summaryrefslogtreecommitdiff
path: root/drivers/misc
diff options
context:
space:
mode:
authorDani Liberman <dliberman@habana.ai>2022-10-31 23:04:14 +0200
committerOded Gabbay <ogabbay@kernel.org>2022-11-23 16:13:47 +0200
commitd3027f4a625063c18becd6953b4a2a273033b071 (patch)
treecbefd082408702e73b03e5254039ba358bfa27be /drivers/misc
parenta63de89bee7ff01dc184fbe289eade5b5ab5f49a (diff)
habanalabs/gaudi2: add page fault notify event
Each time page fault happens, besides capturing its data, also notify the user about it. Signed-off-by: Dani Liberman <dliberman@habana.ai> Reviewed-by: Oded Gabbay <ogabbay@kernel.org> Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Diffstat (limited to 'drivers/misc')
-rw-r--r--drivers/misc/habanalabs/gaudi2/gaudi2.c17
1 files changed, 9 insertions, 8 deletions
diff --git a/drivers/misc/habanalabs/gaudi2/gaudi2.c b/drivers/misc/habanalabs/gaudi2/gaudi2.c
index 61960fa059e0..65c9b535aa69 100644
--- a/drivers/misc/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/misc/habanalabs/gaudi2/gaudi2.c
@@ -8253,7 +8253,8 @@ static void gaudi2_handle_hif_fatal(struct hl_device *hdev, u16 event_type, u64
}
}
-static void gaudi2_handle_page_error(struct hl_device *hdev, u64 mmu_base, bool is_pmmu)
+static void gaudi2_handle_page_error(struct hl_device *hdev, u64 mmu_base, bool is_pmmu,
+ u64 *event_mask)
{
u32 valid, val;
u64 addr;
@@ -8270,7 +8271,7 @@ static void gaudi2_handle_page_error(struct hl_device *hdev, u64 mmu_base, bool
dev_err_ratelimited(hdev->dev, "%s page fault on va 0x%llx\n",
is_pmmu ? "PMMU" : "HMMU", addr);
- hl_capture_page_fault(hdev, addr, 0, is_pmmu);
+ hl_handle_page_fault(hdev, addr, 0, is_pmmu, event_mask);
WREG32(mmu_base + MMU_OFFSET(mmDCORE0_HMMU0_MMU_PAGE_ERROR_CAPTURE), 0);
}
@@ -8296,7 +8297,7 @@ static void gaudi2_handle_access_error(struct hl_device *hdev, u64 mmu_base, boo
}
static void gaudi2_handle_mmu_spi_sei_generic(struct hl_device *hdev, const char *mmu_name,
- u64 mmu_base, bool is_pmmu)
+ u64 mmu_base, bool is_pmmu, u64 *event_mask)
{
u32 spi_sei_cause, interrupt_clr = 0x0;
int i;
@@ -8309,7 +8310,7 @@ static void gaudi2_handle_mmu_spi_sei_generic(struct hl_device *hdev, const char
mmu_name, gaudi2_mmu_spi_sei[i].cause);
if (i == 0)
- gaudi2_handle_page_error(hdev, mmu_base, is_pmmu);
+ gaudi2_handle_page_error(hdev, mmu_base, is_pmmu, event_mask);
else if (i == 1)
gaudi2_handle_access_error(hdev, mmu_base, is_pmmu);
@@ -8381,7 +8382,7 @@ static bool gaudi2_handle_sm_err(struct hl_device *hdev, u8 sm_index)
return reset;
}
-static void gaudi2_handle_mmu_spi_sei_err(struct hl_device *hdev, u16 event_type)
+static void gaudi2_handle_mmu_spi_sei_err(struct hl_device *hdev, u16 event_type, u64 *event_mask)
{
bool is_pmmu = false;
char desc[32];
@@ -8439,7 +8440,7 @@ static void gaudi2_handle_mmu_spi_sei_err(struct hl_device *hdev, u16 event_type
return;
}
- gaudi2_handle_mmu_spi_sei_generic(hdev, desc, mmu_base, is_pmmu);
+ gaudi2_handle_mmu_spi_sei_generic(hdev, desc, mmu_base, is_pmmu, event_mask);
}
@@ -8969,7 +8970,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
case GAUDI2_EVENT_HMMU_0_AXI_ERR_RSP ... GAUDI2_EVENT_HMMU_12_AXI_ERR_RSP:
case GAUDI2_EVENT_PMMU0_PAGE_FAULT_WR_PERM ... GAUDI2_EVENT_PMMU0_SECURITY_ERROR:
case GAUDI2_EVENT_PMMU_AXI_ERR_RSP_0:
- gaudi2_handle_mmu_spi_sei_err(hdev, event_type);
+ gaudi2_handle_mmu_spi_sei_err(hdev, event_type, &event_mask);
reset_flags |= HL_DRV_RESET_FW_FATAL_ERR;
event_mask |= HL_NOTIFIER_EVENT_USER_ENGINE_ERR;
break;
@@ -10206,7 +10207,7 @@ static void gaudi2_ack_mmu_error(struct hl_device *hdev, u64 mmu_id)
if (gaudi2_get_mmu_base(hdev, mmu_id, &mmu_base))
return;
- gaudi2_handle_page_error(hdev, mmu_base, is_pmmu);
+ gaudi2_handle_page_error(hdev, mmu_base, is_pmmu, NULL);
gaudi2_handle_access_error(hdev, mmu_base, is_pmmu);
}