* proposed relayfs changes and systemtap
@ 2006-02-21 14:57 Tom Zanussi
2006-02-21 15:06 ` Tom Zanussi
0 siblings, 1 reply; 5+ messages in thread
From: Tom Zanussi @ 2006-02-21 14:57 UTC (permalink / raw)
To: systemtap
Hi,
Someone over the weekend posted a patch to remove the filesystem part
of relayfs and instead have applications' relayfs files live in other
filesytems instead:
http://marc.theaimsgroup.com/?l=linux-kernel&m=114036950922176&w=2
It looks like there's support for making this change (this idea has
come up multiple times before and I thought it made sense then too,
but it's never really gotten legs until now).
Anyway, in case anyone was worrying about how it would affect
systemtap, not much. The poster of the patches has concentrated on
using sysfs to house relay files, but since systemtap already uses
/proc, it would seem to make sense for systemtap to move its relay
files there instead.
So, assuming we want move things over to procfs, the changes that
would need to be made for systemtap would be:
1) runtime/transport/relayfs.c - add file creation/removal relayfs
callback handlers. These will create/remove the proc files to
represent the relayfs buffers.
2) runtime/transport/relayfs.c - change the relayfs_create_dir() and
relayfs_remove_dir() calls to their procfs counterparts.
3) runtime/stpd/stpd.c - change "/mnt/relay" to "/proc" in main()
So that's it - the changes would be pretty minimal, and of course I'll
be happy to make them.
Tom
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: proposed relayfs changes and systemtap
2006-02-21 14:57 proposed relayfs changes and systemtap Tom Zanussi
@ 2006-02-21 15:06 ` Tom Zanussi
2006-03-16 15:16 ` William Cohen
0 siblings, 1 reply; 5+ messages in thread
From: Tom Zanussi @ 2006-02-21 15:06 UTC (permalink / raw)
To: Tom Zanussi; +Cc: systemtap
Tom Zanussi writes:
> Hi,
>
> Someone over the weekend posted a patch to remove the filesystem part
> of relayfs and instead have applications' relayfs files live in other
> filesytems instead:
>
> http://marc.theaimsgroup.com/?l=linux-kernel&m=114036950922176&w=2
Actually, this is probably a better link to the patches and thread:
http://marc.theaimsgroup.com/?l=linux-kernel&m=114038358624583&w=2
Tom
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: proposed relayfs changes and systemtap
2006-02-21 15:06 ` Tom Zanussi
@ 2006-03-16 15:16 ` William Cohen
2006-03-16 15:22 ` Tom Zanussi
0 siblings, 1 reply; 5+ messages in thread
From: William Cohen @ 2006-03-16 15:16 UTC (permalink / raw)
To: Tom Zanussi; +Cc: systemtap
Tom Zanussi wrote:
> Tom Zanussi writes:
> > Hi,
> >
> > Someone over the weekend posted a patch to remove the filesystem part
> > of relayfs and instead have applications' relayfs files live in other
> > filesytems instead:
> >
> > http://marc.theaimsgroup.com/?l=linux-kernel&m=114036950922176&w=2
>
> Actually, this is probably a better link to the patches and thread:
>
> http://marc.theaimsgroup.com/?l=linux-kernel&m=114038358624583&w=2
>
> Tom
>
>
Hi Tom,
The nightly tests buildok and pass-5 tests failed due to the following
checkin:
2006-03-15 Tom Zanussi <zanussi@us.ibm.com>
* configure.ac: Add check for relayfs version.
* configure: Regenerated.
* stp_check.in: Remove bundled relayfs loading/compilation, add
check for relay (vs relayfs).
The parser tests won't find relayfs-config.h nor will the pass-5 tests
find the include file.
-Will
$ stap -vv
/home/wcohen/stap_testing_200603160930/src/testsuite/buildok/context_test.stp
Created temporary directory "/tmp/staphd43Lr"
Searched
'/home/wcohen/stap_testing_200603160930/src/tapset/2.6.15-1.2054_FC5smp/i686/*.stp',
match count 0
Searched
'/home/wcohen/stap_testing_200603160930/src/tapset/2.6.15-1.2054_FC5smp/*.stp',
match count 0
Searched
'/home/wcohen/stap_testing_200603160930/src/tapset/2.6.15/i686/*.stp',
match count 0
Searched
'/home/wcohen/stap_testing_200603160930/src/tapset/2.6.15/*.stp', match
count 0
Searched
'/home/wcohen/stap_testing_200603160930/src/tapset/2.6/i686/*.stp',
match count 0
Searched '/home/wcohen/stap_testing_200603160930/src/tapset/2.6/*.stp',
match count 0
Searched '/home/wcohen/stap_testing_200603160930/src/tapset/i686/*.stp',
match count 1
Searched '/home/wcohen/stap_testing_200603160930/src/tapset/*.stp',
match count 14
Pass 1: parsed user script and 15 library script(s) in
660usr/10sys/675real ms.
probe uptime_read_proc@fs/proc/proc_misc.c:101 pc=0xc0192ed6
probe uptime_read_proc@fs/proc/proc_misc.c:101 pc=0xc0192ec9
Pass 2: analyzed script: 4 probe(s), 18 function(s), 0 global(s) in
710usr/30sys/740real ms.
Probe #0 locks nothing
Probe #1 locks nothing
Probe #2 locks nothing
Probe #3 locks nothing
Running grep " [tT] " /proc/kallsyms | sort -k 1,8 -s -o
/tmp/staphd43Lr/symbols.sorted
Pass 3: translated to C into "/tmp/staphd43Lr/stap_2498.c" in
230usr/100sys/321real ms.
Running make -C "/lib/modules/2.6.15-1.2054_FC5smp/build"
M="/tmp/staphd43Lr" modules V=1
make: Entering directory `/usr/src/kernels/2.6.15-1.2054_FC5-smp-i686'
mkdir -p /tmp/staphd43Lr/.tmp_versions
make -f scripts/Makefile.build obj=/tmp/staphd43Lr
gcc -m32 -Wp,-MD,/tmp/staphd43Lr/.stap_2498.o.d -nostdinc -isystem
/usr/lib/gcc/i386-redhat-linux/4.1.0/include -D__KERNEL__ -Iinclude
-include include/linux/autoconf.h -Wall -Wundef -Wstrict-prototypes
-Wno-trigraphs -fno-strict-aliasing -fno-common -ffreestanding -Os
-fomit-frame-pointer -g -pipe -msoft-float -fno-builtin-sprintf
-fno-builtin-log2 -fno-builtin-puts -mpreferred-stack-boundary=2
-march=i686 -mtune=pentium4 -mregparm=3 -Iinclude/asm-i386/mach-generic
-Iinclude/asm-i386/mach-default -Wdeclaration-after-statement
-Wno-pointer-sign -Wno-unused -Werror -I
"/home/wcohen/stap_testing_200603160930/src/runtime" -I
"/home/wcohen/stap_testing_200603160930/src/runtime/relayfs" -DMODULE
-D"KBUILD_STR(s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(stap_2498)"
-D"KBUILD_MODNAME=KBUILD_STR(stap_2498)" -c -o
/tmp/staphd43Lr/.tmp_stap_2498.o /tmp/staphd43Lr/stap_2498.c
In file included from
/home/wcohen/stap_testing_200603160930/src/runtime/transport/transport.h:8,
from
/home/wcohen/stap_testing_200603160930/src/runtime/transport/transport.c:17,
from
/home/wcohen/stap_testing_200603160930/src/runtime/io.c:14,
from
/home/wcohen/stap_testing_200603160930/src/runtime/print.c:16,
from
/home/wcohen/stap_testing_200603160930/src/runtime/runtime.h:61,
from /tmp/staphd43Lr/stap_2498.c:29:
/home/wcohen/stap_testing_200603160930/src/runtime/transport/relayfs.h:8:28:
error: relayfs-config.h: No such file or directory
make[1]: *** [/tmp/staphd43Lr/stap_2498.o] Error 1
make: *** [_module_/tmp/staphd43Lr] Error 2
make: Leaving directory `/usr/src/kernels/2.6.15-1.2054_FC5-smp-i686'
Pass 4: compiled C into "stap_2498.ko" in 3230usr/340sys/3306real ms.
Pass 4: compilation failed. Try again with more '-v' (verbose) options.
