Hi, Please ignore the v2 series. Please review the v3 series instead. Thanks! v2->v3: In v2 Persist Through Power Loss(PTPL) is enable default. In v3 PTPL is supported, which is passed as a parameter. v1->v2: - Add sg_persist --report-capabilities for SCSI protocol and enable oncs and rescap for NVMe protocol. - Add persistent reservation capabilities constants and helper functions for SCSI and NVMe protocol. - Add comments for necessary APIs. v1: - Add seven APIs about persistent reservation command for block layer. These APIs including reading keys, reading reservations, registering, reserving, releasing, clearing and preempting. - Add the necessary pr-related operation APIs for both the SCSI protocol and NVMe protocol at the device layer. - Add scsi driver at the driver layer to verify the functions. Changqi Lu (11): block: add persistent reservation in/out api block/raw: add persistent reservation in/out driver scsi/constant: add persistent reservation in/out protocol constants scsi/util: add helper functions for persistent reservation types conversion hw/scsi: add persistent reservation in/out api for scsi device block/nvme: add reservation command protocol constants hw/nvme: add helper functions for converting reservation types hw/nvme: enable ONCS reservations hw/nvme: enable namespace rescap function hw/nvme: add reservation protocal command block/iscsi: add persistent reservation in/out driver block/block-backend.c | 397 ++++++++++++++++++++++++++ block/io.c | 163 +++++++++++ block/iscsi.c | 443 ++++++++++++++++++++++++++++++ block/raw-format.c | 56 ++++ hw/nvme/ctrl.c | 324 +++++++++++++++++++++- hw/nvme/ns.c | 8 + hw/nvme/nvme.h | 44 +++ hw/scsi/scsi-disk.c | 352 ++++++++++++++++++++++++ include/block/block-common.h | 40 +++ include/block/block-io.h | 20 ++ include/block/block_int-common.h | 84 ++++++ include/block/nvme.h | 99 +++++++ include/scsi/constants.h | 52 ++++ include/scsi/utils.h | 8 + include/sysemu/block-backend-io.h | 24 ++ scsi/utils.c | 81 ++++++ 16 files changed, 2193 insertions(+), 2 deletions(-) -- 2.20.1 Add persistent reservation in/out operations at the block level. The following operations are included: - read_keys: retrieves the list of registered keys. - read_reservation: retrieves the current reservation status. - register: registers a new reservation key. - reserve: initiates a reservation for a specific key. - release: releases a reservation for a specific key. - clear: clears all existing reservations. - preempt: preempts a reservation held by another key. Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com> Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> --- block/block-backend.c | 397 ++++++++++++++++++++++++++++++ block/io.c | 163 ++++++++++++ include/block/block-common.h | 40 +++ include/block/block-io.h | 20 ++ include/block/block_int-common.h | 84 +++++++ include/sysemu/block-backend-io.h | 24 ++ 6 files changed, 728 insertions(+) diff --git a/block/block-backend.c b/block/block-backend.c index XXXXXXX..XXXXXXX 100644 --- a/block/block-backend.c +++ b/block/block-backend.c @@ -XXX,XX +XXX,XX @@ BlockAIOCB *blk_aio_ioctl(BlockBackend *blk, unsigned long int req, void *buf, return blk_aio_prwv(blk, req, 0, buf, blk_aio_ioctl_entry, 0, cb, opaque); } +typedef struct BlkPrInCo { + BlockBackend *blk; + uint32_t *generation; + uint32_t num_keys; + BlockPrType *type; + uint64_t *keys; + int ret; +} BlkPrInCo; + +typedef struct BlkPrInCB { + BlockAIOCB common; + BlkPrInCo prco; + bool has_returned; +} BlkPrInCB; + +static const AIOCBInfo blk_pr_in_aiocb_info = { + .aiocb_size = sizeof(BlkPrInCB), +}; + +static void blk_pr_in_complete(BlkPrInCB *acb) +{ + if (acb->has_returned) { + acb->common.cb(acb->common.opaque, acb->prco.ret); + blk_dec_in_flight(acb->prco.blk); + qemu_aio_unref(acb); + } +} + +static void blk_pr_in_complete_bh(void *opaque) +{ + BlkPrInCB *acb = opaque; + assert(acb->has_returned); + blk_pr_in_complete(acb); +} + +static BlockAIOCB *blk_aio_pr_in(BlockBackend *blk, uint32_t *generation, + uint32_t num_keys, BlockPrType *type, + uint64_t *keys, CoroutineEntry co_entry, + BlockCompletionFunc *cb, void *opaque) +{ + BlkPrInCB *acb; + Coroutine *co; + + blk_inc_in_flight(blk); + acb = blk_aio_get(&blk_pr_in_aiocb_info, blk, cb, opaque); + acb->prco = (BlkPrInCo) { + .blk = blk, + .generation = generation, + .num_keys = num_keys, + .type = type, + .ret = NOT_DONE, + .keys = keys, + }; + acb->has_returned = false; + + co = qemu_coroutine_create(co_entry, acb); + aio_co_enter(qemu_get_current_aio_context(), co); + + acb->has_returned = true; + if (acb->prco.ret != NOT_DONE) { + replay_bh_schedule_oneshot_event(qemu_get_current_aio_context(), + blk_pr_in_complete_bh, acb); + } + + return &acb->common; +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_read_keys(BlockBackend *blk, uint32_t *generation, + uint32_t num_keys, uint64_t *keys) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_read_keys(blk_bs(blk), generation, num_keys, keys); +} + +static void coroutine_fn blk_aio_pr_read_keys_entry(void *opaque) +{ + BlkPrInCB *acb = opaque; + BlkPrInCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_read_keys(prco->blk, prco->generation, + prco->num_keys, prco->keys); + blk_pr_in_complete(acb); +} + +BlockAIOCB *blk_aio_pr_read_keys(BlockBackend *blk, uint32_t *generation, + uint32_t num_keys, uint64_t *keys, + BlockCompletionFunc *cb, void *opaque) +{ + IO_CODE(); + return blk_aio_pr_in(blk, generation, num_keys, NULL, keys, + blk_aio_pr_read_keys_entry, cb, opaque); +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_read_reservation(BlockBackend *blk, uint32_t *generation, + uint64_t *key, BlockPrType *type) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_read_reservation(blk_bs(blk), generation, key, type); +} + +static void coroutine_fn blk_aio_pr_read_reservation_entry(void *opaque) +{ + BlkPrInCB *acb = opaque; + BlkPrInCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_read_reservation(prco->blk, prco->generation, + prco->keys, prco->type); + blk_pr_in_complete(acb); +} + +BlockAIOCB *blk_aio_pr_read_reservation(BlockBackend *blk, uint32_t *generation, + uint64_t *key, BlockPrType *type, + BlockCompletionFunc *cb, void *opaque) +{ + IO_CODE(); + return blk_aio_pr_in(blk, generation, 0, type, key, + blk_aio_pr_read_reservation_entry, cb, opaque); +} + +typedef struct BlkPrOutCo { + BlockBackend *blk; + uint64_t old_key; + uint64_t new_key; + bool ptpl; + BlockPrType type; + bool ignore_key; + bool abort; + int ret; +} BlkPrOutCo; + +typedef struct BlkPrOutCB { + BlockAIOCB common; + BlkPrOutCo prco; + bool has_returned; +} BlkPrOutCB; + +static const AIOCBInfo blk_pr_out_aiocb_info = { + .aiocb_size = sizeof(BlkPrOutCB), +}; + +static void blk_pr_out_complete(BlkPrOutCB *acb) +{ + if (acb->has_returned) { + acb->common.cb(acb->common.opaque, acb->prco.ret); + blk_dec_in_flight(acb->prco.blk); + qemu_aio_unref(acb); + } +} + +static void blk_pr_out_complete_bh(void *opaque) +{ + BlkPrOutCB *acb = opaque; + assert(acb->has_returned); + blk_pr_out_complete(acb); +} + +static BlockAIOCB *blk_aio_pr_out(BlockBackend *blk, uint64_t old_key, + uint64_t new_key, bool ptpl, + BlockPrType type, bool ignore_key, + bool abort, CoroutineEntry co_entry, + BlockCompletionFunc *cb, void *opaque) +{ + BlkPrOutCB *acb; + Coroutine *co; + + blk_inc_in_flight(blk); + acb = blk_aio_get(&blk_pr_out_aiocb_info, blk, cb, opaque); + acb->prco = (BlkPrOutCo) { + .blk = blk, + .old_key = old_key, + .new_key = new_key, + .ptpl = ptpl, + .type = type, + .ignore_key = ignore_key, + .abort = abort, + .ret = NOT_DONE, + }; + acb->has_returned = false; + + co = qemu_coroutine_create(co_entry, acb); + aio_co_enter(qemu_get_current_aio_context(), co); + + acb->has_returned = true; + if (acb->prco.ret != NOT_DONE) { + replay_bh_schedule_oneshot_event(qemu_get_current_aio_context(), + blk_pr_out_complete_bh, acb); + } + + return &acb->common; +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_register(BlockBackend *blk, uint64_t old_key, + uint64_t new_key, BlockPrType type, + bool ptpl, bool ignore_key) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_register(blk_bs(blk), old_key, new_key, type, + ptpl, ignore_key); +} + +static void coroutine_fn blk_aio_pr_register_entry(void *opaque) +{ + BlkPrOutCB *acb = opaque; + BlkPrOutCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_register(prco->blk, prco->old_key, prco->new_key, + prco->type, prco->ptpl, + prco->ignore_key); + blk_pr_out_complete(acb); +} + +BlockAIOCB *blk_aio_pr_register(BlockBackend *blk, uint64_t old_key, + uint64_t new_key, BlockPrType type, + bool ptpl, bool ignore_key, + BlockCompletionFunc *cb, + void *opaque) +{ + IO_CODE(); + return blk_aio_pr_out(blk, old_key, new_key, ptpl, type, ignore_key, false, + blk_aio_pr_register_entry, cb, opaque); +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_reserve(BlockBackend *blk, uint64_t key, BlockPrType type) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_reserve(blk_bs(blk), key, type); +} + +static void coroutine_fn blk_aio_pr_reserve_entry(void *opaque) +{ + BlkPrOutCB *acb = opaque; + BlkPrOutCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_reserve(prco->blk, prco->old_key, + prco->type); + blk_pr_out_complete(acb); +} + + +BlockAIOCB *blk_aio_pr_reserve(BlockBackend *blk, uint64_t key, + BlockPrType type, + BlockCompletionFunc *cb, + void *opaque) +{ + IO_CODE(); + return blk_aio_pr_out(blk, key, 0, false, type, false, false, + blk_aio_pr_reserve_entry, cb, opaque); +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_release(BlockBackend *blk, uint64_t key, BlockPrType type) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_release(blk_bs(blk), key, type); +} + +static void coroutine_fn blk_aio_pr_release_entry(void *opaque) +{ + BlkPrOutCB *acb = opaque; + BlkPrOutCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_release(prco->blk, prco->old_key, prco->type); + blk_pr_out_complete(acb); +} + + +BlockAIOCB *blk_aio_pr_release(BlockBackend *blk, uint64_t key, + BlockPrType type, BlockCompletionFunc *cb, + void *opaque) +{ + IO_CODE(); + return blk_aio_pr_out(blk, key, 0, false, type, false, false, + blk_aio_pr_release_entry, cb, opaque); +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_clear(BlockBackend *blk, uint64_t key) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_clear(blk_bs(blk), key); +} + +static void coroutine_fn blk_aio_pr_clear_entry(void *opaque) +{ + BlkPrOutCB *acb = opaque; + BlkPrOutCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_clear(prco->blk, prco->old_key); + blk_pr_out_complete(acb); +} + + +BlockAIOCB *blk_aio_pr_clear(BlockBackend *blk, uint64_t key, + BlockCompletionFunc *cb, void *opaque) +{ + IO_CODE(); + return blk_aio_pr_out(blk, key, 0, false, 0, false, false, + blk_aio_pr_clear_entry, cb, opaque); +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_preempt(BlockBackend *blk, uint64_t cr_key, + uint64_t pr_key, BlockPrType type, bool abort) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_preempt(blk_bs(blk), cr_key, pr_key, type, abort); +} + +static void coroutine_fn blk_aio_pr_preempt_entry(void *opaque) +{ + BlkPrOutCB *acb = opaque; + BlkPrOutCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_preempt(prco->blk, prco->old_key, + prco->new_key, prco->type, + prco->abort); + blk_pr_out_complete(acb); +} + + +BlockAIOCB *blk_aio_pr_preempt(BlockBackend *blk, uint64_t cr_key, + uint64_t pr_key, BlockPrType type, + bool abort, BlockCompletionFunc *cb, + void *opaque) +{ + IO_CODE(); + return blk_aio_pr_out(blk, cr_key, pr_key, false, type, false, abort, + blk_aio_pr_preempt_entry, cb, opaque); +} + /* To be called between exactly one pair of blk_inc/dec_in_flight() */ static int coroutine_fn blk_co_do_pdiscard(BlockBackend *blk, int64_t offset, int64_t bytes) diff --git a/block/io.c b/block/io.c index XXXXXXX..XXXXXXX 100644 --- a/block/io.c +++ b/block/io.c @@ -XXX,XX +XXX,XX @@ out: return co.ret; } +int coroutine_fn bdrv_co_pr_read_keys(BlockDriverState *bs, + uint32_t *generation, uint32_t num_keys, + uint64_t *keys) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_read_keys) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_read_keys(bs, generation, num_keys, keys); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + +int coroutine_fn bdrv_co_pr_read_reservation(BlockDriverState *bs, + uint32_t *generation, uint64_t *key, BlockPrType *type) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_read_reservation) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_read_reservation(bs, generation, key, type); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + +int coroutine_fn bdrv_co_pr_register(BlockDriverState *bs, uint64_t old_key, + uint64_t new_key, BlockPrType type, bool ptpl, + bool ignore_key) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_register) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_register(bs, old_key, new_key, type, + ptpl, ignore_key); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + +int coroutine_fn bdrv_co_pr_reserve(BlockDriverState *bs, uint64_t key, + BlockPrType type) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_reserve) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_reserve(bs, key, type); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + +int coroutine_fn bdrv_co_pr_release(BlockDriverState *bs, uint64_t key, + BlockPrType type) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_release) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_release(bs, key, type); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + +int coroutine_fn bdrv_co_pr_clear(BlockDriverState *bs, uint64_t key) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_clear) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_clear(bs, key); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + +int coroutine_fn bdrv_co_pr_preempt(BlockDriverState *bs, uint64_t cr_key, + uint64_t pr_key, BlockPrType type, bool abort) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_preempt) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_preempt(bs, cr_key, pr_key, type, abort); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + int coroutine_fn bdrv_co_zone_report(BlockDriverState *bs, int64_t offset, unsigned int *nr_zones, BlockZoneDescriptor *zones) diff --git a/include/block/block-common.h b/include/block/block-common.