[Buildroot] boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Buildroot] boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Mikhail B. WproxyM]

We are using binutils-2.35.2+ (later then 2.29.1), grub-2.04,
cross-compiling on recent buildroot under i386/i686 and getting error:

> lib/i386/relocator64.S:66: Error: unknown pseudo-op: `.code64'

So, the working patch is attached. Please add it to buildroot.

Sources:
1) https://debbugs.gnu.org/cgi/bugreport.cgi?bug=41982 (Jul 2020)
2)
http://archive.linuxfromscratch.org/mail-archives/lfs-support/2018-January/051723.html

--
Best regards
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20210420/a08994a2/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0150-grub-core-build-fixes-for-i386.patch
Type: text/x-patch
Size: 1722 bytes
Desc: not available
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20210420/a08994a2/attachment.bin>

[Buildroot] boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Thomas Petazzoni]

Hello Mikhail,

On Tue, 20 Apr 2021 12:35:16 +0300
"Mikhail B. WproxyM" <w.proxy.m@gmail.com> wrote:

> We are using binutils-2.35.2+ (later then 2.29.1), grub-2.04,
> cross-compiling on recent buildroot under i386/i686 and getting error:
> 
> > lib/i386/relocator64.S:66: Error: unknown pseudo-op: `.code64'  
> 
> So, the working patch is attached. Please add it to buildroot.

Do you have some details on the conditions to reproduce this issue?
Indeed, I just tried to build grub2, with the following configuration:

BR2_x86_core2=y
BR2_TOOLCHAIN_EXTERNAL=y
BR2_INIT_NONE=y
BR2_SYSTEM_BIN_SH_NONE=y
# BR2_PACKAGE_BUSYBOX is not set
# BR2_TARGET_ROOTFS_TAR is not set
BR2_TARGET_GRUB2=y
BR2_TARGET_GRUB2_I386_EFI=y

So it's using a i686 target system, with a toolchain that uses binutils
2.34 (i.e later than the 2.29 you're pointing). And it builds fine,
including relocator64.S, which gets built as part of the build process.

Note: I am building *from* an x86-64 system. But how can that matter,
as relocator64.S is built using the cross-compiler?

Also, why isn't this patch in upstream Grub ?

Thanks a lot for the additional details you could provide.

Best regards,

Thomas
-- 
Thomas Petazzoni, co-owner and CEO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

[Buildroot] boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Mikhail B. WproxyM]

Hello, Thomas.

I have the build error, when building *from i686* to i686.

You do not have the build error, when building *from x86-64* to i686.
Looks like it is the reason.

Grub2 developers do not run build tests under i386/i686. Even you do
not run build tests under i686.

P.S.: Common build platform for most developers is x86-64, so I am not
surprised. It is too simple to lose compatibility by using too many
levels of assumptions, facades and abstractions.

--
Best regards

--
Best regards, Mikhail B.


??, 26 ???. 2021 ?. ? 23:56, Thomas Petazzoni <thomas.petazzoni@bootlin.com>:
>
> Hello Mikhail,
>
> On Tue, 20 Apr 2021 12:35:16 +0300
> "Mikhail B. WproxyM" <w.proxy.m@gmail.com> wrote:
>
> > We are using binutils-2.35.2+ (later then 2.29.1), grub-2.04,
> > cross-compiling on recent buildroot under i386/i686 and getting error:
> >
> > > lib/i386/relocator64.S:66: Error: unknown pseudo-op: `.code64'
> >
> > So, the working patch is attached. Please add it to buildroot.
>
> Do you have some details on the conditions to reproduce this issue?
> Indeed, I just tried to build grub2, with the following configuration:
>
> BR2_x86_core2=y
> BR2_TOOLCHAIN_EXTERNAL=y
> BR2_INIT_NONE=y
> BR2_SYSTEM_BIN_SH_NONE=y
> # BR2_PACKAGE_BUSYBOX is not set
> # BR2_TARGET_ROOTFS_TAR is not set
> BR2_TARGET_GRUB2=y
> BR2_TARGET_GRUB2_I386_EFI=y
>
> So it's using a i686 target system, with a toolchain that uses binutils
> 2.34 (i.e later than the 2.29 you're pointing). And it builds fine,
> including relocator64.S, which gets built as part of the build process.
>
> Note: I am building *from* an x86-64 system. But how can that matter,
> as relocator64.S is built using the cross-compiler?
>
> Also, why isn't this patch in upstream Grub ?
>
> Thanks a lot for the additional details you could provide.
>
> Best regards,
>
> Thomas
> --
> Thomas Petazzoni, co-owner and CEO, Bootlin
> Embedded Linux and Kernel engineering
> https://bootlin.com

