* [PATCH 2/3] habanalabs: remove condition that is always true
2019-05-01 17:44 [PATCH 1/3] habanalabs: remove redundant member from parser struct Oded Gabbay
@ 2019-05-01 17:44 ` Oded Gabbay
2019-05-01 17:44 ` [PATCH 3/3] habanalabs: increase timeout if working with simulator Oded Gabbay
1 sibling, 0 replies; 3+ messages in thread
From: Oded Gabbay @ 2019-05-01 17:44 UTC (permalink / raw)
To: linux-kernel; +Cc: gregkh, Dalit Ben Zoor
From: Dalit Ben Zoor <dbenzoor@habana.ai>
After removing the parsing of the command submission
when doing memset of the device memory, goya_validate_dma_pkt_host
is never called by the kernel, so there is no need to check
context id.
Signed-off-by: Dalit Ben Zoor <dbenzoor@habana.ai>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
---
drivers/misc/habanalabs/goya/goya.c | 48 ++++++++++++++---------------
1 file changed, 23 insertions(+), 25 deletions(-)
diff --git a/drivers/misc/habanalabs/goya/goya.c b/drivers/misc/habanalabs/goya/goya.c
index ba6790f9ec6b..9bf572a2d292 100644
--- a/drivers/misc/habanalabs/goya/goya.c
+++ b/drivers/misc/habanalabs/goya/goya.c
@@ -3213,31 +3213,29 @@ static int goya_validate_dma_pkt_host(struct hl_device *hdev,
return -EFAULT;
}
- if (parser->ctx_id != HL_KERNEL_ASID_ID) {
- if (sram_addr) {
- if (!hl_mem_area_inside_range(device_memory_addr,
- le32_to_cpu(user_dma_pkt->tsize),
- hdev->asic_prop.sram_user_base_address,
- hdev->asic_prop.sram_end_address)) {
-
- dev_err(hdev->dev,
- "SRAM address 0x%llx + 0x%x is invalid\n",
- device_memory_addr,
- user_dma_pkt->tsize);
- return -EFAULT;
- }
- } else {
- if (!hl_mem_area_inside_range(device_memory_addr,
- le32_to_cpu(user_dma_pkt->tsize),
- hdev->asic_prop.dram_user_base_address,
- hdev->asic_prop.dram_end_address)) {
-
- dev_err(hdev->dev,
- "DRAM address 0x%llx + 0x%x is invalid\n",
- device_memory_addr,
- user_dma_pkt->tsize);
- return -EFAULT;
- }
+ if (sram_addr) {
+ if (!hl_mem_area_inside_range(device_memory_addr,
+ le32_to_cpu(user_dma_pkt->tsize),
+ hdev->asic_prop.sram_user_base_address,
+ hdev->asic_prop.sram_end_address)) {
+
+ dev_err(hdev->dev,
+ "SRAM address 0x%llx + 0x%x is invalid\n",
+ device_memory_addr,
+ user_dma_pkt->tsize);
+ return -EFAULT;
+ }
+ } else {
+ if (!hl_mem_area_inside_range(device_memory_addr,
+ le32_to_cpu(user_dma_pkt->tsize),
+ hdev->asic_prop.dram_user_base_address,
+ hdev->asic_prop.dram_end_address)) {
+
+ dev_err(hdev->dev,
+ "DRAM address 0x%llx + 0x%x is invalid\n",
+ device_memory_addr,
+ user_dma_pkt->tsize);
+ return -EFAULT;
}
}
--
2.17.1
^ permalink raw reply related [flat|nested] 3+ messages in thread* [PATCH 3/3] habanalabs: increase timeout if working with simulator
2019-05-01 17:44 [PATCH 1/3] habanalabs: remove redundant member from parser struct Oded Gabbay
2019-05-01 17:44 ` [PATCH 2/3] habanalabs: remove condition that is always true Oded Gabbay
@ 2019-05-01 17:44 ` Oded Gabbay
1 sibling, 0 replies; 3+ messages in thread
From: Oded Gabbay @ 2019-05-01 17:44 UTC (permalink / raw)
To: linux-kernel; +Cc: gregkh, Dalit Ben Zoor
From: Dalit Ben Zoor <dbenzoor@habana.ai>
Where there is a spike in the CPU consumption, it may cause
random failures in the C/I since the KMD timeout for CPU
and/or QMAN0 jobs expires and it stops communicating to the simulator.
This commit fixes it by increasing timeout on polling functions
if working with simulator.
Signed-off-by: Dalit Ben Zoor <dbenzoor@habana.ai>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
---
drivers/misc/habanalabs/device.c | 8 +++++++-
drivers/misc/habanalabs/habanalabs.h | 7 ++++++-
2 files changed, 13 insertions(+), 2 deletions(-)
diff --git a/drivers/misc/habanalabs/device.c b/drivers/misc/habanalabs/device.c
index 0e0b9ec71c80..91a9e47a3482 100644
--- a/drivers/misc/habanalabs/device.c
+++ b/drivers/misc/habanalabs/device.c
@@ -1147,7 +1147,13 @@ int hl_poll_timeout_memory(struct hl_device *hdev, u64 addr,
* either by the direct access of the device or by another core
*/
u32 *paddr = (u32 *) (uintptr_t) addr;
- ktime_t timeout = ktime_add_us(ktime_get(), timeout_us);
+ ktime_t timeout;
+
+ /* timeout should be longer when working with simulator */
+ if (!hdev->pdev)
+ timeout_us *= 10;
+
+ timeout = ktime_add_us(ktime_get(), timeout_us);
might_sleep();
diff --git a/drivers/misc/habanalabs/habanalabs.h b/drivers/misc/habanalabs/habanalabs.h
index 0da80e8eab42..71243b319920 100644
--- a/drivers/misc/habanalabs/habanalabs.h
+++ b/drivers/misc/habanalabs/habanalabs.h
@@ -1042,7 +1042,12 @@ void hl_wreg(struct hl_device *hdev, u32 reg, u32 val);
#define hl_poll_timeout(hdev, addr, val, cond, sleep_us, timeout_us) \
({ \
- ktime_t __timeout = ktime_add_us(ktime_get(), timeout_us); \
+ ktime_t __timeout; \
+ /* timeout should be longer when working with simulator */ \
+ if (hdev->pdev) \
+ __timeout = ktime_add_us(ktime_get(), timeout_us); \
+ else \
+ __timeout = ktime_add_us(ktime_get(), (timeout_us * 10)); \
might_sleep_if(sleep_us); \
for (;;) { \
(val) = RREG32(addr); \
--
2.17.1
^ permalink raw reply related [flat|nested] 3+ messages in thread