Libcudest

Reverse engineering of the CUDA system. CUDA primarily communicates with the NVIDIA closed-source driver via several dozen undocumented ioctl()s. My open source implementation, libcudest, is located at GitHub. It began as a project for Hyesoon Kim's CS4803DGC at the Georgia Institute of Technology.

Driver versions

Newer drivers can be used with older CUDA versions, but the converse is not true. The "CUDA macroversion" listed below is the first CUDA release designed explicitly for use with the listed drivers.

Version	CUDA macroversion	Notes
195.36.15	3.0
195.36.24	3.0
195.36.31	3.0
256.22	3.1-beta
256.29	3.1-beta

CUDA Environment variables

Discovered via binary analysis and a shimmed getenv(3). Effects determined via blackbox and binary analyses:

Variable	Notes	Documented?	Effects
__RM_NO_VERSION_CHECK		N
COMPUTE_PROFILE		Y	If set to 1, profiling will be performed. Implies CUDA_LAUNCH_BLOCKING.
COMPUTE_PROFILE_CONFIG		Y	Specifies a profiler configuration file. Only checked if COMPUTE_PROFILE is set.
COMPUTE_PROFILE_CSV		Y	If set to 1, a profiling data will be written in CSV format. Only checked if COMPUTE_PROFILE is set.
COMPUTE_PROFILE_LOG		Y	Specifies profiler output file (default: "./cuda_profile.log"). Only checked if COMPUTE_PROFILE is set.
CUDA_AMODEL_DLL		N
CUDA_AMODEL_GPU		N
CUDA_API_TRACE_PTR		N
CUDA_CACHE_DISABLE		Y	If this is unset, the code cache will be used.
CUDA_CACHE_MAXSIZE		Y
CUDA_CACHE_PATH		Y	If this is set, it overrides the code cache's default path of $HOME/.nv/ComputeCache
CUDA_DEVCODE_CACHE		Y	PTX compilation cache.
CUDA_DEVCODE_PATH		Y	Search path for fat binaries.
CUDA_EMULATION_MODE
CUDA_FORCE_PTX_JIT
CUDA_HEAP_RANGE	Checked each time a context is created
CUDA_INJECTION64_PATH
CUDA_LAUNCH_BLOCKING		Y (CUDA 3.0 Programmer's Guide, 3.2.6.1)	Forces synchronization of host threads on GPU kernels.
CUDA_MEMCHECK	Checked each time a context is created
CUDA_MEMORY_LOG	Checked each time a context is created
CUDA_VISIBLE_DEVICES

Maps

Ordered from highest to lowest locations in x86 memory. These are architecture-, and to a lesser degree driver- and kernel version-specific. Applications and libraries can of course create many more maps than these.

• vsyscalls. read-execute-private, very few pages, topmost area of memory, usually highest mapping
• VDSO. read-execute-private, one page, high in memory (SYSENTER/SYSEXIT)
• Userspace stack. read-write-private, many pages, high in memory
• Anonymous map, 3 read-write-private pages, high in memory.
  • Possibly associated with nvidia driver's NV_STACK_SIZE stack. read-write-private, (3 * 4096 on amd64, 2 * 4096 on i686)
• Two sets of /dev/nvidiaX maps for each bound device. Sets are usually continguous, and contain:
  • an anonymous page, read-write-private
  • several mappings of the device, having variable number of pages, all read-write-shared
• Libraries. variable, middle of memory.
• Userspace heap. read-write-private, many pages, low in memory
• Application (data region). read-write-private, variable, low in memory
• Application (text region). read-execute-private, variable, usually lowest mapping

mmap()s

offset	size	notes	Nouveau name	block range
reg_addr + 0x0000	0x2000	not mapped by libcuda	PMC functional block	0x000000--0x001fff
reg_addr + 0x9000	0x1000	[Rwxs] mapped in cuInit(). first mapping. per-device.	PTIMER functional block	0x009000--0x009fff
reg_addr + 0xc0a000 / 0xc0c000	0x1000	[RWxs] location is acquired from ioctl 4e	PFIFO command submission interface	0xc00000--0xcfffff

ioctls

An ioctl (on x86) is 32 bits. The following definition comes from linux/asm-generic/ioctl.h in a 2.6.34 kernel:

• Bit 31: Read?
• Bit 30: Write?
• Bits 29-16: Parameter size
• Bits 15-8: Type (module)
• Bits 7-0: Number (command)

Looking at the source of the 195.36.15 kernel driver's OS interface, we see that NVIDIA uses the standard ioctl-creation macros from ioctl.h, and can be expected to adhere to this format. The type code used (NV_IOCTL_MAGIC) is 'F' (0x46), which overlaps with the framebuffer ioctl range as registered in 2.6.34. We further see that only _IOWR() is used to declare ioctls, implying that the first two bits will always be '11'. Both of these deductions are borne out observing strace output of a CUDA process.

Code	Param size	Param location(s)	Driver API call sites	Notes
/dev/nvidiactl
0xc8 NV_ESC_CARD_INFO	0x600 (1536)	anonymous page	cuInit	Largest parameter by far. Possibly scaled? Shifted 3 bits left, this is 0x3000, the size of the amd64 anonymous mapping. More likely we support returning up to 32x 48-byte descriptors, and... Wants the first 32 bits to be 1, all others 0. ...this is most likely a mask indicating which card IDs we want information for! typedef struct nv_ioctl_card_info { NvU16 flags; /* see below */ NvU8 bus; /* bus number (PCI, AGP, etc) */ NvU8 slot; /* card slot */ NvU16 vendor_id; /* PCI vendor id */ NvU16 device_id; NvU16 interrupt_line; NvU64 reg_address NV_ALIGN_BYTES(8); NvU64 reg_size NV_ALIGN_BYTES(8); NvU64 fb_address NV_ALIGN_BYTES(8); NvU64 fb_size NV_ALIGN_BYTES(8); } nv_ioctl_card_info_t; Returns (all subsequent bytes are 0): 0x00010001 0x0cb110de 0x00000026 0x00000000 0xf2000000 0x00000000 0x01000000 0x00000000 0xe0000000 0x00000000 0x10000000 0x00000000 0x0001: flag (NV_IOCTL_CARD_INFO_FLAG_PRESENT) 0x0001: bus/slot 0x0cb110de: vendor + device IDs lspci -n: 01:00.0 0300: 10de:0cb1 (rev a2) lspci -t -v: \-[0000:00]-+-03.0-[01]--+-00.0 nVidia Corporation GT215 [GeForce GTS 360M] 0x26: IRQ line (here, #38) 0xf2000000 00000000: reg_address 0x01000000 00000000: reg_size 0xe0000000 00000000: fb_address 0x10000000 00000000: fb_size these are all system memory references, see /proc/iomem: e0000000-f30fffff : PCI Bus 0000:01 e0000000-efffffff : 0000:01:00.0 f0000000-f1ffffff : 0000:01:00.0 f2000000-f2ffffff : 0000:01:00.0 f2000000-f2ffffff : nvidia f3000000-f307ffff : 0000:01:00.0 f3080000-f3083fff : 0000:01:00.1 f3080000-f3083fff : ICH HD audio
0xca NV_ESC_ENV_INFO	0x004	anonymous page	cuInit	Seems to ignore input value. Writes result value (0x00000001). typedef struct nv_ioctl_env_info { NvU32 pat_supported; } nv_ioctl_env_info_t;
0xce NV_ESC_ALLOC_OS_EVENT	0x14
0xcf NV_ESC_FREE_OS_EVENT
0xd1 NV_ESC_STATUS_CODE
0xd2 NV_ESC_CHECK_VERSION_STR	0x048	stack	cuInit	Performed immediately following opening of the nvidiactl device typedef struct nv_ioctl_rm_api_version { NvU32 cmd; NvU32 reply; char versionString[NV_RM_API_VERSION_STRING_LENGTH]; } nv_ioctl_rm_api_version_t; #define NV_RM_API_VERSION_CMD_STRICT 0 #define NV_RM_API_VERSION_CMD_RELAXED '1' #define NV_RM_API_VERSION_CMD_OVERRIDE '2' #define NV_RM_API_VERSION_REPLY_UNRECOGNIZED 0 #define NV_RM_API_VERSION_REPLY_RECOGNIZED 1 0x312e 3633 2e35 3931 35ull == 195.36.15 '1' '.' '6' '3' '.' '5' '9' '1', '5' looks like: all version chars in ascii. first 8 reversed, then any left follow? All other bytes are 0. Writes result to first 8 bytes (0x00000001), leaves others untouched
0x22	0x00c	stack	cuInit	Inputs set to 0. Outputs (example): 3251635025 65 0 First value is used as first input word to the majority of subsequent ioctls Second value ranges over (at least) 41--65... Not sent in 256.22/3.10...
0x2a	0x020	stack	cuInit	Inputs: 0x7fffffffd310: 3251635025 3251635025 533 0 0x7fffffffd320: 4294955888 32767 132 0 Outputs are unchanged
0x2b	0x020	stack	cuInit
0x4d	0x048	stack	cuInit	Performed following opening of nvidiaX device
0x2d	0x014	stack	cuInit	Performed following read of /proc/interrupts
0x4e	0x030		cuInit	Immediately prior to first mmap()
0x4f	0x020		cuInit	Invoked if mmap() returns MAP_FAILED, prior to failing out
0x54	0x30
0x57	0x038
0x58	0x28
0x59	0x10
/dev/nvidiaX
0x32	0x014	stack	cuInit	Performed several times in succession
0x37	0x020	stack	cuInit	Follows burst of 3x 0x32's, then interwoven with bursts of 2a's