[Buildroot] boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Thomas Petazzoni]

Hello Mikhail,

On Wed, 28 Apr 2021 11:07:13 +0300
"Mikhail B. WproxyM" <w.proxy.m@gmail.com> wrote:

> I have the build error, when building *from i686* to i686.
> 
> You do not have the build error, when building *from x86-64* to i686.
> Looks like it is the reason.
> 
> Grub2 developers do not run build tests under i386/i686. Even you do
> not run build tests under i686.

I was worried the issue would be related from the difference of the
build machine architecture. In theory it shouldn't matter, as the file
being built is built with the cross-compiler, but it seems like there
is an actual difference.

What about having this patch submitted to upstream Grub ?

Best regards,

Thomas
-- 
Thomas Petazzoni, co-owner and CEO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

[Buildroot] boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Mikhail B. WproxyM]

>What about having this patch submitted to upstream Grub ?

Idea is nice, but I am not sure that it will work.
The patch was already ignored some time before by grub2 developers,
because it is not included in upstream. I will try.

--
Best regards, Mikhail B.

??, 28 ???. 2021 ?. ? 14:44, Thomas Petazzoni <thomas.petazzoni@bootlin.com>:
>
> Hello Mikhail,
>
> On Wed, 28 Apr 2021 11:07:13 +0300
> "Mikhail B. WproxyM" <w.proxy.m@gmail.com> wrote:
>
> > I have the build error, when building *from i686* to i686.
> >
> > You do not have the build error, when building *from x86-64* to i686.
> > Looks like it is the reason.
> >
> > Grub2 developers do not run build tests under i386/i686. Even you do
> > not run build tests under i686.
>
> I was worried the issue would be related from the difference of the
> build machine architecture. In theory it shouldn't matter, as the file
> being built is built with the cross-compiler, but it seems like there
> is an actual difference.
>
> What about having this patch submitted to upstream Grub ?
>
> Best regards,
>
> Thomas
> --
> Thomas Petazzoni, co-owner and CEO, Bootlin
> Embedded Linux and Kernel engineering
> https://bootlin.com

[Buildroot] boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Mikhail B. WproxyM]

As expected, they do not want to include fix patch, because "The GRUB 2.04
release is a few years old."

--
Best regards, Mikhail B.


??, 28 ???. 2021 ?. ? 16:23, Mikhail B. WproxyM <w.proxy.m@gmail.com>:

> >What about having this patch submitted to upstream Grub ?
>
> Idea is nice, but I am not sure that it will work.
> The patch was already ignored some time before by grub2 developers,
> because it is not included in upstream. I will try.
>
> --
> Best regards, Mikhail B.
>
> ??, 28 ???. 2021 ?. ? 14:44, Thomas Petazzoni <
> thomas.petazzoni at bootlin.com>:
> >
> > Hello Mikhail,
> >
> > On Wed, 28 Apr 2021 11:07:13 +0300
> > "Mikhail B. WproxyM" <w.proxy.m@gmail.com> wrote:
> >
> > > I have the build error, when building *from i686* to i686.
> > >
> > > You do not have the build error, when building *from x86-64* to i686.
> > > Looks like it is the reason.
> > >
> > > Grub2 developers do not run build tests under i386/i686. Even you do
> > > not run build tests under i686.
> >
> > I was worried the issue would be related from the difference of the
> > build machine architecture. In theory it shouldn't matter, as the file
> > being built is built with the cross-compiler, but it seems like there
> > is an actual difference.
> >
> > What about having this patch submitted to upstream Grub ?
> >
> > Best regards,
> >
> > Thomas
> > --
> > Thomas Petazzoni, co-owner and CEO, Bootlin
> > Embedded Linux and Kernel engineering
> > https://bootlin.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20210513/0e749c49/attachment.html>

Re: [Buildroot] boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Mikhail B. WproxyM]

[-- Attachment #1: Type: text/plain, Size: 1534 bytes --]

As expected, they do not want to include fix patch, because "The GRUB 2.04
release is a few years old."

