public inbox for gentoo-dev@lists.gentoo.org
 help / color / mirror / Atom feed
From: "Michał Górny" <mgorny@gentoo.org>
To: gentoo-dev@lists.gentoo.org
Cc: "Robin H. Johnson" <robbat2@gentoo.org>
Subject: Re: [gentoo-dev] [PATCH v2 1/2] check-reqs.eclass: runtime disk checks for any path.
Date: Mon, 26 Feb 2024 14:50:43 +0100	[thread overview]
Message-ID: <2d7cf2ad7b4ccd33dc2cf5d56f1315b289fa599d.camel@gentoo.org> (raw)
In-Reply-To: <20240226063105.1767-1-robbat2@gentoo.org>

[-- Attachment #1: Type: text/plain, Size: 3336 bytes --]

On Sun, 2024-02-25 at 22:31 -0800, Robin H. Johnson wrote:
> Allow checking any runtime path for installing ever-larger packages.
> 
> CHECKREQS_DISK_RUNTIME=( /boot:40M /:350M /opt:500M )
> 
> Recent example of large packages:
> 
> gentoo-kernel-bin:
> /     >=350MB/version (in /lib/modules)
> /boot >=40MB/version
> 
> rust-bin:
> /opt  >=450MB/version
> 
> Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
> ---
>  eclass/check-reqs.eclass | 23 +++++++++++++++++++++++
>  1 file changed, 23 insertions(+)
> 
> diff --git a/eclass/check-reqs.eclass b/eclass/check-reqs.eclass
> index fac2f4553d74..1c59c69489a9 100644
> --- a/eclass/check-reqs.eclass
> +++ b/eclass/check-reqs.eclass
> @@ -30,6 +30,13 @@
>  # # install will need this much space in /var
>  # CHECKREQS_DISK_VAR="1024M"
>  #
> +# # install will need this much space in listed paths.
> +# CHECKREQS_DISK_RUNTIME=(
> +#   /var:1G
> +#   /boot/efi:32M

I'd avoid listing /boot/efi as an example, as /boot is a bit special
and might need special handling in the eclass.  In particular,
on the system here I have EFI mounted at /boot, so there
is no /boot/efi.

A possible generic solution would be to "fall back" from non-existing
locations to a "higher" directory, assuming they would normally be
created as subdirectories.

> +#   /opt/giant-package-with-dedicated-disk:100G
> +# )
> +#
>  # @CODE
>  #
>  # If you don't specify a value for, say, CHECKREQS_MEMORY, then the test is not
> @@ -66,6 +73,11 @@ _CHECK_REQS_ECLASS=1
>  # @DESCRIPTION:
>  # How much space is needed in /var? Eg.: CHECKREQS_DISK_VAR=3000M
>  
> +# @ECLASS_VARIABLE: CHECKREQS_DISK_RUNTIME
> +# @DEFAULT_UNSET
> +# @DESCRIPTION:
> +# How much space is needed in paths? Eg.: CHECKREQS_DISK_RUNTIME=( /:1G /var:5G )
> +
>  # @ECLASS_VARIABLE: CHECKREQS_DONOTHING
>  # @USER_VARIABLE
>  # @DEFAULT_UNSET
> @@ -120,6 +132,7 @@ _check-reqs_prepare() {
>  	debug-print-function ${FUNCNAME} "$@"
>  
>  	if [[ -z ${CHECKREQS_MEMORY} &&
> +			"${#CHECKREQS_DISK_RUNTIME[@]}" -eq 0 &&
>  			-z ${CHECKREQS_DISK_BUILD} &&
>  			-z ${CHECKREQS_DISK_USR} &&
>  			-z ${CHECKREQS_DISK_VAR} ]]; then

Considering all the extra logic discussed in this thread, it might be
reasonable to implicitly convert CHECKREQS_DISK_* into
CHECKREQS_DISK_RUNTIME, so they'd share all the solutions discussed.

So ideally the logic would be something like:

1. Append CHECKREQS_DISK_* into CHECKREQS_DISK_RUNTIME.

2. Replace missing paths with the first parent directory that exists.

3. Replace paths with their respective mount points.

4. Sum the values corresponding to the same mount point.

> @@ -161,6 +174,16 @@ _check-reqs_run() {
>  	fi
>  
>  	if [[ ${MERGE_TYPE} != buildonly ]]; then
> +		if [[ "${#CHECKREQS_DISK_RUNTIME[@]}" -gt 0 ]]; then
> +			for _path_size in "${CHECKREQS_DISK_RUNTIME[@]}"; do
> +				_path=${_path_size/:*}
> +				_size=${_path_size/*:}
> +				_check-reqs_disk \
> +					"${EROOT%/}${_path}" "${_size}"
> +			done
> +			unset _path_size _path _size

Instead of setting them globally, then unsetting, you should use local
variables.

> +		fi
> +
>  		[[ -n ${CHECKREQS_DISK_USR} ]] && \
>  			_check-reqs_disk \
>  				"${EROOT%/}/usr" \

-- 
Best regards,
Michał Górny


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 512 bytes --]

  parent reply	other threads:[~2024-02-26 13:50 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-02-19 22:08 [gentoo-dev] [PATCH] check-reqs.eclass: more disk checks Robin H. Johnson
2024-02-19 22:14 ` [gentoo-dev] " Robin H. Johnson
2024-02-20  6:16   ` Michał Górny
2024-02-20  4:38 ` [gentoo-dev] " Sam James
2024-02-20  5:53 ` Andrew Ammerlaan
2024-02-26  6:31 ` [gentoo-dev] [PATCH v2 1/2] check-reqs.eclass: runtime disk checks for any path Robin H. Johnson
2024-02-26  7:01   ` zzam
2024-02-26  9:46     ` Robin H. Johnson
2024-02-26 13:50   ` Michał Górny [this message]
2024-02-26  6:31 ` [gentoo-dev] [PATCH v2 2/2] sys-firmware/intel-microcode: check-reqs for /boot space Robin H. Johnson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=2d7cf2ad7b4ccd33dc2cf5d56f1315b289fa599d.camel@gentoo.org \
    --to=mgorny@gentoo.org \
    --cc=gentoo-dev@lists.gentoo.org \
    --cc=robbat2@gentoo.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox