NetBSD Problem Report #2675
From gnats Mon Aug 5 16:12:44 1996
Received: from Collatz.McRCIM.McGill.EDU by pain.lcs.mit.edu (8.7.5/8.7.3) with SMTP id PAA22148 for <gnats-bugs@gnats.netbsd.org>; Mon, 5 Aug 1996 15:53:45 -0400 (EDT)
Message-Id: <199608051953.PAA03580@Collatz.McRCIM.McGill.EDU>
Date: Mon, 5 Aug 1996 15:53:38 -0400
From: der Mouse <mouse@Collatz.McRCIM.McGill.EDU>
Reply-To: mouse@Collatz.McRCIM.McGill.EDU
To: gnats-bugs@NetBSD.ORG
Subject: [dM] 8 partitions is awfully few
X-Send-Pr-Version: 3.2
>Number: 2675
>Category: kern
>Synopsis: [dM] 8 partitions is awfully few
>Confidential: no
>Severity: non-critical
>Priority: low
>Responsible: port-sparc-maintainer
>State: open
>Class: change-request
>Submitter-Id: net
>Arrival-Date: Mon Aug 05 16:20:03 +0000 1996
>Closed-Date:
>Last-Modified: Mon Feb 09 09:11:58 +0000 2015
>Originator: der Mouse
>Release: 1.2
>Organization:
Dis-
>Environment:
Any SPARC (I use an IPC)
>Description:
Especially with the advent of modern disks with sizes up around
4G and 9G and the like, 8 partitions per disk seems awfully
few.
>How-To-Repeat:
Try to split up a big disk. Find yourself constrained by the
seven-partition limit (eight including c). Sigh.
>Fix:
Here's what I've done. It seems to work reasonably well. I've
tucked the "other" eight partitions in the spare space in the
disklabel between the ASCII label string and the RPM value,
with a magic number and a checksum. If either magic or
checksum is wrong, the "extended" eight partitions are all 0
offset 0 size.
The disksubr.c patch includes a fix to bounds_check_with_label;
I think I saw a commit message indicating this - or something
like it - went into the main branch, but it hasn't yet showed
up in the sups. I'm including it here on the theory that
anyone who is into fiddling with disk labeling to the point of
sixteen partitions probably also wants this fix. :-)
I'm not sure I've done all that's necessary to MAKEDEV; I'd
appreciate any remarks from those who know.
--- OLD/etc/etc.sparc/MAKEDEV Thu Jan 1 00:00:00 1970
+++ NEW/etc/etc.sparc/MAKEDEV Thu Jan 1 00:00:00 1970
@@ -167,27 +167,42 @@
vnd*) name=vnd; blk=8; chr=110;;
esac
case $unit in
- 0|1|2|3|4|5|6|7|8|9|10|11|12|13|14|15|16|\
- 17|18|19|20|21|22|23|24|25|26|27|28|29|30|31)
- rm -f ${name}${unit}[a-h] r${name}${unit}[a-h]
- mknod ${name}${unit}c b $blk `expr $unit '*' 8 + 2`
- mknod r${name}${unit}c c $chr `expr $unit '*' 8 + 2`
- mknod ${name}${unit}a b $blk `expr $unit '*' 8 + 0`
- mknod ${name}${unit}b b $blk `expr $unit '*' 8 + 1`
- mknod ${name}${unit}d b $blk `expr $unit '*' 8 + 3`
- mknod ${name}${unit}e b $blk `expr $unit '*' 8 + 4`
- mknod ${name}${unit}f b $blk `expr $unit '*' 8 + 5`
- mknod ${name}${unit}g b $blk `expr $unit '*' 8 + 6`
- mknod ${name}${unit}h b $blk `expr $unit '*' 8 + 7`
- mknod r${name}${unit}a c $chr `expr $unit '*' 8 + 0`
- mknod r${name}${unit}b c $chr `expr $unit '*' 8 + 1`
- mknod r${name}${unit}d c $chr `expr $unit '*' 8 + 3`
- mknod r${name}${unit}e c $chr `expr $unit '*' 8 + 4`
- mknod r${name}${unit}f c $chr `expr $unit '*' 8 + 5`
- mknod r${name}${unit}g c $chr `expr $unit '*' 8 + 6`
- mknod r${name}${unit}h c $chr `expr $unit '*' 8 + 7`
- chgrp operator ${name}${unit}[a-h] r${name}${unit}[a-h]
- chmod 640 ${name}${unit}[a-h] r${name}${unit}[a-h]
+ 0|1|2|3|4|5|6|7|8|9|10|11|12|13|14|15)
+ rm -f ${name}${unit}[a-p] r${name}${unit}[a-p]
+ mknod ${name}${unit}c b $blk `expr $unit '*' 16 + 2`
+ mknod r${name}${unit}c c $chr `expr $unit '*' 16 + 2`
+ mknod ${name}${unit}a b $blk `expr $unit '*' 16 + 0`
+ mknod ${name}${unit}b b $blk `expr $unit '*' 16 + 1`
+ mknod ${name}${unit}d b $blk `expr $unit '*' 16 + 3`
+ mknod ${name}${unit}e b $blk `expr $unit '*' 16 + 4`
+ mknod ${name}${unit}f b $blk `expr $unit '*' 16 + 5`
+ mknod ${name}${unit}g b $blk `expr $unit '*' 16 + 6`
+ mknod ${name}${unit}h b $blk `expr $unit '*' 16 + 7`
+ mknod ${name}${unit}i b $blk `expr $unit '*' 16 + 8`
+ mknod ${name}${unit}j b $blk `expr $unit '*' 16 + 9`
+ mknod ${name}${unit}k b $blk `expr $unit '*' 16 + 10`
+ mknod ${name}${unit}l b $blk `expr $unit '*' 16 + 11`
+ mknod ${name}${unit}m b $blk `expr $unit '*' 16 + 12`
+ mknod ${name}${unit}n b $blk `expr $unit '*' 16 + 13`
+ mknod ${name}${unit}o b $blk `expr $unit '*' 16 + 14`
+ mknod ${name}${unit}p b $blk `expr $unit '*' 16 + 15`
+ mknod r${name}${unit}a c $chr `expr $unit '*' 16 + 0`
+ mknod r${name}${unit}b c $chr `expr $unit '*' 16 + 1`
+ mknod r${name}${unit}d c $chr `expr $unit '*' 16 + 3`
+ mknod r${name}${unit}e c $chr `expr $unit '*' 16 + 4`
+ mknod r${name}${unit}f c $chr `expr $unit '*' 16 + 5`
+ mknod r${name}${unit}g c $chr `expr $unit '*' 16 + 6`
+ mknod r${name}${unit}h c $chr `expr $unit '*' 16 + 7`
+ mknod r${name}${unit}i c $chr `expr $unit '*' 16 + 8`
+ mknod r${name}${unit}j c $chr `expr $unit '*' 16 + 9`
+ mknod r${name}${unit}k c $chr `expr $unit '*' 16 + 10`
+ mknod r${name}${unit}l c $chr `expr $unit '*' 16 + 11`
+ mknod r${name}${unit}m c $chr `expr $unit '*' 16 + 12`
+ mknod r${name}${unit}n c $chr `expr $unit '*' 16 + 13`
+ mknod r${name}${unit}o c $chr `expr $unit '*' 16 + 14`
+ mknod r${name}${unit}p c $chr `expr $unit '*' 16 + 15`
+ chgrp operator ${name}${unit}[a-p] r${name}${unit}[a-p]
+ chmod 640 ${name}${unit}[a-p] r${name}${unit}[a-p]
;;
*)
echo bad unit for disk in: $i
--- OLD/sys/arch/sparc/include/disklabel.h Thu Jan 1 00:00:00 1970
+++ NEW/sys/arch/sparc/include/disklabel.h Thu Jan 1 00:00:00 1970
@@ -35,7 +35,7 @@
#define LABELSECTOR 0 /* sector containing label */
#define LABELOFFSET 64 /* offset of label in sector */
-#define MAXPARTITIONS 8 /* number of partitions */
+#define MAXPARTITIONS 16 /* number of partitions */
#define RAW_PART 2 /* raw partition: xx?c */
struct cpu_disklabel {
--- OLD/sys/arch/sparc/include/sun_disklabel.h Thu Jan 1 00:00:00 1970
+++ NEW/sys/arch/sparc/include/sun_disklabel.h Thu Jan 1 00:00:00 1970
@@ -88,7 +88,17 @@
struct sun_disklabel { /* total size = 512 bytes */
char sl_text[128];
- char sl_xxx1[292];
+#if MAXPARTITIONS > 8
+#define SUNXPART (MAXPARTITIONS-8)
+ u_long sl_xpsum; /* additive cksum, [sl_xpmag,sl_xxx1) */
+ u_long sl_xpmag; /* "extended" magic number */
+#define SL_XPMAG (0x199d1fe2+SUNXPART)
+ struct sun_dkpart sl_xpart[SUNXPART]; /* "extended" partitions, i through p */
+ char sl_xxx1[292-8-(8*SUNXPART)]; /* [292] including sl_x* */
+#else
+#define SUNXPART 0
+ char sl_xxx1[292];
+#endif
u_short sl_rpm; /* rotational speed */
u_short sl_pcylinders; /* number of physical cyls */
u_short sl_sparespercyl; /* spare sectors per cylinder */
--- OLD/sys/arch/sparc/sparc/disksubr.c Thu Jan 1 00:00:00 1970
+++ NEW/sys/arch/sparc/sparc/disksubr.c Thu Jan 1 00:00:00 1970
@@ -285,14 +285,16 @@
#define dkpart(dev) (minor(dev) & 7)
struct partition *p = lp->d_partitions + dkpart(bp->b_dev);
- int labelsect = lp->d_partitions[0].p_offset;
int maxsz = p->p_size;
int sz = (bp->b_bcount + DEV_BSIZE - 1) >> DEV_BSHIFT;
/* overwriting disk label ? */
/* XXX should also protect bootstrap in first 8K */
- if (bp->b_blkno + p->p_offset <= LABELSECTOR + labelsect &&
- (bp->b_flags & B_READ) == 0 && wlabel == 0) {
+ /* XXX this assumes everything <=LABELSECTOR is label! */
+ /* But since LABELSECTOR is 0, that's ok for now. */
+ if ( (bp->b_blkno + p->p_offset <= LABELSECTOR) &&
+ ((bp->b_flags & B_READ) == 0) &&
+ (wlabel == 0) ) {
bp->b_error = EROFS;
goto bad;
}
@@ -330,7 +332,7 @@
/* What partition types to assume for Sun disklabels: */
static u_char
-sun_fstypes[8] = {
+sun_fstypes[16] = {
FS_BSDFFS, /* a */
FS_SWAP, /* b */
FS_OTHER, /* c - whole disk */
@@ -339,8 +341,31 @@
FS_BSDFFS, /* f */
FS_BSDFFS, /* g */
FS_BSDFFS, /* h */
+ FS_BSDFFS, /* i */
+ FS_BSDFFS, /* j */
+ FS_BSDFFS, /* k */
+ FS_BSDFFS, /* l */
+ FS_BSDFFS, /* m */
+ FS_BSDFFS, /* n */
+ FS_BSDFFS, /* o */
+ FS_BSDFFS, /* p */
};
+/* Given a struct sun_disklabel, assume it has an extended partition
+ table and compute the correct value for sl_xpsum. */
+static u_long sun_extended_sum(struct sun_disklabel *sl)
+{
+ u_long lsum;
+ u_long *xp;
+ u_long *ep;
+
+ xp = (u_long *) &sl->sl_xpmag;
+ ep = (u_long *) &sl->sl_xxx1[0];
+ lsum = 0;
+ for (;xp<ep;xp++) lsum += *xp;
+ return(lsum);
+}
+
/*
* Given a SunOS disk label, set lp to a BSD disk label.
* Returns NULL on success, else an error string.
@@ -388,7 +413,7 @@
lp->d_rpm = sl->sl_rpm;
lp->d_interleave = sl->sl_interleave;
- lp->d_npartitions = 8;
+ lp->d_npartitions = MAXPARTITIONS;
/* These are as defined in <ufs/ffs/fs.h> */
lp->d_bbsize = 8192; /* XXX */
lp->d_sbsize = 8192; /* XXX */
@@ -414,6 +439,43 @@
}
}
+#if SUNXPART > 0
+
+ /* Clear "extended" partition info, tentatively */
+ for (i=0;i<SUNXPART;i++) {
+ npp = &lp->d_partitions[i+8];
+ npp->p_offset = 0;
+ npp->p_size = 0;
+ npp->p_fstype = FS_UNUSED;
+ }
+
+ /* Check to see if there's an "extended" partition table */
+ if (sl->sl_xpmag == SL_XPMAG) { /* probably... */
+ if (sun_extended_sum(sl) == sl->sl_xpsum) { /* ...yes! */
+ /* There is. Copy over the "extended" partitions.
+ This code parallels the loop for partitions a-h. */
+ /* XXX Abstract the common code? */
+ for (i=0;i<SUNXPART;i++) {
+ spp = &sl->sl_xpart[i];
+ npp = &lp->d_partitions[i+8];
+ npp->p_offset = spp->sdkp_cyloffset * secpercyl;
+ npp->p_size = spp->sdkp_nsectors;
+ if (npp->p_size == 0) {
+ npp->p_fstype = FS_UNUSED;
+ } else {
+ npp->p_fstype = sun_fstypes[i+8];
+ if (npp->p_fstype == FS_BSDFFS) {
+ npp->p_fsize = 1024;
+ npp->p_frag = 8;
+ npp->p_cpg = 16;
+ }
+ }
+ }
+ }
+ }
+
+#endif /* if SUNXPART > 0 */
+
lp->d_checksum = 0;
lp->d_checksum = dkcksum(lp);
return (NULL);
@@ -463,6 +525,39 @@
spp->sdkp_nsectors = npp->p_size;
}
sl->sl_magic = SUN_DKMAGIC;
+
+#if SUNXPART > 0
+
+ /* The reason we load the extended table stuff only conditionally
+ is so that a label that doesn't need it will have NULs there, like
+ a "traditional" Sun label. Since Suns seem to ignore everything
+ between sl_text and sl_rpm, this probably doesn't matter, but it
+ certainly doesn't hurt anything and it's easy to do. */
+
+ /* Do we need to load the extended table? */
+ for (i=0;i<SUNXPART;i++) {
+ if (lp->d_partitions[i+8].p_offset || lp->d_partitions[i+8].p_size) break;
+ }
+ if (i < SUNXPART) { /* we do */
+ sl->sl_xpmag = SL_XPMAG;
+ for (i=0;i<SUNXPART;i++) {
+ spp = &sl->sl_xpart[i];
+ npp = &lp->d_partitions[i+8];
+ if (npp->p_offset % secpercyl) return(EINVAL);
+ sl->sl_xpart[i].sdkp_cyloffset = npp->p_offset / secpercyl;
+ sl->sl_xpart[i].sdkp_nsectors = npp->p_size;
+ }
+ sl->sl_xpsum = sun_extended_sum(sl);
+ } else {
+ sl->sl_xpmag = 0;
+ for (i=0;i<SUNXPART;i++) {
+ sl->sl_xpart[i].sdkp_cyloffset = 0;
+ sl->sl_xpart[i].sdkp_nsectors = 0;
+ }
+ sl->sl_xpsum = 0;
+ }
+
+#endif /* if SUNXPART > 0 */
/* Correct the XOR check. */
sp1 = (u_short *)sl;
der Mouse
mouse@collatz.mcrcim.mcgill.edu
01 EE 31 F6 BB 0C 34 36 00 F3 7C 5A C1 A0 67 1D
>Release-Note:
>Audit-Trail:
From: der Mouse <mouse@Collatz.McRCIM.McGill.EDU>
To: gnats-bugs@NetBSD.ORG
Cc: netbsd-bugs@NetBSD.ORG
Subject: Re: kern/2675: [dM] 8 partitions is awfully few
Date: Mon, 5 Aug 1996 16:28:07 -0400
> Synopsis: [dM] 8 partitions is awfully few
> --- OLD/sys/arch/sparc/sparc/disksubr.c Thu Jan 1 00:00:00 1970
> +++ NEW/sys/arch/sparc/sparc/disksubr.c Thu Jan 1 00:00:00 1970
> @@ -285,14 +285,16 @@
> #define dkpart(dev) (minor(dev) & 7)
>
> struct partition *p = lp->d_partitions + dkpart(bp->b_dev);
> - int labelsect = lp->d_partitions[0].p_offset;
> int maxsz = p->p_size;
Duh. Fix that up to read
#define dkpart(dev) (((unsigned int)minor(dev)) % MAXPARTITIONS)
which is probably what it should have been anyway.
der Mouse
mouse@collatz.mcrcim.mcgill.edu
01 EE 31 F6 BB 0C 34 36 00 F3 7C 5A C1 A0 67 1D
State-Changed-From-To: open->feedback
State-Changed-By: jdolecek
State-Changed-When: Fri Jun 29 03:57:41 PDT 2001
State-Changed-Why:
Ports are being switched to higher number of partitions. Is this
still a problem for you?
From: Jaromír Dolecek <jdolecek@netbsd.org>
To: der Mouse <mouse@Rodents.Montreal.QC.CA>
Cc: gnats-bugs@netbsd.org
Subject: Re: kern/2675
Date: Fri, 6 Jul 2001 14:39:07 +0200 (CEST)
der Mouse wrote:
> The limit is still an _issue_ for me. It hasn't been a _problem_ for
> me since approximately the time I sent the PR; the fixes have been in
> my private patch tree for a long time.
Maybe, if you'd make necessary patches to make the change for sparc similar
way this was handled on i386 and submit the patch via the gnats to this
PR, the issue could be solved for you and rest of sparc users as well.
Jaromir
--
Jaromir Dolecek <jdolecek@NetBSD.org> http://www.ics.muni.cz/~dolecek/
NetBSD - just plain best OS! -=*=- Got spare MCA cards or docs? Hand me them!
From: Jaromír Dolecek <jdolecek@netbsd.org>
To: der Mouse <mouse@Rodents.Montreal.QC.CA>
Cc: gnats-bugs@netbsd.org
Subject: Re: kern/2675
Date: Sun, 15 Jul 2001 16:26:15 +0200 (CEST)
der Mouse wrote:
> I don't see how it isn't _already_ "solved for [me] and [the] rest of
> [the] sparc users" (or at least those who care to bother)...is there
> something wrong with the patches I provided? They sure work for me.
Well, I don't know whether they are good or wrong (haven't looked
on them in that much detail), but since the i386 solution has already
went through quite big discussion and was found out as acceptable,
I'd be reluctant to use different thing for other ports.
Unless the solution would be clearly superiour of course :)
Jaromir
--
Jaromir Dolecek <jdolecek@NetBSD.org> http://www.ics.muni.cz/~dolecek/
NetBSD - just plain best OS! -=*=- Got spare MCA cards or docs? Hand me them!
From: der Mouse <mouse@Rodents.Montreal.QC.CA>
To: Jaromír Dolecek <jdolecek@netbsd.org>
Cc: gnats-bugs@netbsd.org
Subject: Re: kern/2675
Date: Sat, 6 Oct 2001 15:36:41 -0400 (EDT)
>> I don't see how it isn't _already_ "solved for [me] and [the] rest
>> of [the] sparc users" (or at least those who care to bother)...is
>> there something wrong with the patches I provided? They sure work
>> for me.
> Well, I don't know whether they are good or wrong (haven't looked on
> them in that much detail), but since the i386 solution has already
> went through quite big discussion and was found out as acceptable,
> I'd be reluctant to use different thing for other ports.
Well, of *course* the i386 solution is best for all the other ports!
Let's port fdisk immediately!
More seriously, what *is* the i386 solution? As of my freeze point,
the i386 port is still at MAXPARTITIONS=8.
Not that it matters, I suppose. I've seen no evidence anyone is in
favor of what I suggested; most people couldn't give a damn and the few
that have responded seem opposed. (Like almost everything else I've
tried to do for NetBSD.) So there's probably no point in keeping the
PR open.
/~\ The ASCII der Mouse
\ / Ribbon Campaign
X Against HTML mouse@rodents.montreal.qc.ca
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
From: Jaromír Dolecek <jdolecek@netbsd.org>
To: der Mouse <mouse@Rodents.Montreal.QC.CA>
Cc: gnats-bugs@netbsd.org
Subject: Re: kern/2675
Date: Sun, 7 Oct 2001 21:04:18 +0200 (CEST)
der Mouse wrote:
> Well, of *course* the i386 solution is best for all the other ports!
> Let's port fdisk immediately!
Nono. It's just that that I'm reluctant to use something different
if solution on one arch is good and suitable for other archs too.
> More seriously, what *is* the i386 solution? As of my freeze point,
> the i386 port is still at MAXPARTITIONS=8.
See <i386/disklabel.h>. The i386 solution is to use 'big' minors for
partitions >8. Seems quite simple, does the job and maintains
backward compatibility well. I think it could be adopted by
other archs with only minor efford.
> Not that it matters, I suppose. I've seen no evidence anyone is in
> favor of what I suggested; most people couldn't give a damn and the few
> that have responded seem opposed. (Like almost everything else I've
> tried to do for NetBSD.) So there's probably no point in keeping the
> PR open.
Well, there is - sparc still has MAXPARTITIONS=8 :)
Jaromir
--
Jaromir Dolecek <jdolecek@NetBSD.org> http://www.ics.muni.cz/~dolecek/
NetBSD - just plain best OS! -=*=- Got spare MCA cards or docs? Hand me them!
From: der Mouse <mouse@Rodents.Montreal.QC.CA>
To: jdolecek@netbsd.org
Cc: gnats-bugs@netbsd.org
Subject: Re: kern/2675
Date: Sat, 5 Jan 2002 15:38:44 -0500 (EST)
> See <i386/disklabel.h>. The i386 solution is to use 'big' minors for
> partitions >8. Seems quite simple, does the job and maintains
> backward compatibility well.
Oh, that's a solution to only the minor-numbering issue; it's got
nothing to do with the rest of it, such as on-disk labels.
I think it's really ugly (not to mention comparatively expensive,
though I suppose converting between minor number and <unit,part> pairs
doesn't happen all that often). Having the mapping between <unit,part>
pairs and minor numbers be that grossly nonlinear for all future time
seems worse, to me, than one flag day in /dev. That way lies a system
that's little but an accretion of scar tissue upon scar tissue as each
new feature doesn't dare break backward compatability in the slightest.
But if that's is the way NetBSD has decided It Shall Be, to the point
of refusing a working, burnt-in addressing of the issue that takes a
different tack, then I guess it's that or nothing. So far it's been
nothing, and since it appears SPARC users are content with that, and
the issue is a non-problem for me, it's no skin off my nose either way.
I don't think it affects what I wrote except for the definition of
dkpart() in disksubr.c (see the addon I sent just after the patches)
and the corresponding MAKEDEV changes.
/~\ The ASCII der Mouse
\ / Ribbon Campaign
X Against HTML mouse@rodents.montreal.qc.ca
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
State-Changed-From-To: feedback->open
State-Changed-By: fair
State-Changed-When: Sun Jan 6 14:28:34 PST 2002
State-Changed-Why:
Feedback has been provided, and we should answer it properly.
From: der Mouse <mouse@Rodents.Montreal.QC.CA>
To: gnats-bugs@netbsd.org
Cc:
Subject: Re: kern/2675
Date: Sun, 6 Jan 2002 19:21:30 -0500 (EST)
> Feedback has been provided, and we should answer it properly.
Arguably, I'm also the one who should be answering it, as I'm
theoretically here as a PR cruncher. But ISTM there is insufficient
consensus that any part of the patchset in the PR is a Right enough
Thing to commit. :-)
In particular, ISTR seeing it said that /sparc -current hides a
full-blown native NetBSD disklabel in the spare space I put the
extended partitions in. If this is true (it could be just confused
memory), there is insufficient room to expand it to 16 partitions. (By
my count, an N-partition NetBSD disklabel occupies 148+(16*N) bytes.
This means that the most partitions that can be fit into that 292-byte
unused area by this method is 9. By contrast, there's space for 35
"extended" partitions, 43 total, the way I did it (and four bytes left
over) - but those partitions have Sun-disklabel constraints. This is
more than MAXMAXPARTITIONS, but given the reason MAXMAXPARTITIONS is as
low as it is, it's not clear that its limit should apply to a port that
doesn't always have a full-blown NetBSD disklabel on-disk. What, if
anything, is MAXMAXPARTITIONS used for, anyway?)
At any rate, if the decision is to go with a relatively high partition
count by extending the Sun label, my patches may at least be a useful
starting point. (But more likely from historical evidence, I'd say, is
that people will argue about what the right thing is and end up doing
nothing because the argument is never settled. Sigh.)
/~\ The ASCII der Mouse
\ / Ribbon Campaign
X Against HTML mouse@rodents.montreal.qc.ca
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
From: matthew green <mrg@eterna.com.au>
To: gnats-bugs@netbsd.org
Cc:
Subject: re: kern/2675
Date: Sat, 03 Apr 2004 02:49:22 +1000
BTW, when fixing this problem for sparc one should be aware of PR#20277.
State-Changed-From-To: open->closed
State-Changed-By: mycroft
State-Changed-When: Tue Jul 6 06:17:38 UTC 2004
State-Changed-Why:
This was increased to 16 a long time ago.
Responsible-Changed-From-To: kern-bug-people->port-sparc-maintainer
Responsible-Changed-By: jnemeth@netbsd.org
Responsible-Changed-When: Wed, 19 Jul 2006 02:59:14 +0000
Responsible-Changed-Why:
This is sparc specific.
State-Changed-From-To: closed->open
State-Changed-By: jnemeth@netbsd.org
State-Changed-When: Wed, 19 Jul 2006 02:59:14 +0000
State-Changed-Why:
The problem has only been solved for i386/amd64 and not generally.
State-Changed-From-To: open->feedback
State-Changed-By: dholland@NetBSD.org
State-Changed-When: Mon, 09 Feb 2015 02:10:38 +0000
State-Changed-Why:
are GPT tables an adequate solution?
State-Changed-From-To: feedback->open
State-Changed-By: dholland@NetBSD.org
State-Changed-When: Mon, 09 Feb 2015 09:11:58 +0000
State-Changed-Why:
Not for boot disks.
>Unformatted:
(Contact us)
$NetBSD: query-full-pr,v 1.39 2013/11/01 18:47:49 spz Exp $
$NetBSD: gnats_config.sh,v 1.8 2006/05/07 09:23:38 tsutsui Exp $
Copyright © 1994-2014
The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.