diff options
Diffstat (limited to 'drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c')
| -rw-r--r-- | drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c | 41 |
1 files changed, 37 insertions, 4 deletions
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c index 03c29bdd89a1..0b75a37b689b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c @@ -90,7 +90,7 @@ enum SQ_INTERRUPT_ERROR_TYPE { #define KFD_SQ_INT_DATA__ERR_TYPE_MASK 0xF00000 #define KFD_SQ_INT_DATA__ERR_TYPE__SHIFT 20 -static void event_interrupt_poison_consumption(struct kfd_dev *dev, +static void event_interrupt_poison_consumption_v9(struct kfd_dev *dev, uint16_t pasid, uint16_t client_id) { int old_poison, ret = -EINVAL; @@ -141,6 +141,25 @@ static void event_interrupt_poison_consumption(struct kfd_dev *dev, } } +static bool context_id_expected(struct kfd_dev *dev) +{ + switch (KFD_GC_VERSION(dev)) { + case IP_VERSION(9, 0, 1): + return dev->mec_fw_version >= 0x817a; + case IP_VERSION(9, 1, 0): + case IP_VERSION(9, 2, 1): + case IP_VERSION(9, 2, 2): + case IP_VERSION(9, 3, 0): + case IP_VERSION(9, 4, 0): + return dev->mec_fw_version >= 0x17a; + default: + /* Other GFXv9 and later GPUs always sent valid context IDs + * on legitimate events + */ + return KFD_GC_VERSION(dev) >= IP_VERSION(9, 4, 1); + } +} + static bool event_interrupt_isr_v9(struct kfd_dev *dev, const uint32_t *ih_ring_entry, uint32_t *patched_ihre, @@ -206,6 +225,20 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev, if (WARN_ONCE(pasid == 0, "Bug: No PASID in KFD interrupt")) return false; + /* Workaround CP firmware sending bogus signals with 0 context_id. + * Those can be safely ignored on hardware and firmware versions that + * include a valid context_id on legitimate signals. This avoids the + * slow path in kfd_signal_event_interrupt that scans all event slots + * for signaled events. + */ + if (source_id == SOC15_INTSRC_CP_END_OF_PIPE) { + uint32_t context_id = + SOC15_CONTEXT_ID0_FROM_IH_ENTRY(ih_ring_entry); + + if (context_id == 0 && context_id_expected(dev)) + return false; + } + /* Interrupt types we care about: various signals and faults. * They will be forwarded to a work queue (see below). */ @@ -283,7 +316,7 @@ static void event_interrupt_wq_v9(struct kfd_dev *dev, sq_intr_err); if (sq_intr_err != SQ_INTERRUPT_ERROR_TYPE_ILLEGAL_INST && sq_intr_err != SQ_INTERRUPT_ERROR_TYPE_MEMVIOL) { - event_interrupt_poison_consumption(dev, pasid, client_id); + event_interrupt_poison_consumption_v9(dev, pasid, client_id); return; } break; @@ -304,7 +337,7 @@ static void event_interrupt_wq_v9(struct kfd_dev *dev, if (source_id == SOC15_INTSRC_SDMA_TRAP) { kfd_signal_event_interrupt(pasid, context_id0 & 0xfffffff, 28); } else if (source_id == SOC15_INTSRC_SDMA_ECC) { - event_interrupt_poison_consumption(dev, pasid, client_id); + event_interrupt_poison_consumption_v9(dev, pasid, client_id); return; } } else if (client_id == SOC15_IH_CLIENTID_VMC || @@ -315,7 +348,7 @@ static void event_interrupt_wq_v9(struct kfd_dev *dev, if (client_id == SOC15_IH_CLIENTID_UTCL2 && amdgpu_amdkfd_ras_query_utcl2_poison_status(dev->adev)) { - event_interrupt_poison_consumption(dev, pasid, client_id); + event_interrupt_poison_consumption_v9(dev, pasid, client_id); return; } |