1 What: /sys/kernel/debug/accel/<paren !! 1 What: /sys/kernel/debug/habanalabs/hl<n>/addr
2 Date: Jan 2019 2 Date: Jan 2019
3 KernelVersion: 5.1 3 KernelVersion: 5.1
4 Contact: ogabbay@kernel.org 4 Contact: ogabbay@kernel.org
5 Description: Sets the device address to be 5 Description: Sets the device address to be used for read or write through
6 PCI bar, or the device VA of a 6 PCI bar, or the device VA of a host mapped memory to be read or
7 written directly from the host 7 written directly from the host. The latter option is allowed
8 only when the IOMMU is disable 8 only when the IOMMU is disabled.
9 The acceptable value is a stri 9 The acceptable value is a string that starts with "0x"
10 10
11 What: /sys/kernel/debug/accel/<paren !! 11 What: /sys/kernel/debug/habanalabs/hl<n>/clk_gate
12 Date: May 2020 12 Date: May 2020
13 KernelVersion: 5.8 13 KernelVersion: 5.8
14 Contact: ogabbay@kernel.org 14 Contact: ogabbay@kernel.org
15 Description: This setting is now deprecated !! 15 Description: Allow the root user to disable/enable in runtime the clock
>> 16 gating mechanism in Gaudi. Due to how Gaudi is built, the
>> 17 clock gating needs to be disabled in order to access the
>> 18 registers of the TPC and MME engines. This is sometimes needed
>> 19 during debug by the user and hence the user needs this option.
>> 20 The user can supply a bitmask value, each bit represents
>> 21 a different engine to disable/enable its clock gating feature.
>> 22 The bitmask is composed of 20 bits:
>> 23
>> 24 ======= ============
>> 25 0 - 7 DMA channels
>> 26 8 - 11 MME engines
>> 27 12 - 19 TPC engines
>> 28 ======= ============
>> 29
>> 30 The bit's location of a specific engine can be determined
>> 31 using (1 << GAUDI_ENGINE_ID_*). GAUDI_ENGINE_ID_* values
>> 32 are defined in uapi habanalabs.h file in enum gaudi_engine_id
16 33
17 What: /sys/kernel/debug/accel/<paren !! 34 What: /sys/kernel/debug/habanalabs/hl<n>/command_buffers
18 Date: Jan 2019 35 Date: Jan 2019
19 KernelVersion: 5.1 36 KernelVersion: 5.1
20 Contact: ogabbay@kernel.org 37 Contact: ogabbay@kernel.org
21 Description: Displays a list with informati 38 Description: Displays a list with information about the currently allocated
22 command buffers 39 command buffers
23 40
24 What: /sys/kernel/debug/accel/<paren !! 41 What: /sys/kernel/debug/habanalabs/hl<n>/command_submission
25 Date: Jan 2019 42 Date: Jan 2019
26 KernelVersion: 5.1 43 KernelVersion: 5.1
27 Contact: ogabbay@kernel.org 44 Contact: ogabbay@kernel.org
28 Description: Displays a list with informati 45 Description: Displays a list with information about the currently active
29 command submissions 46 command submissions
30 47
31 What: /sys/kernel/debug/accel/<paren !! 48 What: /sys/kernel/debug/habanalabs/hl<n>/command_submission_jobs
32 Date: Jan 2019 49 Date: Jan 2019
33 KernelVersion: 5.1 50 KernelVersion: 5.1
34 Contact: ogabbay@kernel.org 51 Contact: ogabbay@kernel.org
35 Description: Displays a list with detailed 52 Description: Displays a list with detailed information about each JOB (CB) of
36 each active command submission 53 each active command submission
37 54
38 What: /sys/kernel/debug/accel/<paren !! 55 What: /sys/kernel/debug/habanalabs/hl<n>/data32
39 Date: Jan 2019 56 Date: Jan 2019
40 KernelVersion: 5.1 57 KernelVersion: 5.1
41 Contact: ogabbay@kernel.org 58 Contact: ogabbay@kernel.org
42 Description: Allows the root user to read o 59 Description: Allows the root user to read or write directly through the
43 device's PCI bar. Writing to t 60 device's PCI bar. Writing to this file generates a write
44 transaction while reading from 61 transaction while reading from the file generates a read
45 transaction. This custom inter 62 transaction. This custom interface is needed (instead of using
46 the generic Linux user-space P 63 the generic Linux user-space PCI mapping) because the DDR bar
47 is very small compared to the 64 is very small compared to the DDR memory and only the driver can
48 move the bar before and after 65 move the bar before and after the transaction.
