From b30168647fea0fc173a0e244d2943fc5575ca511 Mon Sep 17 00:00:00 2001 From: Vladimir Sementsov-Ogievskiy Date: Fri, 11 Oct 2019 12:07:09 +0300 Subject: block/block-backend: add blk_co_pwritev_part Add blk write function with qiov_offset parameter. It's needed for the following commit. Signed-off-by: Vladimir Sementsov-Ogievskiy Reviewed-by: Max Reitz Message-id: 20191011090711.19940-4-vsementsov@virtuozzo.com Signed-off-by: Max Reitz --- include/sysemu/block-backend.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include') diff --git a/include/sysemu/block-backend.h b/include/sysemu/block-backend.h index 368d53a..73f2cef 100644 --- a/include/sysemu/block-backend.h +++ b/include/sysemu/block-backend.h @@ -121,6 +121,10 @@ void blk_set_dev_ops(BlockBackend *blk, const BlockDevOps *ops, void *opaque); int coroutine_fn blk_co_preadv(BlockBackend *blk, int64_t offset, unsigned int bytes, QEMUIOVector *qiov, BdrvRequestFlags flags); +int coroutine_fn blk_co_pwritev_part(BlockBackend *blk, int64_t offset, + unsigned int bytes, + QEMUIOVector *qiov, size_t qiov_offset, + BdrvRequestFlags flags); int coroutine_fn blk_co_pwritev(BlockBackend *blk, int64_t offset, unsigned int bytes, QEMUIOVector *qiov, BdrvRequestFlags flags); -- cgit v1.1 From f16ba00de946517e5cca4edeeec3e21f8f77132d Mon Sep 17 00:00:00 2001 From: Vladimir Sementsov-Ogievskiy Date: Tue, 22 Oct 2019 14:18:03 +0300 Subject: util: introduce SharedResource Introduce an API for some shared splittable resource, like memory. It's going to be used by backup. Backup uses both read/write io and copy_range. copy_range may consume memory implictly, so the new API is abstract: it doesn't allocate any real memory but only hands out tickets. The idea is that we have some total amount of something and callers should wait in coroutine queue if there is not enough of the resource at the moment. Signed-off-by: Vladimir Sementsov-Ogievskiy Reviewed-by: Max Reitz Message-id: 20191022111805.3432-5-vsementsov@virtuozzo.com Signed-off-by: Max Reitz --- include/qemu/co-shared-resource.h | 71 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 71 insertions(+) create mode 100644 include/qemu/co-shared-resource.h (limited to 'include') diff --git a/include/qemu/co-shared-resource.h b/include/qemu/co-shared-resource.h new file mode 100644 index 0000000..4e45030 --- /dev/null +++ b/include/qemu/co-shared-resource.h @@ -0,0 +1,71 @@ +/* + * Helper functionality for distributing a fixed total amount of + * an abstract resource among multiple coroutines. + * + * Copyright (c) 2019 Virtuozzo International GmbH + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ + +#ifndef QEMU_CO_SHARED_RESOURCE_H +#define QEMU_CO_SHARED_RESOURCE_H + + +typedef struct SharedResource SharedResource; + +/* + * Create SharedResource structure + * + * @total: total amount of some resource to be shared between clients + * + * Note: this API is not thread-safe. + */ +SharedResource *shres_create(uint64_t total); + +/* + * Release SharedResource structure + * + * This function may only be called once everything allocated by all + * clients has been deallocated. + */ +void shres_destroy(SharedResource *s); + +/* + * Try to allocate an amount of @n. Return true on success, and false + * if there is too little left of the collective resource to fulfill + * the request. + */ +bool co_try_get_from_shres(SharedResource *s, uint64_t n); + +/* + * Allocate an amount of @n, and, if necessary, yield until + * that becomes possible. + */ +void coroutine_fn co_get_from_shres(SharedResource *s, uint64_t n); + +/* + * Deallocate an amount of @n. The total amount allocated by a caller + * does not need to be deallocated/released with a single call, but may + * be split over several calls. For example, get(4), get(3), and then + * put(5), put(2). + */ +void coroutine_fn co_put_to_shres(SharedResource *s, uint64_t n); + + +#endif /* QEMU_CO_SHARED_RESOURCE_H */ -- cgit v1.1 From 7f739d0e5375a934f0eade3cd38e6d33673beec0 Mon Sep 17 00:00:00 2001 From: Vladimir Sementsov-Ogievskiy Date: Tue, 22 Oct 2019 14:18:04 +0300 Subject: block/block-copy: add memory limit Currently total allocation for parallel requests to block-copy instance is unlimited. Let's limit it to 128 MiB. For now block-copy is used only in backup, so actually we limit total allocation for backup job. Signed-off-by: Vladimir Sementsov-Ogievskiy Reviewed-by: Max Reitz Message-id: 20191022111805.3432-6-vsementsov@virtuozzo.com Signed-off-by: Max Reitz --- include/block/block-copy.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include') diff --git a/include/block/block-copy.h b/include/block/block-copy.h index e2e135ff..edcdf00 100644 --- a/include/block/block-copy.h +++ b/include/block/block-copy.h @@ -16,6 +16,7 @@ #define BLOCK_COPY_H #include "block/block.h" +#include "qemu/co-shared-resource.h" typedef struct BlockCopyInFlightReq { int64_t start_byte; @@ -69,6 +70,8 @@ typedef struct BlockCopyState { */ ProgressResetCallbackFunc progress_reset_callback; void *progress_opaque; + + SharedResource *mem; } BlockCopyState; BlockCopyState *block_copy_state_new(BdrvChild *source, BdrvChild *target, -- cgit v1.1 From 0e2402452f1f2042923a5206b9ff3e9d70c77811 Mon Sep 17 00:00:00 2001 From: Vladimir Sementsov-Ogievskiy Date: Tue, 22 Oct 2019 14:18:05 +0300 Subject: block/block-copy: increase buffered copy request No reason to limit buffered copy to one cluster. Let's allow up to 1 MiB. Signed-off-by: Vladimir Sementsov-Ogievskiy Reviewed-by: Max Reitz Message-id: 20191022111805.3432-7-vsementsov@virtuozzo.com Signed-off-by: Max Reitz --- include/block/block-copy.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/block/block-copy.h b/include/block/block-copy.h index edcdf00..0a16172 100644 --- a/include/block/block-copy.h +++ b/include/block/block-copy.h @@ -38,7 +38,7 @@ typedef struct BlockCopyState { BdrvDirtyBitmap *copy_bitmap; int64_t cluster_size; bool use_copy_range; - int64_t copy_range_size; + int64_t copy_size; uint64_t len; QLIST_HEAD(, BlockCopyInFlightReq) inflight_reqs; -- cgit v1.1 From e0dd95e3738c5868ddea932aa0f1998900eeaf58 Mon Sep 17 00:00:00 2001 From: Maxim Levitsky Date: Fri, 13 Sep 2019 16:36:26 +0300 Subject: block/nvme: add support for write zeros Signed-off-by: Maxim Levitsky Message-id: 20190913133627.28450-2-mlevitsk@redhat.com Reviewed-by: John Snow Signed-off-by: Max Reitz --- include/block/nvme.h | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/block/nvme.h b/include/block/nvme.h index 3ec8efc..ab5943b 100644 --- a/include/block/nvme.h +++ b/include/block/nvme.h @@ -653,12 +653,29 @@ typedef struct NvmeIdNs { uint8_t mc; uint8_t dpc; uint8_t dps; - uint8_t res30[98]; + + uint8_t nmic; + uint8_t rescap; + uint8_t fpi; + uint8_t dlfeat; + + uint8_t res34[94]; NvmeLBAF lbaf[16]; uint8_t res192[192]; uint8_t vs[3712]; } NvmeIdNs; + +/*Deallocate Logical Block Features*/ +#define NVME_ID_NS_DLFEAT_GUARD_CRC(dlfeat) ((dlfeat) & 0x10) +#define NVME_ID_NS_DLFEAT_WRITE_ZEROES(dlfeat) ((dlfeat) & 0x08) + +#define NVME_ID_NS_DLFEAT_READ_BEHAVIOR(dlfeat) ((dlfeat) & 0x7) +#define NVME_ID_NS_DLFEAT_READ_BEHAVIOR_UNDEFINED 0 +#define NVME_ID_NS_DLFEAT_READ_BEHAVIOR_ZEROES 1 +#define NVME_ID_NS_DLFEAT_READ_BEHAVIOR_ONES 2 + + #define NVME_ID_NS_NSFEAT_THIN(nsfeat) ((nsfeat & 0x1)) #define NVME_ID_NS_FLBAS_EXTENDED(flbas) ((flbas >> 4) & 0x1) #define NVME_ID_NS_FLBAS_INDEX(flbas) ((flbas & 0xf)) -- cgit v1.1 From 5d5b33c08031cfe0e9872bde1f9dcb2215f9b30a Mon Sep 17 00:00:00 2001 From: Max Reitz Date: Fri, 11 Oct 2019 17:27:59 +0200 Subject: include: Move endof() up from hw/virtio/virtio.h endof() is a useful macro, we can make use of it outside of virtio. Signed-off-by: Max Reitz Reviewed-by: Eric Blake Message-id: 20191011152814.14791-2-mreitz@redhat.com Signed-off-by: Max Reitz --- include/hw/virtio/virtio.h | 7 ------- include/qemu/compiler.h | 7 +++++++ 2 files changed, 7 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/hw/virtio/virtio.h b/include/hw/virtio/virtio.h index 48e8d04..ef083af 100644 --- a/include/hw/virtio/virtio.h +++ b/include/hw/virtio/virtio.h @@ -37,13 +37,6 @@ static inline hwaddr vring_align(hwaddr addr, return QEMU_ALIGN_UP(addr, align); } -/* - * Calculate the number of bytes up to and including the given 'field' of - * 'container'. - */ -#define virtio_endof(container, field) \ - (offsetof(container, field) + sizeof_field(container, field)) - typedef struct VirtIOFeature { uint64_t flags; size_t end; diff --git a/include/qemu/compiler.h b/include/qemu/compiler.h index 7b93c73..85c02c1 100644 --- a/include/qemu/compiler.h +++ b/include/qemu/compiler.h @@ -60,6 +60,13 @@ #define sizeof_field(type, field) sizeof(((type *)0)->field) +/* + * Calculate the number of bytes up to and including the given 'field' of + * 'container'. + */ +#define endof(container, field) \ + (offsetof(container, field) + sizeof_field(container, field)) + /* Convert from a base type to a parent type, with compile time checking. */ #ifdef __GNUC__ #define DO_UPCAST(type, field, dev) ( __extension__ ( { \ -- cgit v1.1 From c80d8b06cfa59f5d8229379c85dcb2c3bb9c881b Mon Sep 17 00:00:00 2001 From: Max Reitz Date: Wed, 18 Sep 2019 11:51:40 +0200 Subject: block: Add @exact parameter to bdrv_co_truncate() We have two drivers (iscsi and file-posix) that (in some cases) return success from their .bdrv_co_truncate() implementation if the block device is larger than the requested offset, but cannot be shrunk. Some callers do not want that behavior, so this patch adds a new parameter that they can use to turn off that behavior. This patch just adds the parameter and lets the block/io.c and block/block-backend.c functions pass it around. All other callers always pass false and none of the implementations evaluate it, so that this patch does not change existing behavior. Future patches take care of that. Suggested-by: Maxim Levitsky Signed-off-by: Max Reitz Message-id: 20190918095144.955-5-mreitz@redhat.com Reviewed-by: Maxim Levitsky Signed-off-by: Max Reitz --- include/block/block.h | 6 +++--- include/block/block_int.h | 17 ++++++++++++++++- include/sysemu/block-backend.h | 4 ++-- 3 files changed, 21 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/block/block.h b/include/block/block.h index 89606bd..1df9848 100644 --- a/include/block/block.h +++ b/include/block/block.h @@ -346,10 +346,10 @@ BlockDriverState *bdrv_find_backing_image(BlockDriverState *bs, const char *backing_file); void bdrv_refresh_filename(BlockDriverState *bs); -int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, +int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, bool exact, PreallocMode prealloc, Error **errp); -int bdrv_truncate(BdrvChild *child, int64_t offset, PreallocMode prealloc, - Error **errp); +int bdrv_truncate(BdrvChild *child, int64_t offset, bool exact, + PreallocMode prealloc, Error **errp); int64_t bdrv_nb_sectors(BlockDriverState *bs); int64_t bdrv_getlength(BlockDriverState *bs); diff --git a/include/block/block_int.h b/include/block/block_int.h index ca4ccac4..02dc003 100644 --- a/include/block/block_int.h +++ b/include/block/block_int.h @@ -334,8 +334,23 @@ struct BlockDriver { * bdrv_parse_filename. */ const char *protocol_name; + + /* + * Truncate @bs to @offset bytes using the given @prealloc mode + * when growing. Modes other than PREALLOC_MODE_OFF should be + * rejected when shrinking @bs. + * + * If @exact is true, @bs must be resized to exactly @offset. + * Otherwise, it is sufficient for @bs (if it is a host block + * device and thus there is no way to resize it) to be at least + * @offset bytes in length. + * + * If @exact is true and this function fails but would succeed + * with @exact = false, it should return -ENOTSUP. + */ int coroutine_fn (*bdrv_co_truncate)(BlockDriverState *bs, int64_t offset, - PreallocMode prealloc, Error **errp); + bool exact, PreallocMode prealloc, + Error **errp); int64_t (*bdrv_getlength)(BlockDriverState *bs); bool has_variable_length; diff --git a/include/sysemu/block-backend.h b/include/sysemu/block-backend.h index 73f2cef..b198dec 100644 --- a/include/sysemu/block-backend.h +++ b/include/sysemu/block-backend.h @@ -237,8 +237,8 @@ int coroutine_fn blk_co_pwrite_zeroes(BlockBackend *blk, int64_t offset, int bytes, BdrvRequestFlags flags); int blk_pwrite_compressed(BlockBackend *blk, int64_t offset, const void *buf, int bytes); -int blk_truncate(BlockBackend *blk, int64_t offset, PreallocMode prealloc, - Error **errp); +int blk_truncate(BlockBackend *blk, int64_t offset, bool exact, + PreallocMode prealloc, Error **errp); int blk_pdiscard(BlockBackend *blk, int64_t offset, int bytes); int blk_save_vmstate(BlockBackend *blk, const uint8_t *buf, int64_t pos, int size); -- cgit v1.1