--
Best regards, Mikhail B.


ср, 28 апр. 2021 г. в 16:23, Mikhail B. WproxyM <w.proxy.m@gmail.com>:

> >What about having this patch submitted to upstream Grub ?
>
> Idea is nice, but I am not sure that it will work.
> The patch was already ignored some time before by grub2 developers,
> because it is not included in upstream. I will try.
>
> --
> Best regards, Mikhail B.
>
> ср, 28 апр. 2021 г. в 14:44, Thomas Petazzoni <
> thomas.petazzoni@bootlin.com>:
> >
> > Hello Mikhail,
> >
> > On Wed, 28 Apr 2021 11:07:13 +0300
> > "Mikhail B. WproxyM" <w.proxy.m@gmail.com> wrote:
> >
> > > I have the build error, when building *from i686* to i686.
> > >
> > > You do not have the build error, when building *from x86-64* to i686.
> > > Looks like it is the reason.
> > >
> > > Grub2 developers do not run build tests under i386/i686. Even you do
> > > not run build tests under i686.
> >
> > I was worried the issue would be related from the difference of the
> > build machine architecture. In theory it shouldn't matter, as the file
> > being built is built with the cross-compiler, but it seems like there
> > is an actual difference.
> >
> > What about having this patch submitted to upstream Grub ?
> >
> > Best regards,
> >
> > Thomas
> > --
> > Thomas Petazzoni, co-owner and CEO, Bootlin
> > Embedded Linux and Kernel engineering
> > https://bootlin.com
>

[-- Attachment #2: Type: text/html, Size: 2541 bytes --]

Re: [Buildroot] boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[John Paul Adrian Glaubitz]

Hello Mikhail!

On 5/13/21 9:19 AM, Mikhail B. WproxyM wrote:
> As expected, they do not want to include fix patch, because "The GRUB 2.04
> release is a few years old."

Since GRUB 2.06 will hopefully be released within the next weeks, you should
be able to upgrade to the new upstream release, aren't you?

Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer - glaubitz@debian.org
`. `'   Freie Universitaet Berlin - glaubitz@physik.fu-berlin.de
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913

Fwd: boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Mikhail B. WproxyM]

[-- Attachment #1: Type: text/plain, Size: 442 bytes --]

We are using binutils-2.35.2+ (later then 2.29.1), grub-2.04,
cross-compiling on recent buildroot under i386/i686 and getting error:

> lib/i386/relocator64.S:66: Error: unknown pseudo-op: `.code64'

So, the working patch is attached. Please add it to grub2.

Sources:
1) https://debbugs.gnu.org/cgi/bugreport.cgi?bug=41982 (Jul 2020)
2) http://archive.linuxfromscratch.org/mail-archives/lfs-support/2018-January/051723.html

--
Best regards

[-- Attachment #2: 0150-grub-core-build-fixes-for-i386.patch --]
[-- Type: application/x-patch, Size: 1722 bytes --]

boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Mikhail B. WproxyM]

[-- Attachment #1: Type: text/plain, Size: 442 bytes --]

We are using binutils-2.35.2+ (later then 2.29.1), grub-2.04,
cross-compiling on recent buildroot under i386/i686 and getting error:

> lib/i386/relocator64.S:66: Error: unknown pseudo-op: `.code64'

So, the working patch is attached. Please add it to grub2.

Sources:
1) https://debbugs.gnu.org/cgi/bugreport.cgi?bug=41982 (Jul 2020)
2) http://archive.linuxfromscratch.org/mail-archives/lfs-support/2018-January/051723.html

--
Best regards

[-- Attachment #2: 0150-grub-core-build-fixes-for-i386.patch --]
[-- Type: application/x-patch, Size: 1722 bytes --]

Re: boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Daniel Kiper]

Hi,

Thank you for the report...

On Wed, May 05, 2021 at 12:44:07PM +0300, Mikhail B. WproxyM wrote:
> We are using binutils-2.35.2+ (later then 2.29.1), grub-2.04,
> cross-compiling on recent buildroot under i386/i686 and getting error:

May I ask you to try latest GRUB master git branch [1]? The GRUB 2.04
release is a few years old. We are going to release 2.06 soon.

