| Seq | Timestamp | Type | Name | Duration |
|---|---|---|---|---|
| 0/0 | 2026/10/27 19:44 | flow | patch-triage |
1mArgs:null Results: {
"EnableConfigs": null,
"FocusSymbols": [],
"KMSANReasoning": "",
"NeedsKMSAN": false,
"Reasoning": "The modified code is in the Intel iavf and ice Ethernet drivers (drivers/net/ethernet/intel/iavf/ and drivers/net/ethernet/intel/ice/). These drivers require physical Intel PCIe hardware (E800 series NICs or SR-IOV virtual functions) that are not emulated in standard QEMU or virtualized test environments. Consequently, the modified TSO transmit paths (iavf_tso and ice_tso) are structurally unreachable during fuzzing.",
"WorthFuzzing": false
} |
| 1/1 | 2026/10/27 19:44 | action | read-patch-diff |
0mArgs:null Results: {
"PatchDiff": "commit 4b1f001bd81457984dc0c266f30695c260d722c2\nAuthor: syz-cluster \u003ctriage@syzkaller.com\u003e\nDate: Tue Oct 27 19:44:17 2026 +0000\n\n syz-cluster: applied patch under review\n\ndiff --git a/drivers/net/ethernet/intel/iavf/iavf_txrx.c b/drivers/net/ethernet/intel/iavf/iavf_txrx.c\nindex 8982cf75466faa..2c290d7788d27d 100644\n--- a/drivers/net/ethernet/intel/iavf/iavf_txrx.c\n+++ b/drivers/net/ethernet/intel/iavf/iavf_txrx.c\n@@ -1807,7 +1807,7 @@ static int iavf_tso(struct iavf_tx_buffer *first, u8 *hdr_len,\n \t\tcsum_replace_by_diff(\u0026l4.udp-\u003echeck,\n \t\t\t\t (__force __wsum)htonl(paylen));\n \t\t/* compute length of UDP segmentation header */\n-\t\t*hdr_len = (u8)sizeof(l4.udp) + l4_offset;\n+\t\t*hdr_len = sizeof(*l4.udp) + l4_offset;\n \t} else {\n \t\tcsum_replace_by_diff(\u0026l4.tcp-\u003echeck,\n \t\t\t\t (__force __wsum)htonl(paylen));\ndiff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c\nindex a2c7c4962882da..d9ea812c5c4e81 100644\n--- a/drivers/net/ethernet/intel/ice/ice_txrx.c\n+++ b/drivers/net/ethernet/intel/ice/ice_txrx.c\n@@ -1925,7 +1925,7 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off)\n \t\tcsum_replace_by_diff(\u0026l4.udp-\u003echeck,\n \t\t\t\t (__force __wsum)htonl(paylen));\n \t\t/* compute length of UDP segmentation header */\n-\t\toff-\u003eheader_len = (u8)sizeof(l4.udp) + l4_start;\n+\t\toff-\u003eheader_len = sizeof(*l4.udp) + l4_start;\n \t} else {\n \t\tcsum_replace_by_diff(\u0026l4.tcp-\u003echeck,\n \t\t\t\t (__force __wsum)htonl(paylen));\n"
} |
| 2/1 | 2026/10/27 19:44 | agent | patch-evaluator |
1mModel:core Args: null Results: {
"EnableConfigs": null,
"FocusSymbols": [],
"Reasoning": "The modified code is in the Intel iavf and ice Ethernet drivers (drivers/net/ethernet/intel/iavf/ and drivers/net/ethernet/intel/ice/). These drivers require physical Intel PCIe hardware (E800 series NICs or SR-IOV virtual functions) that are not emulated in standard QEMU or virtualized test environments. Consequently, the modified TSO transmit paths (iavf_tso and ice_tso) are structurally unreachable during fuzzing.",
"WorthFuzzing": false
}Instruction: You are an expert Linux kernel maintainer and security engineer.
Your job is to review a provided patch series and evaluate whether it warrants fuzzing with syzkaller.
IMPORTANT: The changes have ALREADY been applied and committed as the HEAD commit in
your workspace. Do NOT rely on internal assumptions. You must actively use your code access
tools to inspect the actual source code, callers, and surrounding context.
