x86: Remove a.out support
Subsystems (4)
Borislav Petkov
maintainerPATCH 1/1 x86: Remove a.out support
From: Borislav Petkov <[email protected]>
Commit
eac616557050 ("x86: Deprecate a.out support")
deprecated a.out support with the promise to remove it a couple of releases later. That commit landed in v5.1.
Now it is more than a couple of releases later, no one has complained so remove it.
--- arch/x86/Kconfig | 7 - arch/x86/ia32/Makefile | 2 - arch/x86/ia32/ia32_aout.c | 325 -------------------------------------- 3 files changed, 334 deletions(-) delete mode 100644 arch/x86/ia32/ia32_aout.c
arch/x86/Kconfig +0 -7
@@ -2835,13 +2835,6 @@ config IA32_EMULATION 64-bit kernel. You should likely turn this on, unless you're 100% sure that you don't have any 32-bit programs left. -config IA32_AOUT - tristate "IA32 a.out support" - depends on IA32_EMULATION - depends on BROKEN - help - Support old a.out binaries in the 32bit emulation. - config X86_X32 bool "x32 ABI for 64-bit mode" depends on X86_64
arch/x86/ia32/Makefile +0 -2
@@ -5,7 +5,5 @@ obj-$(CONFIG_IA32_EMULATION) := ia32_signal.o -obj-$(CONFIG_IA32_AOUT) += ia32_aout.o - audit-class-$(CONFIG_AUDIT) := audit.o obj-$(CONFIG_IA32_EMULATION) += $(audit-class-y)
arch/x86/ia32/ia32_aout.c deleted +0 -325
325 lines not on this page · raw message
Signature · 1 line
2.29.2
Quoted 6 lines · Borislav Petkov
From: Borislav Petkov <[email protected]>
Commit
eac616557050("x86: Deprecate a.out support")
Quoted 7 lines
deprecated a.out support with the promise to remove it a couple of releases later. That commit landed in v5.1.
Now it is more than a couple of releases later, no one has complained so remove it.
Signed-off-by Borislav Petkov <[email protected]>
Yes please. :)
Signature · 1 line
Kees Cook
Quoted 6 lines · Borislav Petkov
From: Borislav Petkov <[email protected]>
Commit
eac616557050("x86: Deprecate a.out support")
Quoted 7 lines
deprecated a.out support with the promise to remove it a couple of releases later. That commit landed in v5.1.
Now it is more than a couple of releases later, no one has complained so remove it.
Signed-off-by Borislav Petkov <[email protected]>
Let's pour one out for being able to run Minix binaries on Linux. :-)
Speaking of which, if we're doing this, are there some old system calls that we could remove at the same time?
- Ted
Borislav Petkov
maintainerQuoted 1 line · Theodore Ts'o
Let's pour one out for being able to run Minix binaries on Linux. :-)
https://en.wikipedia.org/wiki/Minix_3:
"2013-02-21 ... Dropped support for a.out binaries"
They themselves killed that support a lot earlier. :-)
Quoted 2 lines
Speaking of which, if we're doing this, are there some old system calls that we could remove at the same time?
Good question.
Unfortunately, I'm just the janitor removing this and a.out is waaay before my time. :-)
But I'm open to suggestions on how to determine which are those obsolete syscalls.
Btw, pls do note that this is removing only the x86 a.out support - other arches (m68k and alpha) still select HAVE_AOUT and thus the respective glue in fs/binfmt_aout.c
Signature · 4 lines
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquetteQuoted 8 lines · Borislav Petkov
Quoted 1 line · Theodore Ts'o
Let's pour one out for being able to run Minix binaries on Linux. :-)
https://en.wikipedia.org/wiki/Minix_3:
"2013-02-21 ... Dropped support for a.out binaries"
They themselves killed that support a lot earlier. :-)
Well, Minix 3 is a very different beast than the Minix 1.x that Linus used when he was boot-strapping Linux back in 1991. :-)
Among other things Minux 1.x is system call compatible with the original V7 Unix from AT&T, whereas Minux 2.x and higher switch to having system calls that were Posix comaptible --- and Minix 2, being at teaching OS, wouldn't have kept any backwards compatibility when Prof. Tannenbaum released the 2nd edition of his Operating Systems Design and Implementation book in 1997.
Quoted 10 lines
Quoted 2 lines
Speaking of which, if we're doing this, are there some old system calls that we could remove at the same time?
Good question.
Unfortunately, I'm just the janitor removing this and a.out is waaay before my time. :-)
But I'm open to suggestions on how to determine which are those obsolete syscalls.
