diff options
-rw-r--r-- | README | 2 | ||||
-rw-r--r-- | block/block-backend.c | 51 | ||||
-rw-r--r-- | cpus.c | 16 | ||||
-rw-r--r-- | exec.c | 1 | ||||
-rw-r--r-- | fpu/softfloat.c | 4 | ||||
-rw-r--r-- | hw/block/dataplane/virtio-blk.c | 39 | ||||
-rw-r--r-- | hw/scsi/virtio-scsi-dataplane.c | 9 | ||||
-rw-r--r-- | include/block/aio-wait.h | 13 | ||||
-rw-r--r-- | include/sysemu/iothread.h | 1 | ||||
-rw-r--r-- | include/sysemu/sysemu.h | 1 | ||||
-rw-r--r-- | iothread.c | 31 | ||||
-rw-r--r-- | target/m68k/fpu_helper.c | 35 | ||||
-rw-r--r-- | target/m68k/helper.h | 7 | ||||
-rw-r--r-- | target/m68k/softfloat.c | 1019 | ||||
-rw-r--r-- | target/m68k/softfloat.h | 8 | ||||
-rw-r--r-- | target/m68k/softfloat_fpsp_tables.h | 374 | ||||
-rw-r--r-- | target/m68k/translate.c | 21 | ||||
-rw-r--r-- | util/aio-wait.c | 31 | ||||
-rw-r--r-- | vl.c | 15 |
19 files changed, 1591 insertions, 87 deletions
@@ -73,7 +73,7 @@ The QEMU website is also maintained under source control. git clone git://git.qemu.org/qemu-web.git https://www.qemu.org/2017/02/04/the-new-qemu-website-is-up/ -A 'git-profile' utility was created to make above process less +A 'git-publish' utility was created to make above process less cumbersome, and is highly recommended for making regular contributions, or even just for sending consecutive patch series revisions. It also requires a working 'git send-email' setup, and by default doesn't diff --git a/block/block-backend.c b/block/block-backend.c index b3c790e2bd..f2e0a855ff 100644 --- a/block/block-backend.c +++ b/block/block-backend.c @@ -1150,7 +1150,7 @@ int coroutine_fn blk_co_pwritev(BlockBackend *blk, int64_t offset, typedef struct BlkRwCo { BlockBackend *blk; int64_t offset; - QEMUIOVector *qiov; + void *iobuf; int ret; BdrvRequestFlags flags; } BlkRwCo; @@ -1158,17 +1158,19 @@ typedef struct BlkRwCo { static void blk_read_entry(void *opaque) { BlkRwCo *rwco = opaque; + QEMUIOVector *qiov = rwco->iobuf; - rwco->ret = blk_co_preadv(rwco->blk, rwco->offset, rwco->qiov->size, - rwco->qiov, rwco->flags); + rwco->ret = blk_co_preadv(rwco->blk, rwco->offset, qiov->size, + qiov, rwco->flags); } static void blk_write_entry(void *opaque) { BlkRwCo *rwco = opaque; + QEMUIOVector *qiov = rwco->iobuf; - rwco->ret = blk_co_pwritev(rwco->blk, rwco->offset, rwco->qiov->size, - rwco->qiov, rwco->flags); + rwco->ret = blk_co_pwritev(rwco->blk, rwco->offset, qiov->size, + qiov, rwco->flags); } static int blk_prw(BlockBackend *blk, int64_t offset, uint8_t *buf, @@ -1188,7 +1190,7 @@ static int blk_prw(BlockBackend *blk, int64_t offset, uint8_t *buf, rwco = (BlkRwCo) { .blk = blk, .offset = offset, - .qiov = &qiov, + .iobuf = &qiov, .flags = flags, .ret = NOT_DONE, }; @@ -1296,7 +1298,7 @@ static void blk_aio_complete_bh(void *opaque) } static BlockAIOCB *blk_aio_prwv(BlockBackend *blk, int64_t offset, int bytes, - QEMUIOVector *qiov, CoroutineEntry co_entry, + void *iobuf, CoroutineEntry co_entry, BdrvRequestFlags flags, BlockCompletionFunc *cb, void *opaque) { @@ -1308,7 +1310,7 @@ static BlockAIOCB *blk_aio_prwv(BlockBackend *blk, int64_t offset, int bytes, acb->rwco = (BlkRwCo) { .blk = blk, .offset = offset, - .qiov = qiov, + .iobuf = iobuf, .flags = flags, .ret = NOT_DONE, }; @@ -1331,10 +1333,11 @@ static void blk_aio_read_entry(void *opaque) { BlkAioEmAIOCB *acb = opaque; BlkRwCo *rwco = &acb->rwco; + QEMUIOVector *qiov = rwco->iobuf; - assert(rwco->qiov->size == acb->bytes); + assert(qiov->size == acb->bytes); rwco->ret = blk_co_preadv(rwco->blk, rwco->offset, acb->bytes, - rwco->qiov, rwco->flags); + qiov, rwco->flags); blk_aio_complete(acb); } @@ -1342,10 +1345,11 @@ static void blk_aio_write_entry(void *opaque) { BlkAioEmAIOCB *acb = opaque; BlkRwCo *rwco = &acb->rwco; + QEMUIOVector *qiov = rwco->iobuf; - assert(!rwco->qiov || rwco->qiov->size == acb->bytes); + assert(!qiov || qiov->size == acb->bytes); rwco->ret = blk_co_pwritev(rwco->blk, rwco->offset, acb->bytes, - rwco->qiov, rwco->flags); + qiov, rwco->flags); blk_aio_complete(acb); } @@ -1474,8 +1478,10 @@ int blk_co_ioctl(BlockBackend *blk, unsigned long int req, void *buf) static void blk_ioctl_entry(void *opaque) { BlkRwCo *rwco = opaque; + QEMUIOVector *qiov = rwco->iobuf; + rwco->ret = blk_co_ioctl(rwco->blk, rwco->offset, - rwco->qiov->iov[0].iov_base); + qiov->iov[0].iov_base); } int blk_ioctl(BlockBackend *blk, unsigned long int req, void *buf) @@ -1488,24 +1494,15 @@ static void blk_aio_ioctl_entry(void *opaque) BlkAioEmAIOCB *acb = opaque; BlkRwCo *rwco = &acb->rwco; - rwco->ret = blk_co_ioctl(rwco->blk, rwco->offset, - rwco->qiov->iov[0].iov_base); + rwco->ret = blk_co_ioctl(rwco->blk, rwco->offset, rwco->iobuf); + blk_aio_complete(acb); } BlockAIOCB *blk_aio_ioctl(BlockBackend *blk, unsigned long int req, void *buf, BlockCompletionFunc *cb, void *opaque) { - QEMUIOVector qiov; - struct iovec iov; - - iov = (struct iovec) { - .iov_base = buf, - .iov_len = 0, - }; - qemu_iovec_init_external(&qiov, &iov, 1); - - return blk_aio_prwv(blk, req, 0, &qiov, blk_aio_ioctl_entry, 0, cb, opaque); + return blk_aio_prwv(blk, req, 0, buf, blk_aio_ioctl_entry, 0, cb, opaque); } int blk_co_pdiscard(BlockBackend *blk, int64_t offset, int bytes) @@ -1949,7 +1946,9 @@ int blk_truncate(BlockBackend *blk, int64_t offset, PreallocMode prealloc, static void blk_pdiscard_entry(void *opaque) { BlkRwCo *rwco = opaque; - rwco->ret = blk_co_pdiscard(rwco->blk, rwco->offset, rwco->qiov->size); + QEMUIOVector *qiov = rwco->iobuf; + + rwco->ret = blk_co_pdiscard(rwco->blk, rwco->offset, qiov->size); } int blk_pdiscard(BlockBackend *blk, int64_t offset, int bytes) @@ -993,7 +993,7 @@ void cpu_synchronize_all_pre_loadvm(void) } } -static int do_vm_stop(RunState state) +static int do_vm_stop(RunState state, bool send_stop) { int ret = 0; @@ -1002,7 +1002,9 @@ static int do_vm_stop(RunState state) pause_all_vcpus(); runstate_set(state); vm_state_notify(0, state); - qapi_event_send_stop(&error_abort); + if (send_stop) { + qapi_event_send_stop(&error_abort); + } } bdrv_drain_all(); @@ -1012,6 +1014,14 @@ static int do_vm_stop(RunState state) return ret; } +/* Special vm_stop() variant for terminating the process. Historically clients + * did not expect a QMP STOP event and so we need to retain compatibility. + */ +int vm_shutdown(void) +{ + return do_vm_stop(RUN_STATE_SHUTDOWN, false); +} + static bool cpu_can_run(CPUState *cpu) { if (cpu->stop) { @@ -1994,7 +2004,7 @@ int vm_stop(RunState state) return 0; } - return do_vm_stop(state); + return do_vm_stop(state, true); } /** @@ -3393,7 +3393,6 @@ static bool flatview_access_valid(FlatView *fv, hwaddr addr, int len, if (!memory_access_is_direct(mr, is_write)) { l = memory_access_size(mr, l, addr); if (!memory_region_access_valid(mr, xlat, l, is_write)) { - rcu_read_unlock(); return false; } } diff --git a/fpu/softfloat.c b/fpu/softfloat.c index e124df9f7e..6e16284e66 100644 --- a/fpu/softfloat.c +++ b/fpu/softfloat.c @@ -1342,6 +1342,8 @@ static int64_t round_to_int_and_pack(FloatParts in, int rmode, switch (p.cls) { case float_class_snan: case float_class_qnan: + case float_class_dnan: + case float_class_msnan: return max; case float_class_inf: return p.sign ? min : max; @@ -1430,6 +1432,8 @@ static uint64_t round_to_uint_and_pack(FloatParts in, int rmode, uint64_t max, switch (p.cls) { case float_class_snan: case float_class_qnan: + case float_class_dnan: + case float_class_msnan: s->float_exception_flags = orig_flags | float_flag_invalid; return max; case float_class_inf: diff --git a/hw/block/dataplane/virtio-blk.c b/hw/block/dataplane/virtio-blk.c index 2cb990997e..101f32cf66 100644 --- a/hw/block/dataplane/virtio-blk.c +++ b/hw/block/dataplane/virtio-blk.c @@ -34,6 +34,7 @@ struct VirtIOBlockDataPlane { VirtIODevice *vdev; QEMUBH *bh; /* bh for guest notification */ unsigned long *batch_notify_vqs; + bool batch_notifications; /* Note that these EventNotifiers are assigned by value. This is * fine as long as you do not call event_notifier_cleanup on them @@ -47,8 +48,12 @@ struct VirtIOBlockDataPlane { /* Raise an interrupt to signal guest, if necessary */ void virtio_blk_data_plane_notify(VirtIOBlockDataPlane *s, VirtQueue *vq) { - set_bit(virtio_get_queue_index(vq), s->batch_notify_vqs); - qemu_bh_schedule(s->bh); + if (s->batch_notifications) { + set_bit(virtio_get_queue_index(vq), s->batch_notify_vqs); + qemu_bh_schedule(s->bh); + } else { + virtio_notify_irqfd(s->vdev, vq); + } } static void notify_guest_bh(void *opaque) @@ -177,6 +182,12 @@ int virtio_blk_data_plane_start(VirtIODevice *vdev) s->starting = true; + if (!virtio_vdev_has_feature(vdev, VIRTIO_RING_F_EVENT_IDX)) { + s->batch_notifications = true; + } else { + s->batch_notifications = false; + } + /* Set up guest notifier (irq) */ r = k->set_guest_notifiers(qbus->parent, nvqs, true); if (r != 0) { @@ -229,6 +240,22 @@ int virtio_blk_data_plane_start(VirtIODevice *vdev) return -ENOSYS; } +/* Stop notifications for new requests from guest. + * + * Context: BH in IOThread + */ +static void virtio_blk_data_plane_stop_bh(void *opaque) +{ + VirtIOBlockDataPlane *s = opaque; + unsigned i; + + for (i = 0; i < s->conf->num_queues; i++) { + VirtQueue *vq = virtio_get_queue(s->vdev, i); + + virtio_queue_aio_set_host_notifier_handler(vq, s->ctx, NULL); + } +} + /* Context: QEMU global mutex held */ void virtio_blk_data_plane_stop(VirtIODevice *vdev) { @@ -253,13 +280,7 @@ void virtio_blk_data_plane_stop(VirtIODevice *vdev) trace_virtio_blk_data_plane_stop(s); aio_context_acquire(s->ctx); - - /* Stop notifications for new requests from guest */ - for (i = 0; i < nvqs; i++) { - VirtQueue *vq = virtio_get_queue(s->vdev, i); - - virtio_queue_aio_set_host_notifier_handler(vq, s->ctx, NULL); - } + aio_wait_bh_oneshot(s->ctx, virtio_blk_data_plane_stop_bh, s); /* Drain and switch bs back to the QEMU main loop */ blk_set_aio_context(s->conf->conf.blk, qemu_get_aio_context()); diff --git a/hw/scsi/virtio-scsi-dataplane.c b/hw/scsi/virtio-scsi-dataplane.c index 1c33322ba6..912e5005d8 100644 --- a/hw/scsi/virtio-scsi-dataplane.c +++ b/hw/scsi/virtio-scsi-dataplane.c @@ -107,9 +107,10 @@ static int virtio_scsi_vring_init(VirtIOSCSI *s, VirtQueue *vq, int n, return 0; } -/* assumes s->ctx held */ -static void virtio_scsi_clear_aio(VirtIOSCSI *s) +/* Context: BH in IOThread */ +static void virtio_scsi_dataplane_stop_bh(void *opaque) { + VirtIOSCSI *s = opaque; VirtIOSCSICommon *vs = VIRTIO_SCSI_COMMON(s); int i; @@ -171,7 +172,7 @@ int virtio_scsi_dataplane_start(VirtIODevice *vdev) return 0; fail_vrings: - virtio_scsi_clear_aio(s); + aio_wait_bh_oneshot(s->ctx, virtio_scsi_dataplane_stop_bh, s); aio_context_release(s->ctx); for (i = 0; i < vs->conf.num_queues + 2; i++) { virtio_bus_set_host_notifier(VIRTIO_BUS(qbus), i, false); @@ -207,7 +208,7 @@ void virtio_scsi_dataplane_stop(VirtIODevice *vdev) s->dataplane_stopping = true; aio_context_acquire(s->ctx); - virtio_scsi_clear_aio(s); + aio_wait_bh_oneshot(s->ctx, virtio_scsi_dataplane_stop_bh, s); aio_context_release(s->ctx); blk_drain_all(); /* ensure there are no in-flight requests */ diff --git a/include/block/aio-wait.h b/include/block/aio-wait.h index a48c744fa8..f7a3972200 100644 --- a/include/block/aio-wait.h +++ b/include/block/aio-wait.h @@ -113,4 +113,17 @@ typedef struct { */ void aio_wait_kick(AioWait *wait); +/** + * aio_wait_bh_oneshot: + * @ctx: the aio context + * @cb: the BH callback function + * @opaque: user data for the BH callback function + * + * Run a BH in @ctx and wait for it to complete. + * + * Must be called from the main loop thread with @ctx acquired exactly once. + * Note that main loop event processing may occur. + */ +void aio_wait_bh_oneshot(AioContext *ctx, QEMUBHFunc *cb, void *opaque); + #endif /* QEMU_AIO_WAIT */ diff --git a/include/sysemu/iothread.h b/include/sysemu/iothread.h index 799614ffd2..8a7ac2c528 100644 --- a/include/sysemu/iothread.h +++ b/include/sysemu/iothread.h @@ -45,7 +45,6 @@ typedef struct { char *iothread_get_id(IOThread *iothread); IOThread *iothread_by_id(const char *id); AioContext *iothread_get_aio_context(IOThread *iothread); -void iothread_stop_all(void); GMainContext *iothread_get_g_main_context(IOThread *iothread); /* diff --git a/include/sysemu/sysemu.h b/include/sysemu/sysemu.h index d24ad09f37..356bfdc1c1 100644 --- a/include/sysemu/sysemu.h +++ b/include/sysemu/sysemu.h @@ -56,6 +56,7 @@ void vm_start(void); int vm_prepare_start(void); int vm_stop(RunState state); int vm_stop_force_state(RunState state); +int vm_shutdown(void); typedef enum WakeupReason { /* Always keep QEMU_WAKEUP_REASON_NONE = 0 */ diff --git a/iothread.c b/iothread.c index 2ec5a3bffe..1b3463cb00 100644 --- a/iothread.c +++ b/iothread.c @@ -101,18 +101,6 @@ void iothread_stop(IOThread *iothread) qemu_thread_join(&iothread->thread); } -static int iothread_stop_iter(Object *object, void *opaque) -{ - IOThread *iothread; - - iothread = (IOThread *)object_dynamic_cast(object, TYPE_IOTHREAD); - if (!iothread) { - return 0; - } - iothread_stop(iothread); - return 0; -} - static void iothread_instance_init(Object *obj) { IOThread *iothread = IOTHREAD(obj); @@ -333,25 +321,6 @@ IOThreadInfoList *qmp_query_iothreads(Error **errp) return head; } -void iothread_stop_all(void) -{ - Object *container = object_get_objects_root(); - BlockDriverState *bs; - BdrvNextIterator it; - - for (bs = bdrv_first(&it); bs; bs = bdrv_next(&it)) { - AioContext *ctx = bdrv_get_aio_context(bs); - if (ctx == qemu_get_aio_context()) { - continue; - } - aio_context_acquire(ctx); - bdrv_set_aio_context(bs, qemu_get_aio_context()); - aio_context_release(ctx); - } - - object_child_foreach(container, iothread_stop_iter, NULL); -} - static gpointer iothread_g_main_context_init(gpointer opaque) { AioContext *ctx; diff --git a/target/m68k/fpu_helper.c b/target/m68k/fpu_helper.c index cdb9b50462..62cbb0dff1 100644 --- a/target/m68k/fpu_helper.c +++ b/target/m68k/fpu_helper.c @@ -557,3 +557,38 @@ void HELPER(fscale)(CPUM68KState *env, FPReg *res, FPReg *val0, FPReg *val1) { res->d = floatx80_scale(val1->d, val0->d, &env->fp_status); } + +void HELPER(flognp1)(CPUM68KState *env, FPReg *res, FPReg *val) +{ + res->d = floatx80_lognp1(val->d, &env->fp_status); +} + +void HELPER(flogn)(CPUM68KState *env, FPReg *res, FPReg *val) +{ + res->d = floatx80_logn(val->d, &env->fp_status); +} + +void HELPER(flog10)(CPUM68KState *env, FPReg *res, FPReg *val) +{ + res->d = floatx80_log10(val->d, &env->fp_status); +} + +void HELPER(flog2)(CPUM68KState *env, FPReg *res, FPReg *val) +{ + res->d = floatx80_log2(val->d, &env->fp_status); +} + +void HELPER(fetox)(CPUM68KState *env, FPReg *res, FPReg *val) +{ + res->d = floatx80_etox(val->d, &env->fp_status); +} + +void HELPER(ftwotox)(CPUM68KState *env, FPReg *res, FPReg *val) +{ + res->d = floatx80_twotox(val->d, &env->fp_status); +} + +void HELPER(ftentox)(CPUM68KState *env, FPReg *res, FPReg *val) +{ + res->d = floatx80_tentox(val->d, &env->fp_status); +} diff --git a/target/m68k/helper.h b/target/m68k/helper.h index c348dced3a..9a9734c196 100644 --- a/target/m68k/helper.h +++ b/target/m68k/helper.h @@ -68,6 +68,13 @@ DEF_HELPER_4(frem, void, env, fp, fp, fp) DEF_HELPER_3(fgetexp, void, env, fp, fp) DEF_HELPER_3(fgetman, void, env, fp, fp) DEF_HELPER_4(fscale, void, env, fp, fp, fp) +DEF_HELPER_3(flognp1, void, env, fp, fp) +DEF_HELPER_3(flogn, void, env, fp, fp) +DEF_HELPER_3(flog10, void, env, fp, fp) +DEF_HELPER_3(flog2, void, env, fp, fp) +DEF_HELPER_3(fetox, void, env, fp, fp) +DEF_HELPER_3(ftwotox, void, env, fp, fp) +DEF_HELPER_3(ftentox, void, env, fp, fp) DEF_HELPER_3(mac_move, void, env, i32, i32) DEF_HELPER_3(macmulf, i64, env, i32, i32) diff --git a/target/m68k/softfloat.c b/target/m68k/softfloat.c index 9cb141900c..4bd5b9e6b7 100644 --- a/target/m68k/softfloat.c +++ b/target/m68k/softfloat.c @@ -21,6 +21,7 @@ #include "qemu/osdep.h" #include "softfloat.h" #include "fpu/softfloat-macros.h" +#include "softfloat_fpsp_tables.h" static floatx80 propagateFloatx80NaNOneArg(floatx80 a, float_status *status) { @@ -247,3 +248,1021 @@ floatx80 floatx80_scale(floatx80 a, floatx80 b, float_status *status) return roundAndPackFloatx80(status->floatx80_rounding_precision, aSign, aExp, aSig, 0, status); } + +floatx80 floatx80_move(floatx80 a, float_status *status) +{ + flag aSign; + int32_t aExp; + uint64_t aSig; + + aSig = extractFloatx80Frac(a); + aExp = extractFloatx80Exp(a); + aSign = extractFloatx80Sign(a); + + if (aExp == 0x7FFF) { + if ((uint64_t)(aSig << 1)) { + return propagateFloatx80NaNOneArg(a, status); + } + return a; + } + if (aExp == 0) { + if (aSig == 0) { + return a; + } + normalizeRoundAndPackFloatx80(status->floatx80_rounding_precision, + aSign, aExp, aSig, 0, status); + } + return