After a couple of cvsups and a complete make clean
it still gives me this:
anyone else seeing this?.
ref3# make
cc -c -O -pipe -Wall -Wredundant-decls -Wnested-externs
-Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Winline
-Wcast-qual -fformat-extensions -ansi -g -nostdinc -I-
../../../dev/usb/uhci.c: In function `uhci_dump_all':
../../../dev/usb/uhci.c:694: structure has no member named `hlink'
../../../dev/usb/uhci.c: At top level:
../../../dev/usb/uhci.c:1270: warning: `uhci_reset' defined but not used
../../../dev/usb/uhci.c:261: warning: `uhci_dump_ii' used but
Lots of my old programs get:
/usr/libexec/ld-elf.so.1: /usr/local/lib/libpng.so.3: Undefined symbol
__stderrp
this is NOT the error in UPDATING
and doing what it says for a similar problem:
20010924:
The buildworld has been fixed. You may need to install
the 4.x
On Mon, 21 Jan 2002, Emiel Kollof wrote:
Julian Elischer heeft op maandag 21 januari 2002 om 08:34 het volgende
geschreven:
Lots of my old programs get:
/usr/libexec/ld-elf.so.1: /usr/local/lib/libpng.so.3: Undefined symbol
__stderrp
this is NOT the error in UPDATING
this machine has never been 3.x.
the binaries worked fine up until about 5 months ago.
On Mon, 21 Jan 2002, M. Warner Losh wrote:
Some -current binaries can have this, but recompiling usually fixes
it. Also, I have some older 3.x binaries that I had to install
COMPAT3 to get working.
no, should I?
On Mon, 21 Jan 2002, Alexander Kabaev wrote:
and I've made world a lot of times like that.
and if I do it by hand as sugested, it doesn;t make any difference
either.
Just a guess - have you removed existing old libraries from /usr/lib?
To Unsubscribe: send mail to
: [EMAIL PROTECTED]
: Julian Elischer [EMAIL PROTECTED] writes:
: : this machine has never been 3.x.
: : the binaries worked fine up until about 5 months ago.
:
: All bets are off if it ran -current. You need to rebuild everything.
:
:
: Ah yes, I remember this problem
step 8
a summary for the rest of us?
I got lost there in the middle soemwhere.
On 21 Jan 2002, Dag-Erling Smorgrav wrote:
Andrey A. Chernov [EMAIL PROTECTED] writes:
Thanks to all, especially to Dag-Erling.
Thanks to *you* for pointing out and explaining the issues, submitting
patches,
control processes).
I will need people to help add support for multiple threads to ps, top, etc.
(once we decide what the new interface is).
just a heads-up
julian
p.s diffs from
http://www.freebsd.org/~julian/
--
++ __ _ __
| __--_|\ Julian
Could a usb afficionado fix this?
uhci.o: In function `uhci_idone':
uhci.o(.text+0x1330): undefined reference to `uhci_dump_ii'
uhci.o: In function `uhci_device_isoc_done':
uhci.o(.text+0x2fab): undefined reference to `uhci_dump_ii'
*** Error code 1
To Unsubscribe: send mail to [EMAIL
`
while trying to connect.
--
++ __ _ __
| __--_|\ Julian Elischer | \ U \/ / hard at work in
| / \ [EMAIL PROTECTED] +--x USA\ a very strange
| ( OZ)\___ ___ | country
or two after that.
thanks for your patience..
Julian
--
++ __ _ __
| __--_|\ Julian Elischer | \ U \/ / hard at work in
| / \ [EMAIL PROTECTED] +--x USA\ a very strange
| ( OZ
Non x86 FreeBSD developers:
I have posted to http://www.freebsd.org/~julian/adiff
a set of 'semi-cosmetic' mechanical diffs.
The effect of the diffs is to remove all the places in the code
which assume that the thread structure is embedded in the process
structure, without doing any major
ata-raid.o: In function `ar_attach':
ata-raid.o(.text+0xc0): multiple definition of `ar_attach'
if_ar.o(.text+0x10): first defined here
/usr/libexec/elf/ld: Warning: size of symbol `ar_attach' changed from
663 to 777 in ata-raid.o
The ata-raid author should have checked whether there was
What I think mergemaster should do is compere the file with the original
checked out one it claims to be and if it's the same, it should just
update silently.. i.e. if the user didn't change anything in th old one
he's unlikely to want to change anything in the new one
(maybe a datafile
for the set of patches at:
http://www.freebsd.org/~julian/adiff
these patches SHOULD NOT EFFECT your system except to do some
slight re-aranging of stuff in the kernel.
