All of lore.kernel.org
 help / color / mirror / Atom feed
From: Felix Kuehling <Felix.Kuehling-5C7GfCeVMHo@public.gmane.org>
To: oded.gabbay-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org,
	amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org
Cc: Felix Kuehling <Felix.Kuehling-5C7GfCeVMHo@public.gmane.org>
Subject: [PATCH 12/12] drm/amdkfd: Add sanity checks in IRQ handlers
Date: Tue,  1 May 2018 17:56:12 -0400	[thread overview]
Message-ID: <1525211772-19683-13-git-send-email-Felix.Kuehling@amd.com> (raw)
In-Reply-To: <1525211772-19683-1-git-send-email-Felix.Kuehling-5C7GfCeVMHo@public.gmane.org>

Only accept interrupts from KFD VMIDs. Just checking for a PASID may
not be enough because amdgpu started using PASIDs to map VM faults
to processes.

Warn if an IRQ doesn't have a valid PASID (indicating a firmware bug).

Suggested-by: Shaoyun Liu <Shaoyun.Liu@amd.com>
Suggested-by: Oak Zeng <Oak.Zeng@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c | 20 +++++++++---
 drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c  | 40 ++++++++++++++----------
 2 files changed, 39 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
index 3d5ccb3..49df6c7 100644
--- a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
+++ b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
@@ -27,18 +27,28 @@
 static bool cik_event_interrupt_isr(struct kfd_dev *dev,
 					const uint32_t *ih_ring_entry)
 {
-	unsigned int pasid;
 	const struct cik_ih_ring_entry *ihre =
 			(const struct cik_ih_ring_entry *)ih_ring_entry;
+	unsigned int vmid, pasid;
+
+	/* Only handle interrupts from KFD VMIDs */
+	vmid  = (ihre->ring_id & 0x0000ff00) >> 8;
+	if (vmid < dev->vm_info.first_vmid_kfd ||
+	    vmid > dev->vm_info.last_vmid_kfd)
+		return 0;
 
+	/* If there is no valid PASID, it's likely a firmware bug */
 	pasid = (ihre->ring_id & 0xffff0000) >> 16;
+	if (WARN_ONCE(pasid == 0, "FW bug: No PASID in KFD interrupt"))
+		return 0;
 
-	/* Do not process in ISR, just request it to be forwarded to WQ. */
-	return (pasid != 0) &&
-		(ihre->source_id == CIK_INTSRC_CP_END_OF_PIPE ||
+	/* Interrupt types we care about: various signals and faults.
+	 * They will be forwarded to a work queue (see below).
+	 */
+	return ihre->source_id == CIK_INTSRC_CP_END_OF_PIPE ||
 		ihre->source_id == CIK_INTSRC_SDMA_TRAP ||
 		ihre->source_id == CIK_INTSRC_SQ_INTERRUPT_MSG ||
-		ihre->source_id == CIK_INTSRC_CP_BAD_OPCODE);
+		ihre->source_id == CIK_INTSRC_CP_BAD_OPCODE;
 }
 
 static void cik_event_interrupt_wq(struct kfd_dev *dev,
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 39d4115..37029ba 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
@@ -29,27 +29,35 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev,
 					const uint32_t *ih_ring_entry)
 {
 	uint16_t source_id, client_id, pasid, vmid;
+	const uint32_t *data = ih_ring_entry;
 
-	source_id = SOC15_SOURCE_ID_FROM_IH_ENTRY(ih_ring_entry);
-	client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry);
-	pasid = SOC15_PASID_FROM_IH_ENTRY(ih_ring_entry);
+	/* Only handle interrupts from KFD VMIDs */
 	vmid = SOC15_VMID_FROM_IH_ENTRY(ih_ring_entry);
+	if (vmid < dev->vm_info.first_vmid_kfd ||
+	    vmid > dev->vm_info.last_vmid_kfd)
+		return 0;
+
+	/* If there is no valid PASID, it's likely a firmware bug */
+	pasid = SOC15_PASID_FROM_IH_ENTRY(ih_ring_entry);
+	if (WARN_ONCE(pasid == 0, "FW bug: No PASID in KFD interrupt"))
+		return 0;
 