roundAndPackFloatx80(status->floatx80_rounding_precision, aSign, + aExp, aSig, 0, status); +} + +/*---------------------------------------------------------------------------- +| Algorithms for transcendental functions supported by MC68881 and MC68882 +| mathematical coprocessors. The functions are derived from FPSP library. +*----------------------------------------------------------------------------*/ + +#define one_exp 0x3FFF +#define one_sig LIT64(0x8000000000000000) + +/*---------------------------------------------------------------------------- + | Function for compactifying extended double-precision floating point values. + *----------------------------------------------------------------------------*/ + +static int32_t floatx80_make_compact(int32_t aExp, uint64_t aSig) +{ + return (aExp << 16) | (aSig >> 48); +} + +/*---------------------------------------------------------------------------- + | Log base e of x plus 1 + *----------------------------------------------------------------------------*/ + +floatx80 floatx80_lognp1(floatx80 a, float_status *status) +{ + flag aSign; + int32_t aExp; + uint64_t aSig, fSig; + + int8_t user_rnd_mode, user_rnd_prec; + + int32_t compact, j, k; + floatx80 fp0, fp1, fp2, fp3, f, logof2, klog2, saveu; + + aSig = extractFloatx80Frac(a); + aExp = extractFloatx80Exp(a); + aSign = extractFloatx80Sign(a); + + if (aExp == 0x7FFF) { + if ((uint64_t) (aSig << 1)) { + propagateFloatx80NaNOneArg(a, status); + } + if (aSign) { + float_raise(float_flag_invalid, status); + return floatx80_default_nan(status); + } + return packFloatx80(0, floatx80_infinity.high, floatx80_infinity.low); + } + + if (aExp == 0 && aSig == 0) { + return packFloatx80(aSign, 0, 0); + } + + if (aSign && aExp >= one_exp) { + if (aExp == one_exp && aSig == one_sig) { + float_raise(float_flag_divbyzero, status); + packFloatx80(aSign, floatx80_infinity.high, floatx80_infinity.low); + } + float_raise(float_flag_invalid, status); + return floatx80_default_nan(status); + } + + if (aExp < 0x3f99 || (aExp == 0x3f99 && aSig == one_sig)) { + /* <= min threshold */ + float_raise(float_flag_inexact, status); + return floatx80_move(a, status); + } + + user_rnd_mode = status->float_rounding_mode; + user_rnd_prec = status->floatx80_rounding_precision; + status->float_rounding_mode = float_round_nearest_even; + status->floatx80_rounding_precision = 80; + + compact = floatx80_make_compact(aExp, aSig); + + fp0 = a; /* Z */ + fp1 = a; + + fp0 = floatx80_add(fp0, float32_to_floatx80(make_float32(0x3F800000), + status), status); /* X = (1+Z) */ + + aExp = extractFloatx80Exp(fp0); + aSig = extractFloatx80Frac(fp0); + + compact = floatx80_make_compact(aExp, aSig); + + if (compact < 0x3FFE8000 || compact > 0x3FFFC000) { + /* |X| < 1/2 or |X| > 3/2 */ + k = aExp - 0x3FFF; + fp1 = int32_to_floatx80(k, status); + + fSig = (aSig & LIT64(0xFE00000000000000)) | LIT64(0x0100000000000000); + j = (fSig >> 56) & 0x7E; /* DISPLACEMENT FOR 1/F */ + + f = packFloatx80(0, 0x3FFF, fSig); /* F */ + fp0 = packFloatx80(0, 0x3FFF, aSig); /* Y */ + + fp0 = floatx80_sub(fp0, f, status); /* Y-F */ + + lp1cont1: + /* LP1CONT1 */ + fp0 = floatx80_mul(fp0, log_tbl[j], status); /* FP0 IS U = (Y-F)/F */ + logof2 = packFloatx80(0, 0x3FFE, LIT64(0xB17217F7D1CF79AC)); + klog2 = floatx80_mul(fp1, logof2, status); /* FP1 IS K*LOG2 */ + fp2 = floatx80_mul(fp0, fp0, status); /* FP2 IS V=U*U */ + + fp3 = fp2; + fp1 = fp2; + + fp1 = floatx80_mul(fp1, float64_to_floatx80( + make_float64(0x3FC2499AB5E4040B), status), + status); /* V*A6 */ + fp2 = floatx80_mul(fp2, float64_to_floatx80( + make_float64(0xBFC555B5848CB7DB), status), + status); /* V*A5 */ + fp1 = floatx80_add(fp1, float64_to_floatx80( + make_float64(0x3FC99999987D8730), status), + status); /* A4+V*A6 */ + fp2 = floatx80_add(fp2, float64_to_floatx80( + make_float64(0xBFCFFFFFFF6F7E97), status), + status); /* A3+V*A5 */ + fp1 = floatx80_mul(fp1, fp3, status); /* V*(A4+V*A6) */ + fp2 = floatx80_mul(fp2, fp3, status); /* V*(A3+V*A5) */ + fp1 = floatx80_add(fp1, float64_to_floatx80( + make_float64(0x3FD55555555555A4), status), + status); /* A2+V*(A4+V*A6) */ + fp2 = floatx80_add(fp2, float64_to_floatx80( + make_float64(0xBFE0000000000008), status), + status); /* A1+V*(A3+V*A5) */ + fp1 = floatx80_mul(fp1, fp3, status); /* V*(A2+V*(A4+V*A6)) */ + fp2 = floatx80_mul(fp2, fp3, status); /* V*(A1+V*(A3+V*A5)) */ + fp1 = floatx80_mul(fp1, fp0, status); /* U*V*(A2+V*(A4+V*A6)) */ + fp0 = floatx80_add(fp0, fp2, status); /* U+V*(A1+V*(A3+V*A5)) */ + + fp1 = floatx80_add(fp1, log_tbl[j + 1], + status); /* LOG(F)+U*V*(A2+V*(A4+V*A6)) */ + fp0 = floatx80_add(fp0, fp1, status); /* FP0 IS LOG(F) + LOG(1+U) */ + + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + a = floatx80_add(fp0, klog2, status); + + float_raise(float_flag_inexact, status); + + return a; + } else if (compact < 0x3FFEF07D || compact > 0x3FFF8841) { + /* |X| < 1/16 or |X| > -1/16 */ + /* LP1CARE */ + fSig = (aSig & LIT64(0xFE00000000000000)) | LIT64(0x0100000000000000); + f = packFloatx80(0, 0x3FFF, fSig); /* F */ + j = (fSig >> 56) & 0x7E; /* DISPLACEMENT FOR 1/F */ + + if (compact >= 0x3FFF8000) { /* 1+Z >= 1 */ + /* KISZERO */ + fp0 = floatx80_sub(float32_to_floatx80(make_float32(0x3F800000), + status), f, status); /* 1-F */ + fp0 = floatx80_add(fp0, fp1, status); /* FP0 IS Y-F = (1-F)+Z */ + fp1 = packFloatx80(0, 0, 0); /* K = 0 */ + } else { + /* KISNEG */ + fp0 = floatx80_sub(float32_to_floatx80(make_float32(0x40000000), + status), f, status); /* 2-F */ + fp1 = floatx80_add(fp1, fp1, status); /* 2Z */ + fp0 = floatx80_add(fp0, fp1, status); /* FP0 IS Y-F = (2-F)+2Z */ + fp1 = packFloatx80(1, one_exp, one_sig); /* K = -1 */ + } + goto lp1cont1; + } else { + /* LP1ONE16 */ + fp1 = floatx80_add(fp1, fp1, status); /* FP1 IS 2Z */ + fp0 = floatx80_add(fp0, float32_to_floatx80(make_float32(0x3F800000), + status), status); /* FP0 IS 1+X */ + + /* LP1CONT2 */ + fp1 = floatx80_div(fp1, fp0, status); /* U */ + saveu = fp1; + fp0 = floatx80_mul(fp1, fp1, status); /* FP0 IS V = U*U */ + fp1 = floatx80_mul(fp0, fp0, status); /* FP1 IS W = V*V */ + + fp3 = float64_to_floatx80(make_float64(0x3F175496ADD7DAD6), + status); /* B5 */ + fp2 = float64_to_floatx80(make_float64(0x3F3C71C2FE80C7E0), + status); /* B4 */ + fp3 = floatx80_mul(fp3, fp1, status); /* W*B5 */ + fp2 = floatx80_mul(fp2, fp1, status); /* W*B4 */ + fp3 = floatx80_add(fp3, float64_to_floatx80( + make_float64(0x3F624924928BCCFF), status), + status); /* B3+W*B5 */ + fp2 = floatx80_add(fp2, float64_to_floatx80( + make_float64(0x3F899999999995EC), status), + status); /* B2+W*B4 */ + fp1 = floatx80_mul(fp1, fp3, status); /* W*(B3+W*B5) */ + fp2 = floatx80_mul(fp2, fp0, status); /* V*(B2+W*B4) */ + fp1 = floatx80_add(fp1, float64_to_floatx80( + make_float64(0x3FB5555555555555), status), + status); /* B1+W*(B3+W*B5) */ + + fp0 = floatx80_mul(fp0, saveu, status); /* FP0 IS U*V */ + fp1 = floatx80_add(fp1, fp2, + status); /* B1+W*(B3+W*B5) + V*(B2+W*B4) */ + fp0 = floatx80_mul(fp0, fp1, + status); /* U*V*([B1+W*(B3+W*B5)] + [V*(B2+W*B4)]) */ + + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + a = floatx80_add(fp0, saveu, status); + + /*if (!floatx80_is_zero(a)) { */ + float_raise(float_flag_inexact, status); + /*} */ + + return a; + } +} + +/*---------------------------------------------------------------------------- + | Log base e + *----------------------------------------------------------------------------*/ + +floatx80 floatx80_logn(floatx80 a, float_status *status) +{ + flag aSign; + int32_t aExp; + uint64_t aSig, fSig; + + int8_t user_rnd_mode, user_rnd_prec; + + int32_t compact, j, k, adjk; + floatx80 fp0, fp1, fp2, fp3, f, logof2, klog2, saveu; + + aSig = extractFloatx80Frac(a); + aExp = extractFloatx80Exp(a); + aSign = extractFloatx80Sign(a); + + if (aExp == 0x7FFF) { + if ((uint64_t) (aSig << 1)) { + propagateFloatx80NaNOneArg(a, status); + } + if (aSign == 0) { + return packFloatx80(0, floatx80_infinity.high, + floatx80_infinity.low); + } + } + + adjk = 0; + + if (aExp == 0) { + if (aSig == 0) { /* zero */ + float_raise(float_flag_divbyzero, status); + return packFloatx80(1, floatx80_infinity.high, + floatx80_infinity.low); + } + if ((aSig & one_sig) == 0) { /* denormal */ + normalizeFloatx80Subnormal(aSig, &aExp, &aSig); + adjk = -100; + aExp += 100; + a = packFloatx80(aSign, aExp, aSig); + } + } + + if (aSign) { + float_raise(float_flag_invalid, status); + return floatx80_default_nan(status); + } + + user_rnd_mode = status->float_rounding_mode; + user_rnd_prec = status->floatx80_rounding_precision; + status->float_rounding_mode = float_round_nearest_even; + status->floatx80_rounding_precision = 80; + + compact = floatx80_make_compact(aExp, aSig); + + if (compact < 0x3FFEF07D || compact > 0x3FFF8841) { + /* |X| < 15/16 or |X| > 17/16 */ + k = aExp - 0x3FFF; + k += adjk; + fp1 = int32_to_floatx80(k, status); + + fSig = (aSig & LIT64(0xFE00000000000000)) | LIT64(0x0100000000000000); + j = (fSig >> 56) & 0x7E; /* DISPLACEMENT FOR 1/F */ + + f = packFloatx80(0, 0x3FFF, fSig); /* F */ + fp0 = packFloatx80(0, 0x3FFF, aSig); /* Y */ + + fp0 = floatx80_sub(fp0, f, status); /* Y-F */ + + /* LP1CONT1 */ + fp0 = floatx80_mul(fp0, log_tbl[j], status); /* FP0 IS U = (Y-F)/F */ + logof2 = packFloatx80(0, 0x3FFE, LIT64(0xB17217F7D1CF79AC)); + klog2 = floatx80_mul(fp1, logof2, status); /* FP1 IS K*LOG2 */ + fp2 = floatx80_mul(fp0, fp0, status); /* FP2 IS V=U*U */ + + fp3 = fp2; + fp1 = fp2; + + fp1 = floatx80_mul(fp1, float64_to_floatx80( + make_float64(0x3FC2499AB5E4040B), status), + status); /* V*A6 */ + fp2 = floatx80_mul(fp2, float64_to_floatx80( + make_float64(0xBFC555B5848CB7DB), status), + status); /* V*A5 */ + fp1 = floatx80_add(fp1, float64_to_floatx80( + make_float64(0x3FC99999987D8730), status), + status); /* A4+V*A6 */ + fp2 = floatx80_add(fp2, float64_to_floatx80( + make_float64(0xBFCFFFFFFF6F7E97), status), + status); /* A3+V*A5 */ + fp1 = floatx80_mul(fp1, fp3, status); /* V*(A4+V*A6) */ + fp2 = floatx80_mul(fp2, fp3, status); /* V*(A3+V*A5) */ + fp1 = floatx80_add(fp1, float64_to_floatx80( + make_float64(0x3FD55555555555A4), status), + status); /* A2+V*(A4+V*A6) */ + fp2 = floatx80_add(fp2, float64_to_floatx80( + make_float64(0xBFE0000000000008), status), + status); /* A1+V*(A3+V*A5) */ + fp1 = floatx80_mul(fp1, fp3, status); /* V*(A2+V*(A4+V*A6)) */ + fp2 = floatx80_mul(fp2, fp3, status); /* V*(A1+V*(A3+V*A5)) */ + fp1 = floatx80_mul(fp1, fp0, status); /* U*V*(A2+V*(A4+V*A6)) */ + fp0 = floatx80_add(fp0, fp2, status); /* U+V*(A1+V*(A3+V*A5)) */ + + fp1 = floatx80_add(fp1, log_tbl[j + 1], + status); /* LOG(F)+U*V*(A2+V*(A4+V*A6)) */ + fp0 = floatx80_add(fp0, fp1, status); /* FP0 IS LOG(F) + LOG(1+U) */ + + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + a = floatx80_add(fp0, klog2, status); + + float_raise(float_flag_inexact, status); + + return a; + } else { /* |X-1| >= 1/16 */ + fp0 = a; + fp1 = a; + fp1 = floatx80_sub(fp1, float32_to_floatx80(make_float32(0x3F800000), + status), status); /* FP1 IS X-1 */ + fp0 = floatx80_add(fp0, float32_to_floatx80(make_float32(0x3F800000), + status), status); /* FP0 IS X+1 */ + fp1 = floatx80_add(fp1, fp1, status); /* FP1 IS 2(X-1) */ + + /* LP1CONT2 */ + fp1 = floatx80_div(fp1, fp0, status); /* U */ + saveu = fp1; + fp0 = floatx80_mul(fp1, fp1, status); /* FP0 IS V = U*U */ + fp1 = floatx80_mul(fp0, fp0, status); /* FP1 IS W = V*V */ + + fp3 = float64_to_floatx80(make_float64(0x3F175496ADD7DAD6), + status); /* B5 */ + fp2 = float64_to_floatx80(make_float64(0x3F3C71C2FE80C7E0), + status); /* B4 */ + fp3 = floatx80_mul(fp3, fp1, status); /* W*B5 */ + fp2 = floatx80_mul(fp2, fp1, status); /* W*B4 */ + fp3 = floatx80_add(fp3, float64_to_floatx80( + make_float64(0x3F624924928BCCFF), status), + status); /* B3+W*B5 */ + fp2 = floatx80_add(fp2, float64_to_floatx80( + make_float64(0x3F899999999995EC), status), + status); /* B2+W*B4 */ + fp1 = floatx80_mul(fp1, fp3, status); /* W*(B3+W*B5) */ + fp2 = floatx80_mul(fp2, fp0, status); /* V*(B2+W*B4) */ + fp1 = floatx80_add(fp1, float64_to_floatx80( + make_float64(0x3FB5555555555555), status), + status); /* B1+W*(B3+W*B5) */ + + fp0 = floatx80_mul(fp0, saveu, status); /* FP0 IS U*V */ + fp1 = floatx80_add(fp1, fp2, status); /* B1+W*(B3+W*B5) + V*(B2+W*B4) */ + fp0 = floatx80_mul(fp0, fp1, + status); /* U*V*([B1+W*(B3+W*B5)] + [V*(B2+W*B4)]) */ + + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + a = floatx80_add(fp0, saveu, status); + + /*if (!floatx80_is_zero(a)) { */ + float_raise(float_flag_inexact, status); + /*} */ + + return a; + } +} + +/*---------------------------------------------------------------------------- + | Log base 10 + *----------------------------------------------------------------------------*/ + +floatx80 floatx80_log10(floatx80 a, float_status *status) +{ + flag aSign; + int32_t aExp; + uint64_t aSig; + + int8_t user_rnd_mode, user_rnd_prec; + + floatx80 fp0, fp1; + + aSig = extractFloatx80Frac(a); + aExp = extractFloatx80Exp(a); + aSign = extractFloatx80Sign(a); + + if (aExp == 0x7FFF) { + if ((uint64_t) (aSig << 1)) { + propagateFloatx80NaNOneArg(a, status); + } + if (aSign == 0) { + return packFloatx80(0, floatx80_infinity.high, + floatx80_infinity.low); + } + } + + if (aExp == 0 && aSig == 0) { + float_raise(float_flag_divbyzero, status); + return packFloatx80(1, floatx80_infinity.high, + floatx80_infinity.low); + } + + if (aSign) { + float_raise(float_flag_invalid, status); + return floatx80_default_nan(status); + } + + user_rnd_mode = status->float_rounding_mode; + user_rnd_prec = status->floatx80_rounding_precision; + status->float_rounding_mode = float_round_nearest_even; + status->floatx80_rounding_precision = 80; + + fp0 = floatx80_logn(a, status); + fp1 = packFloatx80(0, 0x3FFD, LIT64(0xDE5BD8A937287195)); /* INV_L10 */ + + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + a = floatx80_mul(fp0, fp1, status); /* LOGN(X)*INV_L10 */ + + float_raise(float_flag_inexact, status); + + return a; +} + +/*---------------------------------------------------------------------------- + | Log base 2 + *----------------------------------------------------------------------------*/ + +floatx80 floatx80_log2(floatx80 a, float_status *status) +{ + flag aSign; + int32_t aExp; + uint64_t aSig; + + int8_t user_rnd_mode, user_rnd_prec; + + floatx80 fp0, fp1; + + aSig = extractFloatx80Frac(a); + aExp = extractFloatx80Exp(a); + aSign = extractFloatx80Sign(a); + + if (aExp == 0x7FFF) { + if ((uint64_t) (aSig << 1)) { + propagateFloatx80NaNOneArg(a, status); + } + if (aSign == 0) { + return packFloatx80(0, floatx80_infinity.high, + floatx80_infinity.low); + } + } + + if (aExp == 0) { + if (aSig == 0) { + float_raise(float_flag_divbyzero, status); + return packFloatx80(1, floatx80_infinity.high, + floatx80_infinity.low); + } + normalizeFloatx80Subnormal(aSig, &aExp, &aSig); + } + + if (aSign) { + float_raise(float_flag_invalid, status); + return floatx80_default_nan(status); + } + + user_rnd_mode = status->float_rounding_mode; + user_rnd_prec = status->floatx80_rounding_precision; + status->float_rounding_mode = float_round_nearest_even; + status->floatx80_rounding_precision = 80; + + if (aSig == one_sig) { /* X is 2^k */ + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + a = int32_to_floatx80(aExp - 0x3FFF, status); + } else { + fp0 = floatx80_logn(a, status); + fp1 = packFloatx80(0, 0x3FFF, LIT64(0xB8AA3B295C17F0BC)); /* INV_L2 */ + + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + a = floatx80_mul(fp0, fp1, status); /* LOGN(X)*INV_L2 */ + } + + float_raise(float_flag_inexact, status); + + return a; +} + +/*---------------------------------------------------------------------------- + | e to x + *----------------------------------------------------------------------------*/ + +floatx80 floatx80_etox(floatx80 a, float_status *status) +{ + flag aSign; + int32_t aExp; + uint64_t aSig; + + int8_t user_rnd_mode, user_rnd_prec; + + int32_t compact, n, j, k, m, m1; + floatx80 fp0, fp1, fp2, fp3, l2, scale, adjscale; + flag adjflag; + + aSig = extractFloatx80Frac(a); + aExp = extractFloatx80Exp(a); + aSign = extractFloatx80Sign(a); + + if (aExp == 0x7FFF) { + if ((uint64_t) (aSig << 1)) { + return propagateFloatx80NaNOneArg(a, status); + } + if (aSign) { + return packFloatx80(0, 0, 0); + } + return packFloatx80(0, floatx80_infinity.high, + floatx80_infinity.low); + } + + if (aExp == 0 && aSig == 0) { + return packFloatx80(0, one_exp, one_sig); + } + + user_rnd_mode = status->float_rounding_mode; + user_rnd_prec = status->floatx80_rounding_precision; + status->float_rounding_mode = float_round_nearest_even; + status->floatx80_rounding_precision = 80; + + adjflag = 