THe aim is to get this committed to 'clarify' the upcoming
KSE commit in http://www.freebsd.org/~julian/thediff
which
I don't know the alpha at all but
how about
leaq t0, thread0
or some similar thing
or even..
ldq t0,$thread0
or something similar..
who's assembler is being used?
On Wed, 6 Feb 2002, Andrew Gallatin wrote:
Andrew Gallatin writes:
Since thread0 is no longer a pointer, this
Ok so since I don't know alpha assembly.
can it be changed to teh address of the thread0 structure?
(I'll bet the same thing needs to be done on the other
architectures)
THANKS!
On Wed, 6 Feb 2002, Andrew Gallatin wrote:
Julian Elischer writes:
for the set of patches at:
http
eni_transmit.o: In function `eni_output':
eni_transmit.o(.text+0x3aa): undefined reference to `MFREE'
eni_transmit.o(.text+0x3be): undefined reference to `MFREE'
eni_transmit.o(.text+0x47a): undefined reference to `MFREE'
fore_output.o: In function `fore_xmit_segment':
THANKS!
(3 down, 2 to go..)
On Wed, 6 Feb 2002, Jake Burkholder wrote:
Apparently, On Wed, Feb 06, 2002 at 12:01:44PM -0800,
Julian Elischer said words to the effect of;
for the set of patches at:
http://www.freebsd.org/~julian/adiff
these patches SHOULD NOT EFFECT your
thread0 is not the structure itself rather than a pointer
that should be the only difference that should affect MD code
On 7 Feb 2002, Benno Rice wrote:
On Thu, 2002-02-07 at 08:01, Julian Elischer wrote:
for the set of patches at:
http://www.freebsd.org/~julian/adiff
As part of the KSe stuff I ended up changing ht ebehaviour of threads with
respect to their ucreds. Previously, they freed their ucred reference when
they entered user space and picked them up again when they re-entered the
kernel. It there was an AST then they re-loaded teh already freed ucred
for a process with one thread but the program
will need more help than that to handle more than one thread.
kvread and KREAD are basically the same thing.
basically you have read in teh thread structure and then read the pcb from
there..
On Thu, 7 Feb 2002, Andrew Gallatin wrote:
Julian Elischer
On Thu, 7 Feb 2002, Terry Lambert wrote:
Julian Elischer wrote:
In the KSE code I have:
in trap(), and syscall()
if (td-td_ucred != p-p_ucred) {
PROC_LOCK(p);
if (td-td_ucred) {
crfree(td-td_ucred
there may be greakage in gdb due th the commit today..
I'm waiting for my buildworld to get sround to it so I can test my fix..
julian
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
phk, is this you?
c++ -g -O -pipe -DHAVE_STDLIB_H=1 -DHAVE_UNISTD_H=1 -DHAVE_DIRENT_H=1
-DHAVE_LIMITS_H=1 -DHAVE_STRING_H=1 -DHAVE_STRINGS_H=1 -DHAVE_MATH_H=1
-DRET_TYPE_SRAND_IS_VOID=1 -DHAVE_SYS_NERR=1 -DHAVE_SYS_ERRLIST=1
-DHAVE_CC_LIMITS_H=1 -DRETSIGTYPE=void -DHAVE_STRUCT_EXCEPTION=1
On Thu, 7 Feb 2002, Julian Elischer wrote:
there may be greakage in gdb due th the commit today..
I'm waiting for my buildworld to get sround to it so I can test my fix..
julian
the fix I am looking at testing is:
Index: i386/kvm-fbsd.c
, Terry Lambert wrote:
Julian Elischer wrote:
yes.. well in booting yuo can have a null ucred. ( I know,s
I've hit it), but in general you are correct.
[ ... ]
What is the default state of td-td_ucred?
on creation, NULL followed very rapidly with being set to
p-p_ucred. (via crhold
On Thu, 7 Feb 2002, John Baldwin wrote:
This code is not safe on SMP non-i386 machines (i.e. ia64, sparc64, alpha, and
possibly p3 and later i386's) because the p_ucred value you read could easily
be a stale value, thus rendering the test invalid. You need the lock for the
compare.