================================================================================
1. CORE TRIAGE PHILOSOPHY
================================================================================
The goal of patch fuzzing is to discover crashes, regressions, exposed latent bugs,
and newly triggered assertions introduced by the patch series.
- REACHABILITY IS THE PRIMARY GATE:
Fuzzing can only discover bugs in code that can actually execute in standard virtualized
environments (GCE or QEMU, utilizing software-emulated devices like USB gadgets, netdev, tun/tap).
If the modified code is structurally unreachable (see Section 2), it MUST NOT be fuzzed,
regardless of whether it adds assertions or complex logic.
- DO NOT BLINDLY TRUST "NO FUNCTIONAL CHANGE" (NFCI) OR "REFACTORING" CLAIMS:
Patch authors routinely label changes as "cleanups", "refactorings", or state
"No functional change intended". Do NOT take these claims at face value.
Code refactorings that rearrange logic, introduce helper functions, or alter state management
in core subsystems frequently introduce subtle semantic shifts or uncover latent kernel bugs.
If reachable executable code is modified or refactored, it MUST be fuzzed.
- NEW OR MODIFIED ASSERTIONS IN REACHABLE CODE MUST BE FUZZED:
When a patch introduces or modifies runtime checks or assertions (e.g., WARN_ON*, VM_WARN_ON*,
BUG_ON*, lockdep_assert*) in reachable code paths, it enforces new or stricter invariants.
Even if the author believes the invariant always holds, fuzzing is essential to verify whether
an unusual sequence of operations can violate it.
================================================================================
2. WHEN TO RETURN WorthFuzzing=false (NEGATIVE CRITERIA)
================================================================================
Return WorthFuzzing=false ONLY IF all modified code falls strictly into one or more of these categories:
- Non-kernel and non-executable changes:
* Modifications to Documentation/, comments, or spelling fixes.
* User-space directories, self-tests, samples, or scripts (e.g., tools/, samples/, scripts/, usr/)
that do not affect the compiled kernel image (vmlinux) or kernel modules.
* Purely decorative logging (e.g., message strings in pr_err, printk, dev_info) or tracepoints
that do not alter control flow or data structures.
* Build system or Kconfig changes that do not alter compiled C logic.
- Structurally unreachable hardware:
* Vendor-specific PCIe switches, SmartNICs, or GPU drivers (e.g., mlxsw, pds_core, qed,
ionic, amdgpu) requiring physical ASIC/PCIe cards not emulated in standard QEMU.
- Unreachable execution paths:
* Driver teardown callbacks (.remove, .shutdown, pci_unregister_driver) executed only during
physical PCI hot-unplug or manual sysfs driver unbinding.
* Code paths exclusive to architectures other than the target architecture.
================================================================================
3. WHEN TO RETURN WorthFuzzing=true (POSITIVE CRITERIA)
================================================================================
Return WorthFuzzing=true whenever the patch touches reachable executable code, including:
- Core Subsystems:
* Any logic modifications in memory management (mm/), synchronization/locking (kernel/locking/),
BPF, scheduler, core networking, VFS, or syscall handling.
- Refactorings and Code Cleanups:
* Any restructuring of reachable data structures, helper abstractions, or algorithm flows.
- Runtime Assertions and Defensive Checks:
* Any introduction or alteration of assertions (WARN_ON*, VM_WARN_ON*, BUG_ON*, etc.) in reachable paths.
- Reachable Drivers and Protocols:
* Drivers accessible via virtual buses (virtio, USB gadget, loopback, netlink, binder, sockets, etc.).
================================================================================
4. EXTRACTING FocusSymbols (PREVENTING DILUTION)
================================================================================
When WorthFuzzing=true, you must extract specific kernel functions into FocusSymbols to guide the fuzzer:
- AVOID UBIQUITOUS LIFECYCLE HOT-PATHS:
Do NOT list generic, ubiquitous functions called by almost every program in the corpus
(including, but not limited to: general memory allocators and deallocators, page fault
and trap handlers, or core synchronization primitives; this is not an exhaustive list).
Listing ubiquitous functions causes the fuzzer to classify thousands of unrelated tests as "focused",
which severely dilutes fuzzing effort away from the actual changes.