> > lib/i386/relocator64.S:66: Error: unknown pseudo-op: `.code64'
>
> So, the working patch is attached. Please add it to grub2.

You can find a few comments below...

> From 270667540146f8ef9ea7a44258a71b3837a7af4a Mon Sep 17 00:00:00 2001
> From: "Jan (janneke) Nieuwenhuizen" <janneke@gnu.org>
> Date: Sun, 21 Jun 2020 15:10:40 +0200
> Subject: [PATCH] grub-core: Build fixes for i386
>
> * grub-core/lib/i386/relocator64.S: Avoid x86_64 instructions on i386.

Please describe in the commit message why this patch is needed and how
the issue can be reproduced.

Please add your "Signed-off-by: ..." line at the end of commit message.
You can find good examples of commit messages in the GRUB git repository.

> ---
>  grub-core/lib/i386/relocator64.S | 26 +++++++++++++++++++++++++-
>  1 file changed, 25 insertions(+), 1 deletion(-)
>
> diff --git a/grub-core/lib/i386/relocator64.S b/grub-core/lib/i386/relocator64.S
> index 148f38adb..45fed9444 100644
> --- a/grub-core/lib/i386/relocator64.S
> +++ b/grub-core/lib/i386/relocator64.S
> @@ -63,7 +63,9 @@ VARIABLE(grub_relocator64_cr3)
>  	movq	%rax, %cr3
>  #endif
>
> +#ifdef __x86_64__
>  	.code64
> +#endif
>
>  	/* mov imm64, %rax */
>  	.byte 	0x48
> @@ -71,7 +73,14 @@ VARIABLE(grub_relocator64_cr3)
>  VARIABLE(grub_relocator64_rsp)
>  	.quad	0
>
> +#ifndef __x86_64__
> +	/* movq	%rax, %rsp */
> +	.byte 	0x48
> +	.byte	0x89
> +	.byte	0xc4
> +#else
>  	movq	%rax, %rsp
> +#endif

I would do it other way around to increase readability...

#ifdef __x86_64__
     movq    %rax, %rsp
#else
     /* movq %rax, %rsp */
     .byte   0x48
     ...

>  #ifdef GRUB_MACHINE_EFI
>  	jmp	LOCAL(skip_efi_stack_align)
> @@ -95,8 +104,15 @@ LOCAL(skip_efi_stack_align):
>  VARIABLE(grub_relocator64_rsi)
>  	.quad	0
>
> +#ifndef	__x86_64__
> +	/* movq	%rax, %rsi */
> +	.byte 	0x48
> +	.byte	0x89
> +	.byte 	0xc6
> +#else
>  	movq	%rax, %rsi
> -
> +#endif

Ditto...

>  	/* mov imm64, %rax */
>  	.byte 	0x48
>  	.byte	0xb8
> @@ -128,6 +144,14 @@ VARIABLE(grub_relocator64_rdx)
>  #ifdef __APPLE__
>  	.byte 0xff, 0x25
>  	.quad 0
> +#elif !defined (__x86_64__)
> +	/* jmp *LOCAL(jump_addr) (%rip) */
> +	.byte 0xff
> +        .byte 0x25
> +	.byte 0
> +	.byte 0
> +	.byte 0
> +	.byte 0
>  #else
>  	jmp *LOCAL(jump_addr) (%rip)
>  #endif

#if defined(__APPLE__) || !defined (__x86_64__)
     .byte 0xff, 0x25
     .quad 0
#else
     jmp *LOCAL(jump_addr) (%rip)
#endif

A patch should be prepared on top of latest GRUB master git branch [1].
And please use "git-send-email" to send the patch as Adrian said...

Daniel

[1] https://git.savannah.gnu.org/gitweb/?p=grub.git&view=view+git+repository

Re: boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Jan Nieuwenhuizen]

[-- Attachment #1: Type: text/plain, Size: 6127 bytes --]

Daniel Kiper writes:

Hello Daniel,

> May I ask you to try latest GRUB master git branch [1]? The GRUB 2.04
> release is a few years old. We are going to release 2.06 soon.