Historically there were a couple of different C libraries which x86 binaries for Linux could use:
A) minix's libc -- used by binaries compiled using the Minix 1 toolchain
B) Libc 4 -- a fork of GNU Libc version 1, for a.out execuables
C) Libc 5 -- a fork of GNU Libc version 1, for ELF execuables
D) Libc 6 -- GNU Libc version 2, for ELF executables
So I think what this would boil down to is finding those system calls that were used by (A) and (B), but not referenced in (C) and (D). I guess we'd also have to check to make sure that some of these ancient system calls might still be used by some of the micro libc's, such as dietlibc, musl, etc.
Is it worth doing? Perhaps; perhaps not. It probably wouldn't reduce the size of the kernel by all *that* much, but getting rid of cruft is kind of a good thing for its own sake.
Cheers.
- Ted
Arnd Bergmann
maintainerQuoted 19 lines · Theodore Ts'o
Quoted 3 lines · Borislav Petkov
But I'm open to suggestions on how to determine which are those obsolete syscalls.
Historically there were a couple of different C libraries which x86 binaries for Linux could use:
A) minix's libc -- used by binaries compiled using the Minix 1 toolchain
B) Libc 4 -- a fork of GNU Libc version 1, for a.out execuables
C) Libc 5 -- a fork of GNU Libc version 1, for ELF execuables
D) Libc 6 -- GNU Libc version 2, for ELF executablesSo I think what this would boil down to is finding those system calls that were used by (A) and (B), but not referenced in (C) and (D). I guess we'd also have to check to make sure that some of these ancient system calls might still be used by some of the micro libc's, such as dietlibc, musl, etc.
Quoted 3 lines
Is it worth doing? Perhaps; perhaps not. It probably wouldn't reduce the size of the kernel by all *that* much, but getting rid of cruft is kind of a good thing for its own sake.
For reference, these are the system calls that are implemented on i386 as of 5.16 but have been replaced on modern architectures, from ooking only in the linux sources:
2 i386 fork sys_fork
5 i386 open sys_open compat_sys_open
7 i386 waitpid sys_waitpid
8 i386 creat sys_creat
9 i386 link sys_link
10 i386 unlink sys_unlink
13 i386 time sys_time32
14 i386 mknod sys_mknod
15 i386 chmod sys_chmod
16 i386 lchown sys_lchown16
16 i386 lchown sys_lchown16
18 i386 oldstat sys_stat
22 i386 umount sys_oldumount
23 i386 setuid sys_setuid16
24 i386 getuid sys_getuid16
25 i386 stime sys_stime32
27 i386 alarm sys_alarm
28 i386 oldfstat sys_fstat
29 i386 pause sys_pause
30 i386 utime sys_utime32
33 i386 access sys_access
34 i386 nice sys_nice
38 i386 rename sys_rename
39 i386 mkdir sys_mkdir
40 i386 rmdir sys_rmdir
42 i386 pipe sys_pipe
46 i386 setgid sys_setgid16
47 i386 getgid sys_getgid16
48 i386 signal sys_signal
49 i386 geteuid sys_geteuid16
50 i386 getegid sys_getegid16
59 i386 oldolduname sys_olduname
62 i386 ustat sys_ustat compat_sys_ustat
63 i386 dup2 sys_dup2
65 i386 getpgrp sys_getpgrp
67 i386 sigaction sys_sigaction compat_sys_sigaction 68 i386 sgetmask sys_sgetmask
69 i386 ssetmask sys_ssetmask
70 i386 setreuid sys_setreuid16
71 i386 setregid sys_setregid16
72 i386 sigsuspend sys_sigsuspend
73 i386 sigpending sys_sigpending compat_sys_sigpending 80 i386 getgroups sys_getgroups16
81 i386 setgroups sys_setgroups16
82 i386 select sys_old_select compat_sys_old_select 83 i386 symlink sys_symlink
84 i386 oldlstat sys_lstat
85 i386 readlink sys_readlink
86 i386 uselib sys_uselib
89 i386 readdir sys_old_readdir compat_sys_old_readdir 95 i386 fchown sys_fchown16
101 i386 ioperm sys_ioperm
102 i386 socketcall sys_socketcall compat_sys_socketcall 109 i386 olduname sys_uname
110 i386 iopl sys_iopl
117 i386 ipc sys_ipc compat_sys_ipc
119 i386 sigreturn sys_sigreturn compat_sys_sigreturn 123 i386 modify_ldt sys_modify_ldt
126 i386 sigprocmask sys_sigprocmask compat_sys_sigprocmask 135 i386 sysfs sys_sysfs
138 i386 setfsuid sys_setfsuid16
139 i386 setfsgid sys_setfsgid16
141 i386 getdents sys_getdents compat_sys_getdents 142 i386 _newselect sys_select compat_sys_select 164 i386 setresuid sys_setresuid16
165 i386 getresuid sys_getresuid16
168 i386 poll sys_poll
170 i386 setresgid sys_setresgid16
171 i386 getresgid sys_getresgid
182 i386 chown sys_chown16
190 i386 vfork sys_vfork
243 i386 set_thread_area sys_set_thread_area
244 i386 get_thread_area sys_get_thread_area
254 i386 epoll_create sys_epoll_create
256 i386 epoll_wait sys_epoll_wait
271 i386 utimes sys_utimes_time32
291 i386 inotify_init sys_inotify_init
299 i386 futimesat sys_futimesat_time32
321 i386 signalfd sys_signalfd compat_sys_signalfd 323 i386 eventfd sys_eventfd
384 i386 arch_prctl sys_arch_prctl compat_sys_arch_prctl
Looking at libc-4.7.2 from Slackware-3.0, the highest syscall number used there is 182 (chown16), anything later is presumably added after the ELF transition was done. All of the syscalls referenced in libc-4.7 are still referenced in libc-5.0, so nothing is obviously changed between those.