- TARGET SPECIFIC FEATURE LOGIC AND ENTRYPOINTS:
List functions that specifically implement the logic being added or altered, or direct API entrypoints
for the subsystem feature under review.
- HANDLING STATIC INLINE FUNCTIONS IN HEADERS (.h):
Compiler-inlined static functions (such as static inlines in mm/*.h or include/linux/*.h) lack
distinct symbol addresses in vmlinux and cannot be targeted directly by symbol coverage filters.
If the changes are primarily in static inline helpers, identify non-static, feature-specific caller
functions in .c files that exercise them (avoiding ubiquitous lifecycle wrappers).
================================================================================
5. IDENTIFYING EnableConfigs
================================================================================
Identify any specific CONFIG_ options required to properly compile and reach the modified code:
- Inspect Kconfig files and #ifdef guards; do not make assumptions.
- Check "depends on" lines in Kconfig to include any non-standard parent subsystem configs needed.
- Strip any 'CONFIG_' prefix (e.g., return "NET_IPV4" instead of "CONFIG_NET_IPV4").
Prefer calling several tools at the same time to save round-trips.
Use set-results tool to provide results of the analysis.
It must be called exactly once before the final reply.
Ignore results of this tool.
Prompt: Target architecture: amd64
For your convenience, here is the diff of the changes:
commit 4b1f001bd81457984dc0c266f30695c260d722c2
Author: syz-cluster <triage@syzkaller.com>
Date: Tue Oct 27 19:44:17 2026 +0000
syz-cluster: applied patch under review
diff --git a/drivers/net/ethernet/intel/iavf/iavf_txrx.c b/drivers/net/ethernet/intel/iavf/iavf_txrx.c
index 8982cf75466faa..2c290d7788d27d 100644
--- a/drivers/net/ethernet/intel/iavf/iavf_txrx.c
+++ b/drivers/net/ethernet/intel/iavf/iavf_txrx.c
@@ -1807,7 +1807,7 @@ static int iavf_tso(struct iavf_tx_buffer *first, u8 *hdr_len,
csum_replace_by_diff(&l4.udp->check,
(__force __wsum)htonl(paylen));
/* compute length of UDP segmentation header */
- *hdr_len = (u8)sizeof(l4.udp) + l4_offset;
+ *hdr_len = sizeof(*l4.udp) + l4_offset;
} else {
csum_replace_by_diff(&l4.tcp->check,
(__force __wsum)htonl(paylen));
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c
index a2c7c4962882da..d9ea812c5c4e81 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.c
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.c
@@ -1925,7 +1925,7 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off)
csum_replace_by_diff(&l4.udp->check,
(__force __wsum)htonl(paylen));
/* compute length of UDP segmentation header */
- off->header_len = (u8)sizeof(l4.udp) + l4_start;
+ off->header_len = sizeof(*l4.udp) + l4_start;
} else {
csum_replace_by_diff(&l4.tcp->check,
(__force __wsum)htonl(paylen));
|
| 3/2 | 2026/10/27 19:44 | llm | patch-evaluator |
0mModel:gemini-3.8-flash Args: null Results: null Tokens: input: 3401 output: 53 thoughts: 993 Thoughts: **Evaluating the Patch Series** Okay, so I need to evaluate this patch series with the syzkaller fuzzing triage criteria. Let's see... the diff shows modifications in `iavf_txrx.c` and `ice_txrx.c`. I need to understand what these drivers are. Ah, `iavf` is the Intel Adaptive Virtual Function Network Driver, used for SR-IOV VFs on newer Intel NICs, and `ice` is the Intel Ethernet Controller E800 Series driver – physical PCIe devices, I'd wager. Now I remember... the instructions explicitly say to exclude "Structurally unreachable hardware: Vendor-specific PCIe switches, SmartNICs, or GPU drivers." I need to determine if `iavf` and `ice` fall into this category. Can I get these devices instantiated in QEMU? No... wait, does QEMU even emulate `ice` or `iavf`? No. QEMU's networking support leans towards things like `e1000`, `vmxnet3`, and `virtio-net` – none of which are equivalent. Looks like `iavf` and `ice` won't be easily emulated without specific hardware or detailed emulation of these NICs, making this likely a no-go for standard syzkaller triage. I'll need to dig into the code a little to be sure, though. Let's see what the drivers themselves look like. |
