rpms/grub2
3
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
f31
grub2/0047-Replace-a-lot-of-man-pages-with-slightly-nicer-ones.patch

1961 lines
50 KiB
Diff
Raw Permalink Normal View History

Another arm building fix. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-07-12 14:56:34 +00:00
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
From: Peter Jones <pjones@redhat.com>
Date: Tue, 4 Mar 2014 11:00:23 -0500
Add some features to do-rebase Signed-off-by: Peter Jones <pjones@redhat.com>
2018-07-10 18:39:10 +00:00
Subject: [PATCH] Replace a lot of man pages with slightly nicer ones.
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
Update to be newer than f24's branch. - Add grub2-get-kernel-settings Related: rhbz#1226325 Signed-off-by: Peter Jones <pjones@redhat.com>
2016-08-25 18:42:57 +00:00
Replace a bunch of machine generated ones with ones that look nicer.
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
---
configure.ac | 23 ++++++
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
conf/Makefile.extra-dist | 1 -
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
docs/Makefile.am | 2 -
docs/man/grub-bios-setup.h2m | 6 --
docs/man/grub-editenv.h2m | 5 --
docs/man/grub-emu.h2m | 6 --
docs/man/grub-file.h2m | 2 -
docs/man/grub-fstest.h2m | 4 -
docs/man/grub-glue-efi.h2m | 4 -
docs/man/grub-install.h2m | 6 --
docs/man/grub-kbdcomp.h2m | 10 ---
docs/man/grub-macbless.h2m | 4 -
docs/man/grub-macho2img.h2m | 4 -
docs/man/grub-menulst2cfg.h2m | 4 -
docs/man/grub-mkconfig.h2m | 4 -
docs/man/grub-mkfont.h2m | 4 -
docs/man/grub-mkimage.h2m | 6 --
docs/man/grub-mklayout.h2m | 10 ---
docs/man/grub-mknetdir.h2m | 4 -
docs/man/grub-mkpasswd-pbkdf2.h2m | 4 -
docs/man/grub-mkrelpath.h2m | 4 -
docs/man/grub-mkrescue.h2m | 4 -
docs/man/grub-mkstandalone.h2m | 4 -
docs/man/grub-mount.h2m | 2 -
docs/man/grub-ofpathname.h2m | 4 -
docs/man/grub-pe2elf.h2m | 4 -
docs/man/grub-probe.h2m | 4 -
docs/man/grub-reboot.h2m | 5 --
docs/man/grub-render-label.h2m | 3 -
docs/man/grub-script-check.h2m | 4 -
docs/man/grub-set-default.h2m | 5 --
docs/man/grub-sparc64-setup.h2m | 6 --
docs/man/grub-syslinux2cfg.h2m | 4 -
gentpl.py | 5 +-
util/grub-bios-setup.8 | 54 +++++++++++++
util/grub-editenv.1 | 46 +++++++++++
util/grub-file.1 | 165 ++++++++++++++++++++++++++++++++++++++
util/grub-fstest.1 | 99 +++++++++++++++++++++++
util/grub-glue-efi.1 | 31 +++++++
util/grub-install.8 | 129 +++++++++++++++++++++++++++++
util/grub-kbdcomp.1 | 19 +++++
util/grub-macbless.1 | 22 +++++
util/grub-menulst2cfg.1 | 12 +++
util/grub-mkconfig.8 | 17 ++++
util/grub-mkfont.1 | 87 ++++++++++++++++++++
util/grub-mkimage.1 | 95 ++++++++++++++++++++++
util/grub-mklayout.1 | 27 +++++++
util/grub-mknetdir.1 | 12 +++
util/grub-mkpasswd-pbkdf2.1 | 27 +++++++
util/grub-mkrelpath.1 | 12 +++
util/grub-mkrescue.1 | 123 ++++++++++++++++++++++++++++
util/grub-mkstandalone.1 | 100 +++++++++++++++++++++++
util/grub-ofpathname.8 | 12 +++
util/grub-probe.8 | 80 ++++++++++++++++++
util/grub-reboot.8 | 21 +++++
util/grub-render-label.1 | 51 ++++++++++++
util/grub-script-check.1 | 21 +++++
util/grub-set-default.8 | 21 +++++
util/grub-sparc64-setup.8 | 12 +++
59 files changed, 1319 insertions(+), 147 deletions(-)
delete mode 100644 docs/man/grub-bios-setup.h2m
delete mode 100644 docs/man/grub-editenv.h2m
delete mode 100644 docs/man/grub-emu.h2m
delete mode 100644 docs/man/grub-file.h2m
delete mode 100644 docs/man/grub-fstest.h2m
delete mode 100644 docs/man/grub-glue-efi.h2m
delete mode 100644 docs/man/grub-install.h2m
delete mode 100644 docs/man/grub-kbdcomp.h2m
delete mode 100644 docs/man/grub-macbless.h2m
delete mode 100644 docs/man/grub-macho2img.h2m
delete mode 100644 docs/man/grub-menulst2cfg.h2m
delete mode 100644 docs/man/grub-mkconfig.h2m
delete mode 100644 docs/man/grub-mkfont.h2m
delete mode 100644 docs/man/grub-mkimage.h2m
delete mode 100644 docs/man/grub-mklayout.h2m
delete mode 100644 docs/man/grub-mknetdir.h2m
delete mode 100644 docs/man/grub-mkpasswd-pbkdf2.h2m
delete mode 100644 docs/man/grub-mkrelpath.h2m
delete mode 100644 docs/man/grub-mkrescue.h2m
delete mode 100644 docs/man/grub-mkstandalone.h2m
delete mode 100644 docs/man/grub-mount.h2m
delete mode 100644 docs/man/grub-ofpathname.h2m
delete mode 100644 docs/man/grub-pe2elf.h2m
delete mode 100644 docs/man/grub-probe.h2m
delete mode 100644 docs/man/grub-reboot.h2m
delete mode 100644 docs/man/grub-render-label.h2m
delete mode 100644 docs/man/grub-script-check.h2m
delete mode 100644 docs/man/grub-set-default.h2m
delete mode 100644 docs/man/grub-sparc64-setup.h2m
delete mode 100644 docs/man/grub-syslinux2cfg.h2m
create mode 100644 util/grub-bios-setup.8
create mode 100644 util/grub-editenv.1
create mode 100644 util/grub-file.1
create mode 100644 util/grub-fstest.1
create mode 100644 util/grub-glue-efi.1
create mode 100644 util/grub-install.8
create mode 100644 util/grub-kbdcomp.1
create mode 100644 util/grub-macbless.1
create mode 100644 util/grub-menulst2cfg.1
create mode 100644 util/grub-mkconfig.8
create mode 100644 util/grub-mkfont.1
create mode 100644 util/grub-mkimage.1
create mode 100644 util/grub-mklayout.1
create mode 100644 util/grub-mknetdir.1
create mode 100644 util/grub-mkpasswd-pbkdf2.1
create mode 100644 util/grub-mkrelpath.1
create mode 100644 util/grub-mkrescue.1
create mode 100644 util/grub-mkstandalone.1
create mode 100644 util/grub-ofpathname.8
create mode 100644 util/grub-probe.8
create mode 100644 util/grub-reboot.8
create mode 100644 util/grub-render-label.1
create mode 100644 util/grub-script-check.1
create mode 100644 util/grub-set-default.8
create mode 100644 util/grub-sparc64-setup.8
diff --git a/configure.ac b/configure.ac
Rebased to newer upstream for fedora-29 Signed-off-by: Peter Jones <pjones@redhat.com>
2018-07-10 19:08:14 +00:00
index 783118ccdcd..d5db2803ec4 100644
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/configure.ac
+++ b/configure.ac
@@ -70,6 +70,29 @@ grub_TRANSFORM([grub-set-default])
grub_TRANSFORM([grub-sparc64-setup])
grub_TRANSFORM([grub-render-label])
grub_TRANSFORM([grub-file])
+grub_TRANSFORM([grub-bios-setup.3])
+grub_TRANSFORM([grub-editenv.1])
+grub_TRANSFORM([grub-fstest.3])
+grub_TRANSFORM([grub-glue-efi.3])
+grub_TRANSFORM([grub-install.1])
+grub_TRANSFORM([grub-kbdcomp.3])
+grub_TRANSFORM([grub-menulst2cfg.1])
+grub_TRANSFORM([grub-mkconfig.1])
+grub_TRANSFORM([grub-mkfont.3])
+grub_TRANSFORM([grub-mkimage.1])
+grub_TRANSFORM([grub-mklayout.3])
+grub_TRANSFORM([grub-mknetdir.3])
+grub_TRANSFORM([grub-mkpasswd-pbkdf2.3])
+grub_TRANSFORM([grub-mkrelpath.3])
+grub_TRANSFORM([grub-mkrescue.1])
+grub_TRANSFORM([grub-mkstandalone.3])
+grub_TRANSFORM([grub-ofpathname.3])
+grub_TRANSFORM([grub-probe.3])
+grub_TRANSFORM([grub-reboot.3])
+grub_TRANSFORM([grub-render-label.3])
+grub_TRANSFORM([grub-script-check.3])
+grub_TRANSFORM([grub-set-default.1])
+grub_TRANSFORM([grub-sparc64-setup.3])
