Re: tcsh.cat

< said: > Maybe, but this doesn't seem to be permitted by POSIX.1-200x: The response I got to a similar question about symbolic links suggests that the definition of ``path name resolution'' is supposed to be the final word on the meaning and interpretation of symbolic links. -GAWollman To Un

Re: tcsh.cat

< said: > Here's an example of a complication: what is the semantics of /tmp/foo/bar > where foo is a symlink to ""? I think the pathname resolves to > /tmp//bar and then to /tmp/bar, but this is surprising since foo doesn't > point anywhere. But this is at least consistent with the historic (p

Re: fts_open() (was: Re: Patch to restore WARNS feature)

< said: > Why not do something like the rpc code does? Check if threaded, if > so cons up a thread specific key otherwise use a global. The Standard does not appear to say whether qsort() is reentrant, but I believe that it ought to be. fts() was clearly intended to be reentrant. (I don't kno

Re: fts_open() (was: Re: Patch to restore WARNS feature)

< said: > How should we call this function? > (I'll implement this tomorrow.) I was would call it `qsort_with_arg' or something similar. There is a namespace issue here; reserves very little namespace for the implementation, so anything we add will have to be hidden from strictly-compliant app

Re: fts_open() (was: Re: Patch to restore WARNS feature)

< said: > 3. Provide an alternative to qsort() that takes an comparison function >that takes an additional function pointer arg (use this arg to avoid >the global in (1)). Actually, doing this would solve a number of similar problems, and it doesn't look to be too difficult to do. I wou

Patch to restore WARNS feature

< said: > + qsort((void *)sp->fts_array, nitems, sizeof(FTSENT *), > + (int (*) __P((const void *, const void *)))sp->fts_compar); This is wrong. The declaration of the comparison function should be fixed, rather than papering over the mistake here. (This is arguably a deficiency i

Re: compiler bug or source bug?

< said: > We could make off_t a long long on the alpha. Otherwise the formats need: > printf("%lld", (long long) pos); /* or %qd to shut up the ANSI warning */ They always need that, until printf learns about the `j' width specifier (at which time they all change to `printf("%jd", (intmax_t)

Correctness of UIO_MAXIOV definition?

< said: >The second question I have is more standards based. > Should we consider changing UIO_MAXIOV to IOV_MAX or > _XOPEN_IOV_MAX and deprecating the 1st? I am unclear > on what the standard is for this. UIO_MAXIOV is what the kernel is willing to do. IOV_MAX being standardized is what s

Re: pgm to kill 4.3 via vm

< said: > The process and signal-related structures may be inconsistent if the > debugger disregards existing locks held over those structures. It does > not matter if code is currently still executing, it matters that > preemption can occur. The choices appear to be: Preemption should never o

Re: pgm to kill 4.3 via vm

< said: > I followed everything here fine until you asserted that the debugger > shouldn't need any locks. When the debugger is running, everything else should have been forcibly halted. -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body o

Re: Boot messages

< said: > Actually, it is a bug. The drivers in the tree should grok these pnp > ids. Actually, no, it is not a bug. The FreeBSD drivers for these devices manage their resources differently from the way the Windows drivers do, and the result is not unexpected if you look closely at the dump in

Re: Boot messages

< said: > This means that I can remove this lines? Sure? > device atkbdc 1 No, I said nothing of the sort. -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message

Re: Boot messages

[Followups to -questions, please.] < said: > unknown: can't assign resources Keyboard controller. > unknown: can't assign resources PS/2 mouse port. > unknown: can't assign resources Serial port whose settings conflict with one of your configured serial ports. > unknown: can't assign

Re: Boot messages

< said: > This is not a bug. This is an FAQ. So much that it's actually > documented in (*gasp!*) the FAQ: Unfortunately, the A in the FAQ is wrong. The ``can't assign resources'' messages indicate that the devices are legacy ISA devices for which a non-PnP-aware driver is compiled into the k

vm_mtx

< said: > You can find the work I've done so far to make a giant vm mutex > here: The Mach code we originally inherited was supposed to already by multiprocessor safe. Did we manage to eliminate that capability? -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe free

Re: cp -d dir patch for review (or 'xargs'?)