Excluding the syscalls that are referenced in libc5 brings the candidates down to
7 i386 waitpid sys_waitpid 16 i386 lchown sys_lchown16 18 i386 oldstat sys_stat 28 i386 oldfstat sys_fstat 48 i386 signal sys_signal 59 i386 oldolduname sys_olduname 68 i386 sgetmask sys_sgetmask 69 i386 ssetmask sys_ssetmask 84 i386 oldlstat sys_lstat 109 i386 olduname sys_uname 119 i386 sigreturn sys_sigreturn 123 i386 modify_ldt sys_modify_ldt 135 i386 sysfs sys_sysfs 141 i386 getdents sys_getdents 142 i386 _newselect sys_select 164 i386 setresuid sys_setresuid16 165 i386 getresuid sys_getresuid16 168 i386 poll sys_poll 170 i386 setresgid sys_setresgid16 171 i386 getresgid sys_getresgid16
This is probably a mix of some things that are too old for libc5, some that are too new, and some that were never part of libc but are called from other places.
Some may be misrepresented here because the macros got
renamed: libc5 calls 'uname' but expects it to be what the kernel now calls '__NR_olduname'. I could not find out which of the 'stat' family it expects because the headers are incomplete in the slackware source package.
Arnd
Borislav Petkov
maintainerJust to answer the "do we still wanna support running original a.out binaries" aspect...
Quoted 9 lines · Theodore Ts'o
Quoted 7 lines
Quoted 1 line
Let's pour one out for being able to run Minix binaries on Linux. :-)
https://en.wikipedia.org/wiki/Minix_3:
"2013-02-21 ... Dropped support for a.out binaries"
They themselves killed that support a lot earlier. :-)
Well, Minix 3 is a very different beast than the Minix 1.x
Right, but if the Minix 1.x successor has deemed a.out support obsolete and has removed in 2013 I figure we can too. :-)
And, btw, Alan had a good, practical idea at the time we talked about deprecating a.out support:
https://lore.kernel.org/all/20190305134347.4be2449c@alans-desktop/
And that is probably the best thing to do: if people wanna run old a.out binaries, they can either write an a.out loader as an ELF program or slap an old linux distro in a VM and do that there just fine.
If you look at the original thread which started this:
https://lore.kernel.org/all/CAG48ez1RVd5mQ_Pb6eygQESaZhpQz765OAZYSoPE0kPqfZEXQg@mail.gmail.com/
it looked even back then like a.out support is so rusty, bitrotten and broken that we're probably even doing ourselves a favor to remove it.
Especially if not even the toolchains would even create an a.out executable and no one even noticed.
So...
Quoted 8 lines
that Linus used when he was boot-strapping Linux back in 1991. :-)
Among other things Minux 1.x is system call compatible with the original V7 Unix from AT&T, whereas Minux 2.x and higher switch to having system calls that were Posix comaptible --- and Minix 2, being at teaching OS, wouldn't have kept any backwards compatibility when Prof. Tannenbaum released the 2nd edition of his Operating Systems Design and Implementation book in 1997.
I had a lot of fun reading that book. :-)
Signature · 4 lines
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquetteFrom: Borislav Petkov
Quoted 1 line
Sent: 13 January 2022 22:57
...
Quoted 4 lines
Btw, pls do note that this is removing only the x86 a.out support - other arches (m68k and alpha) still select
HAVE_AOUTand thus the respective glue infs/binfmt_aout.c
I didn't think alpha was that old :-)
m68k is another matter.
But I've probably dismembered the only old m68k system disk I've had to extract the magnets.