# Optimization flag. Allow user to override.
if test "x$TARGET_CFLAGS" = x; then
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
diff --git a/conf/Makefile.extra-dist b/conf/Makefile.extra-dist
index b16bd925320..39eb94bded6 100644
--- a/conf/Makefile.extra-dist
+++ b/conf/Makefile.extra-dist
@@ -11,7 +11,6 @@ EXTRA_DIST += unicode
EXTRA_DIST += util/import_gcry.py
EXTRA_DIST += util/import_unicode.py
-EXTRA_DIST += docs/man
EXTRA_DIST += docs/autoiso.cfg
EXTRA_DIST += docs/grub.cfg
EXTRA_DIST += docs/osdetect.cfg
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
diff --git a/docs/Makefile.am b/docs/Makefile.am
Work around some issues with older automake found in CentOS. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-04-11 18:42:03 +00:00
index 93eb3962765..ab28f199694 100644
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/Makefile.am
+++ b/docs/Makefile.am
@@ -5,5 +5,3 @@ info_TEXINFOS = grub.texi grub-dev.texi
grub_TEXINFOS = fdl.texi
EXTRA_DIST = font_char_metrics.png font_char_metrics.txt
-
-
diff --git a/docs/man/grub-bios-setup.h2m b/docs/man/grub-bios-setup.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index ac6ede36296..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-bios-setup.h2m
+++ /dev/null
@@ -1,6 +0,0 @@
-[NAME]
-grub-bios-setup \- set up a device to boot using GRUB
-[SEE ALSO]
-.BR grub-install (8),
-.BR grub-mkimage (1),
-.BR grub-mkrescue (1)
diff --git a/docs/man/grub-editenv.h2m b/docs/man/grub-editenv.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 3859d3d4c4f..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-editenv.h2m
+++ /dev/null
@@ -1,5 +0,0 @@
-[NAME]
-grub-editenv \- edit GRUB environment block
-[SEE ALSO]
-.BR grub-reboot (8),
-.BR grub-set-default (8)
diff --git a/docs/man/grub-emu.h2m b/docs/man/grub-emu.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index ef1c000656a..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-emu.h2m
+++ /dev/null
@@ -1,6 +0,0 @@
-[NAME]
-grub-emu \- GRUB emulator
-[SEE ALSO]
-If you are trying to install GRUB, then you should use
-.BR grub-install (8)
-rather than this program.
diff --git a/docs/man/grub-file.h2m b/docs/man/grub-file.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index e09bb4d3101..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-file.h2m
+++ /dev/null
@@ -1,2 +0,0 @@
-[NAME]
-grub-file \- check file type
diff --git a/docs/man/grub-fstest.h2m b/docs/man/grub-fstest.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 9676b159afd..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-fstest.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-fstest \- debug tool for GRUB filesystem drivers
-[SEE ALSO]
-.BR grub-probe (8)
diff --git a/docs/man/grub-glue-efi.h2m b/docs/man/grub-glue-efi.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index c1c6ded49ff..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-glue-efi.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-glue-efi \- generate a fat binary for EFI
-[DESCRIPTION]
-grub-glue-efi processes ia32 and amd64 EFI images and glues them according to Apple format.
diff --git a/docs/man/grub-install.h2m b/docs/man/grub-install.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 8cbbc87a0f2..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-install.h2m
+++ /dev/null
@@ -1,6 +0,0 @@
-[NAME]
-grub-install \- install GRUB to a device
-[SEE ALSO]
-.BR grub-mkconfig (8),
-.BR grub-mkimage (1),
-.BR grub-mkrescue (1)
diff --git a/docs/man/grub-kbdcomp.h2m b/docs/man/grub-kbdcomp.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index d81f9157e01..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-kbdcomp.h2m
+++ /dev/null
@@ -1,10 +0,0 @@
-[NAME]
-grub-kbdcomp \- generate a GRUB keyboard layout file
-[DESCRIPTION]
-grub-kbdcomp processes a X keyboard layout description in
-.BR keymaps (5)
-format into a format that can be used by GRUB's
-.B keymap
-command.
-[SEE ALSO]
-.BR grub-mklayout (8)
diff --git a/docs/man/grub-macbless.h2m b/docs/man/grub-macbless.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 0197c0087d7..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-macbless.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-macbless \- bless a mac file/directory
-[SEE ALSO]
-.BR grub-install (1)
diff --git a/docs/man/grub-macho2img.h2m b/docs/man/grub-macho2img.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index d79aaeed8f9..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-macho2img.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-macho2img \- convert Mach-O to raw image
-[SEE ALSO]
-.BR grub-mkimage (1)
diff --git a/docs/man/grub-menulst2cfg.h2m b/docs/man/grub-menulst2cfg.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index c2e0055ed7e..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-menulst2cfg.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-menulst2cfg \- transform legacy menu.lst into grub.cfg
-[SEE ALSO]
-.BR grub-mkconfig (8)
diff --git a/docs/man/grub-mkconfig.h2m b/docs/man/grub-mkconfig.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 9b42f813010..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-mkconfig.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-mkconfig \- generate a GRUB configuration file
-[SEE ALSO]
-.BR grub-install (8)
diff --git a/docs/man/grub-mkfont.h2m b/docs/man/grub-mkfont.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index d46fe600eca..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-mkfont.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-mkfont \- make GRUB font files
-[SEE ALSO]
-.BR grub-mkconfig (8)
diff --git a/docs/man/grub-mkimage.h2m b/docs/man/grub-mkimage.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index f0fbc2bb197..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-mkimage.h2m
+++ /dev/null
@@ -1,6 +0,0 @@
-[NAME]
-grub-mkimage \- make a bootable image of GRUB
-[SEE ALSO]
-.BR grub-install (8),
-.BR grub-mkrescue (1),
-.BR grub-mknetdir (8)
diff --git a/docs/man/grub-mklayout.h2m b/docs/man/grub-mklayout.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 1e43409c0ab..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-mklayout.h2m
+++ /dev/null
@@ -1,10 +0,0 @@
-[NAME]
-grub-mklayout \- generate a GRUB keyboard layout file
-[DESCRIPTION]
-grub-mklayout processes a keyboard layout description in
-.BR keymaps (5)
-format into a format that can be used by GRUB's
-.B keymap
-command.
-[SEE ALSO]
-.BR grub-mkconfig (8)
diff --git a/docs/man/grub-mknetdir.h2m b/docs/man/grub-mknetdir.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index a2ef13ec111..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-mknetdir.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-mknetdir \- prepare a GRUB netboot directory.