< said: > '-y[replstr]' (no blank after -y). Prohibited by POSIX. The `xargs' utility ``shall'' follow the Utility Syntax Guidelines. > so I don't know how we go about adding options to it. POSIX is clear on this issue: the implementation may add any options it wishes, provided that those opt

Re: FIO* doc added to tty.4 (review)

< said: > Does POSIX now specify select() and/or poll() precisely enough to > show that the current behaviour is wrong? In addition to more explicit requirements for sockets, draft 6 has the following to say about select() and pselect(): A descriptor shall be considered ready for readin

Re: FIO* doc added to tty.4 (review)

< said: > ie: FIONREAD will succeed from a backgrounded process, but the > data read itself with hang (tty driver). You are describing a general property of the terminal subsystem, which any process dealing with terminals needs to be aware of. (AFAIK, ioctl(tty, FIONREAD, &n) should set n t

FIO* doc added to tty.4 (review)

< said: >I've been doing some tty related work and found the FIO* > requests don't seem to be documented. >I've added some reasonable doc to tty.4 and put up the > the diff and html forms: As the initial letter suggests, they are generic file ioctls and not specific to the teletype driv

Re: SOMAXCONN -- not tunable?

< said: > Actually, the correct fix is most likely to redefine it to -1. :) POSIX.1-200x draft 5 has this to say: Implementations shall support values of backlog up to SOMAXCONN, defined in If listen( ) is called with a backlog argument value that is less than

Re: Who is maintainer of kerberos/heimdall/sendmail?

< said: tlambert> The value of SOMAXCONN is not valis; the valid limit is only tlambert> obtainable from sysctl (kern.ipc.somaxconn). > We (Sendmail) will look at integrating your fix into 8.12 (which will be > the first to actually use it -- it's #ifdef'ed out in 8.11). No code should ever exa

RE: selwakeup()

< said: > As a safety check we should probably zero the pid right before zfree()'ing a > proc in wait() however, so that a stale pointer to a free'd process doesn't > have a valid pid if we do this. Should not be necessary. Here is the logic: p = sip->si_p; mtx_lock_spin(&sched

selwakeup()

< said: > If I'm reading this backtrace right, the thread handling the sound > hardware called selwakeup() (frame #19). This called pfind() (frame > #18), which tries to lock allproc. selwakeup() shouldn't need to call pfind(). Because the process table is in type-stable memory, it should be

Re: Request for review [Re: /bin/ls patch round #2]

< said: > This particular case is different from what you say. There is no strict > POSIX/ISO C equivalent of functionality you describe, Certainly there is. The daemon(3) function is implemented entirely on top of POSIX interfaces: fork(), setsid(), chdir(), open(), dup2(), and close(). It i

Re: Request for review [Re: /bin/ls patch round #2]

< Sorry I'm not sure but rune API is slightly different > between 4.4BSD and Plan9, isn't it? Nobody runs Plan 9, whereas hundreds of thousands of machines run *BSD. > Sources of the standard commands are often used as a living > textbook to other programmers. They should be as `good' as > poss

Re: Request for review [Re: /bin/ls patch round #2]

< In general direct manipulation of rune is evil. > It is an internal data structure in libc; Not true. The `rune' API was developed by the Plan 9 people by intention to be different from (in their view, superior to) the ISO C multibyte/wide character API. > Actually NetBSD does not export . N

RE: Latest version of mega header file POSIX update

< said: > I don't think the sys/conf/Makefile.i386 change is needed. :) Oops. Sorry, that one leaked out > Nothing else jumped out at me while I glanced over it however, and > it seems fine at first glance. But did you *test* it? I know it compiles. -GAWollman To Unsubscribe: send ma

Latest version of mega header file POSIX update

I have finally produced another version of the big header file patch for people to look at. I'm still running an older (January) kernel and world based on these changes. I have verified that these still build on i386-architecture systems. I'm looking for some people to do the following: 1) Loo

Re: random reboots...

