Activity for Roderick W. Smith

  • Roderick W. Smith Roderick W. Smith committed [94ba63] on Code

    Do more to fix the refind-mkdefault bug involving a bad BootOrder variable

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    rEFInd 0.14.2 introduced some significant changes to the way it scans for disk-based tools. I did this for various reasons -- partly to simplify the code, partly to deal with changes in the names and locations of some tools (most notably, various memtest tools). One consequence, however, is that rEFInd may detect a slightly different set of tools, potentially increasing or decreasing the number of tools, depending on what's available on a system. That said, some of what you've reported seems strange,...

  • Roderick W. Smith Roderick W. Smith committed [6a00db] on Code

    Fixed bug in refind-mkdefault if BootOrder has invalid entry.

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    rEFInd relies on various on-disk files for much of its functionality, so if it's started over a network, it won't know where to look to find those files -- and they may not be available at all. I tested this years ago, and IIRC, at the time it would sort-of work, but without icons and with only the defaults from refind.conf (since of course it couldn't load that file). It may be that it's gotten worse -- or maybe not. What happens when you do as you're prompted and press a key to continue? If rEFInd...

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    rEFInd automatically filters out duplicate partitions based on the filesystem's UUID or serial number. This is done so as to prevent the display of duplicate entries in software RAID 1 arrays. Chances are this is what's causing your issue -- your backup partition likely contains a duplicate serial number. You might want to check your backup software to see if it has an option to not duplicate the serial number when cloning a partition. As the point of a filesystem serial number is to uniquely identify...

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    I do use git tag on release versions; the latest has a tag of "v.0.14.2".

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    First, you can always set icons_dir in refind.conf to point to a replacement or supplemental directory for icons. These will not be replaced when rEFInd is updated. This option is most commonly used in theming operations, but you can use it to add one or two icons, if you like. Second, I've just added an ArcaOS icon to the rEFInd git repository. I picked a different one than you presented for licensing reasons and because it better fills the square; however, the original was just 80x80, and so it...

  • Roderick W. Smith Roderick W. Smith committed [6c66e9] on Code

    Version 0.14.2 release

  • Roderick W. Smith Roderick W. Smith committed [49e4fc] on Code

    Makefile changes for GNU-EFI:

  • Roderick W. Smith Roderick W. Smith committed [f61d7b] on Code

    Theoretically add gzip.lib to 'make tiano' build target; but that target is deprecated and doesn't work for me.

  • Roderick W. Smith Roderick W. Smith committed [95e71c] on Code

    Remove commented-out code; reformat a few lines.

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    I've pushed my changes to the git repository. Could you please try them, @nl6720? They work for me on Ubuntu 22.04 on x86-64, Ubuntu 24.04 pre-release on ARM64, and Arch Linux on ARM64. Both the Ubuntus used a locally-built GNU-EFI 3.0.18, but the Arch used the Arch GNU-EFI 3.0.18 package. I'm afraid I don't have a working Arch installation on x86-64 for testing. Note that the ARM64 build change requires editing Make.common or passing GNUEFI_ARM64_TARGET_SUPPORT=y when using GNU-EFI 3.0.17 or later;...

  • Roderick W. Smith Roderick W. Smith committed [c1a97f] on Code

    Accommodate change in GNU-EFI 3.0.17 and later build options for ARM64

  • Roderick W. Smith Roderick W. Smith committed [189e40] on Code

    Worked around problems with GNU-EFI 3.0.18

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    I think I've got this sorted out. There are actually three problems: * The change you identified that causes failures to launch if -j .rodata is not used with objcopy. This was the easiest to deal with. * The commit I identified, which changed the way CopyMem() was defined. This one's weird; it was as if the rEFInd build process was ignoring the EFIAPI option on that function, thus calling it with the wrong ABI. Maybe it was; the rEFInd build process is pretty convoluted. I've worked around the problem...