David
-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK Registration No: 1397386 (Wales)
Geert Uytterhoeven
maintainerHi Boris,
Quoted 3 lines · Borislav Petkov
Btw, pls do note that this is removing only the x86 a.out support - other arches (m68k and alpha) still select
HAVE_AOUTand thus the respective glue infs/binfmt_aout.c
Quoting myself (do I need a canned response for this? ;-):
"I think it's safe to assume no one still runs a.out binaries on m68k."
https://lore.kernel.org/r/CAMuHMdVGT1QN8WUqNcU8ihPLncUGfrjV49wb-8nHUgHhOzLeNw@mail.gmail.com/
https://lore.kernel.org/r/CAMuHMdW+m0Q+j3rsQdMXnrEPm+XB5Y2AQrxW5sD1mZAKgmEqoA@mail.gmail.com
Of course, m68k is not encumbered by a.out Netscape binaries, unlike ia32 and alpha ;-)
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Quoted 7 lines · Geert Uytterhoeven
Hi Boris,
Quoted 3 lines · Borislav Petkov
Btw, pls do note that this is removing only the x86 a.out support - other arches (m68k and alpha) still select
HAVE_AOUTand thus the respective glue infs/binfmt_aout.c
Quoted 8 lines
Quoting myself (do I need a canned response for this? ;-):
"I think it's safe to assume no one still runs a.out binaries on m68k."https://lore.kernel.org/r/CAMuHMdVGT1QN8WUqNcU8ihPLncUGfrjV49wb-8nHUgHhOzLeNw@mail.gmail.com/
https://lore.kernel.org/r/CAMuHMdW+m0Q+j3rsQdMXnrEPm+XB5Y2AQrxW5sD1mZAKgmEqoA@mail.gmail.comOf course, m68k is not encumbered by a.out Netscape binaries, unlike ia32 and alpha ;-)
And Alpha's a.out Netscape binary is what required OSF/1 compatible system calls, so if and when we ever can nuke a.out Alpha support, so perhaps there might also be an opporunity to remove OSF/1 compatibility support...
- Ted
Geert Uytterhoeven
maintainerHi Ted,
Quoted 12 lines · Theodore Ts'o
Quoted 11 lines · Geert Uytterhoeven
Quoted 3 lines · Borislav Petkov
Btw, pls do note that this is removing only the x86 a.out support - other arches (m68k and alpha) still select
HAVE_AOUTand thus the respective glue infs/binfmt_aout.cQuoting myself (do I need a canned response for this? ;-):
"I think it's safe to assume no one still runs a.out binaries on m68k."https://lore.kernel.org/r/CAMuHMdVGT1QN8WUqNcU8ihPLncUGfrjV49wb-8nHUgHhOzLeNw@mail.gmail.com/
https://lore.kernel.org/r/CAMuHMdW+m0Q+j3rsQdMXnrEPm+XB5Y2AQrxW5sD1mZAKgmEqoA@mail.gmail.com
Quoted 7 lines
Quoted 2 lines
Of course, m68k is not encumbered by a.out Netscape binaries, unlike ia32 and alpha ;-)
And Alpha's a.out Netscape binary is what required OSF/1 compatible system calls, so if and when we ever can nuke a.out Alpha support, so perhaps there might also be an opporunity to remove OSF/1 compatibility support...
As we seem to be reiterating a discussion from 2019, let's read last time's summary first:
https://www.linuxjournal.com/content/deprecating-aout-binaries
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Borislav Petkov
maintainerQuoted 2 lines · Geert Uytterhoeven
Quoting myself (do I need a canned response for this? ;-):
"I think it's safe to assume no one still runs a.out binaries on m68k."
I remember now that I've looked at the thread from back then but what are you trying to say with your quote?
Do you wanna see another patch which rips out a.out support from m68k?
I could do one ontop but besides build-testing I can't test any other way on m68k...
:-)
Signature · 4 lines
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquetteGeert Uytterhoeven
maintainerHi Boris,
Quoted 7 lines · Borislav Petkov
Quoted 2 lines · Geert Uytterhoeven
Quoting myself (do I need a canned response for this? ;-):
"I think it's safe to assume no one still runs a.out binaries on m68k."I remember now that I've looked at the thread from back then but what are you trying to say with your quote?
Quoted 6 lines
Do you wanna see another patch which rips out a.out support from m68k?
I could do one ontop but besides build-testing I can't test any other way on m68k...
:-)
I mean that I can rip it out on m68k.
Who's gonna take care of alpha?
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Borislav Petkov
maintainerQuoted 3 lines · Geert Uytterhoeven
I mean that I can rip it out on m68k.
Who's gonna take care of alpha?
I'm sceptical anyone would:
$ ./scripts/get_maintainer.pl -f arch/alpha/ Richard Henderson <[email protected]> (odd fixer:ALPHA PORT) Ivan Kokshaysky <[email protected]> (odd fixer:ALPHA PORT)
which leaves me. I could give it a try.
Signature · 4 lines
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquetteBorislav Petkov
maintainerQuoted 5 lines · Borislav Petkov
Quoted 3 lines · Geert Uytterhoeven
I mean that I can rip it out on m68k.
Who's gonna take care of alpha?
