Re: [PATCH 00/22] avoid redundant pipelines

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



"Beat Bolli" <bb@xxxxxxxxx> writes:

> As follow-up to [1], here's a series that eliminates redundant
> pipelines, mostly under t/.
>
> Spawning a process is notoriously slow on Windows, so this will improve
> test performance.

Yay!

> 1/22 to 14/22 remove redundant uses of "cat" where the shell handles
> input redirection from a file or the command can take a file argument.
>
> 15/22 to 20/22 merge redundant uses of "grep" into the following "sed"
> or "awk" command. Both of these are capable of filtering themselves. I'm
> a bit on the fence about this part because readability suffers in some
> cases. It was a fun exercise, though :-)
>
> 21/22 merges multiple "sed" calls into one and replaces a whole "cat |
> grep | awk" construct with a single "awk" call and uses "sort -u"
> instead of "sort | uniq".
>
> 22/22 finally is a bit of an outlier in that it replaces a subshell with
> a shell compound command.
>
> In the tests, I have completely arbitrarily separated the commits into
> thousands blocks to keep the commits and their reviews manageable.
>
> All tests that I was able to run still pass. I don't have p4 or cvs
> installed.
>
> [1] https://lore.kernel.org/git/20240216171046.927552-1-dev+git@xxxxxxxxx/

Thanks.




[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]

  Powered by Linux