On Fri, 8 Feb 2002, Poul-Henning Kamp wrote:
In message [EMAIL PROTECTED], Ju
lian Elischer writes:
phk, is this you?
/usr/include/sys/time.h:137: integer constant out of range
/usr/include/sys/time.h:137: warning: decimal integer constant is so large
that it is unsigned
Yes, that
for what it's worth, doing a make includes
before the make buildworld seems to have got me going again.
On Fri, 8 Feb 2002, Julian Elischer wrote:
On Fri, 8 Feb 2002, Poul-Henning Kamp wrote:
In message [EMAIL PROTECTED], Ju
lian Elischer writes:
phk, is this you?
/usr
can you try a kernel from JUST BEFORE I did the KSE commit yesterday?
I heard someone else complain of thisyesterday afternoon. At that time I
wascertain it was too soon after my commit for him to already have got it,
but it would be nice tio know if I screwed something...
On Fri, 8 Feb 2002,
This succeeded, but looks suspicious to me.
syncing disks... 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1
I've never seen it like that before..
It might be related to the problem some have seen with syncing..
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current
I'd like to commit the code to keep the ucred across userland,
with the code to clear it to NULL kept under DEBUG ifdefs.
i.e.
in trap(), ast() and syscall()
if (td-td_ucred != p-p_ucred) {
PROC_LOCK(p);
if (td-td_ucred) {
Is it possible to make devices enumerated by acpi to take notice of the
flags given to them in device.hints?
for example if I allow acpi then I get:
sio0 port 0x3f8-0x3ff irq 4 on acpi0
sio0: type 16550A, console
sio1 port 0x2f8-0x2ff irq 3 on acpi0
sio1: type 16550A
despite the fact that I
On Fri, 8 Feb 2002, David Wolfskill wrote:
FreeBSD/i386 (freebeast.catwhisker.org) (cuaa0)
login: boot() called on cpu#0
Waiting (max 60 seconds) for system process `vnlru' to stop...stopped
Waiting (max 60 seconds) for system process `bufdaemon' to
stop...stopped
Waiting (max 60
On Fri, 8 Feb 2002, Eugene M. Kim wrote:
On Fri, Feb 08, 2002 at 01:43:54PM -0800, Julian Elischer wrote:
On Fri, 8 Feb 2002, David Wolfskill wrote:
Waiting (max 60 seconds) for system process `vnlru' to stop...stopped
Waiting (max 60 seconds) for system process `bufdaemon
On Fri, 8 Feb 2002, Bruce Evans wrote:
[...]
if (td-td_ucred != p-p_ucred) {
PROC_LOCK(p);
if (td-td_ucred) {
crfree(td-td_ucred);
td-td_ucred = NULL;
}
if
, Eugene M. Kim wrote:
Attached is the requested DDB log (I guessed pid 7 `syncer' is the
process doing the sync; if this is wrong let me know).
Eugene
PS. I used the serial console, so don't feel sorry to ask. =)
On Fri, Feb 08, 2002 at 02:41:30PM -0800, Julian Elischer wrote
david,
On Fri, 8 Feb 2002, Julian Elischer wrote:
cool..
ok, how about adding show witness
[...]
can you do a cvs diff -D{time1] -D[time2] /sys
where time2 is the earliest kernel that has the problem and time1 is the
last kernel that doesn't?
there may be some other diffs in there I'm
Eh; I suspect that's showing the entry to ddb.
Hmmm... gow about:
db show witness
Sleep locks:
0 Giant -- last acquired @ /usr/src/sys/kern/kern_intr.c:532
1 eventhandler -- last acquired @ /usr/src/sys/kern/subr_eventhandler.c:162
3lockmgr -- last acquired @
. =)
On Fri, Feb 08, 2002 at 02:41:30PM -0800, Julian Elischer wrote:
On Fri, 8 Feb 2002, Eugene M. Kim wrote:
On Fri, Feb 08, 2002 at 01:43:54PM -0800, Julian Elischer wrote:
On Fri, 8 Feb 2002, David Wolfskill wrote:
Waiting (max 60 seconds) for system process `vnlru
yes but is it a SMP or UP kernel? (SMP kernel can run on some UP h/w)
thanks!
