HomeSort by relevance Sort by last modified time
    Searched refs:located (Results 1 - 25 of 54) sorted by null

1 2 3

  /device/linaro/bootloader/edk2/Vlv2TbltDevicePkg/
Build_IFWI.sh 103 echo The final IFWI file is located in Stitch
Build_IFWI.bat 162 echo The final IFWI file is located in %ROOT_DIR%\Vlv2TbltDevicePkg\Stitch\
  /external/autotest/bin/
python_venv 51 ERROR: create_venv script could not be located.
  /external/swiftshader/third_party/llvm-7.0/llvm/test/MC/AsmParser/
altmacro_expression.s 16 # The altmacro percent '%' must be located before the first argument
17 # If a percent is located in the middle of the estimated argument without any
  /external/curl/docs/cmdline-opts/
key.d 11 then a PKCS#11 URI (RFC 7512) can be used to specify a private key located in a
  /external/doclava/res/assets/templates/
sample.cs 20 <p>The file containing the source code shown below is located in the corresponding directory in <code>&lt;sdk&gt;/samples/android-&lt;version&gt;/...</code></p>
  /device/linaro/bootloader/edk2/UefiCpuPkg/ResetVector/Vtf0/Ia16/
ResetVectorVtf0.asm 27 ; located just below 0x100000000 (4GB) in the firmware device.
  /cts/tests/tests/neuralnetworks/
Android.mk 18 # The sources are located at frameworks/ml/nn/runtime/test
  /development/build/tools/
windows_sdk.mk 23 # Windows executables. All the targets specified here are located in
  /external/swiftshader/third_party/SPIRV-Headers/include/spirv/unified1/
spv.d 16 + HEADER INFORMATION ARE LOCATED AT https://www.khronos.org/registry/
41 + - C# will use enum classes in the Specification class located in the "Spv" namespace,
  /external/tensorflow/tensorflow/contrib/verbs/
README.md 34 1. Tensor writes will be done directly from the source tensor to the **result** tensor, with no memory copies in between. This should be done for all DMAable tensors which are located either on CPU or on a RDMA compatible GPU device (GPU direct).
36 3. Tensors which are located on a non-RDMA-compatible GPU, will be RDMA written to a registered CPU **proxy** buffer on the receiver side, and then copied to GPU by the receiver.
55 The tensor transfer process is initiated when the receiver requests a tensor. In code it is done by calling **Rendezvous::Recv()** or **Rendezvous::RecvAsync()**. The TensorFlow base implementation handles the case where the requested tensor is located on the same node. The more interesting case where the requested tensor is located on a remote node (receiver != sender) is to be handled in a derivation of the pure virtual **BaseRemoteRendezvous::RecvFromRemoteAsync()**. TensorFlow provides a default GRPC based implementation which comes in the vanilla version but suffers in scalability when running large models. Our RDMA based implementation presumes to be more scalable. HKUST's contrib GDR implementation is more scalable than GRPC, and less scalable than ours, only because we did our evolution based on it.
57 Our entry point is the implementation of **RdmaRemoteRendezvous::RecvFromRemoteAsync()**, located in rdma_rendezvous_mgr.cc. The implementation creates a new **RdmaTensorRequest** object, keyed by request index (uint32_t), stores it in a list of pending requests, and calls its **Start()** method. The **Start()** method basically does 2 things:
68 3. If the source tensor is located on a device which cannot be DMA written from, copy it to CPU.
patch_notes_verbs_with_0_copies.md 13 1. Tensor writes need to be done directly from the source Tensor to the destination Tensor, with no memory copies in between. This should be done for all DMAble tensors which are located either on CPU or on a RDMA compatible GPU device (GPU direct).
15 3. Tensors which are located on a non-RDMA-compatible GPU, will be RDMA written to a registered CPU proxy buffer on the receiver side, and then copied to GPU by the receiver.
  /build/soong/java/
prebuilt_apis.go 185 // API txt files found under the directory where the Android.bp is located.
186 // Specifically, an API file located at ./<ver>/<scope>/api/<module>.txt
hiddenapi_singleton.go 63 // These rules depend on files located in frameworks/base, skip them if running in a tree that doesn't have them.
  /device/linaro/bootloader/edk2/
Edk2Setup.bat 16 @REM Three other scripts, located in the BaseTools\Scripts directory, may be called
305 @echo if python.exe is located in the C:\Python27 directory, then:
316 @echo located in the C:\Python27 directory, then:
322 @echo ERROR : Unable to locate cx_Freeze 4.2.3. The cxfreeze.bat file must be located
487 @REM python.exe has been located, now make sure it's in the PATH
  /build/soong/symbol_inject/
pe_test.go 48 // Test when symbol has the same value as the target symbol but is located afterwards in the list
  /device/linaro/bootloader/arm-trusted-firmware/
Makefile 233 $(error Error: No services/spd/${SPD}/${SPD}.mk located)
292 $(error Error: No bl32/${AARCH32_SP}/${AARCH32_SP}.mk located)
  /device/linaro/bootloader/edk2/ArmPlatformPkg/PrePi/Arm/
ModuleEntryPoint.asm 48 // to install the stacks at the bottom of the Firmware Device (case the FD is located
  /external/autotest/docs/
test-that.md 132 instance located at 100.96.51.136
  /external/libchrome/mojo/public/js/
base.js 30 // located at:
  /external/libffi/src/metag/
sysv.S 177 ! fn was in Arg1 so it's located in in A0FrP+#-0xC
  /external/python/cpython2/Modules/_ctypes/libffi/src/metag/
sysv.S 177 ! fn was in Arg1 so it's located in in A0FrP+#-0xC
  /external/syzkaller/vendor/cloud.google.com/go/storage/
doc.go 66 fmt.Printf("bucket %s, created at %s, is located in %s with storage class %s\n",
  /external/syzkaller/vendor/golang.org/x/oauth2/google/
sdk.go 149 // sdkConfigPath tries to guess where the gcloud config is located.
  /external/u-boot/arch/arm/lib/
lib1funcs.S 41 @ set curbit accordingly. This allows for curbit to be located

Completed in 1414 milliseconds

1 2 3