[GH-ISSUE #48] drvCirrusLogicGD5434: install fails in movehelp because the help folder is named Help, not DriverHelp #41

New issue

Closed

opened 2026-11-03 22:03:05 -07:00 by gitea-mirror · 1 comment

gitea-mirror commented

2026-11-03 22:03:05 -07:00

Owner

Originally created by @raynorpat on GitHub (Oct 14, 2026).
Original GitHub issue: https://github.com/RhapsodiOS/RhapsodiOS/issues/48

rbuild buildpackage of src/drivers-i386/video/drvCirrusLogicGD5434 fails at install time, after the driver compiles and links:

Copied Localizable.strings TwoMeg.strings PCIOneMB.strings PCITwoMB.strings GD5446_PCIOneMB.strings GD5446_PCITwoMB.strings Help
/bin/mv .../CirrusLogicGD5434DisplayDriver.config/English.lproj/DriverHelp .../CirrusLogicGD5434DisplayDriver.config/English.lproj/Help
/bin/mv: rename .../English.lproj/DriverHelp to .../English.lproj/Help: No such file or directory
make[1]: *** [movehelp] Error 1
make: *** [install@CirrusLogicGD5434.drvproj] Error 2
rbuild: failed with status 2

Cause. driverTools-1/DriverProjectType/driver.make sets SOURCE_HELP_DIR = DriverHelp and HELP_FILE = Help (lines 70–71). Its movehelp rule (line 126) renames each language's DriverHelp to Help in the installed bundle. The Cirrus project ships its help already named Help, at CirrusLogicGD5434.drvproj/English.lproj/Help (CL_GD5434.rtfd, CL_GD5446.rtfd, TableOfContents.rtf). So there is no DriverHelp to move, and the rule's unconditional mv fails.

Most driver projects (33 under src/drivers-i386) ship English.lproj/DriverHelp. drvSB8Sound overrides SOURCE_HELP_DIR in its Makefile.preamble instead.

Possible fixes:

rename English.lproj/Help to English.lproj/DriverHelp in the project (and in PB.project if it lists the resource); or
set SOURCE_HELP_DIR = Help in the project's Makefile.preamble, as drvSB8Sound does.

Context. Seen on 2026-10-14 while rebuilding drivers on the i386 build guest (rhap-i386-bootstrapped-v2.img, rbuild with --toolchain gcc-darwin-i386.conf --arch i386), on branch i386-flat. That branch's only change to this project is one config-table key, which is not involved here; the help layout is the same on master. The compile and kl_ld steps succeed, so the driver binary itself is fine.

Originally created by @raynorpat on GitHub (Oct 14, 2026). Original GitHub issue: https://github.com/RhapsodiOS/RhapsodiOS/issues/48 `rbuild buildpackage` of `src/drivers-i386/video/drvCirrusLogicGD5434` fails at install time, after the driver compiles and links: ``` Copied Localizable.strings TwoMeg.strings PCIOneMB.strings PCITwoMB.strings GD5446_PCIOneMB.strings GD5446_PCITwoMB.strings Help /bin/mv .../CirrusLogicGD5434DisplayDriver.config/English.lproj/DriverHelp .../CirrusLogicGD5434DisplayDriver.config/English.lproj/Help /bin/mv: rename .../English.lproj/DriverHelp to .../English.lproj/Help: No such file or directory make[1]: *** [movehelp] Error 1 make: *** [install@CirrusLogicGD5434.drvproj] Error 2 rbuild: failed with status 2 ``` **Cause.** `driverTools-1/DriverProjectType/driver.make` sets `SOURCE_HELP_DIR = DriverHelp` and `HELP_FILE = Help` (lines 70–71). Its `movehelp` rule (line 126) renames each language's `DriverHelp` to `Help` in the installed bundle. The Cirrus project ships its help already named `Help`, at `CirrusLogicGD5434.drvproj/English.lproj/Help` (`CL_GD5434.rtfd`, `CL_GD5446.rtfd`, `TableOfContents.rtf`). So there is no `DriverHelp` to move, and the rule's unconditional `mv` fails. Most driver projects (33 under `src/drivers-i386`) ship `English.lproj/DriverHelp`. `drvSB8Sound` overrides `SOURCE_HELP_DIR` in its `Makefile.preamble` instead. **Possible fixes:** - rename `English.lproj/Help` to `English.lproj/DriverHelp` in the project (and in `PB.project` if it lists the resource); or - set `SOURCE_HELP_DIR = Help` in the project's `Makefile.preamble`, as drvSB8Sound does. **Context.** Seen on 2026-10-14 while rebuilding drivers on the i386 build guest (`rhap-i386-bootstrapped-v2.img`, rbuild with `--toolchain gcc-darwin-i386.conf --arch i386`), on branch `i386-flat`. That branch's only change to this project is one config-table key, which is not involved here; the help layout is the same on master. The compile and `kl_ld` steps succeed, so the driver binary itself is fine.

gitea-mirror referenced this issue

2026-11-03 22:03:06 -07:00

[GH-ISSUE #47] kernel: i386 raw disk I/O drops the low 512 bytes of the offset (odd-sector reads/writes hit the sector below) #42

gitea-mirror closed this issue

2026-11-03 22:03:06 -07:00

gitea-mirror commented

2026-11-03 22:03:07 -07:00

Author

Owner

@raynorpat commented on GitHub (Oct 14, 2026):

Fixed in 1d3b04e40 on master: the Cirrus GD5434 help folder is now English.lproj/DriverHelp (with its LOCAL_RESOURCES entry updated), matching the other i386 drivers, and SB8's redundant movehelp preamble line is removed. The default movehelp step in driver.make.preamble now handles both. Verified by reading the rule, not by a build.

@raynorpat commented on GitHub (Oct 14, 2026): Fixed in 1d3b04e40 on master: the Cirrus GD5434 help folder is now `English.lproj/DriverHelp` (with its `LOCAL_RESOURCES` entry updated), matching the other i386 drivers, and SB8's redundant `movehelp` preamble line is removed. The default `movehelp` step in `driver.make.preamble` now handles both. Verified by reading the rule, not by a build.

gitea-mirror referenced this issue

2026-11-03 22:03:16 -07:00

[GH-ISSUE #59] kernel: mapped-file shrink then regrow loses retained data on UFS and resurrects stale data on msdosfs #54

gitea-mirror referenced this issue

2026-11-03 22:03:18 -07:00

[GH-ISSUE #59] kernel: mapped-file shrink then regrow loses retained data on UFS and resurrects stale data on msdosfs #54

gitea-mirror referenced this issue

2026-11-03 22:03:31 -07:00

[GH-ISSUE #66] nfs: loopback client loses O_APPEND data, returns ESTALE/EEXIST after local renames, and gives no SIGBUS past a truncated mapping #58