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    Further information: It's also necessary to add -j .rodata as an objcopy option. This may well be related to the commit you identified, @nl6720. (I'd been experimenting with that and forgot I had it set when I found the EFIAPI commit.)

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    To quote Marge Gunderson ("Fargo"), "I'm not sure I agree with you a hundred percent on your police work there, Lou." Poking through the Sourceforge git repo for GNU-EFI, I find that the following commit is the troublesome one: https://sourceforge.net/p/gnu-efi/code/ci/deb8a7f267d96299b9aa41300539f617db54c2a9 All commits prior to that one work fine for me, but once I build with that one, it fails in the way you describe. The commit in question changes SetMem() and CopyMem() to use the EFIAPI calling...

  • Roderick W. Smith Roderick W. Smith committed [a9bf0f] on Code

    Fixed 'fat' Apple binary detection bug

  • Roderick W. Smith Roderick W. Smith committed [99db08] on Code

    Minor logging bug fix & optimization in IsValidLoader()

  • Roderick W. Smith Roderick W. Smith committed [68e9b4] on Code

    Fix typo in refind-sb-healthcheck message

  • Roderick W. Smith Roderick W. Smith committed [426d5e] on Code

    Forgot to save NEWS.txt after editing it....

  • Roderick W. Smith Roderick W. Smith committed [786e0c] on Code

    Added Shim information to Info/About screen

  • Roderick W. Smith Roderick W. Smith committed [fba47c] on Code

    Code optimizations, mostly in scan.c's FindTool() and IsValidTool()

  • Roderick W. Smith Roderick W. Smith committed [6cbc40] on Code

    Restructured tool-scanning code; added also_scan_tool_dirs token

  • Roderick W. Smith Roderick W. Smith committed [e43875] on Code

    Remove a few commented-out lines

  • Roderick W. Smith Roderick W. Smith committed [fce19f] on Code

    Improved RPM & Debian packaging, particularly related to Secure Boot

  • Roderick W. Smith Roderick W. Smith committed [e33baf] on Code

    Version 0.14.1 final release

  • Roderick W. Smith Roderick W. Smith committed [f5f5a7] on Code

    Updated LodePNG to version 20230410

  • Roderick W. Smith Roderick W. Smith committed [2f6cab] on Code

    Remove bootx64.efi (etc.) as valid memtest filenames

  • Roderick W. Smith Roderick W. Smith committed [cea1b8] on Code

    Update of copyright dates in multiple files

  • Roderick W. Smith Roderick W. Smith committed [63ce72] on Code

    Remove duplicate tools caused by duplicate tool locations

  • Roderick W. Smith Roderick W. Smith committed [e85de9] on Code

    Move memtest from boot loader to tools menu if found in a boot loader location

  • Roderick W. Smith Roderick W. Smith committed [f8ec4d] on Code

    Fixed bug in refind-install that caused --ownhfs option to fail

  • Roderick W. Smith Roderick W. Smith committed [0e13e9] on Code

    Fixed bug in display of Unicode characters greater than 0x10000

  • Roderick W. Smith Roderick W. Smith posted a comment on merge request #49

    Thanks for the submission. I've accepted it, but added a drop shadow to the icon, to match other rEFInd icons.

  • Roderick W. Smith Roderick W. Smith updated merge request #49

    Adds icon for EndeavourOS

  • Roderick W. Smith Roderick W. Smith committed [10fb85] on Code

    add the EndeavourOS icon

  • Roderick W. Smith Roderick W. Smith committed [de851b] on Code

    Document recent Btrfs RAID code submission in NEWS.txt

  • Roderick W. Smith Roderick W. Smith posted a comment on merge request #48

    Thanks for the contribution!

  • Roderick W. Smith Roderick W. Smith merged merge request #48 on Code

    port RAID1C3/4 support from grub2

  • Roderick W. Smith Roderick W. Smith committed [53ac72] on Code

    Version 1.0.10 release

  • Roderick W. Smith Roderick W. Smith updated merge request #32

    implement k/--move-backup-table/MoveSecondTable

  • Roderick W. Smith Roderick W. Smith posted a comment on merge request #32

    Sorry for the late reply; I've been pretty busy lately. I've merged this and added the changes I've suggested myself.

  • Roderick W. Smith Roderick W. Smith committed [904c35] on Code

    Fine-tuning and documentation of new function to move the secondary/backup partition table

  • Roderick W. Smith Roderick W. Smith committed [4c956f] on Code

    implement k/--move-backup-table/MoveSecondTable

  • Roderick W. Smith Roderick W. Smith committed [bffbf3] on Code

    Fixed accidental inclusion of stray code lines

  • Roderick W. Smith Roderick W. Smith committed [e80ef3] on Code

    New partition type codes from the Discoverable Partitions Specification

  • Roderick W. Smith Roderick W. Smith committed [cd92ef] on Code

    Updated NEWS for recent merge

  • Roderick W. Smith Roderick W. Smith merged merge request #36 on Code

    Minor enhancments

  • Roderick W. Smith Roderick W. Smith posted a comment on merge request #32

    Thanks for the submission; however, I see two problems with it as-is: I don't see an obvious way to detect when the backup partition table is in an odd location. When parted is fed a partition table that's been moved by your code, for instance, it warns that this has been done and offers to fix it. Currently, gdisk includes no such checks, so it won't detect such a move, even though improper use of the facility might result in severely reduced available disk space. I suggest adding a check to the...

  • Roderick W. Smith Roderick W. Smith committed [d5c0a4] on Code

    Add linux_prefixes config file option & change bzImage to Image as kernel ID on ARM64 systems

  • Roderick W. Smith Roderick W. Smith committed [f9a750] on Code

    More logging about finding filesystem types

  • Roderick W. Smith Roderick W. Smith committed [0a4395] on Code

    Excise CSM boot support from non-x86/x86-64 builds

  • Roderick W. Smith Roderick W. Smith committed [5e2d0a] on Code

    Fixed bug that could cause freeze when doing a self-install

  • Roderick W. Smith Roderick W. Smith committed [c97c27] on Code

    Improve log file rotation error checking

  • Roderick W. Smith Roderick W. Smith committed [314dc3] on Code

    Properly log gDS warning

  • Roderick W. Smith Roderick W. Smith committed [9b125c] on Code

    Improve detection of EFI program validity

  • Roderick W. Smith Roderick W. Smith committed [ed4b31] on Code

    Revert most of the previous commit, since it turns out it doesn't work. :(

  • Roderick W. Smith Roderick W. Smith committed [6d8770] on Code

    Fixed problems with Tow-Boot that could cause crashes when loading drivers or re-scanning for boot loaders

  • Roderick W. Smith Roderick W. Smith committed [cb4bf3] on Code

    Minor code cleanup based on valgrind analysis

  • Roderick W. Smith Roderick W. Smith committed [8825e5] on Code

    Fixed JPEG problems: Memory leak & garbled JPEG when compiled on some distributions

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    I have essentially no experience with BitLocker, so I can't offer anything more than speculative advice. That said, my questions and thoughts are these: What happens if you boot from rEFInd to a Linux kernel, vs. rEFInd to GRUB to a Linux kernel? I wouldn't expect there to be a difference, but if there is, that would be informative. Does booting Windows through rEFInd, and then rebooting Windows again through rEFInd, without booting Fedora, cause the recovery key to be needed? In normal operation...

  • Roderick W. Smith Roderick W. Smith committed [8aca5d] on Code

    Fix bug in refind-install discovered after major release.

  • Roderick W. Smith Roderick W. Smith committed [7f918c] on Code

    Added information on Fedora's rEFInd package to the documentation

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    To quote Homer Simpson, "d'oh!" That bug actually existed in 0.13.3 and forced the release of 0.13.3.1. It recurred because, in both cases, it was caused by my accepting bad advice from the shellcheck shell script analysis tool. I've just released a 0.14.0.2 version to address this problem. I've also added comments to the script to deter me from making the same mistake a third time.