Sure.  The bug is still there (see attached patch to reproduce using
Guix):

--8<---------------cut here---------------start------------->8---
$ ./pre-inst-env guix build --system=i686-linux --target=i586-pc-gnu grub 
starting phase `unpack'
grub-2.06~rc1/
[..]
starting phase `configure'
source directory: "/tmp/guix-build-grub-2.06-rc1.drv-0/grub-2.06~rc1" (relative from build: ".")
build directory: "/tmp/guix-build-grub-2.06-rc1.drv-0/grub-2.06~rc1"
configure flags: ("CC_FOR_BUILD=gcc" "CONFIG_SHELL=/gnu/store/v1g7f3p4f0851mywrla8qmr9hb8jgfjr-bash-minimal-5.0.16/bin/bash" "SHELL=/gnu/store/v1g7f3p4f0851mywrla8qmr9hb8jgfjr-bash-minimal-5.0.16/bin/bash" "--prefix=/gnu/store/753q8z0xa9ijap3p3phpsvns7287c7sn-grub-2.06-rc1" "--enable-fast-install" "--build=i686-unknown-linux-gnu" "--host=i586-pc-gnu" "PYTHON=true")
[..]
starting phase `build'
[..]
i586-pc-gnu-gcc -DHAVE_CONFIG_H -I. -I..  -Wall -W  -DGRUB_MACHINE_PCBIOS=1 -DGRUB_MACHINE=I386_PC -m32 -nostdinc -isystem /gnu/store/q8af3pvwsknn132pl1nzrb5281i4pgij-gcc-cross-i586-pc-gnu-7.5.0-lib/lib/gcc/i586-pc-gnu/7.5.0/include -I../include -I../include -DGRUB_FILE=\"lib/i386/relocator.c\" -I. -I. -I.. -I.. -I../include -I../include -I../grub-core/lib/libgcrypt-grub/src/    -D_FILE_OFFSET_BITS=64 -std=gnu99 -Os -m32 -Wall -W -Wshadow -Wpointer-arith -Wundef -Wchar-subscripts -Wcomment -Wdeprecated-declarations -Wdisabled-optimization -Wdiv-by-zero -Wfloat-equal -Wformat-extra-args -Wformat-security -Wformat-y2k -Wimplicit -Wimplicit-function-declaration -Wimplicit-int -Wmain -Wmissing-braces -Wmissing-format-attribute -Wmultichar -Wparentheses -Wreturn-type -Wsequence-point -Wshadow -Wsign-compare -Wswitch -Wtrigraphs -Wunknown-pragmas -Wunused -Wunused-function -Wunused-label -Wunused-parameter -Wunused-value  -Wunused-variable -Wwrite-strings -Wnested-externs -Wstrict-prototypes -g -Wredundant-decls -Wmissing-prototypes -Wmissing-declarations  -Wextra -Wattributes -Wendif-labels -Winit-self -Wint-to-pointer-cast -Winvalid-pch -Wmissing-field-initializers -Wnonnull -Woverflow -Wvla -Wpointer-to-int-cast -Wstrict-aliasing -Wvariadic-macros -Wvolatile-register-var -Wpointer-sign -Wmissing-include-dirs -Wmissing-prototypes -Wmissing-declarations -Wformat=2 -march=i386 -mrtd -mregparm=3 -falign-jumps=1 -falign-loops=1 -falign-functions=1 -freg-struct-return -mno-mmx -mno-sse -mno-sse2 -mno-sse3 -mno-3dnow -msoft-float -fno-dwarf2-cfi-asm -mno-stack-arg-probe -fno-asynchronous-unwind-tables -fno-unwind-tables -fno-ident -fno-stack-protector -Wtrampolines -Werror   -ffreestanding   -MT lib/i386/relocator_module-relocator.o -MD -MP -MF lib/i386/.deps-core/relocator_module-relocator.Tpo -c -o lib/i386/relocator_module-relocator.o `test -f 'lib/i386/relocator.c' || echo './'`lib/i386/relocator.c
lib/i386/relocator64.S: Assembler messages:
lib/i386/relocator64.S:66: Error: unknown pseudo-op: `.code64'
lib/i386/relocator64.S:74: Error: bad register name `%rax'
lib/i386/relocator64.S:98: Error: bad register name `%rax'
lib/i386/relocator64.S:132: Error: bad register name `%rip)'
--8<---------------cut here---------------end--------------->8---

>> > lib/i386/relocator64.S:66: Error: unknown pseudo-op: `.code64'
>>
>> So, the working patch is attached. Please add it to grub2.
>
> You can find a few comments below...