Quoted 7 lines
I'm sceptical anyone would:
$ ./scripts/get_maintainer.pl -f arch/alpha/ Richard Henderson <[email protected]> (odd fixer:ALPHA PORT) Ivan Kokshaysky <[email protected]> (odd fixer:ALPHA PORT)which leaves me. I could give it a try.
Well, on that old thread from 2019 people expressed the wish to be able to run a.out binaries on Alpha. Let's see whether that has changed in the meantime.
/me CCs them.
Signature · 4 lines
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquetteHi Borislav!
Quoted 12 lines · Borislav Petkov
Quoted 11 lines · Borislav Petkov
Quoted 3 lines · Geert Uytterhoeven
I mean that I can rip it out on m68k.
Who's gonna take care of alpha?
I'm sceptical anyone would:
$ ./scripts/get_maintainer.pl -f arch/alpha/ Richard Henderson <[email protected]> (odd fixer:ALPHA PORT) Ivan Kokshaysky <[email protected]> (odd fixer:ALPHA PORT)
Quoted 5 lines
Quoted 1 line
which leaves me. I could give it a try.
Well, on that old thread from 2019 people expressed the wish to be able to run a.out binaries on Alpha. Let's see whether that has changed in the meantime.
Checking on the message I sent three years ago, it didn't seem that I was opposed to removing a.out support, but that I would be happy to help test the patches.
I don't have any particular use case for a.out support either.
Adrian
Signature · 4 lines
.''`. John Paul Adrian Glaubitz : :' : Debian Developer - [email protected] `. `' Freie Universitaet Berlin - [email protected] `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
PATCH 1/1 a.out: Stop building a.out/osf1 support on alpha and m68k
There has been repeated discussion on removing a.out support, the most recent was[1]. Having read through a bunch of the discussion it looks like no one has see any reason why we need to keep a.out support.
The m68k maintainer has even come out in favor of removing a.out support[2].
At a practical level with only two rarely used architectures building a.out support, it gets increasingly hard to test and to care about. Which means the code will almost certainly bit-rot.
Let's see if anyone cares about a.out support on the last two architectures that build it, by disabling the build of the support in Kconfig. If anyone cares, this can be easily reverted, and we can then have a discussion about what it is going to take to support a.out binaries in the long term.
Cc: Richard Henderson <[email protected]>
Cc: Ivan Kokshaysky <[email protected]>
Cc: Matt Turner <[email protected]>
Cc: [email protected]
Cc: Geert Uytterhoeven <[email protected]>
Cc: [email protected]
[1] https://lkml.kernel.org/r/[email protected]
[2] https://lkml.kernel.org/r/CAMuHMdUbTNNr16YY1TFe=-uRLjg6yGzgw_RqtAFpyhnOMM5Pvw@mail.gmail.com
---
Kees do you think you could add this to your execve branch?
I think this should be a complimentary patch to Borislav Petkov's patch at the top of this tread to remove a.out support on x86.
arch/alpha/Kconfig | 1 - arch/m68k/Kconfig | 1 - 2 files changed, 2 deletions(-)
arch/alpha/Kconfig +0 -1
@@ -12,7 +12,6 @@ config ALPHA select FORCE_PCI if !ALPHA_JENSEN select PCI_DOMAINS if PCI select PCI_SYSCALL if PCI - select HAVE_AOUT select HAVE_ASM_MODVERSIONS select HAVE_PCSPKR_PLATFORM select HAVE_PERF_EVENTS
arch/m68k/Kconfig +0 -1
@@ -17,7 +17,6 @@ config M68K select GENERIC_CPU_DEVICES select GENERIC_IOMAP select GENERIC_IRQ_SHOW - select HAVE_AOUT if MMU select HAVE_ASM_MODVERSIONS select HAVE_DEBUG_BUGVERBOSE select HAVE_EFFICIENT_UNALIGNED_ACCESS if !CPU_HAS_NO_UNALIGNED
Signature · 1 line
2.29.2
Re: [PATCH] a.out: Stop building a.out/osf1 support on alpha and m68k
Quoted 4 lines · Eric W. Biederman
Let's see if anyone cares about a.out support on the last two architectures that build it, by disabling the build of the support in
Kconfig.
Oh, I'm pretty sure we can't do this.
a.out on alpha is afaik still very much alive - well, as alive as anything alpha is - although it's called "ECOFF".
It's the native Tru64 (aka "DEC OSF/1", aka "Digital UNIX") format, so it's more than some old legacy Linux thing.
We still call it "a.out", but the history is that a.out with some extensions became COFF, which then became ECOFF, so our a.out code really covers the whole gamut.
