feat: Add automatic external display switching - #133
Conversation
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
📝 WalkthroughWalkthroughThe change adds external-display auto-switching. The daemon detects connected external displays, changes GPU and mode state, restores prior state after disconnection, and exposes the setting through D-Bus, CLI, GUI, Nix, and documentation. ChangesExternal display auto-switching
Estimated code review effort: 4 (Complex) | ~60 minutes Sequence Diagram(s)sequenceDiagram
participant Display as External display
participant Monitor as Display monitor
participant Mode as ModeInterface
participant GPU as GpuInterface
Display->>Monitor: topology change
Monitor->>Mode: reconcile connection state
Mode->>GPU: update required and block state
Mode-->>Monitor: apply Hybrid or restore mode
Possibly related PRs
Suggested reviewers: 🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
41163fe to
c5ed665
Compare
Signed-off-by: JuanDelPueblo <edyancruz@outlook.com>
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 5
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@crates/cardwire-cli/src/args.rs`:
- Line 114: Update the automatic Hybrid mode descriptions to explicitly say they
apply to dGPU-owned external displays, not internal displays: change the command
description in crates/cardwire-cli/src/args.rs at lines 114-114 and the matching
completion description in crates/cardwire-cli/src/completion.rs at lines 31-31.
In `@crates/cardwire-daemon/src/core/gpu/discover.rs`:
- Line 372: Update the default-GPU selection using max_by_key in the discovery
flow to include the GPU id as a secondary tie-break after
default_gpu_rank(stats). Ensure equal-ranked GPUs consistently select the same
highest or lowest GPU id according to the intended ordering, while preserving
the existing primary rank ordering.
In `@crates/cardwire-daemon/src/interface/config.rs`:
- Around line 148-168: Update set_external_display_auto_switch to avoid leaving
runtime state changed when save_to_file or ObjectServer::interface fails:
resolve the object-server interface before invoking
external_display_setting_changed, and restore the previous mode state if
persistence fails. Preserve the existing error propagation and mode-change
emission behavior for successful updates.
In `@crates/cardwire-daemon/src/interface/mode.rs`:
- Around line 233-246: Update apply_at_startup in
crates/cardwire-daemon/src/interface/mode.rs (lines 233-246) to treat
required_external_cards failures as no external display: log a warning, use
current_mode_value(), and continue without returning the topology error; only
propagate set_mode_value failures. In crates/cardwire-daemon/src/models.rs
(lines 136-141), keep apply_startup_fallback limited to mode-application
failures so topology read errors do not overwrite the persisted mode.
In `@crates/cardwire-daemon/src/tasks/monitor_display.rs`:
- Around line 188-245: Wrap the existing monitor lifecycle in a supervising loop
so transient setup or readiness errors do not terminate display monitoring.
Rename the current body to run_display_monitor, accepting references to mode and
mode_interface, and have monitor_display_changes repeatedly call it, log
failures, wait using a fixed restart backoff, then retry; preserve the existing
event and reconciliation logic inside the worker.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: 35b1a3f2-8b8b-4a1e-8616-ebe6443769b4
📒 Files selected for processing (24)
crates/cardwire-cli/src/args.rscrates/cardwire-cli/src/completion.rscrates/cardwire-cli/src/dbus.rscrates/cardwire-cli/src/main.rscrates/cardwire-daemon/src/core/gpu/discover.rscrates/cardwire-daemon/src/core/gpu/mod.rscrates/cardwire-daemon/src/core/inode.rscrates/cardwire-daemon/src/daemon.rscrates/cardwire-daemon/src/file/config.rscrates/cardwire-daemon/src/interface/config.rscrates/cardwire-daemon/src/interface/mode.rscrates/cardwire-daemon/src/models.rscrates/cardwire-daemon/src/tasks/mod.rscrates/cardwire-daemon/src/tasks/monitor_display.rscrates/cardwire-gui/src/app.rscrates/cardwire-gui/src/helpers/dbus.rscrates/cardwire-gui/src/message.rscrates/cardwire-gui/src/models.rscrates/cardwire-gui/src/subscription.rscrates/cardwire-gui/src/ui.rsdocs/development/dbus.mddocs/getting-started/installation.mddocs/getting-started/usage.mdnix/nixos-module.nix
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
crates/cardwire-daemon/src/interface/config.rs (1)
148-181: 🗄️ Data Integrity & Integration | 🔴 Critical | 🏗️ Heavy liftSnapshot
previous_modeandprevious_auto_switchbefore the lock: race with concurrent mode changes.
previous_mode(Line 158) andprevious_auto_switch(Line 159-162) are read before any lock is held.external_display_setting_changed(mode.rs) only acquirestransition_lockafter this snapshot. If a concurrent mode transition (for example a client-drivenModeproperty write, or the display monitor callingapply_external_display_mode) completes between this snapshot and the internal lock acquisition, andsave_to_filethen fails, the rollback at Line 167-175 reverts to the staleprevious_mode, silently discarding the concurrent transition's result.Capture the previous mode atomically as part of the same locked operation in
ModeInterface, and have it returned to the caller, instead of reading it here before the lock.As per the concurrency-issue guidance for TOCTOU races, this needs a fix that closes the pre-lock read window.
🛠️ Proposed fix: capture previous mode inside the lock
- pub async fn external_display_setting_changed(&self, enabled: bool) -> fdo::Result<bool> { + pub async fn external_display_setting_changed(&self, enabled: bool) -> fdo::Result<(bool, Modes)> { let _transition = self.transition_lock.lock().await; + let previous_mode = self.current_mode_value().await; let was_enabled = self.external_display_auto_switch_enabled(); self.config .external_display_auto_switch .store(enabled, Ordering::Relaxed); if !enabled || was_enabled { - return Ok(false); + return Ok((false, previous_mode)); } let res = async { let cards = self.required_external_cards().await?; let connected = !cards.is_empty(); let mode = self.external_display_target_mode(connected).await?; let changed = self.set_mode_value_locked(mode, false).await?; if connected && changed { Self::notify_drm_change(&cards).await; } Ok(changed) } .await; - if res.is_err() { - self.config - .external_display_auto_switch - .store(was_enabled, Ordering::Relaxed); + match &res { + Ok(changed) => return Ok((*changed, previous_mode)), + Err(_) => { + self.config + .external_display_auto_switch + .store(was_enabled, Ordering::Relaxed); + } } - res + res.map(|changed| (changed, previous_mode)) }pub async fn set_external_display_auto_switch( &self, state: bool, #[zbus(object_server)] object_server: &ObjectServer, ) -> fdo::Result<()> { let interface = object_server .interface::<_, ModeInterface>("/com/github/opengamingcollective/cardwire") .await .map_err(|err| fdo::Error::Failed(err.to_string()))?; - let previous_mode = self.mode_interface.current_mode_value().await; let previous_auto_switch = self .config .external_display_auto_switch .load(Ordering::Relaxed); - let changed = self + let (changed, previous_mode) = self .mode_interface .external_display_setting_changed(state) .await?;🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@crates/cardwire-daemon/src/interface/config.rs` around lines 148 - 181, Update ModeInterface::external_display_setting_changed to acquire transition_lock before capturing the current mode, and return the atomically captured previous mode alongside its changed result. In set_external_display_auto_switch, remove the pre-lock current_mode_value snapshot and use the mode returned by external_display_setting_changed for rollback, while preserving the existing save failure restoration behavior.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Outside diff comments:
In `@crates/cardwire-daemon/src/interface/config.rs`:
- Around line 148-181: Update ModeInterface::external_display_setting_changed to
acquire transition_lock before capturing the current mode, and return the
atomically captured previous mode alongside its changed result. In
set_external_display_auto_switch, remove the pre-lock current_mode_value
snapshot and use the mode returned by external_display_setting_changed for
rollback, while preserving the existing save failure restoration behavior.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: 61d8d940-3bf7-42d7-b1fd-36cfd89a8004
📒 Files selected for processing (6)
crates/cardwire-cli/src/args.rscrates/cardwire-cli/src/completion.rscrates/cardwire-daemon/src/core/gpu/discover.rscrates/cardwire-daemon/src/interface/config.rscrates/cardwire-daemon/src/interface/mode.rscrates/cardwire-daemon/src/tasks/monitor_display.rs
…play_setting_changed
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
crates/cardwire-daemon/src/interface/config.rs (2)
185-191: 🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick winRestore the previous mode after a failed save.