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/block-common.h +++ b/include/block/block-common.h @@ -XXX,XX +XXX,XX @@ typedef enum { BDRV_FIX_ERRORS = 2, } BdrvCheckMode; +/** + * According SCSI protocol(chapter 5.9 of SCSI Primary Commands - 4) + * and NVMe protocol(chapter 7.2 of NVMe Base Specification 2.0), + * the persistent reservation types and persistent capabilities of + * the public layer block are abstracted. + */ +typedef enum { + BLK_PR_WRITE_EXCLUSIVE = 0x1, + BLK_PR_EXCLUSIVE_ACCESS = 0x2, + BLK_PR_WRITE_EXCLUSIVE_REGS_ONLY = 0x3, + BLK_PR_EXCLUSIVE_ACCESS_REGS_ONLY = 0x4, + BLK_PR_WRITE_EXCLUSIVE_ALL_REGS = 0x5, + BLK_PR_EXCLUSIVE_ACCESS_ALL_REGS = 0x6, +} BlockPrType; + +typedef enum BLKPrCap { + /* Persist Through Power Loss */ + BLK_PR_CAP_PTPL = 1 << 0, + /* Write Exclusive reservation type */ + BLK_PR_CAP_WR_EX = 1 << 1, + /* Exclusive Access reservation type */ + BLK_PR_CAP_EX_AC = 1 << 2, + /* Write Exclusive Registrants Only reservation type */ + BLK_PR_CAP_WR_EX_RO = 1 << 3, + /* Exclusive Access Registrants Only reservation type */ + BLK_PR_CAP_EX_AC_RO = 1 << 4, + /* Write Exclusive All Registrants reservation type */ + BLK_PR_CAP_WR_EX_AR = 1 << 5, + /* Exclusive Access All Registrants reservation type */ + BLK_PR_CAP_EX_AC_AR = 1 << 6, + + BLK_PR_CAP_ALL = (BLK_PR_CAP_PTPL | + BLK_PR_CAP_WR_EX | + BLK_PR_CAP_EX_AC | + BLK_PR_CAP_WR_EX_RO | + BLK_PR_CAP_EX_AC_RO | + BLK_PR_CAP_WR_EX_AR | + BLK_PR_CAP_EX_AC_AR), +} BLKPrCap; + typedef struct BlockSizes { uint32_t phys; uint32_t log; diff --git a/include/block/block-io.h b/include/block/block-io.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/block-io.h +++ b/include/block/block-io.h @@ -XXX,XX +XXX,XX @@ void bdrv_aio_cancel_async(BlockAIOCB *acb); int coroutine_fn GRAPH_RDLOCK bdrv_co_ioctl(BlockDriverState *bs, int req, void *buf); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_read_keys(BlockDriverState *bs, uint32_t *generation, + uint32_t num_keys, uint64_t *keys); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_read_reservation(BlockDriverState *bs, uint32_t *generation, + uint64_t *key, BlockPrType *type); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_register(BlockDriverState *bs, uint64_t old_key, + uint64_t new_key, BlockPrType type, + bool ptpl, bool ignore_key); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_reserve(BlockDriverState *bs, uint64_t key, BlockPrType type); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_release(BlockDriverState *bs, uint64_t key, BlockPrType type); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_clear(BlockDriverState *bs, uint64_t key); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_preempt(BlockDriverState *bs, uint64_t cr_key, uint64_t pr_key, + BlockPrType type, bool abort); + /* Ensure contents are flushed to disk. */ int coroutine_fn GRAPH_RDLOCK bdrv_co_flush(BlockDriverState *bs); diff --git a/include/block/block_int-common.h b/include/block/block_int-common.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/block_int-common.h +++ b/include/block/block_int-common.h @@ -XXX,XX +XXX,XX @@ struct BlockDriver { int coroutine_fn GRAPH_RDLOCK_PTR (*bdrv_co_ioctl)( BlockDriverState *bs, unsigned long int req, void *buf); + /* + * Persistent reservation series api. + * Please refer to chapter 5.9 of SCSI Primary Commands - 4 or + * chapter 7 of NVMe Base Specification 2.0. + * + * The block layer driver should implement all the following APIs + * or none at all, including: bdrv_co_pr_read_keys, + * bdrv_co_pr_read_reservation, bdrv_co_pr_register, + * bdrv_co_pr_reserve, bdrv_co_pr_release, + * bdrv_co_pr_clear and bdrv_co_pr_preempt. + * + * Read the registered keys and return them in the @keys. + * @generation: The generation of the reservation key. + * @num_keys: The maximum number of keys that can be transmitted. + * @keys: Registered keys array. + * + * On success, store generation in @generation and store keys @keys + * and return the number of @keys. + * On failure return -errno. + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_read_keys)( + BlockDriverState *bs, uint32_t *generation, + uint32_t num_keys, uint64_t *keys); + /* + * Read the reservation key and store it in the @key. + * @generation: The generation of the reservation key. + * @key: The reservation key. + * @type: Type of the reservation key. + * + * On success, store generation in @generation, store the + * reservation key in @key and return the number of @key + * which used to determine whether the reservation key exists. + * On failure return -errno. + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_read_reservation)( + BlockDriverState *bs, uint32_t *generation, + uint64_t *key, BlockPrType *type); + /* + * Register, unregister, or replace a reservation key. + * @old_key: The current reservation key associated with the host. + * @new_key: The new reservation Key. + * @type: Type of the reservation key. + * @ignore_key: Ignore or not @old_key. + * @ptpl: Whether to support Persist Through Power Loss(PTPL). + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_register)( + BlockDriverState *bs, uint64_t old_key, + uint64_t new_key, BlockPrType type, + bool ptpl, bool ignore_key); + /* + * Acquire a reservation on a host. + * @key: The current reservation key associated with the host. + * @type: Type of the reservation key. + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_reserve)( + BlockDriverState *bs, uint64_t key, BlockPrType type); + /* + * Release a reservation on a host. + * @key: The current reservation key associated with the host. + * @type: Type of the reservation key. + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_release)( + BlockDriverState *bs, uint64_t key, BlockPrType type); + /** + * Clear reservations on a host. + * @key: The current reservation key associated with the host. + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_clear)( + BlockDriverState *bs, uint64_t key); + /* + * Preempt a reservation held on a host. + * @cr_key: The current reservation key associated with the host. + * @pr_key: The preempt reservation Key which to be + * unregistered from the namespace. + * @type: Type of the reservation key. + * @abort: Whether to abort a reservation held on a host. + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_preempt)( + BlockDriverState *bs, uint64_t cr_key, + uint64_t pr_key, BlockPrType type, bool abort); + /* * Returns 0 for completed check, -errno for internal errors. * The check results are stored in result. @@ -XXX,XX +XXX,XX @@ typedef struct BlockLimits { uint32_t max_active_zones; uint32_t write_granularity; + + /* Persistent reservation capacities. */ + uint8_t pr_cap; } BlockLimits; typedef struct BdrvOpBlocker BdrvOpBlocker; diff --git a/include/sysemu/block-backend-io.h b/include/sysemu/block-backend-io.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/block-backend-io.h +++ b/include/sysemu/block-backend-io.h @@ -XXX,XX +XXX,XX @@ void blk_aio_cancel_async(BlockAIOCB *acb); BlockAIOCB *blk_aio_ioctl(BlockBackend *blk, unsigned long int req, void *buf, BlockCompletionFunc *cb, void *opaque); +BlockAIOCB *blk_aio_pr_read_keys(BlockBackend *blk, uint32_t *generation, + uint32_t num_keys, uint64_t *keys, + BlockCompletionFunc *cb, void *opaque); +BlockAIOCB *blk_aio_pr_read_reservation(BlockBackend *blk, uint32_t *generation, + uint64_t *key, BlockPrType *type, + BlockCompletionFunc *cb, void *opaque); +BlockAIOCB *blk_aio_pr_register(BlockBackend *blk, uint64_t old_key, + uint64_t new_key, BlockPrType type, + bool ptpl, bool ignore_key, + BlockCompletionFunc *cb, + void *opaque); +BlockAIOCB *blk_aio_pr_reserve(BlockBackend *blk, uint64_t key, + BlockPrType type, + BlockCompletionFunc *cb, + void *opaque); +BlockAIOCB *blk_aio_pr_release(BlockBackend *blk, uint64_t key, + BlockPrType type, BlockCompletionFunc *cb, + void *opaque); +BlockAIOCB *blk_aio_pr_clear(BlockBackend *blk, uint64_t key, + BlockCompletionFunc *cb, void *opaque); +BlockAIOCB *blk_aio_pr_preempt(BlockBackend *blk, uint64_t cr_key, + uint64_t pr_key, BlockPrType type, bool abort, + BlockCompletionFunc *cb, void *opaque); + void blk_inc_in_flight(BlockBackend *blk); void blk_dec_in_flight(BlockBackend *blk); -- 2.20.1 Add persistent reservation in/out operations for raw driver. The following methods are implemented: bdrv_co_pr_read_keys, bdrv_co_pr_read_reservation, bdrv_co_pr_register, bdrv_co_pr_reserve, bdrv_co_pr_release, bdrv_co_pr_clear and bdrv_co_pr_preempt. Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com> Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> --- block/raw-format.c | 56 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 56 insertions(+) diff --git a/block/raw-format.c b/block/raw-format.c index XXXXXXX..XXXXXXX 100644 --- a/block/raw-format.c +++ b/block/raw-format.c @@ -XXX,XX +XXX,XX @@ raw_co_ioctl(BlockDriverState *bs, unsigned long int req, void *buf) return bdrv_co_ioctl(bs->file->bs, req, buf); } +static int coroutine_fn GRAPH_RDLOCK +raw_co_pr_read_keys(BlockDriverState *bs, uint32_t *generation, + uint32_t num_keys, uint64_t *keys) +{ + + return bdrv_co_pr_read_keys(bs->file->bs, generation, num_keys, keys); +} + +static int coroutine_fn GRAPH_RDLOCK +raw_co_pr_read_reservation(BlockDriverState *bs, uint32_t *generation, + uint64_t *key, BlockPrType *type) +{ + return bdrv_co_pr_read_reservation(bs->file->bs, generation, key, type); +} + +static int coroutine_fn GRAPH_RDLOCK +raw_co_pr_register(BlockDriverState *bs, uint64_t old_key, + uint64_t new_key, BlockPrType type, + bool ptpl, bool ignore_key) +{ + return bdrv_co_pr_register(bs->file->bs, old_key, new_key, + type, ptpl, ignore_key); +} + +static int coroutine_fn GRAPH_RDLOCK +raw_co_pr_reserve(BlockDriverState *bs, uint64_t key, BlockPrType type) +{ + return bdrv_co_pr_reserve(bs->file->bs, key, type); +} + +static int coroutine_fn GRAPH_RDLOCK +raw_co_pr_release(BlockDriverState *bs, uint64_t key, BlockPrType type) +{ + return bdrv_co_pr_release(bs->file->bs, key, type); +} + +static int coroutine_fn GRAPH_RDLOCK +raw_co_pr_clear(BlockDriverState *bs, uint64_t key) +{ + return bdrv_co_pr_clear(bs->file->bs, key); +} + +static int coroutine_fn GRAPH_RDLOCK +raw_co_pr_preempt(BlockDriverState *bs, uint64_t old_key, + uint64_t new_key, BlockPrType type, bool abort) +{ + return bdrv_co_pr_preempt(bs->file->bs, old_key, new_key, type, abort); +} + static int GRAPH_RDLOCK raw_has_zero_init(BlockDriverState *bs) { return bdrv_has_zero_init(bs->file->bs); @@ -XXX,XX +XXX,XX @@ BlockDriver bdrv_raw = { .strong_runtime_opts = raw_strong_runtime_opts, .mutable_opts = mutable_opts, .bdrv_cancel_in_flight = raw_cancel_in_flight, + .bdrv_co_pr_read_keys = raw_co_pr_read_keys, + .bdrv_co_pr_read_reservation = raw_co_pr_read_reservation, + .bdrv_co_pr_register = raw_co_pr_register, + .bdrv_co_pr_reserve = raw_co_pr_reserve, + .bdrv_co_pr_release = raw_co_pr_release, + .bdrv_co_pr_clear = raw_co_pr_clear, + .bdrv_co_pr_preempt = raw_co_pr_preempt, }; static void bdrv_raw_init(void) -- 2.20.1 Add constants for the persistent reservation in/out protocol in the scsi/constant module. The constants include the persistent reservation command, type, and scope values defined in sections 6.13 and 6.14 of the SCSI Primary Commands-4 (SPC-4) specification. Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com> Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> --- include/scsi/constants.h | 52 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 52 insertions(+) diff --git a/include/scsi/constants.h b/include/scsi/constants.h index XXXXXXX..XXXXXXX 100644 --- a/include/scsi/constants.h +++ b/include/scsi/constants.h @@ -XXX,XX +XXX,XX @@ #define IDENT_DESCR_TGT_DESCR_SIZE 32 #define XCOPY_BLK2BLK_SEG_DESC_SIZE 28 +typedef enum { + SCSI_PR_WRITE_EXCLUSIVE = 0x01, + SCSI_PR_EXCLUSIVE_ACCESS = 0x03, + SCSI_PR_WRITE_EXCLUSIVE_REGS_ONLY = 0x05, + SCSI_PR_EXCLUSIVE_ACCESS_REGS_ONLY = 0x06, + SCSI_PR_WRITE_EXCLUSIVE_ALL_REGS = 0x07, + SCSI_PR_EXCLUSIVE_ACCESS_ALL_REGS = 0x08, +} SCSIPrType; + +typedef enum { + SCSI_PR_LU_SCOPE = 0x00, +} SCSIPrScope; + +typedef enum { + SCSI_PR_OUT_REGISTER = 0x0, + SCSI_PR_OUT_RESERVE = 0x1, + SCSI_PR_OUT_RELEASE = 0x2, + SCSI_PR_OUT_CLEAR = 0x3, + SCSI_PR_OUT_PREEMPT = 0x4, + SCSI_PR_OUT_PREEMPT_AND_ABORT = 0x5, + SCSI_PR_OUT_REG_AND_IGNORE_KEY = 0x6, + SCSI_PR_OUT_REG_AND_MOVE = 0x7, +} SCSIPrOutAction; + +typedef enum { + SCSI_PR_IN_READ_KEYS = 0x0, + SCSI_PR_IN_READ_RESERVATION = 0x1, + SCSI_PR_IN_REPORT_CAPABILITIES = 0x2, +} SCSIPrInAction; + +typedef enum { + /* Exclusive Access All Registrants reservation type */ + SCSI_PR_CAP_EX_AC_AR = 1 << 0, + /* Write Exclusive reservation type */ + SCSI_PR_CAP_WR_EX = 1 << 9, + /* Exclusive Access reservation type */ + SCSI_PR_CAP_EX_AC = 1 << 11, + /* Write Exclusive Registrants Only reservation type */ + SCSI_PR_CAP_WR_EX_RO = 1 << 13, + /* Exclusive Access Registrants Only reservation type */ + SCSI_PR_CAP_EX_AC_RO = 1 << 14, + /* Write Exclusive All Registrants reservation type */ + SCSI_PR_CAP_WR_EX_AR = 1 << 15, + + SCSI_PR_CAP_ALL = (SCSI_PR_CAP_EX_AC_AR | + SCSI_PR_CAP_WR_EX | + SCSI_PR_CAP_EX_AC | + SCSI_PR_CAP_WR_EX_RO | + SCSI_PR_CAP_EX_AC_RO | + SCSI_PR_CAP_WR_EX_AR), +} SCSIPrCap; + #endif -- 2.20.1 This commit introduces two helper functions that facilitate the conversion between the persistent reservation types used in the SCSI protocol and those used in the block layer. Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com> Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> --- include/scsi/utils.h | 8 +++++ scsi/utils.c | 81 ++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 89 insertions(+) diff --git a/include/scsi/utils.h b/include/scsi/utils.h index XXXXXXX..XXXXXXX 100644 --- a/include/scsi/utils.h +++ b/include/scsi/utils.h @@ -XXX,XX +XXX,XX @@ #ifndef SCSI_UTILS_H #define SCSI_UTILS_H +#include "block/block-common.h" +#include "scsi/constants.h" #ifdef CONFIG_LINUX #include <scsi/sg.h> #endif @@ -XXX,XX +XXX,XX @@ uint32_t scsi_data_cdb_xfer(uint8_t *buf); uint32_t scsi_cdb_xfer(uint8_t *buf); int scsi_cdb_length(uint8_t *buf); +BlockPrType scsi_pr_type_to_block(SCSIPrType type); +SCSIPrType block_pr_type_to_scsi(BlockPrType type); + +uint8_t scsi_pr_cap_to_block(uint16_t scsi_pr_cap); +uint16_t block_pr_cap_to_scsi(uint8_t block_pr_cap); + /* Linux SG_IO interface. */ #ifdef CONFIG_LINUX #define SG_ERR_DRIVER_TIMEOUT 0x06 diff --git a/scsi/utils.c b/scsi/utils.c index XXXXXXX..XXXXXXX 100644 --- a/scsi/utils.c +++ b/scsi/utils.c @@ -XXX,XX +XXX,XX @@ int scsi_sense_from_host_status(uint8_t host_status, } return GOOD; } + +BlockPrType scsi_pr_type_to_block(SCSIPrType type) +{ + switch (type) { + case SCSI_PR_WRITE_EXCLUSIVE: + return BLK_PR_WRITE_EXCLUSIVE; + case SCSI_PR_EXCLUSIVE_ACCESS: + return BLK_PR_EXCLUSIVE_ACCESS; + case SCSI_PR_WRITE_EXCLUSIVE_REGS_ONLY: + return BLK_PR_WRITE_EXCLUSIVE_REGS_ONLY; + case SCSI_PR_EXCLUSIVE_ACCESS_REGS_ONLY: + return BLK_PR_EXCLUSIVE_ACCESS_REGS_ONLY; + case SCSI_PR_WRITE_EXCLUSIVE_ALL_REGS: + return BLK_PR_WRITE_EXCLUSIVE_ALL_REGS; + case SCSI_PR_EXCLUSIVE_ACCESS_ALL_REGS: + return BLK_PR_EXCLUSIVE_ACCESS_ALL_REGS; + } + + return 0; +} + +SCSIPrType block_pr_type_to_scsi(BlockPrType type) +{ + switch (type) { + case BLK_PR_WRITE_EXCLUSIVE: + return SCSI_PR_WRITE_EXCLUSIVE; + case BLK_PR_EXCLUSIVE_ACCESS: + return SCSI_PR_EXCLUSIVE_ACCESS; + case BLK_PR_WRITE_EXCLUSIVE_REGS_ONLY: + return SCSI_PR_WRITE_EXCLUSIVE_REGS_ONLY; + case BLK_PR_EXCLUSIVE_ACCESS_REGS_ONLY: + return SCSI_PR_EXCLUSIVE_ACCESS_REGS_ONLY; + case BLK_PR_WRITE_EXCLUSIVE_ALL_REGS: + return SCSI_PR_WRITE_EXCLUSIVE_ALL_REGS; + case BLK_PR_EXCLUSIVE_ACCESS_ALL_REGS: + return SCSI_PR_EXCLUSIVE_ACCESS_ALL_REGS; + } + + return 0; +} + + +uint8_t scsi_pr_cap_to_block(uint16_t scsi_pr_cap) +{ + uint8_t res = 0; + + res |= (scsi_pr_cap & SCSI_PR_CAP_WR_EX) ? + BLK_PR_CAP_WR_EX : 0; + res |= (scsi_pr_cap & SCSI_PR_CAP_EX_AC) ? + BLK_PR_CAP_EX_AC : 0; + res |= (scsi_pr_cap & SCSI_PR_CAP_WR_EX_RO) ? + BLK_PR_CAP_WR_EX_RO : 0; + res |= (scsi_pr_cap & SCSI_PR_CAP_EX_AC_RO) ? + BLK_PR_CAP_EX_AC_RO : 0; + res |= (scsi_pr_cap & SCSI_PR_CAP_WR_EX_AR) ? + BLK_PR_CAP_WR_EX_AR : 0; + res |= (scsi_pr_cap & SCSI_PR_CAP_EX_AC_AR) ? + BLK_PR_CAP_EX_AC_AR : 0; + + return res; +} + +uint16_t block_pr_cap_to_scsi(uint8_t block_pr_cap) +{ + uint16_t res = 0; + + res |= (block_pr_cap & BLK_PR_CAP_WR_EX) ? + SCSI_PR_CAP_WR_EX : 0; + res |= (block_pr_cap & BLK_PR_CAP_EX_AC) ? + SCSI_PR_CAP_EX_AC : 0; + res |= (block_pr_cap & BLK_PR_CAP_WR_EX_RO) ? + SCSI_PR_CAP_WR_EX_RO : 0; + res |= (block_pr_cap & BLK_PR_CAP_EX_AC_RO) ? + SCSI_PR_CAP_EX_AC_RO : 0; + res |= (block_pr_cap & BLK_PR_CAP_WR_EX_AR) ? + SCSI_PR_CAP_WR_EX_AR : 0; + res |= (block_pr_cap & BLK_PR_CAP_EX_AC_AR) ? + SCSI_PR_CAP_EX_AC_AR : 0; + + return res; +} -- 2.20.1 Add persistent reservation in/out operations in the SCSI device layer. By introducing the persistent reservation in/out api, this enables the SCSI device to perform reservation-related tasks, including querying keys, querying reservation status, registering reservation keys, initiating and releasing reservations, as well as clearing and preempting reservations held by other keys. These operations are crucial for management and control of shared storage resources in a persistent manner. Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com> Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> --- hw/scsi/scsi-disk.c | 352 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 352 insertions(+) diff --git a/hw/scsi/scsi-disk.c b/hw/scsi/scsi-disk.c index XXXXXXX..XXXXXXX 100644 --- a/hw/scsi/scsi-disk.c +++ b/hw/scsi/scsi-disk.c @@ -XXX,XX +XXX,XX @@ #include "migration/vmstate.h" #include "hw/scsi/emulation.h" #include "scsi/constants.h" +#include "scsi/utils.h" #include "sysemu/block-backend.h" #include "sysemu/blockdev.h" #include "hw/block/block.h" @@ -XXX,XX +XXX,XX @@ #include "qemu/cutils.h" #include "trace.h" #include "qom/object.h" +#include "block/block_int.h" #ifdef __linux #include <scsi/sg.h> @@ -XXX,XX +XXX,XX @@ static void scsi_disk_emulate_read_data(SCSIRequest *req) scsi_req_complete(&r->req, GOOD); } +typedef struct SCSIPrReadKeys { + uint32_t generation; + uint32_t num_keys; + uint64_t *keys; + void *req; +} SCSIPrReadKeys; + +typedef struct SCSIPrReadReservation { + uint32_t generation; + uint64_t key; + BlockPrType type; + void *req; +} SCSIPrReadReservation; + +static void scsi_pr_read_keys_complete(void *opaque, int ret) +{ + int num_keys; + uint8_t *buf; + SCSIPrReadKeys *blk_keys = (SCSIPrReadKeys *)opaque; + SCSIDiskReq *r = (SCSIDiskReq *)blk_keys->req; + SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, r->req.dev); + + assert(blk_get_aio_context(s->qdev.conf.blk) == + qemu_get_current_aio_context()); + + assert(r->req.aiocb != NULL); + r->req.aiocb = NULL; + + if (scsi_disk_req_check_error(r, ret, true)) { + goto done; + } + + buf = scsi_req_get_buf(&r->req); + num_keys = MIN(blk_keys->num_keys, ret); + blk_keys->generation = cpu_to_be32(blk_keys->generation); + memcpy(&buf[0], &blk_keys->generation, 4); + for (int i = 0; i < num_keys; i++) { + blk_keys->keys[i] = cpu_to_be64(blk_keys->keys[i]); + memcpy(&buf[8 + i * 8], &blk_keys->keys[i], 8); + } + num_keys = cpu_to_be32(num_keys * 8); + memcpy(&buf[4], &num_keys, 4); + + scsi_req_data(&r->req, r->buflen); +done: + scsi_req_unref(&r->req); + g_free(blk_keys->keys); + g_free(blk_keys); +} + +static int scsi_disk_emulate_pr_read_keys(SCSIRequest *req) +{ + SCSIPrReadKeys *blk_keys; + SCSIDiskReq *r = DO_UPCAST(SCSIDiskReq, req, req); + SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, req->dev); + int buflen = MIN(r->req.cmd.xfer, r->buflen); + int num_keys = (buflen - sizeof(uint32_t) * 2) / sizeof(uint64_t); + + blk_keys = g_new0(SCSIPrReadKeys, 1); + blk_keys->generation = 0; + /* num_keys is the maximum number of keys that can be transmitted */ + blk_keys->num_keys = num_keys; + blk_keys->keys = g_malloc(sizeof(uint64_t) * num_keys); + blk_keys->req = r; + + /* The request is used as the AIO opaque value, so add a ref. */ + scsi_req_ref(&r->req); + r->req.aiocb = blk_aio_pr_read_keys(s->qdev.conf.blk, &blk_keys->generation, + blk_keys->num_keys, blk_keys->keys, + scsi_pr_read_keys_complete, blk_keys); + return 0; +} + +static void scsi_pr_read_reservation_complete(void *opaque, int ret) +{ + uint8_t *buf; + uint32_t additional_len = 0; + SCSIPrReadReservation *blk_rsv = (SCSIPrReadReservation *)opaque; + SCSIDiskReq *r = (SCSIDiskReq *)blk_rsv->req; + SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, r->req.dev); + + assert(blk_get_aio_context(s->qdev.conf.blk) == + qemu_get_current_aio_context()); + + assert(r->req.aiocb != NULL); + r->req.aiocb = NULL; + + if (scsi_disk_req_check_error(r, ret, true)) { + goto done; + } + + buf = scsi_req_get_buf(&r->req); + blk_rsv->generation = cpu_to_be32(blk_rsv->generation); + memcpy(&buf[0], &blk_rsv->generation, 4); + if (ret) { + additional_len = cpu_to_be32(16); + blk_rsv->key = cpu_to_be64(blk_rsv->key); + memcpy(&buf[8], &blk_rsv->key, 8); + buf[21] = block_pr_type_to_scsi(blk_rsv->type) & 0xf; + } else { + additional_len = cpu_to_be32(0); + } + + memcpy(&buf[4], &additional_len, 4); + scsi_req_data(&r->req, r->buflen); + +done: + scsi_req_unref(&r->req); + g_free(blk_rsv); +} + +static int scsi_disk_emulate_pr_read_reservation(SCSIRequest *req) +{ + SCSIPrReadReservation *blk_rsv; + SCSIDiskReq *r = DO_UPCAST(SCSIDiskReq, req, req); + SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, req->dev); + + blk_rsv = g_malloc(sizeof(*blk_rsv)); + blk_rsv->generation = 0; + blk_rsv->key = 0; + blk_rsv->type = 0; + blk_rsv->req = r; + + /* The request is used as the AIO opaque value, so add a ref. */ + scsi_req_ref(&r->req); + r->req.aiocb = blk_aio_pr_read_reservation(s->qdev.conf.blk, + &blk_rsv->generation, &blk_rsv->key, &blk_rsv->type, + scsi_pr_read_reservation_complete, blk_rsv); + return 0; +} + +static int scsi_disk_emulate_pr_report_capabilities(SCSIRequest *req) +{ + SCSIDiskReq *r = DO_UPCAST(SCSIDiskReq, req, req); + SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, req->dev); + BlockDriverState *bs = blk_bs(s->qdev.conf.blk); + uint8_t blk_pr_cap = bs->file->bs->bl.pr_cap; + uint8_t *buf = scsi_req_get_buf(req); + uint16_t pr_cap; + /* + * according to response of report_capabilities + * command of chapter 6.13 of spc4 + */ + int len = sizeof(uint16_t) * 2 + sizeof(uint8_t) * 4; + + if (len > r->buflen) { + return -EINVAL; + } + + len = cpu_to_be16(len); + memcpy(&buf[0], &len, 2); + /* enable PTPL_C */ + buf[2] = 1; + buf[3] = (blk_pr_cap & BLK_PR_CAP_PTPL) ? 