cann 版本：7.0.0.alpha0001
设备：Altlas 500 Pro
首先使用 atc 工具完成 onnx -> om 的模型转换，之后通过 python -m ais_bench --model xx.om --display_all_summary true --loop 10 --warmup_count 2 来运行模型，报错信息如下：
EE1001: The argument is invalid.Reason: rtMemcpy execute failed, reason=[context pointer null]
Solution: 1.Check the input parameter range of the function. 2.Check the function invocation relationship.
TraceBack (most recent call last):
The error from device(0), serial number is 25, there is an aicore error, core id is 0, error code = 0x10, dump info: pc start: 0x12404183a000, current: 0x12404183a18c, vec error info: 0xa3f57ff, mte error info: 0xb8, ifu error info: 0x2287b31a93900, ccu error info: 0x702612520024575d, cube error info: 0xa2, biu error info: 0, aic error mask: 0x65000200d000288, para base: 0x12404002e338, errorStr: Illegal instruction, which is usually caused by unaligned UUB addresses.[FUNC:PrintCoreErrorInfo][FILE:device_error_proc.cc][LINE:533]
The extend info from device(0), serial number is 25, there is aicore error, core id is 0, aicore int: 0x81, aicore error2: 0, axi clamp ctrl: 0, axi clamp state: 0x1717, biu status0: 0x1c00800000000, biu status1: 0x940002092a0000, clk gate mask: 0, dbg addr: 0, ecc en: 0x3f3f, mte ccu ecc 1bit error: 0x1000, vector cube ecc 1bit error: 0, run stall: 0x1, dbg data0: 0, dbg data1: 0, dbg data2: 0, dbg data3: 0x1, dfx data: 0x1[FUNC:PrintCoreErrorInfo][FILE:device_error_proc.cc][LINE:564]
The device(0), core list[0-0], error code is:[FUNC:PrintCoreInfoErrMsg][FILE:device_error_proc.cc][LINE:587]
coreId( 0): 0x10 [FUNC:PrintCoreInfoErrMsg][FILE:device_error_proc.cc][LINE:601]
ctx is NULL![FUNC:MemCopySync][FILE:api_impl.cc][LINE:1594]
The argument is invalid.Reason: rtMemcpy execute failed, reason=[context pointer null]
ctx is NULL![FUNC:HostMalloc][FILE:api_impl.cc][LINE:1397]
The argument is invalid.Reason: rtMallocHost execute failed, reason=[context pointer null]
Call rtMallocHost failed, size:96, ret:0x1a1fa[FUNC:LogExceptionArgs][FILE:exception_dumper.cc][LINE:293]
alloc host memory failed, runtime result = 107002[FUNC:ReportCallError][FILE:log_inner.cpp][LINE:161]
ctx is NULL![FUNC:GetDevErrMsg][FILE:api_impl.cc][LINE:4448]
The argument is invalid.Reason: rtGetDevMsg execute failed, reason=[context pointer null]请问该如何确定是哪个算子报的错？
点赞
收藏
回复
分享
举报
已解决
cann 版本：7.0.0.alpha0001
设备：Altlas 500 Pro
首先使用 atc 工具完成 onnx -> om 的模型转换，之后通过 python -m ais_bench --model xx.om --display_all_summary true --loop 10 --warmup_count 2 来运行模型，报错信息如下：
EE1001: The argument is invalid.Reason: rtMemcpy execute failed, reason=[context pointer null] Solution: 1.Check the input parameter range of the function. 2.Check the function invocation relationship. TraceBack (most recent call last): The error from device(0), serial number is 25, there is an aicore error, core id is 0, error code = 0x10, dump info: pc start: 0x12404183a000, current: 0x12404183a18c, vec error info: 0xa3f57ff, mte error info: 0xb8, ifu error info: 0x2287b31a93900, ccu error info: 0x702612520024575d, cube error info: 0xa2, biu error info: 0, aic error mask: 0x65000200d000288, para base: 0x12404002e338, errorStr: Illegal instruction, which is usually caused by unaligned UUB addresses.[FUNC:PrintCoreErrorInfo][FILE:device_error_proc.cc][LINE:533] The extend info from device(0), serial number is 25, there is aicore error, core id is 0, aicore int: 0x81, aicore error2: 0, axi clamp ctrl: 0, axi clamp state: 0x1717, biu status0: 0x1c00800000000, biu status1: 0x940002092a0000, clk gate mask: 0, dbg addr: 0, ecc en: 0x3f3f, mte ccu ecc 1bit error: 0x1000, vector cube ecc 1bit error: 0, run stall: 0x1, dbg data0: 0, dbg data1: 0, dbg data2: 0, dbg data3: 0x1, dfx data: 0x1[FUNC:PrintCoreErrorInfo][FILE:device_error_proc.cc][LINE:564] The device(0), core list[0-0], error code is:[FUNC:PrintCoreInfoErrMsg][FILE:device_error_proc.cc][LINE:587] coreId( 0): 0x10 [FUNC:PrintCoreInfoErrMsg][FILE:device_error_proc.cc][LINE:601] ctx is NULL![FUNC:MemCopySync][FILE:api_impl.cc][LINE:1594] The argument is invalid.Reason: rtMemcpy execute failed, reason=[context pointer null] ctx is NULL![FUNC:HostMalloc][FILE:api_impl.cc][LINE:1397] The argument is invalid.Reason: rtMallocHost execute failed, reason=[context pointer null] Call rtMallocHost failed, size:96, ret:0x1a1fa[FUNC:LogExceptionArgs][FILE:exception_dumper.cc][LINE:293] alloc host memory failed, runtime result = 107002[FUNC:ReportCallError][FILE:log_inner.cpp][LINE:161] ctx is NULL![FUNC:GetDevErrMsg][FILE:api_impl.cc][LINE:4448] The argument is invalid.Reason: rtGetDevMsg execute failed, reason=[context pointer null]请问该如何确定是哪个算子报的错？