| 4/2 | 2026/10/27 19:44 | tool | grepper |
0mArgs:{
"Expression": "MODULE_DEVICE_TABLE\\(pci",
"PathPrefixes": [
"drivers/net/ethernet/intel/iavf",
"drivers/net/ethernet/intel/ice"
]
}Results: {
"Output": "drivers/net/ethernet/intel/iavf/iavf_main.c=38=static const struct pci_device_id iavf_pci_tbl[] = {\n--\ndrivers/net/ethernet/intel/iavf/iavf_main.c-46-\ndrivers/net/ethernet/intel/iavf/iavf_main.c:47:MODULE_DEVICE_TABLE(pci, iavf_pci_tbl);\ndrivers/net/ethernet/intel/iavf/iavf_main.c-48-\n--\ndrivers/net/ethernet/intel/ice/ice_main.c=5793=static const struct pci_device_id ice_pci_tbl[] = {\n--\ndrivers/net/ethernet/intel/ice/ice_main.c-5845-};\ndrivers/net/ethernet/intel/ice/ice_main.c:5846:MODULE_DEVICE_TABLE(pci, ice_pci_tbl);\ndrivers/net/ethernet/intel/ice/ice_main.c-5847-\n"
} |
| 5/2 | 2026/10/27 19:44 | llm | patch-evaluator |
0mModel:gemini-3.8-flash Args: null Results: null Tokens: input: 5028 output: 45 thoughts: 26 Thoughts: Okay, here's my take on those thoughts, tailored for an expert audience: **Examining PCI Device Tables in i40e and Ice Drivers** Alright, let's dive into the core of this investigation. My immediate thought is to cross-reference the PCI device tables, specifically `iavf_pci_tbl` in the i40e driver and `ice_pci_tbl` in the Ice driver. We're looking for subtle differences or potential overlaps in supported hardware. I want to quickly ascertain if a specific device is supported in one driver but not the other, or if there's any ambiguity that could lead to unexpected behavior during device probe or initialization. Considering the complexities of these drivers and their respective families, a thorough comparison here could quickly pinpoint potential compatibility issues or areas where one driver might have a more robust device detection strategy. I'm thinking about edge cases, perhaps specific device revisions or vendor customizations; are they handled consistently across both? This is where a quick side-by-side comparison of the relevant data structures will be invaluable. Time to get those tables open and see what they reveal. |
| 6/2 | 2026/10/27 19:44 | tool | read-file |
0mArgs:{
"File": "drivers/net/ethernet/intel/iavf/iavf_main.c",
"FirstLine": 35,
"LineCount": 20
}Results: {
"Contents": " 35:\t * { Vendor ID, Device ID, SubVendor ID, SubDevice ID,\n 36:\t * Class, Class Mask, private data (not used) }\n 37:\t */\n 38:\tstatic const struct pci_device_id iavf_pci_tbl[] = {\n 39:\t\t{ PCI_VDEVICE(INTEL, IAVF_DEV_ID_VF) },\n 40:\t\t{ PCI_VDEVICE(INTEL, IAVF_DEV_ID_VF_HV) },\n 41:\t\t{ PCI_VDEVICE(INTEL, IAVF_DEV_ID_X722_VF) },\n 42:\t\t{ PCI_VDEVICE(INTEL, IAVF_DEV_ID_ADAPTIVE_VF) },\n 43:\t\t/* required last entry */\n 44:\t\t{ }\n 45:\t};\n 46:\t\n 47:\tMODULE_DEVICE_TABLE(pci, iavf_pci_tbl);\n 48:\t\n 49:\tMODULE_ALIAS(\"i40evf\");\n 50:\tMODULE_DESCRIPTION(\"Intel(R) Ethernet Adaptive Virtual Function Network Driver\");\n 51:\tMODULE_IMPORT_NS(\"LIBETH\");\n 52:\tMODULE_IMPORT_NS(\"LIBIE\");\n 53:\tMODULE_IMPORT_NS(\"LIBIE_ADMINQ\");\n 54:\tMODULE_LICENSE(\"GPL v2\");\n"
} |
| 7/2 | 2026/10/27 19:44 | llm | patch-evaluator |
