PSOD in Heap_ManagesExtent() or HeapManagesExtentInt() via Heap_BackMapGet()

Products

• VMware vSphere ESXi

Issue/Introduction

Symptoms:

PSOD backtraces is similar to the following:

Panic Message: @BlueScreen: #GP Exception 13 in world 2097473:lpage-est @ 0x420017936474

Backtrace:

 0x4538ca09bf00:[0x420017936474]Heap_BackMapGet@vmkernel#nover+0x78 stack: 0x135200, 0x420017956ec8, 0x3010f860, 0x4200179619ad, 0x3010f860

 0x4538ca09bf20:[0x420017956ec7]BackMap_Lookup@vmkernel#nover+0x90 stack: 0x3010f860, 0x135200, 0x5634bcd3, 0x420017b86107, 0x6

 0x4538ca09bf50:[0x420017b86106]MemSchedEst_LPageLoop@vmkernel#nover+0xab stack: 0x1, 0x100002a7e00, 0x4538ca021980, 0x3, 0x31

 0x4538ca09bfe0:[0x420017b69481]CpuSched_StartWorld@vmkernel#nover+0x82 stack: 0x0, 0x4200178be6a0, 0x0, 0x0, 0x0

 0x4538ca09c000:[0x4200178be69f]Debug_IsInitialized@vmkernel#nover+0xc stack: 0x0, 0x0, 0x0, 0x0, 0x0

Environment

VMware vSphere ESXi 7.0.1

Resolution

If the ESXi host uses Marvell NICs, please double check the management firmware version. For firmware before 7.16.10, there is a bug that can cause wrong DMA when iBFT is enabled. The wrong DMA leads to a memory corruption causing crash. Please update to firmware 7.16.10 or later if applicable.

If the ESXi host uses other types of NICs, please consider applying the following boot option:

dmaMapperPolicy=debug

The boot option enables more debugging capability in ESXi and will help to analyze the problem when it is reproduced again.

Workaround:
No workaround