Yeah, we don't actually parse any of the extensions that make COFF what it is, or ECOFF what _it_ is, but the a.out loader ends up working "well enough" for simple binaries by using ugly code like
#define N_TXTOFF(x) \
((long) N_MAGIC(x) == ZMAGIC ? 0 : \
(sizeof(struct exec) + (x).fh.f_nscns*SCNHSZ + SCNROUND - 1) &
~(SCNROUND - 1))
which jumps over all the section headers.
But sure, it would be interesting to know if any alpha people care - I just have this suspicion that we can't drop it that easily because of the non-Linux legacy.
Linus
Re: [PATCH] a.out: Stop building a.out/osf1 support on alpha and m68k
Quoted 9 lines · Linus Torvalds
Quoted 4 lines · Eric W. Biederman
Let's see if anyone cares about a.out support on the last two architectures that build it, by disabling the build of the support in
Kconfig.[...]
But sure, it would be interesting to know if any alpha people care - I just have this suspicion that we can't drop it that easily because of the non-Linux legacy.
It looks like the only distro supporting Alpha is Gentoo. I pulled down the installation media, and everything is ELF except for firmware COFF files:
$ find . -type f | xargs file | grep -Ei ':.*(out|coff)' ./lib/firmware/meson/vdec/g12a_hevc_mmu.bin: MIPSEB MIPS-III ECOFF executable not stripped - version 0.0 ./lib/firmware/meson/vdec/g12a_vp9.bin: MIPSEB MIPS-III ECOFF executable not stripped - version 0.0 ./lib/firmware/meson/vdec/gxl_hevc_mmu.bin: MIPSEB MIPS-III ECOFF executable not stripped - version 0.0 ./lib/firmware/meson/vdec/gxl_vp9.bin: MIPSEB MIPS-III ECOFF executable not stripped - version 0.0 ./lib/firmware/meson/vdec/sm1_hevc_mmu.bin: MIPSEB MIPS-III ECOFF executable not stripped - version 0.0 ./lib/firmware/meson/vdec/sm1_vp9_mmu.bin: MIPSEB MIPS-III ECOFF executable not stripped - version 0.0 ./lib/firmware/qca/crbtfw32.tlv: mc68k COFF object not stripped
So, since it's an easy revert, sure. Let's do it.
Signature · 1 line
Kees Cook
Kees Cook
AppliedRe: [PATCH] a.out: Stop building a.out/osf1 support on alpha and m68k
In reply to Eric W. BiedermanQuoted 8 lines · Eric W. Biederman
There has been repeated discussion on removing a.out support, the most recent was[1]. Having read through a bunch of the discussion it looks like no one has see any reason why we need to keep a.out support.
The m68k maintainer has even come out in favor of removing a.out support[2].
[...]
Applied to for-next/execve, thanks!
[1/1] a.out: Stop building a.out/osf1 support on alpha and m68k
https://git.kernel.org/kees/c/f451569b9da8
Signature · 1 line
Kees Cook
PATCH 1/1 x86: Remove a.out support
From: Borislav Petkov <[email protected]>
Date: Thu, 13 Jan 2022 17:01:15 +0100
Subject: [PATCH] x86: Remove a.out support
Commit
eac616557050 ("x86: Deprecate a.out support")
deprecated a.out support with the promise to remove it a couple of releases later. That commit landed in v5.1.
Now it is more than a couple of releases later, no one has complained so remove it.
---
Kees can you pick this one up in for-next/execve as well?
It still applies cleanly and the actual patch seems to have gotten lost in the conversation about what more we could do.
Thanks,
Eric
arch/x86/Kconfig | 7 - arch/x86/ia32/Makefile | 2 - arch/x86/ia32/ia32_aout.c | 325 -------------------------------------- 3 files changed, 334 deletions(-) delete mode 100644 arch/x86/ia32/ia32_aout.c
arch/x86/Kconfig +0 -7
@@ -2827,13 +2827,6 @@ config IA32_EMULATION 64-bit kernel. You should likely turn this on, unless you're 100% sure that you don't have any 32-bit programs left. -config IA32_AOUT - tristate "IA32 a.out support" - depends on IA32_EMULATION - depends on BROKEN - help - Support old a.out binaries in the 32bit emulation. - config X86_X32 bool "x32 ABI for 64-bit mode" depends on X86_64
arch/x86/ia32/Makefile +0 -2
@@ -5,7 +5,5 @@ obj-$(CONFIG_IA32_EMULATION) := ia32_signal.o -obj-$(CONFIG_IA32_AOUT) += ia32_aout.o - audit-class-$(CONFIG_AUDIT) := audit.o obj-$(CONFIG_IA32_EMULATION) += $(audit-class-y)
arch/x86/ia32/ia32_aout.c deleted +0 -325
325 lines not on this page · raw message
Signature · 1 line
2.29.2
Quoted 5 lines · Eric W. Biederman
Kees can you pick this one up in for-next/execve as well?
It still applies cleanly and the actual patch seems to have gotten lost in the conversation about what more we could do.