0mModel:gemini-3.8-flash Args: null Results: null Tokens: input: 5696 output: 24 thoughts: 122 Thoughts: **Investigating Intel PCIe Ethernet Driver Fuzzing Availability** Okay, so I'm thinking about fuzzing Intel PCIe Ethernet drivers, specifically `iavf` and `ice`. I know they're both Intel drivers; `ice` is for the E800 series physical NICs, and `iavf` is for the Intel 700 and 800 series SR-IOV Virtual Functions. The challenge here is that neither of these drivers are typically emulated by standard QEMU installations, nor are they readily available in your typical GCE virtualized fuzzing environment. Hold on, before I get too far down this road, I should just quickly double-check that assumption. Are there any virtual device implementations, perhaps some sort of mock device, or software emulation for either `ice` or `iavf` that I might have missed? Let's see... I need to check for anything like `netdevsim`, or the use of a `dummy` network interface, or any other related mechanisms for simulating or abstracting the network card's behavior within these drivers. Gotta cover all my bases. |
| 8/2 | 2026/10/27 19:45 | tool | grepper |
0mArgs:{
"Expression": "iavf_tso|ice_tso"
}Results: {
"Output": "drivers/net/ethernet/intel/iavf/iavf_txrx.c=1696=static void iavf_tx_prepare_vlan_flags(struct sk_buff *skb,\n--\ndrivers/net/ethernet/intel/iavf/iavf_txrx.c-1721-/**\ndrivers/net/ethernet/intel/iavf/iavf_txrx.c:1722: * iavf_tso - set up the tso context descriptor\ndrivers/net/ethernet/intel/iavf/iavf_txrx.c-1723- * @first: pointer to first Tx buffer for xmit\n--\ndrivers/net/ethernet/intel/iavf/iavf_txrx.c-1728- **/\ndrivers/net/ethernet/intel/iavf/iavf_txrx.c:1729:static int iavf_tso(struct iavf_tx_buffer *first, u8 *hdr_len,\ndrivers/net/ethernet/intel/iavf/iavf_txrx.c-1730-\t\t u64 *cd_type_cmd_tso_mss)\n--\ndrivers/net/ethernet/intel/iavf/iavf_txrx.c=2291=static netdev_tx_t iavf_xmit_frame_ring(struct sk_buff *skb,\n--\ndrivers/net/ethernet/intel/iavf/iavf_txrx.c-2352-\ndrivers/net/ethernet/intel/iavf/iavf_txrx.c:2353:\ttso = iavf_tso(first, \u0026hdr_len, \u0026cd_type_cmd_tso_mss);\ndrivers/net/ethernet/intel/iavf/iavf_txrx.c-2354-\n--\ndrivers/net/ethernet/intel/ice/ice_txrx.c=1810=ice_tx_prepare_vlan_flags(struct ice_tx_ring *tx_ring, struct ice_tx_buf *first)\n--\ndrivers/net/ethernet/intel/ice/ice_txrx.c-1833-/**\ndrivers/net/ethernet/intel/ice/ice_txrx.c:1834: * ice_tso - computes mss and TSO length to prepare for TSO\ndrivers/net/ethernet/intel/ice/ice_txrx.c-1835- * @first: pointer to struct ice_tx_buf\n--\ndrivers/net/ethernet/intel/ice/ice_txrx.c=1840=static\ndrivers/net/ethernet/intel/ice/ice_txrx.c:1841:int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off)\ndrivers/net/ethernet/intel/ice/ice_txrx.c-1842-{\n--\ndrivers/net/ethernet/intel/ice/ice_txrx.c=2157=ice_xmit_frame_ring(struct sk_buff *skb, struct ice_tx_ring *tx_ring)\n--\ndrivers/net/ethernet/intel/ice/ice_txrx.c-2211-\t/* set up TSO offload */\ndrivers/net/ethernet/intel/ice/ice_txrx.c:2212:\ttso = ice_tso(first, \u0026offload);\ndrivers/net/ethernet/intel/ice/ice_txrx.c-2213-\tif (tso \u003c 0)\n"
} |
| 9/2 | 2026/10/27 19:45 | llm | patch-evaluator |
0mModel:gemini-3.8-flash Args: null Results: null Tokens: input: 6954 output: 135 thoughts: 978 Thoughts: **Analysis of Potentially Unfuzzable Code**
Okay, so I've been given some code changes and I need to determine if they're relevant for fuzzing. Let's break this down. First, the code modifications are within the `iavf` and `ice` drivers, specifically `iavf_tso` and `ice_tso`. These functions, as I understand it, are part of the transmit datapath for Intel's Adaptive Virtual Function (iavf) and Ethernet Controller E800 series (ice) network interface controllers.
Now, a key point: these drivers, and specifically these functions, are tied to *specific* Intel hardware. They rely on the presence of either physical Intel PCI/PCIe cards or SR-IOV Virtual Functions instantiated by a physical Intel NIC. Importantly, I know from experience that standard QEMU (and by extension, environments like Google Compute Engine) do *not* emulate this hardware. The standard emulated network devices are limited to things like e1000, e1000e, virtio-net, etc. - nothing that models the low-level behavior of these high-performance 100G/200G NICs.
I checked the instructions, and they seem to agree. Section 2 explicitly calls out "Structurally unreachable hardware" including vendor-specific PCIe switches and drivers that can't be emulated in standard QEMU. I'm confident that `iavf` and `ice` fall squarely into this category.
The code change itself, going from `sizeof(l4.udp)` to `sizeof(*l4.udp)`, while semantically important, is a red herring in this context. It appears to be fixing a potential bug on 32-bit systems where the size of a pointer and the size of the `udphdr` structure could be different. However, the driver is still unreachable for the stated reasons.
Therefore, because this code is in device drivers for hardware that is not emulated in standard fuzzing environments, this modified code is structurally unreachable. The tool should be informed that this is *not* worth fuzzing. I need to format my `set-results` to reflect this:
* `WorthFuzzing`: `false`
* `Reasoning`:
1. The changes are within `drivers/net/ethernet/intel/iavf` and `drivers/net/ethernet/intel/ice`, which are device drivers for Intel Adaptive Virtual Function (iavf) and Intel Ethernet Controller E800 series (ice).
2. These drivers require physical Intel PCI/PCIe hardware (or SR-IOV virtual functions) that are not emulated by standard QEMU or available in standard GCE virtualized environments.
3. The modified functions (`iavf_tso` and `ice_tso`) are in the transmit datapath of these specific hardware devices and cannot be reached in standard fuzzing environments.
* `FocusSymbols`: `[]`
* `EnableConfigs`: `[]`
This looks good, and this aligns with the provided instructions. Let's send those results.
|
| 10/2 | 2026/10/27 19:45 | tool | set-results |
0mArgs:{
"EnableConfigs": [],
"FocusSymbols": [],
"Reasoning": "The modified code is in the Intel iavf and ice Ethernet drivers (drivers/net/ethernet/intel/iavf/ and drivers/net/ethernet/intel/ice/). These drivers require physical Intel PCIe hardware (E800 series NICs or SR-IOV virtual functions) that are not emulated in standard QEMU or virtualized test environments. Consequently, the modified TSO transmit paths (iavf_tso and ice_tso) are structurally unreachable during fuzzing.",
"WorthFuzzing": false
}Results: {
"EnableConfigs": null,
"FocusSymbols": [],
"Reasoning": "The modified code is in the Intel iavf and ice Ethernet drivers (drivers/net/ethernet/intel/iavf/ and drivers/net/ethernet/intel/ice/). These drivers require physical Intel PCIe hardware (E800 series NICs or SR-IOV virtual functions) that are not emulated in standard QEMU or virtualized test environments. Consequently, the modified TSO transmit paths (iavf_tso and ice_tso) are structurally unreachable during fuzzing.",
"WorthFuzzing": false
} |
| 11/1 | 2026/10/27 19:45 | action | If |
0mArgs:{
"WorthFuzzing": false
}Results: null |
| Total Calls | Total Tokens | Avg Tokens | Total Duration (Seconds) | Avg Duration (Seconds) |
|---|
| Total Calls | Total Duration (Seconds) | Avg Duration (Seconds) |
|---|