disassembly

This disassembly makes use of libcuda.so.195.36.15 (0867d66be617faab3782fa0ba19ec9ba, 7404990 bytes). Symbols were extracted via objdump -T. AMD64 ABI:

• Integer arguments via RDI, RSI, RDX, RCX, R8 and R9, then stack
• FP arguments in XMM0..XMM7, then stack
• Return value in RAX

location (.text is 000000000007fd10)	length	symbol	notes
ebce0			ebce0: 41 55 push %r13 ebce2: 49 89 fd mov %rdi,%r13 ebce5: 41 54 push %r12 ebce7: 49 89 f4 mov %rsi,%r12 ebcea: 55 push %rbp ebceb: 53 push %rbx ebcec: 48 83 ec 08 sub $0x8,%rsp ebcf0: 8b 35 2a 49 72 00 mov 0x72492a(%rip),%esi # 810620 <__isinff@plt+0x790920> ebcf6: 85 f6 test %esi,%esi ebcf8: 0f 84 16 01 00 00 je ebe14 <__isinff@plt+0x6c114> ebcfe: 8b 0d 2c 49 72 00 mov 0x72492c(%rip),%ecx # 810630 <__isinff@plt+0x790930> ebd04: 85 c9 test %ecx,%ecx ebd06: 0f 8e 08 01 00 00 jle ebe14 <__isinff@plt+0x6c114> ebd0c: 48 8d 1d 0d 49 72 00 lea 0x72490d(%rip),%rbx # 810620 <__isinff@plt+0x790920> ebd13: 31 ed xor %ebp,%ebp ebd15: 48 8b 7b 28 mov 0x28(%rbx),%rdi ebd19: 4c 89 e2 mov %r12,%rdx ebd1c: 4c 89 ee mov %r13,%rsi ebd1f: ff c5 inc %ebp ebd21: ff 53 20 callq *0x20(%rbx) ebd24: 48 83 c3 18 add $0x18,%rbx ebd28: 39 2d 02 49 72 00 cmp %ebp,0x724902(%rip) # 810630 <__isinff@plt+0x790930> ebd2e: 0f 8e e0 00 00 00 jle ebe14 <__isinff@plt+0x6c114> ebd34: 48 8b 7b 28 mov 0x28(%rbx),%rdi ebd38: 4c 89 e2 mov %r12,%rdx ebd3b: 4c 89 ee mov %r13,%rsi ebd3e: ff 53 20 callq *0x20(%rbx) ebd41: 8d 45 01 lea 0x1(%rbp),%eax ebd44: 39 05 e6 48 72 00 cmp %eax,0x7248e6(%rip) # 810630 <__isinff@plt+0x790930> ebd4a: 48 8d 4b 18 lea 0x18(%rbx),%rcx ebd4e: 0f 8e c0 00 00 00 jle ebe14 <__isinff@plt+0x6c114> ebd54: 48 8b 79 28 mov 0x28(%rcx),%rdi ebd58: 4c 89 e2 mov %r12,%rdx ebd5b: 4c 89 ee mov %r13,%rsi ebd5e: ff 51 20 callq *0x20(%rcx) ebd61: 8d 55 02 lea 0x2(%rbp),%edx ebd64: 39 15 c6 48 72 00 cmp %edx,0x7248c6(%rip) # 810630 <__isinff@plt+0x790930> ebd6a: 48 8d 4b 30 lea 0x30(%rbx),%rcx ebd6e: 0f 8e a0 00 00 00 jle ebe14 <__isinff@plt+0x6c114> ebd74: 48 8b 79 28 mov 0x28(%rcx),%rdi ebd78: 4c 89 e2 mov %r12,%rdx ebd7b: 4c 89 ee mov %r13,%rsi ebd7e: ff 51 20 callq *0x20(%rcx) ebd81: 8d 7d 03 lea 0x3(%rbp),%edi ebd84: 39 3d a6 48 72 00 cmp %edi,0x7248a6(%rip) # 810630 <__isinff@plt+0x790930> ebd8a: 48 8d 4b 48 lea 0x48(%rbx),%rcx ebd8e: 0f 8e 80 00 00 00 jle ebe14 <__isinff@plt+0x6c114> ebd94: 48 8b 79 28 mov 0x28(%rcx),%rdi ebd98: 4c 89 e2 mov %r12,%rdx ebd9b: 4c 89 ee mov %r13,%rsi ebd9e: ff 51 20 callq *0x20(%rcx) ebda1: 44 8d 45 04 lea 0x4(%rbp),%r8d ebda5: 44 39 05 84 48 72 00 cmp %r8d,0x724884(%rip) # 810630 <__isinff@plt+0x790930> ebdac: 48 8d 4b 60 lea 0x60(%rbx),%rcx ebdb0: 7e 62 jle ebe14 <__isinff@plt+0x6c114> ebdb2: 48 8b 79 28 mov 0x28(%rcx),%rdi ebdb6: 4c 89 e2 mov %r12,%rdx ebdb9: 4c 89 ee mov %r13,%rsi ebdbc: ff 51 20 callq *0x20(%rcx) ebdbf: 44 8d 4d 05 lea 0x5(%rbp),%r9d ebdc3: 44 39 0d 66 48 72 00 cmp %r9d,0x724866(%rip) # 810630 <__isinff@plt+0x790930> ebdca: 48 8d 4b 78 lea 0x78(%rbx),%rcx ebdce: 7e 44 jle ebe14 <__isinff@plt+0x6c114> ebdd0: 48 8b 79 28 mov 0x28(%rcx),%rdi ebdd4: 4c 89 e2 mov %r12,%rdx ebdd7: 4c 89 ee mov %r13,%rsi ebdda: ff 51 20 callq *0x20(%rcx) ebddd: 44 8d 55 06 lea 0x6(%rbp),%r10d ebde1: 44 39 15 48 48 72 00 cmp %r10d,0x724848(%rip) # 810630 <__isinff@plt+0x790930> ebde8: 48 8d 8b 90 00 00 00 lea 0x90(%rbx),%rcx ebdef: 7e 23 jle ebe14 <__isinff@plt+0x6c114> ebdf1: 48 8b 79 28 mov 0x28(%rcx),%rdi ebdf5: 4c 89 e2 mov %r12,%rdx ebdf8: 4c 89 ee mov %r13,%rsi ebdfb: 83 c5 07 add $0x7,%ebp ebdfe: 48 81 c3 a8 00 00 00 add $0xa8,%rbx ebe05: ff 51 20 callq *0x20(%rcx) ebe08: 39 2d 22 48 72 00 cmp %ebp,0x724822(%rip) # 810630 <__isinff@plt+0x790930> ebe0e: 0f 8f 01 ff ff ff jg ebd15 <__isinff@plt+0x6c015> ebe14: 48 83 c4 08 add $0x8,%rsp ebe18: 5b pop %rbx ebe19: 5d pop %rbp ebe1a: 41 5c pop %r12 ebe1c: 41 5d pop %r13 ebe1e: c3 retq
5c7850		Pthread_setspecific()	5c7850: ff cf dec %edi 5c7852: e9 e9 7c ab ff jmpq 7f540 <pthread_setspecific@plt>
5c7860		Pthread_getspecific()	5c7860: ff cf dec %edi 5c7862: e9 49 83 ab ff jmpq 7fbb0 <pthread_getspecific@plt>