1 : 0; + /* enable TMV */ + buf[3] |= 1 << 7; + + /* enable persistent reservation type */ + pr_cap = block_pr_cap_to_scsi(blk_pr_cap); + buf[4] = pr_cap >> 8 & 0xff; + buf[5] = pr_cap & 0xff; + + scsi_req_data(&r->req, len); + return 0; +} + +static void scsi_aio_pr_complete(void *opaque, int ret) +{ + SCSIDiskReq *r = (SCSIDiskReq *)opaque; + SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, r->req.dev); + + /* The request must only run in the BlockBackend's AioContext */ + assert(blk_get_aio_context(s->qdev.conf.blk) == + qemu_get_current_aio_context()); + + assert(r->req.aiocb != NULL); + r->req.aiocb = NULL; + + if (scsi_disk_req_check_error(r, ret, true)) { + goto done; + } + + scsi_req_complete(&r->req, GOOD); + +done: + scsi_req_unref(&r->req); +} + +static int scsi_disk_emulate_pr_register(SCSIDiskReq *r, uint64_t r_key, + uint64_t sa_key, SCSIPrType type, + uint8_t aptpl, bool ignore_key) +{ + SCSIRequest *req = &r->req; + SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, req->dev); + + /* The request is used as the AIO opaque value, so add a ref. */ + scsi_req_ref(&r->req); + r->req.aiocb = blk_aio_pr_register(s->qdev.conf.blk, r_key, sa_key, + scsi_pr_type_to_block(type), + aptpl ? true : false, + ignore_key, scsi_aio_pr_complete, r); + + return 0; +} + +static int scsi_disk_emulate_pr_reserve(SCSIDiskReq *r, uint64_t r_key, + SCSIPrType type) +{ + SCSIRequest *req = &r->req; + SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, req->dev); + + /* The request is used as the AIO opaque value, so add a ref. */ + scsi_req_ref(&r->req); + r->req.aiocb = blk_aio_pr_reserve(s->qdev.conf.blk, r_key, + scsi_pr_type_to_block(type), + scsi_aio_pr_complete, r); + + return 0; +} + +static int scsi_disk_emulate_pr_release(SCSIDiskReq *r, uint64_t r_key, + SCSIPrType type) +{ + SCSIRequest *req = &r->req; + SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, req->dev); + + /* The request is used as the AIO opaque value, so add a ref. */ + scsi_req_ref(&r->req); + r->req.aiocb = blk_aio_pr_release(s->qdev.conf.blk, r_key, + scsi_pr_type_to_block(type), + scsi_aio_pr_complete, r); + + return 0; +} + +static int scsi_disk_emulate_pr_clear(SCSIDiskReq *r, uint64_t r_key) +{ + SCSIRequest *req = &r->req; + SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, req->dev); + + /* The request is used as the AIO opaque value, so add a ref. */ + scsi_req_ref(&r->req); + r->req.aiocb = blk_aio_pr_clear(s->qdev.conf.blk, r_key, + scsi_aio_pr_complete, r); + + return 0; +} + +static int scsi_disk_emulate_pr_preempt(SCSIDiskReq *r, uint64_t r_key, + uint64_t sa_key, SCSIPrType type, + bool abort) +{ + SCSIRequest *req = &r->req; + SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, req->dev); + + /* The request is used as the AIO opaque value, so add a ref. */ + scsi_req_ref(&r->req); + r->req.aiocb = blk_aio_pr_preempt(s->qdev.conf.blk, r_key, sa_key, + scsi_pr_type_to_block(type), abort, + scsi_aio_pr_complete, r); + + return 0; +} + +static int scsi_disk_emulate_pr_in(SCSIRequest *req) +{ + int rc; + SCSIPrInAction action = req->cmd.buf[1] & 0x1f; + + switch (action) { + case SCSI_PR_IN_READ_KEYS: + rc = scsi_disk_emulate_pr_read_keys(req); + break; + case SCSI_PR_IN_READ_RESERVATION: + rc = scsi_disk_emulate_pr_read_reservation(req); + break; + case SCSI_PR_IN_REPORT_CAPABILITIES: + rc = scsi_disk_emulate_pr_report_capabilities(req); + break; + default: + return -ENOTSUP; + } + + return rc; +} + +static int scsi_disk_emulate_pr_out(SCSIDiskReq *r, uint8_t *inbuf) +{ + int rc; + uint8_t aptpl; + uint64_t r_key, sa_key; + SCSIPrOutAction action; + SCSIPrScope scope; + SCSIPrType type; + SCSIRequest *req = &r->req; + + memcpy(&r_key, &inbuf[0], 8); + r_key = be64_to_cpu(r_key); + memcpy(&sa_key, &inbuf[8], 8); + sa_key = be64_to_cpu(sa_key); + memcpy(&aptpl, &inbuf[20], 1); + aptpl = aptpl & 0x01; + action = req->cmd.buf[1] & 0x1f; + scope = (req->cmd.buf[2] >> 4) & 0x0f; + type = req->cmd.buf[2] & 0x0f; + + if (scope != SCSI_PR_LU_SCOPE) { + return -ENOTSUP; + } + + switch (action) { + case SCSI_PR_OUT_REGISTER: + rc = scsi_disk_emulate_pr_register(r, r_key, sa_key, type, + aptpl, false); + break; + case SCSI_PR_OUT_REG_AND_IGNORE_KEY: + rc = scsi_disk_emulate_pr_register(r, r_key, sa_key, type, aptpl, true); + break; + case SCSI_PR_OUT_RESERVE: + rc = scsi_disk_emulate_pr_reserve(r, r_key, type); + break; + case SCSI_PR_OUT_RELEASE: + rc = scsi_disk_emulate_pr_release(r, r_key, type); + break; + case SCSI_PR_OUT_CLEAR: + rc = scsi_disk_emulate_pr_clear(r, r_key); + break; + case SCSI_PR_OUT_PREEMPT: + rc = scsi_disk_emulate_pr_preempt(r, r_key, sa_key, type, false); + break; + case SCSI_PR_OUT_PREEMPT_AND_ABORT: + rc = scsi_disk_emulate_pr_preempt(r, r_key, sa_key, type, true); + break; + default: + return -ENOTSUP; + } + + return rc; +} + static int scsi_disk_check_mode_select(SCSIDiskState *s, int page, uint8_t *inbuf, int inlen) { @@ -XXX,XX +XXX,XX @@ static void scsi_disk_emulate_write_data(SCSIRequest *req) scsi_req_complete(&r->req, GOOD); break; + case PERSISTENT_RESERVE_OUT: + scsi_disk_emulate_pr_out(r, r->iov.iov_base); + break; default: abort(); } @@ -XXX,XX +XXX,XX @@ static int32_t scsi_disk_emulate_command(SCSIRequest *req, uint8_t *buf) case FORMAT_UNIT: trace_scsi_disk_emulate_command_FORMAT_UNIT(r->req.cmd.xfer); break; + case PERSISTENT_RESERVE_OUT: + break; + case PERSISTENT_RESERVE_IN: + scsi_disk_emulate_pr_in(req); + return 0; default: trace_scsi_disk_emulate_command_UNKNOWN(buf[0], scsi_command_name(buf[0])); @@ -XXX,XX +XXX,XX @@ static const SCSIReqOps *const scsi_disk_reqops_dispatch[256] = { [VERIFY_12] = &scsi_disk_emulate_reqops, [VERIFY_16] = &scsi_disk_emulate_reqops, [FORMAT_UNIT] = &scsi_disk_emulate_reqops, + [PERSISTENT_RESERVE_IN] = &scsi_disk_emulate_reqops, + [PERSISTENT_RESERVE_OUT] = &scsi_disk_emulate_reqops, [READ_6] = &scsi_disk_dma_reqops, [READ_10] = &scsi_disk_dma_reqops, -- 2.20.1 Add constants for the NVMe persistent command protocol. The constants include the reservation command opcode and reservation type values defined in section 7 of the NVMe 2.0 specification. Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com> Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> --- include/block/nvme.h | 61 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 61 insertions(+) diff --git a/include/block/nvme.h b/include/block/nvme.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/nvme.h +++ b/include/block/nvme.h @@ -XXX,XX +XXX,XX @@ enum NvmeIoCommands { NVME_CMD_WRITE_ZEROES = 0x08, NVME_CMD_DSM = 0x09, NVME_CMD_VERIFY = 0x0c, + NVME_CMD_RESV_REGISTER = 0x0d, + NVME_CMD_RESV_REPORT = 0x0e, + NVME_CMD_RESV_ACQUIRE = 0x11, + NVME_CMD_RESV_RELEASE = 0x15, NVME_CMD_IO_MGMT_RECV = 0x12, NVME_CMD_COPY = 0x19, NVME_CMD_IO_MGMT_SEND = 0x1d, @@ -XXX,XX +XXX,XX @@ enum NvmeIoCommands { NVME_CMD_ZONE_APPEND = 0x7d, }; +typedef enum { + NVME_RESV_REGISTER_ACTION_REGISTER = 0x00, + NVME_RESV_REGISTER_ACTION_UNREGISTER = 0x01, + NVME_RESV_REGISTER_ACTION_REPLACE = 0x02, +} NVME_RESV_REGISTER_ACTION; + +typedef enum { + NVME_RESV_RELEASE_ACTION_RELEASE = 0x00, + NVME_RESV_RELEASE_ACTION_CLEAR = 0x01, +} NVME_RESV_RELEASE_ACTION; + +typedef enum { + NVME_RESV_ACQUIRE_ACTION_ACQUIRE = 0x00, + NVME_RESV_ACQUIRE_ACTION_PREEMPT = 0x01, + NVME_RESV_ACQUIRE_ACTION_PREEMPT_AND_ABORT = 0x02, +} NVME_RESV_ACQUIRE_ACTION; + +typedef enum { + NVME_RESV_WRITE_EXCLUSIVE = 0x01, + NVME_RESV_EXCLUSIVE_ACCESS = 0x02, + NVME_RESV_WRITE_EXCLUSIVE_REGS_ONLY = 0x03, + NVME_RESV_EXCLUSIVE_ACCESS_REGS_ONLY = 0x04, + NVME_RESV_WRITE_EXCLUSIVE_ALL_REGS = 0x05, + NVME_RESV_EXCLUSIVE_ACCESS_ALL_REGS = 0x06, +} NVMEResvType; + +typedef enum { + NVME_RESV_PTPL_NO_CHANGE = 0x00, + NVME_RESV_PTPL_DISABLE = 0x02, + NVME_RESV_PTPL_ENABLE = 0x03, +} NVMEResvPTPL; + +typedef enum NVMEPrCap { + /* Persist Through Power Loss */ + NVME_PR_CAP_PTPL = 1 << 0, + /* Write Exclusive reservation type */ + NVME_PR_CAP_WR_EX = 1 << 1, + /* Exclusive Access reservation type */ + NVME_PR_CAP_EX_AC = 1 << 2, + /* Write Exclusive Registrants Only reservation type */ + NVME_PR_CAP_WR_EX_RO = 1 << 3, + /* Exclusive Access Registrants Only reservation type */ + NVME_PR_CAP_EX_AC_RO = 1 << 4, + /* Write Exclusive All Registrants reservation type */ + NVME_PR_CAP_WR_EX_AR = 1 << 5, + /* Exclusive Access All Registrants reservation type */ + NVME_PR_CAP_EX_AC_AR = 1 << 6, + + NVME_PR_CAP_ALL = (NVME_PR_CAP_PTPL | + NVME_PR_CAP_WR_EX | + NVME_PR_CAP_EX_AC | + NVME_PR_CAP_WR_EX_RO | + NVME_PR_CAP_EX_AC_RO | + NVME_PR_CAP_WR_EX_AR | + NVME_PR_CAP_EX_AC_AR), +} NVMEPrCap; + typedef struct QEMU_PACKED NvmeDeleteQ { uint8_t opcode; uint8_t flags; -- 2.20.1 This commit introduces two helper functions that facilitate the conversion between the reservation types used in the NVME protocol and those used in the block layer. Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com> Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> --- hw/nvme/nvme.h | 40 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 40 insertions(+) diff --git a/hw/nvme/nvme.h b/hw/nvme/nvme.h index XXXXXXX..XXXXXXX 100644 --- a/hw/nvme/nvme.h +++ b/hw/nvme/nvme.h @@ -XXX,XX +XXX,XX @@ static inline const char *nvme_io_opc_str(uint8_t opc) } } +static inline NVMEResvType block_pr_type_to_nvme(BlockPrType type) +{ + switch (type) { + case BLK_PR_WRITE_EXCLUSIVE: + return NVME_RESV_WRITE_EXCLUSIVE; + case BLK_PR_EXCLUSIVE_ACCESS: + return NVME_RESV_EXCLUSIVE_ACCESS; + case BLK_PR_WRITE_EXCLUSIVE_REGS_ONLY: + return NVME_RESV_WRITE_EXCLUSIVE_REGS_ONLY; + case BLK_PR_EXCLUSIVE_ACCESS_REGS_ONLY: + return NVME_RESV_EXCLUSIVE_ACCESS_REGS_ONLY; + case BLK_PR_WRITE_EXCLUSIVE_ALL_REGS: + return NVME_RESV_WRITE_EXCLUSIVE_ALL_REGS; + case BLK_PR_EXCLUSIVE_ACCESS_ALL_REGS: + return NVME_RESV_EXCLUSIVE_ACCESS_ALL_REGS; + } + + return 0; +} + +static inline BlockPrType nvme_pr_type_to_block(NVMEResvType type) +{ + switch (type) { + case NVME_RESV_WRITE_EXCLUSIVE: + return BLK_PR_WRITE_EXCLUSIVE; + case NVME_RESV_EXCLUSIVE_ACCESS: + return BLK_PR_EXCLUSIVE_ACCESS; + case NVME_RESV_WRITE_EXCLUSIVE_REGS_ONLY: + return BLK_PR_WRITE_EXCLUSIVE_REGS_ONLY; + case NVME_RESV_EXCLUSIVE_ACCESS_REGS_ONLY: + return BLK_PR_EXCLUSIVE_ACCESS_REGS_ONLY; + case NVME_RESV_WRITE_EXCLUSIVE_ALL_REGS: + return BLK_PR_WRITE_EXCLUSIVE_ALL_REGS; + case NVME_RESV_EXCLUSIVE_ACCESS_ALL_REGS: + return BLK_PR_EXCLUSIVE_ACCESS_ALL_REGS; + } + + return 0; +} + typedef struct NvmeSQueue { struct NvmeCtrl *ctrl; uint16_t sqid; -- 2.20.1 This commit enables ONCS to support the reservation function at the controller level. It also lays the groundwork for detecting and enabling the reservation function on a per-namespace basis in RESCAP. Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com> Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> --- hw/nvme/ctrl.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/hw/nvme/ctrl.c b/hw/nvme/ctrl.c index XXXXXXX..XXXXXXX 100644 --- a/hw/nvme/ctrl.c +++ b/hw/nvme/ctrl.c @@ -XXX,XX +XXX,XX @@ static void nvme_init_ctrl(NvmeCtrl *n, PCIDevice *pci_dev) id->nn = cpu_to_le32(NVME_MAX_NAMESPACES); id->oncs = cpu_to_le16(NVME_ONCS_WRITE_ZEROES | NVME_ONCS_TIMESTAMP | NVME_ONCS_FEATURES | NVME_ONCS_DSM | - NVME_ONCS_COMPARE | NVME_ONCS_COPY); + NVME_ONCS_COMPARE | NVME_ONCS_COPY | + NVME_ONCS_RESRVATIONS); /* * NOTE: If this device ever supports a command set that does NOT use 0x0 -- 2.20.1 This commit enables the rescap function in the namespace by detecting the supported reservation function in the backend driver. Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com> Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> --- hw/nvme/ns.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/hw/nvme/ns.c b/hw/nvme/ns.c index XXXXXXX..XXXXXXX 100644 --- a/hw/nvme/ns.c +++ b/hw/nvme/ns.c @@ -XXX,XX +XXX,XX @@ #include "qemu/bitops.h" #include "sysemu/sysemu.h" #include "sysemu/block-backend.h" +#include "block/block_int.h" #include "nvme.h" #include "trace.h" @@ -XXX,XX +XXX,XX @@ void nvme_ns_init_format(NvmeNamespace *ns) } id_ns->npda = id_ns->npdg = npdg - 1; + + /* + * The persistent reservation capacities of block + * and nvme are currently defined the same. + * If there are subsequent changes, this part needs to be changed. + */ + id_ns->rescap = blk_bs(ns->blkconf.blk)->file->bs->bl.pr_cap; } static int nvme_ns_init(NvmeNamespace *ns, Error **errp) -- 2.20.1 Add reservation acquire, reservation register, reservation release and reservation report commands in the nvme device layer. By introducing these commands, this enables the nvme device to perform reservation-related tasks, including querying keys, querying reservation status, registering reservation keys, initiating and releasing reservations, as well as clearing and preempting reservations held by other keys. These commands are crucial for management and control of shared storage resources in a persistent manner. Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com> Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> --- hw/nvme/ctrl.c | 321 ++++++++++++++++++++++++++++++++++++++++++- hw/nvme/nvme.h | 4 + include/block/nvme.h | 38 +++++ 3 files changed, 362 insertions(+), 1 deletion(-) diff --git a/hw/nvme/ctrl.c b/hw/nvme/ctrl.c index XXXXXXX..XXXXXXX 100644 --- a/hw/nvme/ctrl.c +++ b/hw/nvme/ctrl.c @@ -XXX,XX +XXX,XX @@ static const uint32_t nvme_cse_iocs_nvm[256] = { [NVME_CMD_COMPARE] = NVME_CMD_EFF_CSUPP, [NVME_CMD_IO_MGMT_RECV] = NVME_CMD_EFF_CSUPP, [NVME_CMD_IO_MGMT_SEND] = NVME_CMD_EFF_CSUPP | NVME_CMD_EFF_LBCC, + [NVME_CMD_RESV_REGISTER] = NVME_CMD_EFF_CSUPP, + [NVME_CMD_RESV_REPORT] = NVME_CMD_EFF_CSUPP, + [NVME_CMD_RESV_ACQUIRE] = NVME_CMD_EFF_CSUPP, + [NVME_CMD_RESV_RELEASE] = NVME_CMD_EFF_CSUPP, }; static const uint32_t nvme_cse_iocs_zoned[256] = { @@ -XXX,XX +XXX,XX @@ static const uint32_t nvme_cse_iocs_zoned[256] = { [NVME_CMD_ZONE_APPEND] = NVME_CMD_EFF_CSUPP | NVME_CMD_EFF_LBCC, [NVME_CMD_ZONE_MGMT_SEND] = NVME_CMD_EFF_CSUPP | NVME_CMD_EFF_LBCC, [NVME_CMD_ZONE_MGMT_RECV] = NVME_CMD_EFF_CSUPP, + [NVME_CMD_RESV_REGISTER] = NVME_CMD_EFF_CSUPP, + [NVME_CMD_RESV_REPORT] = NVME_CMD_EFF_CSUPP, + [NVME_CMD_RESV_ACQUIRE] = NVME_CMD_EFF_CSUPP, + [NVME_CMD_RESV_RELEASE] = NVME_CMD_EFF_CSUPP, }; static void nvme_process_sq(void *opaque); @@ -XXX,XX +XXX,XX @@ static void nvme_aio_err(NvmeRequest *req, int ret) switch (req->cmd.opcode) { case NVME_CMD_READ: + case NVME_CMD_RESV_REPORT: status = NVME_UNRECOVERED_READ; break; case NVME_CMD_FLUSH: @@ -XXX,XX +XXX,XX @@ static void nvme_aio_err(NvmeRequest *req, int ret) case NVME_CMD_WRITE_ZEROES: case NVME_CMD_ZONE_APPEND: case NVME_CMD_COPY: + case NVME_CMD_RESV_REGISTER: + case NVME_CMD_RESV_ACQUIRE: + case NVME_CMD_RESV_RELEASE: status = NVME_WRITE_FAULT; break; default: @@ -XXX,XX +XXX,XX @@ static inline bool nvme_is_write(NvmeRequest *req) return rw->opcode == NVME_CMD_WRITE || rw->opcode == NVME_CMD_ZONE_APPEND || - rw->opcode == NVME_CMD_WRITE_ZEROES; + rw->opcode == NVME_CMD_WRITE_ZEROES || + rw->opcode == NVME_CMD_RESV_REGISTER || + rw->opcode == NVME_CMD_RESV_ACQUIRE || + rw->opcode == NVME_CMD_RESV_RELEASE; } static void nvme_misc_cb(void *opaque, int ret) @@ -XXX,XX +XXX,XX @@ static uint16_t nvme_verify(NvmeCtrl *n, NvmeRequest *req) return NVME_NO_COMPLETE; } +typedef struct NvmeKeyInfo { + uint64_t cr_key; + uint64_t nr_key; +} NvmeKeyInfo; + +static uint16_t nvme_resv_register(NvmeCtrl *n, NvmeRequest *req) +{ + int ret; + NvmeKeyInfo key_info; + NvmeNamespace *ns = req->ns; + uint32_t cdw10 = le32_to_cpu(req->cmd.cdw10); + bool ignore_key = cdw10 >> 3 & 0x1; + uint8_t action = cdw10 & 0x7; + uint8_t ptpl = cdw10 >> 30 & 0x3; + bool aptpl; + + switch (ptpl) { + case NVME_RESV_PTPL_NO_CHANGE: + aptpl = (ns->id_ns.rescap & NVME_PR_CAP_PTPL) ? true : false; + break; + case NVME_RESV_PTPL_DISABLE: + aptpl = false; + break; + case NVME_RESV_PTPL_ENABLE: + aptpl = true; + break; + default: + return NVME_INVALID_FIELD; + } + + ret = nvme_h2c(n, (uint8_t *)&key_info, sizeof(NvmeKeyInfo), req); + if (ret) { + return ret; + } + + switch (action) { + case NVME_RESV_REGISTER_ACTION_REGISTER: + req->aiocb = blk_aio_pr_register(ns->blkconf.blk, 0, + key_info.nr_key, 0, aptpl, + ignore_key, nvme_misc_cb, + req); + break; + case NVME_RESV_REGISTER_ACTION_UNREGISTER: + req->aiocb = blk_aio_pr_register(ns->blkconf.blk, key_info.cr_key, 0, + 0, aptpl, ignore_key, + nvme_misc_cb, req); + break; + case NVME_RESV_REGISTER_ACTION_REPLACE: + req->aiocb = blk_aio_pr_register(ns->blkconf.blk, key_info.cr_key, + key_info.nr_key, 0, aptpl, ignore_key, + nvme_misc_cb, req); + break; + default: + return NVME_INVALID_FIELD; + } + + return NVME_NO_COMPLETE; +} + +static uint16_t nvme_resv_release(NvmeCtrl *n, NvmeRequest *req) +{ + int ret; + uint64_t cr_key; + NvmeNamespace *ns = req->ns; + uint32_t cdw10 = le32_to_cpu(req->cmd.cdw10); + uint8_t action = cdw10 & 0x7; + NVMEResvType type = cdw10 >> 8 & 0xff; + + ret = nvme_h2c(n, (uint8_t *)&cr_key, sizeof(cr_key), req); + if (ret) { + return ret; + } + + switch (action) { + case NVME_RESV_RELEASE_ACTION_RELEASE: + req->aiocb = blk_aio_pr_release(ns->blkconf.blk, cr_key, + nvme_pr_type_to_block(type), + nvme_misc_cb, req); + break; + case NVME_RESV_RELEASE_ACTION_CLEAR: + req->aiocb = blk_aio_pr_clear(ns->blkconf.blk, cr_key, + nvme_misc_cb, req); + break; + default: + return NVME_INVALID_FIELD; + } + + return NVME_NO_COMPLETE; +} + +static uint16_t nvme_resv_acquire(NvmeCtrl *n, NvmeRequest *req) +{ + int ret; + NvmeKeyInfo key_info; + NvmeNamespace *ns = req->ns; + uint32_t cdw10 = le32_to_cpu(req->cmd.cdw10); + uint8_t action = cdw10 & 0x7; + NVMEResvType type = cdw10 >> 8 & 0xff; + + ret = nvme_h2c(n, (uint8_t *)&key_info, sizeof(NvmeKeyInfo), req); + if (ret) { + return ret; + } + + switch (action) { + case NVME_RESV_ACQUIRE_ACTION_ACQUIRE: + req->aiocb = blk_aio_pr_reserve(ns->blkconf.blk, key_info.cr_key, + nvme_pr_type_to_block(type), + nvme_misc_cb, req); + break; + case NVME_RESV_ACQUIRE_ACTION_PREEMPT: + req->aiocb = blk_aio_pr_preempt(ns->blkconf.blk, + key_info.cr_key, key_info.nr_key, + nvme_pr_type_to_block(type), + false, nvme_misc_cb, req); + break; + case NVME_RESV_ACQUIRE_ACTION_PREEMPT_AND_ABORT: + req->aiocb = blk_aio_pr_preempt(ns->blkconf.blk, key_info.cr_key, + key_info.nr_key, type, true, + nvme_misc_cb, req); + break; + default: + return NVME_INVALID_FIELD; + } + + return NVME_NO_COMPLETE; +} + +typedef struct NvmeResvKeys { + uint32_t generation; + uint32_t num_keys; + uint64_t *keys; + NvmeRequest *req; +} NvmeResvKeys; + +typedef struct NvmeReadReservation { + uint32_t generation; + uint64_t key; + BlockPrType type; + NvmeRequest *req; + NvmeResvKeys *keys_info; +} NvmeReadReservation; + +static int _nvme_resv_read_reservation_cb(NvmeReadReservation *reservation) +{ + int rc; + NvmeReservationStatus *nvme_status; + NvmeRequest *req = reservation->req; + NvmeCtrl *n = req->sq->ctrl; + NvmeResvKeys *keys_info = reservation->keys_info; + int len = sizeof(NvmeReservationStatusHeader) + + sizeof(NvmeRegisteredCtrl) * keys_info->num_keys; + + nvme_status = g_malloc(len); + nvme_status->header.gen = reservation->generation; + nvme_status->header.rtype = block_pr_type_to_nvme(reservation->type); + nvme_status->header.regctl = keys_info->num_keys; + for (int i = 0; i < keys_info->num_keys; i++) { + nvme_status->regctl_ds[i].cntlid = nvme_ctrl(req)->cntlid; + nvme_status->regctl_ds[i].rkey = keys_info->keys[i]; + nvme_status->regctl_ds[i].rcsts = keys_info->keys[i] == + reservation->key ? 1 : 0; + /* hostid is not supported currently */ + memset(&nvme_status->regctl_ds[i].hostid, 0, 8); + } + + rc = nvme_c2h(n, (uint8_t *)nvme_status, len, req); + g_free(nvme_status); + return rc; +} + +static int _nvme_resv_read_reservation_ext_cb(NvmeReadReservation *reservation) +{ + int rc; + NvmeReservationStatusExt *nvme_status_ext; + NvmeRequest *req = reservation->req; + NvmeCtrl *n = req->sq->ctrl; + NvmeResvKeys *keys_info = reservation->keys_info; + int len = sizeof(NvmeReservationStatusHeader) + + sizeof(uint8_t) * 40 + + sizeof(NvmeRegisteredCtrlExt) * keys_info->num_keys; + + nvme_status_ext = g_malloc(len); + nvme_status_ext->header.gen = cpu_to_be32(reservation->generation); + nvme_status_ext->header.rtype = block_pr_type_to_nvme(reservation->type); + nvme_status_ext->header.regctl = cpu_to_be16(keys_info->num_keys); + + for (int i = 0; i < keys_info->num_keys; i++) { + uint16_t ctnlid = nvme_ctrl(req)->cntlid; + nvme_status_ext->regctl_eds[i].cntlid = cpu_to_be16(ctnlid); + nvme_status_ext->regctl_eds[i].rkey = cpu_to_be64(keys_info->keys[i]); + nvme_status_ext->regctl_eds[i].rcsts = keys_info->keys[i] == + reservation->key ? 1 : 0; + /* hostid is not supported currently */ + memset(&nvme_status_ext->regctl_eds[i].hostid, 0, 16); + } + + rc = nvme_c2h(n, (uint8_t *)nvme_status_ext, len, req); + g_free(nvme_status_ext); + return rc; +} + +static void nvme_resv_read_reservation_cb(void *opaque, int ret) +{ + NvmeReadReservation *reservation = opaque; + NvmeRequest *req = reservation->req; + bool eds = req->cmd.cdw11 & 0x1; + NvmeResvKeys *keys_info = reservation->keys_info; + + if (ret < 0) { + goto out; + } + + if (eds) { + ret = _nvme_resv_read_reservation_ext_cb(reservation); + } else { + ret = _nvme_resv_read_reservation_cb(reservation); + } + +out: + g_free(keys_info); + g_free(reservation); + nvme_misc_cb(req, ret); +} + +static void nvme_resv_read_keys_cb(void *opaque, int ret) +{ + NvmeResvKeys *keys_info = opaque; + NvmeRequest *req = keys_info->req; + NvmeNamespace *ns = req->ns; + NvmeReadReservation *reservation; + + if (ret < 0) { + goto out; + } + + keys_info->num_keys = MIN(ret, keys_info->num_keys); + reservation = g_new0(NvmeReadReservation, 1); + memset(reservation, 0, sizeof(*reservation)); + reservation->req = req; + reservation->keys_info = keys_info; + + req->aiocb = blk_aio_pr_read_reservation(ns->blkconf.blk, + &reservation->generation, &reservation->key, + &reservation->type, nvme_resv_read_reservation_cb, + reservation); + return; + +out: + g_free(keys_info); + nvme_misc_cb(req, ret); +} + + +static uint16_t nvme_resv_report(NvmeCtrl *n, NvmeRequest *req) +{ + int num_keys; + uint32_t cdw10 = req->cmd.cdw10; + uint32_t cdw11 = req->cmd.cdw11; + int buflen = (cdw10 + 1) * sizeof(uint32_t); + bool eds = cdw11 & 0x1; + NvmeNamespace *ns = req->ns; + NvmeResvKeys *keys_info; + + if (eds) { + if (buflen < sizeof(NvmeReservationStatusHeader) + + sizeof(uint8_t) * 40) { + return NVME_INVALID_FIELD; + } + + num_keys = (buflen - sizeof(NvmeReservationStatusHeader) - + sizeof(uint8_t) * 40) / + sizeof(struct NvmeRegisteredCtrlExt); + } else { + if (buflen < sizeof(NvmeReservationStatusHeader)) { + return NVME_INVALID_FIELD; + } + + num_keys = (buflen - sizeof(NvmeReservationStatusHeader)) / + sizeof(struct NvmeRegisteredCtrl); + } + + keys_info = g_new0(NvmeResvKeys, 1); + keys_info->generation = 0; + /* num_keys is the maximum number of keys that can be transmitted */ + keys_info->num_keys = num_keys; + keys_info->keys = g_malloc(sizeof(uint64_t) * num_keys); + keys_info->req = req; + + req->aiocb = blk_aio_pr_read_keys(ns->blkconf.blk, &keys_info->generation, + keys_info->num_keys, keys_info->keys, + nvme_resv_read_keys_cb, keys_info); + + return NVME_NO_COMPLETE; +} + typedef struct NvmeCopyAIOCB { BlockAIOCB common; BlockAIOCB *aiocb; @@ -XXX,XX +XXX,XX @@ static uint16_t nvme_io_cmd(NvmeCtrl *n, NvmeRequest *req) return nvme_dsm(n, req); case NVME_CMD_VERIFY: return nvme_verify(n, req); + case NVME_CMD_RESV_REGISTER: + return nvme_resv_register(n, req); + case NVME_CMD_RESV_REPORT: + return nvme_resv_report(n, req); + case NVME_CMD_RESV_ACQUIRE: + return nvme_resv_acquire(n, req); + case NVME_CMD_RESV_RELEASE: + return nvme_resv_release(n, req); case NVME_CMD_COPY: return nvme_copy(n, req); case NVME_CMD_ZONE_MGMT_SEND: diff --git a/hw/nvme/nvme.h b/hw/nvme/nvme.h index XXXXXXX..XXXXXXX 100644 --- a/hw/nvme/nvme.h +++ b/hw/nvme/nvme.h @@ -XXX,XX +XXX,XX @@ static inline const char *nvme_io_opc_str(uint8_t opc) case NVME_CMD_ZONE_MGMT_SEND: return "NVME_ZONED_CMD_MGMT_SEND"; case NVME_CMD_ZONE_MGMT_RECV: return "NVME_ZONED_CMD_MGMT_RECV"; case NVME_CMD_ZONE_APPEND: return "NVME_ZONED_CMD_ZONE_APPEND"; + case NVME_CMD_RESV_REGISTER: return "NVME_CMD_RESV_REGISTER"; + case NVME_CMD_RESV_REPORT: return "NVME_CMD_RESV_REPORT"; + case NVME_CMD_RESV_ACQUIRE: return "NVME_CMD_RESV_ACQUIRE"; + case NVME_CMD_RESV_RELEASE: return "NVME_CMD_RESV_RELEASE"; default: return "NVME_NVM_CMD_UNKNOWN"; } } diff --git a/include/block/nvme.h b/include/block/nvme.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/nvme.h +++ b/include/block/nvme.h @@ -XXX,XX +XXX,XX @@ typedef enum NVMEPrCap { NVME_PR_CAP_EX_AC_AR), } NVMEPrCap; +typedef struct QEMU_PACKED NvmeRegisteredCtrl { + uint16_t cntlid; + uint8_t rcsts; + uint8_t rsvd3[5]; + uint8_t hostid[8]; + uint64_t rkey; +} NvmeRegisteredCtrl; + +typedef struct QEMU_PACKED NvmeRegisteredCtrlExt { + uint16_t cntlid; + uint8_t rcsts; + uint8_t rsvd3[5]; + uint64_t rkey; + uint8_t hostid[16]; + uint8_t rsvd32[32]; +} NvmeRegisteredCtrlExt; + +typedef struct QEMU_PACKED NvmeReservationStatusHeader { + uint32_t gen; + uint8_t rtype; + uint16_t regctl; + uint16_t resv5; + uint8_t ptpls; + uint8_t resv10[14]; +} NvmeReservationStatusHeader; + +typedef struct QEMU_PACKED NvmeReservationStatus { + struct NvmeReservationStatusHeader header; + struct NvmeRegisteredCtrl regctl_ds[]; +} NvmeReservationStatus; + +typedef struct QEMU_PACKED NvmeReservationStatusExt { + struct NvmeReservationStatusHeader header; + uint8_t rsvd24[40]; + struct NvmeRegisteredCtrlExt regctl_eds[]; +} NvmeReservationStatusExt; + + typedef struct QEMU_PACKED NvmeDeleteQ { uint8_t opcode; uint8_t flags; -- 2.20.1 Add persistent reservation in/out operations for iscsi driver. The following methods are implemented: bdrv_co_pr_read_keys, bdrv_co_pr_read_reservation, bdrv_co_pr_register, bdrv_co_pr_reserve, bdrv_co_pr_release, bdrv_co_pr_clear and bdrv_co_pr_preempt. Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com> Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> --- block/iscsi.c | 443 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 443 insertions(+) diff --git a/block/iscsi.c b/block/iscsi.c index XXXXXXX..XXXXXXX 100644 --- a/block/iscsi.c +++ b/block/iscsi.c @@ -XXX,XX +XXX,XX @@ typedef struct IscsiLun { unsigned long *allocmap_valid; long allocmap_size; int cluster_size; + uint8_t pr_cap; bool use_16_for_rw; bool write_protected; bool lbpme; @@ -XXX,XX +XXX,XX @@ iscsi_co_generic_cb(struct iscsi_context *iscsi, int status, iTask->err_code = -error; iTask->err_str = g_strdup(iscsi_get_error(iscsi)); } + } else if (status == SCSI_STATUS_RESERVATION_CONFLICT) { + iTask->err_code = -EBADE; } } } @@ -XXX,XX +XXX,XX @@ static void iscsi_save_designator(IscsiLun *lun, } } +static void iscsi_get_pr_cap_sync(IscsiLun *iscsilun, Error **errp) +{ + struct scsi_task *task = NULL; + struct scsi_persistent_reserve_in_report_capabilities *rc = NULL; + int retries = ISCSI_CMD_RETRIES; + int xferlen = sizeof(struct scsi_persistent_reserve_in_report_capabilities); + + do { + if (task != NULL) { + scsi_free_scsi_task(task); + task = NULL; + } + + task = iscsi_persistent_reserve_in_sync(iscsilun->iscsi, + iscsilun->lun, SCSI_PR_IN_REPORT_CAPABILITIES, xferlen); + if (task != NULL && task->status == SCSI_STATUS_GOOD) { + rc = scsi_datain_unmarshall(task); + if (rc == NULL) { + error_setg(errp, + "iSCSI: Failed to unmarshall report capabilities data."); + } else { + iscsilun->pr_cap = + scsi_pr_cap_to_block(rc->persistent_reservation_type_mask); + iscsilun->pr_cap |= (rc->ptpl_a) ? BLK_PR_CAP_PTPL : 0; + } + break; + } + + if (task != NULL && task->status == SCSI_STATUS_CHECK_CONDITION + && task->sense.key == SCSI_SENSE_UNIT_ATTENTION) { + break; + } + + } while (task != NULL && task->status == SCSI_STATUS_CHECK_CONDITION + && task->sense.key == SCSI_SENSE_UNIT_ATTENTION + && retries-- > 0); + + if (task == NULL || task->status != SCSI_STATUS_GOOD) { + error_setg(errp, "iSCSI: failed to send report capabilities command"); + } + + if (task) { + scsi_free_scsi_task(task); + } +} + static int iscsi_open(BlockDriverState *bs, QDict *options, int flags, Error **errp) { @@ -XXX,XX +XXX,XX @@ static int iscsi_open(BlockDriverState *bs, QDict *options, int flags, bs->supported_zero_flags = BDRV_REQ_MAY_UNMAP; } + iscsi_get_pr_cap_sync(iscsilun, &local_err); + if (local_err != NULL) { + error_propagate(errp, local_err); + ret = -EINVAL; + } out: qemu_opts_del(opts); g_free(initiator_name); @@ -XXX,XX +XXX,XX @@ static void iscsi_refresh_limits(BlockDriverState *bs, Error **errp) bs->bl.opt_transfer = pow2floor(iscsilun->bl.opt_xfer_len * iscsilun->block_size); } + + bs->bl.pr_cap = iscsilun->pr_cap; } /* Note that this will not re-establish a connection with an iSCSI target - it @@ -XXX,XX +XXX,XX @@ out_unlock: return r; } +static int coroutine_fn +iscsi_co_pr_read_keys(BlockDriverState *bs, uint32_t *generation, + uint32_t num_keys, uint64_t *keys) +{ + IscsiLun *iscsilun = bs->opaque; + QEMUIOVector qiov; + struct IscsiTask iTask; + int xferlen = sizeof(struct scsi_persistent_reserve_in_read_keys) + + sizeof(uint64_t) * num_keys; + uint8_t *buf = g_malloc0(xferlen); + int32_t num_collect_keys = 0; + int r = 0; + + qemu_iovec_init_buf(&qiov, buf, xferlen); + iscsi_co_init_iscsitask(iscsilun, &iTask); + qemu_mutex_lock(&iscsilun->mutex); +retry: + iTask.task = iscsi_persistent_reserve_in_task(iscsilun->iscsi, + iscsilun->lun, SCSI_PR_IN_READ_KEYS, xferlen, + iscsi_co_generic_cb, &iTask); + + if (iTask.task == NULL) { + qemu_mutex_unlock(&iscsilun->mutex); + return -ENOMEM; + } + + scsi_task_set_iov_in(iTask.task, (struct scsi_iovec *)qiov.iov, qiov.niov); + iscsi_co_wait_for_task(&iTask, iscsilun); + + if (iTask.task != NULL) { + scsi_free_scsi_task(iTask.task); + iTask.task = NULL; + } + + if (iTask.do_retry) { + iTask.complete = 0; + goto retry; + } + + if (iTask.status != SCSI_STATUS_GOOD) { + error_report("iSCSI PERSISTENT_RESERVE_IN failed: %s", iTask.err_str); + r = iTask.err_code; + goto out; + } + + memcpy(generation, &buf[0], 4); + *generation = be32_to_cpu(*generation); + memcpy(&num_collect_keys, &buf[4], 4); + num_collect_keys = be32_to_cpu(num_collect_keys) / sizeof(uint64_t); + if (num_collect_keys > num_keys) { + r = -EINVAL; + goto out; + } + + for (int i = 0; i < num_collect_keys; i++) { + memcpy(&keys[i], &buf[8 + i * 8], 8); + keys[i] = be64_to_cpu(keys[i]); + } + r = num_collect_keys; + +out: + qemu_mutex_unlock(&iscsilun->mutex); + g_free(iTask.err_str); + g_free(buf); + return r; +} + +static int coroutine_fn +iscsi_co_pr_read_reservation(BlockDriverState *bs, uint32_t *generation, + uint64_t *key, BlockPrType *type) +{ + IscsiLun *iscsilun = bs->opaque; + QEMUIOVector qiov; + struct IscsiTask iTask; + int xferlen = sizeof(struct scsi_persistent_reserve_in_read_reservation); + uint8_t *buf = g_malloc0(xferlen); + uint8_t scope_type = 0; + int32_t num_collect_keys = 0; + int r = 0; + + qemu_iovec_init_buf(&qiov, buf, xferlen); + iscsi_co_init_iscsitask(iscsilun, &iTask); + qemu_mutex_lock(&iscsilun->mutex); +retry: + iTask.task = iscsi_persistent_reserve_in_task(iscsilun->iscsi, + iscsilun->lun, SCSI_PR_IN_READ_RESERVATION, + xferlen, iscsi_co_generic_cb, &iTask); + + if (iTask.task == NULL) { + qemu_mutex_unlock(&iscsilun->mutex); + return -ENOMEM; + } + + scsi_task_set_iov_in(iTask.task, (struct scsi_iovec *)qiov.iov, qiov.niov); + iscsi_co_wait_for_task(&iTask, iscsilun); + + if (iTask.task != NULL) { + scsi_free_scsi_task(iTask.task); + iTask.task = NULL; + } + + if (iTask.do_retry) { + iTask.complete = 0; + goto retry; + } + + if (iTask.status != SCSI_STATUS_GOOD) { + error_report("iSCSI PERSISTENT_RESERVE_IN failed: %s", iTask.err_str); + r = iTask.err_code; + goto out; + } + + memcpy(generation, &buf[0], 4); + *generation = be32_to_cpu(*generation); + memcpy(key, &buf[8], 8); + *key = be64_to_cpu(*key); + memcpy(&scope_type, &buf[21], 1); + *type = scsi_pr_type_to_block(scope_type & 0xf); + memcpy(&num_collect_keys, &buf[4], 4); + r = be32_to_cpu(num_collect_keys) / sizeof(uint64_t); +out: + qemu_mutex_unlock(&iscsilun->mutex); + g_free(iTask.err_str); + g_free(buf); + return r; +} + +static int coroutine_fn +iscsi_co_pr_register(BlockDriverState *bs, uint64_t old_key, + uint64_t new_key, BlockPrType type, + bool ptpl, bool ignore_key) +{ + IscsiLun *iscsilun = bs->opaque; + struct IscsiTask iTask; + struct scsi_persistent_reserve_out_basic *basic; + SCSIPrOutAction action = ignore_key ? SCSI_PR_OUT_REG_AND_IGNORE_KEY : + SCSI_PR_OUT_REGISTER; + int r = 0; + + basic = g_new0(struct scsi_persistent_reserve_out_basic, 1); + basic->reservation_key = old_key; + basic->service_action_reservation_key = new_key; + basic->aptpl = ptpl ? 1 : 0; + + iscsi_co_init_iscsitask(iscsilun, &iTask); + qemu_mutex_lock(&iscsilun->mutex); +retry: + iTask.task = iscsi_persistent_reserve_out_task(iscsilun->iscsi, + iscsilun->lun, action, 0, block_pr_type_to_scsi(type), + basic, iscsi_co_generic_cb, &iTask); + + if (iTask.task == NULL) { + qemu_mutex_unlock(&iscsilun->mutex); + return -ENOMEM; + } + + iscsi_co_wait_for_task(&iTask, iscsilun); + + if (iTask.task != NULL) { + scsi_free_scsi_task(iTask.task); + iTask.task = NULL; + } + + if (iTask.do_retry) { + iTask.complete = 0; + goto retry; + } + + if (iTask.status != SCSI_STATUS_GOOD) { + error_report("iSCSI PERSISTENT_RESERVE_OUT failed: %s", iTask.err_str); + r = iTask.err_code; + } + + qemu_mutex_unlock(&iscsilun->mutex); + + g_free(iTask.err_str); + g_free(basic); + return r; +} + +static int coroutine_fn +iscsi_co_pr_reserve(BlockDriverState *bs, uint64_t key, BlockPrType type) +{ + IscsiLun *iscsilun = bs->opaque; + struct IscsiTask iTask; + struct scsi_persistent_reserve_out_basic *basic; + int r = 0; + + basic = g_new0(struct scsi_persistent_reserve_out_basic, 1); + basic->reservation_key = key; + + iscsi_co_init_iscsitask(iscsilun, &iTask); + qemu_mutex_lock(&iscsilun->mutex); +retry: + iTask.task = iscsi_persistent_reserve_out_task(iscsilun->iscsi, + iscsilun->lun, SCSI_PR_OUT_RESERVE, 0, + block_pr_type_to_scsi(type), basic, + iscsi_co_generic_cb, &iTask); + + if (iTask.task == NULL) { + qemu_mutex_unlock(&iscsilun->mutex); + return -ENOMEM; + } + + + iscsi_co_wait_for_task(&iTask, iscsilun); + + if (iTask.task != NULL) { + scsi_free_scsi_task(iTask.task); + iTask.task = NULL; + } + + if (iTask.do_retry) { + iTask.complete = 0; + goto retry; + } + + if (iTask.status != SCSI_STATUS_GOOD) { + error_report("iSCSI PERSISTENT_RESERVE_OUT failed: %s", iTask.err_str); + r = iTask.err_code; + } + + qemu_mutex_unlock(&iscsilun->mutex); + + g_free(iTask.err_str); + g_free(basic); + return r; +} + +static int coroutine_fn +iscsi_co_pr_release(BlockDriverState *bs, uint64_t key, BlockPrType type) +{ + IscsiLun *iscsilun = bs->opaque; + struct IscsiTask iTask; + struct scsi_persistent_reserve_out_basic *basic; + int r = 0; + + basic = g_new0(struct scsi_persistent_reserve_out_basic, 1); + basic->reservation_key = key; + + iscsi_co_init_iscsitask(iscsilun, &iTask); + qemu_mutex_lock(&iscsilun->mutex); +retry: + iTask.task = iscsi_persistent_reserve_out_task(iscsilun->iscsi, + iscsilun->lun, SCSI_PR_OUT_RELEASE, 0, + block_pr_type_to_scsi(type), basic, + iscsi_co_generic_cb, &iTask); + + if (iTask.task == NULL) { + qemu_mutex_unlock(&iscsilun->mutex); + return -ENOMEM; + } + + + iscsi_co_wait_for_task(&iTask, iscsilun); + + if (iTask.task != NULL) { + scsi_free_scsi_task(iTask.task); + iTask.task = NULL; + } + + if (iTask.do_retry) { + iTask.complete = 0; + goto retry; + } + + if (iTask.status != SCSI_STATUS_GOOD) { + error_report("iSCSI PERSISTENT_RESERVE_OUT failed: %s", iTask.err_str); + r = iTask.err_code; + } + + qemu_mutex_unlock(&iscsilun->mutex); + + g_free(iTask.err_str); + g_free(basic); + return r; +} + +static int coroutine_fn +iscsi_co_pr_clear(BlockDriverState *bs, uint64_t key) +{ + IscsiLun *iscsilun = bs->opaque; + struct IscsiTask iTask; + struct scsi_persistent_reserve_out_basic *basic; + int r = 0; + + basic = g_new0(struct scsi_persistent_reserve_out_basic, 1); + basic->reservation_key = key; + + iscsi_co_init_iscsitask(iscsilun, &iTask); + qemu_mutex_lock(&iscsilun->mutex); +retry: + iTask.task = iscsi_persistent_reserve_out_task(iscsilun->iscsi, + iscsilun->lun, SCSI_PR_OUT_CLEAR, 0, 0, basic, + iscsi_co_generic_cb, &iTask); + + if (iTask.task == NULL) { + qemu_mutex_unlock(&iscsilun->mutex); + return -ENOMEM; + } + + + iscsi_co_wait_for_task(&iTask, iscsilun); + + if (iTask.task != NULL) { + scsi_free_scsi_task(iTask.task); + iTask.task = NULL; + } + + if (iTask.do_retry) { + iTask.complete = 0; + goto retry; + } + + if (iTask.status != SCSI_STATUS_GOOD) { + error_report("iSCSI PERSISTENT_RESERVE_OUT failed: %s", iTask.err_str); + r = iTask.err_code; + } + + qemu_mutex_unlock(&iscsilun->mutex); + + g_free(iTask.err_str); + g_free(basic); + return r; +} + +static int coroutine_fn +iscsi_co_pr_preempt(BlockDriverState *bs, uint64_t old_key, + uint64_t new_key, BlockPrType type, bool abort) +{ + IscsiLun *iscsilun = bs->opaque; + struct IscsiTask iTask; + struct scsi_persistent_reserve_out_basic *basic; + SCSIPrOutAction action = abort ? SCSI_PR_OUT_PREEMPT_AND_ABORT : + SCSI_PR_OUT_PREEMPT; + int r = 0; + + basic = g_new0(struct scsi_persistent_reserve_out_basic, 1); + basic->reservation_key = old_key; + basic->service_action_reservation_key = new_key; + + iscsi_co_init_iscsitask(iscsilun, &iTask); + qemu_mutex_lock(&iscsilun->mutex); +retry: + iTask.task = iscsi_persistent_reserve_out_task(iscsilun->iscsi, + iscsilun->lun, action, 0, block_pr_type_to_scsi(type), + basic, iscsi_co_generic_cb, &iTask); + + if (iTask.task == NULL) { + qemu_mutex_unlock(&iscsilun->mutex); + return -ENOMEM; + } + + + iscsi_co_wait_for_task(&iTask, iscsilun); + + if (iTask.task != NULL) { + scsi_free_scsi_task(iTask.task); + iTask.task = NULL; + } + + if (iTask.do_retry) { + iTask.complete = 0; + goto retry; + } + + if (iTask.status != SCSI_STATUS_GOOD) { + error_report("iSCSI PERSISTENT_RESERVE_OUT failed: %s", iTask.err_str); + r = iTask.err_code; + } + + qemu_mutex_unlock(&iscsilun->mutex); + + + g_free(iTask.err_str); + g_free(basic); + return r; +} + static const char *const iscsi_strong_runtime_opts[] = { "transport", @@ -XXX,XX +XXX,XX @@ static BlockDriver bdrv_iscsi = { .bdrv_co_writev = iscsi_co_writev, .bdrv_co_flush_to_disk = iscsi_co_flush, + .bdrv_co_pr_read_keys = iscsi_co_pr_read_keys, + .bdrv_co_pr_read_reservation = iscsi_co_pr_read_reservation, + .bdrv_co_pr_register = iscsi_co_pr_register, + .bdrv_co_pr_reserve = iscsi_co_pr_reserve, + .bdrv_co_pr_release = iscsi_co_pr_release, + .bdrv_co_pr_clear = iscsi_co_pr_clear, + .bdrv_co_pr_preempt = iscsi_co_pr_preempt, + #ifdef __linux__ .bdrv_aio_ioctl = iscsi_aio_ioctl, #endif -- 2.20.1 Stefan, the issue you mentioned has been fixed. Almost all patches have been reviewed, thank you very much to Stefan and Klaus. v7->v8: - Fix num_keys may be less than 0 at scsi_pr_read_keys_complete(). - Fix buf memory leak at iscsi driver. v6->v7: - Add buferlen size check at SCSI layer. - Add pr_cap calculation in bdrv_merge_limits() function at block layer, so the ugly bs->file->bs->bl.pr_cap in scsi and nvme layers was changed to bs->bl.pr_cap. - Fix memory leak at iscsi driver, and some other spelling errors. v5->v6: - Add relevant comments in the io layer. v4->v5: - Fixed a memory leak bug at hw/nvme/ctrl.c. v3->v4: - At the nvme layer, the two patches of enabling the ONCS function and enabling rescap are combined into one. - At the nvme layer, add helper functions for pr capacity conversion between the block layer and the nvme layer. v2->v3: In v2 Persist Through Power Loss(PTPL) is enable default. In v3 PTPL is supported, which is passed as a parameter. v1->v2: - Add sg_persist --report-capabilities for SCSI protocol and enable oncs and rescap for NVMe protocol. - Add persistent reservation capabilities constants and helper functions for SCSI and NVMe protocol. - Add comments for necessary APIs. v1: - Add seven APIs about persistent reservation command for block layer. These APIs including reading keys, reading reservations, registering, reserving, releasing, clearing and preempting. - Add the necessary pr-related operation APIs for both the SCSI protocol and NVMe protocol at the device layer. - Add scsi driver at the driver layer to verify the functions Changqi Lu (10): block: add persistent reservation in/out api block/raw: add persistent reservation in/out driver scsi/constant: add persistent reservation in/out protocol constants scsi/util: add helper functions for persistent reservation types conversion hw/scsi: add persistent reservation in/out api for scsi device block/nvme: add reservation command protocol constants hw/nvme: add helper functions for converting reservation types hw/nvme: enable ONCS and rescap function hw/nvme: add reservation protocal command block/iscsi: add persistent reservation in/out driver block/block-backend.c | 403 ++++++++++++++++++++++++++++ block/io.c | 164 ++++++++++++ block/iscsi.c | 425 ++++++++++++++++++++++++++++++ block/raw-format.c | 56 ++++ hw/nvme/ctrl.c | 326 ++++++++++++++++++++++- hw/nvme/ns.c | 5 + hw/nvme/nvme.h | 88 +++++++ hw/scsi/scsi-disk.c | 368 ++++++++++++++++++++++++++ include/block/block-common.h | 40 +++ include/block/block-io.h | 20 ++ include/block/block_int-common.h | 84 ++++++ include/block/nvme.h | 100 ++++++- include/scsi/constants.h | 52 ++++ include/scsi/utils.h | 8 + include/sysemu/block-backend-io.h | 24 ++ scsi/utils.c | 81 ++++++ 16 files changed, 2241 insertions(+), 3 deletions(-) -- 2.20.1 Add persistent reservation in/out operations at the block level. The following operations are included: - read_keys: retrieves the list of registered keys. - read_reservation: retrieves the current reservation status. - register: registers a new reservation key. - reserve: initiates a reservation for a specific key. - release: releases a reservation for a specific key. - clear: clears all existing reservations. - preempt: preempts a reservation held by another key. Signed-off-by: Changqi Lu <luchangqi.123@bytedance.com> Signed-off-by: zhenwei pi <pizhenwei@bytedance.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> --- block/block-backend.c | 403 ++++++++++++++++++++++++++++++ block/io.c | 164 ++++++++++++ include/block/block-common.h | 40 +++ include/block/block-io.h | 20 ++ include/block/block_int-common.h | 84 +++++++ include/sysemu/block-backend-io.h | 24 ++ 6 files changed, 735 insertions(+) diff --git a/block/block-backend.c b/block/block-backend.c index XXXXXXX..XXXXXXX 100644 --- a/block/block-backend.c +++ b/block/block-backend.c @@ -XXX,XX +XXX,XX @@ BlockAIOCB *blk_aio_ioctl(BlockBackend *blk, unsigned long int req, void *buf, return blk_aio_prwv(blk, req, 0, buf, blk_aio_ioctl_entry, 0, cb, opaque); } +typedef struct BlkPrInCo { + BlockBackend *blk; + uint32_t *generation; + uint32_t num_keys; + BlockPrType *type; + uint64_t *keys; + int ret; +} BlkPrInCo; + +typedef struct BlkPrInCB { + BlockAIOCB common; + BlkPrInCo prco; + bool has_returned; +} BlkPrInCB; + +static const AIOCBInfo blk_pr_in_aiocb_info = { + .aiocb_size = sizeof(BlkPrInCB), +}; + +static void blk_pr_in_complete(BlkPrInCB *acb) +{ + if (acb->has_returned) { + acb->common.cb(acb->common.opaque, acb->prco.ret); + + /* This is paired with blk_inc_in_flight() in blk_aio_pr_in(). */ + blk_dec_in_flight(acb->prco.blk); + qemu_aio_unref(acb); + } +} + +static void blk_pr_in_complete_bh(void *opaque) +{ + BlkPrInCB *acb = opaque; + assert(acb->has_returned); + blk_pr_in_complete(acb); +} + +static BlockAIOCB *blk_aio_pr_in(BlockBackend *blk, uint32_t *generation, + uint32_t num_keys, BlockPrType *type, + uint64_t *keys, CoroutineEntry co_entry, + BlockCompletionFunc *cb, void *opaque) +{ + BlkPrInCB *acb; + Coroutine *co; + + /* This is paired with blk_dec_in_flight() in blk_pr_in_complete(). */ + blk_inc_in_flight(blk); + acb = blk_aio_get(&blk_pr_in_aiocb_info, blk, cb, opaque); + acb->prco = (BlkPrInCo) { + .blk = blk, + .generation = generation, + .num_keys = num_keys, + .type = type, + .ret = NOT_DONE, + .keys = keys, + }; + acb->has_returned = false; + + co = qemu_coroutine_create(co_entry, acb); + aio_co_enter(qemu_get_current_aio_context(), co); + + acb->has_returned = true; + if (acb->prco.ret != NOT_DONE) { + replay_bh_schedule_oneshot_event(qemu_get_current_aio_context(), + blk_pr_in_complete_bh, acb); + } + + return &acb->common; +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_read_keys(BlockBackend *blk, uint32_t *generation, + uint32_t num_keys, uint64_t *keys) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_read_keys(blk_bs(blk), generation, num_keys, keys); +} + +static void coroutine_fn blk_aio_pr_read_keys_entry(void *opaque) +{ + BlkPrInCB *acb = opaque; + BlkPrInCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_read_keys(prco->blk, prco->generation, + prco->num_keys, prco->keys); + blk_pr_in_complete(acb); +} + +BlockAIOCB *blk_aio_pr_read_keys(BlockBackend *blk, uint32_t *generation, + uint32_t num_keys, uint64_t *keys, + BlockCompletionFunc *cb, void *opaque) +{ + IO_CODE(); + return blk_aio_pr_in(blk, generation, num_keys, NULL, keys, + blk_aio_pr_read_keys_entry, cb, opaque); +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_read_reservation(BlockBackend *blk, uint32_t *generation, + uint64_t *key, BlockPrType *type) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_read_reservation(blk_bs(blk), generation, key, type); +} + +static void coroutine_fn blk_aio_pr_read_reservation_entry(void *opaque) +{ + BlkPrInCB *acb = opaque; + BlkPrInCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_read_reservation(prco->blk, prco->generation, + prco->keys, prco->type); + blk_pr_in_complete(acb); +} + +BlockAIOCB *blk_aio_pr_read_reservation(BlockBackend *blk, uint32_t *generation, + uint64_t *key, BlockPrType *type, + BlockCompletionFunc *cb, void *opaque) +{ + IO_CODE(); + return blk_aio_pr_in(blk, generation, 0, type, key, + blk_aio_pr_read_reservation_entry, cb, opaque); +} + +typedef struct BlkPrOutCo { + BlockBackend *blk; + uint64_t old_key; + uint64_t new_key; + bool ptpl; + BlockPrType type; + bool ignore_key; + bool abort; + int ret; +} BlkPrOutCo; + +typedef struct BlkPrOutCB { + BlockAIOCB common; + BlkPrOutCo prco; + bool has_returned; +} BlkPrOutCB; + +static const AIOCBInfo blk_pr_out_aiocb_info = { + .aiocb_size = sizeof(BlkPrOutCB), +}; + +static void blk_pr_out_complete(BlkPrOutCB *acb) +{ + if (acb->has_returned) { + acb->common.cb(acb->common.opaque, acb->prco.ret); + + /* This is paired with blk_inc_in_flight() in blk_aio_pr_out(). */ + blk_dec_in_flight(acb->prco.blk); + qemu_aio_unref(acb); + } +} + +static void blk_pr_out_complete_bh(void *opaque) +{ + BlkPrOutCB *acb = opaque; + assert(acb->has_returned); + blk_pr_out_complete(acb); +} + +static BlockAIOCB *blk_aio_pr_out(BlockBackend *blk, uint64_t old_key, + uint64_t new_key, bool ptpl, + BlockPrType type, bool ignore_key, + bool abort, CoroutineEntry co_entry, + BlockCompletionFunc *cb, void *opaque) +{ + BlkPrOutCB *acb; + Coroutine *co; + + /* This is paired with blk_dec_in_flight() in blk_pr_out_complete(). */ + blk_inc_in_flight(blk); + acb = blk_aio_get(&blk_pr_out_aiocb_info, blk, cb, opaque); + acb->prco = (BlkPrOutCo) { + .blk = blk, + .old_key = old_key, + .new_key = new_key, + .ptpl = ptpl, + .type = type, + .ignore_key = ignore_key, + .abort = abort, + .ret = NOT_DONE, + }; + acb->has_returned = false; + + co = qemu_coroutine_create(co_entry, acb); + aio_co_enter(qemu_get_current_aio_context(), co); + + acb->has_returned = true; + if (acb->prco.ret != NOT_DONE) { + replay_bh_schedule_oneshot_event(qemu_get_current_aio_context(), + blk_pr_out_complete_bh, acb); + } + + return &acb->common; +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_register(BlockBackend *blk, uint64_t old_key, + uint64_t new_key, BlockPrType type, + bool ptpl, bool ignore_key) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_register(blk_bs(blk), old_key, new_key, type, + ptpl, ignore_key); +} + +static void coroutine_fn blk_aio_pr_register_entry(void *opaque) +{ + BlkPrOutCB *acb = opaque; + BlkPrOutCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_register(prco->blk, prco->old_key, prco->new_key, + prco->type, prco->ptpl, + prco->ignore_key); + blk_pr_out_complete(acb); +} + +BlockAIOCB *blk_aio_pr_register(BlockBackend *blk, uint64_t old_key, + uint64_t new_key, BlockPrType type, + bool ptpl, bool ignore_key, + BlockCompletionFunc *cb, + void *opaque) +{ + IO_CODE(); + return blk_aio_pr_out(blk, old_key, new_key, ptpl, type, ignore_key, false, + blk_aio_pr_register_entry, cb, opaque); +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_reserve(BlockBackend *blk, uint64_t key, BlockPrType type) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_reserve(blk_bs(blk), key, type); +} + +static void coroutine_fn blk_aio_pr_reserve_entry(void *opaque) +{ + BlkPrOutCB *acb = opaque; + BlkPrOutCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_reserve(prco->blk, prco->old_key, + prco->type); + blk_pr_out_complete(acb); +} + + +BlockAIOCB *blk_aio_pr_reserve(BlockBackend *blk, uint64_t key, + BlockPrType type, + BlockCompletionFunc *cb, + void *opaque) +{ + IO_CODE(); + return blk_aio_pr_out(blk, key, 0, false, type, false, false, + blk_aio_pr_reserve_entry, cb, opaque); +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_release(BlockBackend *blk, uint64_t key, BlockPrType type) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_release(blk_bs(blk), key, type); +} + +static void coroutine_fn blk_aio_pr_release_entry(void *opaque) +{ + BlkPrOutCB *acb = opaque; + BlkPrOutCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_release(prco->blk, prco->old_key, prco->type); + blk_pr_out_complete(acb); +} + + +BlockAIOCB *blk_aio_pr_release(BlockBackend *blk, uint64_t key, + BlockPrType type, BlockCompletionFunc *cb, + void *opaque) +{ + IO_CODE(); + return blk_aio_pr_out(blk, key, 0, false, type, false, false, + blk_aio_pr_release_entry, cb, opaque); +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_clear(BlockBackend *blk, uint64_t key) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_clear(blk_bs(blk), key); +} + +static void coroutine_fn blk_aio_pr_clear_entry(void *opaque) +{ + BlkPrOutCB *acb = opaque; + BlkPrOutCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_clear(prco->blk, prco->old_key); + blk_pr_out_complete(acb); +} + + +BlockAIOCB *blk_aio_pr_clear(BlockBackend *blk, uint64_t key, + BlockCompletionFunc *cb, void *opaque) +{ + IO_CODE(); + return blk_aio_pr_out(blk, key, 0, false, 0, false, false, + blk_aio_pr_clear_entry, cb, opaque); +} + +/* To be called between exactly one pair of blk_inc/dec_in_flight() */ +static int coroutine_fn +blk_aio_pr_do_preempt(BlockBackend *blk, uint64_t cr_key, + uint64_t pr_key, BlockPrType type, bool abort) +{ + IO_CODE(); + + blk_wait_while_drained(blk); + GRAPH_RDLOCK_GUARD(); + + if (!blk_co_is_available(blk)) { + return -ENOMEDIUM; + } + + return bdrv_co_pr_preempt(blk_bs(blk), cr_key, pr_key, type, abort); +} + +static void coroutine_fn blk_aio_pr_preempt_entry(void *opaque) +{ + BlkPrOutCB *acb = opaque; + BlkPrOutCo *prco = &acb->prco; + + prco->ret = blk_aio_pr_do_preempt(prco->blk, prco->old_key, + prco->new_key, prco->type, + prco->abort); + blk_pr_out_complete(acb); +} + + +BlockAIOCB *blk_aio_pr_preempt(BlockBackend *blk, uint64_t cr_key, + uint64_t pr_key, BlockPrType type, + bool abort, BlockCompletionFunc *cb, + void *opaque) +{ + IO_CODE(); + return blk_aio_pr_out(blk, cr_key, pr_key, false, type, false, abort, + blk_aio_pr_preempt_entry, cb, opaque); +} + /* To be called between exactly one pair of blk_inc/dec_in_flight() */ static int coroutine_fn blk_co_do_pdiscard(BlockBackend *blk, int64_t offset, int64_t bytes) diff --git a/block/io.c b/block/io.c index XXXXXXX..XXXXXXX 100644 --- a/block/io.c +++ b/block/io.c @@ -XXX,XX +XXX,XX @@ static void bdrv_merge_limits(BlockLimits *dst, const BlockLimits *src) src->min_mem_alignment); dst->max_iov = MIN_NON_ZERO(dst->max_iov, src->max_iov); dst->max_hw_iov = MIN_NON_ZERO(dst->max_hw_iov, src->max_hw_iov); + dst->pr_cap |= src->pr_cap; } typedef struct BdrvRefreshLimitsState { @@ -XXX,XX +XXX,XX @@ out: return co.ret; } +int coroutine_fn bdrv_co_pr_read_keys(BlockDriverState *bs, + uint32_t *generation, uint32_t num_keys, + uint64_t *keys) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_read_keys) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_read_keys(bs, generation, num_keys, keys); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + +int coroutine_fn bdrv_co_pr_read_reservation(BlockDriverState *bs, + uint32_t *generation, uint64_t *key, BlockPrType *type) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_read_reservation) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_read_reservation(bs, generation, key, type); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + +int coroutine_fn bdrv_co_pr_register(BlockDriverState *bs, uint64_t old_key, + uint64_t new_key, BlockPrType type, bool ptpl, + bool ignore_key) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_register) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_register(bs, old_key, new_key, type, + ptpl, ignore_key); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + +int coroutine_fn bdrv_co_pr_reserve(BlockDriverState *bs, uint64_t key, + BlockPrType type) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_reserve) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_reserve(bs, key, type); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + +int coroutine_fn bdrv_co_pr_release(BlockDriverState *bs, uint64_t key, + BlockPrType type) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_release) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_release(bs, key, type); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + +int coroutine_fn bdrv_co_pr_clear(BlockDriverState *bs, uint64_t key) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_clear) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_clear(bs, key); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + +int coroutine_fn bdrv_co_pr_preempt(BlockDriverState *bs, uint64_t cr_key, + uint64_t pr_key, BlockPrType type, bool abort) +{ + BlockDriver *drv = bs->drv; + CoroutineIOCompletion co = { + .coroutine = qemu_coroutine_self(), + }; + + IO_CODE(); + assert_bdrv_graph_readable(); + + bdrv_inc_in_flight(bs); + if (!drv || !drv->bdrv_co_pr_preempt) { + co.ret = -ENOTSUP; + goto out; + } + + co.ret = drv->bdrv_co_pr_preempt(bs, cr_key, pr_key, type, abort); +out: + bdrv_dec_in_flight(bs); + return co.ret; +} + int coroutine_fn bdrv_co_zone_report(BlockDriverState *bs, int64_t offset, unsigned int *nr_zones, BlockZoneDescriptor *zones) diff --git a/include/block/block-common.h b/include/block/block-common.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/block-common.h +++ b/include/block/block-common.h @@ -XXX,XX +XXX,XX @@ typedef enum { BDRV_FIX_ERRORS = 2, } BdrvCheckMode; +/** + * According SCSI protocol(chapter 5.9 of SCSI Primary Commands - 4) + * and NVMe protocol(chapter 7.2 of NVMe Base Specification 2.0), + * the persistent reservation types and persistent capabilities of + * the public layer block are abstracted. + */ +typedef enum { + BLK_PR_WRITE_EXCLUSIVE = 0x1, + BLK_PR_EXCLUSIVE_ACCESS = 0x2, + BLK_PR_WRITE_EXCLUSIVE_REGS_ONLY = 0x3, + BLK_PR_EXCLUSIVE_ACCESS_REGS_ONLY = 0x4, + BLK_PR_WRITE_EXCLUSIVE_ALL_REGS = 0x5, + BLK_PR_EXCLUSIVE_ACCESS_ALL_REGS = 0x6, +} BlockPrType; + +typedef enum BLKPrCap { + /* Persist Through Power Loss */ + BLK_PR_CAP_PTPL = 1 << 0, + /* Write Exclusive reservation type */ + BLK_PR_CAP_WR_EX = 1 << 1, + /* Exclusive Access reservation type */ + BLK_PR_CAP_EX_AC = 1 << 2, + /* Write Exclusive Registrants Only reservation type */ + BLK_PR_CAP_WR_EX_RO = 1 << 3, + /* Exclusive Access Registrants Only reservation type */ + BLK_PR_CAP_EX_AC_RO = 1 << 4, + /* Write Exclusive All Registrants reservation type */ + BLK_PR_CAP_WR_EX_AR = 1 << 5, + /* Exclusive Access All Registrants reservation type */ + BLK_PR_CAP_EX_AC_AR = 1 << 6, + + BLK_PR_CAP_ALL = (BLK_PR_CAP_PTPL | + BLK_PR_CAP_WR_EX | + BLK_PR_CAP_EX_AC | + BLK_PR_CAP_WR_EX_RO | + BLK_PR_CAP_EX_AC_RO | + BLK_PR_CAP_WR_EX_AR | + BLK_PR_CAP_EX_AC_AR), +} BLKPrCap; + typedef struct BlockSizes { uint32_t phys; uint32_t log; diff --git a/include/block/block-io.h b/include/block/block-io.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/block-io.h +++ b/include/block/block-io.h @@ -XXX,XX +XXX,XX @@ void bdrv_aio_cancel_async(BlockAIOCB *acb); int coroutine_fn GRAPH_RDLOCK bdrv_co_ioctl(BlockDriverState *bs, int req, void *buf); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_read_keys(BlockDriverState *bs, uint32_t *generation, + uint32_t num_keys, uint64_t *keys); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_read_reservation(BlockDriverState *bs, uint32_t *generation, + uint64_t *key, BlockPrType *type); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_register(BlockDriverState *bs, uint64_t old_key, + uint64_t new_key, BlockPrType type, + bool ptpl, bool ignore_key); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_reserve(BlockDriverState *bs, uint64_t key, BlockPrType type); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_release(BlockDriverState *bs, uint64_t key, BlockPrType type); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_clear(BlockDriverState *bs, uint64_t key); +int coroutine_fn GRAPH_RDLOCK +bdrv_co_pr_preempt(BlockDriverState *bs, uint64_t cr_key, uint64_t pr_key, + BlockPrType type, bool abort); + /* Ensure contents are flushed to disk. */ int coroutine_fn GRAPH_RDLOCK bdrv_co_flush(BlockDriverState *bs); diff --git a/include/block/block_int-common.h b/include/block/block_int-common.h index XXXXXXX..XXXXXXX 100644 --- a/include/block/block_int-common.h +++ b/include/block/block_int-common.h @@ -XXX,XX +XXX,XX @@ struct BlockDriver { int coroutine_fn GRAPH_RDLOCK_PTR (*bdrv_co_ioctl)( BlockDriverState *bs, unsigned long int req, void *buf); + /* + * Persistent reservation series api. + * Please refer to chapter 5.9 of SCSI Primary Commands - 4 or + * chapter 7 of NVMe Base Specification 2.0. + * + * The block layer driver should implement all the following APIs + * or none at all, including: bdrv_co_pr_read_keys, + * bdrv_co_pr_read_reservation, bdrv_co_pr_register, + * bdrv_co_pr_reserve, bdrv_co_pr_release, + * bdrv_co_pr_clear and bdrv_co_pr_preempt. + * + * Read the registered keys and return them in the @keys. + * @generation: The generation of the reservation key. + * @num_keys: The maximum number of keys that can be transmitted. + * @keys: Registered keys array. + * + * On success, store generation in @generation and store keys @keys + * and return the number of @keys. + * On failure return -errno. + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_read_keys)( + BlockDriverState *bs, uint32_t *generation, + uint32_t num_keys, uint64_t *keys); + /* + * Read the reservation key and store it in the @key. + * @generation: The generation of the reservation key. + * @key: The reservation key. + * @type: Type of the reservation key. + * + * On success, store generation in @generation, store the + * reservation key in @key and return the number of @key + * which used to determine whether the reservation key exists. + * On failure return -errno. + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_read_reservation)( + BlockDriverState *bs, uint32_t *generation, + uint64_t *key, BlockPrType *type); + /* + * Register, unregister, or replace a reservation key. + * @old_key: The current reservation key associated with the host. + * @new_key: The new reservation Key. + * @type: Type of the reservation key. + * @ignore_key: Ignore or not @old_key. + * @ptpl: Whether to support Persist Through Power Loss(PTPL). + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_register)( + BlockDriverState *bs, uint64_t old_key, + uint64_t new_key, BlockPrType type, + bool ptpl, bool ignore_key); + /* + * Acquire a reservation on a host. + * @key: The current reservation key associated with the host. + * @type: Type of the reservation key. + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_reserve)( + BlockDriverState *bs, uint64_t key, BlockPrType type); + /* + * Release a reservation on a host. + * @key: The current reservation key associated with the host. + * @type: Type of the reservation key. + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_release)( + BlockDriverState *bs, uint64_t key, BlockPrType type); + /** + * Clear reservations on a host. + * @key: The current reservation key associated with the host. + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_clear)( + BlockDriverState *bs, uint64_t key); + /* + * Preempt a reservation held on a host. + * @cr_key: The current reservation key associated with the host. + * @pr_key: The preempt reservation Key which to be + * unregistered from the namespace. + * @type: Type of the reservation key. + * @abort: Whether to abort a reservation held on a host. + */ + int coroutine_fn GRAPH_RDLOCK_PTR(*bdrv_co_pr_preempt)( + BlockDriverState *bs, uint64_t cr_key, + uint64_t pr_key, BlockPrType type, bool abort); + /* * Returns 0 for completed check, -errno for internal errors. * The check results are stored in result. @@ -XXX,XX +XXX,XX @@ typedef struct BlockLimits { uint32_t max_active_zones; uint32_t write_granularity; + + /* Persistent reservation capacities. */ + uint8_t pr_cap; } BlockLimits; typedef struct BdrvOpBlocker BdrvOpBlocker; diff --git a/include/sysemu/block-backend-io.h b/include/sysemu/block-backend-io.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/block-backend-io.h +++ b/include/sysemu/block-backend-io.h @@ -XXX,XX +XXX,XX @@ void blk_aio_cancel_async(BlockAIOCB *acb); BlockAIOCB *blk_aio_ioctl(BlockBackend *blk, unsigned long int req, void *buf, BlockCompletionFunc *cb, void *opaque); +BlockAIOCB *blk_aio_pr_read_keys(BlockBackend *blk, uint32_t *generation, + uint32_t num_keys, uint64_t *keys, + BlockCompletionFunc *cb, void *opaque); +BlockAIOCB *blk_aio_pr_read_reservation(BlockBackend *blk, uint32_t *generation, + uint64_t *key, BlockPrType *type, + BlockCompletionFunc *cb, void *opaque); +BlockAIOCB *blk_aio_pr_register(BlockBackend *blk, uint64_t old_key, + uint64_t new_key, BlockPrType type, + bool ptpl, bool ignore_key, + BlockCompletionFunc *cb, + void *opaque); +BlockAIOCB *blk_aio_pr_reserve(BlockBackend *blk, uint64_t key, + BlockPrType type, + BlockCompletionFunc *cb, + void *opaque); +BlockAIOCB *blk_aio_pr_release(BlockBackend *blk, uint64_t key, + BlockPrType type, BlockCompletionFunc *cb, + void *opaque); +BlockAIOCB *blk_aio_pr_clear(BlockBackend *blk, uint64_t key, + BlockCompletionFunc *cb, void *opaque); +BlockAIOCB *blk_aio_pr_preempt(BlockBackend *blk, uint64_t cr_key, + uint64_t pr_key, BlockPrType type, bool abort, + BlockCompletionFunc *cb, void *opaque); + void blk_inc_in_flight(BlockBackend *blk); void blk_dec_in_flight(BlockBackend *blk); -- 2.20.1