< said: > You need options LIBMCHAIN as well. We don't have mechanism for > specifying dependancies between options as of yet. (sorry, should put a > note in the NOTES). Actually, yes we do, although it's not often used. If the relevant sources are listed twice in `files', conditional on

Re: cp MAKEDEV /dev - on a system with devfs

< said: > There's one real oddity in FreeBSD: > lrwxrwxrwx 1 root wheel 13 Jan 28 13:42 rmt -> /usr/sbin/rmt* The pathname of the `rmt' program is a fundamental part of the `rmt' ``protocol'' such as it is. We've been over this one many times. -GAWollman To Unsubscribe: send

Posix feature tests update

< said: > Critique, please. I have almost completely finished this work. Please join the [EMAIL PROTECTED] mailing-list, where the patches were posted several months ago, and where hopefully more discussion can still take place. I have not had time recently to update the patches but will hopef

find / -fstype local traverses non-local filesystems

< said: > find seems to be traversing all file systems (local and non-local) > but just not reporting the found file when its on a non-local > filesystem. As has been discussed many times before, this is correct behavior. If you want `find' to not traverse a directory, use the `-prune' primary.

Re: Patch for FILE problems (was Re: -CURRENT is bad for me...)

< said: > There is no concept of "forward" or "backward" as far as the toolchain and > runtime support goes. There is only "filename exists" or "file not found". This is a human-factors issue, not a code issue. People expect to see version numbers increasing. -GAWollman To Unsubscribe: sen

Re: Patch for FILE problems (was Re: -CURRENT is bad for me...)

< said: > If we had taken -current to 500, we could go to 501, 502, etc as > required to stop killing our developers, and prior to entering 5.0-BETA we > go back to the next sequentially available major number (be it 5, or 6 > if RELENG_4 bumps again). Shared library version numbers going backw

Re: HEADS UP: installworld gotchas

< said: > That is (1) against Handbook documented policy, (2) too hackish (we > aren't Linux). And who came up with that policy in the first place? -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message

Re: HEADS UP: installworld gotchas

< said: > The major number has already been bumped, I thought. If this is true > then we've only broken compatibility with older versions of -current > after the version number was bumped but before this change, right? However, this may turn out to be so painful that we need to bump it again. -

Re: /etc/shells #include syntax support patch

< said: > I would rather that a separate configuration file be read, for example, > with a list of shells(5) format files to consult. I would rather have a single file, located in a directory intended for configuration files. Perhaps we could call it ``/etc/shells'' which seems to be popular.

patch for test: /etc/shells -> /usr/local/etc/shells

< said: > The patch below (against 4-stable but it will probably apply easily > to -current) moves /etc/shells to /usr/local/etc/shells. Bad idea. No base component (never mind libc!) should hard-code a pathname in /usr/local. -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED]

Re: current panics in mount(2)

< said: > I looked at fixing this once, but got scared off because of binary > compatibility issues. Would 'fixing' mount to use cmsgcred be > acceptable? No, it should use a structure appropriately named and designed for its own purpose. (By preference, it should be binary-compatible with 4.x

current panics in mount(2)

< said: > Somehow there were few problems when `struct mtx' was added to > `struct ucred'. The critical args were probably usually 0. It's a bug that mount(2) uses a bare `struct ucred' and not a well-defined, user-exportable credential structure (like struct cmsgcred used for SCM_CREDS ancilla

RE: Cannot build emulators/vmware2

< said: > work on stable, just do a #ifdef __FreeBSD_version > 4 use selinfo, else use > select.h. Make that: #if __FreeBSD_version >= 500014 #include #else #include #endif -GAWollman -- Garrett A. Wollman | O Siem / We are all family / O Siem / We're all the same [EMAIL PROTECTED] | O S

sys/time.h w/ timespec stuff

< said: > So now, maybe someone can answer my question: why is timespec _KERNEL? It's not. Some of the namespace pollution associated with it is. -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message

Re: excessive paranoia in syslogd(8)?

< said: > If you want to or need to use network sockets, > # syslogd -a localhost > Should provide the behavior you want. I.e., no security whatsoever. -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message

Re: Atomic breakage?

< said: > There are SMP machines using both 386 and 486 processors. There is > no support in FreeBSD for SMP on pre-Pentium processors. Yes, I well recall the Sequent. I wish for it to remain a memory. -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-curr

Re: Atomic breakage?

< said: > person was referring to: on Pentiums with stepping < 0xc, a cmpxchg8b that > crosses a page boundary triggers an illegel opcode fault rather than a page > fault if the second page is missing. This is (part of) the famous `F00F bug'. -GAWollman To Unsubscribe: send mail to [EMAIL P

Re: Atomic breakage?

< said: > Just wondering, can't you use 'LOCK addl' and then use 'LOCK addc'? > add longword, add longword with carry? I know it would be pretty > ugly, but it should work, no? The two bus cycles are independent, so there is a race condition. OTOH, it's a fairly *unlikely* race condition, and

Re: Atomic breakage?

< said: > To support multiple masters, you need proper locks. On older processors, yes. On processors with the CX8 feature bit set, you can do it without any sort of locking (indeed, this is a primitive that semaphores can be built upon). Consider the following: atomic_increment: ; pr

Re: /boot/kernel/kernel: swap_pager_getswapspace: failed

< said: > No I want mfs to grow and shrink its filesystem dynamically. MFS has never done so. MFS is simply a UFS which uses (swap-backed) memory instead of a physical disk; it relies on the filesystem to avoid touching blocks that it doesn't need, and on the VM system to avoid wasting memory o

Making {open,close,read,tell,seek}dir thread-safe.

< said: > I started a cleanup of libc to make it thread-safe. Just as a matter of information The seekdir/telldir interface was debated recently by the Austin Group. The Open Group wanted to include it as part of the XSI extension to 1003.1-200x; other people were strongly opposed to its in

Re: Confusing error messages from shell image activation

[Please watch your carbon copies!] < said: > However, FreeBSD is still the only vendor distribution I know of that > installs software in /usr/local. That's the problem - software that > comes from the vendor doesn't belong in the local administrative > regime. No software that is a part of Fre

Re: Confusing error messages from shell image activation

< said: > There are other places where FreeBSD doesn't comply with the > appropriate standard - packages vs. FHS I have never heard of ``FHS''. What is its ANSI, FIPS, IEEE, IEC, or ISO number? -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in t

Re: Confusing error messages from shell image activation

< said: > Um - compliance with what, exactly? IEEE Std.1003.1-1990 et seq. -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message

Re: RFC: /dev/console -> /var/log/messages idea/patch

< said: > In message <[EMAIL PROTECTED]>, Garrett Wollman write >> They can try, but by the time they do the console has already been >> revoke()d, so they no longer have access to the real console. > I don't know what you consider "the real console", b

Re: RFC: /dev/console -> /var/log/messages idea/patch

< said: > Another particular thing I remember was that some syslog-challenged > daemons whine on /dev/console long after /etc/rc has finished. They can try, but by the time they do the console has already been revoke()d, so they no longer have access to the real console. I've thought about wri

Re: Proper permissons on /var/mail

< said: > I have a similar problem -- every time I make world, perms on /var/mail > get set to 775. Mutt considers my mailbox read-only until I change it > to 1777. It is misconfigured (or perhaps just broken). 1777 mode for /var/mail is insecure, but was necessary in the mists of ancient past

Interrupt allocation

< said: > - pci drivers "know" that pci interrupts are shareable and force > RF_SHAREABLE. Is this required by the pci spec? Yes. > - the isa compatibility code and pcvt force RF_SHAREABLE although isa > irqs are rarely shareable. This is a bug. I suspect it may be due to code rot, relat

Re: ABI is broken??

< said: > I also gather it has to do with the Austin project Garrett? Yes and no. The errors have been there since the beginning of time, but I actually noticed them doing review of our implementation wrt the new standard. For example, the System V IPC implementation uses a data structure bogu

Re: ABI is broken??

< said: > Any reason to not get [libc ABI changes] in -current now and make > the bump? Mostly because they're too small to be worth the pain. I'm waiting for something more significant that I can piggy-back on. -GAWollman -- Garrett A. Wollman | O Siem / We are all family / O Siem / We're

Re: ABI is broken??

< said: > Huh, why we can't just bump libc_r version number and put older (buggy) version into > lib/compat as usually? This would not require any ugly hacks at all. If you want to bump libc_r's version, we should do it to libc as well, and in that case there are a large number of ABI fixes that

Re: XFree86 3.3.6_3 build dies on -current

< said: > Eventually yes, but not this way. According to Bruce sys/types is a > prerequisite for sys/wait. This is currently true, but should be fixed this year (probably not this month -- it depends on how much energy I have). Draft 4 (draft 5 isn't out yet) of POSIX.1-200x says the following

Re: new rc.network6 and rc.firewall6

Grrr !@#%$^ Reply-To: header < said: >> (i.e., a topological sort). Does `rcorder' call `tsort' or does it >> reinvent the wheel? > UTSL You could have simply answered the question. For the benefit of everyone else: yes, it reinvents the wheel. -GAWollman To Unsubscribe: send mai

Re: new rc.network6 and rc.firewall6

Grrr !@#$^& Reply-To:... < said: > Nope. All the /etc/rc.d/ files are scanned by `rcorder'. `rcorder' then > creates a dependacy graph from information in each /etc/rc.d/ file. A > walk of the graph is done to output the list of scripts in the order they > should run in. Hmmm. We alread

Re: "make release" breakage on today's -current

< said: > I agree with this sentiment.. please leave INET6 support in the GENERIC > kernel. I'm sure there are better things to disable, like MFS, SYSV*, > P1003_P1B and friends, and ICMP_BANDLIM. Um, let's only disable things that give us a useful amount of space back. (We may just be screwed

Re: more endian.h breakage; patch included.

< said: > It is strictly correct for POSIX.1-1990, but FreeBSD-2 never had the > requirement until now. POSIX.1-200x is relaxing similar requirements > (I'm not sure about this one), so it is too late to start enforcing it. Yes. All POSIX headers which are defined to use specific foo_t types a

cvs servers load

< said: > Any running load information on the CVS servers available? I keep stats for cvsup3 (thanks to cricket and ucd-snmp). cvsup3 peaks out in the early morning with a five-minute load between 12 and 17. Rarely does the load average get below 2. Cvsup3's swap utilization hovers around 200

TI1225 CardBus controller

< said: > From reading the mailing list archives, I figured that the TI1225 > driver does actually work in some laptops, because the BIOS does > some magic initialization which the driver misses when running on a > non-laptop box. It's not necessarily just because of what the BIOS does. I am of

mtree verification output format

< said: > make "extra" and "missing" attributes in the output > rather than prefixes which can be confused with filenames. > Don't do the "run-in" of the first attribute with a short > filename This looks like a good change, but while you're there: > size

Re: pw_class in _pw_passwd is null if __hashpw() is not called in prior

< said: > Here is another possible trouble. While libc.so.4 with nsswitch no > longer requires the magic '+' entry, libc.so.3 and earlier still > require '+'. IMHO, This Is A Bug. -GAWollman -- Garrett A. Wollman | O Siem / We are all family / O Siem / We're all the same [EMAIL PROTECTED]

Re: mtree again

< said: > What POSIX.1-200x says about this thing? I don't have this book in hand. The rationale includes a table of options used to controlling symlink following in various programs, and suggests that `-H', `-L', and `-P' be used in new programs which implement filesystem traversal. -GAWollman

Re: mtree again

< said: > Is their any harm in just keeping the -P flag as a no-op and optionally > remove it at some later time (for backward compatibility)? We should try to be consistent with POSIX.1-200x as much as possible. -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe fre

Permissions for /var/mail

< said: > Pine 4.21 complains that /var/mail is vulnerable, that the perms should be > 1777 Pine is a steaming, festering pile of Let's try that again. The developers of Pine have certain erroneous ideas about how mail systems work; this is one of them. Ignore Pine's plaints and things should

Re: SMPNG kernel on UP

< It would only take a small amount of Makefile magic to fix > this... something like: > perl -ne 'split(/:/); print ("#define\tUID_", uc($_[0]), "\t", \ >$_[2], "\n");' ${PASSWD} >${.TARGET} Oh, I forgot to point out -- this would be easy to do in AWK a

Re: SMPNG kernel on UP

< said: > Hmmm, they look good to me. Maybe Mark's system doesn't have group > operator at gid 5. That's one bad thing about the new DEVFS: it > appears to enshrine things like this in the kernel... It would only take a small amount of Makefile magic to fix this... something like: PASSWD?=

cvsup3 now carries rsa_eay.c

MIT's Technology Licensing Office has given me the OK to distribute rsa_eay.c, so it will now be available from cvsup3.freebsd.org (aka freebsd.lcs.mit.edu). -GAWollman -- Garrett A. Wollman | O Siem / We are all family / O Siem / We're all the same [EMAIL PROTECTED] | O Siem / The fires of f

Re: Fdescfs updates--coming to a devfs near you!

< said: > I must admit that I think in general that /dev/std{in,out,err} and /dev/fd > is bogus. It looks like something which happened "because we can" more > than something which has a legitimate need. It's required if we ever get around to supporting secure set-id shell scripts. (I think th

Re: Another broken buildworld

< said: > Garrett, now that RSA has released the patent, would you be willing to > add this file to the mirror on cvsup3? I have asked MIT's Technology Licensing Office what their position is in this regard, and hope to have an answer soon. -GAWollman To Unsubscribe: send mail to [EMAIL PROT

Re: USA_RESIDENT variable is no longer needed ?

< said: >> Moreover, can we also throw USA_RESIDENT variable away from ports ? > Perhaps..I'm not immediately sure. This is almost certainly not the last patent issue we'll have to deal with. Unfortunately, it is probably the last one which can be dealt with so simply (since the U.S. has now m

Proposal to clarify mbuf handling rules

< said: > However, many routines that take an mbuf parameter assume that the > mbuf given to them is modifiable and proceed to write all over it. s/assume/require as a necessary precondition/ It's not a coding error, it's part of the specification. No, it's not documented -- but it's pretty cl

Re: yarrow & /dev/random

< said: > You probably don't want to chose RC6 or MARS because their authors > will probably patent them if they lose, and then you'll have to back > off using them fast. If they were going to be patented, the application has already been filed, so you might as well assume that they are patented

Re: make buildworld br0ken in libutil

< said: > -On [2822 17:30], Ollivier Robert ([EMAIL PROTECTED]) wrote: >> Brian, I'm afraid you broke libutil... Every program using libutil now must >> depend on libcrypt too. No. This is precisely why shared libraries have dependencies. For static linking, what Brian has done Just Works.

Re: LINT doesn't compile

< said: > Somebody remind me again why we don't make memcmp(), memset(), > and memmove() available in the kernel? To keep the compiler from pessimizing them. -GAWollman -- Garrett A. Wollman | O Siem / We are all family / O Siem / We're all the same [EMAIL PROTECTED] | O Siem / The fires o

inheriting certificate trust

< said: > I've used this certificate to sign a new certificate, and Microsoft > recognizes it and the trust chain, and will use it for verifying the > servers identity, but not for mail. My guess is that (regardless of the software issues involved) you've just violated your agreement with Veri$i

Re: sendmail 8.11.0 trouble

< said: > According to Leif Neland: >> /etc/pwd.db: Berkeley DB Hash file (Version 2, Little Endian, Bucket Size > /etc/pwd.db should NOT be in DB 2.x format. Recompile pwd_mkdb to be sure to > use DB 1.85 and use it to regenerate /etc/*pwd.db. DB 1.85's file format *is* ``Version 2''. -GAWoll

Re: sendmail 8.11.0 trouble

< said: >> On Sunday, July 30, 2000, Leif Neland wrote: >> > Sendmail 8.11.0 >> > >> > When invoking sendmail or newaliases I get this message: >> > /etc/pwd.db: Invalid argument >> > >> > Sendmail can't read /etc/pwd.db, and therefore cannot deliver any local >> > mail. >> > >> > Howtofixitp

Re: Re[2]: Journaling Filesystem ?

< said: > I didn't even know that background fsck was supported at all. I > remember hearing Kirk talk about it as a future feature at FreeBSD CON > last year, but I havn't heard anything about it since. How do you > use it? It is not quite there yet. It requires the snapshot code, which just

Re[2]: Journaling Filesystem ?

< said: > With Softupdates, you still have to fsck. On a large FS (say half a > terabyte) that can take hours. No you don't. Please read the paper. -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message

Journaling Filesystem ?

< said: > Are there plans for something along this line for FreeBSD? Is there a > project underway? No. Soft Updates provides most of the benefits without requiring changes to the on-disk layout. See . -GAWollman To Unsubscribe: send mail to [EM

cer/b7b/pfc -> pem

< said: > I can export the key as a .cer, .p7b or .pfx, but openssl seems to want it > in .pem format. Of course, you haven't really told us what the format of these things is, so it's difficult to say. The ``standard'' export format is something called PKCS#12. You can use `openssl pkcs12' wi

Re: Request for comments: new `lpd' suite feature

< said: > Huh? Security through ignorance? Remember that `lpr' is setuid-root and uses a ``privileged'' port for its communications. Many sites may still be using trusted-host ``authentication'' internally, and LPRng's ``feature'' may enable a compromise of some such service. (Got enough scar

Re: dc driver and underruns (was: Strangeness with 4.0-S)

< said: > Ohh... and a finally note, DEC blew the chip design by only including > a 160byte threshold point given that PCI 2.0 spec says it should have > been 500bytes!! It wouldn't be the first thing DEC had screwed up in the design of these NICs. On the other hand, Intel has owned the silicon

Re: Request for comments: new `lpd' suite feature

< said: > How would this work with printers on local networks? > Say, a print server 192.168.1.73? I don't see that the IP address has any relevance. -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message

Request for comments: new `lpd' suite feature

Around here, we have a convention that each printer has a record in the DNS for printername.lpd-spooler which points to the print server for that printer. It occurred to me that, if there are no local printers, no additional information is needed for lpr and lpd to operate -- thus obviating the n

Re: /sys hierarchy

< said: > This must pass through -arch before any implementation. Remember, not > every committer reads current. Also remember, not every committer reads arch. -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message

Re: HEADS UP: /etc/rc.shutdown calls local scripts now

< said: > Sounds like a good enough reason to me to port the newer NetBSD LFS code > to FreeBSD. Or, even better, for someone to implement background fsck for soft updates. -GAWollman -- Garrett A. Wollman | O Siem / We are all family / O Siem / We're all the same [EMAIL PROTECTED] | O Siem

Large disks (was Re: bin/19635: add -c for grand total to df(1))

< said: > Maybe this isn't the right list to ask, but stepping into this: > I bought a 30G drive recently, and I was wondering if the 10% 'rule' > for performance is still really needed. I mean, I lose 3 _gigs_ of > storage space, and otherwise the performance detoriates? That > doesn't make sens

Re: cvs commit: src/sys/sys bus.h bus_private.h src/sys/kern subr_bus.c

< said: > In fact the best idea may be to make this a per device variable: > "if your hardware disappears, stay around" > or > "if your hardware disappears, go away" Since there needs to be a method call into the driver either way (to tell it that its hardware has disappeared), it c

Re: /sys hierarchy

< said: > encapsulation. Of course, someone more familiar with the actual code > in the tree might provide some better insight on the feasibility of > splitting these up. Don't, or else legions of network people will curse you to the end of your days. -GAWollman -- Garrett A. Wollman | O Si

cvs commit: ports/textproc/libxml2 Makefile ports/textproc/libxml2/files md5 ports/textproc/libxml2/pkg PLIST

< said: > Bring libxml2 2.1.1 into the fold after a repo-copy. This will > eventually replace libxml for GNOME. About a month ago, I was looking for a reasonable XML library with an eye towards bringing one into the tree (to be used for config files which have grown too complicated in syn

/sys hierarchy

< said: > Can somebody move thing around in sys? I mean put all fs code under let > say '/sys/fs' subdir. And all network protocols code under /sys/net > (or netproto)? Why? What benefit would that have? -GAWollman PS: Be careful how you answer this question. To Unsubscribe: send mail to

Re: HEADS UP: new fetch(1)

< said: > You're not one for constructive criticism, are you? I don't know how What part of YOU MAY NOT CLAIM COPYRIGHT ON MY TEXT don't you understand? -GAWollman To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message

<    1   2   3   4   5   6   >