905d0		"lookupContext()"	// Check for 0x321cbda00, the "driver deinitialized" magic constant. Return 4 if set. 905d0: 48 8b 05 91 e0 77 00 mov 0x77e091(%rip),%rax # 80e668 <__isinff@plt+0x78e968> push %rbx 905d8: b9 04 00 00 00 mov $0x4,%ecx 905dd: 48 89 fb mov %rdi,%rbx 905e0: 8b 90 88 01 00 00 mov 0x188(%rax),%edx 905e6: 81 fa 00 ba 1c 32 cmp $0x321cba00,%edx 905ec: 74 0a je 905f8 <__isinff@plt+0x108f8> // Check for 0xabc123, the "driver initialized" magic constant. Return 3 if unset. 905ee: 81 fa 23 c1 ab 00 cmp $0xabc123,%edx 905f4: b1 03 mov $0x3,%cl 905f6: 74 08 je 90600 <__isinff@plt+0x10900> 905f8: 5b pop %rbx 905f9: 89 c8 mov %ecx,%eax 905fb: c3 retq 905fc: 66 66 66 90 xchg %ax,%ax // Also return 3 (DRIVER_UNINITIALIZED) if we were passed an uninitialized key. 90600: 30 c9 xor %cl,%cl 90602: 48 85 ff test %rdi,%rdi 90605: 74 f1 je 905f8 <__isinff@plt+0x108f8> // Call pthread_getspecific() 90607: 8b 38 mov (%rax),%edi 90609: e8 52 72 53 00 callq 5c7860 <__isinff@plt+0x547b60> // Check for NULL from pthread_getspecific(). Return 201 (INVALID_CONTEXT) in that case. 9060e: 48 85 c0 test %rax,%rax 90611: 48 89 03 mov %rax,(%rbx) 90614: b9 c9 00 00 00 mov $0xc9,%ecx 90619: 74 dd je 905f8 <__isinff@plt+0x108f8> // We had a valid context object. Return 0x338 bytes into it. 9061b: 8b 88 38 03 00 00 mov 0x338(%rax),%ecx 90621: 5b pop %rbx 90622: 89 c8 mov %ecx,%eax 90624: c3 retq
ebcd0			ebcd0: 31 c0 xor %eax,%eax ebcd2: 83 3d 4b 49 72 00 00 cmpl $0x0,0x72494b(%rip) # 810624 <__isinff@plt+0x790924> ebcd9: 0f 95 c0 setne %al ebcdc: c3 retq
81460			81460: 53 push %rbx 81461: 89 fb mov %edi,%ebx 81463: e8 b8 f2 00 00 callq 90720 <__isinff@plt+0x10a20> 81468: 85 db test %ebx,%ebx 8146a: 74 07 je 81473 <__isinff@plt+0x1773> 8146c: 5b pop %rbx 8146d: b8 01 00 00 00 mov $0x1,%eax 81472: c3 retq 81473: 5b pop %rbx 81474: e9 07 f4 00 00 jmpq 90880 <__isinff@plt+0x10b80>
90720			90720: 53 push %rbx 90721: 48 8b 1d 40 df 77 00 mov 0x77df40(%rip),%rbx # 80e668 <__isinff@plt+0x78e968> 90728: 8b 83 80 01 00 00 mov 0x180(%rbx),%eax 9072e: 85 c0 test %eax,%eax 90730: 74 02 je 90734 <__isinff@plt+0x10a34> 90732: 5b pop %rbx 90733: c3 retq 90734: 48 8d bb 78 01 00 00 lea 0x178(%rbx),%rdi 9073b: 31 d2 xor %edx,%edx 9073d: be 01 00 00 00 mov $0x1,%esi 90742: e8 b9 69 53 00 callq 5c7100 <__isinff@plt+0x547400> 90747: 85 c0 test %eax,%eax 90749: 0f 84 9a 00 00 00 je 907e9 <__isinff@plt+0x10ae9> 9074f: 44 8b 9b 80 01 00 00 mov 0x180(%rbx),%r11d 90756: 45 85 db test %r11d,%r11d 90759: 75 d7 jne 90732 <__isinff@plt+0x10a32> 9075b: e8 20 70 53 00 callq 5c7780 <__isinff@plt+0x547a80> 90760: 8b 93 80 01 00 00 mov 0x180(%rbx),%edx 90766: 85 d2 test %edx,%edx 90768: 75 c8 jne 90732 <__isinff@plt+0x10a32> 9076a: e8 11 70 53 00 callq 5c7780 <__isinff@plt+0x547a80> 9076f: 8b 8b 80 01 00 00 mov 0x180(%rbx),%ecx 90775: 85 c9 test %ecx,%ecx 90777: 75 b9 jne 90732 <__isinff@plt+0x10a32> 90779: e8 02 70 53 00 callq 5c7780 <__isinff@plt+0x547a80> 9077e: 8b b3 80 01 00 00 mov 0x180(%rbx),%esi 90784: 85 f6 test %esi,%esi 90786: 75 aa jne 90732 <__isinff@plt+0x10a32> 90788: e8 f3 6f 53 00 callq 5c7780 <__isinff@plt+0x547a80> 9078d: 8b bb 80 01 00 00 mov 0x180(%rbx),%edi 90793: 85 ff test %edi,%edi 90795: 75 9b jne 90732 <__isinff@plt+0x10a32> 90797: e8 e4 6f 53 00 callq 5c7780 <__isinff@plt+0x547a80> 9079c: 44 8b 83 80 01 00 00 mov 0x180(%rbx),%r8d 907a3: 45 85 c0 test %r8d,%r8d 907a6: 75 8a jne 90732 <__isinff@plt+0x10a32> 907a8: e8 d3 6f 53 00 callq 5c7780 <__isinff@plt+0x547a80> 907ad: 44 8b 8b 80 01 00 00 mov 0x180(%rbx),%r9d 907b4: 45 85 c9 test %r9d,%r9d 907b7: 0f 85 75 ff ff ff jne 90732 <__isinff@plt+0x10a32> 907bd: e8 be 6f 53 00 callq 5c7780 <__isinff@plt+0x547a80> 907c2: 44 8b 93 80 01 00 00 mov 0x180(%rbx),%r10d 907c9: 45 85 d2 test %r10d,%r10d 907cc: 0f 85 60 ff ff ff jne 90732 <__isinff@plt+0x10a32> 907d2: e8 a9 6f 53 00 callq 5c7780 <__isinff@plt+0x547a80> 907d7: 44 8b 9b 80 01 00 00 mov 0x180(%rbx),%r11d 907de: 45 85 db test %r11d,%r11d 907e1: 0f 84 74 ff ff ff je 9075b <__isinff@plt+0x10a5b> 907e7: 5b pop %rbx 907e8: c3 retq
0x90880			90880: 41 54 push %r12 90882: 55 push %rbp 90883: 53 push %rbx 90884: 31 db xor %ebx,%ebx 90886: 48 81 ec 00 08 00 00 sub $0x800,%rsp 9088d: e8 8e fe ff ff callq 90720 <__isinff@plt+0x10a20> 90892: 48 8b 2d cf dd 77 00 mov 0x77ddcf(%rip),%rbp # 80e668 <__isinff@plt+0x78e968>8 lea 0x8(%rbp),%rdi 9089d: e8 1e 6f 53 00 callq 5c77c0 <__isinff@plt+0x547ac0> 908a2: 81 bd 88 01 00 00 23 cmpl $0xabc123,0x188(%rbp) 908a9: c1 ab 00 908ac: 0f 84 16 01 00 00 je 909c8 <__isinff@plt+0x10cc8> 908b2: 4c 8d a4 24 00 04 00 lea 0x400(%rsp),%r12 908b9: 00 908ba: 48 8d 3d eb 77 53 00 lea 0x5377eb(%rip),%rdi # 5c80ac <__isinff@plt+0x5483ac>0 00 mov $0x400,%edx 908c6: c6 84 24 00 04 00 00 movb $0x0,0x400(%rsp) 908cd: 00 908ce: c6 04 24 00 movb $0x0,(%rsp) 908d2: 4c 89 e6 mov %r12,%rsi 908d5: e8 76 70 53 00 callq 5c7950 <__isinff@plt+0x547c50> 908da: 48 8d 3d cb 76 53 00 lea 0x5376cb(%rip),%rdi # 5c7fac <__isinff@plt+0x5482ac> mov %rsp,%rsi 908e4: ba 00 04 00 00 mov $0x400,%edx 908e9: e8 62 70 53 00 callq 5c7950 <__isinff@plt+0x547c50> 908ee: 80 3c 24 00 cmpb $0x0,(%rsp) 