diff options
| author | Omer Shpigelman <oshpigelman@habana.ai> | 2020-05-24 23:06:59 +0300 |
|---|---|---|
| committer | Oded Gabbay <oded.gabbay@gmail.com> | 2020-05-25 08:17:57 +0300 |
| commit | 8ff5f4fd40df9525675ea0e512da4cec65d646eb (patch) | |
| tree | fc7ae7c7cae4bfda1d96d9fc075d61414980a011 /drivers/misc/habanalabs/goya | |
| parent | 36fafe87edd636292a4ed6a3af9608f2c7d0d0fb (diff) | |
| download | linux-8ff5f4fd40df9525675ea0e512da4cec65d646eb.tar.bz2 | |
habanalabs: handle MMU cache invalidation timeout
MMU cache invalidation timeout indicates that the device is unstable and
therefore unusable.
Hence in such case do hard reset and return an error to the user if was
called from ioctl.
In addition, change the print to error level and rephrase its text.
Signed-off-by: Omer Shpigelman <oshpigelman@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Diffstat (limited to 'drivers/misc/habanalabs/goya')
| -rw-r--r-- | drivers/misc/habanalabs/goya/goya.c | 34 |
1 files changed, 21 insertions, 13 deletions
diff --git a/drivers/misc/habanalabs/goya/goya.c b/drivers/misc/habanalabs/goya/goya.c index 152418dfe20c..0d2952bb58df 100644 --- a/drivers/misc/habanalabs/goya/goya.c +++ b/drivers/misc/habanalabs/goya/goya.c @@ -4884,7 +4884,7 @@ static void goya_mmu_prepare(struct hl_device *hdev, u32 asid) goya_mmu_prepare_reg(hdev, goya_mmu_regs[i], asid); } -static void goya_mmu_invalidate_cache(struct hl_device *hdev, bool is_hard, +static int goya_mmu_invalidate_cache(struct hl_device *hdev, bool is_hard, u32 flags) { struct goya_device *goya = hdev->asic_specific; @@ -4893,11 +4893,11 @@ static void goya_mmu_invalidate_cache(struct hl_device *hdev, bool is_hard, if (!(goya->hw_cap_initialized & HW_CAP_MMU) || hdev->hard_reset_pending) - return; + return 0; /* no need in L1 only invalidation in Goya */ if (!is_hard) - return; + return 0; if (hdev->pldm) timeout_usec = GOYA_PLDM_MMU_TIMEOUT_USEC; @@ -4919,13 +4919,17 @@ static void goya_mmu_invalidate_cache(struct hl_device *hdev, bool is_hard, mutex_unlock(&hdev->mmu_cache_lock); - if (rc) - dev_notice_ratelimited(hdev->dev, - "Timeout when waiting for MMU cache invalidation\n"); + if (rc) { + dev_err_ratelimited(hdev->dev, + "MMU cache invalidation timeout\n"); + hl_device_reset(hdev, true, false); + } + + return rc; } -static void goya_mmu_invalidate_cache_range(struct hl_device *hdev, - bool is_hard, u32 asid, u64 va, u64 size) +static int goya_mmu_invalidate_cache_range(struct hl_device *hdev, + bool is_hard, u32 asid, u64 va, u64 size) { struct goya_device *goya = hdev->asic_specific; u32 status, timeout_usec, inv_data, pi; @@ -4933,11 +4937,11 @@ static void goya_mmu_invalidate_cache_range(struct hl_device *hdev, if (!(goya->hw_cap_initialized & HW_CAP_MMU) || hdev->hard_reset_pending) - return; + return 0; /* no need in L1 only invalidation in Goya */ if (!is_hard) - return; + return 0; if (hdev->pldm) timeout_usec = GOYA_PLDM_MMU_TIMEOUT_USEC; @@ -4970,9 +4974,13 @@ static void goya_mmu_invalidate_cache_range(struct hl_device *hdev, mutex_unlock(&hdev->mmu_cache_lock); - if (rc) - dev_notice_ratelimited(hdev->dev, - "Timeout when waiting for MMU cache invalidation\n"); + if (rc) { + dev_err_ratelimited(hdev->dev, + "MMU cache invalidation timeout\n"); + hl_device_reset(hdev, true, false); + } + + return rc; } int goya_send_heartbeat(struct hl_device *hdev) |