Thanks!

>> From 270667540146f8ef9ea7a44258a71b3837a7af4a Mon Sep 17 00:00:00 2001
>> From: "Jan (janneke) Nieuwenhuizen" <janneke@gnu.org>
>> Date: Sun, 21 Jun 2020 15:10:40 +0200
>> Subject: [PATCH] grub-core: Build fixes for i386
>>
>> * grub-core/lib/i386/relocator64.S: Avoid x86_64 instructions on i386.
>
> Please describe in the commit message why this patch is needed and how
> the issue can be reproduced.

I have added:

--8<---------------cut here---------------start------------->8---
This fixes cross-compiling to x86 (e.g., the Hurd) from x86-linux.

To reproduce, update the Grub source description in your local Guix
archive and run

   ./pre-inst-env guix build --system=i686-linux --target=i586-pc-gnu grub 

or install an x86 cross-build environment on x86-linux (32bit!) and
configure to cross build and make, e.g., do something like

    ./configure \
       CC_FOR_BUILD=gcc \
       --build=i686-unknown-linux-gnu
       --host=i586-pc-gnu
    make
--8<---------------cut here---------------end--------------->8---

>> diff --git a/grub-core/lib/i386/relocator64.S b/grub-core/lib/i386/relocator64.S
>> index 148f38adb..45fed9444 100644
[..]
>> +#ifndef __x86_64__
>> +	/* movq	%rax, %rsp */
>> +	.byte 	0x48
>> +	.byte	0x89
>> +	.byte	0xc4
>> +#else
>>  	movq	%rax, %rsp
>> +#endif
>
> I would do it other way around to increase readability...
>
> #ifdef __x86_64__
>      movq    %rax, %rsp
> #else
>      /* movq %rax, %rsp */
>      .byte   0x48
>      ...

Done.

>> +#ifndef	__x86_64__
>> +	/* movq	%rax, %rsi */
>> +	.byte 	0x48
>> +	.byte	0x89
>> +	.byte 	0xc6
>> +#else
>>  	movq	%rax, %rsi
>> -
>> +#endif
>
> Ditto...

Ok.

>>  #ifdef __APPLE__
>>  	.byte 0xff, 0x25
>>  	.quad 0
>> +#elif !defined (__x86_64__)
>> +	/* jmp *LOCAL(jump_addr) (%rip) */
>> +	.byte 0xff
>> +        .byte 0x25
>> +	.byte 0
>> +	.byte 0
>> +	.byte 0
>> +	.byte 0
>>  #else
>>  	jmp *LOCAL(jump_addr) (%rip)
>>  #endif
>
> #if defined(__APPLE__) || !defined (__x86_64__)
>      .byte 0xff, 0x25
>      .quad 0
> #else
>      jmp *LOCAL(jump_addr) (%rip)
> #endif

Ah, yes; that's better.  Also updated.

> A patch should be prepared on top of latest GRUB master git branch [1].

Sure, but well..

    https://lists.gnu.org/archive/html/bug-grub/2020-06/msg00013.html

it was when I sent the report ;)

Luckily it still applied cleanly to current master.

> And please use "git-send-email" to send the patch as Adrian said...

Sure, sent separately to bug-grub.

Greetings,
Janneke


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: grub-hurd-cross-build-reproducer.patch --]
[-- Type: text/x-patch, Size: 1185 bytes --]

diff --git a/gnu/packages/bootloaders.scm b/gnu/packages/bootloaders.scm
index 56bef57cb7..78923e4303 100644
--- a/gnu/packages/bootloaders.scm
+++ b/gnu/packages/bootloaders.scm
@@ -88,18 +88,15 @@
 (define-public grub
   (package
     (name "grub")
-    (version "2.04")
+    (version "2.06-rc1")
     (source (origin
              (method url-fetch)
-             (uri (string-append "mirror://gnu/grub/grub-" version ".tar.xz"))
+             (uri (string-append "https://alpha.gnu.org/pub/gnu/grub/"
+                                 "grub-2.06~rc1" ".tar.xz"))
+             (file-name (string-append "grub-" version ".tar.xz"))
              (sha256
               (base32
-               "0zgp5m3hmc9jh8wpjx6czzkh5id2y8n1k823x2mjvm2sk6b28ag5"))
-             (patches (search-patches
-                       "grub-efi-fat-serial-number.patch"
-                       "grub-setup-root.patch"
-                       "grub-verifiers-Blocklist-fallout-cleanup.patch"
-                       "grub-cross-system-i686.patch"))))
+               "112l2isb0rgdz4zz4ssmf5hklv0viz0674fdwr1h1d9a3vrg31rc"))))
     (build-system gnu-build-system)
     (arguments
      `(#:configure-flags