908f2: 0f 85 e8 00 00 00 jne 909e0 <__isinff@plt+0x10ce0> 908f8: c7 85 c0 03 00 00 00 movl $0x0,0x3c0(%rbp) 908ff: 00 00 00 90902: e8 89 6e 53 00 callq 5c7790 <__isinff@plt+0x547a90> 90907: 48 8d bd a0 01 00 00 lea 0x1a0(%rbp),%rdi 9090e: 31 f6 xor %esi,%esi 90910: ba 00 01 00 00 mov $0x100,%edx 90915: 89 85 90 04 00 00 mov %eax,0x490(%rbp) 9091b: c7 85 a0 04 00 00 00 movl $0x0,0x4a0(%rbp) 90922: 00 00 00 90925: 48 c7 85 98 04 00 00 movq $0x0,0x498(%rbp) 9092c: 00 00 00 00 90930: c7 85 b0 03 00 00 00 movl $0x0,0x3b0(%rbp) 90937: 00 00 00 9093a: c7 85 a0 02 00 00 00 movl $0x0,0x2a0(%rbp) 90941: 00 00 00 90944: c7 85 94 01 00 00 00 movl $0x0,0x194(%rbp) 9094b: 00 00 00 9094e: e8 3d f0 fe ff callq 7f990 <memset@plt> 90953: 48 8d bd c0 04 00 00 lea 0x4c0(%rbp),%rdi 9095a: e8 b1 c4 00 00 callq 9ce10 <__isinff@plt+0x1d110> 9095f: e8 cc 1e 07 00 callq 102830 <__isinff@plt+0x82b30> 90964: 85 c0 test %eax,%eax 90966: 89 c3 mov %eax,%ebx 90968: 0f 84 9a 00 00 00 je 90a08 <__isinff@plt+0x10d08> 9096e: 48 8b bd b8 03 00 00 mov 0x3b8(%rbp),%rdi 90975: 48 85 ff test %rdi,%rdi 90978: 74 10 je 9098a <__isinff@plt+0x10c8a> 9097a: e8 d1 a8 00 00 callq 9b250 <__isinff@plt+0x1b550> 9097f: 48 c7 85 b8 03 00 00 movq $0x0,0x3b8(%rbp) 90986: 00 00 00 00 9098a: 85 db test %ebx,%ebx 9098c: 75 72 jne 90a00 <__isinff@plt+0x10d00> 9098e: c7 85 88 01 00 00 23 movl $0xabc123,0x188(%rbp) 90995: c1 ab 00 90998: c7 85 94 04 00 00 00 movl $0x0,0x494(%rbp) 9099f: 00 00 00 909a2: e8 79 b4 05 00 callq ebe20 <__isinff@plt+0x6c120> 909a7: e8 24 b3 05 00 callq ebcd0 <__isinff@plt+0x6bfd0> 909ac: 84 c0 test %al,%al 909ae: 0f 85 ac 00 00 00 jne 90a60 <__isinff@plt+0x10d60> 909b4: e8 b7 20 07 00 callq 102a70 <__isinff@plt+0x82d70> 909b9: 31 db xor %ebx,%ebx 909bb: e8 10 b3 05 00 callq ebcd0 <__isinff@plt+0x6bfd0> 909c0: 84 c0 test %al,%al 909c2: 0f 85 7f 00 00 00 jne 90a47 <__isinff@plt+0x10d47> 909c8: 48 8d 7d 08 lea 0x8(%rbp),%rdi 909cc: e8 df 6d 53 00 callq 5c77b0 <__isinff@plt+0x547ab0> 909d1: 48 81 c4 00 08 00 00 add $0x800,%rsp 909d8: 89 d8 mov %ebx,%eax 909da: 5b pop %rbx 909db: 5d pop %rbp 909dc: 41 5c pop %r12 909de: c3 retq 909df: 90 nop 909e0: 80 bc 24 00 04 00 00 cmpb $0x0,0x400(%rsp) 909e7: 00 909e8: 0f 84 0a ff ff ff je 908f8 <__isinff@plt+0x10bf8> 909ee: c7 85 c0 03 00 00 01 movl $0x1,0x3c0(%rbp) 909f5: 00 00 00 909f8: e9 05 ff ff ff jmpq 90902 <__isinff@plt+0x10c02> 909fd: 66 66 90 xchg %ax,%ax 90a00: 89 9d 88 01 00 00 mov %ebx,0x188(%rbp) 90a06: eb c0 jmp 909c8 <__isinff@plt+0x10cc8> 90a08: 48 8b 3d 19 e5 77 00 mov 0x77e519(%rip),%rdi # 80ef28 <__isinff@plt+0x78f228> 90a0f: b3 02 mov $0x2,%bl 90a11: e8 6a 6e 53 00 callq 5c7880 <__isinff@plt+0x547b80> 90a16: 85 c0 test %eax,%eax 90a18: 89 45 00 mov %eax,0x0(%rbp) 90a1b: 0f 84 4d ff ff ff je 9096e <__isinff@plt+0x10c6e> 90a21: be 01 00 00 00 mov $0x1,%esi 90a26: 4c 89 e7 mov %r12,%rdi 90a29: e8 a2 66 53 00 callq 5c70d0 <__isinff@plt+0x5473d0> 90a2e: f6 84 24 0a 04 00 00 testb $0x8,0x40a(%rsp) 90a35: 08 90a36: 75 3b jne 90a73 <__isinff@plt+0x10d73> 90a38: e8 23 bc 01 00 callq ac660 <__isinff@plt+0x2c960> 90a3d: e8 ce b5 05 00 callq ec010 <__isinff@plt+0x6c310> 90a42: e9 47 ff ff ff jmpq 9098e <__isinff@plt+0x10c8e> 90a47: 48 8b 3d 0a cd 77 00 mov 0x77cd0a(%rip),%rdi # 80d758 <__isinff@plt+0x78da58> 90a4e: 31 f6 xor %esi,%esi 90a50: e8 8b b2 05 00 callq ebce0 <__isinff@plt+0x6bfe0> 90a55: e9 6e ff ff ff jmpq 909c8 <__isinff@plt+0x10cc8> 90a5a: 66 66 90 xchg %ax,%ax 90a5d: 66 66 90 xchg %ax,%ax 90a60: 48 8b 3d b1 dd 77 00 mov 0x77ddb1(%rip),%rdi # 80e818 <__isinff@plt+0x78eb18> 90a67: 31 f6 xor %esi,%esi 90a69: e8 72 b2 05 00 callq ebce0 <__isinff@plt+0x6bfe0> 90a6e: e9 41 ff ff ff jmpq 909b4 <__isinff@plt+0x10cb4> 90a73: 83 8d a4 04 00 00 01 orl $0x1,0x4a4(%rbp) 90a7a: eb bc jmp 90a38 <__isinff@plt+0x10d38> 90a7c: 66 66 66 90 xchg %ax,%ax 90a80: 53 push %rbx 90a81: 48 8b 1d e0 db 77 00 mov 0x77dbe0(%rip),%rbx # 80e668 <__isinff@plt+0x78e968> 90a88: 8b 3b mov (%rbx),%edi 90a8a: 85 ff test %edi,%edi 90a8c: 75 12 jne 90aa0 <__isinff@plt+0x10da0> 90a8e: c7 83 88 01 00 00 00 movl $0x321cba00,0x188(%rbx) 90a95: ba 1c 32 90a98: 5b pop %rbx 90a99: c3 retq
0x10a6a0	0x18c	cuCtxSynchronize
