UML - Add a .note.SuSE section
PATCH 1/1 UML - Add a .note.SuSE section
[ This is both 2.6.24 and -stable material ]
SuSE seems to require that binaries have a .note.SuSE section. Without it, UML segfaults if any parameters are passed on the command line.
--
arch/um/kernel/dyn.lds.S | 1 + 1 file changed, 1 insertion(+)
arch/um/kernel/dyn.lds.S +1 -0
@@ -10,6 +10,7 @@ SECTIONS PROVIDE (__executable_start = START); . = START + SIZEOF_HEADERS; .interp : { *(.interp) } + .note.SuSE : { *(.note.SuSE) } __binary_start = .; . = ALIGN(4096); /* Init code and data */ _text = .;
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
User-mode-linux-devel mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
Quoted 5 lines · Jeff Dike
[ This is both 2.6.24 and -stable material ]
SuSE seems to require that binaries have a .note.SuSE section. Without it, UML segfaults if any parameters are passed on the command line.
this sounds like something really stupid and bad... why would the kernel need to have a per-distro note section???
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
User-mode-linux-devel mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
Quoted 11 lines · Arjan van de Ven
Quoted 5 lines · Jeff Dike
[ This is both 2.6.24 and -stable material ]
SuSE seems to require that binaries have a .note.SuSE section. Without it, UML segfaults if any parameters are passed on the command line.
this sounds like something really stupid and bad... why would the kernel need to have a per-distro note section???
I agree, what did we mess up in the SuSE kernel to require such a hack? :)
thanks,
greg k-h
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
User-mode-linux-devel mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
Quoted 2 lines · Arjan van de Ven
this sounds like something really stupid and bad... why would the kernel need to have a per-distro note section???
Quoted 1 line · Chris Wright
Huh!? Why do we need a SuSE section?
Quoted 1 line · Greg KH
I agree, what did we mess up in the SuSE kernel to require such a hack?
Beats the crap out of me.
Drop this patch - it looks like it might just be papering over symptoms rather than fixing the real problem - see below.
What I do know is that current UML doesn't run when built on a SuSE host, the UML commit which caused it to break is c35e584c087381aaa5f1ed40a28b978535c18fb2
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=c35e584c087381aaa5f1ed40a28b978535c18fb2;hp=a5bd1786fb30abe663b904f6d79bba413e9ba883
and the difference between a working UML binary and a broken one is this:
+ 1 .note.ABI-tag 00000020 0000000060000254 0000000060000254 00000254 2**2
+ CONTENTS, ALLOC, LOAD, READONLY, DATA + 2 .note.SuSE 00000018 0000000060000274 0000000060000274 00000274 2**2 + CONTENTS, ALLOC, LOAD, READONLY, DATA
and the .note.SuSE section makes the difference.
Looking into it a bit further, the contents of the section are:
objdump --section=.note.SuSE -s uml8796-linux-good uml8796-linux-good: file format elf64-x86-64 Contents of section .note.SuSE: 60000274 05000000 04000000 53755345 53755345 ........SuSESuSE 60000284 00000000 01000a02 ........
which seems kind of pointless, but also harmless.
The crash is in this section:
__uml_setup_start = .;
.uml.setup.init : { *(.uml.setup.init) }
__uml_setup_end = .;with &__uml_setup_start being 8 bytes before the start of the first 16-byte structure in .uml.setup.init, so the structures are misaligned wrt the start symbol.
I don't see any connection between the presence of a section at the start of the binary and this misalignment, so the patch is probably wrong.
Jeff
Signature · 11 lines
Work email - jdike at linux dot intel dot com ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ User-mode-linux-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
Quoted 6 lines · Jeff Dike
The crash is in this section:
__uml_setup_start = .; .uml.setup.init : { *(.uml.setup.init) } __uml_setup_end = .;
This looks like a classic bug.
You wanted this:
.uml.setup.init : {
__uml_setup_start = .;
*(.uml.setup.init)
__uml_setup_end = .;
}In this way you are sure __uml_setup_start has same address as start of section.
With the original code the linker will align the .uml.setup.init section to the alignment required by the included sections and thus the label and the actual start address may differ.
Sam
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
User-mode-linux-devel mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
Quoted 8 lines · Sam Ravnborg
Quoted 6 lines · Jeff Dike
The crash is in this section:
__uml_setup_start = .; .uml.setup.init : { *(.uml.setup.init) } __uml_setup_end = .;
Quoted 7 lines
This looks like a classic bug.