Running rm -rf /tmp/staphd43Lr
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: proposed relayfs changes and systemtap
2006-03-16 15:16 ` William Cohen
@ 2006-03-16 15:22 ` Tom Zanussi
2006-03-16 15:39 ` William Cohen
0 siblings, 1 reply; 5+ messages in thread
From: Tom Zanussi @ 2006-03-16 15:22 UTC (permalink / raw)
To: William Cohen; +Cc: Tom Zanussi, systemtap
William Cohen writes:
> Tom Zanussi wrote:
> > Tom Zanussi writes:
> > > Hi,
> > >
> > > Someone over the weekend posted a patch to remove the filesystem part
> > > of relayfs and instead have applications' relayfs files live in other
> > > filesytems instead:
> > >
> > > http://marc.theaimsgroup.com/?l=linux-kernel&m=114036950922176&w=2
> >
> > Actually, this is probably a better link to the patches and thread:
> >
> > http://marc.theaimsgroup.com/?l=linux-kernel&m=114038358624583&w=2
> >
> > Tom
> >
> >
>
> Hi Tom,
>
> The nightly tests buildok and pass-5 tests failed due to the following
> checkin:
>
> 2006-03-15 Tom Zanussi <zanussi@us.ibm.com>
>
> * configure.ac: Add check for relayfs version.
> * configure: Regenerated.
> * stp_check.in: Remove bundled relayfs loading/compilation, add
> check for relay (vs relayfs).
>
> The parser tests won't find relayfs-config.h nor will the pass-5 tests
> find the include file.
Hi,
I just checked in new changes for that - can you get the latest from
cvs and try again? Sorry for the inconvenience.
Tom
>
> -Will
>
>
>
> $ stap -vv
> /home/wcohen/stap_testing_200603160930/src/testsuite/buildok/context_test.stp
> Created temporary directory "/tmp/staphd43Lr"
> Searched
> '/home/wcohen/stap_testing_200603160930/src/tapset/2.6.15-1.2054_FC5smp/i686/*.stp',
> match count 0
> Searched
> '/home/wcohen/stap_testing_200603160930/src/tapset/2.6.15-1.2054_FC5smp/*.stp',
> match count 0
> Searched
> '/home/wcohen/stap_testing_200603160930/src/tapset/2.6.15/i686/*.stp',
> match count 0
> Searched
> '/home/wcohen/stap_testing_200603160930/src/tapset/2.6.15/*.stp', match
> count 0
> Searched
> '/home/wcohen/stap_testing_200603160930/src/tapset/2.6/i686/*.stp',
> match count 0
> Searched '/home/wcohen/stap_testing_200603160930/src/tapset/2.6/*.stp',
> match count 0
> Searched '/home/wcohen/stap_testing_200603160930/src/tapset/i686/*.stp',
> match count 1
> Searched '/home/wcohen/stap_testing_200603160930/src/tapset/*.stp',
> match count 14
> Pass 1: parsed user script and 15 library script(s) in
> 660usr/10sys/675real ms.
> probe uptime_read_proc@fs/proc/proc_misc.c:101 pc=0xc0192ed6
> probe uptime_read_proc@fs/proc/proc_misc.c:101 pc=0xc0192ec9
> Pass 2: analyzed script: 4 probe(s), 18 function(s), 0 global(s) in
> 710usr/30sys/740real ms.
> Probe #0 locks nothing
> Probe #1 locks nothing
> Probe #2 locks nothing
> Probe #3 locks nothing
> Running grep " [tT] " /proc/kallsyms | sort -k 1,8 -s -o
> /tmp/staphd43Lr/symbols.sorted
> Pass 3: translated to C into "/tmp/staphd43Lr/stap_2498.c" in
> 230usr/100sys/321real ms.
> Running make -C "/lib/modules/2.6.15-1.2054_FC5smp/build"
> M="/tmp/staphd43Lr" modules V=1
> make: Entering directory `/usr/src/kernels/2.6.15-1.2054_FC5-smp-i686'
> mkdir -p /tmp/staphd43Lr/.tmp_versions
> make -f scripts/Makefile.build obj=/tmp/staphd43Lr
> gcc -m32 -Wp,-MD,/tmp/staphd43Lr/.stap_2498.o.d -nostdinc -isystem
> /usr/lib/gcc/i386-redhat-linux/4.1.0/include -D__KERNEL__ -Iinclude
> -include include/linux/autoconf.h -Wall -Wundef -Wstrict-prototypes
> -Wno-trigraphs -fno-strict-aliasing -fno-common -ffreestanding -Os
> -fomit-frame-pointer -g -pipe -msoft-float -fno-builtin-sprintf
> -fno-builtin-log2 -fno-builtin-puts -mpreferred-stack-boundary=2
> -march=i686 -mtune=pentium4 -mregparm=3 -Iinclude/asm-i386/mach-generic
> -Iinclude/asm-i386/mach-default -Wdeclaration-after-statement
> -Wno-pointer-sign -Wno-unused -Werror -I
> "/home/wcohen/stap_testing_200603160930/src/runtime" -I
> "/home/wcohen/stap_testing_200603160930/src/runtime/relayfs" -DMODULE
> -D"KBUILD_STR(s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(stap_2498)"
> -D"KBUILD_MODNAME=KBUILD_STR(stap_2498)" -c -o
> /tmp/staphd43Lr/.tmp_stap_2498.o /tmp/staphd43Lr/stap_2498.c
> In file included from
> /home/wcohen/stap_testing_200603160930/src/runtime/transport/transport.h:8,
> from
> /home/wcohen/stap_testing_200603160930/src/runtime/transport/transport.c:17,
> from
> /home/wcohen/stap_testing_200603160930/src/runtime/io.c:14,
> from
> /home/wcohen/stap_testing_200603160930/src/runtime/print.c:16,
> from
> /home/wcohen/stap_testing_200603160930/src/runtime/runtime.h:61,
> from /tmp/staphd43Lr/stap_2498.c:29:
> /home/wcohen/stap_testing_200603160930/src/runtime/transport/relayfs.h:8:28:
> error: relayfs-config.h: No such file or directory
> make[1]: *** [/tmp/staphd43Lr/stap_2498.o] Error 1
> make: *** [_module_/tmp/staphd43Lr] Error 2
> make: Leaving directory `/usr/src/kernels/2.6.15-1.2054_FC5-smp-i686'
> Pass 4: compiled C into "stap_2498.ko" in 3230usr/340sys/3306real ms.
> Pass 4: compilation failed. Try again with more '-v' (verbose) options.
> Running rm -rf /tmp/staphd43Lr
--
Regards,
Tom Zanussi <zanussi@us.ibm.com>
IBM Linux Technology Center/RAS
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: proposed relayfs changes and systemtap
2006-03-16 15:22 ` Tom Zanussi
@ 2006-03-16 15:39 ` William Cohen
0 siblings, 0 replies; 5+ messages in thread
From: William Cohen @ 2006-03-16 15:39 UTC (permalink / raw)
To: Tom Zanussi; +Cc: systemtap
Tom Zanussi wrote:
> William Cohen writes:
> > Tom Zanussi wrote:
> > > Tom Zanussi writes:
> > > > Hi,
> > > >
> > > > Someone over the weekend posted a patch to remove the filesystem part
> > > > of relayfs and instead have applications' relayfs files live in other
> > > > filesytems instead:
> > > >
> > > > http://marc.theaimsgroup.com/?l=linux-kernel&m=114036950922176&w=2
> > >
> > > Actually, this is probably a better link to the patches and thread:
> > >
> > > http://marc.theaimsgroup.com/?l=linux-kernel&m=114038358624583&w=2
> > >
> > > Tom
> > >
> > >
> >
> > Hi Tom,
> >
> > The nightly tests buildok and pass-5 tests failed due to the following
> > checkin:
> >
> > 2006-03-15 Tom Zanussi <zanussi@us.ibm.com>
> >
> > * configure.ac: Add check for relayfs version.
> > * configure: Regenerated.
> > * stp_check.in: Remove bundled relayfs loading/compilation, add
> > check for relay (vs relayfs).
> >
> > The parser tests won't find relayfs-config.h nor will the pass-5 tests
> > find the include file.
>
> Hi,
>
> I just checked in new changes for that - can you get the latest from
> cvs and try again? Sorry for the inconvenience.
>
> Tom
That fixed the problem. The tests seem to run much better now.
-Will
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2006-03-16 15:39 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2006-02-21 14:57 proposed relayfs changes and systemtap Tom Zanussi
2006-02-21 15:06 ` Tom Zanussi
2006-03-16 15:16 ` William Cohen
2006-03-16 15:22 ` Tom Zanussi
2006-03-16 15:39 ` William Cohen
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).