-[SEE ALSO]
-.BR grub-mkimage (1)
diff --git a/docs/man/grub-mkpasswd-pbkdf2.h2m b/docs/man/grub-mkpasswd-pbkdf2.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 4d202f3da7e..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-mkpasswd-pbkdf2.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-mkpasswd-pbkdf2 \- generate hashed password for GRUB
-[SEE ALSO]
-.BR grub-mkconfig (8)
diff --git a/docs/man/grub-mkrelpath.h2m b/docs/man/grub-mkrelpath.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index d01f3961e3f..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-mkrelpath.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-mkrelpath \- make a system path relative to its root
-[SEE ALSO]
-.BR grub-probe (8)
diff --git a/docs/man/grub-mkrescue.h2m b/docs/man/grub-mkrescue.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index a427f02e3c6..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-mkrescue.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-mkrescue \- make a GRUB rescue image
-[SEE ALSO]
-.BR grub-mkimage (1)
diff --git a/docs/man/grub-mkstandalone.h2m b/docs/man/grub-mkstandalone.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index c77313978ad..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-mkstandalone.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-mkstandalone \- make a memdisk-based GRUB image
-[SEE ALSO]
-.BR grub-mkimage (1)
diff --git a/docs/man/grub-mount.h2m b/docs/man/grub-mount.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 8d168982d72..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-mount.h2m
+++ /dev/null
@@ -1,2 +0,0 @@
-[NAME]
-grub-mount \- export GRUB filesystem with FUSE
diff --git a/docs/man/grub-ofpathname.h2m b/docs/man/grub-ofpathname.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 74b43eea039..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-ofpathname.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-ofpathname \- find OpenBOOT path for a device
-[SEE ALSO]
-.BR grub-probe (8)
diff --git a/docs/man/grub-pe2elf.h2m b/docs/man/grub-pe2elf.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 7ca29bd703c..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-pe2elf.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-pe2elf \- convert PE image to ELF
-[SEE ALSO]
-.BR grub-mkimage (1)
diff --git a/docs/man/grub-probe.h2m b/docs/man/grub-probe.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 6e1ffdcf937..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-probe.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-probe \- probe device information for GRUB
-[SEE ALSO]
-.BR grub-fstest (1)
diff --git a/docs/man/grub-reboot.h2m b/docs/man/grub-reboot.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index e4acace65ce..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-reboot.h2m
+++ /dev/null
@@ -1,5 +0,0 @@
-[NAME]
-grub-reboot \- set the default boot entry for GRUB, for the next boot only
-[SEE ALSO]
-.BR grub-set-default (8),
-.BR grub-editenv (1)
diff --git a/docs/man/grub-render-label.h2m b/docs/man/grub-render-label.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 50ae5247c05..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-render-label.h2m
+++ /dev/null
@@ -1,3 +0,0 @@
-[NAME]
-grub-render-label \- generate a .disk_label for Apple Macs.
-
diff --git a/docs/man/grub-script-check.h2m b/docs/man/grub-script-check.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 3653682671a..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-script-check.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-script-check \- check grub.cfg for syntax errors
-[SEE ALSO]
-.BR grub-mkconfig (8)
diff --git a/docs/man/grub-set-default.h2m b/docs/man/grub-set-default.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 7945001c154..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-set-default.h2m
+++ /dev/null
@@ -1,5 +0,0 @@
-[NAME]
-grub-set-default \- set the saved default boot entry for GRUB
-[SEE ALSO]
-.BR grub-reboot (8),
-.BR grub-editenv (1)
diff --git a/docs/man/grub-sparc64-setup.h2m b/docs/man/grub-sparc64-setup.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 18f803a50db..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-sparc64-setup.h2m
+++ /dev/null
@@ -1,6 +0,0 @@
-[NAME]
-grub-sparc64-setup \- set up a device to boot using GRUB
-[SEE ALSO]
-.BR grub-install (8),
-.BR grub-mkimage (1),
-.BR grub-mkrescue (1)
diff --git a/docs/man/grub-syslinux2cfg.h2m b/docs/man/grub-syslinux2cfg.h2m
deleted file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index ad25c8ab753..00000000000
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/docs/man/grub-syslinux2cfg.h2m
+++ /dev/null
@@ -1,4 +0,0 @@
-[NAME]
-grub-syslinux2cfg \- transform syslinux config into grub.cfg
-[SEE ALSO]
-.BR grub-menulst2cfg (8)
diff --git a/gentpl.py b/gentpl.py
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index da67965a41a..bf8439fa743 100644
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- a/gentpl.py
+++ b/gentpl.py
Rebased to grub 2.02 for fedora-27 Signed-off-by: Peter Jones <pjones@redhat.com>
2017-06-16 19:31:32 +00:00
@@ -802,10 +802,7 @@ def manpage(defn, adddeps):
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
output("if COND_MAN_PAGES\n")
gvar_add("man_MANS", name + "." + mansection)
- rule(name + "." + mansection, name + " " + adddeps, """
-chmod a+x """ + name + """
-PATH=$(builddir):$$PATH pkgdatadir=$(builddir) $(HELP2MAN) --section=""" + mansection + """ -i $(top_srcdir)/docs/man/""" + name + """.h2m -o $@ """ + name + """
-""")
+ rule(name + "." + mansection, name + " " + adddeps, "cat $(top_srcdir)/util/" + name + "." + mansection + " | $(top_builddir)/config.status --file=$@:-")
gvar_add("CLEANFILES", name + "." + mansection)
output("endif\n")
diff --git a/util/grub-bios-setup.8 b/util/grub-bios-setup.8
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..56f582b3d75
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-bios-setup.8
@@ -0,0 +1,54 @@
+.TH GRUB-BIOS-SETUP 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-bios-setup\fR \(em Set up images to boot from a device.
+
+.SH SYNOPSIS
+\fBgrub-bios-setup\fR [-a | --allow-floppy] [-b | --boot-image=\fIFILE\fR]
+.RS 17
+[-c | --core-image=\fIFILE\fR] [-d | --directory=\fIDIR\fR]
+.RE
+.RS 17
+[-f | --force] [-m | --device-map=\fIFILE\fR]
+.RE
+.RS 17
+[-s | --skip-fs-probe] [-v | --verbose] \fIDEVICE\fR
+
+.SH DESCRIPTION
+You should not normally run this program directly. Use grub-install instead.
+
+.SH OPTIONS
+.TP
+\fB--allow-floppy\fR
+Make the device also bootable as a floppy. This option is the default for
+/dev/fdX devices. Some BIOSes will not boot images created with this option.
+
+.TP
+\fB--boot-image\fR=\fIFILE\fR
+Use FILE as the boot image. The default value is \fBboot.img\fR.
+
+.TP
+\fB--core-image\fR=\fIFILE\fR
+Use FILE as ther core image. The default value is \fBcore.img\fR.
+
+.TP
+\fB--directory\fR=\fIDIR\fR
+Use GRUB files in the directory DIR. The default value is \fB/boot/grub\fR.
+
+.TP
+\fB--force\fR
+Install even if problems are detected.
+
+.TP
+\fB--device-map\fR=\fIFILE\fR
+Use FILE as the device map. The default value is /boot/grub/device.map .
+
+.TP
+\fB--skip-fs-probe\fR
+Do not probe DEVICE for filesystems.
+
+.TP
+\fB--verbose\fR
+Print verbose messages.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-editenv.1 b/util/grub-editenv.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..d28ba03ba42
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-editenv.1
@@ -0,0 +1,46 @@
+.TH GRUB-EDITENV 1 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-editenv\fR \(em Manage the GRUB environment block.
+
+.SH SYNOPSIS
+\fBgrub-editenv\fR [-v | --verbose] [\fIFILE\fR]
+.RS 14
+<create | list | set \fINAME\fR=\fIVALUE\fR | unset \fINAME\fR>
+
+.SH DESCRIPTION
+\fBgrub-editenv\fR is a command line tool to manage GRUB's stored environment.
+
+.SH OPTIONS
+.TP
Reconcile with upstream once again. Also include some minor fixes for gcc 5.1.1 Signed-off-by: Peter Jones <pjones@redhat.com>
2015-07-22 13:44:51 +00:00
+\fB--verbose\fR
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
+Print verbose messages.
+
+.TP
+\fBFILE\fR
+.RS 7
+File name to use for grub environment. Default is /boot/grub/grubenv .
+.RE
+
+.SH COMMANDS
+.TP
+\fBcreate\fR
+.RS 7
+Create a blank environment block file.
+.RE
+
+.TP
+\fBlist\fR
+.RS 7
+List the current variables.
+.RE
+
+.TP
+\fBset\fR [\fINAME\fR=\fIVALUE\fR ...]
+Set variables.
+
+.TP
+\fBunset [\fINAME\fR ...]
+Delete variables.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-file.1 b/util/grub-file.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..b29cb327889
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-file.1
@@ -0,0 +1,165 @@
+.TH GRUB-FILE 1 "Web Feb 26 2014"
+.SH NAME
+\fBgrub-file\fR \(em Check if FILE is of specified type.