Side note: there are similar other turds if a.out goes away, ie on alpha it's OSF4_COMPAT, and it enables support for a couple of legacy OSF/1 system calls.
I think that was also discussed in the (old) a.out deprecation thread back in 2019..
Linus
Arnd Bergmann
maintainerQuoted 8 lines · Linus Torvalds
Quoted 5 lines · Eric W. Biederman
Kees can you pick this one up in for-next/execve as well?
It still applies cleanly and the actual patch seems to have gotten lost in the conversation about what more we could do.
Quoted 6 lines
Side note: there are similar other turds if a.out goes away, ie on alpha it's
OSF4_COMPAT, and it enables support for a couple of legacy OSF/1 system calls.I think that was also discussed in the (old) a.out deprecation thread back in 2019..
The only thing that actually depends on CONFIG_OSF4_COMPAT seems to be the custom logic in readv() and writev(). Those are the two that you removed in your patch back then[1].
For the other system calls, I fear we can only try to guess which of them are used in Linux applications and which ones are not. These were always available to normal Linux user space, so the ones that are similar to our syscalls could have been used in glibc, e.g. mmap, wait4, stat.
Arnd
[1] https://lore.kernel.org/all/CAHk-=wgt7M6yA5BJCJo0nF22WgPJnN8CvViL9CAJmd+S+Civ6w@mail.gmail.com/
[2] $ git grep sys_osf arch/alpha/kernel/syscalls/syscall.tbl 7 common osf_wait4 sys_osf_wait4
17 common brk sys_osf_brk
21 common osf_mount sys_osf_mount
43 common osf_set_program_attributes sys_osf_set_program_attributes 48 common osf_sigprocmask sys_osf_sigprocmask
71 common mmap sys_osf_mmap
93 common osf_select sys_osf_select
100 common getpriority sys_osf_getpriority
112 common osf_sigstack sys_osf_sigstack
116 common osf_gettimeofday sys_osf_gettimeofday
117 common osf_getrusage sys_osf_getrusage
120 common readv sys_osf_readv
121 common writev sys_osf_writev
122 common osf_settimeofday sys_osf_settimeofday
138 common osf_utimes sys_osf_utimes
156 common sigaction sys_osf_sigaction
159 common osf_getdirentries sys_osf_getdirentries
160 common osf_statfs sys_osf_statfs
161 common osf_fstatfs sys_osf_fstatfs
165 common osf_getdomainname sys_osf_getdomainname
207 common osf_utsname sys_osf_utsname
224 common osf_stat sys_osf_stat
225 common osf_lstat sys_osf_lstat
226 common osf_fstat sys_osf_fstat
227 common osf_statfs64 sys_osf_statfs64
228 common osf_fstatfs64 sys_osf_fstatfs64
241 common osf_sysinfo sys_osf_sysinfo
244 common osf_proplist_syscall sys_osf_proplist_syscall 251 common osf_usleep_thread sys_osf_usleep_thread
256 common osf_getsysinfo sys_osf_getsysinfo
257 common osf_setsysinfo sys_osf_setsysinfo
Borislav Petkov
maintainerQuoted 1 line · Eric W. Biederman
Kees can you pick this one up in for-next/execve as well?
I can carry it through tip, no probs.
Signature · 4 lines
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquetteRe: [PATCH] a.out: Stop building a.out/osf1 support on alpha and m68k
In reply to Eric W. BiedermanQuoted 25 lines · Eric W. Biederman
There has been repeated discussion on removing a.out support, the most recent was[1]. Having read through a bunch of the discussion it looks like no one has see any reason why we need to keep a.out support.
The m68k maintainer has even come out in favor of removing a.out support[2].
At a practical level with only two rarely used architectures building a.out support, it gets increasingly hard to test and to care about. Which means the code will almost certainly bit-rot.
Let's see if anyone cares about a.out support on the last two architectures that build it, by disabling the build of the support in
Kconfig. If anyone cares, this can be easily reverted, and we can then have a discussion about what it is going to take to support a.out binaries in the long term.Cc: Richard Henderson <[email protected]>
Cc: Ivan Kokshaysky <[email protected]>
Cc: Matt Turner <[email protected]>
Cc: [email protected]
Cc: Geert Uytterhoeven <[email protected]>
Cc: [email protected]
[1] https://lkml.kernel.org/r/[email protected]
[2] https://lkml.kernel.org/r/CAMuHMdUbTNNr16YY1TFe=-uRLjg6yGzgw_RqtAFpyhnOMM5Pvw@mail.gmail.com
Quoted 3 lines
Signed-off-by "Eric W. Biederman" <[email protected]>arch/m68k/Kconfig | 1 -
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
From: Geert Uytterhoeven
Quoted 1 line
Sent: 14 January 2022 19:22
...