You wanted this:.uml.setup.init : { __uml_setup_start = .; *(.uml.setup.init) __uml_setup_end = .; }
Ooh, this sounds promising, thanks.
Jeff
Signature · 11 lines
Work email - jdike at linux dot intel dot com ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ User-mode-linux-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
Quoted 16 lines · Jeff Dike
Quoted 7 lines · Sam Ravnborg
Quoted 5 lines · Jeff Dike
The crash is in this section:
__uml_setup_start = .; .uml.setup.init : { *(.uml.setup.init) } __uml_setup_end = .;Quoted 7 lines
This looks like a classic bug.
You wanted this:.uml.setup.init : { __uml_setup_start = .; *(.uml.setup.init) __uml_setup_end = .; }
Quoted 1 line
Ooh, this sounds promising, thanks.
It's not the first time we hit effects of such bugs, is it? The .note.ABI-tag fix, time ago, may be about the same problem. Can you double-check all UML linker scripts for more instances of this bug?
Bye
Signature · 4 lines
"Doh!" (cit.), I've made another another mistake! Paolo Giarrusso, aka Blaisorblade Linux registered user n. 292729 http://www.user-mode-linux.org/~blaisorblade
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
User-mode-linux-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
Quoted 1 line · Blaisorblade
It's not the first time we hit effects of such bugs, is it?
I don't remember seeing this before.
Quoted 1 line
The .note.ABI-tag fix, time ago, may be about the same problem.
Are you referring to
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=c35e584c087381aaa5f1ed40a28b978535c18fb2;hp=a5bd1786fb30abe663b904f6d79bba413e9ba883?
If so, I never understood that - it just came in saying "this fixes static building", so I sent it along. BTW, that commit was singled out by git-bisect as "causing" this particular problem.
Quoted 2 lines
Can you
double-check all UML linker scripts for more instances of this bug?
I did, I have a patch, and it's been verified to fix the problem.
Jeff
Signature · 11 lines
Work email - jdike at linux dot intel dot com ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ User-mode-linux-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
Quoted 11 lines · Jeff Dike
Quoted 1 line · Blaisorblade
It's not the first time we hit effects of such bugs, is it?
I don't remember seeing this before.
Quoted 1 line
The .note.ABI-tag fix, time ago, may be about the same problem.
Are you referring to
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdi
ff;h=c35e584c087381aaa5f1ed40a28b978535c18fb2;hp=a5bd1786fb30abe663b904f6d79bba413e9ba883?
Yes.
Quoted 2 lines
If so, I never understood that - it just came in saying "this fixes static building", so I sent it along.
In this case, I'm referring to the patch which had a typo, which is yours:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=7632fc8f809a97f9d82ce125e8e3e579390ce2e5
Description follows:
"During a static link, ld has started putting a .note section in the .uml.setup.init section. This has the result that the UML setups begin with 32 bytes of garbage and UML crashes immediately on boot.
This patch creates a specific .note section for ld to drop this stuff into."
My patch only made your change work for real - IIRC you had fixed that exact typo too, but you forgot to run quilt refresh before sending the patch (btw, quilt pop -a will force you to refresh all patches to succeed - I do it frequently).
Quoted 7 lines
BTW, that commit was singled
out by git-bisect as "causing" this particular problem.Quoted 2 lines
Can you
double-check all UML linker scripts for more instances of this bug?I did, I have a patch, and it's been verified to fix the problem.
In this case, we _may_ want to remove the .note section altogether - even if it is likely to shake out more problems.
Good bye!
Signature · 4 lines
"Doh!" (cit.), I've made another another mistake! Paolo Giarrusso, aka Blaisorblade Linux registered user n. 292729 http://www.user-mode-linux.org/~blaisorblade
Quoted 5 lines · * Jeff Dike ([email protected]) wrote:
[ This is both 2.6.24 and -stable material ]
SuSE seems to require that binaries have a .note.SuSE section. Without it, UML segfaults if any parameters are passed on the command line.
Huh!? Why do we need a SuSE section?
thanks,
-chris
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
User-mode-linux-devel mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
Quoted 5 lines · Jeff Dike
[ This is both 2.6.24 and -stable material ]
SuSE seems to require that binaries have a .note.SuSE section. Without it, UML segfaults if any parameters are passed on the command line.
This doesn't make any sense. You must have misanalyzed this.
-Andi
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
User-mode-linux-devel mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel