Jiajia Qin d226e40856 [js/webgpu] set query type in onRunStart (#19202) ... ### Description <!-- Describe your changes. --> `env.webgpu.profiling` is a global flag. It may change before each session.run. So the best place is to update it in `onRunStart` event. After this, we can directly check `this.queryType`'s value. Without this pr, we need to make sure that `getCommandEncoder()` is called before checking `this.queryType`. Otherwise, it may happen that `pendingKernels`'s length is not equal to `pendingDispatchNumber`'s length. See the two ugly workarounds [1)](e630dbf528 (diff-006fc84d3997f96a29b8033bd2075d6a0a9509211bd5812a6b934fc74fedfd9dR267-R268)) and [2)](e630dbf528 (diff-618fe297fbe7a1da586380163b8fd2627311ccc217640a3c5cdc9c17a33472c1R73-R80)) if we don't introduce `onRunStart`. Or we need to call `setQueryType` in each kernel run.		2024-01-22 16:08:55 -08:00
..
webgpu	[js/webgpu] Add HardSigmoid support (#19215)	2024-01-22 15:53:26 -08:00
backend-webgpu.ts	[js/webgpu] set query type in onRunStart (#19202)	2024-01-22 16:08:55 -08:00
init.ts	[js/webgpu] Refactor timestamp-query and introduce timestamp-query-inside-passes (#18894)	2024-01-13 00:23:17 -08:00
log.ts
tensor-view.ts	[js/web] revise TensorView (#17473)	2023-09-14 21:14:44 -07:00
util.ts