On Fri, 8 Feb 2002, Eugene M. Kim wrote:
On Fri, Feb 08, 2002 at 03:56:21PM -0800, Julian Elischer wrote:
In your case we need totrace proc 1 I think..
I got the `reboot' process at this session, so I
I tlooks as if show locks would not show any locks held by anyone..
is this true?
On Fri, 8 Feb 2002, Eugene M. Kim wrote:
On Fri, Feb 08, 2002 at 03:56:21PM -0800, Julian Elischer wrote:
In your case we need totrace proc 1 I think..
I got the `reboot' process at this session, so I
running on an UP box.
Eugene
On Fri, Feb 08, 2002 at 04:53:28PM -0800, Julian Elischer wrote:
yes but is it a SMP or UP kernel? (SMP kernel can run on some UP h/w)
thanks!
On Fri, 8 Feb 2002, Eugene M. Kim wrote:
On Fri, Feb 08, 2002 at 03:56:21PM -0800, Julian
Thats it for sure!
committing now..
On Sat, 9 Feb 2002 [EMAIL PROTECTED] wrote:
h so what is the difference between your kernel and mine that works?
just out of curiosity, have you tried a very latest -current?
do you have your own config? how does GENERIC behave?
(what kind
yes,, this exactly fits the symptoms!
I've committed it.
(it's definitly wrong)
assume this will solv ethe problem.
now why doesn't MINE fail?
On Sat, 9 Feb 2002 [EMAIL PROTECTED] wrote:
h so what is the difference between your kernel and mine that works?
just out of curiosity,
:
On 08-Feb-02 Julian Elischer wrote:
I'd like to commit the code to keep the ucred across userland,
with the code to clear it to NULL kept under DEBUG ifdefs.
Use INVARIANTS for the ifdef macro name, but sure.
--
John Baldwin [EMAIL PROTECTED]http://www.FreeBSD.org/~jhb/
Power
No, I just never went into that particular clause of code
But is has made me rethink the whole issue.
On Fri, 8 Feb 2002, John Baldwin wrote:
On 09-Feb-02 Julian Elischer wrote:
yes,, this exactly fits the symptoms!
I've committed it.
(it's definitly wrong)
assume this will solv
Bruce, for the low-level impared such as myself, can you give a quick
precis on teh difference between fast interrupt handlers in -current
and 'normal' interrupt handlers.
Do fast interrupt handlers enter through trap() ?
if they interrupt a user process, do they take on the cred of the
John, (peter? others?)
How is it that getting a ucred reference is guarded by PROC_LOCK(p)
but freeing it is guarded by mtx_lock(Giant);
?
Call me naive, but shouldn't they be guarded by the same thing?
Julian
On Fri, 8 Feb 2002, Julian Elischer wrote:
I'm a little worried about
On Fri, 8 Feb 2002, Julian Elischer wrote:
John, (peter? others?)
How is it that getting a ucred reference is guarded by PROC_LOCK(p)
but freeing it is guarded by mtx_lock(Giant);
?
Call me naive, but shouldn't they be guarded by the same thing?
In fact, if we had good atomic
Thanks
On Sun, 10 Feb 2002, Bruce Evans wrote:
Yes, anything that reaches doreti checks for ASTs and runs userret() if
necessary and possible (only for returns to user mode).
Hmm, this check seems to be inadequate for fast interrupts. There is
no check for rescheduling if the return is
On Sun, 10 Feb 2002, Bruce Evans wrote:
On Fri, 8 Feb 2002, Julian Elischer wrote:
I'd like to commit the code to keep the ucred across userland,
with the code to clear it to NULL kept under DEBUG ifdefs.
i.e.
in trap(), ast() and syscall()
if (td-td_ucred != p
On Sun, 10 Feb 2002, Bruce Evans wrote:
On Sat, 9 Feb 2002, Julian Elischer wrote:
AST is not always called
and userret is always called, but unfortunatly sometimes multiple times
userret() isn't always called either in my version :-). When I'm
finished, it will never be called
On Sun, 10 Feb 2002, Bruce Evans wrote:
This is a well known bug in the device layer. I reported it on 2001/12/26
and fixed it locally a little later. See the thread in -current about
panic during fdisk'ing a md(4) device for patches.
Can you commit the fix?
To Unsubscribe: send
After comments by jhb and bde
--
++ __ _ __
| __--_|\ Julian Elischer | \ U \/ / hard at work in
| / \ [EMAIL PROTECTED] +--x USA\ a very strange
| ( OZ
On Sun, 10 Feb 2002, Bruce Evans wrote:
On Sat, 9 Feb 2002, Julian Elischer wrote:
On Sun, 10 Feb 2002, Bruce Evans wrote:
On Sat, 9 Feb 2002, Julian Elischer wrote:
AST is not always called
and userret is always called, but unfortunatly sometimes multiple times
userret
Anyone making the usual plans or should I just look for the terminal room?
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
On Mon, 11 Feb 2002, John Baldwin wrote:
On 11-Feb-02 Julian Elischer wrote:
here is the BDE version ready to commit.
Extended to other architectures.
Bruce, John, comments?
As I was adding a prototype to ucred.h I stripped the __Ps of the others in
that
section
On Mon, 11 Feb 2002, Terry Lambert wrote:
Julian Elischer wrote:
This makes little sense to me.
Maybe I'm missing something, but by virtue of ownership we don't
have to worry about the ucred's refcount on entry into the kernel
because it is the owner and no one else is allowed
I've committed both the kernel diff and a libkvm diff that seems to work
for me here.
can you check it with the new -current again ?
On Wed, 6 Feb 2002, Andrew Gallatin wrote:
Jake Burkholder writes:
What's the right way to do this?
I think you want lda, its used to load an
On Thu, 7 Feb 2002, Terry Lambert wrote:
Julian Elischer wrote:
In the KSE code I have:
in trap(), and syscall()
if (td-td_ucred != p-p_ucred) {
PROC_LOCK(p);
if (td-td_ucred) {
crfree(td-td_ucred
, Terry Lambert wrote:
Julian Elischer wrote:
yes.. well in booting yuo can have a null ucred. ( I know,s
I've hit it), but in general you are correct.
[ ... ]
What is the default state of td-td_ucred?
on creation, NULL followed very rapidly with being set to
p-p_ucred. (via crhold
phk, is this you?
c++ -g -O -pipe -DHAVE_STDLIB_H=1 -DHAVE_UNISTD_H=1 -DHAVE_DIRENT_H=1
-DHAVE_LIMITS_H=1 -DHAVE_STRING_H=1 -DHAVE_STRINGS_H=1 -DHAVE_MATH_H=1
-DRET_TYPE_SRAND_IS_VOID=1 -DHAVE_SYS_NERR=1 -DHAVE_SYS_ERRLIST=1
-DHAVE_CC_LIMITS_H=1 -DRETSIGTYPE=void -DHAVE_STRUCT_EXCEPTION=1
On Thu, 7 Feb 2002, Julian Elischer wrote:
there may be greakage in gdb due th the commit today..
I'm waiting for my buildworld to get sround to it so I can test my fix..
julian
the fix I am looking at testing is:
Index: i386/kvm-fbsd.c
there may be greakage in gdb due th the commit today..
I'm waiting for my buildworld to get sround to it so I can test my fix..
julian
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
To Unsubscribe: send mail to [EMAIL PROTECTED]
On Fri, 8 Feb 2002, Poul-Henning Kamp wrote:
In message [EMAIL PROTECTED], Ju
lian Elischer writes:
phk, is this you?
/usr/include/sys/time.h:137: integer constant out of range
/usr/include/sys/time.h:137: warning: decimal integer constant is so large
that it is unsigned
Yes, that
for what it's worth, doing a make includes
before the make buildworld seems to have got me going again.
On Fri, 8 Feb 2002, Julian Elischer wrote:
On Fri, 8 Feb 2002, Poul-Henning Kamp wrote:
In message [EMAIL PROTECTED], Ju
lian Elischer writes:
phk, is this you?
/usr
On Fri, 8 Feb 2002, Eugene M. Kim wrote:
On Fri, Feb 08, 2002 at 01:43:54PM -0800, Julian Elischer wrote:
On Fri, 8 Feb 2002, David Wolfskill wrote:
Waiting (max 60 seconds) for system process `vnlru' to stop...stopped
Waiting (max 60 seconds) for system process `bufdaemon
, Eugene M. Kim wrote:
Attached is the requested DDB log (I guessed pid 7 `syncer' is the
process doing the sync; if this is wrong let me know).