[-- Attachment #3: Type: text/plain, Size: 152 bytes --]


-- 
Jan Nieuwenhuizen <janneke@gnu.org> | GNU LilyPond http://lilypond.org
Freelance IT http://JoyofSource.com | Avatar® http://AvatarAcademy.com

Re: boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Daniel Kiper]

Hi Jan,

On Thu, May 13, 2021 at 12:35:25PM +0200, Jan Nieuwenhuizen wrote:
> Daniel Kiper writes:
>
> Hello Daniel,
>
> > May I ask you to try latest GRUB master git branch [1]? The GRUB 2.04
> > release is a few years old. We are going to release 2.06 soon.
>
> Sure.  The bug is still there (see attached patch to reproduce using
> Guix):
>
> --8<---------------cut here---------------start------------->8---
> $ ./pre-inst-env guix build --system=i686-linux --target=i586-pc-gnu grub
> starting phase `unpack'
> grub-2.06~rc1/
> [..]
> starting phase `configure'
> source directory: "/tmp/guix-build-grub-2.06-rc1.drv-0/grub-2.06~rc1" (relative from build: ".")
> build directory: "/tmp/guix-build-grub-2.06-rc1.drv-0/grub-2.06~rc1"
> configure flags: ("CC_FOR_BUILD=gcc" "CONFIG_SHELL=/gnu/store/v1g7f3p4f0851mywrla8qmr9hb8jgfjr-bash-minimal-5.0.16/bin/bash" "SHELL=/gnu/store/v1g7f3p4f0851mywrla8qmr9hb8jgfjr-bash-minimal-5.0.16/bin/bash" "--prefix=/gnu/store/753q8z0xa9ijap3p3phpsvns7287c7sn-grub-2.06-rc1" "--enable-fast-install" "--build=i686-unknown-linux-gnu" "--host=i586-pc-gnu" "PYTHON=true")
> [..]
> starting phase `build'
> [..]
> i586-pc-gnu-gcc -DHAVE_CONFIG_H -I. -I..  -Wall -W  -DGRUB_MACHINE_PCBIOS=1 -DGRUB_MACHINE=I386_PC -m32 -nostdinc -isystem /gnu/store/q8af3pvwsknn132pl1nzrb5281i4pgij-gcc-cross-i586-pc-gnu-7.5.0-lib/lib/gcc/i586-pc-gnu/7.5.0/include -I../include -I../include -DGRUB_FILE=\"lib/i386/relocator.c\" -I. -I. -I.. -I.. -I../include -I../include -I../grub-core/lib/libgcrypt-grub/src/    -D_FILE_OFFSET_BITS=64 -std=gnu99 -Os -m32 -Wall -W -Wshadow -Wpointer-arith -Wundef -Wchar-subscripts -Wcomment -Wdeprecated-declarations -Wdisabled-optimization -Wdiv-by-zero -Wfloat-equal -Wformat-extra-args -Wformat-security -Wformat-y2k -Wimplicit -Wimplicit-function-declaration -Wimplicit-int -Wmain -Wmissing-braces -Wmissing-format-attribute -Wmultichar -Wparentheses -Wreturn-type -Wsequence-point -Wshadow -Wsign-compare -Wswitch -Wtrigraphs -Wunknown-pragmas -Wunused -Wunused-function -Wunused-label -Wunused-parameter -Wunused-value  -Wunused-variable -Wwrite-strings -Wnested-externs -Wstrict-prototypes -g -Wredundant-decls -Wmissing-prototypes -Wmissing-declarations  -Wextra -Wattributes -Wendif-labels -Winit-self -Wint-to-pointer-cast -Winvalid-pch -Wmissing-field-initializers -Wnonnull -Woverflow -Wvla -Wpointer-to-int-cast -Wstrict-aliasing -Wvariadic-macros -Wvolatile-register-var -Wpointer-sign -Wmissing-include-dirs -Wmissing-prototypes -Wmissing-declarations -Wformat=2 -march=i386 -mrtd -mregparm=3 -falign-jumps=1 -falign-loops=1 -falign-functions=1 -freg-struct-return -mno-mmx -mno-sse -mno-sse2 -mno-sse3 -mno-3dnow -msoft-float -fno-dwarf2-cfi-asm -mno-stack-arg-probe -fno-asynchronous-unwind-tables -fno-unwind-tables -fno-ident -fno-stack-protector -Wtrampolines -Werror   -ffreestanding   -MT lib/i386/relocator_module-relocator.o -MD -MP -MF lib/i386/.deps-core/relocator_module-relocator.Tpo -c -o lib/i386/relocator_module-relocator.o `test -f 'lib/i386/relocator.c' || echo './'`lib/i386/relocator.c
> lib/i386/relocator64.S: Assembler messages:
> lib/i386/relocator64.S:66: Error: unknown pseudo-op: `.code64'
> lib/i386/relocator64.S:74: Error: bad register name `%rax'
> lib/i386/relocator64.S:98: Error: bad register name `%rax'
> lib/i386/relocator64.S:132: Error: bad register name `%rip)'
> --8<---------------cut here---------------end--------------->8---

