDev *blkdev) -{ - RING_IDX rc, rp; - struct ioreq *ioreq; - - blkdev->more_work = 0; - - rc = blkdev->rings.common.req_cons; - rp = blkdev->rings.common.sring->req_prod; - xen_rmb(); /* Ensure we see queued requests up to 'rp'. */ - - blk_send_response_all(blkdev); - while (rc != rp) { - /* pull request from ring */ - if (RING_REQUEST_CONS_OVERFLOW(&blkdev->rings.common, rc)) { - break; - } - ioreq = ioreq_start(blkdev); - if (ioreq == NULL) { - blkdev->more_work++; - break; - } - blk_get_request(blkdev, ioreq, rc); - blkdev->rings.common.req_cons = ++rc; - - /* parse them */ - if (ioreq_parse(ioreq) != 0) { - - switch (ioreq->req.operation) { - case BLKIF_OP_READ: - block_acct_invalid(blk_get_stats(blkdev->blk), - BLOCK_ACCT_READ); - break; - case BLKIF_OP_WRITE: - block_acct_invalid(blk_get_stats(blkdev->blk), - BLOCK_ACCT_WRITE); - break; - case BLKIF_OP_FLUSH_DISKCACHE: - block_acct_invalid(blk_get_stats(blkdev->blk), - BLOCK_ACCT_FLUSH); - default: - break; - }; - - if (blk_send_response_one(ioreq)) { - xen_pv_send_notify(&blkdev->xendev); - } - ioreq_release(ioreq, false); - continue; - } - - ioreq_runio_qemu_aio(ioreq); - } - - if (blkdev->more_work && blkdev->requests_inflight < blkdev->max_requests) { - qemu_bh_schedule(blkdev->bh); - } -} - -/* ------------------------------------------------------------- */ - -static void blk_bh(void *opaque) -{ - struct XenBlkDev *blkdev = opaque; - - aio_context_acquire(blkdev->ctx); - blk_handle_requests(blkdev); - aio_context_release(blkdev->ctx); -} - -static void blk_alloc(struct XenDevice *xendev) -{ - struct XenBlkDev *blkdev = container_of(xendev, struct XenBlkDev, xendev); - Error *err = NULL; - - trace_xen_disk_alloc(xendev->name); - - QLIST_INIT(&blkdev->inflight); - QLIST_INIT(&blkdev->finished); - QLIST_INIT(&blkdev->freelist); - - blkdev->iothread = iothread_create(xendev->name, &err); - assert(!err); - - blkdev->ctx = iothread_get_aio_context(blkdev->iothread); - blkdev->bh = aio_bh_new(blkdev->ctx, blk_bh, blkdev); -} - -static void blk_parse_discard(struct XenBlkDev *blkdev) -{ - struct XenDevice *xendev = &blkdev->xendev; - int enable; - - blkdev->feature_discard = true; - - if (xenstore_read_be_int(xendev, "discard-enable", &enable) == 0) { - blkdev->feature_discard = !!enable; - } - - if (blkdev->feature_discard) { - xenstore_write_be_int(xendev, "feature-discard", 1); - } -} - -static int blk_init(struct XenDevice *xendev) -{ - struct XenBlkDev *blkdev = container_of(xendev, struct XenBlkDev, xendev); - int info = 0; - char *directiosafe = NULL; - - trace_xen_disk_init(xendev->name); - - /* read xenstore entries */ - if (blkdev->params == NULL) { - char *h = NULL; - blkdev->params = xenstore_read_be_str(xendev, "params"); - if (blkdev->params != NULL) { - h = strchr(blkdev->params, ':'); - } - if (h != NULL) { - blkdev->fileproto = blkdev->params; - blkdev->filename = h+1; - *h = 0; - } else { - blkdev->fileproto = "<unset>"; - blkdev->filename = blkdev->params; - } - } - if (!strcmp("aio", blkdev->fileproto)) { - blkdev->fileproto = "raw"; - } - if (!strcmp("vhd", blkdev->fileproto)) { - blkdev->fileproto = "vpc"; - } - if (blkdev->mode == NULL) { - blkdev->mode = xenstore_read_be_str(xendev, "mode"); - } - if (blkdev->type == NULL) { - blkdev->type = xenstore_read_be_str(xendev, "type"); - } - if (blkdev->dev == NULL) { - blkdev->dev = xenstore_read_be_str(xendev, "dev"); - } - if (blkdev->devtype == NULL) { - blkdev->devtype = xenstore_read_be_str(xendev, "device-type"); - } - directiosafe = xenstore_read_be_str(xendev, "direct-io-safe"); - blkdev->directiosafe = (directiosafe && atoi(directiosafe)); - - /* do we have all we need? */ - if (blkdev->params == NULL || - blkdev->mode == NULL || - blkdev->type == NULL || - blkdev->dev == NULL) { - goto out_error; - } - - /* read-only ? */ - if (strcmp(blkdev->mode, "w")) { - info |= VDISK_READONLY; - } - - /* cdrom ? */ - if (blkdev->devtype && !strcmp(blkdev->devtype, "cdrom")) { - info |= VDISK_CDROM; - } - - blkdev->file_blk = BLOCK_SIZE; - - /* fill info - * blk_connect supplies sector-size and sectors - */ - xenstore_write_be_int(xendev, "feature-flush-cache", 1); - xenstore_write_be_int(xendev, "info", info); - - xenstore_write_be_int(xendev, "max-ring-page-order", - MAX_RING_PAGE_ORDER); - - blk_parse_discard(blkdev); - - g_free(directiosafe); - return 0; - -out_error: - g_free(blkdev->params); - blkdev->params = NULL; - g_free(blkdev->mode); - blkdev->mode = NULL; - g_free(blkdev->type); - blkdev->type = NULL; - g_free(blkdev->dev); - blkdev->dev = NULL; - g_free(blkdev->devtype); - blkdev->devtype = NULL; - g_free(directiosafe); - blkdev->directiosafe = false; - return -1; -} - -static int blk_connect(struct XenDevice *xendev) -{ - struct XenBlkDev *blkdev = container_of(xendev, struct XenBlkDev, xendev); - int index, qflags; - bool readonly = true; - bool writethrough = true; - int order, ring_ref; - unsigned int ring_size, max_grants; - unsigned int i; - - trace_xen_disk_connect(xendev->name); - - /* read-only ? */ - if (blkdev->directiosafe) { - qflags = BDRV_O_NOCACHE | BDRV_O_NATIVE_AIO; - } else { - qflags = 0; - writethrough = false; - } - if (strcmp(blkdev->mode, "w") == 0) { - qflags |= BDRV_O_RDWR; - readonly = false; - } - if (blkdev->feature_discard) { - qflags |= BDRV_O_UNMAP; - } - - /* init qemu block driver */ - index = (xendev->dev - 202 * 256) / 16; - blkdev->dinfo = drive_get(IF_XEN, 0, index); - if (!blkdev->dinfo) { - Error *local_err = NULL; - QDict *options = NULL; - - if (strcmp(blkdev->fileproto, "<unset>")) { - options = qdict_new(); - qdict_put_str(options, "driver", blkdev->fileproto); - } - - /* setup via xenbus -> create new block driver instance */ - xen_pv_printf(xendev, 2, "create new bdrv (xenbus setup)\n"); - blkdev->blk = blk_new_open(blkdev->filename, NULL, options, - qflags, &local_err); - if (!blkdev->blk) { - xen_pv_printf(xendev, 0, "error: %s\n", - error_get_pretty(local_err)); - error_free(local_err); - return -1; - } - blk_set_enable_write_cache(blkdev->blk, !writethrough); - } else { - /* setup via qemu cmdline -> already setup for us */ - xen_pv_printf(xendev, 2, - "get configured bdrv (cmdline setup)\n"); - blkdev->blk = blk_by_legacy_dinfo(blkdev->dinfo); - if (blk_is_read_only(blkdev->blk) && !readonly) { - xen_pv_printf(xendev, 0, "Unexpected read-only drive"); - blkdev->blk = NULL; - return -1; - } - /* blkdev->blk is not create by us, we get a reference - * so we can blk_unref() unconditionally */ - blk_ref(blkdev->blk); - } - blk_attach_dev_legacy(blkdev->blk, blkdev); - blkdev->file_size = blk_getlength(blkdev->blk); - if (blkdev->file_size < 0) { - BlockDriverState *bs = blk_bs(blkdev->blk); - const char *drv_name = bs ? bdrv_get_format_name(bs) : NULL; - xen_pv_printf(xendev, 1, "blk_getlength: %d (%s) | drv %s\n", - (int)blkdev->file_size, strerror(-blkdev->file_size), - drv_name ?: "-"); - blkdev->file_size = 0; - } - - xen_pv_printf(xendev, 1, "type \"%s\", fileproto \"%s\", filename \"%s\"," - " size %" PRId64 " (%" PRId64 " MB)\n", - blkdev->type, blkdev->fileproto, blkdev->filename, - blkdev->file_size, blkdev->file_size / MiB); - - /* Fill in number of sector size and number of sectors */ - xenstore_write_be_int(xendev, "sector-size", blkdev->file_blk); - xenstore_write_be_int64(xendev, "sectors", - blkdev->file_size / blkdev->file_blk); - - if (xenstore_read_fe_int(xendev, "ring-page-order", - &order) == -1) { - blkdev->nr_ring_ref = 1; - - if (xenstore_read_fe_int(xendev, "ring-ref", - &ring_ref) == -1) { - return -1; - } - blkdev->ring_ref[0] = ring_ref; - - } else if (order >= 0 && order <= MAX_RING_PAGE_ORDER) { - blkdev->nr_ring_ref = 1 << order; - - for (i = 0; i < blkdev->nr_ring_ref; i++) { - char *key; - - key = g_strdup_printf("ring-ref%u", i); - if (!key) { - return -1; - } - - if (xenstore_read_fe_int(xendev, key, - &ring_ref) == -1) { - g_free(key); - return -1; - } - blkdev->ring_ref[i] = ring_ref; - - g_free(key); - } - } else { - xen_pv_printf(xendev, 0, "invalid ring-page-order: %d\n", - order); - return -1; - } - - if (xenstore_read_fe_int(xendev, "event-channel", - &xendev->remote_port) == -1) { - return -1; - } - - if (!xendev->protocol) { - blkdev->protocol = BLKIF_PROTOCOL_NATIVE; - } else if (strcmp(xendev->protocol, XEN_IO_PROTO_ABI_NATIVE) == 0) { - blkdev->protocol = BLKIF_PROTOCOL_NATIVE; - } else if (strcmp(xendev->protocol, XEN_IO_PROTO_ABI_X86_32) == 0) { - blkdev->protocol = BLKIF_PROTOCOL_X86_32; - } else if (strcmp(xendev->protocol, XEN_IO_PROTO_ABI_X86_64) == 0) { - blkdev->protocol = BLKIF_PROTOCOL_X86_64; - } else { - blkdev->protocol = BLKIF_PROTOCOL_NATIVE; - } - - ring_size = XC_PAGE_SIZE * blkdev->nr_ring_ref; - switch (blkdev->protocol) { - case BLKIF_PROTOCOL_NATIVE: - { - blkdev->max_requests = __CONST_RING_SIZE(blkif, ring_size); - break; - } - case BLKIF_PROTOCOL_X86_32: - { - blkdev->max_requests = __CONST_RING_SIZE(blkif_x86_32, ring_size); - break; - } - case BLKIF_PROTOCOL_X86_64: - { - blkdev->max_requests = __CONST_RING_SIZE(blkif_x86_64, ring_size); - break; - } - default: - return -1; - } - - /* Add on the number needed for the ring pages */ - max_grants = blkdev->nr_ring_ref; - - xen_be_set_max_grant_refs(xendev, max_grants); - blkdev->sring = xen_be_map_grant_refs(xendev, blkdev->ring_ref, - blkdev->nr_ring_ref, - PROT_READ | PROT_WRITE); - if (!blkdev->sring) { - return -1; - } - - switch (blkdev->protocol) { - case BLKIF_PROTOCOL_NATIVE: - { - blkif_sring_t *sring_native = blkdev->sring; - BACK_RING_INIT(&blkdev->rings.native, sring_native, ring_size); - break; - } - case BLKIF_PROTOCOL_X86_32: - { - blkif_x86_32_sring_t *sring_x86_32 = blkdev->sring; - - BACK_RING_INIT(&blkdev->rings.x86_32_part, sring_x86_32, ring_size); - break; - } - case BLKIF_PROTOCOL_X86_64: - { - blkif_x86_64_sring_t *sring_x86_64 = blkdev->sring; - - BACK_RING_INIT(&blkdev->rings.x86_64_part, sring_x86_64, ring_size); - break; - } - } - - blk_set_aio_context(blkdev->blk, blkdev->ctx); - - xen_be_bind_evtchn(xendev); - - xen_pv_printf(xendev, 1, "ok: proto %s, nr-ring-ref %u, " - "remote port %d, local port %d\n", - xendev->protocol, blkdev->nr_ring_ref, - xendev->remote_port, xendev->local_port); - return 0; -} - -static void blk_disconnect(struct XenDevice *xendev) -{ - struct XenBlkDev *blkdev = container_of(xendev, struct XenBlkDev, xendev); - - trace_xen_disk_disconnect(xendev->name); - - aio_context_acquire(blkdev->ctx); - - if (blkdev->blk) { - blk_set_aio_context(blkdev->blk, qemu_get_aio_context()); - blk_detach_dev(blkdev->blk, blkdev); - blk_unref(blkdev->blk); - blkdev->blk = NULL; - } - xen_pv_unbind_evtchn(xendev); - - aio_context_release(blkdev->ctx); - - if (blkdev->sring) { - xen_be_unmap_grant_refs(xendev, blkdev->sring, - blkdev->nr_ring_ref); - blkdev->sring = NULL; - } -} - -static int blk_free(struct XenDevice *xendev) -{ - struct XenBlkDev *blkdev = container_of(xendev, struct XenBlkDev, xendev); - struct ioreq *ioreq; - - trace_xen_disk_free(xendev->name); - - blk_disconnect(xendev); - - while (!QLIST_EMPTY(&blkdev->freelist)) { - ioreq = QLIST_FIRST(&blkdev->freelist); - QLIST_REMOVE(ioreq, list); - qemu_iovec_destroy(&ioreq->v); - g_free(ioreq); - } - - g_free(blkdev->params); - g_free(blkdev->mode); - g_free(blkdev->type); - g_free(blkdev->dev); - g_free(blkdev->devtype); - qemu_bh_delete(blkdev->bh); - iothread_destroy(blkdev->iothread); - return 0; -} - -static void blk_event(struct XenDevice *xendev) -{ - struct XenBlkDev *blkdev = container_of(xendev, struct XenBlkDev, xendev); - - qemu_bh_schedule(blkdev->bh); -} - -struct XenDevOps xen_blkdev_ops = { - .flags = DEVOPS_FLAG_NEED_GNTDEV, - .size = sizeof(struct XenBlkDev), - .alloc = blk_alloc, - .init = blk_init, - .initialise = blk_connect, - .disconnect = blk_disconnect, - .event = blk_event, - .free = blk_free, -}; diff --git a/hw/char/xen_console.c b/hw/char/xen_console.c index b1a1e66d5a..dc6ff0e5b3 100644 --- a/hw/char/xen_console.c +++ b/hw/char/xen_console.c @@ -26,7 +26,7 @@ #include "qapi/error.h" #include "hw/hw.h" #include "chardev/char-fe.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include <xen/io/console.h> @@ -39,7 +39,7 @@ struct buffer { }; struct XenConsole { - struct XenDevice xendev; /* must be first */ + struct XenLegacyDevice xendev; /* must be first */ struct buffer buffer; char console[XEN_BUFSIZE]; int ring_ref; @@ -173,7 +173,7 @@ static void xencons_send(struct XenConsole *con) /* -------------------------------------------------------------------- */ -static int con_init(struct XenDevice *xendev) +static int con_init(struct XenLegacyDevice *xendev) { struct XenConsole *con = container_of(xendev, struct XenConsole, xendev); char *type, *dom, label[32]; @@ -222,7 +222,7 @@ out: return ret; } -static int con_initialise(struct XenDevice *xendev) +static int con_initialise(struct XenLegacyDevice *xendev) { struct XenConsole *con = container_of(xendev, struct XenConsole, xendev); int limit; @@ -259,7 +259,7 @@ static int con_initialise(struct XenDevice *xendev) return 0; } -static void con_disconnect(struct XenDevice *xendev) +static void con_disconnect(struct XenLegacyDevice *xendev) { struct XenConsole *con = container_of(xendev, struct XenConsole, xendev); @@ -276,7 +276,7 @@ static void con_disconnect(struct XenDevice *xendev) } } -static void con_event(struct XenDevice *xendev) +static void con_event(struct XenLegacyDevice *xendev) { struct XenConsole *con = container_of(xendev, struct XenConsole, xendev); diff --git a/hw/display/xenfb.c b/hw/display/xenfb.c index 0330dc6f61..6202f1150e 100644 --- a/hw/display/xenfb.c +++ b/hw/display/xenfb.c @@ -30,7 +30,7 @@ #include "hw/hw.h" #include "ui/input.h" #include "ui/console.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include <xen/event_channel.h> #include <xen/io/fbif.h> @@ -46,7 +46,7 @@ /* -------------------------------------------------------------------- */ struct common { - struct XenDevice xendev; /* must be first */ + struct XenLegacyDevice xendev; /* must be first */ void *page; }; @@ -342,14 +342,14 @@ static QemuInputHandler xenfb_rel_mouse = { .sync = xenfb_mouse_sync, }; -static int input_init(struct XenDevice *xendev) +static int input_init(struct XenLegacyDevice *xendev) { xenstore_write_be_int(xendev, "feature-abs-pointer", 1); xenstore_write_be_int(xendev, "feature-raw-pointer", 1); return 0; } -static int input_initialise(struct XenDevice *xendev) +static int input_initialise(struct XenLegacyDevice *xendev) { struct XenInput *in = container_of(xendev, struct XenInput, c.xendev); int rc; @@ -361,7 +361,7 @@ static int input_initialise(struct XenDevice *xendev) return 0; } -static void input_connected(struct XenDevice *xendev) +static void input_connected(struct XenLegacyDevice *xendev) { struct XenInput *in = container_of(xendev, struct XenInput, c.xendev); @@ -395,7 +395,7 @@ static void input_connected(struct XenDevice *xendev) } } -static void input_disconnect(struct XenDevice *xendev) +static void input_disconnect(struct XenLegacyDevice *xendev) { struct XenInput *in = container_of(xendev, struct XenInput, c.xendev); @@ -410,7 +410,7 @@ static void input_disconnect(struct XenDevice *xendev) common_unbind(&in->c); } -static void input_event(struct XenDevice *xendev) +static void input_event(struct XenLegacyDevice *xendev) { struct XenInput *xenfb = container_of(xendev, struct XenInput, c.xendev); struct xenkbd_page *page = xenfb->c.page; @@ -867,7 +867,7 @@ static void xenfb_handle_events(struct XenFB *xenfb) page->out_cons = cons; } -static int fb_init(struct XenDevice *xendev) +static int fb_init(struct XenLegacyDevice *xendev) { #ifdef XENFB_TYPE_RESIZE xenstore_write_be_int(xendev, "feature-resize", 1); @@ -875,7 +875,7 @@ static int fb_init(struct XenDevice *xendev) return 0; } -static int fb_initialise(struct XenDevice *xendev) +static int fb_initialise(struct XenLegacyDevice *xendev) { struct XenFB *fb = container_of(xendev, struct XenFB, c.xendev); struct xenfb_page *fb_page; @@ -912,7 +912,7 @@ static int fb_initialise(struct XenDevice *xendev) return 0; } -static void fb_disconnect(struct XenDevice *xendev) +static void fb_disconnect(struct XenLegacyDevice *xendev) { struct XenFB *fb = container_of(xendev, struct XenFB, c.xendev); @@ -935,7 +935,8 @@ static void fb_disconnect(struct XenDevice *xendev) fb->bug_trigger = 0; } -static void fb_frontend_changed(struct XenDevice *xendev, const char *node) +static void fb_frontend_changed(struct XenLegacyDevice *xendev, + const char *node) { struct XenFB *fb = container_of(xendev, struct XenFB, c.xendev); @@ -953,7 +954,7 @@ static void fb_frontend_changed(struct XenDevice *xendev, const char *node) } } -static void fb_event(struct XenDevice *xendev) +static void fb_event(struct XenLegacyDevice *xendev) { struct XenFB *xenfb = container_of(xendev, struct XenFB, c.xendev); diff --git a/hw/i2c/i2c-ddc.c b/hw/i2c/i2c-ddc.c index be34fe072c..0a0367ff38 100644 --- a/hw/i2c/i2c-ddc.c +++ b/hw/i2c/i2c-ddc.c @@ -56,7 +56,7 @@ static int i2c_ddc_rx(I2CSlave *i2c) I2CDDCState *s = I2CDDC(i2c); int value; - value = s->edid_blob[s->reg]; + value = s->edid_blob[s->reg % sizeof(s->edid_blob)]; s->reg++; return value; } diff --git a/hw/i386/pc.c b/hw/i386/pc.c index fc65049e1d..73d688f842 100644 --- a/hw/i386/pc.c +++ b/hw/i386/pc.c @@ -112,6 +112,16 @@ struct hpet_fw_config hpet_cfg = {.count = UINT8_MAX}; GlobalProperty pc_compat_3_1[] = { { "intel-iommu", "dma-drain", "off" }, + { "Opteron_G3" "-" TYPE_X86_CPU, "rdtscp", "off" }, + { "Opteron_G4" "-" TYPE_X86_CPU, "rdtscp", "off" }, + { "Opteron_G5" "-" TYPE_X86_CPU, "rdtscp", "off" }, + { "Skylake-Client" "-" TYPE_X86_CPU, "mpx", "on" }, + { "Skylake-Client-IBRS" "-" TYPE_X86_CPU, "mpx", "on" }, + { "Skylake-Server" "-" TYPE_X86_CPU, "mpx", "on" }, + { "Skylake-Server-IBRS" "-" TYPE_X86_CPU, "mpx", "on" }, + { "Cascadelake-Server" "-" TYPE_X86_CPU, "mpx", "on" }, + { "Icelake-Client" "-" TYPE_X86_CPU, "mpx", "on" }, + { "Icelake-Server" "-" TYPE_X86_CPU, "mpx", "on" }, }; const size_t pc_compat_3_1_len = G_N_ELEMENTS(pc_compat_3_1); diff --git a/hw/i386/xen/xen-hvm.c b/hw/i386/xen/xen-hvm.c index 8c8562f359..2939122e7c 100644 --- a/hw/i386/xen/xen-hvm.c +++ b/hw/i386/xen/xen-hvm.c @@ -16,7 +16,8 @@ #include "hw/i386/pc.h" #include "hw/i386/apic-msidef.h" #include "hw/xen/xen_common.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" +#include "hw/xen/xen-bus.h" #include "qapi/error.h" #include "qapi/qapi-commands-misc.h" #include "qemu/error-report.h" @@ -1484,6 +1485,8 @@ void xen_hvm_init(PCMachineState *pcms, MemoryRegion **ram_memory) QLIST_INIT(&state->dev_list); device_listener_register(&state->device_listener); + xen_bus_init(); + /* Initialize backend core & drivers */ if (xen_be_init() != 0) { error_report("xen backend core setup failed"); diff --git a/hw/i386/xen/xen-mapcache.c b/hw/i386/xen/xen-mapcache.c index 02e823c5a2..349f72d00c 100644 --- a/hw/i386/xen/xen-mapcache.c +++ b/hw/i386/xen/xen-mapcache.c @@ -14,7 +14,7 @@ #include <sys/resource.h> -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include "qemu/bitmap.h" #include <xen/hvm/params.h> diff --git a/hw/i386/xen/xen_platform.c b/hw/i386/xen/xen_platform.c index deb7a0c374..16afb54fee 100644 --- a/hw/i386/xen/xen_platform.c +++ b/hw/i386/xen/xen_platform.c @@ -30,7 +30,7 @@ #include "hw/pci/pci.h" #include "hw/irq.h" #include "hw/xen/xen_common.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include "trace.h" #include "exec/address-spaces.h" #include "sysemu/block-backend.h" diff --git a/hw/net/xen_nic.c b/hw/net/xen_nic.c index 46a8dbfc90..37cda8e4be 100644 --- a/hw/net/xen_nic.c +++ b/hw/net/xen_nic.c @@ -28,14 +28,14 @@ #include "net/net.h" #include "net/checksum.h" #include "net/util.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include <xen/io/netif.h> /* ------------------------------------------------------------- */ struct XenNetDev { - struct XenDevice xendev; /* must be first */ + struct XenLegacyDevice xendev; /* must be first */ char *mac; int tx_work; int tx_ring_ref; @@ -276,7 +276,7 @@ static NetClientInfo net_xen_info = { .receive = net_rx_packet, }; -static int net_init(struct XenDevice *xendev) +static int net_init(struct XenLegacyDevice *xendev) { struct XenNetDev *netdev = container_of(xendev, struct XenNetDev, xendev); @@ -308,7 +308,7 @@ static int net_init(struct XenDevice *xendev) return 0; } -static int net_connect(struct XenDevice *xendev) +static int net_connect(struct XenLegacyDevice *xendev) { struct XenNetDev *netdev = container_of(xendev, struct XenNetDev, xendev); int rx_copy; @@ -363,7 +363,7 @@ static int net_connect(struct XenDevice *xendev) return 0; } -static void net_disconnect(struct XenDevice *xendev) +static void net_disconnect(struct XenLegacyDevice *xendev) { struct XenNetDev *netdev = container_of(xendev, struct XenNetDev, xendev); @@ -379,14 +379,14 @@ static void net_disconnect(struct XenDevice *xendev) } } -static void net_event(struct XenDevice *xendev) +static void net_event(struct XenLegacyDevice *xendev) { struct XenNetDev *netdev = container_of(xendev, struct XenNetDev, xendev); net_tx_packets(netdev); qemu_flush_queued_packets(qemu_get_queue(netdev->nic)); } -static int net_free(struct XenDevice *xendev) +static int net_free(struct XenLegacyDevice *xendev) { struct XenNetDev *netdev = container_of(xendev, struct XenNetDev, xendev); diff --git a/hw/usb/xen-usb.c b/hw/usb/xen-usb.c index 5758a105a2..b20d0cfadf 100644 --- a/hw/usb/xen-usb.c +++ b/hw/usb/xen-usb.c @@ -27,7 +27,7 @@ #include "qemu/option.h" #include "hw/sysbus.h" #include "hw/usb.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include "monitor/qdev.h" #include "qapi/qmp/qdict.h" #include "qapi/qmp/qstring.h" @@ -99,7 +99,7 @@ struct usbback_hotplug { }; struct usbback_info { - struct XenDevice xendev; /* must be first */ + struct XenLegacyDevice xendev; /* must be first */ USBBus bus; void *urb_sring; void *conn_sring; @@ -142,7 +142,7 @@ static int usbback_gnttab_map(struct usbback_req *usbback_req) unsigned int nr_segs, i, prot; uint32_t ref[USBIF_MAX_SEGMENTS_PER_REQUEST]; struct usbback_info *usbif = usbback_req->usbif; - struct XenDevice *xendev = &usbif->xendev; + struct XenLegacyDevice *xendev = &usbif->xendev; struct usbif_request_segment *seg; void *addr; @@ -220,7 +220,7 @@ static int usbback_gnttab_map(struct usbback_req *usbback_req) static int usbback_init_packet(struct usbback_req *usbback_req) { - struct XenDevice *xendev = &usbback_req->usbif->xendev; + struct XenLegacyDevice *xendev = &usbback_req->usbif->xendev; USBPacket *packet = &usbback_req->packet; USBDevice *dev = usbback_req->stub->dev; USBEndpoint *ep; @@ -279,7 +279,7 @@ static void usbback_do_response(struct usbback_req *usbback_req, int32_t status, { struct usbback_info *usbif; struct usbif_urb_response *res; - struct XenDevice *xendev; + struct XenLegacyDevice *xendev; unsigned int notify; usbif = usbback_req->usbif; @@ -824,7 +824,7 @@ static void usbback_process_port(struct usbback_info *usbif, unsigned port) g_free(busid); } -static void usbback_disconnect(struct XenDevice *xendev) +static void usbback_disconnect(struct XenLegacyDevice *xendev) { struct usbback_info *usbif; unsigned int i; @@ -853,7 +853,7 @@ static void usbback_disconnect(struct XenDevice *xendev) TR_BUS(xendev, "finished\n"); } -static int usbback_connect(struct XenDevice *xendev) +static int usbback_connect(struct XenLegacyDevice *xendev) { struct usbback_info *usbif; struct usbif_urb_sring *urb_sring; @@ -913,7 +913,8 @@ static int usbback_connect(struct XenDevice *xendev) return 0; } -static void usbback_backend_changed(struct XenDevice *xendev, const char *node) +static void usbback_backend_changed(struct XenLegacyDevice *xendev, + const char *node) { struct usbback_info *usbif; unsigned int i; @@ -926,7 +927,7 @@ static void usbback_backend_changed(struct XenDevice *xendev, const char *node) } } -static int usbback_init(struct XenDevice *xendev) +static int usbback_init(struct XenLegacyDevice *xendev) { struct usbback_info *usbif; @@ -1005,7 +1006,7 @@ static USBPortOps xen_usb_port_ops = { static USBBusOps xen_usb_bus_ops = { }; -static void usbback_alloc(struct XenDevice *xendev) +static void usbback_alloc(struct XenLegacyDevice *xendev) { struct usbback_info *usbif; USBPort *p; @@ -1027,7 +1028,7 @@ static void usbback_alloc(struct XenDevice *xendev) usbif->bh = qemu_bh_new(usbback_bh, usbif); } -static int usbback_free(struct XenDevice *xendev) +static int usbback_free(struct XenLegacyDevice *xendev) { struct usbback_info *usbif; struct usbback_req *usbback_req; @@ -1066,7 +1067,7 @@ static int usbback_free(struct XenDevice *xendev) return 0; } -static void usbback_event(struct XenDevice *xendev) +static void usbback_event(struct XenLegacyDevice *xendev) { struct usbback_info *usbif; diff --git a/hw/xen/Makefile.objs b/hw/xen/Makefile.objs index 9ea5c73423..84df60a928 100644 --- a/hw/xen/Makefile.objs +++ b/hw/xen/Makefile.objs @@ -1,5 +1,5 @@ # xen backend driver support -common-obj-$(CONFIG_XEN) += xen_backend.o xen_devconfig.o xen_pvdev.o xen-common.o +common-obj-$(CONFIG_XEN) += xen-legacy-backend.o xen_devconfig.o xen_pvdev.o xen-common.o xen-bus.o xen-bus-helper.o xen-backend.o obj-$(CONFIG_XEN_PCI_PASSTHROUGH) += xen-host-pci-device.o obj-$(CONFIG_XEN_PCI_PASSTHROUGH) += xen_pt.o xen_pt_config_init.o xen_pt_graphics.o xen_pt_msi.o diff --git a/hw/xen/trace-events b/hw/xen/trace-events index c7e7a3b523..f6944624b2 100644 --- a/hw/xen/trace-events +++ b/hw/xen/trace-events @@ -12,3 +12,29 @@ xen_unmap_portio_range(uint32_t id, uint64_t start_addr, uint64_t end_addr) "id: xen_map_pcidev(uint32_t id, uint8_t bus, uint8_t dev, uint8_t func) "id: %u bdf: %02x.%02x.%02x" xen_unmap_pcidev(uint32_t id, uint8_t bus, uint8_t dev, uint8_t func) "id: %u bdf: %02x.%02x.%02x" xen_domid_restrict(int err) "err: %u" + +# include/hw/xen/xen-bus.c +xen_bus_realize(void) "" +xen_bus_unrealize(void) "" +xen_bus_enumerate(void) "" +xen_bus_type_enumerate(const char *type) "type: %s" +xen_bus_backend_create(const char *type, const char *path) "type: %s path: %s" +xen_bus_add_watch(const char *node, const char *key, char *token) "node: %s key: %s token: %s" +xen_bus_remove_watch(const char *node, const char *key, char *token) "node: %s key: %s token: %s" +xen_bus_watch(const char *token) "token: %s" +xen_device_realize(const char *type, char *name) "type: %s name: %s" +xen_device_unrealize(const char *type, char *name) "type: %s name: %s" +xen_device_backend_state(const char *type, char *name, const char *state) "type: %s name: %s -> %s" +xen_device_backend_online(const char *type, char *name, bool online) "type: %s name: %s -> %u" +xen_device_backend_changed(const char *type, char *name) "type: %s name: %s" +xen_device_frontend_state(const char *type, char *name, const char *state) "type: %s name: %s -> %s" +xen_device_frontend_changed(const char *type, char *name) "type: %s name: %s" +xen_device_unplug(const char *type, char *name) "type: %s name: %s" + +# include/hw/xen/xen-bus-helper.c +xs_node_create(const char *node) "%s" +xs_node_destroy(const char *node) "%s" +xs_node_vprintf(char *path, char *value) "%s %s" +xs_node_vscanf(char *path, char *value) "%s %s" +xs_node_watch(char *path) "%s" +xs_node_unwatch(char *path) "%s" diff --git a/hw/xen/xen-backend.c b/hw/xen/xen-backend.c new file mode 100644 index 0000000000..da065f81b7 --- /dev/null +++ b/hw/xen/xen-backend.c @@ -0,0 +1,165 @@ +/* + * Copyright (c) 2018 Citrix Systems Inc. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#include "qemu/osdep.h" +#include "qemu/error-report.h" +#include "qapi/error.h" +#include "hw/xen/xen-backend.h" +#include "hw/xen/xen-bus.h" + +typedef struct XenBackendImpl { + const char *type; + XenBackendDeviceCreate create; + XenBackendDeviceDestroy destroy; +} XenBackendImpl; + +struct XenBackendInstance { + QLIST_ENTRY(XenBackendInstance) entry; + const XenBackendImpl *impl; + XenBus *xenbus; + char *name; + XenDevice *xendev; +}; + +static GHashTable *xen_backend_table_get(void) +{ + static GHashTable *table; + + if (table == NULL) { + table = g_hash_table_new(g_str_hash, g_str_equal); + } + + return table; +} + +static void xen_backend_table_add(XenBackendImpl *impl) +{ + g_hash_table_insert(xen_backend_table_get(), (void *)impl->type, impl); +} + +static const XenBackendImpl *xen_backend_table_lookup(const char *type) +{ + return g_hash_table_lookup(xen_backend_table_get(), type); +} + +void xen_backend_register(const XenBackendInfo *info) +{ + XenBackendImpl *impl = g_new0(XenBackendImpl, 1); + + g_assert(info->type); + + if (xen_backend_table_lookup(info->type)) { + error_report("attempt to register duplicate Xen backend type '%s'", + info->type); + abort(); + } + + if (!info->create) { + error_report("backend type '%s' has no creator", info->type); + abort(); + } + + impl->type = info->type; + impl->create = info->create; + impl->destroy = info->destroy; + + xen_backend_table_add(impl); +} + +static QLIST_HEAD(, XenBackendInstance) backend_list; + +static void xen_backend_list_add(XenBackendInstance *backend) +{ + QLIST_INSERT_HEAD(&backend_list, backend, entry); +} + +static XenBackendInstance *xen_backend_list_find(XenDevice *xendev) +{ + XenBackendInstance *backend; + + QLIST_FOREACH(backend, &backend_list, entry) { + if (backend->xendev == xendev) { + return backend; + } + } + + return NULL; +} + +static void xen_backend_list_remove(XenBackendInstance *backend) +{ + QLIST_REMOVE(backend, entry); +} + +void xen_backend_device_create(XenBus *xenbus, const char *type, + const char *name, QDict *opts, Error **errp) +{ + const XenBackendImpl *impl = xen_backend_table_lookup(type); + XenBackendInstance *backend; + Error *local_error = NULL; + + if (!impl) { + return; + } + + backend = g_new0(XenBackendInstance, 1); + backend->xenbus = xenbus; + backend->name = g_strdup(name); + + impl->create(backend, opts, &local_error); + if (local_error) { + error_propagate(errp, local_error); + g_free(backend->name); + g_free(backend); + return; + } + + backend->impl = impl; + xen_backend_list_add(backend); +} + +XenBus *xen_backend_get_bus(XenBackendInstance *backend) +{ + return backend->xenbus; +} + +const char *xen_backend_get_name(XenBackendInstance *backend) +{ + return backend->name; +} + +void xen_backend_set_device(XenBackendInstance *backend, + XenDevice *xendev) +{ + g_assert(!backend->xendev); + backend->xendev = xendev; +} + +XenDevice *xen_backend_get_device(XenBackendInstance *backend) +{ + return backend->xendev; +} + + +bool xen_backend_try_device_destroy(XenDevice *xendev, Error **errp) +{ + XenBackendInstance *backend = xen_backend_list_find(xendev); + const XenBackendImpl *impl; + + if (!backend) { + return false; + } + + impl = backend->impl; + impl->destroy(backend, errp); + + xen_backend_list_remove(backend); + g_free(backend->name); + g_free(backend); + + return true; +} diff --git a/hw/xen/xen-bus-helper.c b/hw/xen/xen-bus-helper.c new file mode 100644 index 0000000000..5f7a4b2612 --- /dev/null +++ b/hw/xen/xen-bus-helper.c @@ -0,0 +1,184 @@ +/* + * Copyright (c) 2018 Citrix Systems Inc. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#include "qemu/osdep.h" +#include "hw/hw.h" +#include "hw/sysbus.h" +#include "hw/xen/xen.h" +#include "hw/xen/xen-bus.h" +#include "hw/xen/xen-bus-helper.h" +#include "qapi/error.h" + +#include <glib/gprintf.h> + +struct xs_state { + enum xenbus_state statenum; + const char *statestr; +}; +#define XS_STATE(state) { state, #state } + +static struct xs_state xs_state[] = { + XS_STATE(XenbusStateUnknown), + XS_STATE(XenbusStateInitialising), + XS_STATE(XenbusStateInitWait), + XS_STATE(XenbusStateInitialised), + XS_STATE(XenbusStateConnected), + XS_STATE(XenbusStateClosing), + XS_STATE(XenbusStateClosed), + XS_STATE(XenbusStateReconfiguring), + XS_STATE(XenbusStateReconfigured), +}; + +#undef XS_STATE + +const char *xs_strstate(enum xenbus_state state) +{ + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(xs_state); i++) { + if (xs_state[i].statenum == state) { + return xs_state[i].statestr; + } + } + + return "INVALID"; +} + +void xs_node_create(struct xs_handle *xsh, xs_transaction_t tid, + const char *node, struct xs_permissions perms[], + unsigned int nr_perms, Error **errp) +{ + trace_xs_node_create(node); + + if (!xs_write(xsh, tid, node, "", 0)) { + error_setg_errno(errp, errno, "failed to create node '%s'", node); + return; + } + + if (!xs_set_permissions(xsh, tid, node, perms, nr_perms)) { + error_setg_errno(errp, errno, "failed to set node '%s' permissions", + node); + } +} + +void xs_node_destroy(struct xs_handle *xsh, xs_transaction_t tid, + const char *node, Error **errp) +{ + trace_xs_node_destroy(node); + + if (!xs_rm(xsh, tid, node)) { + error_setg_errno(errp, errno, "failed to destroy node '%s'", node); + } +} + +void xs_node_vprintf(struct xs_handle *xsh, xs_transaction_t tid, + const char *node, const char *key, Error **errp, + const char *fmt, va_list ap) +{ + char *path, *value; + int len; + + path = (strlen(node) != 0) ? g_strdup_printf("%s/%s", node, key) : + g_strdup(key); + len = g_vasprintf(&value, fmt, ap); + + trace_xs_node_vprintf(path, value); + + if (!xs_write(xsh, tid, path, value, len)) { + error_setg_errno(errp, errno, "failed to write '%s' to '%s'", + value, path); + } + + g_free(value); + g_free(path); +} + +void xs_node_printf(struct xs_handle *xsh, xs_transaction_t tid, + const char *node, const char *key, Error **errp, + const char *fmt, ...) +{ + va_list ap; + + va_start(ap, fmt); + xs_node_vprintf(xsh, tid, node, key, errp, fmt, ap); + va_end(ap); +} + +int xs_node_vscanf(struct xs_handle *xsh, xs_transaction_t tid, + const char *node, const char *key, Error **errp, + const char *fmt, va_list ap) +{ + char *path, *value; + int rc; + + path = (strlen(node) != 0) ? g_strdup_printf("%s/%s", node, key) : + g_strdup(key); + value = xs_read(xsh, tid, path, NULL); + + trace_xs_node_vscanf(path, value); + + if (value) { + rc = vsscanf(value, fmt, ap); + } else { + error_setg_errno(errp, errno, "failed to read from '%s'", + path); + rc = EOF; + } + + free(value); + g_free(path); + + return rc; +} + +int xs_node_scanf(struct xs_handle *xsh, xs_transaction_t tid, + const char *node, const char *key, Error **errp, + const char *fmt, ...) +{ + va_list ap; + int rc; + + va_start(ap, fmt); + rc = xs_node_vscanf(xsh, tid, node, key, errp, fmt, ap); + va_end(ap); + + return rc; +} + +void xs_node_watch(struct xs_handle *xsh, const char *node, const char *key, + char *token, Error **errp) +{ + char *path; + + path = (strlen(node) != 0) ? g_strdup_printf("%s/%s", node, key) : + g_strdup(key); + + trace_xs_node_watch(path); + + if (!xs_watch(xsh, path, token)) { + error_setg_errno(errp, errno, "failed to watch node '%s'", path); + } + + g_free(path); +} + +void xs_node_unwatch(struct xs_handle *xsh, const char *node, + const char *key, const char *token, Error **errp) +{ + char *path; + + path = (strlen(node) != 0) ? g_strdup_printf("%s/%s", node, key) : + g_strdup(key); + + trace_xs_node_unwatch(path); + + if (!xs_unwatch(xsh, path, token)) { + error_setg_errno(errp, errno, "failed to unwatch node '%s'", path); + } + + g_free(path); +} diff --git a/hw/xen/xen-bus.c b/hw/xen/xen-bus.c new file mode 100644 index 0000000000..3aeccec69c --- /dev/null +++ b/hw/xen/xen-bus.c @@ -0,0 +1,1199 @@ +/* + * Copyright (c) 2018 Citrix Systems Inc. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#include "qemu/osdep.h" +#include "qemu/main-loop.h" +#include "qemu/uuid.h" +#include "hw/hw.h" +#include "hw/sysbus.h" +#include "hw/xen/xen.h" +#include "hw/xen/xen-backend.h" +#include "hw/xen/xen-bus.h" +#include "hw/xen/xen-bus-helper.h" +#include "monitor/monitor.h" +#include "qapi/error.h" +#include "qapi/qmp/qdict.h" +#include "sysemu/sysemu.h" +#include "trace.h" + +static char *xen_device_get_backend_path(XenDevice *xendev) +{ + XenBus *xenbus = XEN_BUS(qdev_get_parent_bus(DEVICE(xendev))); + XenDeviceClass *xendev_class = XEN_DEVICE_GET_CLASS(xendev); + const char *type = object_get_typename(OBJECT(xendev)); + const char *backend = xendev_class->backend; + + if (!backend) { + backend = type; + } + + return g_strdup_printf("/local/domain/%u/backend/%s/%u/%s", + xenbus->backend_id, backend, xendev->frontend_id, + xendev->name); +} + +static char *xen_device_get_frontend_path(XenDevice *xendev) +{ + XenDeviceClass *xendev_class = XEN_DEVICE_GET_CLASS(xendev); + const char *type = object_get_typename(OBJECT(xendev)); + const char *device = xendev_class->device; + + if (!device) { + device = type; + } + + return g_strdup_printf("/local/domain/%u/device/%s/%s", + xendev->frontend_id, device, xendev->name); +} + +static void xen_device_unplug(XenDevice *xendev, Error **errp) +{ + XenBus *xenbus = XEN_BUS(qdev_get_parent_bus(DEVICE(xendev))); + const char *type = object_get_typename(OBJECT(xendev)); + Error *local_err = NULL; + xs_transaction_t tid; + + trace_xen_device_unplug(type, xendev->name); + + /* Mimic the way the Xen toolstack does an unplug */ +again: + tid = xs_transaction_start(xenbus->xsh); + if (tid == XBT_NULL) { + error_setg_errno(errp, errno, "failed xs_transaction_start"); + return; + } + + xs_node_printf(xenbus->xsh, tid, xendev->backend_path, "online", + &local_err, "%u", 0); + if (local_err) { + goto abort; + } + + xs_node_printf(xenbus->xsh, tid, xendev->backend_path, "state", + &local_err, "%u", XenbusStateClosing); + if (local_err) { + goto abort; + } + + if (!xs_transaction_end(xenbus->xsh, tid, false)) { + if (errno == EAGAIN) { + goto again; + } + + error_setg_errno(errp, errno, "failed xs_transaction_end"); + } + + return; + +abort: + /* + * We only abort if there is already a failure so ignore any error + * from ending the transaction. + */ + xs_transaction_end(xenbus->xsh, tid, true); + error_propagate(errp, local_err); +} + +static void xen_bus_print_dev(Monitor *mon, DeviceState *dev, int indent) +{ + XenDevice *xendev = XEN_DEVICE(dev); + + monitor_printf(mon, "%*sname = '%s' frontend_id = %u\n", + indent, "", xendev->name, xendev->frontend_id); +} + +static char *xen_bus_get_dev_path(DeviceState *dev) +{ + return xen_device_get_backend_path(XEN_DEVICE(dev)); +} + +struct XenWatch { + char *node, *key; + char *token; + XenWatchHandler handler; + void *opaque; + Notifier notifier; +}; + +static void watch_notify(Notifier *n, void *data) +{ + XenWatch *watch = container_of(n, XenWatch, notifier); + const char *token = data; + + if (!strcmp(watch->token, token)) { + watch->handler(watch->opaque); + } +} + +static XenWatch *new_watch(const char *node, const char *key, + XenWatchHandler handler, void *opaque) +{ + XenWatch *watch = g_new0(XenWatch, 1); + QemuUUID uuid; + + qemu_uuid_generate(&uuid); + + watch->token = qemu_uuid_unparse_strdup(&uuid); + watch->node = g_strdup(node); + watch->key = g_strdup(key); + watch->handler = handler; + watch->opaque = opaque; + watch->notifier.notify = watch_notify; + + return watch; +} + +static void free_watch(XenWatch *watch) +{ + g_free(watch->token); + g_free(watch->key); + g_free(watch->node); + + g_free(watch); +} + +static XenWatch *xen_bus_add_watch(XenBus *xenbus, const char *node, + const char *key, XenWatchHandler handler, + void *opaque, Error **errp) +{ + XenWatch *watch = new_watch(node, key, handler, opaque); + Error *local_err = NULL; + + trace_xen_bus_add_watch(watch->node, watch->key, watch->token); + + notifier_list_add(&xenbus->watch_notifiers, &watch->notifier); + + xs_node_watch(xenbus->xsh, node, key, watch->token, &local_err); + if (local_err) { + error_propagate(errp, local_err); + + notifier_remove(&watch->notifier); + free_watch(watch); + + return NULL; + } + + return watch; +} + +static void xen_bus_remove_watch(XenBus *xenbus, XenWatch *watch, + Error **errp) +{ + trace_xen_bus_remove_watch(watch->node, watch->key, watch->token); + + xs_node_unwatch(xenbus->xsh, watch->node, watch->key, watch->token, + errp); + + notifier_remove(&watch->notifier); + free_watch(watch); +} + +static void xen_bus_backend_create(XenBus *xenbus, const char *type, + const char *name, char *path, + Error **errp) +{ + xs_transaction_t tid; + char **key; + QDict *opts; + unsigned int i, n; + Error *local_err = NULL; + + trace_xen_bus_backend_create(type, path); + +again: + tid = xs_transaction_start(xenbus->xsh); + if (tid == XBT_NULL) { + error_setg(errp, "failed xs_transaction_start"); + return; + } + + key = xs_directory(xenbus->xsh, tid, path, &n); + if (!key) { + if (!xs_transaction_end(xenbus->xsh, tid, true)) { + error_setg_errno(errp, errno, "failed xs_transaction_end"); + } + return; + } + + opts = qdict_new(); + for (i = 0; i < n; i++) { + char *val; + + /* + * Assume anything found in the xenstore backend area, other than + * the keys created for a generic XenDevice, are parameters + * to be used to configure the backend. + */ + if (!strcmp(key[i], "state") || + !strcmp(key[i], "online") || + !strcmp(key[i], "frontend") || + !strcmp(key[i], "frontend-id") || + !strcmp(key[i], "hotplug-status")) + continue; + + if (xs_node_scanf(xenbus->xsh, tid, path, key[i], NULL, "%ms", + &val) == 1) { + qdict_put_str(opts, key[i], val); + free(val); + } + } + + free(key); + + if (!xs_transaction_end(xenbus->xsh, tid, false)) { + qobject_unref(opts); + + if (errno == EAGAIN) { + goto again; + } + + error_setg_errno(errp, errno, "failed xs_transaction_end"); + return; + } + + xen_backend_device_create(xenbus, type, name, opts, &local_err); + qobject_unref(opts); + + if (local_err) { + error_propagate_prepend(errp, local_err, + "failed to create '%s' device '%s': ", + type, name); + } +} + +static void xen_bus_type_enumerate(XenBus *xenbus, const char *type) +{ + char *domain_path = g_strdup_printf("backend/%s/%u", type, xen_domid); + char **backend; + unsigned int i, n; + + trace_xen_bus_type_enumerate(type); + + backend = xs_directory(xenbus->xsh, XBT_NULL, domain_path, &n); + if (!backend) { + goto out; + } + + for (i = 0; i < n; i++) { + char *backend_path = g_strdup_printf("%s/%s", domain_path, + backend[i]); + enum xenbus_state backend_state; + + if (xs_node_scanf(xenbus->xsh, XBT_NULL, backend_path, "state", + NULL, "%u", &backend_state) != 1) + backend_state = XenbusStateUnknown; + + if (backend_state == XenbusStateInitialising) { + Error *local_err = NULL; + + xen_bus_backend_create(xenbus, type, backend[i], backend_path, + &local_err); + if (local_err) { + error_report_err(local_err); + } + } + + g_free(backend_path); + } + + free(backend); + +out: + g_free(domain_path); +} + +static void xen_bus_enumerate(void *opaque) +{ + XenBus *xenbus = opaque; + char **type; + unsigned int i, n; + + trace_xen_bus_enumerate(); + + type = xs_directory(xenbus->xsh, XBT_NULL, "backend", &n); + if (!type) { + return; + } + + for (i = 0; i < n; i++) { + xen_bus_type_enumerate(xenbus, type[i]); + } + + free(type); +} + +static void xen_bus_unrealize(BusState *bus, Error **errp) +{ + XenBus *xenbus = XEN_BUS(bus); + + trace_xen_bus_unrealize(); + + if (xenbus->backend_watch) { + xen_bus_remove_watch(xenbus, xenbus->backend_watch, NULL); + xenbus->backend_watch = NULL; + } + + if (!xenbus->xsh) { + return; + } + + qemu_set_fd_handler(xs_fileno(xenbus->xsh), NULL, NULL, NULL); + + xs_close(xenbus->xsh); +} + +static void xen_bus_watch(void *opaque) +{ + XenBus *xenbus = opaque; + char **v; + const char *token; + + g_assert(xenbus->xsh); + + v = xs_check_watch(xenbus->xsh); + if (!v) { + return; + } + + token = v[XS_WATCH_TOKEN]; + + trace_xen_bus_watch(token); + + notifier_list_notify(&xenbus->watch_notifiers, (void *)token); + + free(v); +} + +static void xen_bus_realize(BusState *bus, Error **errp) +{ + XenBus *xenbus = XEN_BUS(bus); + unsigned int domid; + Error *local_err = NULL; + + trace_xen_bus_realize(); + + xenbus->xsh = xs_open(0); + if (!xenbus->xsh) { + error_setg_errno(errp, errno, "failed xs_open"); + goto fail; + } + + if (xs_node_scanf(xenbus->xsh, XBT_NULL, "", /* domain root node */ + "domid", NULL, "%u", &domid) == 1) { + xenbus->backend_id = domid; + } else { + xenbus->backend_id = 0; /* Assume lack of node means dom0 */ + } + + notifier_list_init(&xenbus->watch_notifiers); + qemu_set_fd_handler(xs_fileno(xenbus->xsh), xen_bus_watch, NULL, + xenbus); + + module_call_init(MODULE_INIT_XEN_BACKEND); + + xenbus->backend_watch = + xen_bus_add_watch(xenbus, "", /* domain root node */ + "backend", xen_bus_enumerate, xenbus, &local_err); + if (local_err) { + /* This need not be treated as a hard error so don't propagate */ + error_reportf_err(local_err, + "failed to set up enumeration watch: "); + } + + return; + +fail: + xen_bus_unrealize(bus, &error_abort); +} + +static void xen_bus_unplug_request(HotplugHandler *hotplug, + DeviceState *dev, + Error **errp) +{ + XenDevice *xendev = XEN_DEVICE(dev); + + xen_device_unplug(xendev, errp); +} + +static void xen_bus_class_init(ObjectClass *class, void *data) +{ + BusClass *bus_class = BUS_CLASS(class); + HotplugHandlerClass *hotplug_class = HOTPLUG_HANDLER_CLASS(class); + + bus_class->print_dev = xen_bus_print_dev; + bus_class->get_dev_path = xen_bus_get_dev_path; + bus_class->realize = xen_bus_realize; + bus_class->unrealize = xen_bus_unrealize; + + hotplug_class->unplug_request = xen_bus_unplug_request; +} + +static const TypeInfo xen_bus_type_info = { + .name = TYPE_XEN_BUS, + .parent = TYPE_BUS, + .instance_size = sizeof(XenBus), + .class_size = sizeof(XenBusClass), + .class_init = xen_bus_class_init, + .interfaces = (InterfaceInfo[]) { + { TYPE_HOTPLUG_HANDLER }, + { } + }, +}; + +void xen_device_backend_printf(XenDevice *xendev, const char *key, + const char *fmt, ...) +{ + XenBus *xenbus = XEN_BUS(qdev_get_parent_bus(DEVICE(xendev))); + Error *local_err = NULL; + va_list ap; + + g_assert(xenbus->xsh); + + va_start(ap, fmt); + xs_node_vprintf(xenbus->xsh, XBT_NULL, xendev->backend_path, key, + &local_err, fmt, ap); + va_end(ap); + + if (local_err) { + error_report_err(local_err); + } +} + +static int xen_device_backend_scanf(XenDevice *xendev, const char *key, + const char *fmt, ...) +{ + XenBus *xenbus = XEN_BUS(qdev_get_parent_bus(DEVICE(xendev))); + va_list ap; + int rc; + + g_assert(xenbus->xsh); + + va_start(ap, fmt); + rc = xs_node_vscanf(xenbus->xsh, XBT_NULL, xendev->backend_path, key, + NULL, fmt, ap); + va_end(ap); + + return rc; +} + +void xen_device_backend_set_state(XenDevice *xendev, + enum xenbus_state state) +{ + const char *type = object_get_typename(OBJECT(xendev)); + + if (xendev->backend_state == state) { + return; + } + + trace_xen_device_backend_state(type, xendev->name, + xs_strstate(state)); + + xendev->backend_state = state; + xen_device_backend_printf(xendev, "state", "%u", state); +} + +enum xenbus_state xen_device_backend_get_state(XenDevice *xendev) +{ + return xendev->backend_state; +} + +static void xen_device_backend_set_online(XenDevice *xendev, bool online) +{ + const char *type = object_get_typename(OBJECT(xendev)); + + if (xendev->backend_online == online) { + return; + } + + trace_xen_device_backend_online(type, xendev->name, online); + + xendev->backend_online = online; + xen_device_backend_printf(xendev, "online", "%u", online); +} + +static void xen_device_backend_changed(void *opaque) +{ + XenDevice *xendev = opaque; + const char *type = object_get_typename(OBJECT(xendev)); + enum xenbus_state state; + unsigned int online; + + trace_xen_device_backend_changed(type, xendev->name); + + if (xen_device_backend_scanf(xendev, "state", "%u", &state) != 1) { + state = XenbusStateUnknown; + } + + xen_device_backend_set_state(xendev, state); + + if (xen_device_backend_scanf(xendev, "online", "%u", &online) != 1) { + online = 0; + } + + xen_device_backend_set_online(xendev, !!online); + + /* + * If the toolstack (or unplug request callback) has set the backend + * state to Closing, but there is no active frontend (i.e. the + * state is not Connected) then set the backend state to Closed. + */ + if (xendev->backend_state == XenbusStateClosing && + xendev->frontend_state != XenbusStateConnected) { + xen_device_backend_set_state(xendev, XenbusStateClosed); + } + + /* + * If a backend is still 'online' then its state should be cycled + * back round to InitWait in order for a new frontend instance to + * connect. This may happen when, for example, a frontend driver is + * re-installed or updated. + * If a backend is not 'online' then the device should be destroyed. + */ + if (xendev->backend_online && + xendev->backend_state == XenbusStateClosed) { + xen_device_backend_set_state(xendev, XenbusStateInitWait); + } else if (!xendev->backend_online && + (xendev->backend_state == XenbusStateClosed || + xendev->backend_state == XenbusStateInitialising || + xendev->backend_state == XenbusStateInitWait || + xendev->backend_state == XenbusStateUnknown)) { + Error *local_err = NULL; + + if (!xen_backend_try_device_destroy(xendev, &local_err)) { + object_unparent(OBJECT(xendev)); + } + + if (local_err) { + error_report_err(local_err); + } + } +} + +static void xen_device_backend_create(XenDevice *xendev, Error **errp) +{ + XenBus *xenbus = XEN_BUS(qdev_get_parent_bus(DEVICE(xendev))); + struct xs_permissions perms[2]; + Error *local_err = NULL; + + xendev->backend_path = xen_device_get_backend_path(xendev); + + perms[0].id = xenbus->backend_id; + perms[0].perms = XS_PERM_NONE; + perms[1].id = xendev->frontend_id; + perms[1].perms = XS_PERM_READ; + + g_assert(xenbus->xsh); + + xs_node_create(xenbus->xsh, XBT_NULL, xendev->backend_path, perms, + ARRAY_SIZE(perms), &local_err); + if (local_err) { + error_propagate_prepend(errp, local_err, + "failed to create backend: "); + return; + } + + xendev->backend_state_watch = + xen_bus_add_watch(xenbus, xendev->backend_path, + "state", xen_device_backend_changed, + xendev, &local_err); + if (local_err) { + error_propagate_prepend(errp, local_err, + "failed to watch backend state: "); + return; + } + + xendev->backend_online_watch = + xen_bus_add_watch(xenbus, xendev->backend_path, + "online", xen_device_backend_changed, + xendev, &local_err); + if (local_err) { + error_propagate_prepend(errp, local_err, + "failed to watch backend online: "); + return; + } +} + +static void xen_device_backend_destroy(XenDevice *xendev) +{ + XenBus *xenbus = XEN_BUS(qdev_get_parent_bus(DEVICE(xendev))); + Error *local_err = NULL; + + if (xendev->backend_online_watch) { + xen_bus_remove_watch(xenbus, xendev->backend_online_watch, NULL); + xendev->backend_online_watch = NULL; + } + + if (xendev->backend_state_watch) { + xen_bus_remove_watch(xenbus, xendev->backend_state_watch, NULL); + xendev->backend_state_watch = NULL; + } + + if (!xendev->backend_path) { + return; + } + + g_assert(xenbus->xsh); + + xs_node_destroy(xenbus->xsh, XBT_NULL, xendev->backend_path, + &local_err); + g_free(xendev->backend_path); + xendev->backend_path = NULL; + + if (local_err) { + error_report_err(local_err); + } +} + +void xen_device_frontend_printf(XenDevice *xendev, const char *key, + const char *fmt, ...) +{ + XenBus *xenbus = XEN_BUS(qdev_get_parent_bus(DEVICE(xendev))); + Error *local_err = NULL; + va_list ap; + + g_assert(xenbus->xsh); + + va_start(ap, fmt); + xs_node_vprintf(xenbus->xsh, XBT_NULL, xendev->frontend_path, key, + &local_err, fmt, ap); + va_end(ap); + + if (local_err) { + error_report_err(local_err); + } +} + +int xen_device_frontend_scanf(XenDevice *xendev, const char *key, + const char *fmt, ...) +{ + XenBus *xenbus = XEN_BUS(qdev_get_parent_bus(DEVICE(xendev))); + va_list ap; + int rc; + + g_assert(xenbus->xsh); + + va_start(ap, fmt); + rc = xs_node_vscanf(xenbus->xsh, XBT_NULL, xendev->frontend_path, key, + NULL, fmt, ap); + va_end(ap); + + return rc; +} + +static void xen_device_frontend_set_state(XenDevice *xendev, + enum xenbus_state state) +{ + const char *type = object_get_typename(OBJECT(xendev)); + + if (xendev->frontend_state == state) { + return; + } + + trace_xen_device_frontend_state(type, xendev->name, + xs_strstate(state)); + + xendev->frontend_state = state; + xen_device_frontend_printf(xendev, "state", "%u", state); +} + +static void xen_device_frontend_changed(void *opaque) +{ + XenDevice *xendev = opaque; + XenDeviceClass *xendev_class = XEN_DEVICE_GET_CLASS(xendev); + const char *type = object_get_typename(OBJECT(xendev)); + enum xenbus_state state; + + trace_xen_device_frontend_changed(type, xendev->name); + + if (xen_device_frontend_scanf(xendev, "state", "%u", &state) != 1) { + state = XenbusStateUnknown; + } + + xen_device_frontend_set_state(xendev, state); + + if (xendev_class->frontend_changed) { + Error *local_err = NULL; + + xendev_class->frontend_changed(xendev, state, &local_err); + + if (local_err) { + error_reportf_err(local_err, "frontend change error: "); + } + } +} + +static void xen_device_frontend_create(XenDevice *xendev, Error **errp) +{ + XenBus *xenbus = XEN_BUS(qdev_get_parent_bus(DEVICE(xendev))); + struct xs_permissions perms[2]; + Error *local_err = NULL; + + xendev->frontend_path = xen_device_get_frontend_path(xendev); + + perms[0].id = xendev->frontend_id; + perms[0].perms = XS_PERM_NONE; + perms[1].id = xenbus->backend_id; + perms[1].perms = XS_PERM_READ | XS_PERM_WRITE; + + g_assert(xenbus->xsh); + + xs_node_create(xenbus->xsh, XBT_NULL, xendev->frontend_path, perms, + ARRAY_SIZE(perms), &local_err); + if (local_err) { + error_propagate_prepend(errp, local_err, + "failed to create frontend: "); + return; + } + + xendev->frontend_state_watch = + xen_bus_add_watch(xenbus, xendev->frontend_path, "state", + xen_device_frontend_changed, xendev, &local_err); + if (local_err) { + error_propagate_prepend(errp, local_err, + "failed to watch frontend state: "); + } +} + +static void xen_device_frontend_destroy(XenDevice *xendev) +{ + XenBus *xenbus = XEN_BUS(qdev_get_parent_bus(DEVICE(xendev))); + Error *local_err = NULL; + + if (xendev->frontend_state_watch) { + xen_bus_remove_watch(xenbus, xendev->frontend_state_watch, NULL); + xendev->frontend_state_watch = NULL; + } + + if (!xendev->frontend_path) { + return; + } + + g_assert(xenbus->xsh); + + xs_node_destroy(xenbus->xsh, XBT_NULL, xendev->frontend_path, + &local_err); + g_free(xendev->frontend_path); + xendev->frontend_path = NULL; + + if (local_err) { + error_report_err(local_err); + } +} + +void xen_device_set_max_grant_refs(XenDevice *xendev, unsigned int nr_refs, + Error **errp) +{ + if (xengnttab_set_max_grants(xendev->xgth, nr_refs)) { + error_setg_errno(errp, errno, "xengnttab_set_max_grants failed"); + } +} + +void *xen_device_map_grant_refs(XenDevice *xendev, uint32_t *refs, + unsigned int nr_refs, int prot, + Error **errp) +{ + void *map = xengnttab_map_domain_grant_refs(xendev->xgth, nr_refs, + xendev->frontend_id, refs, + prot); + + if (!map) { + error_setg_errno(errp, errno, + "xengnttab_map_domain_grant_refs failed"); + } + + return map; +} + +void xen_device_unmap_grant_refs(XenDevice *xendev, void *map, + unsigned int nr_refs, Error **errp) +{ + if (xengnttab_unmap(xendev->xgth, map, nr_refs)) { + error_setg_errno(errp, errno, "xengnttab_unmap failed"); + } +} + +static void compat_copy_grant_refs(XenDevice *xendev, bool to_domain, + XenDeviceGrantCopySegment segs[], + unsigned int nr_segs, Error **errp) +{ + uint32_t *refs = g_new(uint32_t, nr_segs); + int prot = to_domain ? PROT_WRITE : PROT_READ; + void *map; + unsigned int i; + + for (i = 0; i < nr_segs; i++) { + XenDeviceGrantCopySegment *seg = &segs[i]; + + refs[i] = to_domain ? seg->dest.foreign.ref : + seg->source.foreign.ref; + } + + map = xengnttab_map_domain_grant_refs(xendev->xgth, nr_segs, + xendev->frontend_id, refs, + prot); + if (!map) { + error_setg_errno(errp, errno, + "xengnttab_map_domain_grant_refs failed"); + goto done; + } + + for (i = 0; i < nr_segs; i++) { + XenDeviceGrantCopySegment *seg = &segs[i]; + void *page = map + (i * XC_PAGE_SIZE); + + if (to_domain) { + memcpy(page + seg->dest.foreign.offset, seg->source.virt, + seg->len); + } else { + memcpy(seg->dest.virt, page + seg->source.foreign.offset, + seg->len); + } + } + + if (xengnttab_unmap(xendev->xgth, map, nr_segs)) { + error_setg_errno(errp, errno, "xengnttab_unmap failed"); + } + +done: + g_free(refs); +} + +void xen_device_copy_grant_refs(XenDevice *xendev, bool to_domain, + XenDeviceGrantCopySegment segs[], + unsigned int nr_segs, Error **errp) +{ + xengnttab_grant_copy_segment_t *xengnttab_segs; + unsigned int i; + + if (!xendev->feature_grant_copy) { + compat_copy_grant_refs(xendev, to_domain, segs, nr_segs, errp); + return; + } + + xengnttab_segs = g_new0(xengnttab_grant_copy_segment_t, nr_segs); + + for (i = 0; i < nr_segs; i++) { + XenDeviceGrantCopySegment *seg = &segs[i]; + xengnttab_grant_copy_segment_t *xengnttab_seg = &xengnttab_segs[i]; + + if (to_domain) { + xengnttab_seg->flags = GNTCOPY_dest_gref; + xengnttab_seg->dest.foreign.domid = xendev->frontend_id; + xengnttab_seg->dest.foreign.ref = seg->dest.foreign.ref; + xengnttab_seg->dest.foreign.offset = seg->dest.foreign.offset; + xengnttab_seg->source.virt = seg->source.virt; + } else { + xengnttab_seg->flags = GNTCOPY_source_gref; + xengnttab_seg->source.foreign.domid = xendev->frontend_id; + xengnttab_seg->source.foreign.ref = seg->source.foreign.ref; + xengnttab_seg->source.foreign.offset = + seg->source.foreign.offset; + xengnttab_seg->dest.virt = seg->dest.virt; + } + + xengnttab_seg->len = seg->len; + } + + if (xengnttab_grant_copy(xendev->xgth, nr_segs, xengnttab_segs)) { + error_setg_errno(errp, errno, "xengnttab_grant_copy failed"); + goto done; + } + + for (i = 0; i < nr_segs; i++) { + xengnttab_grant_copy_segment_t *xengnttab_seg = &xengnttab_segs[i]; + + if (xengnttab_seg->status != GNTST_okay) { + error_setg(errp, "xengnttab_grant_copy seg[%u] failed", i); + break; + } + } + +done: + g_free(xengnttab_segs); +} + +struct XenEventChannel { + evtchn_port_t local_port; + XenEventHandler handler; + void *opaque; + Notifier notifier; +}; + +static void event_notify(Notifier *n, void *data) +{ + XenEventChannel *channel = container_of(n, XenEventChannel, notifier); + unsigned long port = (unsigned long)data; + + if (port == channel->local_port) { + channel->handler(channel->opaque); + } +} + +XenEventChannel *xen_device_bind_event_channel(XenDevice *xendev, + unsigned int port, + XenEventHandler handler, + void *opaque, Error **errp) +{ + XenEventChannel *channel = g_new0(XenEventChannel, 1); + xenevtchn_port_or_error_t local_port; + + local_port = xenevtchn_bind_interdomain(xendev->xeh, + xendev->frontend_id, + port); + if (local_port < 0) { + error_setg_errno(errp, errno, "xenevtchn_bind_interdomain failed"); + + g_free(channel); + return NULL; + } + + channel->local_port = local_port; + channel->handler = handler; + channel->opaque = opaque; + channel->notifier.notify = event_notify; + + notifier_list_add(&xendev->event_notifiers, &channel->notifier); + + return channel; +} + +void xen_device_notify_event_channel(XenDevice *xendev, + XenEventChannel *channel, + Error **errp) +{ + if (!channel) { + error_setg(errp, "bad channel"); + return; + } + + if (xenevtchn_notify(xendev->xeh, channel->local_port) < 0) { + error_setg_errno(errp, errno, "xenevtchn_notify failed"); + } +} + +void xen_device_unbind_event_channel(XenDevice *xendev, + XenEventChannel *channel, + Error **errp) +{ + if (!channel) { + error_setg(errp, "bad channel"); + return; + } + + notifier_remove(&channel->notifier); + + if (xenevtchn_unbind(xendev->xeh, channel->local_port) < 0) { + error_setg_errno(errp, errno, "xenevtchn_unbind failed"); + } + + g_free(channel); +} + +static void xen_device_unrealize(DeviceState *dev, Error **errp) +{ + XenDevice *xendev = XEN_DEVICE(dev); + XenDeviceClass *xendev_class = XEN_DEVICE_GET_CLASS(xendev); + const char *type = object_get_typename(OBJECT(xendev)); + + if (!xendev->name) { + return; + } + + trace_xen_device_unrealize(type, xendev->name); + + if (xendev->exit.notify) { + qemu_remove_exit_notifier(&xendev->exit); + xendev->exit.notify = NULL; + } + + if (xendev_class->unrealize) { + xendev_class->unrealize(xendev, errp); + } + + xen_device_frontend_destroy(xendev); + xen_device_backend_destroy(xendev); + + if (xendev->xeh) { + qemu_set_fd_handler(xenevtchn_fd(xendev->xeh), NULL, NULL, NULL); + xenevtchn_close(xendev->xeh); + xendev->xeh = NULL; + } + + if (xendev->xgth) { + xengnttab_close(xendev->xgth); + xendev->xgth = NULL; + } + + g_free(xendev->name); + xendev->name = NULL; +} + +static void xen_device_exit(Notifier *n, void *data) +{ + XenDevice *xendev = container_of(n, XenDevice, exit); + + xen_device_unrealize(DEVICE(xendev), &error_abort); +} + +static void xen_device_event(void *opaque) +{ + XenDevice *xendev = opaque; + unsigned long port = xenevtchn_pending(xendev->xeh); + + notifier_list_notify(&xendev->event_notifiers, (void *)port); + + xenevtchn_unmask(xendev->xeh, port); +} + +static void xen_device_realize(DeviceState *dev, Error **errp) +{ + XenDevice *xendev = XEN_DEVICE(dev); + XenDeviceClass *xendev_class = XEN_DEVICE_GET_CLASS(xendev); + XenBus *xenbus = XEN_BUS(qdev_get_parent_bus(DEVICE(xendev))); + const char *type = object_get_typename(OBJECT(xendev)); + Error *local_err = NULL; + + if (xendev->frontend_id == DOMID_INVALID) { + xendev->frontend_id = xen_domid; + } + + if (xendev->frontend_id >= DOMID_FIRST_RESERVED) { + error_setg(errp, "invalid frontend-id"); + goto unrealize; + } + + if (!xendev_class->get_name) { + error_setg(errp, "get_name method not implemented"); + goto unrealize; + } + + xendev->name = xendev_class->get_name(xendev, &local_err); + if (local_err) { + error_propagate_prepend(errp, local_err, + "failed to get device name: "); + goto unrealize; + } + + trace_xen_device_realize(type, xendev->name); + + xendev->xgth = xengnttab_open(NULL, 0); + if (!xendev->xgth) { + error_setg_errno(errp, errno, "failed xengnttab_open"); + goto unrealize; + } + + xendev->feature_grant_copy = + (xengnttab_grant_copy(xendev->xgth, 0, NULL) == 0); + + xendev->xeh = xenevtchn_open(NULL, 0); + if (!xendev->xeh) { + error_setg_errno(errp, errno, "failed xenevtchn_open"); + goto unrealize; + } + + notifier_list_init(&xendev->event_notifiers); + qemu_set_fd_handler(xenevtchn_fd(xendev->xeh), xen_device_event, NULL, + xendev); + + xen_device_backend_create(xendev, &local_err); + if (local_err) { + error_propagate(errp, local_err); + goto unrealize; + } + + xen_device_frontend_create(xendev, &local_err); + if (local_err) { + error_propagate(errp, local_err); + goto unrealize; + } + + if (xendev_class->realize) { + xendev_class->realize(xendev, &local_err); + if (local_err) { + error_propagate(errp, local_err); + goto unrealize; + } + } + + xen_device_backend_printf(xendev, "frontend", "%s", + xendev->frontend_path); + xen_device_backend_printf(xendev, "frontend-id", "%u", + xendev->frontend_id); + xen_device_backend_printf(xendev, "hotplug-status", "connected"); + + xen_device_backend_set_online(xendev, true); + xen_device_backend_set_state(xendev, XenbusStateInitWait); + + xen_device_frontend_printf(xendev, "backend", "%s", + xendev->backend_path); + xen_device_frontend_printf(xendev, "backend-id", "%u", + xenbus->backend_id); + + xen_device_frontend_set_state(xendev, XenbusStateInitialising); + + xendev->exit.notify = xen_device_exit; + qemu_add_exit_notifier(&xendev->exit); + return; + +unrealize: + xen_device_unrealize(dev, &error_abort); +} + +static Property xen_device_props[] = { + DEFINE_PROP_UINT16("frontend-id", XenDevice, frontend_id, + DOMID_INVALID), + DEFINE_PROP_END_OF_LIST() +}; + +static void xen_device_class_init(ObjectClass *class, void *data) +{ + DeviceClass *dev_class = DEVICE_CLASS(class); + + dev_class->realize = xen_device_realize; + dev_class->unrealize = xen_device_unrealize; + dev_class->props = xen_device_props; + dev_class->bus_type = TYPE_XEN_BUS; +} + +static const TypeInfo xen_device_type_info = { + .name = TYPE_XEN_DEVICE, + .parent = TYPE_DEVICE, + .instance_size = sizeof(XenDevice), + .abstract = true, + .class_size = sizeof(XenDeviceClass), + .class_init = xen_device_class_init, +}; + +typedef struct XenBridge { + SysBusDevice busdev; +} XenBridge; + +#define TYPE_XEN_BRIDGE "xen-bridge" + +static const TypeInfo xen_bridge_type_info = { + .name = TYPE_XEN_BRIDGE, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_size = sizeof(XenBridge), +}; + +static void xen_register_types(void) +{ + type_register_static(&xen_bridge_type_info); + type_register_static(&xen_bus_type_info); + type_register_static(&xen_device_type_info); +} + +type_init(xen_register_types) + +void xen_bus_init(void) +{ + DeviceState *dev = qdev_create(NULL, TYPE_XEN_BRIDGE); + BusState *bus = qbus_create(TYPE_XEN_BUS, dev, NULL); + + qdev_init_nofail(dev); + qbus_set_bus_hotplug_handler(bus, &error_abort); +} diff --git a/hw/xen/xen-common.c b/hw/xen/xen-common.c index 18a9045556..0e9e58f04d 100644 --- a/hw/xen/xen-common.c +++ b/hw/xen/xen-common.c @@ -10,7 +10,7 @@ #include "qemu/osdep.h" #include "qemu/error-report.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include "chardev/char.h" #include "sysemu/accel.h" #include "migration/misc.h" diff --git a/hw/xen/xen_backend.c b/hw/xen/xen-legacy-backend.c index 0bc6b1de60..36fd1e9b09 100644 --- a/hw/xen/xen_backend.c +++ b/hw/xen/xen-legacy-backend.c @@ -30,7 +30,7 @@ #include "hw/boards.h" #include "qemu/log.h" #include "qapi/error.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include "hw/xen/xen_pvdev.h" #include "monitor/qdev.h" @@ -42,49 +42,54 @@ BusState *xen_sysbus; /* ------------------------------------------------------------- */ /* public */ -struct xs_handle *xenstore = NULL; +struct xs_handle *xenstore; const char *xen_protocol; /* private */ static bool xen_feature_grant_copy; static int debug; -int xenstore_write_be_str(struct XenDevice *xendev, const char *node, const char *val) +int xenstore_write_be_str(struct XenLegacyDevice *xendev, const char *node, + const char *val) { return xenstore_write_str(xendev->be, node, val); } -int xenstore_write_be_int(struct XenDevice *xendev, const char *node, int ival) +int xenstore_write_be_int(struct XenLegacyDevice *xendev, const char *node, + int ival) { return xenstore_write_int(xendev->be, node, ival); } -int xenstore_write_be_int64(struct XenDevice *xendev, const char *node, int64_t ival) +int xenstore_write_be_int64(struct XenLegacyDevice *xendev, const char *node, + int64_t ival) { return xenstore_write_int64(xendev->be, node, ival); } -char *xenstore_read_be_str(struct XenDevice *xendev, const char *node) +char *xenstore_read_be_str(struct XenLegacyDevice *xendev, const char *node) { return xenstore_read_str(xendev->be, node); } -int xenstore_read_be_int(struct XenDevice *xendev, const char *node, int *ival) +int xenstore_read_be_int(struct XenLegacyDevice *xendev, const char *node, + int *ival) { return xenstore_read_int(xendev->be, node, ival); } -char *xenstore_read_fe_str(struct XenDevice *xendev, const char *node) +char *xenstore_read_fe_str(struct XenLegacyDevice *xendev, const char *node) { return xenstore_read_str(xendev->fe, node); } -int xenstore_read_fe_int(struct XenDevice *xendev, const char *node, int *ival) +int xenstore_read_fe_int(struct XenLegacyDevice *xendev, const char *node, + int *ival) { return xenstore_read_int(xendev->fe, node, ival); } -int xenstore_read_fe_uint64(struct XenDevice *xendev, const char *node, +int xenstore_read_fe_uint64(struct XenLegacyDevice *xendev, const char *node, uint64_t *uval) { return xenstore_read_uint64(xendev->fe, node, uval); @@ -92,7 +97,7 @@ int xenstore_read_fe_uint64(struct XenDevice *xendev, const char *node, /* ------------------------------------------------------------- */ -int xen_be_set_state(struct XenDevice *xendev, enum xenbus_state state) +int xen_be_set_state(struct XenLegacyDevice *xendev, enum xenbus_state state) { int rc; @@ -106,7 +111,7 @@ int xen_be_set_state(struct XenDevice *xendev, enum xenbus_state state) return 0; } -void xen_be_set_max_grant_refs(struct XenDevice *xendev, +void xen_be_set_max_grant_refs(struct XenLegacyDevice *xendev, unsigned int nr_refs) { assert(xendev->ops->flags & DEVOPS_FLAG_NEED_GNTDEV); @@ -117,7 +122,7 @@ void xen_be_set_max_grant_refs(struct XenDevice *xendev, } } -void *xen_be_map_grant_refs(struct XenDevice *xendev, uint32_t *refs, +void *xen_be_map_grant_refs(struct XenLegacyDevice *xendev, uint32_t *refs, unsigned int nr_refs, int prot) { void *ptr; @@ -135,7 +140,7 @@ void *xen_be_map_grant_refs(struct XenDevice *xendev, uint32_t *refs, return ptr; } -void xen_be_unmap_grant_refs(struct XenDevice *xendev, void *ptr, +void xen_be_unmap_grant_refs(struct XenLegacyDevice *xendev, void *ptr, unsigned int nr_refs) { assert(xendev->ops->flags & DEVOPS_FLAG_NEED_GNTDEV); @@ -146,7 +151,7 @@ void xen_be_unmap_grant_refs(struct XenDevice *xendev, void *ptr, } } -static int compat_copy_grant_refs(struct XenDevice *xendev, +static int compat_copy_grant_refs(struct XenLegacyDevice *xendev, bool to_domain, XenGrantCopySegment segs[], unsigned int nr_segs) @@ -195,7 +200,7 @@ static int compat_copy_grant_refs(struct XenDevice *xendev, return 0; } -int xen_be_copy_grant_refs(struct XenDevice *xendev, +int xen_be_copy_grant_refs(struct XenLegacyDevice *xendev, bool to_domain, XenGrantCopySegment segs[], unsigned int nr_segs) @@ -259,10 +264,11 @@ int xen_be_copy_grant_refs(struct XenDevice *xendev, /* * get xen backend device, allocate a new one if it doesn't exist. */ -static struct XenDevice *xen_be_get_xendev(const char *type, int dom, int dev, - struct XenDevOps *ops) +static struct XenLegacyDevice *xen_be_get_xendev(const char *type, int dom, + int dev, + struct XenDevOps *ops) { - struct XenDevice *xendev; + struct XenLegacyDevice *xendev; xendev = xen_pv_find_xendev(type, dom, dev); if (xendev) { @@ -314,7 +320,8 @@ static struct XenDevice *xen_be_get_xendev(const char *type, int dom, int dev, * Node specifies the changed field. node = NULL means * update all fields (used for initialization). */ -static void xen_be_backend_changed(struct XenDevice *xendev, const char *node) +static void xen_be_backend_changed(struct XenLegacyDevice *xendev, + const char *node) { if (node == NULL || strcmp(node, "online") == 0) { if (xenstore_read_be_int(xendev, "online", &xendev->online) == -1) { @@ -330,7 +337,8 @@ static void xen_be_backend_changed(struct XenDevice *xendev, const char *node) } } -static void xen_be_frontend_changed(struct XenDevice *xendev, const char *node) +static void xen_be_frontend_changed(struct XenLegacyDevice *xendev, + const char *node) { int fe_state; @@ -373,7 +381,7 @@ static void xen_be_frontend_changed(struct XenDevice *xendev, const char *node) * only affects the xendev->be_state variable as xenbus should * already be put into that state by xend. */ -static int xen_be_try_setup(struct XenDevice *xendev) +static int xen_be_try_setup(struct XenLegacyDevice *xendev) { char token[XEN_BUFSIZE]; int be_state; @@ -417,7 +425,7 @@ static int xen_be_try_setup(struct XenDevice *xendev) * * Goes to InitWait on success. */ -static int xen_be_try_init(struct XenDevice *xendev) +static int xen_be_try_init(struct XenLegacyDevice *xendev) { int rc = 0; @@ -446,7 +454,7 @@ static int xen_be_try_init(struct XenDevice *xendev) * * Goes to Connected on success. */ -static int xen_be_try_initialise(struct XenDevice *xendev) +static int xen_be_try_initialise(struct XenLegacyDevice *xendev) { int rc = 0; @@ -487,7 +495,7 @@ static int xen_be_try_initialise(struct XenDevice *xendev) * frontend being Connected. Note that this may be called more * than once since the backend state is not modified. */ -static void xen_be_try_connected(struct XenDevice *xendev) +static void xen_be_try_connected(struct XenLegacyDevice *xendev) { if (!xendev->ops->connected) { return; @@ -510,7 +518,8 @@ static void xen_be_try_connected(struct XenDevice *xendev) * * Goes to Closed when done. */ -static void xen_be_disconnect(struct XenDevice *xendev, enum xenbus_state state) +static void xen_be_disconnect(struct XenLegacyDevice *xendev, + enum xenbus_state state) { if (xendev->be_state != XenbusStateClosing && xendev->be_state != XenbusStateClosed && @@ -529,7 +538,7 @@ static void xen_be_disconnect(struct XenDevice *xendev, enum xenbus_state state) /* * Try to reset xendev, for reconnection by another frontend instance. */ -static int xen_be_try_reset(struct XenDevice *xendev) +static int xen_be_try_reset(struct XenLegacyDevice *xendev) { if (xendev->fe_state != XenbusStateInitialising) { return -1; @@ -543,7 +552,7 @@ static int xen_be_try_reset(struct XenDevice *xendev) /* * state change dispatcher function */ -void xen_be_check_state(struct XenDevice *xendev) +void xen_be_check_state(struct XenLegacyDevice *xendev) { int rc = 0; @@ -587,7 +596,7 @@ void xen_be_check_state(struct XenDevice *xendev) static int xenstore_scan(const char *type, int dom, struct XenDevOps *ops) { - struct XenDevice *xendev; + struct XenLegacyDevice *xendev; char path[XEN_BUFSIZE], token[XEN_BUFSIZE]; char **dev = NULL; unsigned int cdev, j; @@ -620,7 +629,7 @@ static int xenstore_scan(const char *type, int dom, struct XenDevOps *ops) void xenstore_update_be(char *watch, char *type, int dom, struct XenDevOps *ops) { - struct XenDevice *xendev; + struct XenLegacyDevice *xendev; char path[XEN_BUFSIZE], *bepath; unsigned int len, dev; @@ -628,9 +637,9 @@ void xenstore_update_be(char *watch, char *type, int dom, if (strncmp(path, watch, len) != 0) { return; } - if (sscanf(watch+len, "/%u/%255s", &dev, path) != 2) { + if (sscanf(watch + len, "/%u/%255s", &dev, path) != 2) { strcpy(path, ""); - if (sscanf(watch+len, "/%u", &dev) != 1) { + if (sscanf(watch + len, "/%u", &dev) != 1) { dev = -1; } } @@ -651,7 +660,7 @@ void xenstore_update_be(char *watch, char *type, int dom, } } -void xenstore_update_fe(char *watch, struct XenDevice *xendev) +void xenstore_update_fe(char *watch, struct XenLegacyDevice *xendev) { char *node; unsigned int len; @@ -744,7 +753,6 @@ void xen_be_register_common(void) xen_be_register("console", &xen_console_ops); xen_be_register("vkbd", &xen_kbdmouse_ops); - xen_be_register("qdisk", &xen_blkdev_ops); #ifdef CONFIG_VIRTFS xen_be_register("9pfs", &xen_9pfs_ops); #endif @@ -753,7 +761,7 @@ void xen_be_register_common(void) #endif } -int xen_be_bind_evtchn(struct XenDevice *xendev) +int xen_be_bind_evtchn(struct XenLegacyDevice *xendev) { if (xendev->local_port != -1) { return 0; @@ -789,7 +797,7 @@ static const TypeInfo xendev_type_info = { .name = TYPE_XENBACKEND, .parent = TYPE_XENSYSDEV, .class_init = xendev_class_init, - .instance_size = sizeof(struct XenDevice), + .instance_size = sizeof(struct XenLegacyDevice), }; static void xen_sysbus_class_init(ObjectClass *klass, void *data) diff --git a/hw/xen/xen_devconfig.c b/hw/xen/xen_devconfig.c index 3500d88a3e..315dbc9c51 100644 --- a/hw/xen/xen_devconfig.c +++ b/hw/xen/xen_devconfig.c @@ -1,5 +1,5 @@ #include "qemu/osdep.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include "qemu/option.h" #include "sysemu/blockdev.h" diff --git a/hw/xen/xen_pt.c b/hw/xen/xen_pt.c index f1f3a3727c..5539d56c3a 100644 --- a/hw/xen/xen_pt.c +++ b/hw/xen/xen_pt.c @@ -59,7 +59,7 @@ #include "hw/pci/pci.h" #include "hw/xen/xen.h" #include "hw/i386/pc.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include "xen_pt.h" #include "qemu/range.h" #include "exec/address-spaces.h" @@ -847,6 +847,12 @@ static void xen_pt_realize(PCIDevice *d, Error **errp) } machine_irq = s->real_device.irq; + if (machine_irq == 0) { + XEN_PT_LOG(d, "machine irq is 0\n"); + cmd |= PCI_COMMAND_INTX_DISABLE; + goto out; + } + rc = xc_physdev_map_pirq(xen_xc, xen_domid, machine_irq, &pirq); if (rc < 0) { error_setg_errno(errp, errno, "Mapping machine irq %u to" diff --git a/hw/xen/xen_pt_config_init.c b/hw/xen/xen_pt_config_init.c index 47f9010c75..31ec5add1d 100644 --- a/hw/xen/xen_pt_config_init.c +++ b/hw/xen/xen_pt_config_init.c @@ -15,7 +15,7 @@ #include "qemu/osdep.h" #include "qapi/error.h" #include "qemu/timer.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include "xen_pt.h" #define XEN_PT_MERGE_VALUE(value, data, val_mask) \ @@ -300,7 +300,9 @@ static int xen_pt_irqpin_reg_init(XenPCIPassthroughState *s, XenPTRegInfo *reg, uint32_t real_offset, uint32_t *data) { - *data = xen_pt_pci_read_intx(s); + if (s->real_device.irq) { + *data = xen_pt_pci_read_intx(s); + } return 0; } diff --git a/hw/xen/xen_pt_graphics.c b/hw/xen/xen_pt_graphics.c index 135c8df1e7..b69732729b 100644 --- a/hw/xen/xen_pt_graphics.c +++ b/hw/xen/xen_pt_graphics.c @@ -5,7 +5,7 @@ #include "qapi/error.h" #include "xen_pt.h" #include "xen-host-pci-device.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" static unsigned long igd_guest_opregion; static unsigned long igd_host_opregion; @@ -185,8 +185,19 @@ void xen_pt_setup_vga(XenPCIPassthroughState *s, XenHostPCIDevice *dev, return; } + if (bios_size < sizeof(struct rom_header)) { + error_setg(errp, "VGA: VBIOS image corrupt (too small)"); + return; + } + /* Currently we fixed this address as a primary. */ rom = (struct rom_header *)bios; + + if (rom->pcioffset + sizeof(struct pci_data) > bios_size) { + error_setg(errp, "VGA: VBIOS image corrupt (bad pcioffset field)"); + return; + } + pd = (void *)(bios + (unsigned char)rom->pcioffset); /* We may need to fixup Device Identification. */ @@ -194,6 +205,11 @@ void xen_pt_setup_vga(XenPCIPassthroughState *s, XenHostPCIDevice *dev, pd->device = s->real_device.device_id; len = rom->size * 512; + if (len > bios_size) { + error_setg(errp, "VGA: VBIOS image corrupt (bad size field)"); + return; + } + /* Then adjust the bios checksum */ for (c = (char *)bios; c < ((char *)bios + len); c++) { checksum += *c; diff --git a/hw/xen/xen_pt_msi.c b/hw/xen/xen_pt_msi.c index cc514f9157..fb4b887b92 100644 --- a/hw/xen/xen_pt_msi.c +++ b/hw/xen/xen_pt_msi.c @@ -11,7 +11,7 @@ #include "qemu/osdep.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include "xen_pt.h" #include "hw/i386/apic-msidef.h" diff --git a/hw/xen/xen_pvdev.c b/hw/xen/xen_pvdev.c index f026556f62..6ef09cbf9d 100644 --- a/hw/xen/xen_pvdev.c +++ b/hw/xen/xen_pvdev.c @@ -20,7 +20,7 @@ #include "qemu/osdep.h" #include "qemu/log.h" #include "hw/qdev-core.h" -#include "hw/xen/xen_backend.h" +#include "hw/xen/xen-legacy-backend.h" #include "hw/xen/xen_pvdev.h" /* private */ @@ -34,7 +34,7 @@ struct xs_dirs { static QTAILQ_HEAD(, xs_dirs) xs_cleanup = QTAILQ_HEAD_INITIALIZER(xs_cleanup); -static QTAILQ_HEAD(, XenDevice) xendevs = +static QTAILQ_HEAD(, XenLegacyDevice) xendevs = QTAILQ_HEAD_INITIALIZER(xendevs); /* ------------------------------------------------------------- */ @@ -195,7 +195,7 @@ const char *xenbus_strstate(enum xenbus_state state) * 2 == noisy debug messages (logfile only). * 3 == will flood your log (logfile only). */ -void xen_pv_printf(struct XenDevice *xendev, int msg_level, +void xen_pv_printf(struct XenLegacyDevice *xendev, int msg_level, const char *fmt, ...) { va_list args; @@ -230,7 +230,7 @@ void xen_pv_printf(struct XenDevice *xendev, int msg_level, void xen_pv_evtchn_event(void *opaque) { - struct XenDevice *xendev = opaque; + struct XenLegacyDevice *xendev = opaque; evtchn_port_t port; port = xenevtchn_pending(xendev->evtchndev); @@ -247,7 +247,7 @@ void xen_pv_evtchn_event(void *opaque) } } -void xen_pv_unbind_evtchn(struct XenDevice *xendev) +void xen_pv_unbind_evtchn(struct XenLegacyDevice *xendev) { if (xendev->local_port == -1) { return; @@ -258,16 +258,16 @@ void xen_pv_unbind_evtchn(struct XenDevice *xendev) xendev->local_port = -1; } -int xen_pv_send_notify(struct XenDevice *xendev) +int xen_pv_send_notify(struct XenLegacyDevice *xendev) { return xenevtchn_notify(xendev->evtchndev, xendev->local_port); } /* ------------------------------------------------------------- */ -struct XenDevice *xen_pv_find_xendev(const char *type, int dom, int dev) +struct XenLegacyDevice *xen_pv_find_xendev(const char *type, int dom, int dev) { - struct XenDevice *xendev; + struct XenLegacyDevice *xendev; QTAILQ_FOREACH(xendev, &xendevs, next) { if (xendev->dom != dom) { @@ -287,7 +287,7 @@ struct XenDevice *xen_pv_find_xendev(const char *type, int dom, int dev) /* * release xen backend device. */ -void xen_pv_del_xendev(struct XenDevice *xendev) +void xen_pv_del_xendev(struct XenLegacyDevice *xendev) { if (xendev->ops->free) { xendev->ops->free(xendev); @@ -312,7 +312,7 @@ void xen_pv_del_xendev(struct XenDevice *xendev) qdev_unplug(&xendev->qdev, NULL); } -void xen_pv_insert_xendev(struct XenDevice *xendev) +void xen_pv_insert_xendev(struct XenLegacyDevice *xendev) { QTAILQ_INSERT_TAIL(&xendevs, xendev, next); } diff --git a/hw/xenpv/Makefile.objs b/hw/xenpv/Makefile.objs index bbf5873fd1..8bfa4586ab 100644 --- a/hw/xenpv/Makefile.objs +++ b/hw/xenpv/Makefile.objs @@ -1,4 +1,2 @@ # Xen PV machine support obj-$(CONFIG_XEN) += xen_machine_pv.o -# Xen PV machine builder support -obj-$(CONFIG_XEN_PV_DOMAIN_BUILD) += xen_domainbuild.o diff --git a/hw/xenpv/xen_domainbuild.c b/hw/xenpv/xen_domainbuild.c deleted file mode 100644 index 2859280a6a..0000000000 --- a/hw/xenpv/xen_domainbuild.c +++ /dev/null @@ -1,299 +0,0 @@ -#include "qemu/osdep.h" -#include "qemu/units.h" -#include "hw/xen/xen_backend.h" -#include "xen_domainbuild.h" -#include "qemu/timer.h" -#include "qemu/log.h" - -#include <xenguest.h> - -static int xenstore_domain_mkdir(char *path) -{ - struct xs_permissions perms_ro[] = {{ - .id = 0, /* set owner: dom0 */ - },{ - .id = xen_domid, - .perms = XS_PERM_READ, - }}; - struct xs_permissions perms_rw[] = {{ - .id = 0, /* set owner: dom0 */ - },{ - .id = xen_domid, - .perms = XS_PERM_READ | XS_PERM_WRITE, - }}; - const char *writable[] = { "device", "control", "error", NULL }; - char subpath[256]; - int i; - - if (!xs_mkdir(xenstore, 0, path)) { - fprintf(stderr, "%s: xs_mkdir %s: failed\n", __func__, path); - return -1; - } - if (!xs_set_permissions(xenstore, 0, path, perms_ro, 2)) { - fprintf(stderr, "%s: xs_set_permissions failed\n", __func__); - return -1; - } - - for (i = 0; writable[i]; i++) { - snprintf(subpath, sizeof(subpath), "%s/%s", path, writable[i]); - if (!xs_mkdir(xenstore, 0, subpath)) { - fprintf(stderr, "%s: xs_mkdir %s: failed\n", __func__, subpath); - return -1; - } - if (!xs_set_permissions(xenstore, 0, subpath, perms_rw, 2)) { - fprintf(stderr, "%s: xs_set_permissions failed\n", __func__); - return -1; - } - } - return 0; -} - -int xenstore_domain_init1(const char *kernel, const char *ramdisk, - const char *cmdline) -{ - char *dom, uuid_string[42], vm[256], path[256]; - int i; - - qemu_uuid_unparse(&qemu_uuid, uuid_string); - dom = xs_get_domain_path(xenstore, xen_domid); - snprintf(vm, sizeof(vm), "/vm/%s", uuid_string); - - xenstore_domain_mkdir(dom); - - xenstore_write_str(vm, "image/ostype", "linux"); - if (kernel) - xenstore_write_str(vm, "image/kernel", kernel); - if (ramdisk) - xenstore_write_str(vm, "image/ramdisk", ramdisk); - if (cmdline) - xenstore_write_str(vm, "image/cmdline", cmdline); - - /* name + id */ - xenstore_write_str(vm, "name", qemu_name ? qemu_name : "no-name"); - xenstore_write_str(vm, "uuid", uuid_string); - xenstore_write_str(dom, "name", qemu_name ? qemu_name : "no-name"); - xenstore_write_int(dom, "domid", xen_domid); - xenstore_write_str(dom, "vm", vm); - - /* memory */ - xenstore_write_int(dom, "memory/target", ram_size / KiB); - xenstore_write_int(vm, "memory", ram_size / MiB); - xenstore_write_int(vm, "maxmem", ram_size / MiB); - - /* cpus */ - for (i = 0; i < smp_cpus; i++) { - snprintf(path, sizeof(path), "cpu/%d/availability",i); - xenstore_write_str(dom, path, "online"); - } - xenstore_write_int(vm, "vcpu_avail", smp_cpus); - xenstore_write_int(vm, "vcpus", smp_cpus); - - /* vnc password */ - xenstore_write_str(vm, "vncpassword", "" /* FIXME */); - - free(dom); - return 0; -} - -int xenstore_domain_init2(int xenstore_port, int xenstore_mfn, - int console_port, int console_mfn) -{ - char *dom; - - dom = xs_get_domain_path(xenstore, xen_domid); - - /* signal new domain */ - xs_introduce_domain(xenstore, - xen_domid, - xenstore_mfn, - xenstore_port); - - /* xenstore */ - xenstore_write_int(dom, "store/ring-ref", xenstore_mfn); - xenstore_write_int(dom, "store/port", xenstore_port); - - /* console */ - xenstore_write_str(dom, "console/type", "ioemu"); - xenstore_write_int(dom, "console/limit", 128 * KiB); - xenstore_write_int(dom, "console/ring-ref", console_mfn); - xenstore_write_int(dom, "console/port", console_port); - xen_config_dev_console(0); - - free(dom); - return 0; -} - -/* ------------------------------------------------------------- */ - -static QEMUTimer *xen_poll; - -/* check domain state once per second */ -static void xen_domain_poll(void *opaque) -{ - struct xc_dominfo info; - int rc; - - rc = xc_domain_getinfo(xen_xc, xen_domid, 1, &info); - if ((rc != 1) || (info.domid != xen_domid)) { - qemu_log("xen: domain %d is gone\n", xen_domid); - goto quit; - } - if (info.dying) { - qemu_log("xen: domain %d is dying (%s%s)\n", xen_domid, - info.crashed ? "crashed" : "", - info.shutdown ? "shutdown" : ""); - goto quit; - } - - timer_mod(xen_poll, qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + 1000); - return; - -quit: - qemu_system_shutdown_request(SHUTDOWN_CAUSE_GUEST_SHUTDOWN); -} - -static int xen_domain_watcher(void) -{ - int qemu_running = 1; - int fd[2], i, n, rc; - char byte; - - if (pipe(fd) != 0) { - qemu_log("%s: Huh? pipe error: %s\n", __func__, strerror(errno)); - return -1; - } - if (fork() != 0) - return 0; /* not child */ - - /* close all file handles, except stdio/out/err, - * our watch pipe and the xen interface handle */ - n = getdtablesize(); - for (i = 3; i < n; i++) { - if (i == fd[0]) - continue; - close(i); - } - - /* - * Reopen xc interface, since the original is unsafe after fork - * and was closed above. - */ - xen_xc = xc_interface_open(0, 0, 0); - - /* ignore term signals */ - signal(SIGINT, SIG_IGN); - signal(SIGTERM, SIG_IGN); - - /* wait for qemu exiting */ - while (qemu_running) { - rc = read(fd[0], &byte, 1); - switch (rc) { - case -1: - if (errno == EINTR) - continue; - qemu_log("%s: Huh? read error: %s\n", __func__, strerror(errno)); - qemu_running = 0; - break; - case 0: - /* EOF -> qemu exited */ - qemu_running = 0; - break; - default: - qemu_log("%s: Huh? data on the watch pipe?\n", __func__); - break; - } - } - - /* cleanup */ - qemu_log("%s: destroy domain %d\n", __func__, xen_domid); - xc_domain_destroy(xen_xc, xen_domid); - _exit(0); -} - -/* normal cleanup */ -static void xen_domain_cleanup(void) -{ - char *dom; - - dom = xs_get_domain_path(xenstore, xen_domid); - if (dom) { - xs_rm(xenstore, 0, dom); - free(dom); - } - xs_release_domain(xenstore, xen_domid); -} - -int xen_domain_build_pv(const char *kernel, const char *ramdisk, - const char *cmdline) -{ - uint32_t ssidref = 0; - uint32_t flags = 0; - xen_domain_handle_t uuid; - unsigned int xenstore_port = 0, console_port = 0; - unsigned long xenstore_mfn = 0, console_mfn = 0; - int rc; - - memcpy(uuid, &qemu_uuid, sizeof(uuid)); - rc = xen_domain_create(xen_xc, ssidref, uuid, flags, &xen_domid); - if (rc < 0) { - fprintf(stderr, "xen: xc_domain_create() failed\n"); - goto err; - } - qemu_log("xen: created domain %d\n", xen_domid); - atexit(xen_domain_cleanup); - if (xen_domain_watcher() == -1) { - goto err; - } - - xenstore_domain_init1(kernel, ramdisk, cmdline); - - rc = xc_domain_max_vcpus(xen_xc, xen_domid, smp_cpus); - if (rc < 0) { - fprintf(stderr, "xen: xc_domain_max_vcpus() failed\n"); - goto err; - } - -#if 0 - rc = xc_domain_setcpuweight(xen_xc, xen_domid, 256); - if (rc < 0) { - fprintf(stderr, "xen: xc_domain_setcpuweight() failed\n"); - goto err; - } -#endif - - rc = xc_domain_setmaxmem(xen_xc, xen_domid, ram_size / KiB); - if (rc < 0) { - fprintf(stderr, "xen: xc_domain_setmaxmem() failed\n"); - goto err; - } - - xenstore_port = xc_evtchn_alloc_unbound(xen_xc, xen_domid, 0); - console_port = xc_evtchn_alloc_unbound(xen_xc, xen_domid, 0); - - rc = xc_linux_build(xen_xc, xen_domid, ram_size / MiB, - kernel, ramdisk, cmdline, - 0, flags, - xenstore_port, &xenstore_mfn, - console_port, &console_mfn); - if (rc < 0) { - fprintf(stderr, "xen: xc_linux_build() failed\n"); - goto err; - } - - xenstore_domain_init2(xenstore_port, xenstore_mfn, - console_port, console_mfn); - - qemu_log("xen: unpausing domain %d\n", xen_domid); - rc = xc_domain_unpause(xen_xc, xen_domid); - if (rc < 0) { - fprintf(stderr, "xen: xc_domain_unpause() failed\n"); - goto err; - } - - xen_poll = timer_new_ms(QEMU_CLOCK_REALTIME, xen_domain_poll, NULL); - timer_mod(xen_poll, qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + 1000); - return 0; - -err: - return -1; -} diff --git a/hw/xenpv/xen_domainbuild.h b/hw/xenpv/xen_domainbuild.h deleted file mode 100644 index 652d9b410f..0000000000 --- a/hw/xenpv/xen_domainbuild.h +++ /dev/null @@ -1,13 +0,0 @@ -#ifndef QEMU_HW_XEN_DOMAINBUILD_H -#define QEMU_HW_XEN_DOMAINBUILD_H - -#include "hw/xen/xen_common.h" - -int xenstore_domain_init1(const char *kernel, const char *ramdisk, - const char *cmdline); -int xenstore_domain_init2(int xenstore_port, int xenstore_mfn, - int console_port, int console_mfn); -int xen_domain_build_pv(const char *kernel, const char *ramdisk, - const char *cmdline); - -#endif /* QEMU_HW_XEN_DOMAINBUILD_H */ diff --git a/hw/xenpv/xen_machine_pv.c b/hw/xenpv/xen_machine_pv.c index 44d67b87c4..dcaf2a01a3 100644 --- a/hw/xenpv/xen_machine_pv.c +++ b/hw/xenpv/xen_machine_pv.c @@ -26,8 +26,8 @@ #include "qemu/error-report.h" #include "hw/hw.h" #include "hw/boards.h" -#include "hw/xen/xen_backend.h" -#include "xen_domainbuild.h" +#include "hw/xen/xen-legacy-backend.h" +#include "hw/xen/xen-bus.h" #include "sysemu/block-backend.h" static void xen_init_pv(MachineState *machine) @@ -43,21 +43,8 @@ static void xen_init_pv(MachineState *machine) switch (xen_mode) { case XEN_ATTACH: - /* nothing to do, xend handles everything */ + /* nothing to do, libxl handles everything */ break; -#ifdef CONFIG_XEN_PV_DOMAIN_BUILD - case XEN_CREATE: { - const char *kernel_filename = machine->kernel_filename; - const char *kernel_cmdline = machine->kernel_cmdline; - const char *initrd_filename = machine->initrd_filename; - if (xen_domain_build_pv(kernel_filename, initrd_filename, - kernel_cmdline) < 0) { - error_report("xen pv domain creation failed"); - exit(1); - } - break; - } -#endif case XEN_EMULATE: error_report("xen emulation not implemented (yet)"); exit(1); @@ -93,6 +80,8 @@ static void xen_init_pv(MachineState *machine) xen_config_dev_nic(nd_table + i); } + xen_bus_init(); + /* config cleanup hook */ atexit(xen_config_cleanup); } diff --git a/include/block/nbd.h b/include/block/nbd.h index 65402d3396..1971b55789 100644 --- a/include/block/nbd.h +++ b/include/block/nbd.h @@ -295,9 +295,10 @@ typedef struct NBDExport NBDExport; typedef struct NBDClient NBDClient; NBDExport *nbd_export_new(BlockDriverState *bs, off_t dev_offset, off_t size, - uint16_t nbdflags, void (*close)(NBDExport *), - bool writethrough, BlockBackend *on_eject_blk, - Error **errp); + const char *name, const char *description, + const char *bitmap, uint16_t nbdflags, + void (*close)(NBDExport *), bool writethrough, + BlockBackend *on_eject_blk, Error **errp); void nbd_export_close(NBDExport *exp); void nbd_export_remove(NBDExport *exp, NbdServerRemoveMode mode, Error **errp); void nbd_export_get(NBDExport *exp); @@ -306,8 +307,6 @@ void nbd_export_put(NBDExport *exp); BlockBackend *nbd_export_get_blockdev(NBDExport *exp); NBDExport *nbd_export_find(const char *name); -void nbd_export_set_name(NBDExport *exp, const char *name); -void nbd_export_set_description(NBDExport *exp, const char *description); void nbd_export_close_all(void); void nbd_client_new(QIOChannelSocket *sioc, @@ -320,9 +319,6 @@ void nbd_client_put(NBDClient *client); void nbd_server_start(SocketAddress *addr, const char *tls_creds, Error **errp); -void nbd_export_bitmap(NBDExport *exp, const char *bitmap, - const char *bitmap_export_name, Error **errp); - /* nbd_read * Reads @size bytes from @ioc. Returns 0 on success. */ diff --git a/include/hw/xen/xen-backend.h b/include/hw/xen/xen-backend.h new file mode 100644 index 0000000000..010d712638 --- /dev/null +++ b/include/hw/xen/xen-backend.h @@ -0,0 +1,39 @@ +/* + * Copyright (c) 2018 Citrix Systems Inc. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#ifndef HW_XEN_BACKEND_H +#define HW_XEN_BACKEND_H + +#include "hw/xen/xen-bus.h" + +typedef struct XenBackendInstance XenBackendInstance; + +typedef void (*XenBackendDeviceCreate)(XenBackendInstance *backend, + QDict *opts, Error **errp); +typedef void (*XenBackendDeviceDestroy)(XenBackendInstance *backend, + Error **errp); + +typedef struct XenBackendInfo { + const char *type; + XenBackendDeviceCreate create; + XenBackendDeviceDestroy destroy; +} XenBackendInfo; + +XenBus *xen_backend_get_bus(XenBackendInstance *backend); +const char *xen_backend_get_name(XenBackendInstance *backend); + +void xen_backend_set_device(XenBackendInstance *backend, + XenDevice *xendevice); +XenDevice *xen_backend_get_device(XenBackendInstance *backend); + +void xen_backend_register(const XenBackendInfo *info); + +void xen_backend_device_create(XenBus *xenbus, const char *type, + const char *name, QDict *opts, Error **errp); +bool xen_backend_try_device_destroy(XenDevice *xendev, Error **errp); + +#endif /* HW_XEN_BACKEND_H */ diff --git a/include/hw/xen/xen-block.h b/include/hw/xen/xen-block.h new file mode 100644 index 0000000000..11d351b4b3 --- /dev/null +++ b/include/hw/xen/xen-block.h @@ -0,0 +1,94 @@ +/* + * Copyright (c) 2018 Citrix Systems Inc. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#ifndef HW_XEN_BLOCK_H +#define HW_XEN_BLOCK_H + +#include "hw/xen/xen-bus.h" +#include "hw/block/block.h" +#include "hw/block/dataplane/xen-block.h" +#include "sysemu/iothread.h" + +typedef enum XenBlockVdevType { + XEN_BLOCK_VDEV_TYPE_INVALID, + XEN_BLOCK_VDEV_TYPE_DP, + XEN_BLOCK_VDEV_TYPE_XVD, + XEN_BLOCK_VDEV_TYPE_HD, + XEN_BLOCK_VDEV_TYPE_SD, + XEN_BLOCK_VDEV_TYPE__MAX +} XenBlockVdevType; + +typedef struct XenBlockVdev { + XenBlockVdevType type; + unsigned long disk; + unsigned long partition; + unsigned long number; +} XenBlockVdev; + + +typedef struct XenBlockProperties { + XenBlockVdev vdev; + BlockConf conf; + unsigned int max_ring_page_order; + IOThread *iothread; +} XenBlockProperties; + +typedef struct XenBlockDrive { + char *id; + char *node_name; +} XenBlockDrive; + +typedef struct XenBlockIOThread { + char *id; +} XenBlockIOThread; + +typedef struct XenBlockDevice { + XenDevice xendev; + XenBlockProperties props; + const char *device_type; + unsigned int info; + XenBlockDataPlane *dataplane; + XenBlockDrive *drive; + XenBlockIOThread *iothread; +} XenBlockDevice; + +typedef void (*XenBlockDeviceRealize)(XenBlockDevice *blockdev, Error **errp); +typedef void (*XenBlockDeviceUnrealize)(XenBlockDevice *blockdev, Error **errp); + +typedef struct XenBlockDeviceClass { + /*< private >*/ + XenDeviceClass parent_class; + /*< public >*/ + XenBlockDeviceRealize realize; + XenBlockDeviceUnrealize unrealize; +} XenBlockDeviceClass; + +#define TYPE_XEN_BLOCK_DEVICE "xen-block" +#define XEN_BLOCK_DEVICE(obj) \ + OBJECT_CHECK(XenBlockDevice, (obj), TYPE_XEN_BLOCK_DEVICE) +#define XEN_BLOCK_DEVICE_CLASS(class) \ + OBJECT_CLASS_CHECK(XenBlockDeviceClass, (class), TYPE_XEN_BLOCK_DEVICE) +#define XEN_BLOCK_DEVICE_GET_CLASS(obj) \ + OBJECT_GET_CLASS(XenBlockDeviceClass, (obj), TYPE_XEN_BLOCK_DEVICE) + +typedef struct XenDiskDevice { + XenBlockDevice blockdev; +} XenDiskDevice; + +#define TYPE_XEN_DISK_DEVICE "xen-disk" +#define XEN_DISK_DEVICE(obj) \ + OBJECT_CHECK(XenDiskDevice, (obj), TYPE_XEN_DISK_DEVICE) + +typedef struct XenCDRomDevice { + XenBlockDevice blockdev; +} XenCDRomDevice; + +#define TYPE_XEN_CDROM_DEVICE "xen-cdrom" +#define XEN_CDROM_DEVICE(obj) \ + OBJECT_CHECK(XenCDRomDevice, (obj), TYPE_XEN_CDROM_DEVICE) + +#endif /* HW_XEN_BLOCK_H */ diff --git a/include/hw/xen/xen-bus-helper.h b/include/hw/xen/xen-bus-helper.h new file mode 100644 index 0000000000..4c0f747445 --- /dev/null +++ b/include/hw/xen/xen-bus-helper.h @@ -0,0 +1,45 @@ +/* + * Copyright (c) 2018 Citrix Systems Inc. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#ifndef HW_XEN_BUS_HELPER_H +#define HW_XEN_BUS_HELPER_H + +#include "hw/xen/xen_common.h" + +const char *xs_strstate(enum xenbus_state state); + +void xs_node_create(struct xs_handle *xsh, xs_transaction_t tid, + const char *node, struct xs_permissions perms[], + unsigned int nr_perms, Error **errp); +void xs_node_destroy(struct xs_handle *xsh, xs_transaction_t tid, + const char *node, Error **errp); + +/* Write to node/key unless node is empty, in which case write to key */ +void xs_node_vprintf(struct xs_handle *xsh, xs_transaction_t tid, + const char *node, const char *key, Error **errp, + const char *fmt, va_list ap) + GCC_FMT_ATTR(6, 0); +void xs_node_printf(struct xs_handle *xsh, xs_transaction_t tid, + const char *node, const char *key, Error **errp, + const char *fmt, ...) + GCC_FMT_ATTR(6, 7); + +/* Read from node/key unless node is empty, in which case read from key */ +int xs_node_vscanf(struct xs_handle *xsh, xs_transaction_t tid, + const char *node, const char *key, Error **errp, + const char *fmt, va_list ap); +int xs_node_scanf(struct xs_handle *xsh, xs_transaction_t tid, + const char *node, const char *key, Error **errp, + const char *fmt, ...); + +/* Watch node/key unless node is empty, in which case watch key */ +void xs_node_watch(struct xs_handle *xsh, const char *node, const char *key, + char *token, Error **errp); +void xs_node_unwatch(struct xs_handle *xsh, const char *node, const char *key, + const char *token, Error **errp); + +#endif /* HW_XEN_BUS_HELPER_H */ diff --git a/include/hw/xen/xen-bus.h b/include/hw/xen/xen-bus.h new file mode 100644 index 0000000000..3183f10e3c --- /dev/null +++ b/include/hw/xen/xen-bus.h @@ -0,0 +1,137 @@ +/* + * Copyright (c) 2018 Citrix Systems Inc. + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#ifndef HW_XEN_BUS_H +#define HW_XEN_BUS_H + +#include "hw/xen/xen_common.h" +#include "hw/sysbus.h" +#include "qemu/notify.h" + +typedef void (*XenWatchHandler)(void *opaque); + +typedef struct XenWatch XenWatch; + +typedef struct XenDevice { + DeviceState qdev; + domid_t frontend_id; + char *name; + char *backend_path, *frontend_path; + enum xenbus_state backend_state, frontend_state; + Notifier exit; + XenWatch *backend_state_watch, *frontend_state_watch; + bool backend_online; + XenWatch *backend_online_watch; + xengnttab_handle *xgth; + bool feature_grant_copy; + xenevtchn_handle *xeh; + NotifierList event_notifiers; +} XenDevice; + +typedef char *(*XenDeviceGetName)(XenDevice *xendev, Error **errp); +typedef void (*XenDeviceRealize)(XenDevice *xendev, Error **errp); +typedef void (*XenDeviceFrontendChanged)(XenDevice *xendev, + enum xenbus_state frontend_state, + Error **errp); +typedef void (*XenDeviceUnrealize)(XenDevice *xendev, Error **errp); + +typedef struct XenDeviceClass { + /*< private >*/ + DeviceClass parent_class; + /*< public >*/ + const char *backend; + const char *device; + XenDeviceGetName get_name; + XenDeviceRealize realize; + XenDeviceFrontendChanged frontend_changed; + XenDeviceUnrealize unrealize; +} XenDeviceClass; + +#define TYPE_XEN_DEVICE "xen-device" +#define XEN_DEVICE(obj) \ + OBJECT_CHECK(XenDevice, (obj), TYPE_XEN_DEVICE) +#define XEN_DEVICE_CLASS(class) \ + OBJECT_CLASS_CHECK(XenDeviceClass, (class), TYPE_XEN_DEVICE) +#define XEN_DEVICE_GET_CLASS(obj) \ + OBJECT_GET_CLASS(XenDeviceClass, (obj), TYPE_XEN_DEVICE) + +typedef struct XenBus { + BusState qbus; + domid_t backend_id; + struct xs_handle *xsh; + NotifierList watch_notifiers; + XenWatch *backend_watch; +} XenBus; + +typedef struct XenBusClass { + /*< private >*/ + BusClass parent_class; +} XenBusClass; + +#define TYPE_XEN_BUS "xen-bus" +#define XEN_BUS(obj) \ + OBJECT_CHECK(XenBus, (obj), TYPE_XEN_BUS) +#define XEN_BUS_CLASS(class) \ + OBJECT_CLASS_CHECK(XenBusClass, (class), TYPE_XEN_BUS) +#define XEN_BUS_GET_CLASS(obj) \ + OBJECT_GET_CLASS(XenBusClass, (obj), TYPE_XEN_BUS) + +void xen_bus_init(void); + +void xen_device_backend_set_state(XenDevice *xendev, + enum xenbus_state state); +enum xenbus_state xen_device_backend_get_state(XenDevice *xendev); + +void xen_device_backend_printf(XenDevice *xendev, const char *key, + const char *fmt, ...) + GCC_FMT_ATTR(3, 4); +void xen_device_frontend_printf(XenDevice *xendev, const char *key, + const char *fmt, ...) + GCC_FMT_ATTR(3, 4); + +int xen_device_frontend_scanf(XenDevice *xendev, const char *key, + const char *fmt, ...); + +void xen_device_set_max_grant_refs(XenDevice *xendev, unsigned int nr_refs, + Error **errp); +void *xen_device_map_grant_refs(XenDevice *xendev, uint32_t *refs, + unsigned int nr_refs, int prot, + Error **errp); +void xen_device_unmap_grant_refs(XenDevice *xendev, void *map, + unsigned int nr_refs, Error **errp); + +typedef struct XenDeviceGrantCopySegment { + union { + void *virt; + struct { + uint32_t ref; + off_t offset; + } foreign; + } source, dest; + size_t len; +} XenDeviceGrantCopySegment; + +void xen_device_copy_grant_refs(XenDevice *xendev, bool to_domain, + XenDeviceGrantCopySegment segs[], + unsigned int nr_segs, Error **errp); + +typedef struct XenEventChannel XenEventChannel; + +typedef void (*XenEventHandler)(void *opaque); + +XenEventChannel *xen_device_bind_event_channel(XenDevice *xendev, + unsigned int port, + XenEventHandler handler, + void *opaque, Error **errp); +void xen_device_notify_event_channel(XenDevice *xendev, + XenEventChannel *channel, + Error **errp); +void xen_device_unbind_event_channel(XenDevice *xendev, + XenEventChannel *channel, + Error **errp); + +#endif /* HW_XEN_BUS_H */ diff --git a/include/hw/xen/xen_backend.h b/include/hw/xen/xen-legacy-backend.h index 9c17fdd85d..20cb47b5bf 100644 --- a/include/hw/xen/xen_backend.h +++ b/include/hw/xen/xen-legacy-backend.h @@ -11,7 +11,7 @@ #define TYPE_XENBACKEND "xen-backend" #define XENBACKEND_DEVICE(obj) \ - OBJECT_CHECK(XenDevice, (obj), TYPE_XENBACKEND) + OBJECT_CHECK(XenLegacyDevice, (obj), TYPE_XENBACKEND) /* variables */ extern struct xs_handle *xenstore; @@ -20,32 +20,37 @@ extern DeviceState *xen_sysdev; extern BusState *xen_sysbus; int xenstore_mkdir(char *path, int p); -int xenstore_write_be_str(struct XenDevice *xendev, const char *node, const char *val); -int xenstore_write_be_int(struct XenDevice *xendev, const char *node, int ival); -int xenstore_write_be_int64(struct XenDevice *xendev, const char *node, int64_t ival); -char *xenstore_read_be_str(struct XenDevice *xendev, const char *node); -int xenstore_read_be_int(struct XenDevice *xendev, const char *node, int *ival); -void xenstore_update_fe(char *watch, struct XenDevice *xendev); +int xenstore_write_be_str(struct XenLegacyDevice *xendev, const char *node, + const char *val); +int xenstore_write_be_int(struct XenLegacyDevice *xendev, const char *node, + int ival); +int xenstore_write_be_int64(struct XenLegacyDevice *xendev, const char *node, + int64_t ival); +char *xenstore_read_be_str(struct XenLegacyDevice *xendev, const char *node); +int xenstore_read_be_int(struct XenLegacyDevice *xendev, const char *node, + int *ival); +void xenstore_update_fe(char *watch, struct XenLegacyDevice *xendev); void xenstore_update_be(char *watch, char *type, int dom, struct XenDevOps *ops); -char *xenstore_read_fe_str(struct XenDevice *xendev, const char *node); -int xenstore_read_fe_int(struct XenDevice *xendev, const char *node, int *ival); -int xenstore_read_fe_uint64(struct XenDevice *xendev, const char *node, +char *xenstore_read_fe_str(struct XenLegacyDevice *xendev, const char *node); +int xenstore_read_fe_int(struct XenLegacyDevice *xendev, const char *node, + int *ival); +int xenstore_read_fe_uint64(struct XenLegacyDevice *xendev, const char *node, uint64_t *uval); -void xen_be_check_state(struct XenDevice *xendev); +void xen_be_check_state(struct XenLegacyDevice *xendev); /* xen backend driver bits */ int xen_be_init(void); void xen_be_register_common(void); int xen_be_register(const char *type, struct XenDevOps *ops); -int xen_be_set_state(struct XenDevice *xendev, enum xenbus_state state); -int xen_be_bind_evtchn(struct XenDevice *xendev); -void xen_be_set_max_grant_refs(struct XenDevice *xendev, +int xen_be_set_state(struct XenLegacyDevice *xendev, enum xenbus_state state); +int xen_be_bind_evtchn(struct XenLegacyDevice *xendev); +void xen_be_set_max_grant_refs(struct XenLegacyDevice *xendev, unsigned int nr_refs); -void *xen_be_map_grant_refs(struct XenDevice *xendev, uint32_t *refs, +void *xen_be_map_grant_refs(struct XenLegacyDevice *xendev, uint32_t *refs, unsigned int nr_refs, int prot); -void xen_be_unmap_grant_refs(struct XenDevice *xendev, void *ptr, +void xen_be_unmap_grant_refs(struct XenLegacyDevice *xendev, void *ptr, unsigned int nr_refs); typedef struct XenGrantCopySegment { @@ -59,17 +64,17 @@ typedef struct XenGrantCopySegment { size_t len; } XenGrantCopySegment; -int xen_be_copy_grant_refs(struct XenDevice *xendev, +int xen_be_copy_grant_refs(struct XenLegacyDevice *xendev, bool to_domain, XenGrantCopySegment segs[], unsigned int nr_segs); -static inline void *xen_be_map_grant_ref(struct XenDevice *xendev, +static inline void *xen_be_map_grant_ref(struct XenLegacyDevice *xendev, uint32_t ref, int prot) { return xen_be_map_grant_refs(xendev, &ref, 1, prot); } -static inline void xen_be_unmap_grant_ref(struct XenDevice *xendev, +static inline void xen_be_unmap_grant_ref(struct XenLegacyDevice *xendev, void *ptr) { return xen_be_unmap_grant_refs(xendev, ptr, 1); diff --git a/include/hw/xen/xen.h b/include/hw/xen/xen.h index 7efcdaa8fe..ba039c146d 100644 --- a/include/hw/xen/xen.h +++ b/include/hw/xen/xen.h @@ -15,8 +15,7 @@ /* xen-machine.c */ enum xen_mode { XEN_EMULATE = 0, // xen emulation, using xenner (default) - XEN_CREATE, // create xen domain - XEN_ATTACH // attach to xen domain created by xend + XEN_ATTACH // attach to xen domain created by libxl }; extern uint32_t xen_domid; diff --git a/include/hw/xen/xen_common.h b/include/hw/xen/xen_common.h index 93f631e5bf..9a8155e172 100644 --- a/include/hw/xen/xen_common.h +++ b/include/hw/xen/xen_common.h @@ -32,6 +32,7 @@ extern xc_interface *xen_xc; typedef xc_interface xenforeignmemory_handle; typedef xc_evtchn xenevtchn_handle; typedef xc_gnttab xengnttab_handle; +typedef evtchn_port_or_error_t xenevtchn_port_or_error_t; #define xenevtchn_open(l, f) xc_evtchn_open(l, f); #define xenevtchn_close(h) xc_evtchn_close(h) @@ -661,24 +662,6 @@ static inline int xen_set_ioreq_server_state(domid_t dom, #endif -#ifdef CONFIG_XEN_PV_DOMAIN_BUILD -#if CONFIG_XEN_CTRL_INTERFACE_VERSION < 40700 -static inline int xen_domain_create(xc_interface *xc, uint32_t ssidref, - xen_domain_handle_t handle, uint32_t flags, - uint32_t *pdomid) -{ - return xc_domain_create(xc, ssidref, handle, flags, pdomid); -} -#else -static inline int xen_domain_create(xc_interface *xc, uint32_t ssidref, - xen_domain_handle_t handle, uint32_t flags, - uint32_t *pdomid) -{ - return xc_domain_create(xc, ssidref, handle, flags, pdomid, NULL); -} -#endif -#endif - /* Xen before 4.8 */ #if CONFIG_XEN_CTRL_INTERFACE_VERSION < 40800 diff --git a/include/hw/xen/xen_pvdev.h b/include/hw/xen/xen_pvdev.h index d473e9b34d..83e5174d90 100644 --- a/include/hw/xen/xen_pvdev.h +++ b/include/hw/xen/xen_pvdev.h @@ -6,7 +6,7 @@ #define XEN_BUFSIZE 1024 -struct XenDevice; +struct XenLegacyDevice; /* driver uses grant tables -> open gntdev device (xendev->gnttabdev) */ #define DEVOPS_FLAG_NEED_GNTDEV 1 @@ -16,19 +16,21 @@ struct XenDevice; struct XenDevOps { size_t size; uint32_t flags; - void (*alloc)(struct XenDevice *xendev); - int (*init)(struct XenDevice *xendev); - int (*initialise)(struct XenDevice *xendev); - void (*connected)(struct XenDevice *xendev); - void (*event)(struct XenDevice *xendev); - void (*disconnect)(struct XenDevice *xendev); - int (*free)(struct XenDevice *xendev); - void (*backend_changed)(struct XenDevice *xendev, const char *node); - void (*frontend_changed)(struct XenDevice *xendev, const char *node); + void (*alloc)(struct XenLegacyDevice *xendev); + int (*init)(struct XenLegacyDevice *xendev); + int (*initialise)(struct XenLegacyDevice *xendev); + void (*connected)(struct XenLegacyDevice *xendev); + void (*event)(struct XenLegacyDevice *xendev); + void (*disconnect)(struct XenLegacyDevice *xendev); + int (*free)(struct XenLegacyDevice *xendev); + void (*backend_changed)(struct XenLegacyDevice *xendev, + const char *node); + void (*frontend_changed)(struct XenLegacyDevice *xendev, + const char *node); int (*backend_register)(void); }; -struct XenDevice { +struct XenLegacyDevice { DeviceState qdev; const char *type; int dom; @@ -49,7 +51,7 @@ struct XenDevice { xengnttab_handle *gnttabdev; struct XenDevOps *ops; - QTAILQ_ENTRY(XenDevice) next; + QTAILQ_ENTRY(XenLegacyDevice) next; }; /* ------------------------------------------------------------- */ @@ -66,14 +68,14 @@ void xenstore_update(void *unused); const char *xenbus_strstate(enum xenbus_state state); void xen_pv_evtchn_event(void *opaque); -void xen_pv_insert_xendev(struct XenDevice *xendev); -void xen_pv_del_xendev(struct XenDevice *xendev); -struct XenDevice *xen_pv_find_xendev(const char *type, int dom, int dev); +void xen_pv_insert_xendev(struct XenLegacyDevice *xendev); +void xen_pv_del_xendev(struct XenLegacyDevice *xendev); +struct XenLegacyDevice *xen_pv_find_xendev(const char *type, int dom, int dev); -void xen_pv_unbind_evtchn(struct XenDevice *xendev); -int xen_pv_send_notify(struct XenDevice *xendev); +void xen_pv_unbind_evtchn(struct XenLegacyDevice *xendev); +int xen_pv_send_notify(struct XenLegacyDevice *xendev); -void xen_pv_printf(struct XenDevice *xendev, int msg_level, +void xen_pv_printf(struct XenLegacyDevice *xendev, int msg_level, const char *fmt, ...) GCC_FMT_ATTR(3, 4); #endif /* QEMU_HW_XEN_PVDEV_H */ diff --git a/include/qemu/module.h b/include/qemu/module.h index 54300ab6e5..55dd2beea8 100644 --- a/include/qemu/module.h +++ b/include/qemu/module.h @@ -44,6 +44,7 @@ typedef enum { MODULE_INIT_OPTS, MODULE_INIT_QOM, MODULE_INIT_TRACE, + MODULE_INIT_XEN_BACKEND, MODULE_INIT_MAX } module_init_type; @@ -51,6 +52,8 @@ typedef enum { #define opts_init(function) module_init(function, MODULE_INIT_OPTS) #define type_init(function) module_init(function, MODULE_INIT_QOM) #define trace_init(function) module_init(function, MODULE_INIT_TRACE) +#define xen_backend_init(function) module_init(function, \ + MODULE_INIT_XEN_BACKEND) #define block_module_load_one(lib) module_load_one("block-", lib) #define ui_module_load_one(lib) module_load_one("ui-", lib) diff --git a/include/ui/input.h b/include/ui/input.h index 34ebc67c5a..8c8ccb999f 100644 --- a/include/ui/input.h +++ b/include/ui/input.h @@ -49,7 +49,6 @@ int qemu_input_key_value_to_scancode(const KeyValue *value, bool down, int *codes); int qemu_input_linux_to_qcode(unsigned int lnx); -InputEvent *qemu_input_event_new_btn(InputButton btn, bool down); void qemu_input_queue_btn(QemuConsole *src, InputButton btn, bool down); void qemu_input_update_buttons(QemuConsole *src, uint32_t *button_map, uint32_t button_old, uint32_t button_new); @@ -58,8 +57,6 @@ bool qemu_input_is_absolute(void); int qemu_input_scale_axis(int value, int min_in, int max_in, int min_out, int max_out); -InputEvent *qemu_input_event_new_move(InputEventKind kind, - InputAxis axis, int value); void qemu_input_queue_rel(QemuConsole *src, InputAxis axis, int value); void qemu_input_queue_abs(QemuConsole *src, InputAxis axis, int value, int min_in, int max_in); diff --git a/nbd/server.c b/nbd/server.c index 7af0ddffb2..e8c56607ef 100644 --- a/nbd/server.c +++ b/nbd/server.c @@ -1456,9 +1456,10 @@ static void nbd_eject_notifier(Notifier *n, void *data) } NBDExport *nbd_export_new(BlockDriverState *bs, off_t dev_offset, off_t size, - uint16_t nbdflags, void (*close)(NBDExport *), - bool writethrough, BlockBackend *on_eject_blk, - Error **errp) + const char *name, const char *description, + const char *bitmap, uint16_t nbdflags, + void (*close)(NBDExport *), bool writethrough, + BlockBackend *on_eject_blk, Error **errp) { AioContext *ctx; BlockBackend *blk; @@ -1471,6 +1472,7 @@ NBDExport *nbd_export_new(BlockDriverState *bs, off_t dev_offset, off_t size, * that BDRV_O_INACTIVE is cleared and the image is ready for write * access since the export could be available before migration handover. */ + assert(name); ctx = bdrv_get_aio_context(bs); aio_context_acquire(ctx); bdrv_invalidate_cache(bs, NULL); @@ -1494,6 +1496,8 @@ NBDExport *nbd_export_new(BlockDriverState *bs, off_t dev_offset, off_t size, QTAILQ_INIT(&exp->clients); exp->blk = blk; exp->dev_offset = dev_offset; + exp->name = g_strdup(name); + exp->description = g_strdup(description); exp->nbdflags = nbdflags; exp->size = size < 0 ? blk_getlength(blk) : size; if (exp->size < 0) { @@ -1503,6 +1507,43 @@ NBDExport *nbd_export_new(BlockDriverState *bs, off_t dev_offset, off_t size, } exp->size -= exp->size % BDRV_SECTOR_SIZE; + if (bitmap) { + BdrvDirtyBitmap *bm = NULL; + BlockDriverState *bs = blk_bs(blk); + + while (true) { + bm = bdrv_find_dirty_bitmap(bs, bitmap); + if (bm != NULL || bs->backing == NULL) { + break; + } + + bs = bs->backing->bs; + } + + if (bm == NULL) { + error_setg(errp, "Bitmap '%s' is not found", bitmap); + goto fail; + } + + if ((nbdflags & NBD_FLAG_READ_ONLY) && bdrv_is_writable(bs) && + bdrv_dirty_bitmap_enabled(bm)) { + error_setg(errp, + "Enabled bitmap '%s' incompatible with readonly export", + bitmap); + goto fail; + } + + if (bdrv_dirty_bitmap_user_locked(bm)) { + error_setg(errp, "Bitmap '%s' is in use", bitmap); + goto fail; + } + + bdrv_dirty_bitmap_set_qmp_locked(bm, true); + exp->export_bitmap = bm; + exp->export_bitmap_context = g_strdup_printf("qemu:dirty-bitmap:%s", + bitmap); + } + exp->close = close; exp->ctx = blk_get_aio_context(blk); blk_add_aio_context_notifier(blk, blk_aio_attached, blk_aio_detach, exp); @@ -1513,10 +1554,14 @@ NBDExport *nbd_export_new(BlockDriverState *bs, off_t dev_offset, off_t size, exp->eject_notifier.notify = nbd_eject_notifier; blk_add_remove_bs_notifier(on_eject_blk, &exp->eject_notifier); } + QTAILQ_INSERT_TAIL(&exports, exp, next); + nbd_export_get(exp); return exp; fail: blk_unref(blk); + g_free(exp->name); + g_free(exp->description); g_free(exp); return NULL; } @@ -1533,43 +1578,29 @@ NBDExport *nbd_export_find(const char *name) return NULL; } -void nbd_export_set_name(NBDExport *exp, const char *name) -{ - if (exp->name == name) { - return; - } - - nbd_export_get(exp); - if (exp->name != NULL) { - g_free(exp->name); - exp->name = NULL; - QTAILQ_REMOVE(&exports, exp, next); - nbd_export_put(exp); - } - if (name != NULL) { - nbd_export_get(exp); - exp->name = g_strdup(name); - QTAILQ_INSERT_TAIL(&exports, exp, next); - } - nbd_export_put(exp); -} - -void nbd_export_set_description(NBDExport *exp, const char *description) -{ - g_free(exp->description); - exp->description = g_strdup(description); -} - void nbd_export_close(NBDExport *exp) { NBDClient *client, *next; nbd_export_get(exp); + /* + * TODO: Should we expand QMP NbdServerRemoveNode enum to allow a + * close mode that stops advertising the export to new clients but + * still permits existing clients to run to completion? Because of + * that possibility, nbd_export_close() can be called more than + * once on an export. + */ QTAILQ_FOREACH_SAFE(client, &exp->clients, next, next) { client_close(client, true); } - nbd_export_set_name(exp, NULL); - nbd_export_set_description(exp, NULL); + if (exp->name) { + nbd_export_put(exp); + g_free(exp->name); + exp->name = NULL; + QTAILQ_REMOVE(&exports, exp, next); + } + g_free(exp->description); + exp->description = NULL; nbd_export_put(exp); } @@ -2430,44 +2461,3 @@ void nbd_client_new(QIOChannelSocket *sioc, co = qemu_coroutine_create(nbd_co_client_start, client); qemu_coroutine_enter(co); } - -void nbd_export_bitmap(NBDExport *exp, const char *bitmap, - const char *bitmap_export_name, Error **errp) -{ - BdrvDirtyBitmap *bm = NULL; - BlockDriverState *bs = blk_bs(exp->blk); - - if (exp->export_bitmap) { - error_setg(errp, "Export bitmap is already set"); - return; - } - - while (true) { - bm = bdrv_find_dirty_bitmap(bs, bitmap); - if (bm != NULL || bs->backing == NULL) { - break; - } - - bs = bs->backing->bs; - } - - if (bm == NULL) { - error_setg(errp, "Bitmap '%s' is not found", bitmap); - return; - } - - if (bdrv_dirty_bitmap_enabled(bm)) { - error_setg(errp, "Bitmap '%s' is enabled", bitmap); - return; - } - - if (bdrv_dirty_bitmap_user_locked(bm)) { - error_setg(errp, "Bitmap '%s' is in use", bitmap); - return; - } - - bdrv_dirty_bitmap_set_qmp_locked(bm, true); - exp->export_bitmap = bm; - exp->export_bitmap_context = - g_strdup_printf("qemu:dirty-bitmap:%s", bitmap_export_name); -} diff --git a/qapi/block-core.json b/qapi/block-core.json index 762000f31f..91685be6c2 100644 --- a/qapi/block-core.json +++ b/qapi/block-core.json @@ -1806,29 +1806,29 @@ # Currently, all dirty tracking bitmaps are loaded from Qcow2 on # open. # -# @x-disabled: the bitmap is created in the disabled state, which means that -# it will not track drive changes. The bitmap may be enabled with -# x-block-dirty-bitmap-enable. Default is false. (Since: 3.0) +# @disabled: the bitmap is created in the disabled state, which means that +# it will not track drive changes. The bitmap may be enabled with +# block-dirty-bitmap-enable. Default is false. (Since: 4.0) # # Since: 2.4 ## { 'struct': 'BlockDirtyBitmapAdd', 'data': { 'node': 'str', 'name': 'str', '*granularity': 'uint32', - '*persistent': 'bool', '*autoload': 'bool', '*x-disabled': 'bool' } } + '*persistent': 'bool', '*autoload': 'bool', '*disabled': 'bool' } } ## # @BlockDirtyBitmapMerge: # # @node: name of device/node which the bitmap is tracking # -# @dst_name: name of the destination dirty bitmap +# @target: name of the destination dirty bitmap # -# @src_name: name of the source dirty bitmap +# @bitmaps: name(s) of the source dirty bitmap(s) # -# Since: 3.0 +# Since: 4.0 ## { 'struct': 'BlockDirtyBitmapMerge', - 'data': { 'node': 'str', 'dst_name': 'str', 'src_name': 'str' } } + 'data': { 'node': 'str', 'target': 'str', 'bitmaps': ['str'] } } ## # @block-dirty-bitmap-add: @@ -1899,7 +1899,7 @@ 'data': 'BlockDirtyBitmap' } ## -# @x-block-dirty-bitmap-enable: +# @block-dirty-bitmap-enable: # # Enables a dirty bitmap so that it will begin tracking disk changes. # @@ -1907,20 +1907,20 @@ # If @node is not a valid block device, DeviceNotFound # If @name is not found, GenericError with an explanation # -# Since: 3.0 +# Since: 4.0 # # Example: # -# -> { "execute": "x-block-dirty-bitmap-enable", +# -> { "execute": "block-dirty-bitmap-enable", # "arguments": { "node": "drive0", "name": "bitmap0" } } # <- { "return": {} } # ## - { 'command': 'x-block-dirty-bitmap-enable', + { 'command': 'block-dirty-bitmap-enable', 'data': 'BlockDirtyBitmap' } ## -# @x-block-dirty-bitmap-disable: +# @block-dirty-bitmap-disable: # # Disables a dirty bitmap so that it will stop tracking disk changes. # @@ -1928,42 +1928,42 @@ # If @node is not a valid block device, DeviceNotFound # If @name is not found, GenericError with an explanation # -# Since: 3.0 +# Since: 4.0 # # Example: # -# -> { "execute": "x-block-dirty-bitmap-disable", +# -> { "execute": "block-dirty-bitmap-disable", # "arguments": { "node": "drive0", "name": "bitmap0" } } # <- { "return": {} } # ## - { 'command': 'x-block-dirty-bitmap-disable', + { 'command': 'block-dirty-bitmap-disable', 'data': 'BlockDirtyBitmap' } ## -# @x-block-dirty-bitmap-merge: +# @block-dirty-bitmap-merge: # -# FIXME: Rename @src_name and @dst_name to src-name and dst-name. -# -# Merge @src_name dirty bitmap to @dst_name dirty bitmap. @src_name dirty -# bitmap is unchanged. On error, @dst_name is unchanged. +# Merge dirty bitmaps listed in @bitmaps to the @target dirty bitmap. +# The @bitmaps dirty bitmaps are unchanged. +# On error, @target is unchanged. # # Returns: nothing on success # If @node is not a valid block device, DeviceNotFound -# If @dst_name or @src_name is not found, GenericError -# If bitmaps has different sizes or granularities, GenericError +# If any bitmap in @bitmaps or @target is not found, GenericError +# If any of the bitmaps have different sizes or granularities, +# GenericError # -# Since: 3.0 +# Since: 4.0 # # Example: # -# -> { "execute": "x-block-dirty-bitmap-merge", -# "arguments": { "node": "drive0", "dst_name": "bitmap0", -# "src_name": "bitmap1" } } +# -> { "execute": "block-dirty-bitmap-merge", +# "arguments": { "node": "drive0", "target": "bitmap0", +# "bitmaps": ["bitmap1"] } } # <- { "return": {} } # ## - { 'command': 'x-block-dirty-bitmap-merge', + { 'command': 'block-dirty-bitmap-merge', 'data': 'BlockDirtyBitmapMerge' } ## diff --git a/qapi/block.json b/qapi/block.json index 11f01f28ef..5a79d639e8 100644 --- a/qapi/block.json +++ b/qapi/block.json @@ -246,6 +246,10 @@ # # @writable: Whether clients should be able to write to the device via the # NBD connection (default false). + +# @bitmap: Also export the dirty bitmap reachable from @device, so the +# NBD client can use NBD_OPT_SET_META_CONTEXT with +# "qemu:dirty-bitmap:NAME" to inspect the bitmap. (since 4.0) # # Returns: error if the server is not running, or export with the same name # already exists. @@ -253,7 +257,8 @@ # Since: 1.3.0 ## { 'command': 'nbd-server-add', - 'data': {'device': 'str', '*name': 'str', '*writable': 'bool'} } + 'data': {'device': 'str', '*name': 'str', '*writable': 'bool', + '*bitmap': 'str' } } ## # @NbdServerRemoveMode: @@ -297,29 +302,6 @@ 'data': {'name': 'str', '*mode': 'NbdServerRemoveMode'} } ## -# @x-nbd-server-add-bitmap: -# -# Expose a dirty bitmap associated with the selected export. The bitmap search -# starts at the device attached to the export, and includes all backing files. -# The exported bitmap is then locked until the NBD export is removed. -# -# @name: Export name. -# -# @bitmap: Bitmap name to search for. -# -# @bitmap-export-name: How the bitmap will be seen by nbd clients -# (default @bitmap) -# -# Note: the client must use NBD_OPT_SET_META_CONTEXT with a query of -# "qemu:dirty-bitmap:NAME" (where NAME matches @bitmap-export-name) to access -# the exposed bitmap. -# -# Since: 3.0 -## - { 'command': 'x-nbd-server-add-bitmap', - 'data': {'name': 'str', 'bitmap': 'str', '*bitmap-export-name': 'str'} } - -## # @nbd-server-stop: # # Stop QEMU's embedded NBD server, and unregister all devices previously diff --git a/qapi/transaction.json b/qapi/transaction.json index 5875cdb16c..95edb78227 100644 --- a/qapi/transaction.json +++ b/qapi/transaction.json @@ -46,9 +46,9 @@ # - @abort: since 1.6 # - @block-dirty-bitmap-add: since 2.5 # - @block-dirty-bitmap-clear: since 2.5 -# - @x-block-dirty-bitmap-enable: since 3.0 -# - @x-block-dirty-bitmap-disable: since 3.0 -# - @x-block-dirty-bitmap-merge: since 3.1 +# - @block-dirty-bitmap-enable: since 4.0 +# - @block-dirty-bitmap-disable: since 4.0 +# - @block-dirty-bitmap-merge: since 4.0 # - @blockdev-backup: since 2.3 # - @blockdev-snapshot: since 2.5 # - @blockdev-snapshot-internal-sync: since 1.7 @@ -62,9 +62,9 @@ 'abort': 'Abort', 'block-dirty-bitmap-add': 'BlockDirtyBitmapAdd', 'block-dirty-bitmap-clear': 'BlockDirtyBitmap', - 'x-block-dirty-bitmap-enable': 'BlockDirtyBitmap', - 'x-block-dirty-bitmap-disable': 'BlockDirtyBitmap', - 'x-block-dirty-bitmap-merge': 'BlockDirtyBitmapMerge', + 'block-dirty-bitmap-enable': 'BlockDirtyBitmap', + 'block-dirty-bitmap-disable': 'BlockDirtyBitmap', + 'block-dirty-bitmap-merge': 'BlockDirtyBitmapMerge', 'blockdev-backup': 'BlockdevBackup', 'blockdev-snapshot': 'BlockdevSnapshot', 'blockdev-snapshot-internal-sync': 'BlockdevSnapshotInternal', diff --git a/qemu-doc.texi b/qemu-doc.texi index f7ad1dfe4b..16b955cbf9 100644 --- a/qemu-doc.texi +++ b/qemu-doc.texi @@ -37,6 +37,7 @@ * QEMU System emulator for non PC targets:: * QEMU Guest Agent:: * QEMU User space emulator:: +* System requirements:: * Implementation notes:: * Deprecated features:: * Supported build platforms:: @@ -2813,6 +2814,18 @@ Act as if the host page size was 'pagesize' bytes Run the emulation in single step mode. @end table +@node System requirements +@chapter System requirements + +@section KVM kernel module + +On x86_64 hosts, the default set of CPU features enabled by the KVM accelerator +require the host to be running Linux v4.5 or newer. + +The OpteronG[345] CPU models require KVM support for RDTSCP, which was +added with Linux 4.5 which is supported by the major distros. And even +if RHEL7 has kernel 3.10, KVM there has the required functionality there +to make it close to a 4.5 or newer kernel. @include qemu-tech.texi diff --git a/qemu-nbd.c b/qemu-nbd.c index 2807e13239..51b55f2e06 100644 --- a/qemu-nbd.c +++ b/qemu-nbd.c @@ -61,7 +61,7 @@ #define MBR_SIZE 512 -static NBDExport *exp; +static NBDExport *export; static int verbose; static char *srcpath; static SocketAddress *saddr; @@ -95,6 +95,7 @@ static void usage(const char *name) "Exposing part of the image:\n" " -o, --offset=OFFSET offset into the image\n" " -P, --partition=NUM only expose partition NUM\n" +" -B, --bitmap=NAME expose a persistent dirty bitmap\n" "\n" "General purpose options:\n" " --object type,id=ID,... define an object such as 'secret' for providing\n" @@ -335,7 +336,7 @@ static int nbd_can_accept(void) return state == RUNNING && nb_fds < shared; } -static void nbd_export_closed(NBDExport *exp) +static void nbd_export_closed(NBDExport *export) { assert(state == TERMINATING); state = TERMINATED; @@ -509,7 +510,7 @@ int main(int argc, char **argv) off_t fd_size; QemuOpts *sn_opts = NULL; const char *sn_id_or_name = NULL; - const char *sopt = "hVb:o:p:rsnP:c:dvk:e:f:tl:x:T:D:"; + const char *sopt = "hVb:o:p:rsnP:c:dvk:e:f:tl:x:T:D:B:"; struct option lopt[] = { { "help", no_argument, NULL, 'h' }, { "version", no_argument, NULL, 'V' }, @@ -519,6 +520,7 @@ int main(int argc, char **argv) { "offset", required_argument, NULL, 'o' }, { "read-only", no_argument, NULL, 'r' }, { "partition", required_argument, NULL, 'P' }, + { "bitmap", required_argument, NULL, 'B' }, { "connect", required_argument, NULL, 'c' }, { "disconnect", no_argument, NULL, 'd' }, { "snapshot", no_argument, NULL, 's' }, @@ -558,6 +560,7 @@ int main(int argc, char **argv) QDict *options = NULL; const char *export_name = ""; /* Default export name */ const char *export_description = NULL; + const char *bitmap = NULL; const char *tlscredsid = NULL; bool imageOpts = false; bool writethrough = true; @@ -695,6 +698,9 @@ int main(int argc, char **argv) exit(EXIT_FAILURE); } break; + case 'B': + bitmap = optarg; + break; case 'k': sockpath = optarg; if (sockpath[0] != '/') { @@ -1015,10 +1021,10 @@ int main(int argc, char **argv) } } - exp = nbd_export_new(bs, dev_offset, fd_size, nbdflags, nbd_export_closed, - writethrough, NULL, &error_fatal); - nbd_export_set_name(exp, export_name); - nbd_export_set_description(exp, export_description); + export = nbd_export_new(bs, dev_offset, fd_size, export_name, + export_description, bitmap, nbdflags, + nbd_export_closed, writethrough, NULL, + &error_fatal); if (device) { #if HAVE_NBD_DEVICE @@ -1055,9 +1061,9 @@ int main(int argc, char **argv) main_loop_wait(false); if (state == TERMINATE) { state = TERMINATING; - nbd_export_close(exp); - nbd_export_put(exp); - exp = NULL; + nbd_export_close(export); + nbd_export_put(export); + export = NULL; } } while (state != TERMINATED); diff --git a/qemu-nbd.texi b/qemu-nbd.texi index 9a84e81eed..96b1546006 100644 --- a/qemu-nbd.texi +++ b/qemu-nbd.texi @@ -45,6 +45,10 @@ auto-detecting Export the disk as read-only @item -P, --partition=@var{num} Only expose partition @var{num} +@item -B, --bitmap=@var{name} +If @var{filename} has a qcow2 persistent bitmap @var{name}, expose +that bitmap via the ``qemu:dirty-bitmap:@var{name}'' context +accessible through NBD_OPT_SET_META_CONTEXT. @item -s, --snapshot Use @var{filename} as an external snapshot, create a temporary file with backing_file=@var{filename}, redirect the write to diff --git a/qemu-options.hx b/qemu-options.hx index d4f3564b78..521511ec13 100644 --- a/qemu-options.hx +++ b/qemu-options.hx @@ -3394,13 +3394,9 @@ ETEXI DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid, "-xen-domid id specify xen guest domain id\n", QEMU_ARCH_ALL) -DEF("xen-create", 0, QEMU_OPTION_xen_create, - "-xen-create create domain using xen hypercalls, bypassing xend\n" - " warning: should not be used when xend is in use\n", - QEMU_ARCH_ALL) DEF("xen-attach", 0, QEMU_OPTION_xen_attach, "-xen-attach attach to existing xen domain\n" - " xend will use this when starting QEMU\n", + " libxl will use this when starting QEMU\n", QEMU_ARCH_ALL) DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict, "-xen-domid-restrict restrict set of available xen operations\n" @@ -3411,14 +3407,10 @@ STEXI @item -xen-domid @var{id} @findex -xen-domid Specify xen guest domain @var{id} (XEN only). -@item -xen-create -@findex -xen-create -Create domain using xen hypercalls, bypassing xend. -Warning: should not be used when xend is in use (XEN only). @item -xen-attach @findex -xen-attach Attach to existing xen domain. -xend will use this when starting QEMU (XEN only). +libxl will use this when starting QEMU (XEN only). @findex -xen-domid-restrict Restrict set of available xen operations to specified domain id (XEN only). ETEXI diff --git a/target/i386/cpu.c b/target/i386/cpu.c index fa37203d89..2f5412592d 100644 --- a/target/i386/cpu.c +++ b/target/i386/cpu.c @@ -929,6 +929,13 @@ static FeatureWordInfo feature_word_info[FEATURE_WORDS] = { */ .no_autoenable_flags = ~0U, }, + /* + * .feat_names are commented out for Hyper-V enlightenments because we + * don't want to have two different ways for enabling them on QEMU command + * line. Some features (e.g. "hyperv_time", "hyperv_vapic", ...) require + * enabling several feature bits simultaneously, exposing these bits + * individually may just confuse guests. + */ [FEAT_HYPERV_EAX] = { .type = CPUID_FEATURE_WORD, .feat_names = { @@ -980,6 +987,36 @@ static FeatureWordInfo feature_word_info[FEATURE_WORDS] = { }, .cpuid = { .eax = 0x40000003, .reg = R_EDX, }, }, + [FEAT_HV_RECOMM_EAX] = { + .type = CPUID_FEATURE_WORD, + .feat_names = { + NULL /* hv_recommend_pv_as_switch */, + NULL /* hv_recommend_pv_tlbflush_local */, + NULL /* hv_recommend_pv_tlbflush_remote */, + NULL /* hv_recommend_msr_apic_access */, + NULL /* hv_recommend_msr_reset */, + NULL /* hv_recommend_relaxed_timing */, + NULL /* hv_recommend_dma_remapping */, + NULL /* hv_recommend_int_remapping */, + NULL /* hv_recommend_x2apic_msrs */, + NULL /* hv_recommend_autoeoi_deprecation */, + NULL /* hv_recommend_pv_ipi */, + NULL /* hv_recommend_ex_hypercalls */, + NULL /* hv_hypervisor_is_nested */, + NULL /* hv_recommend_int_mbec */, + NULL /* hv_recommend_evmcs */, + NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, + }, + .cpuid = { .eax = 0x40000004, .reg = R_EAX, }, + }, + [FEAT_HV_NESTED_EAX] = { + .type = CPUID_FEATURE_WORD, + .cpuid = { .eax = 0x4000000A, .reg = R_EAX, }, + }, [FEAT_SVM] = { .type = CPUID_FEATURE_WORD, .feat_names = { @@ -2296,7 +2333,7 @@ static X86CPUDefinition builtin_x86_defs[] = { CPUID_7_0_EBX_HLE | CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP | CPUID_7_0_EBX_BMI2 | CPUID_7_0_EBX_ERMS | CPUID_7_0_EBX_INVPCID | CPUID_7_0_EBX_RTM | CPUID_7_0_EBX_RDSEED | CPUID_7_0_EBX_ADX | - CPUID_7_0_EBX_SMAP | CPUID_7_0_EBX_MPX, + CPUID_7_0_EBX_SMAP, /* Missing: XSAVES (not supported by some Linux versions, * including v4.1 to v4.12). * KVM doesn't yet expose any XSAVES state save component, @@ -2343,7 +2380,7 @@ static X86CPUDefinition builtin_x86_defs[] = { CPUID_7_0_EBX_HLE | CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP | CPUID_7_0_EBX_BMI2 | CPUID_7_0_EBX_ERMS | CPUID_7_0_EBX_INVPCID | CPUID_7_0_EBX_RTM | CPUID_7_0_EBX_RDSEED | CPUID_7_0_EBX_ADX | - CPUID_7_0_EBX_SMAP | CPUID_7_0_EBX_MPX, + CPUID_7_0_EBX_SMAP, /* Missing: XSAVES (not supported by some Linux versions, * including v4.1 to v4.12). * KVM doesn't yet expose any XSAVES state save component, @@ -2388,7 +2425,7 @@ static X86CPUDefinition builtin_x86_defs[] = { CPUID_7_0_EBX_HLE | CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP | CPUID_7_0_EBX_BMI2 | CPUID_7_0_EBX_ERMS | CPUID_7_0_EBX_INVPCID | CPUID_7_0_EBX_RTM | CPUID_7_0_EBX_RDSEED | CPUID_7_0_EBX_ADX | - CPUID_7_0_EBX_SMAP | CPUID_7_0_EBX_MPX | CPUID_7_0_EBX_CLWB | + CPUID_7_0_EBX_SMAP | CPUID_7_0_EBX_CLWB | CPUID_7_0_EBX_AVX512F | CPUID_7_0_EBX_AVX512DQ | CPUID_7_0_EBX_AVX512BW | CPUID_7_0_EBX_AVX512CD | CPUID_7_0_EBX_AVX512VL | CPUID_7_0_EBX_CLFLUSHOPT, @@ -2440,7 +2477,7 @@ static X86CPUDefinition builtin_x86_defs[] = { CPUID_7_0_EBX_HLE | CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP | CPUID_7_0_EBX_BMI2 | CPUID_7_0_EBX_ERMS | CPUID_7_0_EBX_INVPCID | CPUID_7_0_EBX_RTM | CPUID_7_0_EBX_RDSEED | CPUID_7_0_EBX_ADX | - CPUID_7_0_EBX_SMAP | CPUID_7_0_EBX_MPX | CPUID_7_0_EBX_CLWB | + CPUID_7_0_EBX_SMAP | CPUID_7_0_EBX_CLWB | CPUID_7_0_EBX_AVX512F | CPUID_7_0_EBX_AVX512DQ | CPUID_7_0_EBX_AVX512BW | CPUID_7_0_EBX_AVX512CD | CPUID_7_0_EBX_AVX512VL, @@ -2490,7 +2527,7 @@ static X86CPUDefinition builtin_x86_defs[] = { CPUID_7_0_EBX_HLE | CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP | CPUID_7_0_EBX_BMI2 | CPUID_7_0_EBX_ERMS | CPUID_7_0_EBX_INVPCID | CPUID_7_0_EBX_RTM | CPUID_7_0_EBX_RDSEED | CPUID_7_0_EBX_ADX | - CPUID_7_0_EBX_SMAP | CPUID_7_0_EBX_MPX | CPUID_7_0_EBX_CLWB | + CPUID_7_0_EBX_SMAP | CPUID_7_0_EBX_CLWB | CPUID_7_0_EBX_AVX512F | CPUID_7_0_EBX_AVX512DQ | CPUID_7_0_EBX_AVX512BW | CPUID_7_0_EBX_AVX512CD | CPUID_7_0_EBX_AVX512VL | CPUID_7_0_EBX_CLFLUSHOPT | @@ -2546,7 +2583,7 @@ static X86CPUDefinition builtin_x86_defs[] = { CPUID_7_0_EBX_HLE | CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP | CPUID_7_0_EBX_BMI2 | CPUID_7_0_EBX_ERMS | CPUID_7_0_EBX_INVPCID | CPUID_7_0_EBX_RTM | CPUID_7_0_EBX_RDSEED | CPUID_7_0_EBX_ADX | - CPUID_7_0_EBX_SMAP | CPUID_7_0_EBX_MPX | CPUID_7_0_EBX_INTEL_PT, + CPUID_7_0_EBX_SMAP | CPUID_7_0_EBX_INTEL_PT, .features[FEAT_7_0_ECX] = CPUID_7_0_ECX_VBMI | CPUID_7_0_ECX_UMIP | CPUID_7_0_ECX_PKU | CPUID_7_0_ECX_OSPKE | CPUID_7_0_ECX_VBMI2 | CPUID_7_0_ECX_GFNI | @@ -2601,7 +2638,7 @@ static X86CPUDefinition builtin_x86_defs[] = { CPUID_7_0_EBX_HLE | CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP | CPUID_7_0_EBX_BMI2 | CPUID_7_0_EBX_ERMS | CPUID_7_0_EBX_INVPCID | CPUID_7_0_EBX_RTM | CPUID_7_0_EBX_RDSEED | CPUID_7_0_EBX_ADX | - CPUID_7_0_EBX_SMAP | CPUID_7_0_EBX_MPX | CPUID_7_0_EBX_CLWB | + CPUID_7_0_EBX_SMAP | CPUID_7_0_EBX_CLWB | CPUID_7_0_EBX_AVX512F | CPUID_7_0_EBX_AVX512DQ | CPUID_7_0_EBX_AVX512BW | CPUID_7_0_EBX_AVX512CD | CPUID_7_0_EBX_AVX512VL | CPUID_7_0_EBX_CLFLUSHOPT | @@ -2706,7 +2743,6 @@ static X86CPUDefinition builtin_x86_defs[] = { CPUID_DE | CPUID_FP87, .features[FEAT_1_ECX] = CPUID_EXT_CX16 | CPUID_EXT_SSE3, - /* Missing: CPUID_EXT2_RDTSCP */ .features[FEAT_8000_0001_EDX] = CPUID_EXT2_LM | CPUID_EXT2_NX | CPUID_EXT2_SYSCALL, .features[FEAT_8000_0001_ECX] = @@ -2730,9 +2766,9 @@ static X86CPUDefinition builtin_x86_defs[] = { .features[FEAT_1_ECX] = CPUID_EXT_POPCNT | CPUID_EXT_CX16 | CPUID_EXT_MONITOR | CPUID_EXT_SSE3, - /* Missing: CPUID_EXT2_RDTSCP */ .features[FEAT_8000_0001_EDX] = - CPUID_EXT2_LM | CPUID_EXT2_NX | CPUID_EXT2_SYSCALL, + CPUID_EXT2_LM | CPUID_EXT2_NX | CPUID_EXT2_SYSCALL | + CPUID_EXT2_RDTSCP, .features[FEAT_8000_0001_ECX] = CPUID_EXT3_MISALIGNSSE | CPUID_EXT3_SSE4A | CPUID_EXT3_ABM | CPUID_EXT3_SVM | CPUID_EXT3_LAHF_LM, @@ -2757,10 +2793,9 @@ static X86CPUDefinition builtin_x86_defs[] = { CPUID_EXT_POPCNT | CPUID_EXT_SSE42 | CPUID_EXT_SSE41 | CPUID_EXT_CX16 | CPUID_EXT_SSSE3 | CPUID_EXT_PCLMULQDQ | CPUID_EXT_SSE3, - /* Missing: CPUID_EXT2_RDTSCP */ .features[FEAT_8000_0001_EDX] = CPUID_EXT2_LM | CPUID_EXT2_PDPE1GB | CPUID_EXT2_NX | - CPUID_EXT2_SYSCALL, + CPUID_EXT2_SYSCALL | CPUID_EXT2_RDTSCP, .features[FEAT_8000_0001_ECX] = CPUID_EXT3_FMA4 | CPUID_EXT3_XOP | CPUID_EXT3_3DNOWPREFETCH | CPUID_EXT3_MISALIGNSSE | @@ -2788,10 +2823,9 @@ static X86CPUDefinition builtin_x86_defs[] = { CPUID_EXT_AES | CPUID_EXT_POPCNT | CPUID_EXT_SSE42 | CPUID_EXT_SSE41 | CPUID_EXT_CX16 | CPUID_EXT_FMA | CPUID_EXT_SSSE3 | CPUID_EXT_PCLMULQDQ | CPUID_EXT_SSE3, - /* Missing: CPUID_EXT2_RDTSCP */ .features[FEAT_8000_0001_EDX] = CPUID_EXT2_LM | CPUID_EXT2_PDPE1GB | CPUID_EXT2_NX | - CPUID_EXT2_SYSCALL, + CPUID_EXT2_SYSCALL | CPUID_EXT2_RDTSCP, .features[FEAT_8000_0001_ECX] = CPUID_EXT3_TBM | CPUID_EXT3_FMA4 | CPUID_EXT3_XOP | CPUID_EXT3_3DNOWPREFETCH | CPUID_EXT3_MISALIGNSSE | @@ -5151,6 +5185,10 @@ static void x86_cpu_realizefn(DeviceState *dev, Error **errp) if (cpu->host_phys_bits) { /* The user asked for us to use the host physical bits */ cpu->phys_bits = host_phys_bits; + if (cpu->host_phys_bits_limit && + cpu->phys_bits > cpu->host_phys_bits_limit) { + cpu->phys_bits = cpu->host_phys_bits_limit; + } } /* Print a warning if the user set it to a value that's not the @@ -5738,6 +5776,7 @@ static Property x86_cpu_properties[] = { DEFINE_PROP_BOOL("kvm", X86CPU, expose_kvm, true), DEFINE_PROP_UINT32("phys-bits", X86CPU, phys_bits, 0), DEFINE_PROP_BOOL("host-phys-bits", X86CPU, host_phys_bits, false), + DEFINE_PROP_UINT8("host-phys-bits-limit", X86CPU, host_phys_bits_limit, 0), DEFINE_PROP_BOOL("fill-mtrr-mask", X86CPU, fill_mtrr_mask, true), DEFINE_PROP_UINT32("level", X86CPU, env.cpuid_level, UINT32_MAX), DEFINE_PROP_UINT32("xlevel", X86CPU, env.cpuid_xlevel, UINT32_MAX), diff --git a/target/i386/cpu.h b/target/i386/cpu.h index ef41a033c5..59656a70e6 100644 --- a/target/i386/cpu.h +++ b/target/i386/cpu.h @@ -497,6 +497,8 @@ typedef enum FeatureWord { FEAT_HYPERV_EAX, /* CPUID[4000_0003].EAX */ FEAT_HYPERV_EBX, /* CPUID[4000_0003].EBX */ FEAT_HYPERV_EDX, /* CPUID[4000_0003].EDX */ + FEAT_HV_RECOMM_EAX, /* CPUID[4000_0004].EAX */ + FEAT_HV_NESTED_EAX, /* CPUID[4000_000A].EAX */ FEAT_SVM, /* CPUID[8000_000A].EDX */ FEAT_XSAVE, /* CPUID[EAX=0xd,ECX=1].EAX */ FEAT_6_EAX, /* CPUID[6].EAX */ @@ -1459,6 +1461,9 @@ struct X86CPU { /* if true override the phys_bits value with a value read from the host */ bool host_phys_bits; + /* if set, limit maximum value for phys_bits when host_phys_bits is true */ + uint8_t host_phys_bits_limit; + /* Stop SMI delivery for migration compatibility with old machines */ bool kvm_no_smi_migration; diff --git a/target/i386/kvm.c b/target/i386/kvm.c index 739cf8c8ea..9af4542fb8 100644 --- a/target/i386/kvm.c +++ b/target/i386/kvm.c @@ -798,6 +798,48 @@ static int hyperv_handle_properties(CPUState *cs) } env->features[FEAT_HYPERV_EAX] |= HV_SYNTIMERS_AVAILABLE; } + if (cpu->hyperv_relaxed_timing) { + env->features[FEAT_HV_RECOMM_EAX] |= HV_RELAXED_TIMING_RECOMMENDED; + } + if (cpu->hyperv_vapic) { + env->features[FEAT_HV_RECOMM_EAX] |= HV_APIC_ACCESS_RECOMMENDED; + } + if (cpu->hyperv_tlbflush) { + if (kvm_check_extension(cs->kvm_state, + KVM_CAP_HYPERV_TLBFLUSH) <= 0) { + fprintf(stderr, "Hyper-V TLB flush support " + "(requested by 'hv-tlbflush' cpu flag) " + " is not supported by kernel\n"); + return -ENOSYS; + } + env->features[FEAT_HV_RECOMM_EAX] |= HV_REMOTE_TLB_FLUSH_RECOMMENDED; + env->features[FEAT_HV_RECOMM_EAX] |= HV_EX_PROCESSOR_MASKS_RECOMMENDED; + } + if (cpu->hyperv_ipi) { + if (kvm_check_extension(cs->kvm_state, + KVM_CAP_HYPERV_SEND_IPI) <= 0) { + fprintf(stderr, "Hyper-V IPI send support " + "(requested by 'hv-ipi' cpu flag) " + " is not supported by kernel\n"); + return -ENOSYS; + } + env->features[FEAT_HV_RECOMM_EAX] |= HV_CLUSTER_IPI_RECOMMENDED; + env->features[FEAT_HV_RECOMM_EAX] |= HV_EX_PROCESSOR_MASKS_RECOMMENDED; + } + if (cpu->hyperv_evmcs) { + uint16_t evmcs_version; + + if (kvm_vcpu_enable_cap(cs, KVM_CAP_HYPERV_ENLIGHTENED_VMCS, 0, + (uintptr_t)&evmcs_version)) { + fprintf(stderr, "Hyper-V Enlightened VMCS " + "(requested by 'hv-evmcs' cpu flag) " + "is not supported by kernel\n"); + return -ENOSYS; + } + env->features[FEAT_HV_RECOMM_EAX] |= HV_ENLIGHTENED_VMCS_RECOMMENDED; + env->features[FEAT_HV_NESTED_EAX] = evmcs_version; + } + return 0; } @@ -879,7 +921,6 @@ int kvm_arch_init_vcpu(CPUState *cs) uint32_t unused; struct kvm_cpuid_entry2 *c; uint32_t signature[3]; - uint16_t evmcs_version; int kvm_base = KVM_CPUID_SIGNATURE; int r; Error *local_err = NULL; @@ -954,44 +995,8 @@ int kvm_arch_init_vcpu(CPUState *cs) c = &cpuid_data.entries[cpuid_i++]; c->function = HV_CPUID_ENLIGHTMENT_INFO; - if (cpu->hyperv_relaxed_timing) { - c->eax |= HV_RELAXED_TIMING_RECOMMENDED; - } - if (cpu->hyperv_vapic) { - c->eax |= HV_APIC_ACCESS_RECOMMENDED; - } - if (cpu->hyperv_tlbflush) { - if (kvm_check_extension(cs->kvm_state, - KVM_CAP_HYPERV_TLBFLUSH) <= 0) { - fprintf(stderr, "Hyper-V TLB flush support " - "(requested by 'hv-tlbflush' cpu flag) " - " is not supported by kernel\n"); - return -ENOSYS; - } - c->eax |= HV_REMOTE_TLB_FLUSH_RECOMMENDED; - c->eax |= HV_EX_PROCESSOR_MASKS_RECOMMENDED; - } - if (cpu->hyperv_ipi) { - if (kvm_check_extension(cs->kvm_state, - KVM_CAP_HYPERV_SEND_IPI) <= 0) { - fprintf(stderr, "Hyper-V IPI send support " - "(requested by 'hv-ipi' cpu flag) " - " is not supported by kernel\n"); - return -ENOSYS; - } - c->eax |= HV_CLUSTER_IPI_RECOMMENDED; - c->eax |= HV_EX_PROCESSOR_MASKS_RECOMMENDED; - } - if (cpu->hyperv_evmcs) { - if (kvm_vcpu_enable_cap(cs, KVM_CAP_HYPERV_ENLIGHTENED_VMCS, 0, - (uintptr_t)&evmcs_version)) { - fprintf(stderr, "Hyper-V Enlightened VMCS " - "(requested by 'hv-evmcs' cpu flag) " - "is not supported by kernel\n"); - return -ENOSYS; - } - c->eax |= HV_ENLIGHTENED_VMCS_RECOMMENDED; - } + + c->eax = env->features[FEAT_HV_RECOMM_EAX]; c->ebx = cpu->hyperv_spinlock_attempts; c = &cpuid_data.entries[cpuid_i++]; @@ -1015,7 +1020,7 @@ int kvm_arch_init_vcpu(CPUState *cs) c = &cpuid_data.entries[cpuid_i++]; c->function = HV_CPUID_NESTED_FEATURES; - c->eax = evmcs_version; + c->eax = env->features[FEAT_HV_NESTED_EAX]; } } diff --git a/target/riscv/Makefile.objs b/target/riscv/Makefile.objs index fcc5d34c1f..4072abe3e4 100644 --- a/target/riscv/Makefile.objs +++ b/target/riscv/Makefile.objs @@ -1 +1 @@ -obj-y += translate.o op_helper.o cpu_helper.o cpu.o fpu_helper.o gdbstub.o pmp.o +obj-y += translate.o op_helper.o cpu_helper.o cpu.o csr.o fpu_helper.o gdbstub.o pmp.o diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index 5e8a2cb2ba..28d7e5302f 100644 --- a/target/riscv/cpu.c +++ b/target/riscv/cpu.c @@ -126,6 +126,7 @@ static void rv32gcsu_priv1_09_1_cpu_init(Object *obj) set_versions(env, USER_VERSION_2_02_0, PRIV_VERSION_1_09_1); set_resetvec(env, DEFAULT_RSTVEC); set_feature(env, RISCV_FEATURE_MMU); + set_feature(env, RISCV_FEATURE_PMP); } static void rv32gcsu_priv1_10_0_cpu_init(Object *obj) @@ -135,6 +136,7 @@ static void rv32gcsu_priv1_10_0_cpu_init(Object *obj) set_versions(env, USER_VERSION_2_02_0, PRIV_VERSION_1_10_0); set_resetvec(env, DEFAULT_RSTVEC); set_feature(env, RISCV_FEATURE_MMU); + set_feature(env, RISCV_FEATURE_PMP); } static void rv32imacu_nommu_cpu_init(Object *obj) @@ -143,6 +145,7 @@ static void rv32imacu_nommu_cpu_init(Object *obj) set_misa(env, RV32 | RVI | RVM | RVA | RVC | RVU); set_versions(env, USER_VERSION_2_02_0, PRIV_VERSION_1_10_0); set_resetvec(env, DEFAULT_RSTVEC); + set_feature(env, RISCV_FEATURE_PMP); } #elif defined(TARGET_RISCV64) @@ -154,6 +157,7 @@ static void rv64gcsu_priv1_09_1_cpu_init(Object *obj) set_versions(env, USER_VERSION_2_02_0, PRIV_VERSION_1_09_1); set_resetvec(env, DEFAULT_RSTVEC); set_feature(env, RISCV_FEATURE_MMU); + set_feature(env, RISCV_FEATURE_PMP); } static void rv64gcsu_priv1_10_0_cpu_init(Object *obj) @@ -163,6 +167,7 @@ static void rv64gcsu_priv1_10_0_cpu_init(Object *obj) set_versions(env, USER_VERSION_2_02_0, PRIV_VERSION_1_10_0); set_resetvec(env, DEFAULT_RSTVEC); set_feature(env, RISCV_FEATURE_MMU); + set_feature(env, RISCV_FEATURE_PMP); } static void rv64imacu_nommu_cpu_init(Object *obj) @@ -171,6 +176,7 @@ static void rv64imacu_nommu_cpu_init(Object *obj) set_misa(env, RV64 | RVI | RVM | RVA | RVC | RVU); set_versions(env, USER_VERSION_2_02_0, PRIV_VERSION_1_10_0); set_resetvec(env, DEFAULT_RSTVEC); + set_feature(env, RISCV_FEATURE_PMP); } #endif diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index 4ee09b9cff..743f02c8b9 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -83,9 +83,10 @@ /* S extension denotes that Supervisor mode exists, however it is possible to have a core that support S mode but does not have an MMU and there is currently no bit in misa to indicate whether an MMU exists or not - so a cpu features bitfield is required */ + so a cpu features bitfield is required, likewise for optional PMP support */ enum { - RISCV_FEATURE_MMU + RISCV_FEATURE_MMU, + RISCV_FEATURE_PMP }; #define USER_VERSION_2_02_0 0x00020200 @@ -289,9 +290,39 @@ static inline void cpu_get_tb_cpu_state(CPURISCVState *env, target_ulong *pc, #endif } -void csr_write_helper(CPURISCVState *env, target_ulong val_to_write, - target_ulong csrno); -target_ulong csr_read_helper(CPURISCVState *env, target_ulong csrno); +int riscv_csrrw(CPURISCVState *env, int csrno, target_ulong *ret_value, + target_ulong new_value, target_ulong write_mask); + +static inline void csr_write_helper(CPURISCVState *env, target_ulong val, + int csrno) +{ + riscv_csrrw(env, csrno, NULL, val, MAKE_64BIT_MASK(0, TARGET_LONG_BITS)); +} + +static inline target_ulong csr_read_helper(CPURISCVState *env, int csrno) +{ + target_ulong val = 0; + riscv_csrrw(env, csrno, &val, 0, 0); + return val; +} + +typedef int (*riscv_csr_predicate_fn)(CPURISCVState *env, int csrno); +typedef int (*riscv_csr_read_fn)(CPURISCVState *env, int csrno, + target_ulong *ret_value); +typedef int (*riscv_csr_write_fn)(CPURISCVState *env, int csrno, + target_ulong new_value); +typedef int (*riscv_csr_op_fn)(CPURISCVState *env, int csrno, + target_ulong *ret_value, target_ulong new_value, target_ulong write_mask); + +typedef struct { + riscv_csr_predicate_fn predicate; + riscv_csr_read_fn read; + riscv_csr_write_fn write; + riscv_csr_op_fn op; +} riscv_csr_operations; + +void riscv_get_csr_ops(int csrno, riscv_csr_operations *ops); +void riscv_set_csr_ops(int csrno, riscv_csr_operations *ops); #include "exec/cpu-all.h" diff --git a/target/riscv/cpu_helper.c b/target/riscv/cpu_helper.c index 0234c2d528..f257050f12 100644 --- a/target/riscv/cpu_helper.c +++ b/target/riscv/cpu_helper.c @@ -404,7 +404,8 @@ int riscv_cpu_handle_mmu_fault(CPUState *cs, vaddr address, int size, qemu_log_mask(CPU_LOG_MMU, "%s address=%" VADDR_PRIx " ret %d physical " TARGET_FMT_plx " prot %d\n", __func__, address, ret, pa, prot); - if (!pmp_hart_has_privs(env, pa, TARGET_PAGE_SIZE, 1 << rw)) { + if (riscv_feature(env, RISCV_FEATURE_PMP) && + !pmp_hart_has_privs(env, pa, TARGET_PAGE_SIZE, 1 << rw)) { ret = TRANSLATE_FAIL; } if (ret == TRANSLATE_SUCCESS) { @@ -528,7 +529,7 @@ void riscv_cpu_do_interrupt(CPUState *cs) get_field(s, MSTATUS_SIE) : get_field(s, MSTATUS_UIE << env->priv)); s = set_field(s, MSTATUS_SPP, env->priv); s = set_field(s, MSTATUS_SIE, 0); - csr_write_helper(env, s, CSR_MSTATUS); + env->mstatus = s; riscv_set_mode(env, PRV_S); } else { /* No need to check MTVEC for misaligned - lower 2 bits cannot be set */ @@ -553,7 +554,7 @@ void riscv_cpu_do_interrupt(CPUState *cs) get_field(s, MSTATUS_MIE) : get_field(s, MSTATUS_UIE << env->priv)); s = set_field(s, MSTATUS_MPP, env->priv); s = set_field(s, MSTATUS_MIE, 0); - csr_write_helper(env, s, CSR_MSTATUS); + env->mstatus = s; riscv_set_mode(env, PRV_M); } /* TODO yield load reservation */ diff --git a/target/riscv/csr.c b/target/riscv/csr.c new file mode 100644 index 0000000000..5e7e7d16b8 --- /dev/null +++ b/target/riscv/csr.c @@ -0,0 +1,863 @@ +/* + * RISC-V Control and Status Registers. + * + * Copyright (c) 2016-2017 Sagar Karandikar, sagark@eecs.berkeley.edu + * Copyright (c) 2017-2018 SiFive, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2 or later, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "qemu/log.h" +#include "cpu.h" +#include "qemu/main-loop.h" +#include "exec/exec-all.h" + +/* CSR function table */ +static riscv_csr_operations csr_ops[]; + +/* CSR function table constants */ +enum { + CSR_TABLE_SIZE = 0x1000 +}; + +/* CSR function table public API */ +void riscv_get_csr_ops(int csrno, riscv_csr_operations *ops) +{ + *ops = csr_ops[csrno & (CSR_TABLE_SIZE - 1)]; +} + +void riscv_set_csr_ops(int csrno, riscv_csr_operations *ops) +{ + csr_ops[csrno & (CSR_TABLE_SIZE - 1)] = *ops; +} + +/* Predicates */ +static int fs(CPURISCVState *env, int csrno) +{ +#if !defined(CONFIG_USER_ONLY) + if (!(env->mstatus & MSTATUS_FS)) { + return -1; + } +#endif + return 0; +} + +static int ctr(CPURISCVState *env, int csrno) +{ +#if !defined(CONFIG_USER_ONLY) + target_ulong ctr_en = env->priv == PRV_U ? env->scounteren : + env->priv == PRV_S ? env->mcounteren : -1U; + if (!(ctr_en & (1 << (csrno & 31)))) { + return -1; + } +#endif + return 0; +} + +#if !defined(CONFIG_USER_ONLY) +static int any(CPURISCVState *env, int csrno) +{ + return 0; +} + +static int smode(CPURISCVState *env, int csrno) +{ + return -!riscv_has_ext(env, RVS); +} + +static int pmp(CPURISCVState *env, int csrno) +{ + return -!riscv_feature(env, RISCV_FEATURE_PMP); +} +#endif + +/* User Floating-Point CSRs */ +static int read_fflags(CPURISCVState *env, int csrno, target_ulong *val) +{ +#if !defined(CONFIG_USER_ONLY) + if (!(env->mstatus & MSTATUS_FS)) { + return -1; + } +#endif + *val = cpu_riscv_get_fflags(env); + return 0; +} + +static int write_fflags(CPURISCVState *env, int csrno, target_ulong val) +{ +#if !defined(CONFIG_USER_ONLY) + if (!(env->mstatus & MSTATUS_FS)) { + return -1; + } + env->mstatus |= MSTATUS_FS; +#endif + cpu_riscv_set_fflags(env, val & (FSR_AEXC >> FSR_AEXC_SHIFT)); + return 0; +} + +static int read_frm(CPURISCVState *env, int csrno, target_ulong *val) +{ +#if !defined(CONFIG_USER_ONLY) + if (!(env->mstatus & MSTATUS_FS)) { + return -1; + } +#endif + *val = env->frm; + return 0; +} + +static int write_frm(CPURISCVState *env, int csrno, target_ulong val) +{ +#if !defined(CONFIG_USER_ONLY) + if (!(env->mstatus & MSTATUS_FS)) { + return -1; + } + env->mstatus |= MSTATUS_FS; +#endif + env->frm = val & (FSR_RD >> FSR_RD_SHIFT); + return 0; +} + +static int read_fcsr(CPURISCVState *env, int csrno, target_ulong *val) +{ +#if !defined(CONFIG_USER_ONLY) + if (!(env->mstatus & MSTATUS_FS)) { + return -1; + } +#endif + *val = (cpu_riscv_get_fflags(env) << FSR_AEXC_SHIFT) + | (env->frm << FSR_RD_SHIFT); + return 0; +} + +static int write_fcsr(CPURISCVState *env, int csrno, target_ulong val) +{ +#if !defined(CONFIG_USER_ONLY) + if (!(env->mstatus & MSTATUS_FS)) { + return -1; + } + env->mstatus |= MSTATUS_FS; +#endif + env->frm = (val & FSR_RD) >> FSR_RD_SHIFT; + cpu_riscv_set_fflags(env, (val & FSR_AEXC) >> FSR_AEXC_SHIFT); + return 0; +} + +/* User Timers and Counters */ +static int read_instret(CPURISCVState *env, int csrno, target_ulong *val) +{ +#if !defined(CONFIG_USER_ONLY) + if (use_icount) { + *val = cpu_get_icount(); + } else { + *val = cpu_get_host_ticks(); + } +#else + *val = cpu_get_host_ticks(); +#endif + return 0; +} + +#if defined(TARGET_RISCV32) +static int read_instreth(CPURISCVState *env, int csrno, target_ulong *val) +{ +#if !defined(CONFIG_USER_ONLY) + if (use_icount) { + *val = cpu_get_icount() >> 32; + } else { + *val = cpu_get_host_ticks() >> 32; + } +#else + *val = cpu_get_host_ticks() >> 32; +#endif + return 0; +} +#endif /* TARGET_RISCV32 */ + +#if defined(CONFIG_USER_ONLY) +static int read_time(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = cpu_get_host_ticks(); + return 0; +} + +#if defined(TARGET_RISCV32) +static int read_timeh(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = cpu_get_host_ticks() >> 32; + return 0; +} +#endif + +#else /* CONFIG_USER_ONLY */ + +/* Machine constants */ + +#define M_MODE_INTERRUPTS (MIP_MSIP | MIP_MTIP | MIP_MEIP) +#define S_MODE_INTERRUPTS (MIP_SSIP | MIP_STIP | MIP_SEIP) + +static const target_ulong delegable_ints = S_MODE_INTERRUPTS; +static const target_ulong all_ints = M_MODE_INTERRUPTS | S_MODE_INTERRUPTS; +static const target_ulong delegable_excps = + (1ULL << (RISCV_EXCP_INST_ADDR_MIS)) | + (1ULL << (RISCV_EXCP_INST_ACCESS_FAULT)) | + (1ULL << (RISCV_EXCP_ILLEGAL_INST)) | + (1ULL << (RISCV_EXCP_BREAKPOINT)) | + (1ULL << (RISCV_EXCP_LOAD_ADDR_MIS)) | + (1ULL << (RISCV_EXCP_LOAD_ACCESS_FAULT)) | + (1ULL << (RISCV_EXCP_STORE_AMO_ADDR_MIS)) | + (1ULL << (RISCV_EXCP_STORE_AMO_ACCESS_FAULT)) | + (1ULL << (RISCV_EXCP_U_ECALL)) | + (1ULL << (RISCV_EXCP_S_ECALL)) | + (1ULL << (RISCV_EXCP_H_ECALL)) | + (1ULL << (RISCV_EXCP_M_ECALL)) | + (1ULL << (RISCV_EXCP_INST_PAGE_FAULT)) | + (1ULL << (RISCV_EXCP_LOAD_PAGE_FAULT)) | + (1ULL << (RISCV_EXCP_STORE_PAGE_FAULT)); +static const target_ulong sstatus_v1_9_mask = SSTATUS_SIE | SSTATUS_SPIE | + SSTATUS_UIE | SSTATUS_UPIE | SSTATUS_SPP | SSTATUS_FS | SSTATUS_XS | + SSTATUS_SUM | SSTATUS_SD; +static const target_ulong sstatus_v1_10_mask = SSTATUS_SIE | SSTATUS_SPIE | + SSTATUS_UIE | SSTATUS_UPIE | SSTATUS_SPP | SSTATUS_FS | SSTATUS_XS | + SSTATUS_SUM | SSTATUS_MXR | SSTATUS_SD; + +#if defined(TARGET_RISCV32) +static const char valid_vm_1_09[16] = { + [VM_1_09_MBARE] = 1, + [VM_1_09_SV32] = 1, +}; +static const char valid_vm_1_10[16] = { + [VM_1_10_MBARE] = 1, + [VM_1_10_SV32] = 1 +}; +#elif defined(TARGET_RISCV64) +static const char valid_vm_1_09[16] = { + [VM_1_09_MBARE] = 1, + [VM_1_09_SV39] = 1, + [VM_1_09_SV48] = 1, +}; +static const char valid_vm_1_10[16] = { + [VM_1_10_MBARE] = 1, + [VM_1_10_SV39] = 1, + [VM_1_10_SV48] = 1, + [VM_1_10_SV57] = 1 +}; +#endif /* CONFIG_USER_ONLY */ + +/* Machine Information Registers */ +static int read_zero(CPURISCVState *env, int csrno, target_ulong *val) +{ + return *val = 0; +} + +static int read_mhartid(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->mhartid; + return 0; +} + +/* Machine Trap Setup */ +static int read_mstatus(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->mstatus; + return 0; +} + +static int validate_vm(CPURISCVState *env, target_ulong vm) +{ + return (env->priv_ver >= PRIV_VERSION_1_10_0) ? + valid_vm_1_10[vm & 0xf] : valid_vm_1_09[vm & 0xf]; +} + +static int write_mstatus(CPURISCVState *env, int csrno, target_ulong val) +{ + target_ulong mstatus = env->mstatus; + target_ulong mask = 0; + target_ulong mpp = get_field(val, MSTATUS_MPP); + + /* flush tlb on mstatus fields that affect VM */ + if (env->priv_ver <= PRIV_VERSION_1_09_1) { + if ((val ^ mstatus) & (MSTATUS_MXR | MSTATUS_MPP | + MSTATUS_MPRV | MSTATUS_SUM | MSTATUS_VM)) { + tlb_flush(CPU(riscv_env_get_cpu(env))); + } + mask = MSTATUS_SIE | MSTATUS_SPIE | MSTATUS_MIE | MSTATUS_MPIE | + MSTATUS_SPP | MSTATUS_FS | MSTATUS_MPRV | MSTATUS_SUM | + MSTATUS_MPP | MSTATUS_MXR | + (validate_vm(env, get_field(val, MSTATUS_VM)) ? + MSTATUS_VM : 0); + } + if (env->priv_ver >= PRIV_VERSION_1_10_0) { + if ((val ^ mstatus) & (MSTATUS_MXR | MSTATUS_MPP | + MSTATUS_MPRV | MSTATUS_SUM)) { + tlb_flush(CPU(riscv_env_get_cpu(env))); + } + mask = MSTATUS_SIE | MSTATUS_SPIE | MSTATUS_MIE | MSTATUS_MPIE | + MSTATUS_SPP | MSTATUS_FS | MSTATUS_MPRV | MSTATUS_SUM | + MSTATUS_MPP | MSTATUS_MXR; + } + + /* silenty discard mstatus.mpp writes for unsupported modes */ + if (mpp == PRV_H || + (!riscv_has_ext(env, RVS) && mpp == PRV_S) || + (!riscv_has_ext(env, RVU) && mpp == PRV_U)) { + mask &= ~MSTATUS_MPP; + } + + mstatus = (mstatus & ~mask) | (val & mask); + + /* Note: this is a workaround for an issue where mstatus.FS + does not report dirty after floating point operations + that modify floating point state. This workaround is + technically compliant with the RISC-V Privileged + specification as it is legal to return only off, or dirty. + at the expense of extra floating point save/restore. */ + + /* FP is always dirty or off */ + if (mstatus & MSTATUS_FS) { + mstatus |= MSTATUS_FS; + } + + int dirty = ((mstatus & MSTATUS_FS) == MSTATUS_FS) | + ((mstatus & MSTATUS_XS) == MSTATUS_XS); + mstatus = set_field(mstatus, MSTATUS_SD, dirty); + env->mstatus = mstatus; + + return 0; +} + +static int read_misa(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->misa; + return 0; +} + +static int read_medeleg(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->medeleg; + return 0; +} + +static int write_medeleg(CPURISCVState *env, int csrno, target_ulong val) +{ + env->medeleg = (env->medeleg & ~delegable_excps) | (val & delegable_excps); + return 0; +} + +static int read_mideleg(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->mideleg; + return 0; +} + +static int write_mideleg(CPURISCVState *env, int csrno, target_ulong val) +{ + env->mideleg = (env->mideleg & ~delegable_ints) | (val & delegable_ints); + return 0; +} + +static int read_mie(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->mie; + return 0; +} + +static int write_mie(CPURISCVState *env, int csrno, target_ulong val) +{ + env->mie = (env->mie & ~all_ints) | (val & all_ints); + return 0; +} + +static int read_mtvec(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->mtvec; + return 0; +} + +static int write_mtvec(CPURISCVState *env, int csrno, target_ulong val) +{ + /* bits [1:0] encode mode; 0 = direct, 1 = vectored, 2 >= reserved */ + if ((val & 3) == 0) { + env->mtvec = val >> 2 << 2; + } else { + qemu_log_mask(LOG_UNIMP, "CSR_MTVEC: vectored traps not supported"); + } + return 0; +} + +static int read_mcounteren(CPURISCVState *env, int csrno, target_ulong *val) +{ + if (env->priv_ver < PRIV_VERSION_1_10_0) { + return -1; + } + *val = env->mcounteren; + return 0; +} + +static int write_mcounteren(CPURISCVState *env, int csrno, target_ulong val) +{ + if (env->priv_ver < PRIV_VERSION_1_10_0) { + return -1; + } + env->mcounteren = val; + return 0; +} + +static int read_mscounteren(CPURISCVState *env, int csrno, target_ulong *val) +{ + if (env->priv_ver > PRIV_VERSION_1_09_1) { + return -1; + } + *val = env->mcounteren; + return 0; +} + +static int write_mscounteren(CPURISCVState *env, int csrno, target_ulong val) +{ + if (env->priv_ver > PRIV_VERSION_1_09_1) { + return -1; + } + env->mcounteren = val; + return 0; +} + +static int read_mucounteren(CPURISCVState *env, int csrno, target_ulong *val) +{ + if (env->priv_ver > PRIV_VERSION_1_09_1) { + return -1; + } + *val = env->scounteren; + return 0; +} + +static int write_mucounteren(CPURISCVState *env, int csrno, target_ulong val) +{ + if (env->priv_ver > PRIV_VERSION_1_09_1) { + return -1; + } + env->scounteren = val; + return 0; +} + +/* Machine Trap Handling */ +static int read_mscratch(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->mscratch; + return 0; +} + +static int write_mscratch(CPURISCVState *env, int csrno, target_ulong val) +{ + env->mscratch = val; + return 0; +} + +static int read_mepc(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->mepc; + return 0; +} + +static int write_mepc(CPURISCVState *env, int csrno, target_ulong val) +{ + env->mepc = val; + return 0; +} + +static int read_mcause(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->mcause; + return 0; +} + +static int write_mcause(CPURISCVState *env, int csrno, target_ulong val) +{ + env->mcause = val; + return 0; +} + +static int read_mbadaddr(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->mbadaddr; + return 0; +} + +static int write_mbadaddr(CPURISCVState *env, int csrno, target_ulong val) +{ + env->mbadaddr = val; + return 0; +} + +static int rmw_mip(CPURISCVState *env, int csrno, target_ulong *ret_value, + target_ulong new_value, target_ulong write_mask) +{ + RISCVCPU *cpu = riscv_env_get_cpu(env); + target_ulong mask = write_mask & delegable_ints; + uint32_t old_mip; + + /* We can't allow the supervisor to control SEIP as this would allow the + * supervisor to clear a pending external interrupt which will result in + * lost a interrupt in the case a PLIC is attached. The SEIP bit must be + * hardware controlled when a PLIC is attached. This should be an option + * for CPUs with software-delegated Supervisor External Interrupts. */ + mask &= ~MIP_SEIP; + + if (mask) { + qemu_mutex_lock_iothread(); + old_mip = riscv_cpu_update_mip(cpu, mask, (new_value & mask)); + qemu_mutex_unlock_iothread(); + } else { + old_mip = atomic_read(&env->mip); + } + + if (ret_value) { + *ret_value = old_mip; + } + + return 0; +} + +/* Supervisor Trap Setup */ +static int read_sstatus(CPURISCVState *env, int csrno, target_ulong *val) +{ + target_ulong mask = ((env->priv_ver >= PRIV_VERSION_1_10_0) ? + sstatus_v1_10_mask : sstatus_v1_9_mask); + *val = env->mstatus & mask; + return 0; +} + +static int write_sstatus(CPURISCVState *env, int csrno, target_ulong val) +{ + target_ulong mask = ((env->priv_ver >= PRIV_VERSION_1_10_0) ? + sstatus_v1_10_mask : sstatus_v1_9_mask); + target_ulong newval = (env->mstatus & ~mask) | (val & mask); + return write_mstatus(env, CSR_MSTATUS, newval); +} + +static int read_sie(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->mie & env->mideleg; + return 0; +} + +static int write_sie(CPURISCVState *env, int csrno, target_ulong val) +{ + target_ulong newval = (env->mie & ~env->mideleg) | (val & env->mideleg); + return write_mie(env, CSR_MIE, newval); +} + +static int read_stvec(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->stvec; + return 0; +} + +static int write_stvec(CPURISCVState *env, int csrno, target_ulong val) +{ + /* bits [1:0] encode mode; 0 = direct, 1 = vectored, 2 >= reserved */ + if ((val & 3) == 0) { + env->stvec = val >> 2 << 2; + } else { + qemu_log_mask(LOG_UNIMP, "CSR_STVEC: vectored traps not supported"); + } + return 0; +} + +static int read_scounteren(CPURISCVState *env, int csrno, target_ulong *val) +{ + if (env->priv_ver < PRIV_VERSION_1_10_0) { + return -1; + } + *val = env->scounteren; + return 0; +} + +static int write_scounteren(CPURISCVState *env, int csrno, target_ulong val) +{ + if (env->priv_ver < PRIV_VERSION_1_10_0) { + return -1; + } + env->scounteren = val; + return 0; +} + +/* Supervisor Trap Handling */ +static int read_sscratch(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->sscratch; + return 0; +} + +static int write_sscratch(CPURISCVState *env, int csrno, target_ulong val) +{ + env->sscratch = val; + return 0; +} + +static int read_sepc(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->sepc; + return 0; +} + +static int write_sepc(CPURISCVState *env, int csrno, target_ulong val) +{ + env->sepc = val; + return 0; +} + +static int read_scause(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->scause; + return 0; +} + +static int write_scause(CPURISCVState *env, int csrno, target_ulong val) +{ + env->scause = val; + return 0; +} + +static int read_sbadaddr(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = env->sbadaddr; + return 0; +} + +static int write_sbadaddr(CPURISCVState *env, int csrno, target_ulong val) +{ + env->sbadaddr = val; + return 0; +} + +static int rmw_sip(CPURISCVState *env, int csrno, target_ulong *ret_value, + target_ulong new_value, target_ulong write_mask) +{ + return rmw_mip(env, CSR_MSTATUS, ret_value, new_value, + write_mask & env->mideleg); +} + +/* Supervisor Protection and Translation */ +static int read_satp(CPURISCVState *env, int csrno, target_ulong *val) +{ + if (!riscv_feature(env, RISCV_FEATURE_MMU)) { + *val = 0; + } else if (env->priv_ver >= PRIV_VERSION_1_10_0) { + *val = env->satp; + } else { + *val = env->sptbr; + } + return 0; +} + +static int write_satp(CPURISCVState *env, int csrno, target_ulong val) +{ + if (!riscv_feature(env, RISCV_FEATURE_MMU)) { + return 0; + } + if (env->priv_ver <= PRIV_VERSION_1_09_1 && (val ^ env->sptbr)) { + tlb_flush(CPU(riscv_env_get_cpu(env))); + env->sptbr = val & (((target_ulong) + 1 << (TARGET_PHYS_ADDR_SPACE_BITS - PGSHIFT)) - 1); + } + if (env->priv_ver >= PRIV_VERSION_1_10_0 && + validate_vm(env, get_field(val, SATP_MODE)) && + ((val ^ env->satp) & (SATP_MODE | SATP_ASID | SATP_PPN))) + { + tlb_flush(CPU(riscv_env_get_cpu(env))); + env->satp = val; + } + return 0; +} + +/* Physical Memory Protection */ +static int read_pmpcfg(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = pmpcfg_csr_read(env, csrno - CSR_PMPCFG0); + return 0; +} + +static int write_pmpcfg(CPURISCVState *env, int csrno, target_ulong val) +{ + pmpcfg_csr_write(env, csrno - CSR_PMPCFG0, val); + return 0; +} + +static int read_pmpaddr(CPURISCVState *env, int csrno, target_ulong *val) +{ + *val = pmpaddr_csr_read(env, csrno - CSR_PMPADDR0); + return 0; +} + +static int write_pmpaddr(CPURISCVState *env, int csrno, target_ulong val) +{ + pmpaddr_csr_write(env, csrno - CSR_PMPADDR0, val); + return 0; +} + +#endif + +/* + * riscv_csrrw - read and/or update control and status register + * + * csrr <-> riscv_csrrw(env, csrno, ret_value, 0, 0); + * csrrw <-> riscv_csrrw(env, csrno, ret_value, value, -1); + * csrrs <-> riscv_csrrw(env, csrno, ret_value, -1, value); + * csrrc <-> riscv_csrrw(env, csrno, ret_value, 0, value); + */ + +int riscv_csrrw(CPURISCVState *env, int csrno, target_ulong *ret_value, + target_ulong new_value, target_ulong write_mask) +{ + int ret; + target_ulong old_value; + + /* check privileges and return -1 if check fails */ +#if !defined(CONFIG_USER_ONLY) + int csr_priv = get_field(csrno, 0x300); + int read_only = get_field(csrno, 0xC00) == 3; + if ((write_mask && read_only) || (env->priv < csr_priv)) { + return -1; + } +#endif + + /* check predicate */ + if (!csr_ops[csrno].predicate || csr_ops[csrno].predicate(env, csrno) < 0) { + return -1; + } + + /* execute combined read/write operation if it exists */ + if (csr_ops[csrno].op) { + return csr_ops[csrno].op(env, csrno, ret_value, new_value, write_mask); + } + + /* if no accessor exists then return failure */ + if (!csr_ops[csrno].read) { + return -1; + } + + /* read old value */ + ret = csr_ops[csrno].read(env, csrno, &old_value); + if (ret < 0) { + return ret; + } + + /* write value if writable and write mask set, otherwise drop writes */ + if (write_mask) { + new_value = (old_value & ~write_mask) | (new_value & write_mask); + if (csr_ops[csrno].write) { + ret = csr_ops[csrno].write(env, csrno, new_value); + if (ret < 0) { + return ret; + } + } + } + + /* return old value */ + if (ret_value) { + *ret_value = old_value; + } + + return 0; +} + +/* Control and Status Register function table */ +static riscv_csr_operations csr_ops[CSR_TABLE_SIZE] = { + /* User Floating-Point CSRs */ + [CSR_FFLAGS] = { fs, read_fflags, write_fflags }, + [CSR_FRM] = { fs, read_frm, write_frm }, + [CSR_FCSR] = { fs, read_fcsr, write_fcsr }, + + /* User Timers and Counters */ + [CSR_CYCLE] = { ctr, read_instret }, + [CSR_INSTRET] = { ctr, read_instret }, +#if defined(TARGET_RISCV32) + [CSR_CYCLEH] = { ctr, read_instreth }, + [CSR_INSTRETH] = { ctr, read_instreth }, +#endif + + /* User-level time CSRs are only available in linux-user + * In privileged mode, the monitor emulates these CSRs */ +#if defined(CONFIG_USER_ONLY) + [CSR_TIME] = { ctr, read_time }, +#if defined(TARGET_RISCV32) + [CSR_TIMEH] = { ctr, read_timeh }, +#endif +#endif + +#if !defined(CONFIG_USER_ONLY) + /* Machine Timers and Counters */ + [CSR_MCYCLE] = { any, read_instret }, + [CSR_MINSTRET] = { any, read_instret }, +#if defined(TARGET_RISCV32) + [CSR_MCYCLEH] = { any, read_instreth }, + [CSR_MINSTRETH] = { any, read_instreth }, +#endif + + /* Machine Information Registers */ + [CSR_MVENDORID] = { any, read_zero }, + [CSR_MARCHID] = { any, read_zero }, + [CSR_MIMPID] = { any, read_zero }, + [CSR_MHARTID] = { any, read_mhartid }, + + /* Machine Trap Setup */ + [CSR_MSTATUS] = { any, read_mstatus, write_mstatus }, + [CSR_MISA] = { any, read_misa }, + [CSR_MIDELEG] = { any, read_mideleg, write_mideleg }, + [CSR_MEDELEG] = { any, read_medeleg, write_medeleg }, + [CSR_MIE] = { any, read_mie, write_mie }, + [CSR_MTVEC] = { any, read_mtvec, write_mtvec }, + [CSR_MCOUNTEREN] = { any, read_mcounteren, write_mcounteren }, + + /* Legacy Counter Setup (priv v1.9.1) */ + [CSR_MUCOUNTEREN] = { any, read_mucounteren, write_mucounteren }, + [CSR_MSCOUNTEREN] = { any, read_mscounteren, write_mscounteren }, + + /* Machine Trap Handling */ + [CSR_MSCRATCH] = { any, read_mscratch, write_mscratch }, + [CSR_MEPC] = { any, read_mepc, write_mepc }, + [CSR_MCAUSE] = { any, read_mcause, write_mcause }, + [CSR_MBADADDR] = { any, read_mbadaddr, write_mbadaddr }, + [CSR_MIP] = { any, NULL, NULL, rmw_mip }, + + /* Supervisor Trap Setup */ + [CSR_SSTATUS] = { smode, read_sstatus, write_sstatus }, + [CSR_SIE] = { smode, read_sie, write_sie }, + [CSR_STVEC] = { smode, read_stvec, write_stvec }, + [CSR_SCOUNTEREN] = { smode, read_scounteren, write_scounteren }, + + /* Supervisor Trap Handling */ + [CSR_SSCRATCH] = { smode, read_sscratch, write_sscratch }, + [CSR_SEPC] = { smode, read_sepc, write_sepc }, + [CSR_SCAUSE] = { smode, read_scause, write_scause }, + [CSR_SBADADDR] = { smode, read_sbadaddr, write_sbadaddr }, + [CSR_SIP] = { smode, NULL, NULL, rmw_sip }, + + /* Supervisor Protection and Translation */ + [CSR_SATP] = { smode, read_satp, write_satp }, + + /* Physical Memory Protection */ + [CSR_PMPCFG0 ... CSR_PMPADDR9] = { pmp, read_pmpcfg, write_pmpcfg }, + [CSR_PMPADDR0 ... CSR_PMPADDR15] = { pmp, read_pmpaddr, write_pmpaddr }, + + /* Performance Counters */ + [CSR_HPMCOUNTER3 ... CSR_HPMCOUNTER31] = { ctr, read_zero }, + [CSR_MHPMCOUNTER3 ... CSR_MHPMCOUNTER31] = { any, read_zero }, + [CSR_MHPMEVENT3 ... CSR_MHPMEVENT31] = { any, read_zero }, +#if defined(TARGET_RISCV32) + [CSR_HPMCOUNTER3H ... CSR_HPMCOUNTER31H] = { ctr, read_zero }, + [CSR_MHPMCOUNTER3H ... CSR_MHPMCOUNTER31H] = { any, read_zero }, +#endif +#endif /* !CONFIG_USER_ONLY */ +}; diff --git a/target/riscv/gdbstub.c b/target/riscv/gdbstub.c index 4f919b6c34..3cabb21cd0 100644 --- a/target/riscv/gdbstub.c +++ b/target/riscv/gdbstub.c @@ -33,7 +33,10 @@ int riscv_cpu_gdb_read_register(CPUState *cs, uint8_t *mem_buf, int n) } else if (n < 65) { return gdb_get_reg64(mem_buf, env->fpr[n - 33]); } else if (n < 4096 + 65) { - return gdb_get_regl(mem_buf, csr_read_helper(env, n - 65)); + target_ulong val = 0; + if (riscv_csrrw(env, n - 65, &val, 0, 0) == 0) { + return gdb_get_regl(mem_buf, val); + } } return 0; } @@ -56,7 +59,10 @@ int riscv_cpu_gdb_write_register(CPUState *cs, uint8_t *mem_buf, int n) env->fpr[n - 33] = ldq_p(mem_buf); /* always 64-bit */ return sizeof(uint64_t); } else if (n < 4096 + 65) { - csr_write_helper(env, ldtul_p(mem_buf), n - 65); + target_ulong val = ldtul_p(mem_buf); + if (riscv_csrrw(env, n - 65, NULL, val, -1) == 0) { + return sizeof(target_ulong); + } } return 0; } diff --git a/target/riscv/op_helper.c b/target/riscv/op_helper.c index 3726299d4a..81bd1a77ea 100644 --- a/target/riscv/op_helper.c +++ b/target/riscv/op_helper.c @@ -24,39 +24,6 @@ #include "exec/exec-all.h" #include "exec/helper-proto.h" -#ifndef CONFIG_USER_ONLY - -#if defined(TARGET_RISCV32) -static const char valid_vm_1_09[16] = { - [VM_1_09_MBARE] = 1, - [VM_1_09_SV32] = 1, -}; -static const char valid_vm_1_10[16] = { - [VM_1_10_MBARE] = 1, - [VM_1_10_SV32] = 1 -}; -#elif defined(TARGET_RISCV64) -static const char valid_vm_1_09[16] = { - [VM_1_09_MBARE] = 1, - [VM_1_09_SV39] = 1, - [VM_1_09_SV48] = 1, -}; -static const char valid_vm_1_10[16] = { - [VM_1_10_MBARE] = 1, - [VM_1_10_SV39] = 1, - [VM_1_10_SV48] = 1, - [VM_1_10_SV57] = 1 -}; -#endif - -static int validate_vm(CPURISCVState *env, target_ulong vm) -{ - return (env->priv_ver >= PRIV_VERSION_1_10_0) ? - valid_vm_1_10[vm & 0xf] : valid_vm_1_09[vm & 0xf]; -} - -#endif - /* Exceptions processing helpers */ void QEMU_NORETURN do_raise_exception_err(CPURISCVState *env, uint32_t exception, uintptr_t pc) @@ -72,584 +39,34 @@ void helper_raise_exception(CPURISCVState *env, uint32_t exception) do_raise_exception_err(env, exception, 0); } -static void validate_mstatus_fs(CPURISCVState *env, uintptr_t ra) -{ -#ifndef CONFIG_USER_ONLY - if (!(env->mstatus & MSTATUS_FS)) { - do_raise_exception_err(env, RISCV_EXCP_ILLEGAL_INST, ra); - } -#endif -} - -/* - * Handle writes to CSRs and any resulting special behavior - * - * Adapted from Spike's processor_t::set_csr - */ -void csr_write_helper(CPURISCVState *env, target_ulong val_to_write, - target_ulong csrno) -{ -#ifndef CONFIG_USER_ONLY - uint64_t delegable_ints = MIP_SSIP | MIP_STIP | MIP_SEIP; - uint64_t all_ints = delegable_ints | MIP_MSIP | MIP_MTIP; -#endif - - switch (csrno) { - case CSR_FFLAGS: - validate_mstatus_fs(env, GETPC()); - cpu_riscv_set_fflags(env, val_to_write & (FSR_AEXC >> FSR_AEXC_SHIFT)); - break; - case CSR_FRM: - validate_mstatus_fs(env, GETPC()); - env->frm = val_to_write & (FSR_RD >> FSR_RD_SHIFT); - break; - case CSR_FCSR: - validate_mstatus_fs(env, GETPC()); - env->frm = (val_to_write & FSR_RD) >> FSR_RD_SHIFT; - cpu_riscv_set_fflags(env, (val_to_write & FSR_AEXC) >> FSR_AEXC_SHIFT); - break; -#ifndef CONFIG_USER_ONLY - case CSR_MSTATUS: { - target_ulong mstatus = env->mstatus; - target_ulong mask = 0; - target_ulong mpp = get_field(val_to_write, MSTATUS_MPP); - - /* flush tlb on mstatus fields that affect VM */ - if (env->priv_ver <= PRIV_VERSION_1_09_1) { - if ((val_to_write ^ mstatus) & (MSTATUS_MXR | MSTATUS_MPP | - MSTATUS_MPRV | MSTATUS_SUM | MSTATUS_VM)) { - helper_tlb_flush(env); - } - mask = MSTATUS_SIE | MSTATUS_SPIE | MSTATUS_MIE | MSTATUS_MPIE | - MSTATUS_SPP | MSTATUS_FS | MSTATUS_MPRV | MSTATUS_SUM | - MSTATUS_MPP | MSTATUS_MXR | - (validate_vm(env, get_field(val_to_write, MSTATUS_VM)) ? - MSTATUS_VM : 0); - } - if (env->priv_ver >= PRIV_VERSION_1_10_0) { - if ((val_to_write ^ mstatus) & (MSTATUS_MXR | MSTATUS_MPP | - MSTATUS_MPRV | MSTATUS_SUM)) { - helper_tlb_flush(env); - } - mask = MSTATUS_SIE | MSTATUS_SPIE | MSTATUS_MIE | MSTATUS_MPIE | - MSTATUS_SPP | MSTATUS_FS | MSTATUS_MPRV | MSTATUS_SUM | - MSTATUS_MPP | MSTATUS_MXR; - } - - /* silenty discard mstatus.mpp writes for unsupported modes */ - if (mpp == PRV_H || - (!riscv_has_ext(env, RVS) && mpp == PRV_S) || - (!riscv_has_ext(env, RVU) && mpp == PRV_U)) { - mask &= ~MSTATUS_MPP; - } - - mstatus = (mstatus & ~mask) | (val_to_write & mask); - - /* Note: this is a workaround for an issue where mstatus.FS - does not report dirty after floating point operations - that modify floating point state. This workaround is - technically compliant with the RISC-V Privileged - specification as it is legal to return only off, or dirty. - at the expense of extra floating point save/restore. */ - - /* FP is always dirty or off */ - if (mstatus & MSTATUS_FS) { - mstatus |= MSTATUS_FS; - } - - int dirty = ((mstatus & MSTATUS_FS) == MSTATUS_FS) | - ((mstatus & MSTATUS_XS) == MSTATUS_XS); - mstatus = set_field(mstatus, MSTATUS_SD, dirty); - env->mstatus = mstatus; - break; - } - case CSR_MIP: { - /* - * Since the writeable bits in MIP are not set asynchrously by the - * CLINT, no additional locking is needed for read-modifiy-write - * CSR operations - */ - qemu_mutex_lock_iothread(); - RISCVCPU *cpu = riscv_env_get_cpu(env); - riscv_cpu_update_mip(cpu, MIP_SSIP | MIP_STIP, - (val_to_write & (MIP_SSIP | MIP_STIP))); - /* - * csrs, csrc on mip.SEIP is not decomposable into separate read and - * write steps, so a different implementation is needed - */ - qemu_mutex_unlock_iothread(); - break; - } - case CSR_MIE: { - env->mie = (env->mie & ~all_ints) | - (val_to_write & all_ints); - break; - } - case CSR_MIDELEG: - env->mideleg = (env->mideleg & ~delegable_ints) - | (val_to_write & delegable_ints); - break; - case CSR_MEDELEG: { - target_ulong mask = 0; - mask |= 1ULL << (RISCV_EXCP_INST_ADDR_MIS); - mask |= 1ULL << (RISCV_EXCP_INST_ACCESS_FAULT); - mask |= 1ULL << (RISCV_EXCP_ILLEGAL_INST); - mask |= 1ULL << (RISCV_EXCP_BREAKPOINT); - mask |= 1ULL << (RISCV_EXCP_LOAD_ADDR_MIS); - mask |= 1ULL << (RISCV_EXCP_LOAD_ACCESS_FAULT); - mask |= 1ULL << (RISCV_EXCP_STORE_AMO_ADDR_MIS); - mask |= 1ULL << (RISCV_EXCP_STORE_AMO_ACCESS_FAULT); - mask |= 1ULL << (RISCV_EXCP_U_ECALL); - mask |= 1ULL << (RISCV_EXCP_S_ECALL); - mask |= 1ULL << (RISCV_EXCP_H_ECALL); - mask |= 1ULL << (RISCV_EXCP_M_ECALL); - mask |= 1ULL << (RISCV_EXCP_INST_PAGE_FAULT); - mask |= 1ULL << (RISCV_EXCP_LOAD_PAGE_FAULT); - mask |= 1ULL << (RISCV_EXCP_STORE_PAGE_FAULT); - env->medeleg = (env->medeleg & ~mask) - | (val_to_write & mask); - break; - } - case CSR_MINSTRET: - /* minstret is WARL so unsupported writes are ignored */ - break; - case CSR_MCYCLE: - /* mcycle is WARL so unsupported writes are ignored */ - break; -#if defined(TARGET_RISCV32) - case CSR_MINSTRETH: - /* minstreth is WARL so unsupported writes are ignored */ - break; - case CSR_MCYCLEH: - /* mcycleh is WARL so unsupported writes are ignored */ - break; -#endif - case CSR_MUCOUNTEREN: - if (env->priv_ver <= PRIV_VERSION_1_09_1) { - env->scounteren = val_to_write; - break; - } else { - goto do_illegal; - } - case CSR_MSCOUNTEREN: - if (env->priv_ver <= PRIV_VERSION_1_09_1) { - env->mcounteren = val_to_write; - break; - } else { - goto do_illegal; - } - case CSR_SSTATUS: { - target_ulong ms = env->mstatus; - target_ulong mask = SSTATUS_SIE | SSTATUS_SPIE | SSTATUS_UIE - | SSTATUS_UPIE | SSTATUS_SPP | SSTATUS_FS | SSTATUS_XS - | SSTATUS_SUM | SSTATUS_SD; - if (env->priv_ver >= PRIV_VERSION_1_10_0) { - mask |= SSTATUS_MXR; - } - ms = (ms & ~mask) | (val_to_write & mask); - csr_write_helper(env, ms, CSR_MSTATUS); - break; - } - case CSR_SIP: { - qemu_mutex_lock_iothread(); - target_ulong next_mip = (env->mip & ~env->mideleg) - | (val_to_write & env->mideleg); - qemu_mutex_unlock_iothread(); - csr_write_helper(env, next_mip, CSR_MIP); - break; - } - case CSR_SIE: { - target_ulong next_mie = (env->mie & ~env->mideleg) - | (val_to_write & env->mideleg); - csr_write_helper(env, next_mie, CSR_MIE); - break; - } - case CSR_SATP: /* CSR_SPTBR */ { - if (!riscv_feature(env, RISCV_FEATURE_MMU)) { - break; - } - if (env->priv_ver <= PRIV_VERSION_1_09_1 && (val_to_write ^ env->sptbr)) - { - helper_tlb_flush(env); - env->sptbr = val_to_write & (((target_ulong) - 1 << (TARGET_PHYS_ADDR_SPACE_BITS - PGSHIFT)) - 1); - } - if (env->priv_ver >= PRIV_VERSION_1_10_0 && - validate_vm(env, get_field(val_to_write, SATP_MODE)) && - ((val_to_write ^ env->satp) & (SATP_MODE | SATP_ASID | SATP_PPN))) - { - helper_tlb_flush(env); - env->satp = val_to_write; - } - break; - } - case CSR_SEPC: - env->sepc = val_to_write; - break; - case CSR_STVEC: - /* bits [1:0] encode mode; 0 = direct, 1 = vectored, 2 >= reserved */ - if ((val_to_write & 3) == 0) { - env->stvec = val_to_write >> 2 << 2; - } else { - qemu_log_mask(LOG_UNIMP, - "CSR_STVEC: vectored traps not supported\n"); - } - break; - case CSR_SCOUNTEREN: - if (env->priv_ver >= PRIV_VERSION_1_10_0) { - env->scounteren = val_to_write; - break; - } else { - goto do_illegal; - } - case CSR_SSCRATCH: - env->sscratch = val_to_write; - break; - case CSR_SCAUSE: - env->scause = val_to_write; - break; - case CSR_SBADADDR: - env->sbadaddr = val_to_write; - break; - case CSR_MEPC: - env->mepc = val_to_write; - break; - case CSR_MTVEC: - /* bits [1:0] indicate mode; 0 = direct, 1 = vectored, 2 >= reserved */ - if ((val_to_write & 3) == 0) { - env->mtvec = val_to_write >> 2 << 2; - } else { - qemu_log_mask(LOG_UNIMP, - "CSR_MTVEC: vectored traps not supported\n"); - } - break; - case CSR_MCOUNTEREN: - if (env->priv_ver >= PRIV_VERSION_1_10_0) { - env->mcounteren = val_to_write; - break; - } else { - goto do_illegal; - } - case CSR_MSCRATCH: - env->mscratch = val_to_write; - break; - case CSR_MCAUSE: - env->mcause = val_to_write; - break; - case CSR_MBADADDR: - env->mbadaddr = val_to_write; - break; - case CSR_MISA: - /* misa is WARL so unsupported writes are ignored */ - break; - case CSR_PMPCFG0: - case CSR_PMPCFG1: - case CSR_PMPCFG2: - case CSR_PMPCFG3: - pmpcfg_csr_write(env, csrno - CSR_PMPCFG0, val_to_write); - break; - case CSR_PMPADDR0: - case CSR_PMPADDR1: - case CSR_PMPADDR2: - case CSR_PMPADDR3: - case CSR_PMPADDR4: - case CSR_PMPADDR5: - case CSR_PMPADDR6: - case CSR_PMPADDR7: - case CSR_PMPADDR8: - case CSR_PMPADDR9: - case CSR_PMPADDR10: - case CSR_PMPADDR11: - case CSR_PMPADDR12: - case CSR_PMPADDR13: - case CSR_PMPADDR14: - case CSR_PMPADDR15: - pmpaddr_csr_write(env, csrno - CSR_PMPADDR0, val_to_write); - break; -#endif -#if !defined(CONFIG_USER_ONLY) - do_illegal: -#endif - default: - do_raise_exception_err(env, RISCV_EXCP_ILLEGAL_INST, GETPC()); - } -} - -/* - * Handle reads to CSRs and any resulting special behavior - * - * Adapted from Spike's processor_t::get_csr - */ -target_ulong csr_read_helper(CPURISCVState *env, target_ulong csrno) -{ -#ifndef CONFIG_USER_ONLY - target_ulong ctr_en = env->priv == PRV_U ? env->scounteren : - env->priv == PRV_S ? env->mcounteren : -1U; -#else - target_ulong ctr_en = -1; -#endif - target_ulong ctr_ok = (ctr_en >> (csrno & 31)) & 1; - - if (csrno >= CSR_HPMCOUNTER3 && csrno <= CSR_HPMCOUNTER31) { - if (ctr_ok) { - return 0; - } - } -#if defined(TARGET_RISCV32) - if (csrno >= CSR_HPMCOUNTER3H && csrno <= CSR_HPMCOUNTER31H) { - if (ctr_ok) { - return 0; - } - } -#endif - if (csrno >= CSR_MHPMCOUNTER3 && csrno <= CSR_MHPMCOUNTER31) { - return 0; - } -#if defined(TARGET_RISCV32) - if (csrno >= CSR_MHPMCOUNTER3 && csrno <= CSR_MHPMCOUNTER31) { - return 0; - } -#endif - if (csrno >= CSR_MHPMEVENT3 && csrno <= CSR_MHPMEVENT31) { - return 0; - } - - switch (csrno) { - case CSR_FFLAGS: - validate_mstatus_fs(env, GETPC()); - return cpu_riscv_get_fflags(env); - case CSR_FRM: - validate_mstatus_fs(env, GETPC()); - return env->frm; - case CSR_FCSR: - validate_mstatus_fs(env, GETPC()); - return (cpu_riscv_get_fflags(env) << FSR_AEXC_SHIFT) - | (env->frm << FSR_RD_SHIFT); - /* rdtime/rdtimeh is trapped and emulated by bbl in system mode */ -#ifdef CONFIG_USER_ONLY - case CSR_TIME: - return cpu_get_host_ticks(); -#if defined(TARGET_RISCV32) - case CSR_TIMEH: - return cpu_get_host_ticks() >> 32; -#endif -#endif - case CSR_INSTRET: - case CSR_CYCLE: - if (ctr_ok) { -#if !defined(CONFIG_USER_ONLY) - if (use_icount) { - return cpu_get_icount(); - } else { - return cpu_get_host_ticks(); - } -#else - return cpu_get_host_ticks(); -#endif - } - break; -#if defined(TARGET_RISCV32) - case CSR_INSTRETH: - case CSR_CYCLEH: - if (ctr_ok) { -#if !defined(CONFIG_USER_ONLY) - if (use_icount) { - return cpu_get_icount() >> 32; - } else { - return cpu_get_host_ticks() >> 32; - } -#else - return cpu_get_host_ticks() >> 32; -#endif - } - break; -#endif -#ifndef CONFIG_USER_ONLY - case CSR_MINSTRET: - case CSR_MCYCLE: - if (use_icount) { - return cpu_get_icount(); - } else { - return cpu_get_host_ticks(); - } - case CSR_MINSTRETH: - case CSR_MCYCLEH: -#if defined(TARGET_RISCV32) - if (use_icount) { - return cpu_get_icount() >> 32; - } else { - return cpu_get_host_ticks() >> 32; - } -#endif - break; - case CSR_MUCOUNTEREN: - if (env->priv_ver <= PRIV_VERSION_1_09_1) { - return env->scounteren; - } else { - break; /* illegal instruction */ - } - case CSR_MSCOUNTEREN: - if (env->priv_ver <= PRIV_VERSION_1_09_1) { - return env->mcounteren; - } else { - break; /* illegal instruction */ - } - case CSR_SSTATUS: { - target_ulong mask = SSTATUS_SIE | SSTATUS_SPIE | SSTATUS_UIE - | SSTATUS_UPIE | SSTATUS_SPP | SSTATUS_FS | SSTATUS_XS - | SSTATUS_SUM | SSTATUS_SD; - if (env->priv_ver >= PRIV_VERSION_1_10_0) { - mask |= SSTATUS_MXR; - } - return env->mstatus & mask; - } - case CSR_SIP: { - qemu_mutex_lock_iothread(); - target_ulong tmp = env->mip & env->mideleg; - qemu_mutex_unlock_iothread(); - return tmp; - } - case CSR_SIE: - return env->mie & env->mideleg; - case CSR_SEPC: - return env->sepc; - case CSR_SBADADDR: - return env->sbadaddr; - case CSR_STVEC: - return env->stvec; - case CSR_SCOUNTEREN: - if (env->priv_ver >= PRIV_VERSION_1_10_0) { - return env->scounteren; - } else { - break; /* illegal instruction */ - } - case CSR_SCAUSE: - return env->scause; - case CSR_SATP: /* CSR_SPTBR */ - if (!riscv_feature(env, RISCV_FEATURE_MMU)) { - return 0; - } - if (env->priv_ver >= PRIV_VERSION_1_10_0) { - return env->satp; - } else { - return env->sptbr; - } - case CSR_SSCRATCH: - return env->sscratch; - case CSR_MSTATUS: - return env->mstatus; - case CSR_MIP: { - qemu_mutex_lock_iothread(); - target_ulong tmp = env->mip; - qemu_mutex_unlock_iothread(); - return tmp; - } - case CSR_MIE: - return env->mie; - case CSR_MEPC: - return env->mepc; - case CSR_MSCRATCH: - return env->mscratch; - case CSR_MCAUSE: - return env->mcause; - case CSR_MBADADDR: - return env->mbadaddr; - case CSR_MISA: - return env->misa; - case CSR_MARCHID: - return 0; /* as spike does */ - case CSR_MIMPID: - return 0; /* as spike does */ - case CSR_MVENDORID: - return 0; /* as spike does */ - case CSR_MHARTID: - return env->mhartid; - case CSR_MTVEC: - return env->mtvec; - case CSR_MCOUNTEREN: - if (env->priv_ver >= PRIV_VERSION_1_10_0) { - return env->mcounteren; - } else { - break; /* illegal instruction */ - } - case CSR_MEDELEG: - return env->medeleg; - case CSR_MIDELEG: - return env->mideleg; - case CSR_PMPCFG0: - case CSR_PMPCFG1: - case CSR_PMPCFG2: - case CSR_PMPCFG3: - return pmpcfg_csr_read(env, csrno - CSR_PMPCFG0); - case CSR_PMPADDR0: - case CSR_PMPADDR1: - case CSR_PMPADDR2: - case CSR_PMPADDR3: - case CSR_PMPADDR4: - case CSR_PMPADDR5: - case CSR_PMPADDR6: - case CSR_PMPADDR7: - case CSR_PMPADDR8: - case CSR_PMPADDR9: - case CSR_PMPADDR10: - case CSR_PMPADDR11: - case CSR_PMPADDR12: - case CSR_PMPADDR13: - case CSR_PMPADDR14: - case CSR_PMPADDR15: - return pmpaddr_csr_read(env, csrno - CSR_PMPADDR0); -#endif - } - /* used by e.g. MTIME read */ - do_raise_exception_err(env, RISCV_EXCP_ILLEGAL_INST, GETPC()); -} - -/* - * Check that CSR access is allowed. - * - * Adapted from Spike's decode.h:validate_csr - */ -static void validate_csr(CPURISCVState *env, uint64_t which, - uint64_t write, uintptr_t ra) -{ -#ifndef CONFIG_USER_ONLY - unsigned csr_priv = get_field((which), 0x300); - unsigned csr_read_only = get_field((which), 0xC00) == 3; - if (((write) && csr_read_only) || (env->priv < csr_priv)) { - do_raise_exception_err(env, RISCV_EXCP_ILLEGAL_INST, ra); - } -#endif -} - target_ulong helper_csrrw(CPURISCVState *env, target_ulong src, target_ulong csr) { - validate_csr(env, csr, 1, GETPC()); - uint64_t csr_backup = csr_read_helper(env, csr); - csr_write_helper(env, src, csr); - return csr_backup; + target_ulong val = 0; + if (riscv_csrrw(env, csr, &val, src, -1) < 0) { + do_raise_exception_err(env, RISCV_EXCP_ILLEGAL_INST, GETPC()); + } + return val; } target_ulong helper_csrrs(CPURISCVState *env, target_ulong src, target_ulong csr, target_ulong rs1_pass) { - validate_csr(env, csr, rs1_pass != 0, GETPC()); - uint64_t csr_backup = csr_read_helper(env, csr); - if (rs1_pass != 0) { - csr_write_helper(env, src | csr_backup, csr); + target_ulong val = 0; + if (riscv_csrrw(env, csr, &val, -1, rs1_pass ? src : 0) < 0) { + do_raise_exception_err(env, RISCV_EXCP_ILLEGAL_INST, GETPC()); } - return csr_backup; + return val; } target_ulong helper_csrrc(CPURISCVState *env, target_ulong src, target_ulong csr, target_ulong rs1_pass) { - validate_csr(env, csr, rs1_pass != 0, GETPC()); - uint64_t csr_backup = csr_read_helper(env, csr); - if (rs1_pass != 0) { - csr_write_helper(env, (~src) & csr_backup, csr); + target_ulong val = 0; + if (riscv_csrrw(env, csr, &val, 0, rs1_pass ? src : 0) < 0) { + do_raise_exception_err(env, RISCV_EXCP_ILLEGAL_INST, GETPC()); } - return csr_backup; + return val; } #ifndef CONFIG_USER_ONLY @@ -674,7 +91,7 @@ target_ulong helper_sret(CPURISCVState *env, target_ulong cpu_pc_deb) mstatus = set_field(mstatus, MSTATUS_SPIE, 0); mstatus = set_field(mstatus, MSTATUS_SPP, PRV_U); riscv_set_mode(env, prev_priv); - csr_write_helper(env, mstatus, CSR_MSTATUS); + env->mstatus = mstatus; return retpc; } @@ -699,7 +116,7 @@ target_ulong helper_mret(CPURISCVState *env, target_ulong cpu_pc_deb) mstatus = set_field(mstatus, MSTATUS_MPIE, 0); mstatus = set_field(mstatus, MSTATUS_MPP, PRV_U); riscv_set_mode(env, prev_priv); - csr_write_helper(env, mstatus, CSR_MSTATUS); + env->mstatus = mstatus; return retpc; } diff --git a/tests/Makefile.include b/tests/Makefile.include index c17f6d5dfa..f403a6571d 100644 --- a/tests/Makefile.include +++ b/tests/Makefile.include @@ -222,18 +222,23 @@ check-qtest-x86_64-y += $(check-qtest-i386-y) check-qtest-x86_64-$(CONFIG_SDHCI) += tests/sdhci-test$(EXESUF) check-qtest-alpha-y += tests/boot-serial-test$(EXESUF) +check-qtest-alpha-$(CONFIG_VGA) += tests/display-vga-test$(EXESUF) check-qtest-hppa-y += tests/boot-serial-test$(EXESUF) +check-qtest-hppa-$(CONFIG_VGA) += tests/display-vga-test$(EXESUF) check-qtest-m68k-y = tests/boot-serial-test$(EXESUF) check-qtest-microblaze-y += tests/boot-serial-test$(EXESUF) check-qtest-mips-$(CONFIG_ISA_TESTDEV) = tests/endianness-test$(EXESUF) +check-qtest-mips-$(CONFIG_VGA) += tests/display-vga-test$(EXESUF) check-qtest-mips64-$(CONFIG_ISA_TESTDEV) = tests/endianness-test$(EXESUF) +check-qtest-mips64-$(CONFIG_VGA) += tests/display-vga-test$(EXESUF) check-qtest-mips64el-$(CONFIG_ISA_TESTDEV) = tests/endianness-test$(EXESUF) +check-qtest-mips64el-$(CONFIG_VGA) += tests/display-vga-test$(EXESUF) check-qtest-moxie-y += tests/boot-serial-test$(EXESUF) diff --git a/tests/atomic64-bench.c b/tests/atomic64-bench.c index 71692560ed..121a8c14f4 100644 --- a/tests/atomic64-bench.c +++ b/tests/atomic64-bench.c @@ -74,16 +74,14 @@ static void *thread_func(void *arg) static void run_test(void) { - unsigned int remaining; unsigned int i; while (atomic_read(&n_ready_threads) != n_threads) { cpu_relax(); } + atomic_set(&test_start, true); - do { - remaining = sleep(duration); - } while (remaining); + g_usleep(duration * G_USEC_PER_SEC); atomic_set(&test_stop, true); for (i = 0; i < n_threads; i++) { diff --git a/tests/atomic_add-bench.c b/tests/atomic_add-bench.c index 2f6c72f63a..5666f6bbff 100644 --- a/tests/atomic_add-bench.c +++ b/tests/atomic_add-bench.c @@ -76,16 +76,14 @@ static void *thread_func(void *arg) static void run_test(void) { - unsigned int remaining; unsigned int i; while (atomic_read(&n_ready_threads) != n_threads) { cpu_relax(); } + atomic_set(&test_start, true); - do { - remaining = sleep(duration); - } while (remaining); + g_usleep(duration * G_USEC_PER_SEC); atomic_set(&test_stop, true); for (i = 0; i < n_threads; i++) { diff --git a/tests/display-vga-test.c b/tests/display-vga-test.c index 2d7d24eee0..bd176dcf3a 100644 --- a/tests/display-vga-test.c +++ b/tests/display-vga-test.c @@ -40,13 +40,11 @@ static void pci_virtio_gpu(void) qtest_end(); } -#ifdef CONFIG_VIRTIO_VGA static void pci_virtio_vga(void) { qtest_start("-vga none -device virtio-vga"); qtest_end(); } -#endif int main(int argc, char **argv) { @@ -62,8 +60,10 @@ int main(int argc, char **argv) qtest_add_func("/display/pci/secondary", pci_secondary); qtest_add_func("/display/pci/multihead", pci_multihead); qtest_add_func("/display/pci/virtio-gpu", pci_virtio_gpu); -#ifdef CONFIG_VIRTIO_VGA - qtest_add_func("/display/pci/virtio-vga", pci_virtio_vga); -#endif + if (g_str_equal(arch, "i386") || g_str_equal(arch, "x86_64") || + g_str_equal(arch, "hppa") || g_str_equal(arch, "ppc64")) { + qtest_add_func("/display/pci/virtio-vga", pci_virtio_vga); + } + return g_test_run(); } diff --git a/tests/docker/Makefile.include b/tests/docker/Makefile.include index 9467e9d088..7032c68895 100644 --- a/tests/docker/Makefile.include +++ b/tests/docker/Makefile.include @@ -98,19 +98,6 @@ docker-image-debian-s390x-cross: docker-image-debian9 docker-image-debian-win32-cross: docker-image-debian8-mxe docker-image-debian-win64-cross: docker-image-debian8-mxe -# Debian SID images - we are tracking a rolling distro so we want to -# force a re-build of the base image if we ever need to build one of -# its children. -ifndef SKIP_DOCKER_BUILD -ifeq ($(HAVE_USER_DOCKER),y) -SID_AGE=$(shell $(DOCKER_SCRIPT) check --checktype=age --olderthan=180 --quiet qemu:debian-sid) -ifeq ($(SID_AGE),) -else -docker-image-debian-sid: NOCACHE=1 -endif -endif -endif - docker-image-debian-alpha-cross: docker-image-debian-sid docker-image-debian-hppa-cross: docker-image-debian-sid docker-image-debian-m68k-cross: docker-image-debian-sid diff --git a/tests/docker/dockerfiles/debian-amd64.docker b/tests/docker/dockerfiles/debian-amd64.docker index 24b113b76f..954fcf9606 100644 --- a/tests/docker/dockerfiles/debian-amd64.docker +++ b/tests/docker/dockerfiles/debian-amd64.docker @@ -24,7 +24,8 @@ RUN DEBIAN_FRONTEND=noninteractive eatmydata \ libegl1-mesa-dev \ libepoxy-dev \ libgbm-dev -RUN git clone https://anongit.freedesktop.org/git/virglrenderer.git /usr/src/virglrenderer +RUN git clone https://anongit.freedesktop.org/git/virglrenderer.git /usr/src/virglrenderer && \ + cd /usr/src/virglrenderer && git checkout virglrenderer-0.7.0 RUN cd /usr/src/virglrenderer && ./autogen.sh && ./configure --with-glx --disable-tests && make install # netmap @@ -35,5 +36,7 @@ RUN git clone https://github.com/luigirizzo/netmap.git /usr/src/netmap RUN cd /usr/src/netmap/LINUX && ./configure --no-drivers --no-apps --kernel-dir=$(ls -d /usr/src/linux-headers-*-amd64) && make install ENV QEMU_CONFIGURE_OPTS --enable-netmap +RUN ldconfig + # gcrypt ENV QEMU_CONFIGURE_OPTS $QEMU_CONFIGURE_OPTS --enable-gcrypt diff --git a/tests/docker/dockerfiles/debian-sid.docker b/tests/docker/dockerfiles/debian-sid.docker index 4e4cda0ba5..676941cb32 100644 --- a/tests/docker/dockerfiles/debian-sid.docker +++ b/tests/docker/dockerfiles/debian-sid.docker @@ -11,7 +11,12 @@ # updated and trigger a re-build. # -FROM debian:sid-slim +# This must be earlier than the snapshot date we are aiming for +FROM debian:sid-20181011-slim + +# Use a snapshot known to work (see http://snapshot.debian.org/#Usage) +ENV DEBIAN_SNAPSHOT_DATE "20181030" +RUN sed -i "s%^deb \(https\?://\)deb.debian.org/debian/\? \(.*\)%deb [check-valid-until=no] \1snapshot.debian.org/archive/debian/${DEBIAN_SNAPSHOT_DATE} \2%" /etc/apt/sources.list # Use a snapshot known to work (see http://snapshot.debian.org/#Usage) ENV DEBIAN_SNAPSHOT_DATE "20181030" diff --git a/tests/docker/dockerfiles/debian.docker b/tests/docker/dockerfiles/debian.docker deleted file mode 100644 index fd32e71b79..0000000000 --- a/tests/docker/dockerfiles/debian.docker +++ /dev/null @@ -1,13 +0,0 @@ -# This template is deprecated and was previously based on Jessie on QEMU 2.9. -# Now than Stretch is out, please use qemu:debian8 as base for Jessie, -# and qemu:debian9 for Stretch. -# -FROM qemu:debian9 - -MAINTAINER Philippe Mathieu-Daudé <f4bug@amsat.org> - -RUN for n in $(seq 8); do echo; done && \ - echo "\n\t\tThis image is deprecated." && echo && \ - echo "\tUse 'FROM qemu:debian9' to use the stable Debian Stretch image" && \ - echo "\tor 'FROM qemu:debian8' to use old Debian Jessie." && \ - for n in $(seq 8); do echo; done diff --git a/tests/docker/dockerfiles/fedora-i386-cross.docker b/tests/docker/dockerfiles/fedora-i386-cross.docker index a4fd895b07..eb8108d118 100644 --- a/tests/docker/dockerfiles/fedora-i386-cross.docker +++ b/tests/docker/dockerfiles/fedora-i386-cross.docker @@ -1,4 +1,4 @@ -FROM fedora:latest +FROM fedora:29 ENV PACKAGES \ gcc \ glib2-devel.i686 \ diff --git a/tests/docker/dockerfiles/fedora.docker b/tests/docker/dockerfiles/fedora.docker index 1d0e3dc4ec..69d4a7f5d7 100644 --- a/tests/docker/dockerfiles/fedora.docker +++ b/tests/docker/dockerfiles/fedora.docker @@ -1,4 +1,4 @@ -FROM fedora:28 +FROM fedora:29 ENV PACKAGES \ bc \ bison \ @@ -82,7 +82,7 @@ ENV PACKAGES \ tar \ usbredir-devel \ virglrenderer-devel \ - vte3-devel \ + vte291-devel \ which \ xen-devel \ zlib-devel diff --git a/tests/docker/dockerfiles/travis.docker b/tests/docker/dockerfiles/travis.docker index 03ebfb0ef2..e72dc85ca7 100644 --- a/tests/docker/dockerfiles/travis.docker +++ b/tests/docker/dockerfiles/travis.docker @@ -1,8 +1,8 @@ -FROM travisci/ci-garnet:packer-1512502276-986baf0 +FROM travisci/ci-sardonyx:packer-1546978056-2c98a19 ENV DEBIAN_FRONTEND noninteractive ENV LANG en_US.UTF-8 ENV LC_ALL en_US.UTF-8 -RUN cat /etc/apt/sources.list | sed "s/# deb-src/deb-src/" >> /etc/apt/sources.list +RUN sed -i "s/# deb-src/deb-src/" /etc/apt/sources.list RUN apt-get update RUN apt-get -y build-dep qemu RUN apt-get -y install device-tree-compiler python2.7 python-yaml dh-autoreconf gdb strace lsof net-tools gcovr diff --git a/tests/qemu-iotests/206 b/tests/qemu-iotests/206 index 128c334c7c..5bb738bf23 100755 --- a/tests/qemu-iotests/206 +++ b/tests/qemu-iotests/206 @@ -26,7 +26,9 @@ from iotests import imgfmt iotests.verify_image_format(supported_fmts=['qcow2']) def blockdev_create(vm, options): - result = vm.qmp_log('blockdev-create', job_id='job0', options=options) + result = vm.qmp_log('blockdev-create', + filters=[iotests.filter_qmp_testfiles], + job_id='job0', options=options) if 'return' in result: assert result['return'] == {} @@ -52,7 +54,9 @@ with iotests.FilePath('t.qcow2') as disk_path, \ 'filename': disk_path, 'size': 0 }) - vm.qmp_log('blockdev-add', driver='file', filename=disk_path, + vm.qmp_log('blockdev-add', + filters=[iotests.filter_qmp_testfiles], + driver='file', filename=disk_path, node_name='imgfile') blockdev_create(vm, { 'driver': imgfmt, diff --git a/tests/qemu-iotests/223 b/tests/qemu-iotests/223 index 397b865d34..773892dbe6 100755 --- a/tests/qemu-iotests/223 +++ b/tests/qemu-iotests/223 @@ -25,6 +25,7 @@ status=1 # failure is the default! _cleanup() { + nbd_server_stop _cleanup_test_img _cleanup_qemu rm -f "$TEST_DIR/nbd" @@ -35,6 +36,7 @@ trap "_cleanup; exit \$status" 0 1 2 3 15 . ./common.rc . ./common.filter . ./common.qemu +. ./common.nbd _supported_fmt qcow2 _supported_proto file # uses NBD as well @@ -61,6 +63,8 @@ echo "=== Create partially sparse image, then add dirty bitmaps ===" echo # Two bitmaps, to contrast granularity issues +# Also note that b will be disabled, while b2 is left enabled, to +# check for read-only interactions _make_test_img -o cluster_size=4k 4M $QEMU_IO -c 'w -P 0x11 1M 2M' "$TEST_IMG" | _filter_qemu_io run_qemu <<EOF @@ -107,26 +111,37 @@ echo _launch_qemu 2> >(_filter_nbd) +# Intentionally provoke some errors as well, to check error handling silent= _send_qemu_cmd $QEMU_HANDLE '{"execute":"qmp_capabilities"}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"blockdev-add", "arguments":{"driver":"qcow2", "node-name":"n", "file":{"driver":"file", "filename":"'"$TEST_IMG"'"}}}' "return" -_send_qemu_cmd $QEMU_HANDLE '{"execute":"x-block-dirty-bitmap-disable", +_send_qemu_cmd $QEMU_HANDLE '{"execute":"block-dirty-bitmap-disable", "arguments":{"node":"n", "name":"b"}}' "return" -_send_qemu_cmd $QEMU_HANDLE '{"execute":"x-block-dirty-bitmap-disable", - "arguments":{"node":"n", "name":"b2"}}' "return" +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", + "arguments":{"device":"n"}}' "error" # Attempt add without server _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-start", "arguments":{"addr":{"type":"unix", "data":{"path":"'"$TEST_DIR/nbd"'"}}}}' "return" +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-start", + "arguments":{"addr":{"type":"unix", + "data":{"path":"'"$TEST_DIR/nbd"1'"}}}}' "error" # Attempt second server +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", + "arguments":{"device":"n", "bitmap":"b"}}' "return" +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", + "arguments":{"device":"nosuch"}}' "error" # Attempt to export missing node _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", - "arguments":{"device":"n"}}' "return" -_send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", - "arguments":{"name":"n", "bitmap":"b"}}' "return" + "arguments":{"device":"n"}}' "error" # Attempt to export same name twice _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", - "arguments":{"device":"n", "name":"n2"}}' "return" -_send_qemu_cmd $QEMU_HANDLE '{"execute":"x-nbd-server-add-bitmap", - "arguments":{"name":"n2", "bitmap":"b2"}}' "return" + "arguments":{"device":"n", "name":"n2", + "bitmap":"b2"}}' "error" # enabled vs. read-only +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", + "arguments":{"device":"n", "name":"n2", + "bitmap":"b3"}}' "error" # Missing bitmap +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-add", + "arguments":{"device":"n", "name":"n2", "writable":true, + "bitmap":"b2"}}' "return" echo echo "=== Contrast normal status to large granularity dirty-bitmap ===" @@ -150,16 +165,33 @@ $QEMU_IMG map --output=json --image-opts \ "$IMG,x-dirty-bitmap=qemu:dirty-bitmap:b2" | _filter_qemu_img_map echo -echo "=== End NBD server ===" +echo "=== End qemu NBD server ===" echo _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", "arguments":{"name":"n"}}' "return" _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", "arguments":{"name":"n2"}}' "return" +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-remove", + "arguments":{"name":"n2"}}' "error" # Attempt duplicate clean _send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-stop"}' "return" +_send_qemu_cmd $QEMU_HANDLE '{"execute":"nbd-server-stop"}' "error" # Again _send_qemu_cmd $QEMU_HANDLE '{"execute":"quit"}' "return" +echo +echo "=== Use qemu-nbd as server ===" +echo + +nbd_server_start_unix_socket -r -f $IMGFMT -B b "$TEST_IMG" +IMG="driver=nbd,server.type=unix,server.path=$nbd_unix_socket" +$QEMU_IMG map --output=json --image-opts \ + "$IMG,x-dirty-bitmap=qemu:dirty-bitmap:b" | _filter_qemu_img_map + +nbd_server_start_unix_socket -f $IMGFMT -B b2 "$TEST_IMG" +IMG="driver=nbd,server.type=unix,server.path=$nbd_unix_socket" +$QEMU_IMG map --output=json --image-opts \ + "$IMG,x-dirty-bitmap=qemu:dirty-bitmap:b2" | _filter_qemu_img_map + # success, all done echo '*** done' rm -f $seq.full diff --git a/tests/qemu-iotests/223.out b/tests/qemu-iotests/223.out index 99ca172fbb..0de5240a75 100644 --- a/tests/qemu-iotests/223.out +++ b/tests/qemu-iotests/223.out @@ -27,11 +27,14 @@ wrote 2097152/2097152 bytes at offset 2097152 {"return": {}} {"return": {}} {"return": {}} +{"error": {"class": "GenericError", "desc": "NBD server not running"}} {"return": {}} +{"error": {"class": "GenericError", "desc": "NBD server already running"}} {"return": {}} -{"return": {}} -{"return": {}} -{"return": {}} +{"error": {"class": "GenericError", "desc": "Cannot find device=nosuch nor node_name=nosuch"}} +{"error": {"class": "GenericError", "desc": "NBD server already has export named 'n'"}} +{"error": {"class": "GenericError", "desc": "Enabled bitmap 'b2' incompatible with readonly export"}} +{"error": {"class": "GenericError", "desc": "Bitmap 'b3' is not found"}} {"return": {}} === Contrast normal status to large granularity dirty-bitmap === @@ -58,10 +61,22 @@ read 2097152/2097152 bytes at offset 2097152 { "start": 1024, "length": 2096128, "depth": 0, "zero": false, "data": true}, { "start": 2097152, "length": 2097152, "depth": 0, "zero": false, "data": false}] -=== End NBD server === +=== End qemu NBD server === {"return": {}} {"return": {}} +{"error": {"class": "GenericError", "desc": "Export 'n2' is not found"}} {"return": {}} +{"error": {"class": "GenericError", "desc": "NBD server not running"}} {"return": {}} + +=== Use qemu-nbd as server === + +[{ "start": 0, "length": 65536, "depth": 0, "zero": false, "data": false}, +{ "start": 65536, "length": 2031616, "depth": 0, "zero": false, "data": true}, +{ "start": 2097152, "length": 2097152, "depth": 0, "zero": false, "data": false}] +[{ "start": 0, "length": 512, "depth": 0, "zero": false, "data": true}, +{ "start": 512, "length": 512, "depth": 0, "zero": false, "data": false}, +{ "start": 1024, "length": 2096128, "depth": 0, "zero": false, "data": true}, +{ "start": 2097152, "length": 2097152, "depth": 0, "zero": false, "data": false}] *** done diff --git a/tests/qemu-iotests/236 b/tests/qemu-iotests/236 new file mode 100755 index 0000000000..79a6381f8e --- /dev/null +++ b/tests/qemu-iotests/236 @@ -0,0 +1,161 @@ +#!/usr/bin/env python +# +# Test bitmap merges. +# +# Copyright (c) 2018 John Snow for Red Hat, Inc. +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see <http://www.gnu.org/licenses/>. +# +# owner=jsnow@redhat.com + +import iotests +from iotests import log + +iotests.verify_image_format(supported_fmts=['generic']) +size = 64 * 1024 * 1024 +granularity = 64 * 1024 + +patterns = [("0x5d", "0", "64k"), + ("0xd5", "1M", "64k"), + ("0xdc", "32M", "64k"), + ("0xcd", "0x3ff0000", "64k")] # 64M - 64K + +overwrite = [("0xab", "0", "64k"), # Full overwrite + ("0xad", "0x00f8000", "64k"), # Partial-left (1M-32K) + ("0x1d", "0x2008000", "64k"), # Partial-right (32M+32K) + ("0xea", "0x3fe0000", "64k")] # Adjacent-left (64M - 128K) + +def query_bitmaps(vm): + res = vm.qmp("query-block") + return { "bitmaps": { device['device']: device.get('dirty-bitmaps', []) for + device in res['return'] } } + +with iotests.FilePath('img') as img_path, \ + iotests.VM() as vm: + + log('--- Preparing image & VM ---\n') + iotests.qemu_img_create('-f', iotests.imgfmt, img_path, str(size)) + vm.add_drive(img_path) + vm.launch() + + log('\n--- Adding preliminary bitmaps A & B ---\n') + vm.qmp_log("block-dirty-bitmap-add", node="drive0", + name="bitmapA", granularity=granularity) + vm.qmp_log("block-dirty-bitmap-add", node="drive0", + name="bitmapB", granularity=granularity) + + # Dirties 4 clusters. count=262144 + log('\n--- Emulating writes ---\n') + for p in patterns: + cmd = "write -P%s %s %s" % p + log(cmd) + log(vm.hmp_qemu_io("drive0", cmd)) + + log(query_bitmaps(vm), indent=2) + + log('\n--- Submitting & Aborting Transaction ---\n') + vm.qmp_log("transaction", indent=2, actions=[ + { "type": "block-dirty-bitmap-disable", + "data": { "node": "drive0", "name": "bitmapB" }}, + { "type": "block-dirty-bitmap-add", + "data": { "node": "drive0", "name": "bitmapC", + "granularity": granularity }}, + { "type": "block-dirty-bitmap-clear", + "data": { "node": "drive0", "name": "bitmapA" }}, + { "type": "abort", "data": {}} + ]) + log(query_bitmaps(vm), indent=2) + + log('\n--- Disabling B & Adding C ---\n') + vm.qmp_log("transaction", indent=2, actions=[ + { "type": "block-dirty-bitmap-disable", + "data": { "node": "drive0", "name": "bitmapB" }}, + { "type": "block-dirty-bitmap-add", + "data": { "node": "drive0", "name": "bitmapC", + "granularity": granularity }}, + # Purely extraneous, but test that it works: + { "type": "block-dirty-bitmap-disable", + "data": { "node": "drive0", "name": "bitmapC" }}, + { "type": "block-dirty-bitmap-enable", + "data": { "node": "drive0", "name": "bitmapC" }}, + ]) + + log('\n--- Emulating further writes ---\n') + # Dirties 6 clusters, 3 of which are new in contrast to "A". + # A = 64 * 1024 * (4 + 3) = 458752 + # C = 64 * 1024 * 6 = 393216 + for p in overwrite: + cmd = "write -P%s %s %s" % p + log(cmd) + log(vm.hmp_qemu_io("drive0", cmd)) + + log('\n--- Disabling A & C ---\n') + vm.qmp_log("transaction", indent=2, actions=[ + { "type": "block-dirty-bitmap-disable", + "data": { "node": "drive0", "name": "bitmapA" }}, + { "type": "block-dirty-bitmap-disable", + "data": { "node": "drive0", "name": "bitmapC" }} + ]) + + # A: 7 clusters + # B: 4 clusters + # C: 6 clusters + log(query_bitmaps(vm), indent=2) + + log('\n--- Submitting & Aborting Merge Transaction ---\n') + vm.qmp_log("transaction", indent=2, actions=[ + { "type": "block-dirty-bitmap-add", + "data": { "node": "drive0", "name": "bitmapD", + "disabled": True, "granularity": granularity }}, + { "type": "block-dirty-bitmap-merge", + "data": { "node": "drive0", "target": "bitmapD", + "bitmaps": ["bitmapB", "bitmapC"] }}, + { "type": "abort", "data": {}} + ]) + log(query_bitmaps(vm), indent=2) + + log('\n--- Creating D as a merge of B & C ---\n') + # Good hygiene: create a disabled bitmap as a merge target. + vm.qmp_log("transaction", indent=2, actions=[ + { "type": "block-dirty-bitmap-add", + "data": { "node": "drive0", "name": "bitmapD", + "disabled": True, "granularity": granularity }}, + { "type": "block-dirty-bitmap-merge", + "data": { "node": "drive0", "target": "bitmapD", + "bitmaps": ["bitmapB", "bitmapC"] }} + ]) + + # A and D should now both have 7 clusters apiece. + # B and C remain unchanged with 4 and 6 respectively. + log(query_bitmaps(vm), indent=2) + + # A and D should be equivalent. + # Some formats round the size of the disk, so don't print the checksums. + check_a = vm.qmp('x-debug-block-dirty-bitmap-sha256', + node="drive0", name="bitmapA")['return']['sha256'] + check_d = vm.qmp('x-debug-block-dirty-bitmap-sha256', + node="drive0", name="bitmapD")['return']['sha256'] + assert(check_a == check_d) + + log('\n--- Removing bitmaps A, B, C, and D ---\n') + vm.qmp_log("block-dirty-bitmap-remove", node="drive0", name="bitmapA") + vm.qmp_log("block-dirty-bitmap-remove", node="drive0", name="bitmapB") + vm.qmp_log("block-dirty-bitmap-remove", node="drive0", name="bitmapC") + vm.qmp_log("block-dirty-bitmap-remove", node="drive0", name="bitmapD") + + log('\n--- Final Query ---\n') + log(query_bitmaps(vm), indent=2) + + log('\n--- Done ---\n') + vm.shutdown() diff --git a/tests/qemu-iotests/236.out b/tests/qemu-iotests/236.out new file mode 100644 index 0000000000..1dad24db0d --- /dev/null +++ b/tests/qemu-iotests/236.out @@ -0,0 +1,351 @@ +--- Preparing image & VM --- + + +--- Adding preliminary bitmaps A & B --- + +{"execute": "block-dirty-bitmap-add", "arguments": {"granularity": 65536, "name": "bitmapA", "node": "drive0"}} +{"return": {}} +{"execute": "block-dirty-bitmap-add", "arguments": {"granularity": 65536, "name": "bitmapB", "node": "drive0"}} +{"return": {}} + +--- Emulating writes --- + +write -P0x5d 0 64k +{"return": ""} +write -P0xd5 1M 64k +{"return": ""} +write -P0xdc 32M 64k +{"return": ""} +write -P0xcd 0x3ff0000 64k +{"return": ""} +{ + "bitmaps": { + "drive0": [ + { + "count": 262144, + "granularity": 65536, + "name": "bitmapB", + "status": "active" + }, + { + "count": 262144, + "granularity": 65536, + "name": "bitmapA", + "status": "active" + } + ] + } +} + +--- Submitting & Aborting Transaction --- + +{ + "execute": "transaction", + "arguments": { + "actions": [ + { + "data": { + "node": "drive0", + "name": "bitmapB" + }, + "type": "block-dirty-bitmap-disable" + }, + { + "data": { + "node": "drive0", + "name": "bitmapC", + "granularity": 65536 + }, + "type": "block-dirty-bitmap-add" + }, + { + "data": { + "node": "drive0", + "name": "bitmapA" + }, + "type": "block-dirty-bitmap-clear" + }, + { + "data": {}, + "type": "abort" + } + ] + } +} +{ + "error": { + "class": "GenericError", + "desc": "Transaction aborted using Abort action" + } +} +{ + "bitmaps": { + "drive0": [ + { + "count": 262144, + "granularity": 65536, + "name": "bitmapB", + "status": "active" + }, + { + "count": 262144, + "granularity": 65536, + "name": "bitmapA", + "status": "active" + } + ] + } +} + +--- Disabling B & Adding C --- + +{ + "execute": "transaction", + "arguments": { + "actions": [ + { + "data": { + "node": "drive0", + "name": "bitmapB" + }, + "type": "block-dirty-bitmap-disable" + }, + { + "data": { + "node": "drive0", + "name": "bitmapC", + "granularity": 65536 + }, + "type": "block-dirty-bitmap-add" + }, + { + "data": { + "node": "drive0", + "name": "bitmapC" + }, + "type": "block-dirty-bitmap-disable" + }, + { + "data": { + "node": "drive0", + "name": "bitmapC" + }, + "type": "block-dirty-bitmap-enable" + } + ] + } +} +{ + "return": {} +} + +--- Emulating further writes --- + +write -P0xab 0 64k +{"return": ""} +write -P0xad 0x00f8000 64k +{"return": ""} +write -P0x1d 0x2008000 64k +{"return": ""} +write -P0xea 0x3fe0000 64k +{"return": ""} + +--- Disabling A & C --- + +{ + "execute": "transaction", + "arguments": { + "actions": [ + { + "data": { + "node": "drive0", + "name": "bitmapA" + }, + "type": "block-dirty-bitmap-disable" + }, + { + "data": { + "node": "drive0", + "name": "bitmapC" + }, + "type": "block-dirty-bitmap-disable" + } + ] + } +} +{ + "return": {} +} +{ + "bitmaps": { + "drive0": [ + { + "count": 393216, + "granularity": 65536, + "name": "bitmapC", + "status": "disabled" + }, + { + "count": 262144, + "granularity": 65536, + "name": "bitmapB", + "status": "disabled" + }, + { + "count": 458752, + "granularity": 65536, + "name": "bitmapA", + "status": "disabled" + } + ] + } +} + +--- Submitting & Aborting Merge Transaction --- + +{ + "execute": "transaction", + "arguments": { + "actions": [ + { + "data": { + "node": "drive0", + "disabled": true, + "name": "bitmapD", + "granularity": 65536 + }, + "type": "block-dirty-bitmap-add" + }, + { + "data": { + "node": "drive0", + "target": "bitmapD", + "bitmaps": [ + "bitmapB", + "bitmapC" + ] + }, + "type": "block-dirty-bitmap-merge" + }, + { + "data": {}, + "type": "abort" + } + ] + } +} +{ + "error": { + "class": "GenericError", + "desc": "Transaction aborted using Abort action" + } +} +{ + "bitmaps": { + "drive0": [ + { + "count": 393216, + "granularity": 65536, + "name": "bitmapC", + "status": "disabled" + }, + { + "count": 262144, + "granularity": 65536, + "name": "bitmapB", + "status": "disabled" + }, + { + "count": 458752, + "granularity": 65536, + "name": "bitmapA", + "status": "disabled" + } + ] + } +} + +--- Creating D as a merge of B & C --- + +{ + "execute": "transaction", + "arguments": { + "actions": [ + { + "data": { + "node": "drive0", + "disabled": true, + "name": "bitmapD", + "granularity": 65536 + }, + "type": "block-dirty-bitmap-add" + }, + { + "data": { + "node": "drive0", + "target": "bitmapD", + "bitmaps": [ + "bitmapB", + "bitmapC" + ] + }, + "type": "block-dirty-bitmap-merge" + } + ] + } +} +{ + "return": {} +} +{ + "bitmaps": { + "drive0": [ + { + "count": 458752, + "granularity": 65536, + "name": "bitmapD", + "status": "disabled" + }, + { + "count": 393216, + "granularity": 65536, + "name": "bitmapC", + "status": "disabled" + }, + { + "count": 262144, + "granularity": 65536, + "name": "bitmapB", + "status": "disabled" + }, + { + "count": 458752, + "granularity": 65536, + "name": "bitmapA", + "status": "disabled" + } + ] + } +} + +--- Removing bitmaps A, B, C, and D --- + +{"execute": "block-dirty-bitmap-remove", "arguments": {"name": "bitmapA", "node": "drive0"}} +{"return": {}} +{"execute": "block-dirty-bitmap-remove", "arguments": {"name": "bitmapB", "node": "drive0"}} +{"return": {}} +{"execute": "block-dirty-bitmap-remove", "arguments": {"name": "bitmapC", "node": "drive0"}} +{"return": {}} +{"execute": "block-dirty-bitmap-remove", "arguments": {"name": "bitmapD", "node": "drive0"}} +{"return": {}} + +--- Final Query --- + +{ + "bitmaps": { + "drive0": [] + } +} + +--- Done --- + diff --git a/tests/qemu-iotests/group b/tests/qemu-iotests/group index 61a6d98ebd..f6b245917a 100644 --- a/tests/qemu-iotests/group +++ b/tests/qemu-iotests/group @@ -233,3 +233,4 @@ 233 auto quick 234 auto quick migration 235 auto quick +236 auto quick diff --git a/tests/qemu-iotests/iotests.py b/tests/qemu-iotests/iotests.py index d537538ba0..cbedfaf1df 100644 --- a/tests/qemu-iotests/iotests.py +++ b/tests/qemu-iotests/iotests.py @@ -30,6 +30,7 @@ import signal import logging import atexit import io +from collections import OrderedDict sys.path.append(os.path.join(os.path.dirname(__file__), '..', '..', 'scripts')) import qtest @@ -63,7 +64,7 @@ socket_scm_helper = os.environ.get('SOCKET_SCM_HELPER', 'socket_scm_helper') debug = False luks_default_secret_object = 'secret,id=keysec0,data=' + \ - os.environ['IMGKEYSECRET'] + os.environ.get('IMGKEYSECRET', '') luks_default_key_secret_opt = 'key-secret=keysec0' @@ -75,6 +76,16 @@ def qemu_img(*args): sys.stderr.write('qemu-img received signal %i: %s\n' % (-exitcode, ' '.join(qemu_img_args + list(args)))) return exitcode +def ordered_kwargs(kwargs): + # kwargs prior to 3.6 are not ordered, so: + od = OrderedDict() + for k, v in sorted(kwargs.items()): + if isinstance(v, dict): + od[k] = ordered_kwargs(v) + else: + od[k] = v + return od + def qemu_img_create(*args): args = list(args) @@ -235,10 +246,36 @@ def filter_qmp_event(event): event['timestamp']['microseconds'] = 'USECS' return event +def filter_qmp(qmsg, filter_fn): + '''Given a string filter, filter a QMP object's values. + filter_fn takes a (key, value) pair.''' + # Iterate through either lists or dicts; + if isinstance(qmsg, list): + items = enumerate(qmsg) + else: + items = qmsg.items() + + for k, v in items: + if isinstance(v, list) or isinstance(v, dict): + qmsg[k] = filter_qmp(v, filter_fn) + else: + qmsg[k] = filter_fn(k, v) + return qmsg + def filter_testfiles(msg): prefix = os.path.join(test_dir, "%s-" % (os.getpid())) return msg.replace(prefix, 'TEST_DIR/PID-') +def filter_qmp_testfiles(qmsg): + def _filter(key, value): + if key == 'filename' or key == 'backing-file': + return filter_testfiles(value) + return value + return filter_qmp(qmsg, _filter) + +def filter_generated_node_ids(msg): + return re.sub("#block[0-9]+", "NODE_NAME", msg) + def filter_img_info(output, filename): lines = [] for line in output.split('\n'): @@ -251,11 +288,18 @@ def filter_img_info(output, filename): lines.append(line) return '\n'.join(lines) -def log(msg, filters=[]): +def log(msg, filters=[], indent=None): + '''Logs either a string message or a JSON serializable message (like QMP). + If indent is provided, JSON serializable messages are pretty-printed.''' for flt in filters: msg = flt(msg) - if type(msg) is dict or type(msg) is list: - print(json.dumps(msg, sort_keys=True)) + if isinstance(msg, dict) or isinstance(msg, list): + # Python < 3.4 needs to know not to add whitespace when pretty-printing: + separators = (', ', ': ') if indent is None else (',', ': ') + # Don't sort if it's already sorted + do_sort = not isinstance(msg, OrderedDict) + print(json.dumps(msg, sort_keys=do_sort, + indent=indent, separators=separators)) else: print(msg) @@ -444,12 +488,14 @@ class VM(qtest.QEMUQtestMachine): result.append(filter_qmp_event(ev)) return result - def qmp_log(self, cmd, filters=[filter_testfiles], **kwargs): - logmsg = '{"execute": "%s", "arguments": %s}' % \ - (cmd, json.dumps(kwargs, sort_keys=True)) - log(logmsg, filters) + def qmp_log(self, cmd, filters=[], indent=None, **kwargs): + full_cmd = OrderedDict(( + ("execute", cmd), + ("arguments", ordered_kwargs(kwargs)) + )) + log(full_cmd, filters, indent=indent) result = self.qmp(cmd, **kwargs) - log(json.dumps(result, sort_keys=True), filters) + log(result, filters, indent=indent) return result def run_job(self, job, auto_finalize=True, auto_dismiss=False): diff --git a/tests/qht-bench.c b/tests/qht-bench.c index ab4e708180..e3b512f26f 100644 --- a/tests/qht-bench.c +++ b/tests/qht-bench.c @@ -398,16 +398,14 @@ static void pr_stats(void) static void run_test(void) { - unsigned int remaining; int i; while (atomic_read(&n_ready_threads) != n_rw_threads + n_rz_threads) { cpu_relax(); } + atomic_set(&test_start, true); - do { - remaining = sleep(duration); - } while (remaining); + g_usleep(duration * G_USEC_PER_SEC); atomic_set(&test_stop, true); for (i = 0; i < n_rw_threads; i++) { diff --git a/ui/input.c b/ui/input.c index 35c7964f64..9494688295 100644 --- a/ui/input.c +++ b/ui/input.c @@ -460,22 +460,18 @@ void qemu_input_event_send_key_delay(uint32_t delay_ms) } } -InputEvent *qemu_input_event_new_btn(InputButton btn, bool down) -{ - InputEvent *evt = g_new0(InputEvent, 1); - evt->u.btn.data = g_new0(InputBtnEvent, 1); - evt->type = INPUT_EVENT_KIND_BTN; - evt->u.btn.data->button = btn; - evt->u.btn.data->down = down; - return evt; -} - void qemu_input_queue_btn(QemuConsole *src, InputButton btn, bool down) { - InputEvent *evt; - evt = qemu_input_event_new_btn(btn, down); - qemu_input_event_send(src, evt); - qapi_free_InputEvent(evt); + InputBtnEvent bevt = { + .button = btn, + .down = down, + }; + InputEvent evt = { + .type = INPUT_EVENT_KIND_BTN, + .u.btn.data = &bevt, + }; + + qemu_input_event_send(src, &evt); } void qemu_input_update_buttons(QemuConsole *src, uint32_t *button_map, @@ -515,37 +511,35 @@ int qemu_input_scale_axis(int value, return ((int64_t)value - min_in) * range_out / range_in + min_out; } -InputEvent *qemu_input_event_new_move(InputEventKind kind, - InputAxis axis, int value) -{ - InputEvent *evt = g_new0(InputEvent, 1); - InputMoveEvent *move = g_new0(InputMoveEvent, 1); - - evt->type = kind; - evt->u.rel.data = move; /* evt->u.rel is the same as evt->u.abs */ - move->axis = axis; - move->value = value; - return evt; -} - void qemu_input_queue_rel(QemuConsole *src, InputAxis axis, int value) { - InputEvent *evt; - evt = qemu_input_event_new_move(INPUT_EVENT_KIND_REL, axis, value); - qemu_input_event_send(src, evt); - qapi_free_InputEvent(evt); + InputMoveEvent move = { + .axis = axis, + .value = value, + }; + InputEvent evt = { + .type = INPUT_EVENT_KIND_REL, + .u.rel.data = &move, + }; + + qemu_input_event_send(src, &evt); } void qemu_input_queue_abs(QemuConsole *src, InputAxis axis, int value, int min_in, int max_in) { - InputEvent *evt; - int scaled = qemu_input_scale_axis(value, min_in, max_in, + InputMoveEvent move = { + .axis = axis, + .value = qemu_input_scale_axis(value, min_in, max_in, INPUT_EVENT_ABS_MIN, - INPUT_EVENT_ABS_MAX); - evt = qemu_input_event_new_move(INPUT_EVENT_KIND_ABS, axis, scaled); - qemu_input_event_send(src, evt); - qapi_free_InputEvent(evt); + INPUT_EVENT_ABS_MAX), + }; + InputEvent evt = { + .type = INPUT_EVENT_KIND_ABS, + .u.abs.data = &move, + }; + + qemu_input_event_send(src, &evt); } void qemu_input_check_mode_change(void) diff --git a/util/aio-posix.c b/util/aio-posix.c index 51c41ed3c9..8640dfde9f 100644 --- a/util/aio-posix.c +++ b/util/aio-posix.c @@ -200,6 +200,31 @@ static AioHandler *find_aio_handler(AioContext *ctx, int fd) return NULL; } +static bool aio_remove_fd_handler(AioContext *ctx, AioHandler *node) +{ + /* If the GSource is in the process of being destroyed then + * g_source_remove_poll() causes an assertion failure. Skip + * removal in that case, because glib cleans up its state during + * destruction anyway. + */ + if (!g_source_is_destroyed(&ctx->source)) { + g_source_remove_poll(&ctx->source, &node->pfd); + } + + /* If a read is in progress, just mark the node as deleted */ + if (qemu_lockcnt_count(&ctx->list_lock)) { + node->deleted = 1; + node->pfd.revents = 0; + return false; + } + /* Otherwise, delete it for real. We can't just mark it as + * deleted because deleted nodes are only cleaned up while + * no one is walking the handlers list. + */ + QLIST_REMOVE(node, node); + return true; +} + void aio_set_fd_handler(AioContext *ctx, int fd, bool is_external, @@ -209,6 +234,7 @@ void aio_set_fd_handler(AioContext *ctx, void *opaque) { AioHandler *node; + AioHandler *new_node = NULL; bool is_new = false; bool deleted = false; int poll_disable_change; @@ -223,50 +249,39 @@ void aio_set_fd_handler(AioContext *ctx, qemu_lockcnt_unlock(&ctx->list_lock); return; } + /* Clean events in order to unregister fd from the ctx epoll. */ + node->pfd.events = 0; - /* If the GSource is in the process of being destroyed then - * g_source_remove_poll() causes an assertion failure. Skip - * removal in that case, because glib cleans up its state during - * destruction anyway. - */ - if (!g_source_is_destroyed(&ctx->source)) { - g_source_remove_poll(&ctx->source, &node->pfd); - } - - /* If a read is in progress, just mark the node as deleted */ - if (qemu_lockcnt_count(&ctx->list_lock)) { - node->deleted = 1; - node->pfd.revents = 0; - } else { - /* Otherwise, delete it for real. We can't just mark it as - * deleted because deleted nodes are only cleaned up while - * no one is walking the handlers list. - */ - QLIST_REMOVE(node, node); - deleted = true; - } poll_disable_change = -!node->io_poll; } else { poll_disable_change = !io_poll - (node && !node->io_poll); if (node == NULL) { - /* Alloc and insert if it's not already there */ - node = g_new0(AioHandler, 1); - node->pfd.fd = fd; - QLIST_INSERT_HEAD_RCU(&ctx->aio_handlers, node, node); - - g_source_add_poll(&ctx->source, &node->pfd); is_new = true; } + /* Alloc and insert if it's not already there */ + new_node = g_new0(AioHandler, 1); /* Update handler with latest information */ - node->io_read = io_read; - node->io_write = io_write; - node->io_poll = io_poll; - node->opaque = opaque; - node->is_external = is_external; + new_node->io_read = io_read; + new_node->io_write = io_write; + new_node->io_poll = io_poll; + new_node->opaque = opaque; + new_node->is_external = is_external; + + if (is_new) { + new_node->pfd.fd = fd; + } else { + new_node->pfd = node->pfd; + } + g_source_add_poll(&ctx->source, &new_node->pfd); + + new_node->pfd.events = (io_read ? G_IO_IN | G_IO_HUP | G_IO_ERR : 0); + new_node->pfd.events |= (io_write ? G_IO_OUT | G_IO_ERR : 0); - node->pfd.events = (io_read ? G_IO_IN | G_IO_HUP | G_IO_ERR : 0); - node->pfd.events |= (io_write ? G_IO_OUT | G_IO_ERR : 0); + QLIST_INSERT_HEAD_RCU(&ctx->aio_handlers, new_node, node); + } + if (node) { + deleted = aio_remove_fd_handler(ctx, node); } /* No need to order poll_disable_cnt writes against other updates; @@ -278,7 +293,12 @@ void aio_set_fd_handler(AioContext *ctx, atomic_set(&ctx->poll_disable_cnt, atomic_read(&ctx->poll_disable_cnt) + poll_disable_change); - aio_epoll_update(ctx, node, is_new); + if (new_node) { + aio_epoll_update(ctx, new_node, is_new); + } else if (node) { + /* Unregister deleted fd_handler */ + aio_epoll_update(ctx, node, false); + } qemu_lockcnt_unlock(&ctx->list_lock); aio_notify(ctx); diff --git a/util/aio-win32.c b/util/aio-win32.c index c58957cc4b..a23b9c364d 100644 --- a/util/aio-win32.c +++ b/util/aio-win32.c @@ -35,6 +35,22 @@ struct AioHandler { QLIST_ENTRY(AioHandler) node; }; +static void aio_remove_fd_handler(AioContext *ctx, AioHandler *node) +{ + /* If aio_poll is in progress, just mark the node as deleted */ + if (qemu_lockcnt_count(&ctx->list_lock)) { + node->deleted = 1; + node->pfd.revents = 0; + } else { + /* Otherwise, delete it for real. We can't just mark it as + * deleted because deleted nodes are only cleaned up after + * releasing the list_lock. + */ + QLIST_REMOVE(node, node); + g_free(node); + } +} + void aio_set_fd_handler(AioContext *ctx, int fd, bool is_external, @@ -44,41 +60,23 @@ void aio_set_fd_handler(AioContext *ctx, void *opaque) { /* fd is a SOCKET in our case */ - AioHandler *node; + AioHandler *old_node; + AioHandler *node = NULL; qemu_lockcnt_lock(&ctx->list_lock); - QLIST_FOREACH(node, &ctx->aio_handlers, node) { - if (node->pfd.fd == fd && !node->deleted) { + QLIST_FOREACH(old_node, &ctx->aio_handlers, node) { + if (old_node->pfd.fd == fd && !old_node->deleted) { break; } } - /* Are we deleting the fd handler? */ - if (!io_read && !io_write) { - if (node) { - /* If aio_poll is in progress, just mark the node as deleted */ - if (qemu_lockcnt_count(&ctx->list_lock)) { - node->deleted = 1; - node->pfd.revents = 0; - } else { - /* Otherwise, delete it for real. We can't just mark it as - * deleted because deleted nodes are only cleaned up after - * releasing the list_lock. - */ - QLIST_REMOVE(node, node); - g_free(node); - } - } - } else { + if (io_read || io_write) { HANDLE event; long bitmask = 0; - if (node == NULL) { - /* Alloc and insert if it's not already there */ - node = g_new0(AioHandler, 1); - node->pfd.fd = fd; - QLIST_INSERT_HEAD_RCU(&ctx->aio_handlers, node, node); - } + /* Alloc and insert if it's not already there */ + node = g_new0(AioHandler, 1); + node->pfd.fd = fd; node->pfd.events = 0; if (node->io_read) { @@ -104,9 +102,13 @@ void aio_set_fd_handler(AioContext *ctx, bitmask |= FD_WRITE | FD_CONNECT; } + QLIST_INSERT_HEAD_RCU(&ctx->aio_handlers, node, node); event = event_notifier_get_handle(&ctx->notifier); WSAEventSelect(node->pfd.fd, event, bitmask); } + if (old_node) { + aio_remove_fd_handler(ctx, old_node); + } qemu_lockcnt_unlock(&ctx->list_lock); aio_notify(ctx); @@ -139,18 +141,7 @@ void aio_set_event_notifier(AioContext *ctx, if (node) { g_source_remove_poll(&ctx->source, &node->pfd); - /* aio_poll is in progress, just mark the node as deleted */ - if (qemu_lockcnt_count(&ctx->list_lock)) { - node->deleted = 1; - node->pfd.revents = 0; - } else { - /* Otherwise, delete it for real. We can't just mark it as - * deleted because deleted nodes are only cleaned up after - * releasing the list_lock. - */ - QLIST_REMOVE(node, node); - g_free(node); - } + aio_remove_fd_handler(ctx, node); } } else { if (node == NULL) { @@ -3856,13 +3856,6 @@ int main(int argc, char **argv, char **envp) } xen_domid = atoi(optarg); break; - case QEMU_OPTION_xen_create: - if (!(xen_available())) { - error_report("Option not supported for this target"); - exit(1); - } - xen_mode = XEN_CREATE; - break; case QEMU_OPTION_xen_attach: if (!(xen_available())) { error_report("Option not supported for this target"); |