49 66
50 If the IOMMU is disabled, it a 67 If the IOMMU is disabled, it also allows the root user to read
51 or write from the host a devic 68 or write from the host a device VA of a host mapped memory
52 69
53 What: /sys/kernel/debug/accel/<paren !! 70 What: /sys/kernel/debug/habanalabs/hl<n>/data64
54 Date: Jan 2020 71 Date: Jan 2020
55 KernelVersion: 5.6 72 KernelVersion: 5.6
56 Contact: ogabbay@kernel.org 73 Contact: ogabbay@kernel.org
57 Description: Allows the root user to read o 74 Description: Allows the root user to read or write 64 bit data directly
58 through the device's PCI bar. 75 through the device's PCI bar. Writing to this file generates a
59 write transaction while readin 76 write transaction while reading from the file generates a read
60 transaction. This custom inter 77 transaction. This custom interface is needed (instead of using
61 the generic Linux user-space P 78 the generic Linux user-space PCI mapping) because the DDR bar
62 is very small compared to the 79 is very small compared to the DDR memory and only the driver can
63 move the bar before and after 80 move the bar before and after the transaction.
64 81
65 If the IOMMU is disabled, it a 82 If the IOMMU is disabled, it also allows the root user to read
66 or write from the host a devic 83 or write from the host a device VA of a host mapped memory
67 84
68 What: /sys/kernel/debug/accel/<paren !! 85 What: /sys/kernel/debug/habanalabs/hl<n>/data_dma
69 Date: Apr 2021 86 Date: Apr 2021
70 KernelVersion: 5.13 87 KernelVersion: 5.13
71 Contact: ogabbay@kernel.org 88 Contact: ogabbay@kernel.org
72 Description: Allows the root user to read f 89 Description: Allows the root user to read from the device's internal
73 memory (DRAM/SRAM) through a D 90 memory (DRAM/SRAM) through a DMA engine.
74 This property is a binary blob 91 This property is a binary blob that contains the result of the
75 DMA transfer. 92 DMA transfer.
76 This custom interface is neede 93 This custom interface is needed (instead of using the generic
77 Linux user-space PCI mapping) 94 Linux user-space PCI mapping) because the amount of internal
78 memory is huge (>32GB) and rea 95 memory is huge (>32GB) and reading it via the PCI bar will take
79 a very long time. 96 a very long time.
80 This interface doesn't support 97 This interface doesn't support concurrency in the same device.
81 In GAUDI and GOYA, this action 98 In GAUDI and GOYA, this action can cause undefined behavior
82 in case it is done while the d !! 99 in case the it is done while the device is executing user
83 workloads. 100 workloads.
84 Only supported on GAUDI at thi 101 Only supported on GAUDI at this stage.
85 102
86 What: /sys/kernel/debug/accel/<paren !! 103 What: /sys/kernel/debug/habanalabs/hl<n>/device
87 Date: Jan 2019 104 Date: Jan 2019
88 KernelVersion: 5.1 105 KernelVersion: 5.1
89 Contact: ogabbay@kernel.org 106 Contact: ogabbay@kernel.org
90 Description: Enables the root user to set t 107 Description: Enables the root user to set the device to specific state.
91 Valid values are "disable", "e 108 Valid values are "disable", "enable", "suspend", "resume".
92 User can read this property to 109 User can read this property to see the valid values
93 110
94 What: /sys/kernel/debug/accel/<paren !! 111 What: /sys/kernel/debug/habanalabs/hl<n>/dma_size
95 Date: Oct 2022 <<
96 KernelVersion: 6.2 <<
97 Contact: ttayar@habana.ai <<
98 Description: The watchdog timeout value in <<
99 certain error cases, after whi <<
100 <<
101 What: /sys/kernel/debug/accel/<paren <<
102 Date: Apr 2021 112 Date: Apr 2021
103 KernelVersion: 5.13 113 KernelVersion: 5.13
104 Contact: ogabbay@kernel.org 114 Contact: ogabbay@kernel.org
105 Description: Specify the size of the DMA tr 115 Description: Specify the size of the DMA transaction when using DMA to read
106 from the device's internal mem 116 from the device's internal memory. The value can not be larger
107 than 128MB. Writing to this va 117 than 128MB. Writing to this value initiates the DMA transfer.
108 When the write is finished, th 118 When the write is finished, the user can read the "data_dma"
109 blob 119 blob
110 120
111 What: /sys/kernel/debug/accel/<paren !! 121 What: /sys/kernel/debug/habanalabs/hl<n>/dump_security_violations
112 Date: Aug 2022 <<
113 KernelVersion: 5.20 <<
114 Contact: fkassabri@habana.ai <<
115 Description: Dumps all razwi events to dmes <<
116 After reading the status regis <<
117 the routine will clear the sta <<
118 Usage: cat dump_razwi_events <<
119 <<
120 What: /sys/kernel/debug/accel/<paren <<
121 Date: Jan 2021 122 Date: Jan 2021
122 KernelVersion: 5.12 123 KernelVersion: 5.12
123 Contact: ogabbay@kernel.org 124 Contact: ogabbay@kernel.org
124 Description: Dumps all security violations 125 Description: Dumps all security violations to dmesg. This will also ack
125 all security violations meanin 126 all security violations meanings those violations will not be
126 dumped next time user calls th 127 dumped next time user calls this API
127 128
128 What: /sys/kernel/debug/accel/<paren !! 129 What: /sys/kernel/debug/habanalabs/hl<n>/engines
129 Date: Jul 2019 130 Date: Jul 2019
130 KernelVersion: 5.3 131 KernelVersion: 5.3
131 Contact: ogabbay@kernel.org 132 Contact: ogabbay@kernel.org
132 Description: Displays the status registers 133 Description: Displays the status registers values of the device engines and
133 their derived idle status 134 their derived idle status
134 135
135 What: /sys/kernel/debug/accel/<paren !! 136 What: /sys/kernel/debug/habanalabs/hl<n>/i2c_addr
136 Date: Jan 2019 137 Date: Jan 2019
137 KernelVersion: 5.1 138 KernelVersion: 5.1
138 Contact: ogabbay@kernel.org 139 Contact: ogabbay@kernel.org
139 Description: Sets I2C device address for I2 140 Description: Sets I2C device address for I2C transaction that is generated
140 by the device's CPU, Not avail !! 141 by the device's CPU
141 firmware <<
142 142
143 What: /sys/kernel/debug/accel/<paren !! 143 What: /sys/kernel/debug/habanalabs/hl<n>/i2c_bus
144 Date: Jan 2019 144 Date: Jan 2019
145 KernelVersion: 5.1 145 KernelVersion: 5.1
146 Contact: ogabbay@kernel.org 146 Contact: ogabbay@kernel.org
147 Description: Sets I2C bus address for I2C t 147 Description: Sets I2C bus address for I2C transaction that is generated by
148 the device's CPU, Not availabl !! 148 the device's CPU
149 firmware <<
150 149
151 What: /sys/kernel/debug/accel/<paren !! 150 What: /sys/kernel/debug/habanalabs/hl<n>/i2c_data
152 Date: Jan 2019 151 Date: Jan 2019
153 KernelVersion: 5.1 152 KernelVersion: 5.1
154 Contact: ogabbay@kernel.org 153 Contact: ogabbay@kernel.org
155 Description: Triggers an I2C transaction th 154 Description: Triggers an I2C transaction that is generated by the device's
156 CPU. Writing to this file gene 155 CPU. Writing to this file generates a write transaction while
157 reading from the file generate !! 156 reading from the file generates a read transaction
158 when device is loaded with sec <<
159 157
160 What: /sys/kernel/debug/accel/<paren !! 158 What: /sys/kernel/debug/habanalabs/hl<n>/i2c_reg
161 Date: Dec 2021 <<
162 KernelVersion: 5.17 <<
163 Contact: obitton@habana.ai <<
164 Description: Sets I2C length in bytes for I <<
165 the device's CPU, Not availabl <<
166 firmware <<
167 <<
168 What: /sys/kernel/debug/accel/<paren <<
169 Date: Jan 2019 159 Date: Jan 2019
170 KernelVersion: 5.1 160 KernelVersion: 5.1
171 Contact: ogabbay@kernel.org 161 Contact: ogabbay@kernel.org
172 Description: Sets I2C register id for I2C t 162 Description: Sets I2C register id for I2C transaction that is generated by
173 the device's CPU, Not availabl !! 163 the device's CPU
174 firmware <<
175 164
176 What: /sys/kernel/debug/accel/<paren !! 165 What: /sys/kernel/debug/habanalabs/hl<n>/led0
177 Date: Jan 2019 166 Date: Jan 2019
178 KernelVersion: 5.1 167 KernelVersion: 5.1
179 Contact: ogabbay@kernel.org 168 Contact: ogabbay@kernel.org
180 Description: Sets the state of the first S/ !! 169 Description: Sets the state of the first S/W led on the device
181 when device is loaded with sec <<
182 170
183 What: /sys/kernel/debug/accel/<paren !! 171 What: /sys/kernel/debug/habanalabs/hl<n>/led1
184 Date: Jan 2019 172 Date: Jan 2019
185 KernelVersion: 5.1 173 KernelVersion: 5.1
186 Contact: ogabbay@kernel.org 174 Contact: ogabbay@kernel.org
187 Description: Sets the state of the second S !! 175 Description: Sets the state of the second S/W led on the device
188 when device is loaded with sec <<
189 176
190 What: /sys/kernel/debug/accel/<paren !! 177 What: /sys/kernel/debug/habanalabs/hl<n>/led2
191 Date: Jan 2019 178 Date: Jan 2019
192 KernelVersion: 5.1 179 KernelVersion: 5.1
193 Contact: ogabbay@kernel.org 180 Contact: ogabbay@kernel.org
194 Description: Sets the state of the third S/ !! 181 Description: Sets the state of the third S/W led on the device
195 when device is loaded with sec <<
196 <<
197 What: /sys/kernel/debug/accel/<paren <<
198 Date: May 2022 <<
199 KernelVersion: 5.19 <<
200 Contact: dhirschfeld@habana.ai <<
201 Description: Allows the root user to scrub <<
202 value can be set using the deb <<
203 182
204 What: /sys/kernel/debug/accel/<paren !! 183 What: /sys/kernel/debug/habanalabs/hl<n>/mmu
205 Date: May 2022 <<
206 KernelVersion: 5.19 <<
207 Contact: dhirschfeld@habana.ai <<
208 Description: The value to which the dram wi <<
209 scrubs the dram using 'memory_ <<
210 the scrubbing value when using <<
211 <<
212 What: /sys/kernel/debug/accel/<paren <<
213 Date: Jan 2019 184 Date: Jan 2019
214 KernelVersion: 5.1 185 KernelVersion: 5.1
215 Contact: ogabbay@kernel.org 186 Contact: ogabbay@kernel.org
216 Description: Displays the hop values and ph 187 Description: Displays the hop values and physical address for a given ASID
217 and virtual address. The user 188 and virtual address. The user should write the ASID and VA into
218 the file and then read the fil 189 the file and then read the file to get the result.
219 e.g. to display info about VA 190 e.g. to display info about VA 0x1000 for ASID 1 you need to do:
220 echo "1 0x1000" > /sys/kernel/ !! 191 echo "1 0x1000" > /sys/kernel/debug/habanalabs/hl0/mmu
221 192
222 What: /sys/kernel/debug/accel/<paren !! 193 What: /sys/kernel/debug/habanalabs/hl<n>/mmu_error
223 Date: Mar 2021 194 Date: Mar 2021
224 KernelVersion: 5.12 195 KernelVersion: 5.12
225 Contact: fkassabri@habana.ai 196 Contact: fkassabri@habana.ai
226 Description: Check and display page fault o 197 Description: Check and display page fault or access violation mmu errors for
227 all MMUs specified in mmu_cap_ 198 all MMUs specified in mmu_cap_mask.