+
+.SH SYNOPSIS
+\fBgrub-file\fR (--is-i386-xen-pae-domu | --is-x86_64-xen-domu |
+.RS 11
+--is-x86-xen-dom0 | --is-x86-multiboot |
+.RE
+.RS 11
+--is-x86-multiboot2 | --is-arm-linux | --is-arm64-linux |
+.RE
+.RS 11
+--is-ia64-linux | --is-mips-linux | --is-mipsel-linux |
+.RE
+.RS 11
+--is-sparc64-linux | --is-powerpc-linux | --is-x86-linux |
+.RE
+.RS 11
+--is-x86-linux32 | --is-x86-kfreebsd | --is-i386-kfreebsd |
+.RE
+.RS 11
+--is-x86_64-kfreebsd | --is-x86-knetbsd |
+.RE
+.RS 11
+--is-i386-knetbsd | --is-x86_64-knetbsd | --is-i386-efi |
+.RE
+.RS 11
+--is-x86_64-efi | --is-ia64-efi | --is-arm64-efi |
+.RE
+.RS 11
+--is-arm-efi | --is-hibernated-hiberfil | --is-x86_64-xnu |
+.RE
+.RS 11
+--is-i386-xnu | --is-xnu-hibr | --is-x86-bios-bootsector)
+.RE
+.RS 11
+\fIFILE\fR
+
+.SH DESCRIPTION
+\fBgrub-file\fR is used to check if \fIFILE\fR is of a specified type.
+
+.SH OPTIONS
+.TP
+--is-i386-xen-pae-domu
+Check if FILE can be booted as i386 PAE Xen unprivileged guest kernel
+
+.TP
+--is-x86_64-xen-domu
+Check if FILE can be booted as x86_64 Xen unprivileged guest kernel
+
+.TP
+--is-x86-xen-dom0
+Check if FILE can be used as Xen x86 privileged guest kernel
+
+.TP
+--is-x86-multiboot
+Check if FILE can be used as x86 multiboot kernel
+
+.TP
+--is-x86-multiboot2
+Check if FILE can be used as x86 multiboot2 kernel
+
+.TP
+--is-arm-linux
+Check if FILE is ARM Linux
+
+.TP
+--is-arm64-linux
+Check if FILE is ARM64 Linux
+
+.TP
+--is-ia64-linux
+Check if FILE is IA64 Linux
+
+.TP
+--is-mips-linux
+Check if FILE is MIPS Linux
+
+.TP
+--is-mipsel-linux
+Check if FILE is MIPSEL Linux
+
+.TP
+--is-sparc64-linux
+Check if FILE is SPARC64 Linux
+
+.TP
+--is-powerpc-linux
+Check if FILE is POWERPC Linux
+
+.TP
+--is-x86-linux
+Check if FILE is x86 Linux
+
+.TP
+--is-x86-linux32
+Check if FILE is x86 Linux supporting 32-bit protocol
+
+.TP
+--is-x86-kfreebsd
+Check if FILE is x86 kFreeBSD
+
+.TP
+--is-i386-kfreebsd
+Check if FILE is i386 kFreeBSD
+
+.TP
+--is-x86_64-kfreebsd
+Check if FILE is x86_64 kFreeBSD
+
+.TP
+--is-x86-knetbsd
+Check if FILE is x86 kNetBSD
+
+.TP
+--is-i386-knetbsd
+Check if FILE is i386 kNetBSD
+
+.TP
+--is-x86_64-knetbsd
+Check if FILE is x86_64 kNetBSD
+
+.TP
+--is-i386-efi
+Check if FILE is i386 EFI file
+
+.TP
+--is-x86_64-efi
+Check if FILE is x86_64 EFI file
+
+.TP
+--is-ia64-efi
+Check if FILE is IA64 EFI file
+
+.TP
+--is-arm64-efi
+Check if FILE is ARM64 EFI file
+
+.TP
+--is-arm-efi
+Check if FILE is ARM EFI file
+
+.TP
+--is-hibernated-hiberfil
+Check if FILE is hiberfil.sys in hibernated state
+
+.TP
+--is-x86_64-xnu
+Check if FILE is x86_64 XNU (Mac OS X kernel)
+
+.TP
+--is-i386-xnu
+Check if FILE is i386 XNU (Mac OS X kernel)
+
+.TP
+--is-xnu-hibr
+Check if FILE is XNU (Mac OS X kernel) hibernated image
+
+.TP
+--is-x86-bios-bootsector
+Check if FILE is BIOS bootsector
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-fstest.1 b/util/grub-fstest.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..792fa78634c
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-fstest.1
@@ -0,0 +1,99 @@
+.TH GRUB-FSTEST 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-fstest\fR — Debug tool for GRUB's filesystem driver.
+
+.SH SYNOPSIS
+\fBgrub-fstest\fR [-c | --diskcount=\fINUM\fR] [-C | --crypto]
+.RS 13
+[-d | --debug=\fISTRING\fR] [-K | --zfs-key=\fIFILE\fR|\fIprompt\fR]
+.RE
+.RS 13
+[-n | --length=\fINUM\fR] [-r | --root=\fIDEVICE_NAME\fR]
+.RE
+.RS 13
+[-s | --skip=\fINUM\fR] [-u | --uncompress] [-v | --verbose]
+.RE
+.RS 13
+\fIIMAGE_PATH\fR <blocklist \fIFILE\fR | cat \fIFILE\fR |
+.RE
+.RS 13
+cmp \fIFILE\fR \fILOCAL\fR | cp \fIFILE\fR \fILOCAL\fR | crc \fIFILE\fR |
+.RE
+.RS 13
+hex \fIFILE\fR | ls \fIPATH\fR | xnu_uuid \fIDEVICE\fR>
+
+.SH DESCRIPTION
+\fBgrub-fstest\fR is a tool for testing GRUB's filesystem drivers. You should not normally need to run this program.
+
+.SH OPTIONS
+.TP
+\fB--diskcount\fR=\fINUM\fR
+Specify the number of input files.
+
+.TP
+\fB--crypto\fR
+Mount cryptographic devices.
+
+.TP
+\fB--debug\fR=\fISTRING\fR
+Set debug environment variable.
+
+.TP
+\fB--zfs-key\fR=\fIFILE\fR|\fIprompt\fR
+Load ZFS cryptographic key.
+
+.TP
+\fB--length\fR=\fINUM\fR
+Handle NUM bytes in output file.
+
+.TP
+\fB--root\fR=\fIDEVICE_NAME\fR
+Set root device.
+
+.TP
+\fB--skip\fR=\fINUM\fR
+Skip NUM bytes from output file.
+
+.TP
+\fB--uncompress\fR
+Uncompress data.
+
+.TP
+\fB--verbose\fR
+Print verbose messages.
+
+.SH COMMANDS
+.TP
+\fBblocklist\fR \fIFILE\fR
+Display block list of \fIFILE\fR.
+
+.TP
+\fBcat\fR \fIFILE\fR
+Display \fIFILE\fR on standard output.
+
+.TP
+\fBcmp\fR \fIFILE\fR \fILOCAL\fR
+Compare \fIFILE\fR with local file \fILOCAL\fR.
+
+.TP
+\fBcp\fR \fIFILE\fR \fILOCAL\fR
+Copy \fIFILE\fR to local file \fILOCAL\fR.
+
+.TP
+\fBcrc\fR \fIFILE\fR
+Display the CRC-32 checksum of \fIFILE\fR.
+
+.TP
+\fBhex\fR \fIFILE\fR
+Display contents of \fIFILE\fR in hexidecimal.
+
+.TP
+\fBls\fR \fIPATH\fR
+List files at \fIPATH\fR.
+
+.TP
+\fBxnu_uuid\fR \fIDEVICE\fR
+Display the XNU UUID of \fIDEVICE\fR.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-glue-efi.1 b/util/grub-glue-efi.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..72bd555d577
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-glue-efi.1
@@ -0,0 +1,31 @@
+.TH GRUB-GLUE-EFI 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-glue-efi\fR \(em Create an Apple fat EFI binary.
+
+.SH SYNOPSIS
+\fBgrub-glue-efi\fR <-3 | --input32=\fIFILE\fR> <-6 | --input64=\fIFILE\fR>
+.RS 15
+<-o | --output=\fIFILE\fR> [-v | --verbose]
+
+.SH DESCRIPTION
+\fBgrub-glue-efi\fR creates an Apple fat EFI binary from two EFI binaries.
+
+.SH OPTIONS
+.TP
+\fB--input32\fR=\fIFILE\fR
+Read 32-bit binary from \fIFILE\fR.
+
+.TP
+\fB--input64\fR=\fIFILE\fR
+Read 64-bit binary from \fIFILE\fR.
+
+.TP
+\fB--output\fR=\fIFILE\fR
+Write resulting fat binary to \fIFILE\fR.
+
+.TP
+\fB--verbose\fR
+Print verbose messages.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-install.8 b/util/grub-install.8
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..76272a39d2e
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-install.8
@@ -0,0 +1,129 @@
+.TH GRUB-INSTALL 1 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-install\fR \(em Install GRUB on a device.