0; + + if (aExp >= 0x3FBE) { /* |X| >= 2^(-65) */ + compact = floatx80_make_compact(aExp, aSig); + + if (compact < 0x400CB167) { /* |X| < 16380 log2 */ + fp0 = a; + fp1 = a; + fp0 = floatx80_mul(fp0, float32_to_floatx80( + make_float32(0x42B8AA3B), status), + status); /* 64/log2 * X */ + adjflag = 0; + n = floatx80_to_int32(fp0, status); /* int(64/log2*X) */ + fp0 = int32_to_floatx80(n, status); + + j = n & 0x3F; /* J = N mod 64 */ + m = n / 64; /* NOTE: this is really arithmetic right shift by 6 */ + if (n < 0 && j) { + /* arithmetic right shift is division and + * round towards minus infinity + */ + m--; + } + m += 0x3FFF; /* biased exponent of 2^(M) */ + + expcont1: + fp2 = fp0; /* N */ + fp0 = floatx80_mul(fp0, float32_to_floatx80( + make_float32(0xBC317218), status), + status); /* N * L1, L1 = lead(-log2/64) */ + l2 = packFloatx80(0, 0x3FDC, LIT64(0x82E308654361C4C6)); + fp2 = floatx80_mul(fp2, l2, status); /* N * L2, L1+L2 = -log2/64 */ + fp0 = floatx80_add(fp0, fp1, status); /* X + N*L1 */ + fp0 = floatx80_add(fp0, fp2, status); /* R */ + + fp1 = floatx80_mul(fp0, fp0, status); /* S = R*R */ + fp2 = float32_to_floatx80(make_float32(0x3AB60B70), + status); /* A5 */ + fp2 = floatx80_mul(fp2, fp1, status); /* fp2 is S*A5 */ + fp3 = floatx80_mul(float32_to_floatx80(make_float32(0x3C088895), + status), fp1, + status); /* fp3 is S*A4 */ + fp2 = floatx80_add(fp2, float64_to_floatx80(make_float64( + 0x3FA5555555554431), status), + status); /* fp2 is A3+S*A5 */ + fp3 = floatx80_add(fp3, float64_to_floatx80(make_float64( + 0x3FC5555555554018), status), + status); /* fp3 is A2+S*A4 */ + fp2 = floatx80_mul(fp2, fp1, status); /* fp2 is S*(A3+S*A5) */ + fp3 = floatx80_mul(fp3, fp1, status); /* fp3 is S*(A2+S*A4) */ + fp2 = floatx80_add(fp2, float32_to_floatx80( + make_float32(0x3F000000), status), + status); /* fp2 is A1+S*(A3+S*A5) */ + fp3 = floatx80_mul(fp3, fp0, status); /* fp3 IS R*S*(A2+S*A4) */ + fp2 = floatx80_mul(fp2, fp1, + status); /* fp2 IS S*(A1+S*(A3+S*A5)) */ + fp0 = floatx80_add(fp0, fp3, status); /* fp0 IS R+R*S*(A2+S*A4) */ + fp0 = floatx80_add(fp0, fp2, status); /* fp0 IS EXP(R) - 1 */ + + fp1 = exp_tbl[j]; + fp0 = floatx80_mul(fp0, fp1, status); /* 2^(J/64)*(Exp(R)-1) */ + fp0 = floatx80_add(fp0, float32_to_floatx80(exp_tbl2[j], status), + status); /* accurate 2^(J/64) */ + fp0 = floatx80_add(fp0, fp1, + status); /* 2^(J/64) + 2^(J/64)*(Exp(R)-1) */ + + scale = packFloatx80(0, m, one_sig); + if (adjflag) { + adjscale = packFloatx80(0, m1, one_sig); + fp0 = floatx80_mul(fp0, adjscale, status); + } + + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + a = floatx80_mul(fp0, scale, status); + + float_raise(float_flag_inexact, status); + + return a; + } else { /* |X| >= 16380 log2 */ + if (compact > 0x400CB27C) { /* |X| >= 16480 log2 */ + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + if (aSign) { + a = roundAndPackFloatx80( + status->floatx80_rounding_precision, + 0, -0x1000, aSig, 0, status); + } else { + a = roundAndPackFloatx80( + status->floatx80_rounding_precision, + 0, 0x8000, aSig, 0, status); + } + float_raise(float_flag_inexact, status); + + return a; + } else { + fp0 = a; + fp1 = a; + fp0 = floatx80_mul(fp0, float32_to_floatx80( + make_float32(0x42B8AA3B), status), + status); /* 64/log2 * X */ + adjflag = 1; + n = floatx80_to_int32(fp0, status); /* int(64/log2*X) */ + fp0 = int32_to_floatx80(n, status); + + j = n & 0x3F; /* J = N mod 64 */ + /* NOTE: this is really arithmetic right shift by 6 */ + k = n / 64; + if (n < 0 && j) { + /* arithmetic right shift is division and + * round towards minus infinity + */ + k--; + } + /* NOTE: this is really arithmetic right shift by 1 */ + m1 = k / 2; + if (k < 0 && (k & 1)) { + /* arithmetic right shift is division and + * round towards minus infinity + */ + m1--; + } + m = k - m1; + m1 += 0x3FFF; /* biased exponent of 2^(M1) */ + m += 0x3FFF; /* biased exponent of 2^(M) */ + + goto expcont1; + } + } + } else { /* |X| < 2^(-65) */ + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + a = floatx80_add(a, float32_to_floatx80(make_float32(0x3F800000), + status), status); /* 1 + X */ + + float_raise(float_flag_inexact, status); + + return a; + } +} + +/*---------------------------------------------------------------------------- + | 2 to x + *----------------------------------------------------------------------------*/ + +floatx80 floatx80_twotox(floatx80 a, float_status *status) +{ + flag aSign; + int32_t aExp; + uint64_t aSig; + + int8_t user_rnd_mode, user_rnd_prec; + + int32_t compact, n, j, l, m, m1; + floatx80 fp0, fp1, fp2, fp3, adjfact, fact1, fact2; + + aSig = extractFloatx80Frac(a); + aExp = extractFloatx80Exp(a); + aSign = extractFloatx80Sign(a); + + if (aExp == 0x7FFF) { + if ((uint64_t) (aSig << 1)) { + return propagateFloatx80NaNOneArg(a, status); + } + if (aSign) { + return packFloatx80(0, 0, 0); + } + return packFloatx80(0, floatx80_infinity.high, + floatx80_infinity.low); + } + + if (aExp == 0 && aSig == 0) { + return packFloatx80(0, one_exp, one_sig); + } + + user_rnd_mode = status->float_rounding_mode; + user_rnd_prec = status->floatx80_rounding_precision; + status->float_rounding_mode = float_round_nearest_even; + status->floatx80_rounding_precision = 80; + + fp0 = a; + + compact = floatx80_make_compact(aExp, aSig); + + if (compact < 0x3FB98000 || compact > 0x400D80C0) { + /* |X| > 16480 or |X| < 2^(-70) */ + if (compact > 0x3FFF8000) { /* |X| > 16480 */ + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + if (aSign) { + return roundAndPackFloatx80(status->floatx80_rounding_precision, + 0, -0x1000, aSig, 0, status); + } else { + return roundAndPackFloatx80(status->floatx80_rounding_precision, + 0, 0x8000, aSig, 0, status); + } + } else { /* |X| < 2^(-70) */ + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + a = floatx80_add(fp0, float32_to_floatx80( + make_float32(0x3F800000), status), + status); /* 1 + X */ + + float_raise(float_flag_inexact, status); + + return a; + } + } else { /* 2^(-70) <= |X| <= 16480 */ + fp1 = fp0; /* X */ + fp1 = floatx80_mul(fp1, float32_to_floatx80( + make_float32(0x42800000), status), + status); /* X * 64 */ + n = floatx80_to_int32(fp1, status); + fp1 = int32_to_floatx80(n, status); + j = n & 0x3F; + l = n / 64; /* NOTE: this is really arithmetic right shift by 6 */ + if (n < 0 && j) { + /* arithmetic right shift is division and + * round towards minus infinity + */ + l--; + } + m = l / 2; /* NOTE: this is really arithmetic right shift by 1 */ + if (l < 0 && (l & 1)) { + /* arithmetic right shift is division and + * round towards minus infinity + */ + m--; + } + m1 = l - m; + m1 += 0x3FFF; /* ADJFACT IS 2^(M') */ + + adjfact = packFloatx80(0, m1, one_sig); + fact1 = exp2_tbl[j]; + fact1.high += m; + fact2.high = exp2_tbl2[j] >> 16; + fact2.high += m; + fact2.low = (uint64_t)(exp2_tbl2[j] & 0xFFFF); + fact2.low <<= 48; + + fp1 = floatx80_mul(fp1, float32_to_floatx80( + make_float32(0x3C800000), status), + status); /* (1/64)*N */ + fp0 = floatx80_sub(fp0, fp1, status); /* X - (1/64)*INT(64 X) */ + fp2 = packFloatx80(0, 0x3FFE, LIT64(0xB17217F7D1CF79AC)); /* LOG2 */ + fp0 = floatx80_mul(fp0, fp2, status); /* R */ + + /* EXPR */ + fp1 = floatx80_mul(fp0, fp0, status); /* S = R*R */ + fp2 = float64_to_floatx80(make_float64(0x3F56C16D6F7BD0B2), + status); /* A5 */ + fp3 = float64_to_floatx80(make_float64(0x3F811112302C712C), + status); /* A4 */ + fp2 = floatx80_mul(fp2, fp1, status); /* S*A5 */ + fp3 = floatx80_mul(fp3, fp1, status); /* S*A4 */ + fp2 = floatx80_add(fp2, float64_to_floatx80( + make_float64(0x3FA5555555554CC1), status), + status); /* A3+S*A5 */ + fp3 = floatx80_add(fp3, float64_to_floatx80( + make_float64(0x3FC5555555554A54), status), + status); /* A2+S*A4 */ + fp2 = floatx80_mul(fp2, fp1, status); /* S*(A3+S*A5) */ + fp3 = floatx80_mul(fp3, fp1, status); /* S*(A2+S*A4) */ + fp2 = floatx80_add(fp2, float64_to_floatx80( + make_float64(0x3FE0000000000000), status), + status); /* A1+S*(A3+S*A5) */ + fp3 = floatx80_mul(fp3, fp0, status); /* R*S*(A2+S*A4) */ + + fp2 = floatx80_mul(fp2, fp1, status); /* S*(A1+S*(A3+S*A5)) */ + fp0 = floatx80_add(fp0, fp3, status); /* R+R*S*(A2+S*A4) */ + fp0 = floatx80_add(fp0, fp2, status); /* EXP(R) - 1 */ + + fp0 = floatx80_mul(fp0, fact1, status); + fp0 = floatx80_add(fp0, fact2, status); + fp0 = floatx80_add(fp0, fact1, status); + + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + a = floatx80_mul(fp0, adjfact, status); + + float_raise(float_flag_inexact, status); + + return a; + } +} + +/*---------------------------------------------------------------------------- + | 10 to x + *----------------------------------------------------------------------------*/ + +floatx80 floatx80_tentox(floatx80 a, float_status *status) +{ + flag aSign; + int32_t aExp; + uint64_t aSig; + + int8_t user_rnd_mode, user_rnd_prec; + + int32_t compact, n, j, l, m, m1; + floatx80 fp0, fp1, fp2, fp3, adjfact, fact1, fact2; + + aSig = extractFloatx80Frac(a); + aExp = extractFloatx80Exp(a); + aSign = extractFloatx80Sign(a); + + if (aExp == 0x7FFF) { + if ((uint64_t) (aSig << 1)) { + return propagateFloatx80NaNOneArg(a, status); + } + if (aSign) { + return packFloatx80(0, 0, 0); + } + return packFloatx80(0, floatx80_infinity.high, + floatx80_infinity.low); + } + + if (aExp == 0 && aSig == 0) { + return packFloatx80(0, one_exp, one_sig); + } + + user_rnd_mode = status->float_rounding_mode; + user_rnd_prec = status->floatx80_rounding_precision; + status->float_rounding_mode = float_round_nearest_even; + status->floatx80_rounding_precision = 80; + + fp0 = a; + + compact = floatx80_make_compact(aExp, aSig); + + if (compact < 0x3FB98000 || compact > 0x400B9B07) { + /* |X| > 16480 LOG2/LOG10 or |X| < 2^(-70) */ + if (compact > 0x3FFF8000) { /* |X| > 16480 */ + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + if (aSign) { + return roundAndPackFloatx80(status->floatx80_rounding_precision, + 0, -0x1000, aSig, 0, status); + } else { + return roundAndPackFloatx80(status->floatx80_rounding_precision, + 0, 0x8000, aSig, 0, status); + } + } else { /* |X| < 2^(-70) */ + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + a = floatx80_add(fp0, float32_to_floatx80( + make_float32(0x3F800000), status), + status); /* 1 + X */ + + float_raise(float_flag_inexact, status); + + return a; + } + } else { /* 2^(-70) <= |X| <= 16480 LOG 2 / LOG 10 */ + fp1 = fp0; /* X */ + fp1 = floatx80_mul(fp1, float64_to_floatx80( + make_float64(0x406A934F0979A371), + status), status); /* X*64*LOG10/LOG2 */ + n = floatx80_to_int32(fp1, status); /* N=INT(X*64*LOG10/LOG2) */ + fp1 = int32_to_floatx80(n, status); + + j = n & 0x3F; + l = n / 64; /* NOTE: this is really arithmetic right shift by 6 */ + if (n < 0 && j) { + /* arithmetic right shift is division and + * round towards minus infinity + */ + l--; + } + m = l / 2; /* NOTE: this is really arithmetic right shift by 1 */ + if (l < 0 && (l & 1)) { + /* arithmetic right shift is division and + * round towards minus infinity + */ + m--; + } + m1 = l - m; + m1 += 0x3FFF; /* ADJFACT IS 2^(M') */ + + adjfact = packFloatx80(0, m1, one_sig); + fact1 = exp2_tbl[j]; + fact1.high += m; + fact2.high = exp2_tbl2[j] >> 16; + fact2.high += m; + fact2.low = (uint64_t)(exp2_tbl2[j] & 0xFFFF); + fact2.low <<= 48; + + fp2 = fp1; /* N */ + fp1 = floatx80_mul(fp1, float64_to_floatx80( + make_float64(0x3F734413509F8000), status), + status); /* N*(LOG2/64LOG10)_LEAD */ + fp3 = packFloatx80(1, 0x3FCD, LIT64(0xC0219DC1DA994FD2)); + fp2 = floatx80_mul(fp2, fp3, status); /* N*(LOG2/64LOG10)_TRAIL */ + fp0 = floatx80_sub(fp0, fp1, status); /* X - N L_LEAD */ + fp0 = floatx80_sub(fp0, fp2, status); /* X - N L_TRAIL */ + fp2 = packFloatx80(0, 0x4000, LIT64(0x935D8DDDAAA8AC17)); /* LOG10 */ + fp0 = floatx80_mul(fp0, fp2, status); /* R */ + + /* EXPR */ + fp1 = floatx80_mul(fp0, fp0, status); /* S = R*R */ + fp2 = float64_to_floatx80(make_float64(0x3F56C16D6F7BD0B2), + status); /* A5 */ + fp3 = float64_to_floatx80(make_float64(0x3F811112302C712C), + status); /* A4 */ + fp2 = floatx80_mul(fp2, fp1, status); /* S*A5 */ + fp3 = floatx80_mul(fp3, fp1, status); /* S*A4 */ + fp2 = floatx80_add(fp2, float64_to_floatx80( + make_float64(0x3FA5555555554CC1), status), + status); /* A3+S*A5 */ + fp3 = floatx80_add(fp3, float64_to_floatx80( + make_float64(0x3FC5555555554A54), status), + status); /* A2+S*A4 */ + fp2 = floatx80_mul(fp2, fp1, status); /* S*(A3+S*A5) */ + fp3 = floatx80_mul(fp3, fp1, status); /* S*(A2+S*A4) */ + fp2 = floatx80_add(fp2, float64_to_floatx80( + make_float64(0x3FE0000000000000), status), + status); /* A1+S*(A3+S*A5) */ + fp3 = floatx80_mul(fp3, fp0, status); /* R*S*(A2+S*A4) */ + + fp2 = floatx80_mul(fp2, fp1, status); /* S*(A1+S*(A3+S*A5)) */ + fp0 = floatx80_add(fp0, fp3, status); /* R+R*S*(A2+S*A4) */ + fp0 = floatx80_add(fp0, fp2, status); /* EXP(R) - 1 */ + + fp0 = floatx80_mul(fp0, fact1, status); + fp0 = floatx80_add(fp0, fact2, status); + fp0 = floatx80_add(fp0, fact1, status); + + status->float_rounding_mode = user_rnd_mode; + status->floatx80_rounding_precision = user_rnd_prec; + + a = floatx80_mul(fp0, adjfact, status); + + float_raise(float_flag_inexact, status); + + return a; + } +} diff --git a/target/m68k/softfloat.h b/target/m68k/softfloat.h index 78fbc0cd0c..d28e49fe9f 100644 --- a/target/m68k/softfloat.h +++ b/target/m68k/softfloat.h @@ -26,4 +26,12 @@ floatx80 floatx80_mod(floatx80 a, floatx80 b, float_status *status); floatx80 floatx80_getman(floatx80 a, float_status *status); floatx80 floatx80_getexp(floatx80 a, float_status *status); floatx80 floatx80_scale(floatx80 a, floatx80 b, float_status *status); +floatx80 floatx80_move(floatx80 a, float_status *status); +floatx80 floatx80_lognp1(floatx80 a, float_status *status); +floatx80 floatx80_logn(floatx80 a, float_status *status); +floatx80 floatx80_log10(floatx80 a, float_status *status); +floatx80 floatx80_log2(floatx80 a, float_status *status); +floatx80 floatx80_etox(floatx80 a, float_status *status); +floatx80 floatx80_twotox(floatx80 a, float_status *status); +floatx80 floatx80_tentox(floatx80 a, float_status *status); #endif diff --git a/target/m68k/softfloat_fpsp_tables.h b/target/m68k/softfloat_fpsp_tables.h new file mode 100644 index 0000000000..dd76dc0373 --- /dev/null +++ b/target/m68k/softfloat_fpsp_tables.h @@ -0,0 +1,374 @@ +/* + * Ported from a work by Andreas Grabher for Previous, NeXT Computer Emulator, + * derived from NetBSD M68040 FPSP functions, + * derived from release 2a of the SoftFloat IEC/IEEE Floating-point Arithmetic + * Package. Those parts of the code (and some later contributions) are + * provided under that license, as detailed below. + * It has subsequently been modified by contributors to the QEMU Project, + * so some portions are provided under: + * the SoftFloat-2a license + * the BSD license + * GPL-v2-or-later + * + * Any future contributions to this file will be taken to be licensed under + * the Softfloat-2a license unless specifically indicated otherwise. + */ + +/* Portions of this work are licensed under the terms of the GNU GPL, + * version 2 or later. See the COPYING file in the top-level directory. + */ + +#ifndef TARGET_M68K_SOFTFLOAT_FPSP_TABLES_H +#define TARGET_M68K_SOFTFLOAT_FPSP_TABLES_H + +static const floatx80 log_tbl[128] = { + make_floatx80_init(0x3FFE, 0xFE03F80FE03F80FE), + make_floatx80_init(0x3FF7, 0xFF015358833C47E2), + make_floatx80_init(0x3FFE, 0xFA232CF252138AC0), + make_floatx80_init(0x3FF9, 0xBDC8D83EAD88D549), + make_floatx80_init(0x3FFE, 0xF6603D980F6603DA), + make_floatx80_init(0x3FFA, 0x9CF43DCFF5EAFD48), + make_floatx80_init(0x3FFE, 0xF2B9D6480F2B9D65), + make_floatx80_init(0x3FFA, 0xDA16EB88CB8DF614), + make_floatx80_init(0x3FFE, 0xEF2EB71FC4345238), + make_floatx80_init(0x3FFB, 0x8B29B7751BD70743), + make_floatx80_init(0x3FFE, 0xEBBDB2A5C1619C8C), + make_floatx80_init(0x3FFB, 0xA8D839F830C1FB49), + make_floatx80_init(0x3FFE, 0xE865AC7B7603A197), + make_floatx80_init(0x3FFB, 0xC61A2EB18CD907AD), + make_floatx80_init(0x3FFE, 0xE525982AF70C880E), + make_floatx80_init(0x3FFB, 0xE2F2A47ADE3A18AF), + make_floatx80_init(0x3FFE, 0xE1FC780E1FC780E2), + make_floatx80_init(0x3FFB, 0xFF64898EDF55D551), + make_floatx80_init(0x3FFE, 0xDEE95C4CA037BA57), + make_floatx80_init(0x3FFC, 0x8DB956A97B3D0148), + make_floatx80_init(0x3FFE, 0xDBEB61EED19C5958), + make_floatx80_init(0x3FFC, 0x9B8FE100F47BA1DE), + make_floatx80_init(0x3FFE, 0xD901B2036406C80E), + make_floatx80_init(0x3FFC, 0xA9372F1D0DA1BD17), + make_floatx80_init(0x3FFE, 0xD62B80D62B80D62C), + make_floatx80_init(0x3FFC, 0xB6B07F38CE90E46B), + make_floatx80_init(0x3FFE, 0xD3680D3680D3680D), + make_floatx80_init(0x3FFC, 0xC3FD032906488481), + make_floatx80_init(0x3FFE, 0xD0B69FCBD2580D0B), + make_floatx80_init(0x3FFC, 0xD11DE0FF15AB18CA), + make_floatx80_init(0x3FFE, 0xCE168A7725080CE1), + make_floatx80_init(0x3FFC, 0xDE1433A16C66B150), + make_floatx80_init(0x3FFE, 0xCB8727C065C393E0), + make_floatx80_init(0x3FFC, 0xEAE10B5A7DDC8ADD), + make_floatx80_init(0x3FFE, 0xC907DA4E871146AD), + make_floatx80_init(0x3FFC, 0xF7856E5EE2C9B291), + make_floatx80_init(0x3FFE, 0xC6980C6980C6980C), + make_floatx80_init(0x3FFD, 0x82012CA5A68206D7), + make_floatx80_init(0x3FFE, 0xC4372F855D824CA6), + make_floatx80_init(0x3FFD, 0x882C5FCD7256A8C5), + make_floatx80_init(0x3FFE, 0xC1E4BBD595F6E947), + make_floatx80_init(0x3FFD, 0x8E44C60B4CCFD7DE), + make_floatx80_init(0x3FFE, 0xBFA02FE80BFA02FF), + make_floatx80_init(0x3FFD, 0x944AD09EF4351AF6), + make_floatx80_init(0x3FFE, 0xBD69104707661AA3), + make_floatx80_init(0x3FFD, 0x9A3EECD4C3EAA6B2), + make_floatx80_init(0x3FFE, 0xBB3EE721A54D880C), + make_floatx80_init(0x3FFD, 0xA0218434353F1DE8), + make_floatx80_init(0x3FFE, 0xB92143FA36F5E02E), + make_floatx80_init(0x3FFD, 0xA5F2FCABBBC506DA), + make_floatx80_init(0x3FFE, 0xB70FBB5A19BE3659), + make_floatx80_init(0x3FFD, 0xABB3B8BA2AD362A5), + make_floatx80_init(0x3FFE, 0xB509E68A9B94821F), + make_floatx80_init(0x3FFD, 0xB1641795CE3CA97B), + make_floatx80_init(0x3FFE, 0xB30F63528917C80B), + make_floatx80_init(0x3FFD, 0xB70475515D0F1C61), + make_floatx80_init(0x3FFE, 0xB11FD3B80B11FD3C), + make_floatx80_init(0x3FFD, 0xBC952AFEEA3D13E1), + make_floatx80_init(0x3FFE, 0xAF3ADDC680AF3ADE), + make_floatx80_init(0x3FFD, 0xC2168ED0F458BA4A), + make_floatx80_init(0x3FFE, 0xAD602B580AD602B6), + make_floatx80_init(0x3FFD, 0xC788F439B3163BF1), + make_floatx80_init(0x3FFE, 0xAB8F69E28359CD11), + make_floatx80_init(0x3FFD, 0xCCECAC08BF04565D), + make_floatx80_init(0x3FFE, 0xA9C84A47A07F5638), + make_floatx80_init(0x3FFD, 0xD24204872DD85160), + make_floatx80_init(0x3FFE, 0xA80A80A80A80A80B), + make_floatx80_init(0x3FFD, 0xD78949923BC3588A), + make_floatx80_init(0x3FFE, 0xA655C4392D7B73A8), + make_floatx80_init(0x3FFD, 0xDCC2C4B49887DACC), + make_floatx80_init(0x3FFE, 0xA4A9CF1D96833751), + make_floatx80_init(0x3FFD, 0xE1EEBD3E6D6A6B9E), + make_floatx80_init(0x3FFE, 0xA3065E3FAE7CD0E0), + make_floatx80_init(0x3FFD, 0xE70D785C2F9F5BDC), + make_floatx80_init(0x3FFE, 0xA16B312EA8FC377D), + make_floatx80_init(0x3FFD, 0xEC1F392C5179F283), + make_floatx80_init(0x3FFE, 0x9FD809FD809FD80A), + make_floatx80_init(0x3FFD, 0xF12440D3E36130E6), + make_floatx80_init(0x3FFE, 0x9E4CAD23DD5F3A20), + make_floatx80_init(0x3FFD, 0xF61CCE92346600BB), + make_floatx80_init(0x3FFE, 0x9CC8E160C3FB19B9), + make_floatx80_init(0x3FFD, 0xFB091FD38145630A), + make_floatx80_init(0x3FFE, 0x9B4C6F9EF03A3CAA), + make_floatx80_init(0x3FFD, 0xFFE97042BFA4C2AD), + make_floatx80_init(0x3FFE, 0x99D722DABDE58F06), + make_floatx80_init(0x3FFE, 0x825EFCED49369330), + make_floatx80_init(0x3FFE, 0x9868C809868C8098), + make_floatx80_init(0x3FFE, 0x84C37A7AB9A905C9), + make_floatx80_init(0x3FFE, 0x97012E025C04B809), + make_floatx80_init(0x3FFE, 0x87224C2E8E645FB7), + make_floatx80_init(0x3FFE, 0x95A02568095A0257), + make_floatx80_init(0x3FFE, 0x897B8CAC9F7DE298), + make_floatx80_init(0x3FFE, 0x9445809445809446), + make_floatx80_init(0x3FFE, 0x8BCF55DEC4CD05FE), + make_floatx80_init(0x3FFE, 0x92F113840497889C), + make_floatx80_init(0x3FFE, 0x8E1DC0FB89E125E5), + make_floatx80_init(0x3FFE, 0x91A2B3C4D5E6F809), + make_floatx80_init(0x3FFE, 0x9066E68C955B6C9B), + make_floatx80_init(0x3FFE, 0x905A38633E06C43B), + make_floatx80_init(0x3FFE, 0x92AADE74C7BE59E0), + make_floatx80_init(0x3FFE, 0x8F1779D9FDC3A219), + make_floatx80_init(0x3FFE, 0x94E9BFF615845643), + make_floatx80_init(0x3FFE, 0x8DDA520237694809), + make_floatx80_init(0x3FFE, 0x9723A1B720134203), + make_floatx80_init(0x3FFE, 0x8CA29C046514E023), + make_floatx80_init(0x3FFE, 0x995899C890EB8990), + make_floatx80_init(0x3FFE, 0x8B70344A139BC75A), + make_floatx80_init(0x3FFE, 0x9B88BDAA3A3DAE2F), + make_floatx80_init(0x3FFE, 0x8A42F8705669DB46), + make_floatx80_init(0x3FFE, 0x9DB4224FFFE1157C), + make_floatx80_init(0x3FFE, 0x891AC73AE9819B50), + make_floatx80_init(0x3FFE, 0x9FDADC268B7A12DA), + make_floatx80_init(0x3FFE, 0x87F78087F78087F8), + make_floatx80_init(0x3FFE, 0xA1FCFF17CE733BD4), + make_floatx80_init(0x3FFE, 0x86D905447A34ACC6), + make_floatx80_init(0x3FFE, 0xA41A9E8F5446FB9F), + make_floatx80_init(0x3FFE, 0x85BF37612CEE3C9B), + make_floatx80_init(0x3FFE, 0xA633CD7E6771CD8B), + make_floatx80_init(0x3FFE, 0x84A9F9C8084A9F9D), + make_floatx80_init(0x3FFE, 0xA8489E600B435A5E), + make_floatx80_init(0x3FFE, 0x839930523FBE3368), + make_floatx80_init(0x3FFE, 0xAA59233CCCA4BD49), + make_floatx80_init(0x3FFE, 0x828CBFBEB9A020A3), + make_floatx80_init(0x3FFE, 0xAC656DAE6BCC4985), + make_floatx80_init(0x3FFE, 0x81848DA8FAF0D277), + make_floatx80_init(0x3FFE, 0xAE6D8EE360BB2468), + make_floatx80_init(0x3FFE, 0x8080808080808081), + make_floatx80_init(0x3FFE, 0xB07197A23C46C654) +}; + +static const floatx80 exp_tbl[64] = { + make_floatx80_init(0x3FFF, 0x8000000000000000), + make_floatx80_init(0x3FFF, 0x8164D1F3BC030774), + make_floatx80_init(0x3FFF, 0x82CD8698AC2BA1D8), + make_floatx80_init(0x3FFF, 0x843A28C3ACDE4048), + make_floatx80_init(0x3FFF, 0x85AAC367CC487B14), + make_floatx80_init(0x3FFF, 0x871F61969E8D1010), + make_floatx80_init(0x3FFF, 0x88980E8092DA8528), + make_floatx80_init(0x3FFF, 0x8A14D575496EFD9C), + make_floatx80_init(0x3FFF, 0x8B95C1E3EA8BD6E8), + make_floatx80_init(0x3FFF, 0x8D1ADF5B7E5BA9E4), + make_floatx80_init(0x3FFF, 0x8EA4398B45CD53C0), + make_floatx80_init(0x3FFF, 0x9031DC431466B1DC), + make_floatx80_init(0x3FFF, 0x91C3D373AB11C338), + make_floatx80_init(0x3FFF, 0x935A2B2F13E6E92C), + make_floatx80_init(0x3FFF, 0x94F4EFA8FEF70960), + make_floatx80_init(0x3FFF, 0x96942D3720185A00), + make_floatx80_init(0x3FFF, 0x9837F0518DB8A970), + make_floatx80_init(0x3FFF, 0x99E0459320B7FA64), + make_floatx80_init(0x3FFF, 0x9B8D39B9D54E5538), + make_floatx80_init(0x3FFF, 0x9D3ED9A72CFFB750), + make_floatx80_init(0x3FFF, 0x9EF5326091A111AC), + make_floatx80_init(0x3FFF, 0xA0B0510FB9714FC4), + make_floatx80_init(0x3FFF, 0xA27043030C496818), + make_floatx80_init(0x3FFF, 0xA43515AE09E680A0), + make_floatx80_init(0x3FFF, 0xA5FED6A9B15138EC), + make_floatx80_init(0x3FFF, 0xA7CD93B4E9653568), + make_floatx80_init(0x3FFF, 0xA9A15AB4EA7C0EF8), + make_floatx80_init(0x3FFF, 0xAB7A39B5A93ED338), + make_floatx80_init(0x3FFF, 0xAD583EEA42A14AC8), + make_floatx80_init(0x3FFF, 0xAF3B78AD690A4374), + make_floatx80_init(0x3FFF, 0xB123F581D2AC2590), + make_floatx80_init(0x3FFF, 0xB311C412A9112488), + make_floatx80_init(0x3FFF, 0xB504F333F9DE6484), + make_floatx80_init(0x3FFF, 0xB6FD91E328D17790), + make_floatx80_init(0x3FFF, 0xB8FBAF4762FB9EE8), + make_floatx80_init(0x3FFF, 0xBAFF5AB2133E45FC), + make_floatx80_init(0x3FFF, 0xBD08A39F580C36C0), + make_floatx80_init(0x3FFF, 0xBF1799B67A731084), + make_floatx80_init(0x3FFF, 0xC12C4CCA66709458), + make_floatx80_init(0x3FFF, 0xC346CCDA24976408), + make_floatx80_init(0x3FFF, 0xC5672A115506DADC), + make_floatx80_init(0x3FFF, 0xC78D74C8ABB9B15C), + make_floatx80_init(0x3FFF, 0xC9B9BD866E2F27A4), + make_floatx80_init(0x3FFF, 0xCBEC14FEF2727C5C), + make_floatx80_init(0x3FFF, 0xCE248C151F8480E4), + make_floatx80_init(0x3FFF, 0xD06333DAEF2B2594), + make_floatx80_init(0x3FFF, 0xD2A81D91F12AE45C), + make_floatx80_init(0x3FFF, 0xD4F35AABCFEDFA20), + make_floatx80_init(0x3FFF, 0xD744FCCAD69D6AF4), + make_floatx80_init(0x3FFF, 0xD99D15C278AFD7B4), + make_floatx80_init(0x3FFF, 0xDBFBB797DAF23754), + make_floatx80_init(0x3FFF, 0xDE60F4825E0E9124), + make_floatx80_init(0x3FFF, 0xE0CCDEEC2A94E110), + make_floatx80_init(0x3FFF, 0xE33F8972BE8A5A50), + make_floatx80_init(0x3FFF, 0xE5B906E77C8348A8), + make_floatx80_init(0x3FFF, 0xE8396A503C4BDC68), + make_floatx80_init(0x3FFF, 0xEAC0C6E7DD243930), + make_floatx80_init(0x3FFF, 0xED4F301ED9942B84), + make_floatx80_init(0x3FFF, 0xEFE4B99BDCDAF5CC), + make_floatx80_init(0x3FFF, 0xF281773C59FFB138), + make_floatx80_init(0x3FFF, 0xF5257D152486CC2C), + make_floatx80_init(0x3FFF, 0xF7D0DF730AD13BB8), + make_floatx80_init(0x3FFF, 0xFA83B2DB722A033C), + make_floatx80_init(0x3FFF, 0xFD3E0C0CF486C174) +}; + +static const float32 