If
save_to_filereturns an error, this setter has already changedexternal_display_auto_switch_mode. A later display disconnect can use the unpersisted mode, while a daemon restart loads the old mode. Restore the previous value before returning the error.🛠️ Proposed fix
pub async fn set_external_display_auto_switch_mode(&self, mode: u32) -> fdo::Result<()> { Modes::try_from(mode).map_err(|err| fdo::Error::InvalidArgs(err.to_string()))?; + let previous_mode = self + .config + .external_display_auto_switch_mode + .load(Ordering::Relaxed); self.config .external_display_auto_switch_mode .store(mode, Ordering::Relaxed); - self.save_to_file().await + if let Err(err) = self.save_to_file().await { + self.config + .external_display_auto_switch_mode + .store(previous_mode, Ordering::Relaxed); + return Err(err); + } + Ok(()) }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@crates/cardwire-daemon/src/interface/config.rs` around lines 185 - 191, Update set_external_display_auto_switch_mode to capture the current external_display_auto_switch_mode before storing the new validated mode, then restore that previous value if save_to_file returns an error before propagating the failure. Preserve the existing validation and successful-save behavior.
163-166: 🗄️ Data Integrity & Integration | 🟠 Major | 🏗️ Heavy liftCast a transactional rollback around
external_display_setting_changed. The setting is restored on its internal error path, butapply_modecan change GPU or BPF state before a later operation fails.mode_stateis saved only afterapply_modesucceeds, and?prevents caller rollback. Restoreprevious_modeand all affected state before returning the error.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@crates/cardwire-daemon/src/interface/config.rs` around lines 163 - 166, Update the apply flow around external_display_setting_changed so failures roll back transactionally: preserve the pre-apply mode and affected GPU/BPF state before apply_mode, restore them when that operation or a later step returns an error, and only commit mode_state after the entire sequence succeeds. Ensure the error is still propagated after rollback rather than allowing ? to bypass restoration.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@crates/cardwire-daemon/src/interface/config.rs`:
- Around line 172-175: Update the rollback handling around
mode_interface.set_mode_value in the save_to_file failure path so restoration
failures are not discarded. Preserve the original save error while propagating
or clearly reporting when restoring previous_mode fails, ensuring callers can
detect that rollback did not complete.
---
Outside diff comments:
In `@crates/cardwire-daemon/src/interface/config.rs`:
- Around line 185-191: Update set_external_display_auto_switch_mode to capture
the current external_display_auto_switch_mode before storing the new validated
mode, then restore that previous value if save_to_file returns an error before
propagating the failure. Preserve the existing validation and successful-save
behavior.
- Around line 163-166: Update the apply flow around
external_display_setting_changed so failures roll back transactionally: preserve
the pre-apply mode and affected GPU/BPF state before apply_mode, restore them
when that operation or a later step returns an error, and only commit mode_state
after the entire sequence succeeds. Ensure the error is still propagated after
rollback rather than allowing ? to bypass restoration.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: 743c32d6-6905-40b7-b3fe-49bd2a15fb81
📒 Files selected for processing (1)
crates/cardwire-daemon/src/interface/config.rs
…ers and mode application
Signed-off-by: JuanDelPueblo <edyancruz@outlook.com>
There was a problem hiding this comment.
Actionable comments posted: 3
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@crates/cardwire-daemon/src/interface/config.rs`:
- Around line 179-181: Update the !state branch in the setting handler to
restore the external display snapshot rather than call
cancel_external_display_snapshot, preserving the user’s prior mode and per-GPU
state when disabling the setting. Use the existing snapshot restoration method
on mode_interface and leave the enabled path unchanged.
In `@crates/cardwire-daemon/src/interface/mode.rs`:
- Around line 212-243: Fix the lock-order inversion between refresh_gpu and
set_required_cards_locked: avoid awaiting block_gpu or unblock_gpu while holding
gpu_list.write(), since those operations acquire pci_list.read() after
refresh_gpu acquires pci_list.write() and then gpu_list.write(). Restructure the
set_required_cards_locked flow to determine and record required GPU updates
under the GPU lock, release that lock, then perform blocking/unblocking
operations in the established pci_list-before-gpu_list order while preserving
the existing state and restore_departed behavior.
- Around line 535-544: Update the mode-change flow around set_mode_value_locked
and external_display_auto_switch_enabled so explicit non-manual changes
(Integrated, Smart, and Hybrid) also restore external-display requirements when
a dGPU display is connected. Preserve the existing manual-mode behavior, and
ensure external_display_required is reapplied before reconciliation can be
skipped due to last_connected being true.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: b3e70897-dc35-4bb5-bfe8-409fe701b1eb
📒 Files selected for processing (23)
crates/cardwire-cli/src/args.rscrates/cardwire-cli/src/completion.rscrates/cardwire-cli/src/dbus.rscrates/cardwire-cli/src/main.rscrates/cardwire-daemon/src/daemon.rscrates/cardwire-daemon/src/file/config.rscrates/cardwire-daemon/src/interface/config.rscrates/cardwire-daemon/src/interface/debug.rscrates/cardwire-daemon/src/interface/gpu.rscrates/cardwire-daemon/src/interface/mode.rscrates/cardwire-daemon/src/models.rscrates/cardwire-daemon/src/tasks/battery_switch.rscrates/cardwire-daemon/src/tasks/monitor_display.rscrates/cardwire-gui/src/app.rscrates/cardwire-gui/src/helpers/dbus.rscrates/cardwire-gui/src/message.rscrates/cardwire-gui/src/models.rscrates/cardwire-gui/src/subscription.rscrates/cardwire-gui/src/ui.rsdocs/development/dbus.mddocs/getting-started/installation.mddocs/getting-started/usage.mdnix/nixos-module.nix
💤 Files with no reviewable changes (7)
- crates/cardwire-cli/src/main.rs
- crates/cardwire-gui/src/message.rs
- crates/cardwire-gui/src/app.rs
- crates/cardwire-gui/src/models.rs
- crates/cardwire-gui/src/subscription.rs
- crates/cardwire-cli/src/dbus.rs
- docs/getting-started/installation.md
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
crates/cardwire-daemon/src/interface/mode.rs (2)
568-582: 🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick winA stumble in bookkeeping turns "Integrated" into "Hybrid" forever.
When
set_modeapplies a non-Manual mode (Integrated/Smart) while a required external display is connected, it callsapply_external_display_cards_locked(&cards, false)at Line 577.capture_snapshot=falseskipscapture_external_display_snapshot_locked, solatest_modestaysNone. Inside that call,target = Hybridgets persisted throughset_mode_value_locked(target, false), overwriting the user's just-saved "Integrated" choice inmode.jsonwith "Hybrid," with no snapshot to restore from once the display disconnects.clear_external_display_snapshot_locked()at Line 573 also wipes any prior snapshot before the reapply, so an existing snapshot cannot survive this call either.The same conjuring happens at startup:
apply_at_startup(Line 423) calls the identicalapply_external_display_cards_locked(&cards, false). If the persisted mode is Integrated/Smart and the display is already connected at boot, the daemon silently rewrites the on-disk mode to Hybrid every time it starts, with no way back.Capture a snapshot of the just-applied mode before handing control to Hybrid, mirroring the connect-time flow in
apply_external_display_mode.🪄 Proposed fix for `set_mode`
pub(crate) async fn set_mode(&self, mode: u32) -> fdo::Result<()> { // Valide inputs and turn into a Modes let mode = Modes::try_from(mode).map_err(|err| fdo::Error::InvalidArgs(err.to_string()))?; let _transition = self.transition_lock.lock().await; self.set_mode_value_locked(mode, false).await?; - self.clear_external_display_snapshot_locked().await; - if self.external_display_auto_switch_enabled() { - let cards = self.required_external_cards().await?; - if !cards.is_empty() { - self.apply_external_display_cards_locked(&cards, false) - .await?; - } - } + if self.external_display_auto_switch_enabled() { + let cards = self.required_external_cards().await?; + if !cards.is_empty() { + let capture_snapshot = !self.external_display_override_active_locked().await; + self.apply_external_display_cards_locked(&cards, capture_snapshot) + .await?; + } else { + self.clear_external_display_snapshot_locked().await; + } + } else { + self.clear_external_display_snapshot_locked().await; + } Ok(()) }Apply the same
capture_snapshotlogic to theapply_at_startupcall at Line 423.Also applies to: 416-431
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@crates/cardwire-daemon/src/interface/mode.rs` around lines 568 - 582, Update the external-display application flows in set_mode and apply_at_startup to call apply_external_display_cards_locked with snapshot capture enabled, matching the connect-time behavior in apply_external_display_mode. Preserve the existing mode transition and card application logic while ensuring the newly selected Integrated/Smart mode is snapshotted before Hybrid is applied.
357-393: 🎯 Functional Correctness | 🟠 Major | ⚡ Quick winDisabling the ward does not release what it was guarding.
external_display_setting_changed's!enabledbranch (Lines 367-369) storesfalseinto the config and returns without touchinglatest_modeor any GPU'sexternal_display_requiredflag. If the setting was actively protecting a GPU (a display is connected andexternal_display_requiredis true) when the user disables the feature, that state never clears.external_display_override_active_locked()keeps returning true afterward, soset_battery_mode_valuekeeps rejecting battery-triggered mode changes even though external-display auto-switch is now off.Restore or clear the snapshot when the feature transitions from enabled to disabled.
🪄 Proposed fix
if !enabled { + if was_enabled { + if let Err(err) = self.restore_external_display_snapshot_locked().await { + warn!("failed to restore external-display state while disabling: {err}"); + } + } return Ok((false, previous_mode)); }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@crates/cardwire-daemon/src/interface/mode.rs` around lines 357 - 393, Update external_display_setting_changed so the enabled-to-disabled transition restores or clears the external-display snapshot before returning, releasing latest_mode and each GPU’s external_display_required state. Preserve the existing no-op behavior when already disabled and return the previous mode with the transition result after cleanup.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Outside diff comments:
In `@crates/cardwire-daemon/src/interface/mode.rs`:
- Around line 568-582: Update the external-display application flows in set_mode
and apply_at_startup to call apply_external_display_cards_locked with snapshot
capture enabled, matching the connect-time behavior in
apply_external_display_mode. Preserve the existing mode transition and card
application logic while ensuring the newly selected Integrated/Smart mode is
snapshotted before Hybrid is applied.
- Around line 357-393: Update external_display_setting_changed so the
enabled-to-disabled transition restores or clears the external-display snapshot
before returning, releasing latest_mode and each GPU’s external_display_required
state. Preserve the existing no-op behavior when already disabled and return the
previous mode with the transition result after cleanup.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: c0e4cef3-308d-4275-880c-3912ec7ece84
📒 Files selected for processing (1)
crates/cardwire-daemon/src/interface/mode.rs
| #[zbus(property)] | ||
| pub async fn set_external_display_auto_switch( | ||
| &self, | ||
| state: bool, | ||
| #[zbus(object_server)] object_server: &ObjectServer, | ||
| ) -> fdo::Result<()> { | ||
| let interface = object_server | ||
| .interface::<_, ModeInterface>("/com/github/opengamingcollective/cardwire") | ||
| .await | ||
| .map_err(|err| fdo::Error::Failed(err.to_string()))?; | ||
| let previous_auto_switch = self | ||
| .config | ||
| .external_display_auto_switch | ||
| .load(Ordering::Relaxed); | ||
| let (changed, previous_mode) = self | ||
| .mode_interface | ||
| .external_display_setting_changed(state) | ||
| .await?; | ||
| if let Err(err) = self.save_to_file().await { | ||
| self.config | ||
| .external_display_auto_switch | ||
| .store(previous_auto_switch, Ordering::Relaxed); | ||
| if state | ||
| && !previous_auto_switch | ||
| && let Err(rollback_err) = self | ||
| .mode_interface | ||
| .restore_external_display_snapshot() | ||
| .await | ||
| { | ||
| warn!( | ||
| "failed to restore display state during config save rollback: {rollback_err}" | ||
| ); | ||
| if changed | ||
| && let Err(mode_err) = self | ||
| .mode_interface | ||
| .set_mode_value(previous_mode, false) | ||
| .await | ||
| { | ||
| warn!("failed to restore mode value during config save rollback: {mode_err}"); | ||
| } | ||
| } | ||
| return Err(err); | ||
| } | ||
| if !state { | ||
| self.mode_interface.cancel_external_display_snapshot().await; | ||
| } | ||
| self.mode_interface | ||
| .emit_mode_change(&interface, changed) | ||
| .await | ||
| .map_err(|err| fdo::Error::Failed(err.to_string()))?; | ||
| Ok(()) | ||
| } |
There was a problem hiding this comment.
This function is doing far too much, config.rs should only be used as simple setters/getters for the config, in this case true or false.
There was a problem hiding this comment.
Understood, simplified to just a boolean setter/getter. Will now close this PR and split it
Signed-off-by: JuanDelPueblo <edyancruz@outlook.com>
2 participants
Description
Add configurable automatic GPU handling for external displays connected through dGPU-owned DRM connectors.
The daemon discovers connector ownership, monitors display topology changes through udev, and remembers the current mode and per-GPU state. Manual mode stays active and temporarily unblocks the required GPU; other modes switch to Hybrid. The previous state is restored after disconnect unless the user changes it manually.
The feature is disabled by default and exposed through D-Bus, the CLI, the GUI, and the NixOS module. External-display handling takes priority over battery auto-switching.
Related to #7.
TODO
Checklist:
Summary by CodeRabbit