+
+.SH SYNOPSIS
+\fBgrub-install\fR [--modules=\fIMODULES\fR] [--install-modules=\fIMODULES\fR]
+.RS 14
+[--themes=\fITHEMES\fR] [--fonts=\fIFONTS\fR] [--locales=\fILOCALES\fR]
+.RE
+.RS 14
+[--compress[=\fIno\fR,\fIxz\fR,\fIgz\fR,\fIlzo\fR]] [-d | --directory=\fIDIR\fR]
+.RE
+.RS 14
+[--grub-mkimage=\fIFILE\fR] [--boot-directory=\fIDIR\fR]
+.RE
+.RS 14
+[--target=\fITARGET\fR] [--grub-setup=\fIFILE\fR]
+.RE
+.RS 14
+[--grub-mkrelpath=\fIFILE\fR] [--grub-probe=\fIFILE\fR]
+.RE
+.RS 14
+[--allow-floppy] [--recheck] [--force] [--force-file-id]
+.RE
+.RS 14
+[--disk-module=\fIMODULE\fR] [--no-nvram] [--removable]
+.RE
+.RS 14
+[--bootloader-id=\fIID\fR] [--efi-directory=\fIDIR\fR] \fIINSTALL_DEVICE\fR
+
+.SH DESCRIPTION
+\fBgrub-install\fR installs GRUB onto a device. This includes copying GRUB images into the target directory (generally \fI/boot/grub\fR), and on some platforms may also include installing GRUB onto a boot sector.
+
+.SH OPTIONS
+.TP
+\fB--modules\fR=\fIMODULES\fR\!
+Pre-load modules specified by \fIMODULES\fR.
+
+.TP
+\fB--install-modules\fR=\fIMODULES\fR
+Install only \fIMODULES\fR and their dependencies. The default is to install all available modules.
+
+.TP
+\fB--themes\fR=\fITHEMES\fR
+Install \fITHEMES\fR. The default is to install the \fIstarfield\fR theme, if available.
+
+.TP
+\fB--fonts\fR=\fIFONTS\fR
+Install \fIFONTS\fR. The default is to install the \fIunicode\fR font.
+
+.TP
+\fB--locales\fR=\fILOCALES\fR
+Install only locales listed in \fILOCALES\fR. The default is to install all available locales.
+
+.TP
+\fB--compress\fR=\fIno\fR,\fIxz\fR,\fIgz\fR,\fIlzo\fR
+Compress GRUB files using the specified compression algorithm.
+
+.TP
+\fB--directory\fR=\fIDIR\fR
+Use images and modules in \fIDIR\fR.
+
+.TP
+\fB--grub-mkimage\fR=\fIFILE\fR
+Use \fIFILE\fR as \fBgrub-mkimage\fR. The default is \fI/usr/bin/grub-mkimage\fR.
+
+.TP
+\fB--boot-directory\fR=\fIDIR\fR
+Use \fIDIR\fR as the boot directory. The default is \fI/boot\fR. GRUB will put its files in a subdirectory of this directory named \fIgrub\fR.
+
+.TP
+\fB--target\fR=\fITARGET\fR
+Install GRUB for \fITARGET\fR platform. The default is the platform \fBgrub-install\fR is running on.
+
+.TP
+\fB--grub-setup\fR=\fIFILE\fR
+Use \fIFILE\fR as \fBgrub-setup\fR. The default is \fI/usr/bin/grub-setup\fR.
+
+.TP
+\fB--grub-mkrelpath\fR=\fIFILE\fR
+Use \fIFILE\fR as \fBgrub-mkrelpath\fR. The default is \fI/usr/bin/grub-mkrelpath\fR.
+
+.TP
+\fB--grub-probe\fR=\fIFILE\fR
+Use \fIFILE\fR as \fBgrub-probe\fR. The default is \fI/usr/bin/grub-mkrelpath\fR.
+
+.TP
+\fB--allow-floppy
+Make the device also bootable as a floppy. This option is the default for /dev/fdX devices. Some BIOSes will not boot images created with this option.
+
+.TP
+\fB--recheck
+Delete any existing device map and create a new one if necessary.
+
+.TP
+\fB--force
+Install even if problems are detected.
+
+.TP
+\fB--force-file-id
+Use identifier file even if UUID is available.
+
+.TP
+\fB--disk-module\fR=\fIMODULE\fR
+Use \fIMODULE\fR for disk access. This allows you to manually specify either \fIbiosdisk\fR or \fInative\fR disk access. This option is only available on the BIOS target platform.
+
+.TP
+\fB--no-nvram
+Do not update the \fIboot-device\fR NVRAM variable. This option is only available on IEEE1275 target platforms.
+
+.TP
+\fB--removable
+Treat the target device as if it is removeable. This option is only available on the EFI target platform.
+
+.TP
+\fB--bootloader-id\fR=\fIID\fR
+Use \fIID\fR as the bootloader ID. This opption is only available on the EFI target platform.
+
+.TP
+\fB--efi-directory\fR=\fIDIR\fR
+Use \fIDIR\fR as the EFI System Partition root. This opption is only available on the EFI ta
+rget platform.
+
+.TP
+\fIINSTALL_DEVICE\fR
+Install GRUB to the block device \fIINSTALL_DEVICE\fR.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-kbdcomp.1 b/util/grub-kbdcomp.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..0bb969a5b43
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-kbdcomp.1
@@ -0,0 +1,19 @@
+.TH GRUB-KBDCOMP 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-kbdcomp\fR \(em Generate a GRUB keyboard layout file.
+
+.SH SYNOPSIS
+\fBgrub-kbdcomp\fR <-o | --output=\fIFILE\fR> \fICKBMAP_ARGUMENTS\fR
+
+.SH DESCRIPTION
+\fBgrub-kbdcomp\fR processes an X keyboard layout description in
+\fBkeymaps\fR(5) format into a format that can be used by GRUB's \fBkeymap\fR
+command.
+
+.SH OPTIONS
+.TP
+\fB--output\fR=\fIFILE\fR
+Write output to \fIFILE\fR.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-macbless.1 b/util/grub-macbless.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..41a96186f70
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-macbless.1
@@ -0,0 +1,22 @@
+.TH GRUB-MACBLESS 1 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-macbless\fR \(em Mac-style bless on HFS or HFS+
+
+.SH SYNOPSIS
+\fBgrub-macbless\fR [-v | --verbose] [-p | --ppc] \fIFILE\fR | [-x | --x86] \fIFILE\fR
+
+.SH OPTIONS
+.TP
+--x86
+Bless for x86 based Macs.
+
+.TP
+--ppc
+Bless for PPC based Macs.
+
+.TP
+--verbose
+Print verbose messages.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-menulst2cfg.1 b/util/grub-menulst2cfg.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..91e2ef87113
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-menulst2cfg.1
@@ -0,0 +1,12 @@
+.TH GRUB-MENULST2CFG 1 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-menulst2cfg\fR \(em Convert a configuration file from GRUB 0.xx to GRUB 2.xx format.
+
+.SH SYNOPSIS
+\fBgrub-menulst2cfg\fR [\fIINFILE\fR [\fIOUTFILE\fR]]
+
+.SH DESCRIPTION
+\fBgrub-menulst2cfg\fR converts a configuration file from GRUB 0.xx to the current format.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-mkconfig.8 b/util/grub-mkconfig.8
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..a2d1f577b9b
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-mkconfig.8
@@ -0,0 +1,17 @@
+.TH GRUB-MKCONFIG 1 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-mkconfig\fR \(em Generate a GRUB configuration file.
+
+.SH SYNOPSIS
+\fBgrub-mkconfig\fR [-o | --output=\fIFILE\fR]
+
+.SH DESCRIPTION
+\fBgrub-mkconfig\fR generates a configuration file for GRUB.
+
+.SH OPTIONS
+.TP
+\fB--output\fR=\fIFILE\fR
+Write generated output to \fIFILE\fR.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-mkfont.1 b/util/grub-mkfont.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..3494857987d
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-mkfont.1
@@ -0,0 +1,87 @@
+.TH GRUB-MKFONT 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-mkfont\fR \(em Convert common font file formats into the PF2 format.
+
+.SH SYNOPSIS
+\fBgrub-mkfont\fR [--ascii-bitmaps] [-a | --force-autohint]
+.RS 13
+[-b | --bold] [-c | --asce=\fINUM\fR] [-d | --desc=\fINUM\fR]
+.RE
+.RS 13
+[-i | --index=\fINUM\fR] [-n | --name=\fINAME\fR] [--no-bitmap]
+.RE
+.RS 13
+[--no-hinting] <-o | --output=\fIFILE\fR>
+.RE
+.RS 13
+[-r | --range=\fIFROM-TO\fR[\fI,FROM-TO\fR]] [-s | --size=\fISIZE\fR]
+.RE
+.RS 13
+[-v | --verbose] [--width-spec] \fIFONT_FILES\fR
+
+.SH DESCRIPTION
+\fBgrub-mkfont\fR converts font files from common formats into the PF2 format used by GRUB.
+
+.SH OPTIONS
+.TP
+--ascii-bitmaps
+Save only bitmaps for ASCII characters.
+
+.TP
+--force-autohint
+Force generation of automatic hinting.
+
+.TP
+--bold
+Convert font to bold.
+
+.TP
+--asce=\fINUM\fR
+Set font ascent to \fINUM\fR.
+
+.TP
+--desc=\fINUM\fR
+Set font descent to \fINUM\fR.
+
+.TP
+--index=\fINUM\fR
+Select face index \fINUM\fR.
+
+.TP
+--name=\fINAME\fR
+Set font family to \fINAME\fR.
+
+.TP
+--no-bitmap
+Ignore bitmap strikes when loading.
+
+.TP
+--no-hinting
+Disable hinting.
+
+.TP
+--output=\fIFILE\fR
+Save ouptut to \fIFILE\fR. This argument is required.
+
+.TP
+--range=\fIFROM-TO\fR\fI,FROM-TO\fR
+Set the font ranges to each pair of \fIFROM\fR,\fITO\fR.
+
+.TP
+--size=\fISIZE\fR
+Set font size to \fISIZE\fR.
+
+.TP
+--verbose
+Print verbose messages.
+
+.TP
+--width-spec
+Create a width summary file.
+
+.TP
+\fIFONT_FILES\fR
+The input files to be converted.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-mkimage.1 b/util/grub-mkimage.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..4dea4f54597
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-mkimage.1
@@ -0,0 +1,95 @@
+.TH GRUB-MKIMAGE 1 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-mkimage\fR \(em Make a bootable GRUB image.
+
+.SH SYNOPSIS
+\fBgrub-mkimage\fR [-c | --config=\fRFILE\fI] [-C | --compression=(\fIxz\fR,\fInone\fR,\fIauto\fR)]
+.RS 14
+[-d | --directory=\fRDIR\fR] [-k | --pubkey=\fIFILE\fR]
+.RE
+.RS 14
+[-m | --memdisk=\fIFILE\fR] [-n | --note] [-o | --output=\fIFILE\fR]
+.RE
+.RS 14
+[-O | --format=\fIFORMAT\fR] [-p | --prefix=\fIDIR\fR]
+.RE
+.RS 14
+[-v | --verbose] \fIMODULES\fR
+
+.SH DESCRIPTION
+\fBgrub-mkimage\fI builds a bootable image of GRUB.
+
+.SH OPTIONS
+.TP
+--config=\fIFILE\fR
+Embed \fIFILE\fR as the image's initial configuration file.
+
+.TP
+--compression=(\fIxz\fR,\fInone\fR,\fIauto\fR)
+Use one of \fIxz\fR, \fInone\fR, or \fIauto\fR as the compression method for the core image.
+
+.TP
+--directory=\fIDIR\fR
+Use images and modules from \fIDIR\fR. The default value is \fB/usr/lib/grub/<platform>\fR.
+
+.TP
+--pubkey=\fIFILE\fR
+Embed the public key \fIFILE\fR for signature checking.
+
+.TP
+--memdisk=\fIFILE\fR
+Embed the memdisk image \fIFILE\fR. If no \fB-p\fR option is also specified, this implies \fI-p (memdisk)/boot/grub\fR.
+
+.TP
+--note
+Add a CHRP \fINOTE\fR section. This option is only valid on IEEE1275 platforms.
+
+.TP
+--output=\fIFILE\fR
+Write the generated file to \fIFILE\fR. The default is to write to standard output.
+
+.TP
+--format=\fIFORMAT\fR
+Generate an image in the specified \fIFORMAT\fR. Valid values are:
+.RS
+.RS 4
+.P
+i386-coreboot,
+i386-multiboot,
+i386-pc,
+i386-pc-pxe,
+i386-efi,
+i386-ieee1275,
+i386-qemu,
+x86_64-efi,
+mipsel-yeeloong-flash,
+mipsel-fuloong2f-flash,
+mipself-loongson-elf,
+powerpc-ieee1275,
+sparc64-ieee1275-raw,
+sparc64-ieee1275-cdcore,
+sparc64-ieee1275-aout,
+ia64-efi,
+mips-arc,
+mipsel-arc,
+mipsel-qemu_mips-elf,
+mips-qemu_mips-flash,
+mipsel-qemu_mips-flash,
+mips-qemu_mips-elf
+.RE
+.RE
+
+.TP
+--prefix=\fIDIR\fR
+Set prefix directory. The default value is \fI/boot/grub\fR.
+
+.TP
+--verbose
+Print verbose messages.
+
+.TP
+\fIMODULES\fR
+Include \fIMODULES\fR.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-mklayout.1 b/util/grub-mklayout.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..d1bbc2ec515
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-mklayout.1
@@ -0,0 +1,27 @@
+.TH GRUB-MKLAYOUT 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-mklayout\fR \(em Generate a GRUB keyboard layout file.
+
+.SH SYNOPSIS
+\fBgrub-mklayout\fR [-i | --input=\fIFILE\fR] [-o | --output=\fIFILE\fR]
+.RS 15
+[-v | --verbose]
+
+.SH DESCRIPTION
+\fBgrub-mklayout\fR generates a GRUB keyboard layout description which corresponds with the Linux console layout description given as input.
+
+.SH OPTIONS
+.TP
+--input=\fIFILE\fR
+Use \fIFILE\fR as the input. The default value is the standard input device.
+
+.TP
+--output=\fIFILE\fR
+Use \fIFILE\fR as the output. The default value is the standard output device.
+
+.TP
+--verbose
+Print verbose messages.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-mknetdir.1 b/util/grub-mknetdir.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..fa7e8d4ef0d
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-mknetdir.1
@@ -0,0 +1,12 @@
+.TH GRUB-MKNETDIR 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-mknetdir\fR \(em Prepare a GRUB netboot directory.
+
+.SH SYNOPSIS
Reconcile with upstream once again. Also include some minor fixes for gcc 5.1.1 Signed-off-by: Peter Jones <pjones@redhat.com>
2015-07-22 13:44:51 +00:00
+\fBgrub-mknetdir\fR
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
+
+.SH DESCRIPTION
+\fBgrub-mknetdir\fR prepares a directory for GRUB to be netbooted from.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-mkpasswd-pbkdf2.1 b/util/grub-mkpasswd-pbkdf2.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..73c437c15d8
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-mkpasswd-pbkdf2.1
@@ -0,0 +1,27 @@
+.TH GRUB-MKPASSWD-PBKDF2 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-mkpasswd-pbkdf2\fR \(em Generate a PBKDF2 password hash.
+
+.SH SYNOPSIS
+\fBgrub-mkpasswd-pbkdf2\fR [-c | --iteration-count=\fINUM\fR] [-l | --buflen=\fINUM\fR]
+.RS 22
+[-s | --salt=\fINUM\fR]
+
+.SH DESCRIPTION
+\fBgrub-mkpasswd-pbkdf2\fR generates a PBKDF2 password string suitable for use in a GRUB configuration file.
+
+.SH OPTIONS
+.TP
+--iteration-count=\fINUM\fR
+Number of PBKDF2 iterations.
+
+.TP
+--buflen=\fINUM\fR
+Length of generated hash.
+
+.TP
+--salt=\fINUM\fR
+Length of salt to use.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-mkrelpath.1 b/util/grub-mkrelpath.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..85f1113621d
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-mkrelpath.1
@@ -0,0 +1,12 @@
+.TH GRUB-MKRELPATH 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-mkrelpath\fR \(em Generate a relative GRUB path given an OS path.
+
+.SH SYNOPSIS
+\fBgrub-mkrelpath\fR \fIFILE\fR
+
+.SH DESCRIPTION
+\fBgrub-mkrelpath\fR takes an OS filesystem path for \fIFILE\fR and returns a relative path suitable for use in a GRUB configuration file.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-mkrescue.1 b/util/grub-mkrescue.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..4ed9fc723fd
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-mkrescue.1
@@ -0,0 +1,123 @@
+.TH GRUB-MKRESCUE 3 "Wed Feb 26 2014"
+.SH NAME
+grub-mkrescue \(em Generate a GRUB rescue image using GNU Xorriso.