0x10a830	0x253	cuInit	10a830: 48 89 5c 24 e0 mov %rbx,-0x20(%rsp) 10a835: 48 89 6c 24 e8 mov %rbp,-0x18(%rsp) 10a83a: 89 fd mov %edi,%ebp 10a83c: 4c 89 64 24 f0 mov %r12,-0x10(%rsp) 10a841: 4c 89 6c 24 f8 mov %r13,-0x8(%rsp) 10a846: 48 81 ec 88 00 00 00 sub $0x88,%rsp 10a84d: 48 c7 44 24 48 00 00 movq $0x0,0x48(%rsp) 10a854: 00 00 10a856: e8 75 14 fe ff callq ebcd0 <__isinff@plt+0x6bfd0> 10a85b: 84 c0 test %al,%al 10a85d: 89 c3 mov %eax,%ebx 10a85f: 0f 85 52 01 00 00 jne 10a9b7 <__isinff@plt+0x8acb7> 10a865: 45 31 ed xor %r13d,%r13d 10a868: 44 8b 0d a5 5f 70 00 mov 0x705fa5(%rip),%r9d # 810814 <__isinff@plt+0x790b14> 10a86f: 45 85 c9 test %r9d,%r9d 10a872: 0f 84 08 01 00 00 je 10a980 <__isinff@plt+0x8ac80> // first-time initialization sequence returns to here 10a878: 89 ef mov %ebp,%edi 10a87a: e8 e1 6b f7 ff callq 81460 <__isinff@plt+0x1760> 10a87f: 44 8b 15 8e 5f 70 00 mov 0x705f8e(%rip),%r10d # 810814 <__isinff@plt+0x790b14> 10a886: 41 89 c4 mov %eax,%r12d 10a889: 45 85 d2 test %r10d,%r10d 10a88c: 0f 84 b4 00 00 00 je 10a946 <__isinff@plt+0x8ac46> 10a892: 84 db test %bl,%bl 10a894: 0f 95 c3 setne %bl 10a897: 44 84 eb test %r13b,%bl 10a89a: 75 24 jne 10a8c0 <__isinff@plt+0x8abc0> 10a89c: 44 89 e0 mov %r12d,%eax 10a89f: 48 8b 5c 24 68 mov 0x68(%rsp),%rbx 10a8a4: 48 8b 6c 24 70 mov 0x70(%rsp),%rbp 10a8a9: 4c 8b 64 24 78 mov 0x78(%rsp),%r12 10a8ae: 4c 8b ac 24 80 00 00 mov 0x80(%rsp),%r13 10a8b5: 00 10a8b6: 48 81 c4 88 00 00 00 add $0x88,%rsp 10a8bd: c3 retq 10a8be: 66 90 xchg %ax,%ax 10a8c0: fc cld 10a8c1: 31 c0 xor %eax,%eax 10a8c3: b9 08 00 00 00 mov $0x8,%ecx 10a8c8: 48 89 e7 mov %rsp,%rdi 10a8cb: f3 48 ab rep stos %rax,%es:(%rdi) 10a8ce: 4c 8b 2d 93 3d 70 00 mov 0x703d93(%rip),%r13 # 80e668 <__isinff@plt+0x78e968> 10a8d5: 89 6c 24 50 mov %ebp,0x50(%rsp) 10a8d9: 48 8d 6c 24 50 lea 0x50(%rsp),%rbp 10a8de: 41 8b 7d 00 mov 0x0(%r13),%edi 10a8e2: e8 79 cf 4b 00 callq 5c7860 <__isinff@plt+0x547b60> 10a8e7: 4c 8d 1d 75 0a 4d 00 lea 0x4d0a75(%rip),%r11 # 5db363 <__isinff@plt+0x55b663> 10a8ee: 48 89 c2 mov %rax,%rdx 10a8f1: 48 89 04 24 mov %rax,(%rsp) 10a8f5: 31 c0 xor %eax,%eax 10a8f7: 48 85 d2 test %rdx,%rdx 10a8fa: 48 c7 44 24 08 00 00 movq $0x0,0x8(%rsp) 10a901: 00 00 10a903: c7 44 24 10 01 00 00 movl $0x1,0x10(%rsp) 10a90a: 00 10a90b: 4c 89 5c 24 18 mov %r11,0x18(%rsp) 10a910: 48 89 6c 24 20 mov %rbp,0x20(%rsp) 10a915: 74 07 je 10a91e <__isinff@plt+0x8ac1e> 10a917: 48 8b 82 f0 03 00 00 mov 0x3f0(%rdx),%rax 10a91e: 48 8b 74 24 48 mov 0x48(%rsp),%rsi 10a923: 48 8b 3d 16 43 70 00 mov 0x704316(%rip),%rdi # 80ec40 <__isinff@plt+0x78ef40> 10a92a: 48 89 44 24 28 mov %rax,0x28(%rsp) 10a92f: 44 89 64 24 38 mov %r12d,0x38(%rsp) 10a934: 48 89 74 24 30 mov %rsi,0x30(%rsp) 10a939: 48 89 e6 mov %rsp,%rsi 10a93c: e8 9f 13 fe ff callq ebce0 <__isinff@plt+0x6bfe0> 10a941: e9 56 ff ff ff jmpq 10a89c <__isinff@plt+0x8ab9c> 10a946: 48 8d 3d 8d 00 4d 00 lea 0x4d008d(%rip),%rdi # 5da9da <__isinff@plt+0x55acda> 10a94d: e8 0e 4b f7 ff callq 7f460 <getenv@plt> 10a952: 48 85 c0 test %rax,%rax 10a955: 74 17 je 10a96e <__isinff@plt+0x8ac6e> 10a957: 31 c9 xor %ecx,%ecx 10a959: ba 0a 00 00 00 mov $0xa,%edx 10a95e: 31 f6 xor %esi,%esi 10a960: 48 89 c7 mov %rax,%rdi 10a963: e8 c8 4f f7 ff callq 7f930 <__strtol_internal@plt> 10a968: 89 05 aa 5e 70 00 mov %eax,0x705eaa(%rip) # 810818 <__isinff@plt+0x790b18> 10a96e: c7 05 9c 5e 70 00 01 movl $0x1,0x705e9c(%rip) # 810814 <__isinff@plt+0x790b14> 10a975: 00 00 00 10a978: e9 15 ff ff ff jmpq 10a892 <__isinff@plt+0x8ab92> 10a97d: 66 66 90 xchg %ax,%ax // first-time initialization sequence. 10a980: 48 8d 3d 53 00 4d 00 lea 0x4d0053(%rip),%rdi # 5da9da <__isinff@plt+0x55acda> "CUDA_API_TRACE_PTR" 10a987: e8 d4 4a f7 ff callq 7f460 <getenv@plt> 10a98c: 48 85 c0 test %rax,%rax 10a98f: 74 17 je 10a9a8 <__isinff@plt+0x8aca8> // we go to 10a9a8 if CUDA_API_TRACE_PTR is unset 10a991: 31 c9 xor %ecx,%ecx 10a993: ba 0a 00 00 00 mov $0xa,%edx 10a998: 31 f6 xor %esi,%esi 10a99a: 48 89 c7 mov %rax,%rdi 10a99d: e8 8e 4f f7 ff callq 7f930 <__strtol_internal@plt> 10a9a2: 89 05 70 5e 70 00 mov %eax,0x705e70(%rip) # 810818 <__isinff@plt+0x790b18> // set cudaInitialized to 1 10a9a8: c7 05 62 5e 70 00 01 movl $0x1,0x705e62(%rip) # 810814 <__isinff@plt+0x790b14> 10a9af: 00 00 00 10a9b2: e9 c1 fe ff ff jmpq 10a878 <__isinff@plt+0x8ab78> 10a9b7: 31 ff xor %edi,%edi 10a9b9: e8 12 5c f8 ff callq 905d0 <__isinff@plt+0x108d0> 10a9be: 85 c0 test %eax,%eax 10a9c0: 0f 85 9f fe ff ff jne 10a865 <__isinff@plt+0x8ab65> 10a9c6: fc cld 10a9c7: 31 c0 xor %eax,%eax 10a9c9: b9 07 00 00 00 mov $0x7,%ecx 10a9ce: 48 89 e7 mov %rsp,%rdi 10a9d1: f3 48 ab rep stos %rax,%es:(%rdi) 10a9d4: 48 8b 05 8d 3c 70 00 mov 0x703c8d(%rip),%rax # 80e668 <__isinff@plt+0x78e968> 10a9db: 89 6c 24 50 mov %ebp,0x50(%rsp) 10a9df: 8b 38 mov (%rax),%edi 10a9e1: e8 7a ce 4b 00 callq 5c7860 <__isinff@plt+0x547b60> 10a9e6: 48 85 c0 test %rax,%rax 10a9e9: 48 89 04 24 mov %rax,(%rsp) 10a9ed: 74 69 je 10aa58 <__isinff@plt+0x8ad58> 10a9ef: 48 ff 80 f0 03 00 00 incq 0x3f0(%rax) 10a9f6: 48 8d 54 24 50 lea 0x50(%rsp),%rdx 10a9fb: 48 8d 35 61 09 4d 00 lea 0x4d0961(%rip),%rsi # 5db363 <__isinff@plt+0x55b663> 10aa02: 48 8b 04 24 mov (%rsp),%rax 10aa06: 48 c7 44 24 08 00 00 movq $0x0,0x8(%rsp) 10aa0d: 00 00 10aa0f: 48 89 54 24 20 mov %rdx,0x20(%rsp) 10aa14: 31 d2 xor %edx,%edx 10aa16: c7 44 24 10 01 00 00 movl $0x1,0x10(%rsp) 10aa1d: 00 10aa1e: 48 89 74 24 18 mov %rsi,0x18(%rsp) 10aa23: 48 85 c0 test %rax,%rax 10aa26: 74 07 je 10aa2f <__isinff@plt+0x8ad2f> 10aa28: 48 8b 90 f0 03 00 00 mov 0x3f0(%rax),%rdx 10aa2f: 48 8b 3d ea 35 70 00 mov 0x7035ea(%rip),%rdi # 80e020 <__isinff@plt+0x78e320> 10aa36: 4c 8d 44 24 48 lea 0x48(%rsp),%r8 10aa3b: 48 89 e6 mov %rsp,%rsi 10aa3e: 41 bd 01 00 00 00 mov $0x1,%r13d 10aa44: 48 89 54 24 28 mov %rdx,0x28(%rsp) 10aa49: 4c 89 44 24 30 mov %r8,0x30(%rsp) 10aa4e: e8 8d 12 fe ff callq ebce0 <__isinff@plt+0x6bfe0> 10aa53: e9 10 fe ff ff jmpq 10a868 <__isinff@plt+0x8ab68> 10aa58: 4c 8d 05 04 09 4d 00 lea 0x4d0904(%rip),%r8 # 5db363 <__isinff@plt+0x55b663> 10aa5f: 4c 8d 64 24 50 lea 0x50(%rsp),%r12 10aa64: 31 d2 xor %edx,%edx 10aa66: 48 c7 44 24 08 00 00 movq $0x0,0x8(%rsp) 10aa6d: 00 00 10aa6f: c7 44 24 10 01 00 00 movl $0x1,0x10(%rsp) 10aa76: 00 10aa77: 4c 89 44 24 18 mov %r8,0x18(%rsp) 10aa7c: 4c 89 64 24 20 mov %r12,0x20(%rsp) 10aa81: eb ac jmp 10aa2f <__isinff@plt+0x8ad2f> 10aa83: 66 66 66 90 xchg %ax,%ax 10aa87: 66 66 90 xchg %ax,%ax 10aa8a: 66 66 90 xchg %ax,%ax 10aa8d: 66 66 90 xchg %ax,%ax
0x10aa90	0x272	cuGetExportTable
0x10ad10	0x286	cuGraphicsUnmapResources
0x10afa0	0x286	cuGraphicsMapResources
0x10b230	0x272	cuGraphicsResourceSetMapFlags
0x10b4b0	0x28e	cuGraphicsResourceGetMappedPointer
0x10b740	0x2b2	cuGraphicsSubResourceGetMappedArray
0x10ba00	0x24d	cuGraphicsUnregisterResource
0x10bc50	0x245	cuStreamDestroy
0x10bea0	0x245	cuStreamSynchronize
0x10c0f0	0x245	cuStreamQuery
0x10c340	0x272	cuStreamCreate
0x10c5c0	0x28e	cuEventElapsedTime
0x10c850	0x24d	cuEventDestroy
0x10caa0	0x24d	cuEventSynchronize
0x10ccf0	0x24d	cuEventQuery
0x10cf40	0x261	cuEventRecord
0x10d1b0	0x272	cuEventCreate
0x10d430	0x2aa	cuLaunchGridAsync
0x10d6e0	0x28e	cuLaunchGrid
0x10d970	0x24d	cuLaunch
0x10dbc0	0x28e	cuParamSetTexRef
0x10de50	0x2b9	cuParamSetv
0x10e110	0x2b5	cuParamSetf
0x10e3d0	0x28e	cuParamSeti
0x10e660	0x272	cuParamSetSize
0x10e8e0	0x272	cuTexRefGetFlags
0x10eb60	0x28e	cuTexRefGetFormat
0x10edf0	0x272	cuTexRefGetFilterMode
0x10f070	0x28e	cuTexRefGetAddressMode
0x10f300	0x272	cuTexRefGetArray
0x10f580	0x272	cuTexRefGetAddress
0x10f800	0x272	cuTexRefSetFlags
0x10fa80	0x272	cuTexRefSetFilterMode
0x10fd00	0x28e	cuTexRefSetAddressMode
0x10ff90	0x28e	cuTexRefSetFormat
0x11a0d0	0x28e	cuCtxCreate
0x11a360	0x28e	cuDeviceGetAttribute
0x11a5f0	0x272	cuDeviceGetProperties
0x11a870	0x272	cuDeviceTotalMem
0x11aaf0	0x28e	cuDeviceComputeCapability
0x11ad80	0x28e	cuDeviceGetName
0x11b010	0x24d	cuDeviceGetCount	11b010: 48 89 5c 24 e0 mov %rbx,-0x20(%rsp) 11b015: 48 89 6c 24 e8 mov %rbp,-0x18(%rsp) 11b01a: 48 89 fd mov %rdi,%rbp 11b01d: 4c 89 64 24 f0 mov %r12,-0x10(%rsp) 11b022: 4c 89 6c 24 f8 mov %r13,-0x8(%rsp) 11b027: 48 83 ec 78 sub $0x78,%rsp 11b02b: 48 c7 44 24 48 00 00 movq $0x0,0x48(%rsp) 11b032: 00 00 11b034: e8 97 0c fd ff callq ebcd0 <__isinff@plt+0x6bfd0> 11b039: 84 c0 test %al,%al 11b03b: 89 c3 mov %eax,%ebx 11b03d: 0f 85 4d 01 00 00 jne 11b190 <__isinff@plt+0x9b490> 11b043: 45 31 ed xor %r13d,%r13d 11b046: 44 8b 0d c7 57 6f 00 mov 0x6f57c7(%rip),%r9d # 810814 <__isinff@plt+0x790b14> 11b04d: 45 85 c9 test %r9d,%r9d 11b050: 0f 84 01 01 00 00 je 11b157 <__isinff@plt+0x9b457> 11b056: 48 89 ef mov %rbp,%rdi // Perform the basic driver-initialization checks 11b059: e8 42 5b f6 ff callq 80ba0 <__isinff@plt+0xea0> 11b05e: 44 8b 15 af 57 6f 00 mov 0x6f57af(%rip),%r10d # 810814 <__isinff@plt+0x790b14> 11b065: 41 89 c4 mov %eax,%r12d 11b068: 45 85 d2 test %r10d,%r10d 11b06b: 0f 84 af 00 00 00 je 11b120 <__isinff@plt+0x9b420> 11b071: 84 db test %bl,%bl 11b073: 0f 95 c3 setne %bl 11b076: 44 84 eb test %r13b,%bl 11b079: 75 1c jne 11b097 <__isinff@plt+0x9b397> 11b07b: 44 89 e0 mov %r12d,%eax 11b07e: 48 8b 5c 24 58 mov 0x58(%rsp),%rbx 11b083: 48 8b 6c 24 60 mov 0x60(%rsp),%rbp 11b088: 4c 8b 64 24 68 mov 0x68(%rsp),%r12 11b08d: 4c 8b 6c 24 70 mov 0x70(%rsp),%r13 11b092: 48 83 c4 78 add $0x78,%rsp 11b096: c3 retq // handle weird case 11b097: fc cld 11b098: 31 c0 xor %eax,%eax 11b09a: b9 08 00 00 00 mov $0x8,%ecx 11b09f: 48 89 e7 mov %rsp,%rdi 11b0a2: f3 48 ab rep stos %rax,%es:(%rdi) 11b0a5: 4c 8b 2d bc 35 6f 00 mov 0x6f35bc(%rip),%r13 # 80e668 <__isinff@plt+0x78e968> 11b0ac: 48 89 6c 24 40 mov %rbp,0x40(%rsp) 11b0b1: 48 8d 6c 24 40 lea 0x40(%rsp),%rbp 11b0b6: 41 8b 7d 00 mov 0x0(%r13),%edi 11b0ba: e8 a1 c7 4a 00 callq 5c7860 <__isinff@plt+0x547b60> 11b0bf: 4c 8d 1d 6a 02 4c 00 lea 0x4c026a(%rip),%r11 # 5db330 <__isinff@plt+0x55b630> 11b0c6: 48 89 c2 mov %rax,%rdx 11b0c9: 48 89 04 24 mov %rax,(%rsp) 11b0cd: 31 c0 xor %eax,%eax 11b0cf: 48 85 d2 test %rdx,%rdx 11b0d2: 48 c7 44 24 08 00 00 movq $0x0,0x8(%rsp) 11b0d9: 00 00 11b0db: c7 44 24 10 04 00 00 movl $0x4,0x10(%rsp) 11b0e2: 00 11b0e3: 4c 89 5c 24 18 mov %r11,0x18(%rsp) 11b0e8: 48 89 6c 24 20 mov %rbp,0x20(%rsp) 11b0ed: 74 07 je 11b0f6 <__isinff@plt+0x9b3f6> 11b0ef: 48 8b 82 f0 03 00 00 mov 0x3f0(%rdx),%rax 11b0f6: 48 8b 74 24 48 mov 0x48(%rsp),%rsi 11b0fb: 48 8b 3d 3e 3b 6f 00 mov 0x6f3b3e(%rip),%rdi # 80ec40 <__isinff@plt+0x78ef40> 11b102: 48 89 44 24 28 mov %rax,0x28(%rsp) 11b107: 44 89 64 24 38 mov %r12d,0x38(%rsp) 11b10c: 48 89 74 24 30 mov %rsi,0x30(%rsp) 11b111: 48 89 e6 mov %rsp,%rsi 11b114: e8 c7 0b fd ff callq ebce0 <__isinff@plt+0x6bfe0> 11b119: e9 5d ff ff ff jmpq 11b07b <__isinff@plt+0x9b37b> 11b11e: 66 90 xchg %ax,%ax // Handler for no-context case 11b120: 48 8d 3d b3 f8 4b 00 lea 0x4bf8b3(%rip),%rdi # 5da9da <__isinff@plt+0x55acda> 11b127: e8 34 43 f6 ff callq 7f460 <getenv@plt> 11b12c: 48 85 c0 test %rax,%rax 11b12f: 74 17 je 11b148 <__isinff@plt+0x9b448> 11b131: 31 c9 xor %ecx,%ecx 11b133: ba 0a 00 00 00 mov $0xa,%edx 11b138: 31 f6 xor %esi,%esi 11b13a: 48 89 c7 mov %rax,%rdi 11b13d: e8 ee 47 f6 ff callq 7f930 <__strtol_internal@plt> 11b142: 89 05 d0 56 6f 00 mov %eax,0x6f56d0(%rip) # 810818 <__isinff@plt+0x790b18> 11b148: c7 05 c2 56 6f 00 01 movl $0x1,0x6f56c2(%rip) # 810814 <__isinff@plt+0x790b14> 11b14f: 00 00 00 11b152: e9 1a ff ff ff jmpq 11b071 <__isinff@plt+0x9b371> 11b157: 48 8d 3d 7c f8 4b 00 lea 0x4bf87c(%rip),%rdi # 5da9da <__isinff@plt+0x55acda> 11b15e: e8 fd 42 f6 ff callq 7f460 <getenv@plt> 11b163: 48 85 c0 test %rax,%rax 11b166: 74 17 je 11b17f <__isinff@plt+0x9b47f> 11b168: 31 c9 xor %ecx,%ecx 11b16a: ba 0a 00 00 00 mov $0xa,%edx 11b16f: 31 f6 xor %esi,%esi 11b171: 48 89 c7 mov %rax,%rdi 11b174: e8 b7 47 f6 ff callq 7f930 <__strtol_internal@plt> 11b179: 89 05 99 56 6f 00 mov %eax,0x6f5699(%rip) # 810818 <__isinff@plt+0x790b18> 11b17f: c7 05 8b 56 6f 00 01 movl $0x1,0x6f568b(%rip) # 810814 <__isinff@plt+0x790b14> 11b186: 00 00 00 11b189: e9 c8 fe ff ff jmpq 11b056 <__isinff@plt+0x9b356> 11b18e: 66 90 xchg %ax,%ax 11b190: 31 ff xor %edi,%edi 11b192: e8 39 54 f7 ff callq 905d0 <__isinff@plt+0x108d0> 11b197: 85 c0 test %eax,%eax 11b199: 0f 85 a4 fe ff ff jne 11b043 <__isinff@plt+0x9b343> 11b19f: fc cld 11b1a0: 31 c0 xor %eax,%eax 11b1a2: b9 07 00 00 00 mov $0x7,%ecx 11b1a7: 48 89 e7 mov %rsp,%rdi 11b1aa: f3 48 ab rep stos %rax,%es:(%rdi) 11b1ad: 48 8b 05 b4 34 6f 00 mov 0x6f34b4(%rip),%rax # 80e668 <__isinff@plt+0x78e968> 11b1b4: 48 89 6c 24 40 mov %rbp,0x40(%rsp) 11b1b9: 8b 38 mov (%rax),%edi 11b1bb: e8 a0 c6 4a 00 callq 5c7860 <__isinff@plt+0x547b60> 11b1c0: 48 85 c0 test %rax,%rax 11b1c3: 48 89 04 24 mov %rax,(%rsp) 11b1c7: 74 69 je 11b232 <__isinff@plt+0x9b532> 11b1c9: 48 ff 80 f0 03 00 00 incq 0x3f0(%rax) 11b1d0: 48 8d 54 24 40 lea 0x40(%rsp),%rdx 11b1d5: 48 8d 35 54 01 4c 00 lea 0x4c0154(%rip),%rsi # 5db330 <__isinff@plt+0x55b630> 11b1dc: 48 8b 04 24 mov (%rsp),%rax 11b1e0: 48 c7 44 24 08 00 00 movq $0x0,0x8(%rsp) 11b1e7: 00 00 11b1e9: 48 89 54 24 20 mov %rdx,0x20(%rsp) 11b1ee: 31 d2 xor %edx,%edx 11b1f0: c7 44 24 10 04 00 00 movl $0x4,0x10(%rsp) 11b1f7: 00 11b1f8: 48 89 74 24 18 mov %rsi,0x18(%rsp) 11b1fd: 48 85 c0 test %rax,%rax 11b200: 74 07 je 11b209 <__isinff@plt+0x9b509> 11b202: 48 8b 90 f0 03 00 00 mov 0x3f0(%rax),%rdx 11b209: 48 8b 3d 10 2e 6f 00 mov 0x6f2e10(%rip),%rdi # 80e020 <__isinff@plt+0x78e320> 11b210: 4c 8d 44 24 48 lea 0x48(%rsp),%r8 11b215: 48 89 e6 mov %rsp,%rsi 11b218: 41 bd 01 00 00 00 mov $0x1,%r13d 11b21e: 48 89 54 24 28 mov %rdx,0x28(%rsp) 11b223: 4c 89 44 24 30 mov %r8,0x30(%rsp) 11b228: e8 b3 0a fd ff callq ebce0 <__isinff@plt+0x6bfe0> 11b22d: e9 14 fe ff ff jmpq 11b046 <__isinff@plt+0x9b346> 11b232: 4c 8d 05 f7 00 4c 00 lea 0x4c00f7(%rip),%r8 # 5db330 <__isinff@plt+0x55b630> 11b239: 4c 8d 64 24 40 lea 0x40(%rsp),%r12 11b23e: 31 d2 xor %edx,%edx 11b240: 48 c7 44 24 08 00 00 movq $0x0,0x8(%rsp) 11b247: 00 00 11b249: c7 44 24 10 04 00 00 movl $0x4,0x10(%rsp) 11b250: 00 11b251: 4c 89 44 24 18 mov %r8,0x18(%rsp) 11b256: 4c 89 64 24 20 mov %r12,0x20(%rsp) 11b25b: eb ac jmp 11b209 <__isinff@plt+0x9b509> 11b25d: 66 66 90 xchg %ax,%ax 11b260: 48 89 5c 24 d8 mov %rbx,-0x28(%rsp) 11b265: 48 89 6c 24 e0 mov %rbp,-0x20(%rsp) 11b26a: 48 89 fd mov %rdi,%rbp 11b26d: 4c 89 64 24 e8 mov %r12,-0x18(%rsp) 