-	if (pasid) {
-		const uint32_t *data = ih_ring_entry;
+	source_id = SOC15_SOURCE_ID_FROM_IH_ENTRY(ih_ring_entry);
+	client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry);
 
-		pr_debug("client id 0x%x, source id %d, pasid 0x%x. raw data:\n",
-			 client_id, source_id, pasid);
-		pr_debug("%8X, %8X, %8X, %8X, %8X, %8X, %8X, %8X.\n",
-			 data[0], data[1], data[2], data[3],
-			 data[4], data[5], data[6], data[7]);
-	}
+	pr_debug("client id 0x%x, source id %d, pasid 0x%x. raw data:\n",
+		 client_id, source_id, pasid);
+	pr_debug("%8X, %8X, %8X, %8X, %8X, %8X, %8X, %8X.\n",
+		 data[0], data[1], data[2], data[3],
+		 data[4], data[5], data[6], data[7]);
 
-	return (pasid != 0) &&
-		(source_id == SOC15_INTSRC_CP_END_OF_PIPE ||
-		 source_id == SOC15_INTSRC_SDMA_TRAP ||
-		 source_id == SOC15_INTSRC_SQ_INTERRUPT_MSG ||
-		 source_id == SOC15_INTSRC_CP_BAD_OPCODE);
+	/* Interrupt types we care about: various signals and faults.
+	 * They will be forwarded to a work queue (see below).
+	 */
+	return source_id == SOC15_INTSRC_CP_END_OF_PIPE ||
+		source_id == SOC15_INTSRC_SDMA_TRAP ||
+		source_id == SOC15_INTSRC_SQ_INTERRUPT_MSG ||
+		source_id == SOC15_INTSRC_CP_BAD_OPCODE;
 }
 
 static void event_interrupt_wq_v9(struct kfd_dev *dev,
-- 
2.7.4

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

  parent reply	other threads:[~2018-05-01 21:56 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-01 21:56 [PATCH 00/12] Assorted KFD fixes Felix Kuehling
     [not found] ` <1525211772-19683-1-git-send-email-Felix.Kuehling-5C7GfCeVMHo@public.gmane.org>
2018-05-01 21:56   ` [PATCH 01/12] drm/amdkfd: Dump HQD of HIQ Felix Kuehling
2018-05-01 21:56   ` [PATCH 02/12] drm/amdkfd: Reduce priority of context-saving waves before spin-wait Felix Kuehling
2018-05-01 21:56   ` [PATCH 03/12] drm/amdkfd: Use volatile MTYPE in default/alternate apertures Felix Kuehling
2018-05-01 21:56   ` [PATCH 04/12] drm/amdkfd: use %px to print user space address instead of %p Felix Kuehling
2018-05-01 21:56   ` [PATCH 05/12] drm/amdkfd: Remove redundant include of amd-iommu.h Felix Kuehling
2018-05-01 21:56   ` [PATCH 06/12] drm/amdkfd: Separate trap handler assembly code and its hex values Felix Kuehling
2018-05-01 21:56   ` [PATCH 07/12] drm/amdkfd: Fix CP soft hang on APUs Felix Kuehling
2018-05-01 21:56   ` [PATCH 08/12] drm/amdkfd: Fix signal handling performance again Felix Kuehling
2018-05-01 21:56   ` [PATCH 09/12] drm/amdkfd: Remove initialization of cp_hqd_ib_control on CIK Felix Kuehling
2018-05-01 21:56   ` [PATCH 10/12] drm/amdkfd: Locking PM mutex while allocating IB buffer Felix Kuehling
2018-05-01 21:56   ` [PATCH 11/12] drm/amdkfd: Remove queue node when destroy queue failed Felix Kuehling
2018-05-01 21:56   ` Felix Kuehling [this message]
2018-05-12 11:13   ` [PATCH 00/12] Assorted KFD fixes Oded Gabbay

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1525211772-19683-13-git-send-email-Felix.Kuehling@amd.com \
    --to=felix.kuehling-5c7gfcevmho@public.gmane.org \
    --cc=amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org \
    --cc=oded.gabbay-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.