[PATCH 2/4] drm/amdkfd: replace source_id with client_id for RAS poison consumption

Zhang, Hawking Hawking.Zhang at amd.com
Wed Mar 16 13:52:46 UTC 2022


[AMD Official Use Only]

+       case SOC15_IH_CLIENTID_SDMA5:
+       case SOC15_IH_CLIENTID_SDMA6:
+       case SOC15_IH_CLIENTID_SDMA7:

Please drop the unnecessary case here since the feature is not available on all the ASICs.

+       case SOC15_IH_CLIENTID_GRBM_CP:
+       case SOC15_IH_CLIENTID_SE0SH:
+       case SOC15_IH_CLIENTID_SE1SH:
+       case SOC15_IH_CLIENTID_SE2SH:
+       case SOC15_IH_CLIENTID_SE3SH:

And I also think we can reduce the cases for SQ interrupt as well

With above addressed, the patch is

Reviewed-by: Hawking Zhang <Hawking.Zhang at amd.com>

Regards,
Hawking

-----Original Message-----
From: Zhou1, Tao <Tao.Zhou1 at amd.com>
Sent: Wednesday, March 16, 2022 17:26
To: amd-gfx at lists.freedesktop.org; Zhang, Hawking <Hawking.Zhang at amd.com>; Kuehling, Felix <Felix.Kuehling at amd.com>; Yang, Stanley <Stanley.Yang at amd.com>; Chai, Thomas <YiPeng.Chai at amd.com>
Cc: Zhou1, Tao <Tao.Zhou1 at amd.com>
Subject: [PATCH 2/4] drm/amdkfd: replace source_id with client_id for RAS poison consumption

client_id is more accruate here and we can deal with more different cases.

Signed-off-by: Tao Zhou <tao.zhou1 at amd.com>
---
 .../gpu/drm/amd/amdkfd/kfd_int_process_v9.c   | 25 ++++++++++++++-----
 1 file changed, 19 insertions(+), 6 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 a992798ff8b6..e5f03f79546f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
@@ -91,7 +91,7 @@ enum SQ_INTERRUPT_ERROR_TYPE {  #define KFD_SQ_INT_DATA__ERR_TYPE__SHIFT 20

 static void event_interrupt_poison_consumption(struct kfd_dev *dev,
-                               uint16_t pasid, uint16_t source_id)
+                               uint16_t pasid, uint16_t client_id)
 {
        int old_poison, ret = -EINVAL;
        struct kfd_process *p = kfd_lookup_process_by_pasid(pasid);
@@ -107,11 +107,24 @@ static void event_interrupt_poison_consumption(struct kfd_dev *dev,

        pr_warn("RAS poison consumption handling\n");

-       switch (source_id) {
-       case SOC15_INTSRC_SQ_INTERRUPT_MSG:
+       switch (client_id) {
+       case SOC15_IH_CLIENTID_GRBM_CP:
+       case SOC15_IH_CLIENTID_SE0SH:
+       case SOC15_IH_CLIENTID_SE1SH:
+       case SOC15_IH_CLIENTID_SE2SH:
+       case SOC15_IH_CLIENTID_SE3SH:
+       case SOC15_IH_CLIENTID_UTCL2:
                ret = kfd_dqm_evict_pasid(dev->dqm, pasid);
                break;
-       case SOC15_INTSRC_SDMA_ECC:
+       case SOC15_IH_CLIENTID_SDMA0:
+       case SOC15_IH_CLIENTID_SDMA1:
+       case SOC15_IH_CLIENTID_SDMA2:
+       case SOC15_IH_CLIENTID_SDMA3:
+       case SOC15_IH_CLIENTID_SDMA4:
+       case SOC15_IH_CLIENTID_SDMA5:
+       case SOC15_IH_CLIENTID_SDMA6:
+       case SOC15_IH_CLIENTID_SDMA7:
+               break;
        default:
                break;
        }
@@ -269,7 +282,7 @@ 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, source_id);
+                                       event_interrupt_poison_consumption(dev, pasid, client_id);
                                        return;
                                }
                                break;
@@ -290,7 +303,7 @@ 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, source_id);
+                       event_interrupt_poison_consumption(dev, pasid, client_id);
                        return;
                }
        } else if (client_id == SOC15_IH_CLIENTID_VMC ||
--
2.35.1



More information about the amd-gfx mailing list