  • Roderick W. Smith Roderick W. Smith committed [0e7d63] on Code

    Document previous merge

  • Roderick W. Smith Roderick W. Smith committed [e1cc65] on Code

    Truncate decimal inputs (e.g., '9.5G' becomes '9G')

  • Roderick W. Smith Roderick W. Smith committed [42eea8] on Code

    Do some explicit casts in gptcurses.cc to eliminate compiler warnings.

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    You launch gdisk against the whole disk (e.g., /dev/sda in Ubuntu or /dev/disk0 is macOS), NOT against an individual partition (e.g., /dev/sda1 in Ubuntu or /dev/disk0s1 in macOS). You did the latter, so gdisk gave no useful output. If Ubuntu is installed on your /dev/disk0 (in macOS), then the diskutil output makes it look like it's correctly marked, but I'm less familiar with what diskutil works, so I can't be sure of that; I'd still like to see confirmation from gdisk. Another possibility is that...

  • Roderick W. Smith Roderick W. Smith committed [e7a566] on Code

    Document recent mergest

  • Roderick W. Smith Roderick W. Smith posted a comment on merge request #31

    Thanks for the improvement!

  • Roderick W. Smith Roderick W. Smith merged merge request #31 on Code

    Allow partition dynamically allocated by --largest-new to be referenced by other options

  • Roderick W. Smith Roderick W. Smith posted a comment on merge request #30

    Thanks for the submission; however, it's not acceptable as-is. It works fine for gdisk, but the cgdisk, the prompt is ugly and accepts no input, thus freezing the program; and it breaks the flow in the command-line, non-interactive sgdisk with an unacceptable prompt. You'll need to change the structure of the prompting for each program (although it's possible you could create one to work for both cgdisk and gdisk and another for sgdisk). For sgdisk in particular, you'll need to add a new command-line...

  • Roderick W. Smith Roderick W. Smith merged merge request #29 on Code

    Use 64bit off_t on 32bit linux systems

  • Roderick W. Smith Roderick W. Smith posted a comment on merge request #29

    LGTM. Thanks.

  • Roderick W. Smith Roderick W. Smith committed [3c4a9f] on Code

    Document recent merge.

  • Roderick W. Smith Roderick W. Smith merged merge request #28 on Code

    Fix NULL dereference when duplicating string argument

  • Roderick W. Smith Roderick W. Smith posted a comment on merge request #28

    LGTM. Thanks!

  • Roderick W. Smith Roderick W. Smith committed [47af96] on Code

    Bump the version up artificially to make Launchpad happy

  • Roderick W. Smith Roderick W. Smith committed [11a49d] on Code

    Fixed missing refind-sb-healthcheck from .zip and native .deb packages

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    I've just released rEFInd 0.14.0. It's available from the downloads section here on Sourceforge, and should be available soon from the Ubuntu PPA (it's built there for Ubuntu 22.04, as of 3:54 PM US Eastern Time on March 4; but builds for other versions are still pending). See the revisions page of the documentation for details of the changes. In brief, new features include: the ability to launch gzipped kernels (which is important on ARM64, but not on x86 or x86-64) Secure Boot Advanced Targeting...

  • Roderick W. Smith Roderick W. Smith committed [bf6c58] on Code

    Final changes for version 0.14.0

  • Roderick W. Smith Roderick W. Smith committed [e59131] on Code

    Minor tweaks, mostly to refind-sb-healthcheck output

  • Roderick W. Smith Roderick W. Smith committed [120159] on Code

    Minor cleanup of mkrlconf script

  • Roderick W. Smith Roderick W. Smith committed [393cf4] on Code

    Light code cleanup on refind-install script

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    Yeah, there's no way you've got both macOS and Ubuntu installed on that one disk. You can probably figure out where Ubuntu is with df, as in: $ df / Filesystem 1K-blocks Used Available Use% Mounted on /dev/sda2 36937728 10000696 26771048 28% / This example shows that Linux is on /dev/sda2. There are exceptions to this rule, though. If you used LVM for your setup, then you won't see a raw disk device filename; instead, you'd see something like /dev/mapper/ubuntu-root, which isn't very informative...