11b272: 4c 89 6c 24 f0 mov %r13,-0x10(%rsp) 11b277: 41 89 f4 mov %esi,%r12d 11b27a: 4c 89 74 24 f8 mov %r14,-0x8(%rsp) 11b27f: 48 81 ec 88 00 00 00 sub $0x88,%rsp 11b286: 48 c7 44 24 58 00 00 movq $0x0,0x58(%rsp) 11b28d: 00 00 11b28f: e8 3c 0a fd ff callq ebcd0 <__isinff@plt+0x6bfd0> 11b294: 84 c0 test %al,%al 11b296: 89 c3 mov %eax,%ebx 11b298: 0f 85 62 01 00 00 jne 11b400 <__isinff@plt+0x9b700> 11b29e: 45 31 f6 xor %r14d,%r14d 11b2a1: 44 8b 0d 6c 55 6f 00 mov 0x6f556c(%rip),%r9d # 810814 <__isinff@plt+0x790b14> 11b2a8: 45 85 c9 test %r9d,%r9d 11b2ab: 0f 84 16 01 00 00 je 11b3c7 <__isinff@plt+0x9b6c7> 11b2b1: 44 89 e6 mov %r12d,%esi 11b2b4: 48 89 ef mov %rbp,%rdi 11b2b7: e8 34 5b f6 ff callq 80df0 <__isinff@plt+0x10f0> 11b2bc: 44 8b 15 51 55 6f 00 mov 0x6f5551(%rip),%r10d # 810814 <__isinff@plt+0x790b14> 11b2c3: 41 89 c5 mov %eax,%r13d 11b2c6: 45 85 d2 test %r10d,%r10d 11b2c9: 0f 84 c1 00 00 00 je 11b390 <__isinff@plt+0x9b690> 11b2cf: 84 db test %bl,%bl 11b2d1: 0f 95 c3 setne %bl 11b2d4: 41 84 de test %bl,%r14b 11b2d7: 75 27 jne 11b300 <__isinff@plt+0x9b600> 11b2d9: 44 89 e8 mov %r13d,%eax 11b2dc: 48 8b 5c 24 60 mov 0x60(%rsp),%rbx 11b2e1: 48 8b 6c 24 68 mov 0x68(%rsp),%rbp 11b2e6: 4c 8b 64 24 70 mov 0x70(%rsp),%r12 11b2eb: 4c 8b 6c 24 78 mov 0x78(%rsp),%r13 11b2f0: 4c 8b b4 24 80 00 00 mov 0x80(%rsp),%r14 11b2f7: 00 11b2f8: 48 81 c4 88 00 00 00 add $0x88,%rsp 11b2ff: c3 retq 11b300: fc cld 11b301: 31 c0 xor %eax,%eax 11b303: b9 08 00 00 00 mov $0x8,%ecx 11b308: 48 89 e7 mov %rsp,%rdi 11b228: e8 b3 0a fd ff callq ebce0 <__isinff@plt+0x6bfe0> 11b22d: e9 14 fe ff ff jmpq 11b046 <__isinff@plt+0x9b346> 11b232: 4c 8d 05 f7 00 4c 00 lea 0x4c00f7(%rip),%r8 # 5db330 <__isinff@plt+0x55b630>24 40 lea 0x40(%rsp),%r12 11b23e: 31 d2 xor %edx,%edx 11b240: 48 c7 44 24 08 00 00 movq $0x0,0x8(%rsp) 11b247: 00 00 11b249: c7 44 24 10 04 00 00 movl $0x4,0x10(%rsp) 11b250: 00 11b251: 4c 89 44 24 18 mov %r8,0x18(%rsp) 11b256: 4c 89 64 24 20 mov %r12,0x20(%rsp) 11b25b: eb ac jmp 11b209 <__isinff@plt+0x9b509>
0x11b260	0x272	cuDeviceGet
0x11b4e0	0x24d	cuDriverGetVersion
0x11b730	0x18c	cuGLInit
0x11b8c0	0x28e	cuGLCtxCreate
0x11bb50	0x261	cuGLUnmapBufferObjectAsync
0x11bdc0	0x2aa	cuGLMapBufferObjectAsync
0x11c070	0x261	cuGLSetBufferObjectMapFlags
0x11c2e0	0x253	cuGLUnregisterBufferObject
0x11c540	0x253	cuGLUnmapBufferObject
0x11c7a0	0x28e	cuGLMapBufferObject
0x11ca30	0x253	cuGLRegisterBufferObject
0x11cc90	0x2b2	cuGraphicsGLRegisterImage
0x11cf50	0x28e	cuGraphicsGLRegisterBuffer
0x81ef0	0x5f	cuMemGetAttribute
0x110a30	0x24d	cuTexRefDestroy
0x110c80	0x24d	cuTexRefCreate
0x110ed0	0x272	cuArray3DGetDescriptor
0x111b20	0x272	cuFuncSetCacheConfig
0x111da0	0x28e	cuFuncGetAttribute
0x112b30	0x2d5	cuMemsetD2D8
0x112e10	0x28d	cuMemsetD32
0x113ac0	0x2cc	cuMemcpyAtoHAsync
0x113d90	0x2cc	cuMemcpyHtoAAsync
0x114ac0	0x24d	cuMemcpy2DUnaligned
0x114d10	0x24d	cuMemcpy2D
0x114f60	0x2d4	cuMemcpyAtoA
0x115a80	0x2b2	cuMemcpyDtoA
0x115d40	0x28d	cuMemcpyDtoD
0x115fd0	0x28e	cuMemcpyDtoH
0x116a00	0x28e	cuMemHostAlloc
0x116c90	0x24d	cuMemFreeHost
0x116ee0	0x272	cuMemAllocHost
0x117bb0	0x272	cuMemGetInfo
0x117e30	0x28e	cuModuleGetTexRef
0x118ae0	0x2d4	cuModuleLoadDataEx
0x118dc0	0x272	cuModuleLoadData
0x119c00	0x272	cuCtxAttach
0x119e80	0x24d	cuCtxDestroy
0x1104e0	0x2b2	cuTexRefSetAddress
0x1107a0	0x28e	cuTexRefSetArray
0x1113d0	0x24d	cuArrayDestroy
0x1118a0	0x272	cuArrayCreate
0x1122b0	0x2b2	cuFuncSetBlockShape
0x1130a0	0x29d	cuMemsetD16
0x1135e0	0x261	cuMemcpy3DAsync
0x1145c0	0x2aa	cuMemcpyHtoDAsync
0x1157c0	0x2b1	cuMemcpyAtoD
0x1164f0	0x272	cuMemHostGetFlags
0x1173f0	0x253	cuMemFree
0x1180c0	0x2b9	cuModuleGetGlobal
0x1192c0	0x24d	cuCtxGetDevice
0x1199b0	0x24d	cuCtxDetach
0x102120	0x119	clGetExtensionFunctionAddress
0x110220	0x2b2	cuTexRefSetAddress2D
0x111150	0x272	cuArray3DCreate
0x111620	0x272	cuArrayGetDescriptor
0x112030	0x272	cuFuncSetSharedSize
0x112570	0x2d3	cuMemsetD2D32
0x112850	0x2d5	cuMemsetD2D16
0x113340	0x29c	cuMemsetD8
0x113850	0x261	cuMemcpy2DAsync
0x114060	0x2aa	cuMemcpyDtoDAsync
0x114310	0x2aa	cuMemcpyDtoHAsync
0x114870	0x24d	cuMemcpy3D
0x115240	0x2b2	cuMemcpyAtoH
0x115500	0x2b9	cuMemcpyHtoA
0x116260	0x28d	cuMemcpyHtoD
0x116770	0x28e	cuMemHostGetDevicePointer
0x117160	0x28e	cuMemGetAddressRange
0x117650	0x2d4	cuMemAllocPitch
0x117930	0x272	cuMemAlloc
0x118380	0x28e	cuModuleGetFunction
0x118610	0x24d	cuModuleUnload
0x118860	0x272	cuModuleLoadFatBinary
0x119040	0x272	cuModuleLoad
0x119510	0x24d	cuCtxPopCurrent
0x119760	0x24d	cuCtxPushCurrent

See Also

• Kernel ioctl numbering documentation
• My CUDA and CUBAR pages
• I develped ptracer to get traces for this project