Eugene
PS. I used the serial console, so don't feel sorry to ask. =)
On Fri, Feb 08, 2002 at 02:41:30PM -0800, Julian Elischer wrote
david,
On Fri, 8 Feb 2002, Julian Elischer wrote:
cool..
ok, how about adding show witness
[...]
can you do a cvs diff -D{time1] -D[time2] /sys
where time2 is the earliest kernel that has the problem and time1 is the
last kernel that doesn't?
there may be some other diffs in there I'm
Eh; I suspect that's showing the entry to ddb.
Hmmm... gow about:
db show witness
Sleep locks:
0 Giant -- last acquired @ /usr/src/sys/kern/kern_intr.c:532
1 eventhandler -- last acquired @ /usr/src/sys/kern/subr_eventhandler.c:162
3lockmgr -- last acquired @
Ok so since I don't know alpha assembly.
can it be changed to teh address of the thread0 structure?
(I'll bet the same thing needs to be done on the other
architectures)
THANKS!
On Wed, 6 Feb 2002, Andrew Gallatin wrote:
Julian Elischer writes:
for the set of patches at:
http
On Fri, 8 Feb 2002, Bruce Evans wrote:
[...]
if (td-td_ucred != p-p_ucred) {
PROC_LOCK(p);
if (td-td_ucred) {
crfree(td-td_ucred);
td-td_ucred = NULL;
}
if
I'd like to commit the code to keep the ucred across userland,
with the code to clear it to NULL kept under DEBUG ifdefs.
i.e.
in trap(), ast() and syscall()
if (td-td_ucred != p-p_ucred) {
PROC_LOCK(p);
if (td-td_ucred) {
No, I just never went into that particular clause of code
But is has made me rethink the whole issue.
On Fri, 8 Feb 2002, John Baldwin wrote:
On 09-Feb-02 Julian Elischer wrote:
yes,, this exactly fits the symptoms!
I've committed it.
(it's definitly wrong)
assume this will solv
John, (peter? others?)
How is it that getting a ucred reference is guarded by PROC_LOCK(p)
but freeing it is guarded by mtx_lock(Giant);
?
Call me naive, but shouldn't they be guarded by the same thing?
Julian
On Fri, 8 Feb 2002, Julian Elischer wrote:
I'm a little worried about
:
On 08-Feb-02 Julian Elischer wrote:
I'd like to commit the code to keep the ucred across userland,
with the code to clear it to NULL kept under DEBUG ifdefs.
Use INVARIANTS for the ifdef macro name, but sure.
--
John Baldwin [EMAIL PROTECTED]http://www.FreeBSD.org/~jhb/
Power
I don't know the alpha at all but
how about
leaq t0, thread0
or some similar thing
or even..
ldq t0,$thread0
or something similar..
who's assembler is being used?
On Wed, 6 Feb 2002, Andrew Gallatin wrote:
Andrew Gallatin writes:
Since thread0 is no longer a pointer, this
Thanks
On Sun, 10 Feb 2002, Bruce Evans wrote:
Yes, anything that reaches doreti checks for ASTs and runs userret() if
necessary and possible (only for returns to user mode).
Hmm, this check seems to be inadequate for fast interrupts. There is
no check for rescheduling if the return is
Anyone making the usual plans or should I just look for the terminal room?
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
On Mon, 11 Feb 2002, Alfred Perlstein wrote:
* Julian Elischer [EMAIL PROTECTED] [020211 15:00] wrote:
In the current world, when the thread enters userland, it does:
lock giant
crfree() (which includes mutexes)
unlock giant
This isn't needed afaik.
Argue with jhb
As part of the KSe stuff I ended up changing ht ebehaviour of threads with
respect to their ucreds. Previously, they freed their ucred reference when
they entered user space and picked them up again when they re-entered the
kernel. It there was an AST then they re-loaded teh already freed ucred
This is a multi-part message in MIME format.
--Boundary_(ID_TA6w1McSahPHMx8+rAaDoQ)
Content-type: text/plain; charset=iso-8859-2
Content-transfer-encoding: 7BIT
After comments by jhb and bde
--
++ __ _ __
| __--_|\ Julian Elischer
Is it possible to make devices enumerated by acpi to take notice of the
flags given to them in device.hints?
for example if I allow acpi then I get:
sio0 port 0x3f8-0x3ff irq 4 on acpi0
sio0: type 16550A, console
sio1 port 0x2f8-0x2ff irq 3 on acpi0
sio1: type 16550A
despite the fact that I
yes,, this exactly fits the symptoms!
I've committed it.
(it's definitly wrong)
assume this will solv ethe problem.
now why doesn't MINE fail?
On Sat, 9 Feb 2002 [EMAIL PROTECTED] wrote:
h so what is the difference between your kernel and mine that works?
just out of curiosity,
On Mon, 11 Feb 2002, John Baldwin wrote:
On 12-Feb-02 Julian Elischer wrote:
The proclock is needed to get the reference,
guarding against other threads, and giant is needed fo rnot to free it
because if it reaches a refcount of 0 it needs to call free(). (which john
assures me
I apparently broke the build of libkvm (again)
I will fix it shortly.
I'm in transit right now.
Julian
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
I made a change recently to make LINT compile because the
ata raid was using the symbol ar_attach and I changed it to be atar_attach
because ar_attach was already used in the if_ar driver.
please check that there isn't a MACRO somewhere that automatically
generates an 'ar_attach reference)
On Mon, 18 Feb 2002, Matthew Dillon wrote:
[...]
It turns out that the two processes got into an extremely non-optimal
contested/sleep/wakeup situation, even though they do not actually have
to sleep on Giant in this situation.
The solution is to allow
On Mon, 18 Feb 2002, Matthew Dillon wrote:
:
:I can't see any major problem with this but I can't help thinking that
:there must be one.. on UP the question is: who is going to
:release the lock if no-one is runnable?
An interrupt, of course. Wakeups don't happen out of thin air!
The fully safe version of this code is:
td-td_retval[0] = td-td_ucred-cr_ruid;
td-td_retval[1] = td-td_ucred-cr_uid;
return (0);
because td-td_ucred is read-only for it's whole existance.
Also, it's not worth arguing about this when Jouhn's not here..
and when he is it's better offline..
I'd prefer to be working on a branch of CVS if it weren't for the people
that would scream whenever I moved my merged tag up.
(eek eek cvsup bloat).
That way i would have a dozen people helping me but with my code in P4 I
have me and occasionally Peter.
P4 is ok but it's strangling me because
exactly
On Tue, 19 Feb 2002, Terry Lambert wrote:
Julian Elischer wrote:
The fully safe version of this code is:
td-td_retval[0] = td-td_ucred-cr_ruid;
td-td_retval[1] = td-td_ucred-cr_uid;
return (0);
because td-td_ucred is read-only for it's whole existance.
???
Are you
that touch multiple files as a unit.
Basically I think tha exported CVS tree form cvsup10 is the way to go...
for working on projects held in P4. :-/
On Tue, 19 Feb 2002, David Xu wrote:
- Original Message -
From: Peter Wemm [EMAIL PROTECTED]
To: Julian Elischer [EMAIL PROTECTED]
Cc
On Tue, 19 Feb 2002, George V. Neville-Neil wrote:
Hi Folks,
Now that Luigi has put in polling support for some ethernet drivers
I was wondering how much work it would be to make the remote kernel debugging
run over the ethernet. I have worked on systems like this before (it's the
On Thu, 21 Feb 2002, Greg Lehey wrote:
On Tuesday, 19 February 2002 at 0:00:19 -0800, Peter Wemm wrote:
Am I the only person who, despite careful scrutiny, missed this
announcement? I would have thought that this would at least have been
worth a HEADS UP and prior discussion in -core.
the probkem is that phk didn't test a buildworld before making the
following change to proc.h:
revision 1.204
date: 2002/02/22 13:32:01; author: phk; state: Exp; lines: +1 -1
Convert p-p_runtime and PCPU(switchtime) to bintime format.
On Fri, 22 Feb 2002, John Hay wrote:
Date: Fri, 22 Feb 2002 08:09:28 -0900
Last night's build for -current failed with the following:
I got -CURRENT built today without problem. Recent CVSup history:
(snip history)
Might try updating again
Already did
401 - 500 of 1563 matches
Mail list logo