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    That might work for you on your installation, but it doesn't do much good for anybody else (unless they happen to read this thread). What I need to know to help other people is whether version 0.13.3.8 of rEFInd (or its ext4fs_x64.efi driver file, to be more precise) fixes the problem for the setup where you encountered the problem but without your modifications. It's worked for me in my tests using a local build of e2fsprogs 1.47.0 on an Ubuntu 22.04 system, but that's different from your use case,...

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    I'm glad you got the boot order issue sorted out! As to macOS seeing the Ubuntu partition(s) as unreadable macOS partitions, that's most likely an issue with the partition type codes. You can change those with gdisk in Ubuntu (or in macOS, if you install it in macOS). For instance: $ sudo gdisk /dev/sda GPT fdisk (gdisk) version 1.0.8 Partition table scan: MBR: protective BSD: not present APM: not present GPT: present Found valid GPT with protective MBR; using GPT. Command (? for help): p Disk /dev/sda:...

  • Roderick W. Smith Roderick W. Smith committed [9a3703] on Code

    Fix refind-install failure to install filesystem drivers on some distributions

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    Can you boot into Ubuntu in some other way (via GRUB, for instance)? If so, then the output of sudo efibootmgr -v might be helpful. You might also try booting rEFInd via a USB drive; there's a USB image linked on the rEFInd downloads page that you can write to a USB flash drive; you should then be able to boot it by holding down the Option/Alt key as you power up the Mac. That's not a good long-term workaround, but it should at least get you into Ubuntu if you can't currently boot in any other way....

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    Could you please try the latest (pre-release) version in the git repository, or available as a .zip file here: https://www.rodsbooks.com/refind-bin-0.13.3.8.zip I did some tests, and that latest version has no problems with an ext4 filesystem created with mkfs.ext4 using e2fsprogs 1.47.0. I suspect that any problems you ran into were not actually caused by FEATURE_C12 (aka orphan_file), but rather by the metadata_csum_seed flag, which was also enabled by e2fsprogs 1.47.0; but the metadata_csum_seed...

  • Roderick W. Smith Roderick W. Smith committed [f4c337] on Code

    Add some Red Hat Secure Boot keys & development tool for extracting them from Shim

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    Hi all, I've recently added a new Linux script to the rEFInd project, refind-sb-healthcheck, to help with managing Shim binaries and Secure Boot keys. This script does three things, in an interactive way: It scans the computer for newer Shim binaries than the one that was used to boot most recently. If found, it offers to replace the current Shim with the newer one. Note that "newer" means "has a more recent file timestamp"; the script does not dig into the code to try to identify the version number....

  • Roderick W. Smith Roderick W. Smith committed [a2101a] on Code

    Added refind-sb-healthcheck script

  • Roderick W. Smith Roderick W. Smith committed [86c532] on Code

    Documentation updates

  • Roderick W. Smith Roderick W. Smith posted a comment on discussion General Discussion

    Could you please try the latest pre-release version, available in the git repository (as source code) or here: https://www.rodsbooks.com/refind-bin-0.13.3.8.zip That should fix the problem. BTW, nice subvolume names. ;)

  • Roderick W. Smith Roderick W. Smith committed [fde811] on Code

    Fix also_scan_dirs volume-specification bug

  • Roderick W. Smith Roderick W. Smith committed [274772] on Code

    Debugging for confused menu listings

  • Roderick W. Smith Roderick W. Smith committed [7bd6b6] on Code

    Documentation wording tweak

  • Roderick W. Smith Roderick W. Smith committed [cce5cd] on Code

    Remove unnecessary comparison

1 >