[PATCH AUTOSEL 5.10 67/93] habanalabs/gaudi: set the correct cpu_id on MME2_QM failure

From: Sasha Levin
Date: Fri Jul 09 2021 - 22:31:10 EST


From: Koby Elbaz <kelbaz@xxxxxxxxx>

[ Upstream commit b92c637c5f5ef7e3e21dbc7bfa7f1999450f3902 ]

This fix was applied since there was an incorrect reported CPU ID to GIC
such that an error in MME2 QMAN aliased to be an arriving from DMA0_QM.

Signed-off-by: Koby Elbaz <kelbaz@xxxxxxxxx>
Reviewed-by: Oded Gabbay <ogabbay@xxxxxxxxxx>
Signed-off-by: Oded Gabbay <ogabbay@xxxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
drivers/misc/habanalabs/gaudi/gaudi.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/misc/habanalabs/gaudi/gaudi.c b/drivers/misc/habanalabs/gaudi/gaudi.c
index 68f661aca3ff..044b2ae196f9 100644
--- a/drivers/misc/habanalabs/gaudi/gaudi.c
+++ b/drivers/misc/habanalabs/gaudi/gaudi.c
@@ -2164,7 +2164,7 @@ static void gaudi_init_mme_qman(struct hl_device *hdev, u32 mme_offset,

/* Configure RAZWI IRQ */
mme_id = mme_offset /
- (mmMME1_QM_GLBL_CFG0 - mmMME0_QM_GLBL_CFG0);
+ (mmMME1_QM_GLBL_CFG0 - mmMME0_QM_GLBL_CFG0) / 2;

mme_qm_err_cfg = MME_QMAN_GLBL_ERR_CFG_MSG_EN_MASK;
if (hdev->stop_on_err) {
--
2.30.2