exp_tbl2[64] = { + const_float32(0x00000000), + const_float32(0x9F841A9B), + const_float32(0x9FC1D5B9), + const_float32(0xA0728369), + const_float32(0x1FC5C95C), + const_float32(0x1EE85C9F), + const_float32(0x9FA20729), + const_float32(0xA07BF9AF), + const_float32(0xA0020DCF), + const_float32(0x205A63DA), + const_float32(0x1EB70051), + const_float32(0x1F6EB029), + const_float32(0xA0781494), + const_float32(0x9EB319B0), + const_float32(0x2017457D), + const_float32(0x1F11D537), + const_float32(0x9FB952DD), + const_float32(0x1FE43087), + const_float32(0x1FA2A818), + const_float32(0x1FDE494D), + const_float32(0x20504890), + const_float32(0xA073691C), + const_float32(0x1F9B7A05), + const_float32(0xA0797126), + const_float32(0xA071A140), + const_float32(0x204F62DA), + const_float32(0x1F283C4A), + const_float32(0x9F9A7FDC), + const_float32(0xA05B3FAC), + const_float32(0x1FDF2610), + const_float32(0x9F705F90), + const_float32(0x201F678A), + const_float32(0x1F32FB13), + const_float32(0x20038B30), + const_float32(0x200DC3CC), + const_float32(0x9F8B2AE6), + const_float32(0xA02BBF70), + const_float32(0xA00BF518), + const_float32(0xA041DD41), + const_float32(0x9FDF137B), + const_float32(0x201F1568), + const_float32(0x1FC13A2E), + const_float32(0xA03F8F03), + const_float32(0x1FF4907D), + const_float32(0x9E6E53E4), + const_float32(0x1FD6D45C), + const_float32(0xA076EDB9), + const_float32(0x9FA6DE21), + const_float32(0x1EE69A2F), + const_float32(0x207F439F), + const_float32(0x201EC207), + const_float32(0x9E8BE175), + const_float32(0x20032C4B), + const_float32(0x2004DFF5), + const_float32(0x1E72F47A), + const_float32(0x1F722F22), + const_float32(0xA017E945), + const_float32(0x1F401A5B), + const_float32(0x9FB9A9E3), + const_float32(0x20744C05), + const_float32(0x1F773A19), + const_float32(0x1FFE90D5), + const_float32(0xA041ED22), + const_float32(0x1F853F3A), +}; + +static const floatx80 exp2_tbl[64] = { + make_floatx80_init(0x3FFF, 0x8000000000000000), + make_floatx80_init(0x3FFF, 0x8164D1F3BC030773), + make_floatx80_init(0x3FFF, 0x82CD8698AC2BA1D7), + make_floatx80_init(0x3FFF, 0x843A28C3ACDE4046), + make_floatx80_init(0x3FFF, 0x85AAC367CC487B15), + make_floatx80_init(0x3FFF, 0x871F61969E8D1010), + make_floatx80_init(0x3FFF, 0x88980E8092DA8527), + make_floatx80_init(0x3FFF, 0x8A14D575496EFD9A), + make_floatx80_init(0x3FFF, 0x8B95C1E3EA8BD6E7), + make_floatx80_init(0x3FFF, 0x8D1ADF5B7E5BA9E6), + make_floatx80_init(0x3FFF, 0x8EA4398B45CD53C0), + make_floatx80_init(0x3FFF, 0x9031DC431466B1DC), + make_floatx80_init(0x3FFF, 0x91C3D373AB11C336), + make_floatx80_init(0x3FFF, 0x935A2B2F13E6E92C), + make_floatx80_init(0x3FFF, 0x94F4EFA8FEF70961), + make_floatx80_init(0x3FFF, 0x96942D3720185A00), + make_floatx80_init(0x3FFF, 0x9837F0518DB8A96F), + make_floatx80_init(0x3FFF, 0x99E0459320B7FA65), + make_floatx80_init(0x3FFF, 0x9B8D39B9D54E5539), + make_floatx80_init(0x3FFF, 0x9D3ED9A72CFFB751), + make_floatx80_init(0x3FFF, 0x9EF5326091A111AE), + make_floatx80_init(0x3FFF, 0xA0B0510FB9714FC2), + make_floatx80_init(0x3FFF, 0xA27043030C496819), + make_floatx80_init(0x3FFF, 0xA43515AE09E6809E), + make_floatx80_init(0x3FFF, 0xA5FED6A9B15138EA), + make_floatx80_init(0x3FFF, 0xA7CD93B4E965356A), + make_floatx80_init(0x3FFF, 0xA9A15AB4EA7C0EF8), + make_floatx80_init(0x3FFF, 0xAB7A39B5A93ED337), + make_floatx80_init(0x3FFF, 0xAD583EEA42A14AC6), + make_floatx80_init(0x3FFF, 0xAF3B78AD690A4375), + make_floatx80_init(0x3FFF, 0xB123F581D2AC2590), + make_floatx80_init(0x3FFF, 0xB311C412A9112489), + make_floatx80_init(0x3FFF, 0xB504F333F9DE6484), + make_floatx80_init(0x3FFF, 0xB6FD91E328D17791), + make_floatx80_init(0x3FFF, 0xB8FBAF4762FB9EE9), + make_floatx80_init(0x3FFF, 0xBAFF5AB2133E45FB), + make_floatx80_init(0x3FFF, 0xBD08A39F580C36BF), + make_floatx80_init(0x3FFF, 0xBF1799B67A731083), + make_floatx80_init(0x3FFF, 0xC12C4CCA66709456), + make_floatx80_init(0x3FFF, 0xC346CCDA24976407), + make_floatx80_init(0x3FFF, 0xC5672A115506DADD), + make_floatx80_init(0x3FFF, 0xC78D74C8ABB9B15D), + make_floatx80_init(0x3FFF, 0xC9B9BD866E2F27A3), + make_floatx80_init(0x3FFF, 0xCBEC14FEF2727C5D), + make_floatx80_init(0x3FFF, 0xCE248C151F8480E4), + make_floatx80_init(0x3FFF, 0xD06333DAEF2B2595), + make_floatx80_init(0x3FFF, 0xD2A81D91F12AE45A), + make_floatx80_init(0x3FFF, 0xD4F35AABCFEDFA1F), + make_floatx80_init(0x3FFF, 0xD744FCCAD69D6AF4), + make_floatx80_init(0x3FFF, 0xD99D15C278AFD7B6), + make_floatx80_init(0x3FFF, 0xDBFBB797DAF23755), + make_floatx80_init(0x3FFF, 0xDE60F4825E0E9124), + make_floatx80_init(0x3FFF, 0xE0CCDEEC2A94E111), + make_floatx80_init(0x3FFF, 0xE33F8972BE8A5A51), + make_floatx80_init(0x3FFF, 0xE5B906E77C8348A8), + make_floatx80_init(0x3FFF, 0xE8396A503C4BDC68), + make_floatx80_init(0x3FFF, 0xEAC0C6E7DD24392F), + make_floatx80_init(0x3FFF, 0xED4F301ED9942B84), + make_floatx80_init(0x3FFF, 0xEFE4B99BDCDAF5CB), + make_floatx80_init(0x3FFF, 0xF281773C59FFB13A), + make_floatx80_init(0x3FFF, 0xF5257D152486CC2C), + make_floatx80_init(0x3FFF, 0xF7D0DF730AD13BB9), + make_floatx80_init(0x3FFF, 0xFA83B2DB722A033A), + make_floatx80_init(0x3FFF, 0xFD3E0C0CF486C175) +}; + +static const uint32_t exp2_tbl2[64] = { + 0x3F738000, 0x3FBEF7CA, 0x3FBDF8A9, 0x3FBCD7C9, + 0xBFBDE8DA, 0x3FBDE85C, 0x3FBEBBF1, 0x3FBB80CA, + 0xBFBA8373, 0xBFBE9670, 0x3FBDB700, 0x3FBEEEB0, + 0x3FBBFD6D, 0xBFBDB319, 0x3FBDBA2B, 0x3FBE91D5, + 0x3FBE8D5A, 0xBFBCDE7B, 0xBFBEBAAF, 0xBFBD86DA, + 0xBFBEBEDD, 0x3FBCC96E, 0xBFBEC90B, 0x3FBBD1DB, + 0x3FBCE5EB, 0xBFBEC274, 0x3FBEA83C, 0x3FBECB00, + 0x3FBE9301, 0xBFBD8367, 0xBFBEF05F, 0x3FBDFB3C, + 0x3FBEB2FB, 0x3FBAE2CB, 0x3FBCDC3C, 0x3FBEE9AA, + 0xBFBEAEFD, 0xBFBCBF51, 0x3FBEF88A, 0x3FBD83B2, + 0x3FBDF8AB, 0xBFBDFB17, 0xBFBEFE3C, 0xBFBBB6F8, + 0xBFBCEE53, 0xBFBDA4AE, 0x3FBC9124, 0x3FBEB243, + 0x3FBDE69A, 0xBFB8BC61, 0x3FBDF610, 0xBFBD8BE1, + 0x3FBACB12, 0x3FBB9BFE, 0x3FBCF2F4, 0x3FBEF22F, + 0xBFBDBF4A, 0x3FBEC01A, 0x3FBE8CAC, 0xBFBCBB3F, + 0x3FBEF73A, 0xBFB8B795, 0x3FBEF84B, 0xBFBEF581 +}; +#endif diff --git a/target/m68k/translate.c b/target/m68k/translate.c index dbb24f8d84..6d5bde0777 100644 --- a/target/m68k/translate.c +++ b/target/m68k/translate.c @@ -5054,6 +5054,27 @@ DISAS_INSN(fpu) case 0x45: /* fdsqrt */ gen_helper_fdsqrt(cpu_env, cpu_dest, cpu_src); break; + case 0x06: /* flognp1 */ + gen_helper_flognp1(cpu_env, cpu_dest, cpu_src); + break; + case 0x10: /* fetox */ + gen_helper_fetox(cpu_env, cpu_dest, cpu_src); + break; + case 0x11: /* ftwotox */ + gen_helper_ftwotox(cpu_env, cpu_dest, cpu_src); + break; + case 0x12: /* ftentox */ + gen_helper_ftentox(cpu_env, cpu_dest, cpu_src); + break; + case 0x14: /* flogn */ + gen_helper_flogn(cpu_env, cpu_dest, cpu_src); + break; + case 0x15: /* flog10 */ + gen_helper_flog10(cpu_env, cpu_dest, cpu_src); + break; + case 0x16: /* flog2 */ + gen_helper_flog2(cpu_env, cpu_dest, cpu_src); + break; case 0x18: /* fabs */ gen_helper_fabs(cpu_env, cpu_dest, cpu_src); break; diff --git a/util/aio-wait.c b/util/aio-wait.c index a487cdb852..975afddf4c 100644 --- a/util/aio-wait.c +++ b/util/aio-wait.c @@ -38,3 +38,34 @@ void aio_wait_kick(AioWait *wait) aio_bh_schedule_oneshot(qemu_get_aio_context(), dummy_bh_cb, NULL); } } + +typedef struct { + AioWait wait; + bool done; + QEMUBHFunc *cb; + void *opaque; +} AioWaitBHData; + +/* Context: BH in IOThread */ +static void aio_wait_bh(void *opaque) +{ + AioWaitBHData *data = opaque; + + data->cb(data->opaque); + + data->done = true; + aio_wait_kick(&data->wait); +} + +void aio_wait_bh_oneshot(AioContext *ctx, QEMUBHFunc *cb, void *opaque) +{ + AioWaitBHData data = { + .cb = cb, + .opaque = opaque, + }; + + assert(qemu_get_current_aio_context() == qemu_get_aio_context()); + + aio_bh_schedule_oneshot(ctx, aio_wait_bh, &data); + AIO_WAIT_WHILE(&data.wait, ctx, !data.done); +} @@ -4722,17 +4722,10 @@ int main(int argc, char **argv, char **envp) os_setup_post(); main_loop(); - replay_disable_events(); - - /* The ordering of the following is delicate. Stop vcpus to prevent new - * I/O requests being queued by the guest. Then stop IOThreads (this - * includes a drain operation and completes all request processing). At - * this point emulated devices are still associated with their IOThreads - * (if any) but no longer have any work to do. Only then can we close - * block devices safely because we know there is no more I/O coming. - */ - pause_all_vcpus(); - iothread_stop_all(); + + /* No more vcpu or device emulation activity beyond this point */ + vm_shutdown(); + bdrv_close_all(); res_free(); |