(https://godoc.org/github.com/tensorflow/tensorflow/tensorflow/go)
*   New checkpoint format becomes the default in `tf.train.Saver`. Old V1
    checkpoints continue to be readable; controlled by the `write_version`
    argument, `tf.train.Saver` now by default writes out in the new V2 format.
    It significantly reduces the peak memory required and latency incurred
    during restore.

* Add official Microsoft project generator for [serving spaCy with FastAPI and Azure Cognitive Skills](https://github.com/microsoft/cookiecutter-spacy-fastapi) to [Project Generators](https://fastapi.tiangolo.com/project-generation/). PR [#1390](https://github.com/tiangolo/fastapi/pull/1390) by [@kabirkhan](https://github.com/kabirkhan).

  * You can now set container requests, container limits, and node allocatable reservations for the new `ephemeral-storage` resource.

  * The `ephemeral-storage` resource includes all the disk space a container might consume with container overlay or scratch.

* [alpha] Mount namespace propagation

  * The `VolumeMount.Propagation` field for `VolumeMount` in pod containers is now available.

### Kubernetes Node system swap support

Every system administrator or Kubernetes user has been in the same boat regarding setting up and using Kubernetes: disable swap space. With the release of Kubernetes 1.22, *alpha* support is available to run nodes with swap memory. This change lets administrators opt in to configuring swap on Linux nodes, treating a portion of block storage as additional virtual memory.

- cloud.google.com/go/security: v1.13.0
- cloud.google.com/go/securitycenter: v1.19.0
- cloud.google.com/go/servicedirectory: v1.9.0
- cloud.google.com/go/shell: v1.6.0
- cloud.google.com/go/spanner: v1.45.0
- cloud.google.com/go/speech: v1.15.0
- cloud.google.com/go/storagetransfer: v1.8.0
- cloud.google.com/go/talent: v1.5.0
- cloud.google.com/go/texttospeech: v1.6.0
- cloud.google.com/go/tpu: v1.5.0

(out *Nat) Exp(x *Nat, e []byte, m *Modulus) *Nat { if !m.odd { panic("bigmod: modulus for Exp must be odd") } // We use a 4 bit window. For our RSA workload, 4 bit windows are faster // than 2 bit windows, but use an extra 12 nats worth of scratch space. // Using bit sizes that don't divide 8 are more complex to implement, but // are likely to be more efficient if necessary. table := [(1 << 4) - 1]*Nat{ // table[i] = x ^ (i+1) // newNat calls are unrolled so they are allocated on the stack. NewNat(),...

(out *Nat) Exp(x *Nat, e []byte, m *Modulus) *Nat { if !m.odd { panic("bigmod: modulus for Exp must be odd") } // We use a 4 bit window. For our RSA workload, 4 bit windows are faster // than 2 bit windows, but use an extra 12 nats worth of scratch space. // Using bit sizes that don't divide 8 are more complex to implement, but // are likely to be more efficient if necessary. table := [(1 << 4) - 1]*Nat{ // table[i] = x ^ (i+1) // newNat calls are unrolled so they are allocated on the stack. NewNat(),...

pkg encoding/xml, type EndElement struct
pkg encoding/xml, type EndElement struct, Name Name
pkg encoding/xml, type Name struct
pkg encoding/xml, type Name struct, Local string
pkg encoding/xml, type Name struct, Space string
pkg encoding/xml, type ProcInst struct
pkg encoding/xml, type ProcInst struct, Inst []uint8
pkg encoding/xml, type ProcInst struct, Target string
pkg encoding/xml, type StartElement struct