If a TCG guest reboots during a running migration HTAB entries are not marked dirty, and the destination boots with an invalid HTAB.
When a reboot occurs reset the state of HTAB migration, and explicitly inform the destination of invalid entries. Signed-off-by: Samuel Mendoza-Jonas <sam...@au1.ibm.com> --- hw/ppc/spapr.c | 59 +++++++++++++++++++++++++++++++++++--------------- include/hw/ppc/spapr.h | 1 + 2 files changed, 42 insertions(+), 18 deletions(-) diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c index 1610c28..9f419e8 100644 --- a/hw/ppc/spapr.c +++ b/hw/ppc/spapr.c @@ -829,26 +829,30 @@ static void spapr_reset_htab(sPAPREnvironment *spapr) shift = kvmppc_reset_htab(spapr->htab_shift); + pthread_mutex_lock(&spapr->htab_mutex); if (shift > 0) { /* Kernel handles htab, we don't need to allocate one */ spapr->htab_shift = shift; kvmppc_kern_htab = true; /* Tell readers to update their file descriptor */ - pthread_mutex_lock(&spapr->htab_mutex); if (spapr->htab_fd > 0) { spapr->htab_fd_stale = true; } - pthread_mutex_unlock(&spapr->htab_mutex); } else { if (!spapr->htab) { /* Allocate an htab if we don't yet have one */ spapr->htab = qemu_memalign(HTAB_SIZE(spapr), HTAB_SIZE(spapr)); + } else { + spapr->htab_mig_full = true; + spapr->htab_first_pass = true; + spapr->htab_save_index = 0; } /* And clear it */ memset(spapr->htab, 0, HTAB_SIZE(spapr)); } + pthread_mutex_unlock(&spapr->htab_mutex); /* Update the RMA size if necessary */ if (spapr->vrma_adjust) { @@ -1019,6 +1023,7 @@ static int htab_save_setup(QEMUFile *f, void *opaque) pthread_mutex_lock(&spapr->htab_mutex); spapr->htab_fd = kvmppc_get_htab_fd(false); spapr->htab_fd_stale = false; + spapr->htab_mig_full = false; pthread_mutex_unlock(&spapr->htab_mutex); if (spapr->htab_fd < 0) { fprintf(stderr, "Unable to open fd for reading hash table from KVM: %s\n", @@ -1034,6 +1039,7 @@ static int htab_save_setup(QEMUFile *f, void *opaque) static void htab_save_first_pass(QEMUFile *f, sPAPREnvironment *spapr, int64_t max_ns) { + bool final = max_ns < 0; int htabslots = HTAB_SIZE(spapr) / HASH_PTE_SIZE_64; int index = spapr->htab_save_index; int64_t starttime = qemu_clock_get_ns(QEMU_CLOCK_REALTIME); @@ -1041,33 +1047,40 @@ static void htab_save_first_pass(QEMUFile *f, sPAPREnvironment *spapr, assert(spapr->htab_first_pass); do { - int chunkstart; + int chunkstart, invalidstart; - /* Consume invalid HPTEs */ - while ((index < htabslots) - && !HPTE_VALID(HPTE(spapr->htab, index))) { + chunkstart = index; + /* Consume valid HPTEs */ + while ((index < htabslots && (index - chunkstart < USHRT_MAX)) + && HPTE_VALID(HPTE(spapr->htab, index))) { index++; CLEAN_HPTE(HPTE(spapr->htab, index)); } - /* Consume valid HPTEs */ - chunkstart = index; - while ((index < htabslots) - && HPTE_VALID(HPTE(spapr->htab, index))) { + invalidstart = index; + /* Consume invalid HPTEs */ + while ((index < htabslots && (index - invalidstart < USHRT_MAX)) + && !HPTE_VALID(HPTE(spapr->htab, index))) { index++; CLEAN_HPTE(HPTE(spapr->htab, index)); } - if (index > chunkstart) { - int n_valid = index - chunkstart; + /* Avoid writing an end marker (0,0,0) */ + if (index > chunkstart + && !(chunkstart == invalidstart && !spapr->htab_mig_full)) { + int n_valid = invalidstart - chunkstart; + /* If a reset has occured we must explicitly overwrite the HTAB + * of the destination */ + int n_invalid = spapr->htab_mig_full ? index - invalidstart : 0; qemu_put_be32(f, chunkstart); qemu_put_be16(f, n_valid); - qemu_put_be16(f, 0); + qemu_put_be16(f, n_invalid); qemu_put_buffer(f, HPTE(spapr->htab, chunkstart), HASH_PTE_SIZE_64 * n_valid); - if ((qemu_clock_get_ns(QEMU_CLOCK_REALTIME) - starttime) > max_ns) { + if ((qemu_clock_get_ns(QEMU_CLOCK_REALTIME) - starttime) > max_ns + && !final) { break; } } @@ -1182,10 +1195,14 @@ static int htab_save_iterate(QEMUFile *f, void *opaque) if (rc < 0) { return rc; } - } else if (spapr->htab_first_pass) { - htab_save_first_pass(f, spapr, MAX_ITERATION_NS); } else { - rc = htab_save_later_pass(f, spapr, MAX_ITERATION_NS); + pthread_mutex_lock(&spapr->htab_mutex); + if (spapr->htab_first_pass) { + htab_save_first_pass(f, spapr, MAX_ITERATION_NS); + } else { + rc = htab_save_later_pass(f, spapr, MAX_ITERATION_NS); + } + pthread_mutex_unlock(&spapr->htab_mutex); } /* End marker */ @@ -1220,7 +1237,13 @@ static int htab_save_complete(QEMUFile *f, void *opaque) close(spapr->htab_fd); spapr->htab_fd = -1; } else { - htab_save_later_pass(f, spapr, -1); + pthread_mutex_lock(&spapr->htab_mutex); + if (spapr->htab_first_pass) { + htab_save_first_pass(f, spapr, -1); + } else { + htab_save_later_pass(f, spapr, -1); + } + pthread_mutex_unlock(&spapr->htab_mutex); } /* End marker */ diff --git a/include/hw/ppc/spapr.h b/include/hw/ppc/spapr.h index 5e29bec..ee95459 100644 --- a/include/hw/ppc/spapr.h +++ b/include/hw/ppc/spapr.h @@ -38,6 +38,7 @@ typedef struct sPAPREnvironment { bool htab_first_pass; int htab_fd; bool htab_fd_stale; + bool htab_mig_full; pthread_mutex_t htab_mutex; } sPAPREnvironment; -- 1.9.3