reduce submodule clone time and size using depth option

[karaketir16]

This will decrease the required clone size and so reduce the overall build time.

[karaketir16]

Let's wait for the workflow completion for comparision

[karaketir16]

Some remotes don’t support cloning with the --depth 1 option. I’ll investigate.

[karaketir16]

Comparison of Cloning the binutils Submodule

Without --depth 1:

With --depth 1:

Results:

• Object Count: Reduced drastically from 1,239,461 to 40,643.
• Download Size: Dropped from approximately 450 MiB (full clone) to 78.15 MiB (shallow clone).

Note: I did not wait this time for cloning of the full history. here an old log from workflow:

[TommyMurphyTM1234]

I presume that you've seen this existing open PR already?

• https://github.com/riscv-collab/riscv-gnu-toolchain/pull/1605

Some remotes don’t support cloning with the --depth 1 option. I’ll investigate.

I thought that uClibc was the only one here that had a problem with it?

• https://github.com/riscv-collab/riscv-gnu-toolchain/pull/1605#issuecomment-2452734283

[karaketir16]

No, I did not see other pull request. I wrote a script, thats try full clone, if shallow fails.

[mickflemm]

This is already done here, for all modules including uclibc: https://github.com/riscv-collab/riscv-gnu-toolchain/pull/1608/commits/efc7cc9c828050d99125b0b6054a4ab80901adc8