+
+.SH SYNOPSIS
+\fBgrub-mkrescue\fR [-o | --output=\fIFILE\fR] [--modules=\fIMODULES\fR]
+.RS 15
+[--install-modules=\fIMODULES\fR] [--themes=\fITHEMES\fR]
+.RE
+.RS 15
+[--fonts=\fIFONTS\fR] [--locales=\fILOCALES\fR]
+.RE
+.RS 15
+[--compress[=\fIno\fR,\fIxz\fR,\fIgz\fR,\fIlzo\fR]] [-d | --directory=\fIDIR\fR]
+.RE
+.RS 15
+[--grub-mkimage=\fIFILE\fR] [--rom-directory=\fIDIR\fR]
+.RE
+.RS 15
+[--xorriso=\fIFILE\fR] [--grub-glue-efi=\fIFILE\fR]
+.RE
+.RS 15
+[--grub-render-label=\fIFILE\fR] [--label-font=\fIFILE\fR]
+.RE
+.RS 15
+[--label-color=\fICOLOR\fR] [--label-bgcolor=\fIFILE\fR]
+.RE
+.RS 15
+[--product-name=\fISTRING\fR] [--product-version=\fISTRING\fR]
+.RE
+.RS 15
+[--sparc-boot] [--arcs-boot]
+
+.SH DESCRIPTION
+\fBgrub-mkrescue\fR can be used to generate a rescue image with the GRUB bootloader.
+
+.SH OPTIONS
+.TP
+\fB--output\fR=\fIFILE\fR
+Write the generated file to \fIFILE\fR. The default is to write to standard output.
+
+.TP
+\fB--modules\fR=\fIMODULES\fR
+Pre-load modules specified by \fIMODULES\fR.
+
+.TP
+\fB--install-modules\fR=\fIMODULES\fR
+Install only \fIMODULES\fR and their dependencies. The default is to install all available modules.
+
+.TP
+\fB--themes\fR=\fITHEMES\fR
+Install \fITHEMES\fR. The default is to install the \fIstarfield\fR theme, if available.
+
+.TP
+\fB--fonts\fR=\fIFONTS\fR
+Install \fIFONTS\fR. The default is to install the \fIunicode\fR font.
+
+.TP
+\fB--locales\fR=\fILOCALES\fR
+Install only locales listed in \fILOCALES\fR. The default is to install all available locales.
+
+.TP
+\fB--compress\fR[=\fIno\fR,\fIxz\fR,\fIgz\fR,\fIlzo\fR]
+Compress GRUB files using the specified compression algorithm.
+
+.TP
+\fB--directory\fR=\fIDIR\fR
+Use images and modules in \fIDIR\fR.
+
+.TP
+\fB--grub-mkimage\fR=\fIFILE\fR
+Use \fIFILE\fR as \fBgrub-mkimage\fR(1). The default is \fI/usr/bin/grub-mkimage\fR.
+
+.TP
+\fB--rom-directory\fR=\fIDIR\fR
+Save ROM images in \fIDIR\fR.
+
+.TP
+\fB--xorriso\fR=\fIFILE\fR
+Use \fIFILE\fR as \fBxorriso\fI.
+
+.TP
+\fB--grub-glue-efi\fR=\fIFILE\fR
+Use \fIFILE\fR as \fBgrub-glue-efi\fR(3).
+
+.TP
+\fB--grub-render-label\fR=\fIFILE\fR
+Use \fIFILE\fR as \fBgrub-render-label\fR(3).
+
+.TP
+\fB--label-font\fR=\fIFILE\fR
+Use \fIFILE\fR as the font file for generated labels.
+
+.TP
+\fB--label-color\fR=\fICOLOR\fR
+Use \fICOLOR\fI as the color for generated labels.
+
+.TP
+\fB--label-bgcolor\fR=\fICOLOR\fR
+Use \fICOLOR\fR as the background color for generated labels.
+
+.TP
+\fB--product-name\fR=\fISTRING\fR
+Use \fISTRING\fR as the product name in generated labels.
+
+.TP
+\fB--product-version\fR=\fISTRING\fR
+Use \fISTRING\fR as the product version in generated labels.
+
+.TP
+\fB--sparc-boot\fR
+Enable booting the SPARC platform. This disables HFS+, APM, ARCS, and "boot as disk image" on the \fIi386-pc\fR target platform.
+
+.TP
+\fB--arcs-boot\fR
+Enable ARCS booting. This is typically for big-endian MIPS machines, and disables HFS+, APM, sparc64, and "boot as disk image" on the \fIi386-pc\fR target platform.
+
+.TP
+\fB--\fR
+All options after a \fB--\fR will be passed directly to xorriso's command line when generating the image.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-mkstandalone.1 b/util/grub-mkstandalone.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..ba2d2bdf279
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-mkstandalone.1
@@ -0,0 +1,100 @@
+.TH GRUB-MKSTANDALONE 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-mkstandalone\fR \(em Generate a standalone image in the selected format.
+
+.SH SYNOPSIS
+\fBgrub-mkstandalone\fR [-o | --output=\fIFILE\fR] [-O | --format=\fIFORMAT\fR]
+.RS 19
+[-C | --compression=(\fIxz\fR|\fInone\fR|\fIauto\fR)]
+.RE
+.RS 19
+[--modules=\fIMODULES\fR] [--install-modules=\fIMODULES\fR]
+.RE
+.RS 19
+[--themes=\fITHEMES\fR] [--fonts=\fIFONTS\fR]
+.RE
+.RS 19
+[--locales=\fILOCALES\fR] [--compress[=\fIno\fR,\fIxz\fR,\fIgz\fR,\fIlzo\fR]]
+.RE
+.RS 19
+[-d | --directory=\fIDIR\fR] [--grub-mkimage=\fIFILE\fR]
+.RE
+.RS 19
+\fISOURCE...\fR
+
+.SH DESCRIPTION
+
+.SH OPTIONS
+.TP
+--output=\fIFILE\fR
+Write the generated file to \fIFILE\fR. The default is to write to standard output.
+
+.TP
+--format=\fIFORMAT\fR
+Generate an image in the specified \fIFORMAT\fR. Valid values are:
+.RS
+.RS 4
+.P
+i386-coreboot,
+i386-multiboot,
+i386-pc,
+i386-pc-pxe,
+i386-efi,
+i386-ieee1275,
+i386-qemu,
+x86_64-efi,
+mipsel-yeeloong-flash,
+mipsel-fuloong2f-flash,
+mipself-loongson-elf,
+powerpc-ieee1275,
+sparc64-ieee1275-raw,
+sparc64-ieee1275-cdcore,
+sparc64-ieee1275-aout,
+ia64-efi,
+mips-arc,
+mipsel-arc,
+mipsel-qemu_mips-elf,
+mips-qemu_mips-flash,
+mipsel-qemu_mips-flash,
+mips-qemu_mips-elf
+.RE
+.RE
+
+.TP
+--compression=(\fIxz\fR|\fInone\fR|\fIauto\fR)
+Use one of \fIxz\fR, \fInone\fR, or \fIauto\fR as the compression method for the core image.
+
+.TP
+--modules=\fIMODULES\fR
+Pre-load modules specified by \fIMODULES\fR.
+
+.TP
+--install-modules=\fIMODULES\fR
+Install only \fIMODULES\fR and their dependencies. The default is to install all available modules.
+
+.TP
+--themes=\fITHEMES\fR
+Install \fITHEMES\fR. The default is to install the \fIstarfield\fR theme, if available.
+
+.TP
+--fonts=\fIFONTS\fR
+Install \fIFONTS\fR. The default is to install the \fIunicode\fR font.
+
+.TP
+--locales=\fILOCALES\fR
+Install only locales listed in \fILOCALES\fR. The default is to install all available locales.
+
+.TP
+--compress[=\fIno\fR,\fIxz\fR,\fIgz\fR,\fIlzo\fR]
+Compress GRUB files using the specified compression algorithm.
+
+.TP
+--directory=\fIDIR\fR
+Use images and modules in \fIDIR\fR.
