# Debugging PCIe Link Speed Negotiation Failures via lspci
Canonical Interactive Tool: https://www.whyitsbroken.com/fix/debugging-pcie-link-speed-negotiation-failures-via-lspci
Subsystem: PCIe Bus and Interconnect Subsystem | Difficulty: Moderate DIY | Time: 15-45 min | DIY Cost: $0.00-$45.00 (vs Pro $150.00-$340.00)
Empirical Verified Fixes: 1 | Evolution Version: v1

## Direct Answer Summary
The most probable cause is physical signal degradation or improper seating of the PCIe device. Confirm by running 'lspci -vvv' to compare LnkSta vs LnkCap. If LnkSta is lower, inspect the physical slot and connectors for debris or oxidation. DIY fixes typically cost under $45, whereas professional service ranges from $150 to $340.

## Ranked Root Causes & Telemetry Signatures
### 1. Physical Signal Integrity / Oxidation (Baseline Probability: 40%)
- **Telemetry Signature:** Normal: 0.8V-1.2V signal swing | Fault: High bit error rate (BER) in PCIe logs
- **Why It Happens:** Oxidation or debris on the gold fingers of the PCIe card or inside the slot increases impedance, forcing the controller to down-train to a lower speed for stability.
- **60-Second Verification Test:** Remove card, clean gold fingers with 99% isopropyl alcohol, and re-seat firmly.
- **Resolution Steps:**
  1. Power off system and unplug power cables.
  2. Remove PCIe card and inspect slot for debris.
  3. Clean contacts with alcohol and re-seat.
  4. Reboot and verify link speed with lspci -vvv.

### 2. Slot Bandwidth / Lane Allocation (Baseline Probability: 25%)
- **Telemetry Signature:** Normal: x16 electrical | Fault: x8 or x4 electrical in x16 physical slot
- **Why It Happens:** The motherboard shares PCIe lanes between slots; populating multiple slots often forces the controller to bifurcate or reduce lane width for all devices.
- **60-Second Verification Test:** Check motherboard manual to see if the slot is electrically wired for fewer lanes than its physical size.
- **Resolution Steps:**
  1. Consult motherboard manual for lane configuration.
  2. Move card to a primary slot if available.
  3. Verify lane width in BIOS/UEFI settings.
  4. Reboot and verify link speed.

### 3. Thermal Throttling (Baseline Probability: 20%)
- **Telemetry Signature:** Normal: < 75°C | Fault: > 90°C during link training
- **Why It Happens:** High temperatures on the PCIe controller or bridge chip trigger protective throttling, which can include reducing link speed to lower power consumption and heat.
- **60-Second Verification Test:** Monitor device temperature during load using 'nvidia-smi' or 'sensors'.
- **Resolution Steps:**
  1. Clean dust from heatsinks and fans.
  2. Verify airflow path is unobstructed.
  3. Re-apply thermal interface material if necessary.
  4. Test under load to confirm stable link.

### 4. BIOS/Firmware Configuration (Baseline Probability: 15%)
- **Telemetry Signature:** Normal: PCIe Gen4/5 enabled | Fault: PCIe Gen2/3 forced in BIOS
- **Why It Happens:** BIOS settings may be hard-coded to a lower PCIe generation for compatibility, preventing the device from negotiating its maximum supported speed.
- **60-Second Verification Test:** Enter BIOS/UEFI and check 'PCIe Link Speed' or 'Gen' settings for the specific slot.
- **Resolution Steps:**
  1. Enter BIOS during boot.
  2. Navigate to Advanced PCIe settings.
  3. Set link speed to 'Auto' or 'Gen4/5'.
  4. Save and exit.

## Frequently Asked Questions
**Q: How do I interpret lspci -vvv output?**
Look for 'LnkCap' (the maximum capability of the hardware) and 'LnkSta' (the current negotiated speed). If LnkSta is lower than LnkCap, a negotiation failure or limitation is present.

**Q: Can a bad PCIe cable cause this?**
Yes, especially with riser cables. Poor shielding or signal loss in low-quality cables often causes the link to train down to Gen1 or Gen2 speeds.

**Q: Does 'downgraded' always mean hardware failure?**
No. It often indicates a configuration mismatch, thermal throttling, or a bandwidth limitation imposed by the motherboard's lane allocation logic.

**Q: What is the LTSSM state?**
The Link Training and Status State Machine (LTSSM) manages the PCIe link. If it is not in the L0 state, the link has failed to train successfully.

## External Technical References
- [PowerEdge: XE9685L - Linux command "lspci -vvv" shows "downgraded" on the NVIDIA CX7 | Dell US](https://www.dell.com/support/kbdoc/en-us/000308325/poweredge-xe9685l-linux-command-lspci-vvv-shows-downgraded-on-the-nvidia-cx7-device)
- [General Debug Checklist — PCIe Debug K-Map 1.0 documentation](https://xilinx.github.io/pcie-debug-kmap/pciedebug/build/html/docs/UltraScale%2B_Devices_Integrated_Block_for_PCIExpress/debug_faq.html)
- [AM5728: PCIe link status - Processors forum - Processors - TI E2E support forums](https://e2e.ti.com/support/processors-group/processors/f/processors-forum/830571/am5728-pcie-link-status)
- [PCIe | BlueField Troubleshooting Guide](https://networking-docs.nvidia.com/bfswtroubleshooting/pcie)
- [lspci(8) - Linux manual page](https://man7.org/linux/man-pages/man8/lspci.8.html)