[...]

> > A patch should be prepared on top of latest GRUB master git branch [1].
>
> Sure, but well..
>
>     https://lists.gnu.org/archive/html/bug-grub/2020-06/msg00013.html
>
> it was when I sent the report ;)

Sorry, I missed it...

> Luckily it still applied cleanly to current master.

Great!

> > And please use "git-send-email" to send the patch as Adrian said...
>
> Sure, sent separately to bug-grub.

I cannot find your updated patch. May I ask you to send it using
"git-send-email" to grub-devel@gnu.org?

Daniel

Re: boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[Jan Nieuwenhuizen]

Daniel Kiper writes:

Hello Daniel,

[..]

>> Sure, sent separately to bug-grub.
>
> I cannot find your updated patch.

It's here:

    https://lists.gnu.org/archive/html/bug-grub/2021-05/msg00009.html

> May I ask you to send it using "git-send-email" to grub-devel@gnu.org?

(waiting with this for a bit, please ping if needed :)

Greetings,
Janneke

-- 
Jan Nieuwenhuizen <janneke@gnu.org> | GNU LilyPond http://lilypond.org
Freelance IT http://JoyofSource.com | Avatar® http://AvatarAcademy.com

Re: boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[John Paul Adrian Glaubitz]

Hello Jan!

On 5/18/21 7:05 AM, Jan Nieuwenhuizen wrote:
>>> Sure, sent separately to bug-grub.
>>
>> I cannot find your updated patch.
> 
> It's here:
> 
>     https://lists.gnu.org/archive/html/bug-grub/2021-05/msg00009.html
> 
>> May I ask you to send it using "git-send-email" to grub-devel@gnu.org?
> 
> (waiting with this for a bit, please ping if needed :)

I think you should post the patch to this mailing list so it can be reviewed
here. Reviewing the patch when it's located in the archives of a different
mailing list is a bit awkward.

Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer - glaubitz@debian.org
`. `'   Freie Universitaet Berlin - glaubitz@physik.fu-berlin.de
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913

Re: Fwd: boot/grub2: grub-core-build-fixes-for-i386 (grub-2.04 + binutils-2.35.2 + pseudo-op .code64)

[John Paul Adrian Glaubitz]

Hello Mikhail!

On 5/5/21 11:39 AM, Mikhail B. WproxyM wrote:
> We are using binutils-2.35.2+ (later then 2.29.1), grub-2.04,
> cross-compiling on recent buildroot under i386/i686 and getting error:
> 
>> lib/i386/relocator64.S:66: Error: unknown pseudo-op: `.code64'
> 
> So, the working patch is attached. Please add it to grub2.

Please re-send your patch using git-send-email, see [1], for example.

Adrian

> [1] https://git-send-email.io/

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer - glaubitz@debian.org
`. `'   Freie Universitaet Berlin - glaubitz@physik.fu-berlin.de
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913