228 e.g. to display error info for 199 e.g. to display error info for MMU hw cap bit 9, you need to do:
229 echo "0x200" > /sys/kernel/deb !! 200 echo "0x200" > /sys/kernel/debug/habanalabs/hl0/mmu_error
230 cat /sys/kernel/debug/accel/<p !! 201 cat /sys/kernel/debug/habanalabs/hl0/mmu_error
231 202
232 What: /sys/kernel/debug/accel/<paren !! 203 What: /sys/kernel/debug/habanalabs/hl<n>/set_power_state
233 Date: Mar 2022 <<
234 KernelVersion: 5.19 <<
235 Contact: osharabi@habana.ai <<
236 Description: Allows the root user to dump m <<
237 protected config space. <<
238 This property is a binary blob <<
239 monitors registers dump. <<
240 This custom interface is neede <<
241 Linux user-space PCI mapping) <<
242 and cannot be accessed using P <<
243 This interface doesn't support <<
244 Only supported on GAUDI. <<
245 <<
246 What: /sys/kernel/debug/accel/<paren <<
247 Date: Mar 2022 <<
248 KernelVersion: 5.19 <<
249 Contact: osharabi@habana.ai <<
250 Description: Triggers dump of monitor data. <<
251 must be 1. Triggering the moni <<
252 current registers values of al <<
253 When the write is finished, th <<
254 blob <<
255 <<
256 What: /sys/kernel/debug/accel/<paren <<
257 Date: Feb 2024 <<
258 KernelVersion: 6.11 <<
259 Contact: trisin@habana.ai <<
260 Description: Exposes the device's server ty <<
261 <<
262 What: /sys/kernel/debug/accel/<paren <<
263 Date: Jan 2019 204 Date: Jan 2019
264 KernelVersion: 5.1 205 KernelVersion: 5.1
265 Contact: ogabbay@kernel.org 206 Contact: ogabbay@kernel.org
266 Description: Sets the PCI power state. Vali 207 Description: Sets the PCI power state. Valid values are "1" for D0 and "2"
267 for D3Hot 208 for D3Hot
268 209
269 What: /sys/kernel/debug/accel/<paren !! 210 What: /sys/kernel/debug/habanalabs/hl<n>/skip_reset_on_timeout
270 Date: Jun 2021 211 Date: Jun 2021
271 KernelVersion: 5.13 212 KernelVersion: 5.13
272 Contact: ynudelman@habana.ai 213 Contact: ynudelman@habana.ai
273 Description: Sets the skip reset on timeout 214 Description: Sets the skip reset on timeout option for the device. Value of
274 "0" means device will be reset 215 "0" means device will be reset in case some CS has timed out,
275 otherwise it will not be reset 216 otherwise it will not be reset.
276 217
277 What: /sys/kernel/debug/accel/<paren !! 218 What: /sys/kernel/debug/habanalabs/hl<n>/stop_on_err
278 Date: Oct 2021 <<
279 KernelVersion: 5.15 <<
280 Contact: ynudelman@habana.ai <<
281 Description: Gets the state dump occurring <<
282 State dump is used for debug a <<
283 a problem in a CS execution, b <<
284 Reading from the node returns <<
285 Writing an integer X discards <<
286 next read would return X+1-st <<
287 <<
288 What: /sys/kernel/debug/accel/<paren <<
289 Date: Mar 2020 219 Date: Mar 2020
290 KernelVersion: 5.6 220 KernelVersion: 5.6
291 Contact: ogabbay@kernel.org 221 Contact: ogabbay@kernel.org
292 Description: Sets the stop-on_error option 222 Description: Sets the stop-on_error option for the device engines. Value of
293 "0" is for disable, otherwise 223 "0" is for disable, otherwise enable.
294 Relevant only for GOYA and GAU <<
295 224
296 What: /sys/kernel/debug/accel/<paren !! 225 What: /sys/kernel/debug/habanalabs/hl<n>/userptr
297 Date: Sep 2021 <<
298 KernelVersion: 5.16 <<
299 Contact: obitton@habana.ai <<
300 Description: Sets the command submission ti <<
301 <<
302 What: /sys/kernel/debug/accel/<paren <<
303 Date: Jan 2019 226 Date: Jan 2019
304 KernelVersion: 5.1 227 KernelVersion: 5.1
305 Contact: ogabbay@kernel.org 228 Contact: ogabbay@kernel.org
306 Description: Displays a list with informati !! 229 Description: Displays a list with information about the currently user
307 pointers (user virtual address 230 pointers (user virtual addresses) that are pinned and mapped
308 to DMA addresses 231 to DMA addresses
309 232
310 What: /sys/kernel/debug/accel/<paren !! 233 What: /sys/kernel/debug/habanalabs/hl<n>/vm
311 Date: Oct 2021 <<
312 KernelVersion: 5.15 <<
313 Contact: ogabbay@kernel.org <<
314 Description: Allows to search for specific <<
315 addresses) that are pinned and <<
316 their resolution to the specif <<
317 <<
318 What: /sys/kernel/debug/accel/<paren <<
319 Date: Jan 2019 234 Date: Jan 2019
320 KernelVersion: 5.1 235 KernelVersion: 5.1
321 Contact: ogabbay@kernel.org 236 Contact: ogabbay@kernel.org
322 Description: Displays a list with informati 237 Description: Displays a list with information about all the active virtual
323 address mappings per ASID and 238 address mappings per ASID and all user mappings of HW blocks
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.