+
+.TP
+--grub-mkimage=\fIFILE\fR
+Use \fIFILE\fR as \fBgrub-mkimage\fR. The default is \fI/usr/bin/grub-mkimage\fR.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-ofpathname.8 b/util/grub-ofpathname.8
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..bf3743aeba1
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-ofpathname.8
@@ -0,0 +1,12 @@
+.TH GRUB-OFPATHNAME 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-ofpathname\fR \(em Generate an IEEE-1275 device path for a specified device.
+
+.SH SYNOPSIS
+\fBgrub-ofpathname\fR \fIDEVICE\fR
+
+.SH DESCRIPTION
+\fBgrub-ofpathname\fR generates an IEEE-1275 device path for the specified \fIDEVICE\fR.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-probe.8 b/util/grub-probe.8
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..04e26c832bb
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-probe.8
@@ -0,0 +1,80 @@
+.TH GRUB-PROBE 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-probe\fR \(em Probe device information for a given path.
+
+.SH SYNOPSIS
+\fBgrub-probe\fR \[-d | --device] [-m | --device-map=\fIFILE\fR]
+.RS 12
+[-t | --target=(fs|fs_uuid|fs_label|drive|device|partmap|
+.RE
+.RS 28
+abstraction|cryptodisk_uuid|
+.RE
+.RS 28
+msdos_parttype)]
+.RE
+.RS 12
+[-v | --verbose] (PATH|DEVICE)
+
+.SH DESCRIPTION
+\fBgrub-probe\fR probes a path or device for filesystem and related information.
+
+.SH OPTIONS
+.TP
+--device
+Final option represents a \fIDEVICE\fR, rather than a filesystem \fIPATH\fR.
+.TP
+--device-map=\fIFILE\fR
+Use \fIFILE\fR as the device map. The default value is \fI/boot/grub/device.map\fR.
+
+.TP
+--target=(fs|fs_uuid|fs_label|drive|device|partmap|msdos_parttype)
+Select among various output definitions. The default is \fIfs\fR.
+.RS
+.TP
+\fIfs\fR
+filesystem module
+
+.TP
+\fIfs_uuid\fR
+filesystem UUID
+
+.TP
+\fIfs_label\fR
+filesystem label
+
+.TP
+\fIdrive\fR
+GRUB drive name
+
+.TP
+\fIdevice\fR
+System device
+
+.TP
+\fIpartmap\fR
+partition map module
+
+.TP
+\fIabstraction\fR
+abstraction module
+
+.TP
+\fIcryptodisk_uuid\fR
+cryptographic container
+
+.TP
+\fImsdos_partmap\fR
+MS-DOS partition map
+.RE
+
+.TP
+--verbose
+Print verbose output.
+
+.TP
+(\fIPATH\fR|\fIDEVICE\fR)
+If --device is passed, a block \fIDEVICE\fR. Otherwise, the \fIPATH\fR of a file on the filesystem.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-reboot.8 b/util/grub-reboot.8
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..faa5e4eece2
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-reboot.8
@@ -0,0 +1,21 @@
+.TH GRUB-REBOOT 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-reboot\fR \(em Set the default boot menu entry for the next boot only.
+
+.SH SYNOPSIS
+\fBgrub-reboot\fR [--boot-directory=\fIDIR\fR] \fIMENU_ENTRY\fR
+
+.SH DESCRIPTION
+\fBgrub-reboot\fR sets the default boot menu entry for the next boot, but not further boots after that. This command only works for GRUB configuration files created with \fIGRUB_DEFAULT=saved\fR in \fI/etc/default/grub\fR.
+
+.SH OPTIONS
+.TP
+--boot-directory=\fIDIR\fR
+Find GRUB images under \fIDIR/grub\fR. The default value is \fI/boot\fR, resulting in grub images being search for at \fI/boot/grub\fR.
+
+.TP
+\fIMENU_ENTRY\fR
+A number, a menu item title or a menu item identifier.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-render-label.1 b/util/grub-render-label.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..4d51c8abf01
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-render-label.1
@@ -0,0 +1,51 @@
+.TH GRUB-RENDER-LABEL 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-render-label\fR \(em Render an Apple disk label.
+
+.SH SYNOPSIS
+\fBgrub-render-label\fR [-b | --bgcolor=\fICOLOR\fR] [-c | --color=\fICOLOR\fR]
+.RS 19
+[-f | --font=\fIFILE\fR] [-i | --input=\fIFILE\fR]
+.RE
+.RS 19
+[-o | --output=\fIFILE\fR] [-t | --text=\fISTRING\fR]
+.RE
+.RS 19
+[-v | --verbose]
+
+.SH DESCRIPTION
+\fBgrub-render-label\fR renders an Apple disk label (.disk_label) file.
+
+
+.SH OPTIONS
+.TP
+\fB--color\fR=\fICOLOR\fR
+Use \fICOLOR\fI as the color for generated labels.
+
+.TP
+\fB--bgcolor\fR=\fICOLOR\fR
+Use \fICOLOR\fR as the background color for generated labels.
+
+.TP
+\fB--font\fR=\fIFILE\fR
+Use \fIFILE\fR as the font file for generated labels.
+
+.TP
+--input=\fIFILE\fR
+Read input text from \fIFILE\fR.
+
+.TP
+--output=\fIFILE\fR
+Render output to \fIFILE\fR.
+
+.TP
+--text=\fISTRING\fR
+Use \fISTRING\fR as input text.
+
+.TP
+--verbose
+Print verbose output.
+
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-script-check.1 b/util/grub-script-check.1
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..0f1f625b05d
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-script-check.1
@@ -0,0 +1,21 @@
+.TH GRUB-SCRIPT-CHECK 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-script-check\fR \(em Check GRUB configuration file for syntax errors.
+
+.SH SYNOPSIS
+\fBgrub-script-check\fR [-v | --verbose] \fIPATH\fR
+
+.SH DESCRIPTION
+\fBgrub-script-check\fR verifies that a specified GRUB configuration file does not contain syntax errors.
+
+.SH OPTIONS
+.TP
+--verbose
+Print verbose output.
+
+.TP
+\fIPATH\fR
+Path of the file to use as input.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-set-default.8 b/util/grub-set-default.8
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..a96265a1509
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-set-default.8
@@ -0,0 +1,21 @@
+.TH GRUB-SET-DEFAULT 1 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-set-default\fR \(em Set the default boot menu entry for GRUB.
+
+.SH SYNOPSIS
+\fBgrub-set-default\fR [--boot-directory=\fIDIR\fR] \fIMENU_ENTRY\fR
+
+.SH DESCRIPTION
+\fBgrub-set-default\fR sets the default boot menu entry for all subsequent boots. This command only works for GRUB configuration files created with \fIGRUB_DEFAULT=saved\fR in \fI/etc/default/grub\fR.
+
+.SH OPTIONS
+.TP
+--boot-directory=\fIDIR\fR
+Find GRUB images under \fIDIR/grub\fR. The default value is \fI/boot\fR, resulting in grub images being search for at \fI/boot/grub\fR.
+
+.TP
+\fIMENU_ENTRY\fR
+A number, a menu item title or a menu item identifier.
+
+.SH SEE ALSO
+.BR "info grub"
diff --git a/util/grub-sparc64-setup.8 b/util/grub-sparc64-setup.8
new file mode 100644
Update grub2 for f28 - Try to fix things for new compiler madness. I really don't know why gcc decided __attribute__((packed)) on a "typedef struct" should imply __attribute__((align (1))) and that it should have a warning that it does so. The obvious behavior would be to keep the alignment of the first element unless it's used in another object or type that /also/ hask the packed attribute. Why should it change the default alignment at all? - Merge in the BLS patches Javier and I wrote. - Attempt to fix pmtimer initialization failures to not be super duper slow. Signed-off-by: Peter Jones <pjones@redhat.com>
2018-02-27 18:56:41 +00:00
index 00000000000..37ea2dd5eaa
Re-merge RHEL 7 changes and ARM works in progress.
2014-07-02 15:49:28 +00:00
--- /dev/null
+++ b/util/grub-sparc64-setup.8
@@ -0,0 +1,12 @@
+.TH GRUB-SPARC64-SETUP 3 "Wed Feb 26 2014"
+.SH NAME
+\fBgrub-sparc64-setup\fR \(em Set up a device to boot a sparc64 GRUB image.
+
+.SH SYNOPSIS
+\fBgrub-sparc64-setup\fR [OPTIONS].
+
+.SH DESCRIPTION
+You should not normally run this program directly. Use grub-install instead.
+
+.SH SEE ALSO
+.BR "info grub"
Reference in New Issue Copy Permalink