Quoted 1 line
Who's gonna take care of alpha?
Would anyone notice if the knife slipped?
David
-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK Registration No: 1397386 (Wales)
The following commit has been merged into the x86/cleanups branch of tip:
The following commit has been merged into the x86/cleanups branch of tip:
Commit-ID: f9444ea5e20847d1dd4a98d4dff4cc97655834bb Gitweb: https://git.kernel.org/tip/f9444ea5e20847d1dd4a98d4dff4cc97655834bb Author: Borislav Petkov <[email protected]> AuthorDate: Thu, 13 Jan 2022 14:26:10 +01:00 Committer: Borislav Petkov <[email protected]> CommitterDate: Fri, 11 Mar 2022 11:06:32 +01:00
x86: Remove a.out support
Commit
eac616557050 ("x86: Deprecate a.out support")
deprecated a.out support with the promise to remove it a couple of releases later. That commit landed in v5.1.
Now it is more than a couple of releases later, no one has complained so remove it.
--- arch/x86/Kconfig | 7 +- arch/x86/ia32/Makefile | 2 +- arch/x86/ia32/ia32_aout.c | 325 +------------------------------------- 3 files changed, 334 deletions(-) delete mode 100644 arch/x86/ia32/ia32_aout.c
arch/x86/Kconfig +0 -7
@@ -2827,13 +2827,6 @@ config IA32_EMULATION 64-bit kernel. You should likely turn this on, unless you're 100% sure that you don't have any 32-bit programs left. -config IA32_AOUT - tristate "IA32 a.out support" - depends on IA32_EMULATION - depends on BROKEN - help - Support old a.out binaries in the 32bit emulation. - config X86_X32 bool "x32 ABI for 64-bit mode" depends on X86_64
arch/x86/ia32/Makefile +0 -2
@@ -5,7 +5,5 @@ obj-$(CONFIG_IA32_EMULATION) := ia32_signal.o -obj-$(CONFIG_IA32_AOUT) += ia32_aout.o - audit-class-$(CONFIG_AUDIT) := audit.o obj-$(CONFIG_IA32_EMULATION) += $(audit-class-y)
arch/x86/ia32/ia32_aout.c deleted +0 -325
325 lines not on this page · raw message
The following commit has been merged into the x86/cleanups branch of tip:
The following commit has been merged into the x86/cleanups branch of tip:
Commit-ID: c7bda0dca98cca351a9bc852b3df8b9b99ffd400 Gitweb: https://git.kernel.org/tip/c7bda0dca98cca351a9bc852b3df8b9b99ffd400 Author: Borislav Petkov <[email protected]> AuthorDate: Thu, 13 Jan 2022 14:26:10 +01:00 Committer: Borislav Petkov <[email protected]> CommitterDate: Mon, 11 Apr 2022 18:04:27 +02:00
x86: Remove a.out support
Commit
eac616557050 ("x86: Deprecate a.out support")
deprecated a.out support with the promise to remove it a couple of releases later. That commit landed in v5.1.
Now it is more than a couple of releases later, no one has complained so remove it.
Fold in a hunk removing the reference to arch/x86/ia32/ia32_aout.c in MAINTAINERS:
https://lore.kernel.org/r/[email protected]
--- MAINTAINERS | 1 +- arch/x86/Kconfig | 7 +- arch/x86/ia32/Makefile | 2 +- arch/x86/ia32/ia32_aout.c | 325 +------------------------------------- 4 files changed, 335 deletions(-) delete mode 100644 arch/x86/ia32/ia32_aout.c
MAINTAINERS +0 -1
@@ -7378,7 +7378,6 @@ L: [email protected] S: Supported T: git git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux.git for-next/execve F: arch/alpha/kernel/binfmt_loader.c -F: arch/x86/ia32/ia32_aout.c F: fs/*binfmt_*.c F: fs/exec.c F: include/linux/binfmts.h
arch/x86/Kconfig +0 -7
@@ -2838,13 +2838,6 @@ config IA32_EMULATION 64-bit kernel. You should likely turn this on, unless you're 100% sure that you don't have any 32-bit programs left. -config IA32_AOUT - tristate "IA32 a.out support" - depends on IA32_EMULATION - depends on BROKEN - help - Support old a.out binaries in the 32bit emulation. - config X86_X32_ABI bool "x32 ABI for 64-bit mode" depends on X86_64
arch/x86/ia32/Makefile +0 -2
@@ -5,7 +5,5 @@ obj-$(CONFIG_IA32_EMULATION) := ia32_signal.o -obj-$(CONFIG_IA32_AOUT) += ia32_aout.o - audit-class-$(CONFIG_AUDIT) := audit.o obj-$(CONFIG_IA32_EMULATION) += $(audit-class-y)
arch/x86/ia32/ia32_aout.c deleted +0 -325
325 lines not on this page · raw message