Commits on Dec 16, 2021

1. perf: use a reusable bytes.Buffer to speed up MutableTree.String …

   The prior code used a naive string concatenation
   str += "..."
   
   which is very slow and inefficient especially when iterating over
   all the keys, but also it performed an unnecessary byteslice->string
   conversion for all arguments inside (nodeDB).traverse* using
   
       str += fmt.Sprintf("%s: %x\n", string(key), value)
   
   notice the `string(key)` passed into fmt.Sprintf("%s"?
   At bare minimum that code should just simply be:
   
       str += fmt.Sprintf("%s: %x\n", key, value)
   
   per https://twitter.com/orijtech/status/1462100381803102213?s=20
   which saves a whole lot of cycles and RAM.
   
   This change uses:
   * (*bytes.Buffer) in combination with fmt.Fprintf
   * A sync.Pool with (*bytes.Buffer) values to reuse buffers
   
   and the results are profound:
   
   ```shell
   $ benchstat before.txt after.txt
   name          old time/op    new time/op    delta
   TreeString-8     111ms ± 8%       4ms ± 4%  -96.01%  (p=0.000 n=20+20)
   
   name          old alloc/op   new alloc/op   delta
   TreeString-8     734MB ± 0%       2MB ± 1%  -99.72%  (p=0.000 n=20+20)
   
   name          old allocs/op  new allocs/op  delta
   TreeString-8     37.0k ± 0%     28.7k ± 0%  -22.40%  (p=0.000 n=20+20)
   ```
   
   Fixes cosmos#455

   

   odeke-em committed Dec 16, 2021
   Configuration menu

   • View commit details
   • Copy full SHA for d2df784
   • Browse repository at this point

   Copy the full SHA

   d2df784 View commit details

   Browse the repository at this point in the history