pixi global list is very slow

[dhirschfeld]

Checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pixi, using pixi --version.

Reproducible example

❯ pixi --version
pixi 0.38.0

❯ time pixi global list
<snip>

real    0m11.746s
user    0m3.372s
sys     0m9.271s

pixi-global.toml

version = 1 [envs.awscli] channels = ["conda-forge"] dependencies = { awscli = "*" } exposed = { aws_completer = "aws_completer", aws_zsh_completer = "aws_zsh_completer", "aws.cmd" = "aws.cmd", aws = "aws", aws_bash_completer = "aws_bash_completer" } [envs.argocd-cli] channels = ["conda-forge"] dependencies = { argocd-cli = "*" } exposed = { argocd = "argocd" } [envs.zoxide] channels = ["conda-forge"] dependencies = { zoxide = "*" } exposed = { zoxide = "zoxide" } [envs.uv] channels = ["conda-forge"] dependencies = { uv = "*" } exposed = { uvx = "uvx", uv = "uv" } [envs.ghostscript] channels = ["conda-forge"] dependencies = { ghostscript = "*" } exposed = { pdf2dsc = "pdf2dsc", ps2ps2 = "ps2ps2", pfbtopfa = "pfbtopfa", ps2epsi = "ps2epsi", ps2pdf = "ps2pdf", unix-lpr = "unix-lpr", printafm = "printafm", gslp = "gslp", pdf2ps = "pdf2ps", ps2pdfwr = "ps2pdfwr", dvipdf = "dvipdf", pf2afm = "pf2afm", ps2ps = "ps2ps", ps2pdf12 = "ps2pdf12", gsnd = "gsnd", gs = "gs", pphs = "pphs", gslj = "gslj", ps2pdf14 = "ps2pdf14", lprsetup = "lprsetup", ps2ascii = "ps2ascii", gsdj500 = "gsdj500", eps2eps = "eps2eps", gsbj = "gsbj", ps2pdf13 = "ps2pdf13", gsdj = "gsdj" } [envs.kubernetes-helm] channels = ["conda-forge"] dependencies = { kubernetes-helm = "*" } exposed = { helm = "helm" } [envs.micromamba] channels = ["conda-forge"] dependencies = { micromamba = "*" } exposed = { conda = "micromamba", micromamba = "micromamba" } [envs.nodejs] channels = ["conda-forge"] dependencies = { nodejs = "*" } exposed = { npm = "npm", npx = "npx", corepack = "corepack", node = "node" } [envs.starship] channels = ["conda-forge"] dependencies = { starship = "*" } exposed = { starship = "starship" } [envs.toolong] channels = ["conda-forge"] dependencies = { toolong = "*" } exposed = { tl = "tl" } [envs.fd-find] channels = ["conda-forge"] dependencies = { fd-find = "*" } exposed = { fd = "fd" } [envs.skopeo] channels = ["conda-forge"] dependencies = { skopeo = "*" } exposed = { skopeo = "skopeo" } [envs.databricks-cli] channels = ["conda-forge"] dependencies = { databricks-cli = "*" } exposed = { databricks = "databricks" } [envs.kubernetes-client] channels = ["conda-forge"] dependencies = { kubernetes-client = "*" } exposed = { kubectl = "kubectl", kubectl-convert = "kubectl-convert" } [envs.terraform] channels = ["conda-forge"] dependencies = { terraform = "*" } exposed = { terraform = "terraform" } [envs.harlequin] channels = ["conda-forge"] dependencies = { harlequin = "*" } exposed = { harlequin = "harlequin" } [envs.git-delta] channels = ["conda-forge"] dependencies = { git-delta = "*" } exposed = { delta = "delta" } [envs.podman] channels = ["conda-forge"] dependencies = { podman = "*" } exposed = { podmansh = "podmansh", podman = "podman", podman-remote = "podman-remote" } [envs.jq] channels = ["conda-forge"] dependencies = { jq = "*" } exposed = { jq = "jq" } [envs.gh] channels = ["conda-forge"] dependencies = { gh = "*" } exposed = { gh = "gh" } [envs.ripgrep] channels = ["conda-forge"] dependencies = { ripgrep = "*" } exposed = { rg = "rg" } [envs.hyperfine] channels = ["conda-forge"] dependencies = { hyperfine = "*" } exposed = { hyperfine = "hyperfine" } [envs.eza] channels = ["conda-forge"] dependencies = { eza = "*" } exposed = { eza = "eza" } [envs.crane] channels = ["conda-forge"] dependencies = { crane = "*" } exposed = { crane = "crane" } [envs.pgcli] channels = ["conda-forge"] dependencies = { pgcli = "*" } exposed = { pgcli = "pgcli" } [envs.opentofu] channels = ["conda-forge"] dependencies = { opentofu = "*" } exposed = { tofu = "tofu" } [envs.go-sqlcmd] channels = ["conda-forge"] dependencies = { go-sqlcmd = "*" } exposed = { sqlcmd = "sqlcmd" } [envs.helix] channels = ["conda-forge"] dependencies = { helix = "*" } exposed = { hx = "hx" } [envs.bat] channels = ["conda-forge"] dependencies = { bat = "*" } exposed = { bat = "bat" } [envs.fzf] channels = ["conda-forge"] dependencies = { fzf = "*" } exposed = { fzf = "fzf" } [envs.pixi-pack] channels = ["conda-forge"] dependencies = { pixi-pack = "*" } exposed = { pixi-pack = "pixi-pack" } [envs.copier] channels = ["conda-forge"] dependencies = { copier = "*", copier-templates-extensions = "*" } exposed = { copier = "copier" } [envs.grayskull] channels = ["conda-forge"] dependencies = { grayskull = "*" } exposed = { grayskull = "grayskull", greyskull = "greyskull", conda-grayskull = "conda-grayskull", conda-greyskull = "conda-greyskull" } [envs.conda-package-handling] channels = ["conda-forge"] dependencies = { conda-package-handling = "*" } exposed = { cph = "cph" } [envs.conda-recipe-manager] channels = ["https://prefix.dev/conda-forge"] dependencies = { conda-recipe-manager = "*" } exposed = { conda-recipe-manager = "conda-recipe-manager", crm = "crm" }

Issue description

Listing the global environment is much slower than it used to be.

Expected behavior

❯ pixi --version
pixi 0.37.0

❯ time pixi global list
<snip>

real    0m2.762s
user    0m1.214s
sys     0m2.927s

[Hofer-Julian]

This is very surprising, thanks for writing this up.

On which platform are you measuring this?

[dhirschfeld]

On which platform are you measuring this?

Ubuntu 22.04.4

[Hofer-Julian]

I can reproduce this.

927 ms with pixi version 0.38
185 ms with pixi version 0.37

[Hofer-Julian]

I cannot reproduce it with current main.
We are planning to release 0.39 today.

@dhirschfeld please let us know if you can then still reproduce this problem

[wolfv]

I tried to fix some of the slowness here and I know how to fix the rest: #2609

[dhirschfeld]

Seems fast again in 0.39.0:

❯ pixi --version
pixi 0.38.0

❯ time pixi global list
<snip>

real    0m13.491s
user    0m3.726s
sys     0m10.620s


❯ pixi --version
pixi 0.39.0

❯ time pixi global list
<snip>

real    0m2.811s
user    0m1.362s
sys     0m3.341s

[Hofer-Julian]

Great!
Let's close this issue then.