documented Oops running big-endian reiserfs on parisc architecture

documented Oops running big-endian reiserfs on parisc architecture

[thunder7]

Well, I watched the number of files in /lost+found grow every time I
had to pull the plug from my HP C200, so I added a 'sync' option in
/etc/fstab and tried to get reiserfs working.

I patched my 2.4.9-pa13 kernel with

endian-safe-reiserfs-for-2.4.8.patch

which went without warnings or errors. A patched
reiserfs-progs-3.x.0k-pre9 also built well (despite what the configure
scripts says, you _need_ automake).

Creating a reiserfs-partition went without a hitch.

running bonnie -s 128 on it did this:

Adding Swap: 1041400k swap-space (priority -1)
eth0: Setting half-duplex based on MII#1 link partner capability of 0021.
reiserfs: checking transaction log (device 08:03) ...
Using r5 hash to sort names
ReiserFS version 3.6.25
bonnie[163]: Unaligned data reference 28

     YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
PSW: 00000000000001001111111100001110
r0-3     00000000 102ec550 10197d0c 26f24838
r4-7     26f24848 27054e40 00000011 00000020
r8-11    0000000a 26f06580 00000020 26f247a0
r12-15   27088d1c ffffffff 26f24838 00000001
r16-19   26f24708 faf02c68 0006a248 26f24852
r20-23   0000000f 00000003 03000000 00000002
r24-27   0000000a 0000000a 26f24848 102d4010
r28-31   26f24848 01000000 26f24b40 10196fc8
sr0-3    00000000 00000004 00000000 00000004
sr4-7    00000000 00000000 00000000 00000000

IASQ: 00000000 00000000 IAOQ: 10197d10 10197d14
 IIR: 0c7c1093    ISR: 00000000  IOR: 26f24846
 ORIG_R28: 00000000
pa8200:/var/log#

After reading in the archives about ksymoops, I looked up this part of
my System.map:

101979e0 T reiserfs_lookup
101979e0 t .L1432
10197ad4 t reiserfs_add_entry
10197e50 t .L1512
10197f28 t .L1510
10198068 T reiserfs_create

objdump -d /boot/linux-2.4.9-pa13 revealed:

10197cbc:       d6 d6 0b 18     depw,z r22,7,8,r22
10197cc0:       d6 b5 0a 10     depw,z r21,15,16,r21
10197cc4:       41 33 01 fe     ldb ff(sr0,r9),r19
10197cc8:       0a d5 02 75     or r21,r22,r21
10197ccc:       d6 94 09 08     depw,z r20,23,24,r20
10197cd0:       0a b4 02 74     or r20,r21,r20
10197cd4:       0a 93 02 73     or r19,r20,r19
10197cd8:       d2 75 18 f8     extrw,u r19,7,8,r21
10197cdc:       d2 74 19 f8     extrw,u r19,15,8,r20
10197ce0:       0c 73 12 16     stb  r19,b(sr0,r3)
10197ce4:       d2 73 1a f8     extrw,u r19,23,8,r19
10197ce8:       0c 75 12 10     stb  r21,8(sr0,r3)
10197cec:       0c 74 12 12     stb  r20,9(sr0,r3)
10197cf0:       e8 86 a9 e4     b,l 102a51e8 <memcpy>,%r2
10197cf4:       0c 73 12 14     stb  r19,a(sr0,r3)
10197cf8:       08 06 02 59     copy r6,r25
10197cfc:       08 04 02 5a     copy r4,r26
10197d00:       d7 20 1c 1d     depwi 0,31,3,r25
10197d04:       e8 0d a5 d8     call 101b1ff8 <padd_item>
10197d08:       08 08 02 58     copy r8,r24
10197d0c:       00 01 0e 74     rsm 1,r20
10197d10:       0c 7c 10 93     ldw  e(sr0,r3),r19
10197d14:       d6 60 1f 1f     depwi 0,7,1,r19
10197d18:       0c 73 12 9c     stw  r19,e(sr0,r3)
10197d1c:       00 14 18 60     mtsm r20
10197d20:       85 e0 26 32     cmpib,=,n 0,r15,10198040 <.L1510+0x118>
10197d24:       00 01 0e 74     rsm 1,r20
10197d28:       0c 7c 10 93     ldw  e(sr0,r3),r19
10197d2c:       d6 7f 1f 5f     depwi -1,5,1,r19
10197d30:       0c 73 12 9c     stw  r19,e(sr0,r3)
10197d34:       00 14 18 60     mtsm r20
10197d38:       08 00 02 54     copy r0,r20
10197d3c:       08 00 02 55     copy r0,r21
10197d40:       37 c6 3e 51     ldo -d8(sp),r6
10197d44:       37 cd 3d c1     ldo -120(sp),r13
10197d48:       6b d4 3e 51     stw r20,-d8(sr0,sp)
10197d4c:       6b d5 3e 59     stw r21,-d4(sr0,sp)
10197d50:       6b d4 3e 61     stw r20,-d0(sr0,sp)
10197d54:       6b d5 3e 69     stw r21,-cc(sr0,sp)
10197d58:       08 0c 02 59     copy r12,r25
10197d5c:       6b c6 3e 01     stw r6,-100(sr0,sp)
10197d60:       6b cd 3f 99     stw r13,-34(sr0,sp)
10197d64:       08 08 02 58     copy r8,r24
10197d68:       08 05 02 5a     copy r5,r26
10197d6c:       eb ff b4 4d     b,l 10197798 <reiserfs_find_entry>,%r2

which makes the error somewhere around here in 
fs/reiserfs/namei.c, function reiserfs_add_entry, after call to
padd_item, before call to reiserfs_find_entry:

    /* copy name */
    memcpy ((char *)(deh + 1), name, namelen);
    /* padd by 0s to the 4 byte boundary */
    padd_item ((char *)(deh + 1), ROUND_UP (namelen), namelen);

    /* entry is ready to be pasted into tree, set 'visibility' and 'stat data in entry' attributes */
    mark_de_without_sd (deh);
    visible ? mark_de_visible (deh) : mark_de_hidden (deh);

    /* find the proper place for the new entry */
    memset (bit_string, 0, sizeof (bit_string));
    de.de_gen_number_bit_string = (char *)bit_string;
    retval = reiserfs_find_entry (dir, name, namelen, &path, &de);


And there I'm stuck. I'm sending this to the parisc-list to document the
failure, and to linux-kernel to reach some reiserfs-experts, perhaps.

I'm willing to test anything!

Good luck,
Jurriaan
-- 
"I have seen her. That is enough, perhaps."
Pug smiled. "You are a rare man."
	Raymond E Feist - The King's Buccaneer
GNU/Linux 2.4.9-ac5 SMP/ReiserFS 2x1402 bogomips load av: 0.04 0.01 0.06

Re: [parisc-linux] documented Oops running big-endian reiserfs on parisc architecture

[Matthew Wilcox]

On Sun, Sep 02, 2001 at 10:55:38AM +0200, thunder7@xs4all.nl wrote:
> ReiserFS version 3.6.25
> bonnie[163]: Unaligned data reference 28

As it says, an unaligned data reference.

> r0-3     00000000 102ec550 10197d0c 26f24838

> IASQ: 00000000 00000000 IAOQ: 10197d10 10197d14

In kernel mode.

> 10197d10:       0c 7c 10 93     ldw  e(sr0,r3),r19

r3 is 26f24838, and offset `e' from that is unaligned.

> which makes the error somewhere around here in 
> fs/reiserfs/namei.c, function reiserfs_add_entry, after call to
> padd_item, before call to reiserfs_find_entry:
> 
>     /* copy name */
>     memcpy ((char *)(deh + 1), name, namelen);
>     /* padd by 0s to the 4 byte boundary */
>     padd_item ((char *)(deh + 1), ROUND_UP (namelen), namelen);
> 
>     /* entry is ready to be pasted into tree, set 'visibility' and 'stat data in entry' attributes */
>     mark_de_without_sd (deh);
>     visible ? mark_de_visible (deh) : mark_de_hidden (deh);
> 
>     /* find the proper place for the new entry */
>     memset (bit_string, 0, sizeof (bit_string));
>     de.de_gen_number_bit_string = (char *)bit_string;
>     retval = reiserfs_find_entry (dir, name, namelen, &path, &de);

I suspect mark_de_without_sd is an inlined function/macro and this will
be where the unaligned data reference is happening.

-- 
Revolutions do not require corporate support.

[SOLVED + PATCH]: documented Oops running big-endian reiserfs on parisc architecture

[thunder7]

On Sun, Sep 02, 2001 at 03:00:23PM +0100, Matthew Wilcox wrote:
> On Sun, Sep 02, 2001 at 10:55:38AM +0200, thunder7@xs4all.nl wrote:
> > ReiserFS version 3.6.25
> > bonnie[163]: Unaligned data reference 28
> 
> > which makes the error somewhere around here in 
> > fs/reiserfs/namei.c, function reiserfs_add_entry, after call to
> > padd_item, before call to reiserfs_find_entry:
> > 
> >     padd_item ((char *)(deh + 1), ROUND_UP (namelen), namelen);
> > 
> >     /* entry is ready to be pasted into tree, set 'visibility' and 'stat data in entry' attributes */
> >     mark_de_without_sd (deh);
> >     visible ? mark_de_visible (deh) : mark_de_hidden (deh);
> > 
> >     /* find the proper place for the new entry */
> >     memset (bit_string, 0, sizeof (bit_string));
> >     de.de_gen_number_bit_string = (char *)bit_string;
> >     retval = reiserfs_find_entry (dir, name, namelen, &path, &de);
> 
> I suspect mark_de_without_sd is an inlined function/macro and this will
> be where the unaligned data reference is happening.
> 
Correct. And the comments just above there about alignment are very
enlightening; it seems that the IBM/S390 architecture has some special
needs, and I just tested that my PA-RISC kernel has the same needs.
Thus I am able to present a real bugfix.

This patch allows me to run bonnie on a reiserfs partition with pa-risc
linux.

--- linux/include/linux/reiserfs_fs.h   Sun Sep  2 21:54:25 2001
+++ linux-new/include/linux/reiserfs_fs.h       Sun Sep  2 20:47:27 2001
@@ -924,7 +924,7 @@
 #define DEH_Visible 2

 /* 64 bit systems (and the S/390) need to be aligned explicitly -jdm */
-#if BITS_PER_LONG == 64 || defined(__s390__)
+#if BITS_PER_LONG == 64 || defined(__s390__) || defined(__hppa__)
 #   define ADDR_UNALIGNED_BITS  (3)
 #endif

This applies to linux-2.4.9-pa13 with
endian-safe-reiserfs-for-2.4.8.patch and to 2.4.9-ac5.

Please apply,
Jurriaan
-- 
It is well to remember, my son, that the entire population of the
universe, with one trifling exception, is composed of others.
        John Andrew Holmes
GNU/Linux 2.4.9-ac5 SMP/ReiserFS 2x1402 bogomips load av: 0.98 0.83 0.37

Re: [SOLVED + PATCH]: documented Oops running big-endian reiserfs on parisc architecture

[Richard Zidlicky]

On Sun, Sep 02, 2001 at 07:57:17PM +0200, thunder7@xs4all.nl wrote:
> 
> --- linux/include/linux/reiserfs_fs.h   Sun Sep  2 21:54:25 2001
> +++ linux-new/include/linux/reiserfs_fs.h       Sun Sep  2 20:47:27 2001
> @@ -924,7 +924,7 @@
>  #define DEH_Visible 2
> 
>  /* 64 bit systems (and the S/390) need to be aligned explicitly -jdm */
> -#if BITS_PER_LONG == 64 || defined(__s390__)
> +#if BITS_PER_LONG == 64 || defined(__s390__) || defined(__hppa__)
>  #   define ADDR_UNALIGNED_BITS  (3)
>  #endif

couldn't reiserfs use asm/unaligned.h like anyone else?
Seems at least sparc and mips may need the same treatment.

Bye
Richard

Re: [parisc-linux] documented Oops running big-endian reiserfs on parisc architecture

[David S. Miller]

   From: Matthew Wilcox <willy@debian.org>
   Date: Sun, 2 Sep 2001 15:00:23 +0100

   On Sun, Sep 02, 2001 at 10:55:38AM +0200, thunder7@xs4all.nl wrote:
   > ReiserFS version 3.6.25
   > bonnie[163]: Unaligned data reference 28
   
   As it says, an unaligned data reference.
   
BTW, you should not be OOPSing on this as unaligned references are
defined as completely normal, especially in the networking.

Is it impossible to handle unaligned access traps properly on
parisc?  If so, well you have some problems...

Later,
David S. Miller
davem@redhat.com

Re: [SOLVED + PATCH]: documented Oops running big-endian reiserfs on parisc architecture

[David S. Miller]

   From: Richard Zidlicky <Richard.Zidlicky@stud.informatik.uni-erlangen.de>
   Date: Mon, 3 Sep 2001 00:34:37 +0200

   On Sun, Sep 02, 2001 at 07:57:17PM +0200, thunder7@xs4all.nl wrote:
   >  /* 64 bit systems (and the S/390) need to be aligned explicitly -jdm */
   > -#if BITS_PER_LONG == 64 || defined(__s390__)
   > +#if BITS_PER_LONG == 64 || defined(__s390__) || defined(__hppa__)
   >  #   define ADDR_UNALIGNED_BITS  (3)
   >  #endif
   
   couldn't reiserfs use asm/unaligned.h like anyone else?
   Seems at least sparc and mips may need the same treatment.

Sparc will act correctly for unaliagned accesses.

It will trap and run very slowly, but it wont' OOPS and
it will give correct results.

This is actually required behavior, I don't know why parisc
is acting differently.

Later,
David S. Miller
davem@redhat.com

Re: [parisc-linux] documented Oops running big-endian reiserfs on parisc architecture

[Matthew Wilcox]

On Sun, Sep 02, 2001 at 04:04:41PM -0700, David S. Miller wrote:
>    From: Matthew Wilcox <willy@debian.org>
>    Date: Sun, 2 Sep 2001 15:00:23 +0100
> 
>    On Sun, Sep 02, 2001 at 10:55:38AM +0200, thunder7@xs4all.nl wrote:
>    > ReiserFS version 3.6.25
>    > bonnie[163]: Unaligned data reference 28
>    
>    As it says, an unaligned data reference.
>    
> BTW, you should not be OOPSing on this as unaligned references are
> defined as completely normal, especially in the networking.
> 
> Is it impossible to handle unaligned access traps properly on
> parisc?  If so, well you have some problems...

No, we just haven't bothered to implement it yet.  Not many people
use IPX these days.

-- 
Revolutions do not require corporate support.

Re: [parisc-linux] documented Oops running big-endian reiserfs on parisc architecture

[David S. Miller]

   From: Matthew Wilcox <willy@debian.org>
   Date: Mon, 3 Sep 2001 00:25:14 +0100
   
   No, we just haven't bothered to implement it yet.  Not many people
   use IPX these days.
   
IPX is not the only way this can happen.  Normal IPv4 can get
it with some ethernet cards on receive.

Later,
David S. Miller
davem@redhat.com

Re: [parisc-linux] Re: [SOLVED + PATCH]: documented Oops running big-endian reiserfs

[Alan Cox]

> It will trap and run very slowly, but it wont' OOPS and
> it will give correct results.
> 
> This is actually required behavior, I don't know why parisc
> is acting differently.

Several of the platforms only have trap fixups for a small subset of operations
that can legitimately occur on things like networking headers. It makes the
recovery code a lot smaller and cleaner.

Alan

Re: [parisc-linux] documented Oops running big-endian reiserfs on parisc architecture

[Alan Cox]

> > Is it impossible to handle unaligned access traps properly on
> > parisc?  If so, well you have some problems...
> 
> No, we just haven't bothered to implement it yet.  Not many people
> use IPX these days.

You also need unaligned trap fixups for

AX.25, NetROM, LAPB, X.25, Appletalk, PPP, Anything over 802.2LLC, Linus
NFS code for some NFS mount options (although not the -ac NFS code)

Alan

Re: documented Oops running big-endian reiserfs on parisc architecture

[Hans Reiser]

Well, I should have read the next email before saying these patches seemed to be
stable.

Jeff?

Hans

thunder7@xs4all.nl wrote:
> 
> Well, I watched the number of files in /lost+found grow every time I
> had to pull the plug from my HP C200, so I added a 'sync' option in
> /etc/fstab and tried to get reiserfs working.
> 
> I patched my 2.4.9-pa13 kernel with
> 
> endian-safe-reiserfs-for-2.4.8.patch
> 
> which went without warnings or errors. A patched
> reiserfs-progs-3.x.0k-pre9 also built well (despite what the configure
> scripts says, you _need_ automake).
> 
> Creating a reiserfs-partition went without a hitch.
> 
> running bonnie -s 128 on it did this:
> 
> Adding Swap: 1041400k swap-space (priority -1)
> eth0: Setting half-duplex based on MII#1 link partner capability of 0021.
> reiserfs: checking transaction log (device 08:03) ...
> Using r5 hash to sort names
> ReiserFS version 3.6.25
> bonnie[163]: Unaligned data reference 28
> 
>      YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
> PSW: 00000000000001001111111100001110
> r0-3     00000000 102ec550 10197d0c 26f24838
> r4-7     26f24848 27054e40 00000011 00000020
> r8-11    0000000a 26f06580 00000020 26f247a0
> r12-15   27088d1c ffffffff 26f24838 00000001
> r16-19   26f24708 faf02c68 0006a248 26f24852
> r20-23   0000000f 00000003 03000000 00000002
> r24-27   0000000a 0000000a 26f24848 102d4010
> r28-31   26f24848 01000000 26f24b40 10196fc8
> sr0-3    00000000 00000004 00000000 00000004
> sr4-7    00000000 00000000 00000000 00000000
> 
> IASQ: 00000000 00000000 IAOQ: 10197d10 10197d14
>  IIR: 0c7c1093    ISR: 00000000  IOR: 26f24846
>  ORIG_R28: 00000000
> pa8200:/var/log#
> 
> After reading in the archives about ksymoops, I looked up this part of
> my System.map:
> 
> 101979e0 T reiserfs_lookup
> 101979e0 t .L1432
> 10197ad4 t reiserfs_add_entry
> 10197e50 t .L1512
> 10197f28 t .L1510
> 10198068 T reiserfs_create
> 
> objdump -d /boot/linux-2.4.9-pa13 revealed:
> 
> 10197cbc:       d6 d6 0b 18     depw,z r22,7,8,r22
> 10197cc0:       d6 b5 0a 10     depw,z r21,15,16,r21
> 10197cc4:       41 33 01 fe     ldb ff(sr0,r9),r19
> 10197cc8:       0a d5 02 75     or r21,r22,r21
> 10197ccc:       d6 94 09 08     depw,z r20,23,24,r20
> 10197cd0:       0a b4 02 74     or r20,r21,r20
> 10197cd4:       0a 93 02 73     or r19,r20,r19
> 10197cd8:       d2 75 18 f8     extrw,u r19,7,8,r21
> 10197cdc:       d2 74 19 f8     extrw,u r19,15,8,r20
> 10197ce0:       0c 73 12 16     stb  r19,b(sr0,r3)
> 10197ce4:       d2 73 1a f8     extrw,u r19,23,8,r19
> 10197ce8:       0c 75 12 10     stb  r21,8(sr0,r3)
> 10197cec:       0c 74 12 12     stb  r20,9(sr0,r3)
> 10197cf0:       e8 86 a9 e4     b,l 102a51e8 <memcpy>,%r2
> 10197cf4:       0c 73 12 14     stb  r19,a(sr0,r3)
> 10197cf8:       08 06 02 59     copy r6,r25
> 10197cfc:       08 04 02 5a     copy r4,r26
> 10197d00:       d7 20 1c 1d     depwi 0,31,3,r25
> 10197d04:       e8 0d a5 d8     call 101b1ff8 <padd_item>
> 10197d08:       08 08 02 58     copy r8,r24
> 10197d0c:       00 01 0e 74     rsm 1,r20
> 10197d10:       0c 7c 10 93     ldw  e(sr0,r3),r19
> 10197d14:       d6 60 1f 1f     depwi 0,7,1,r19
> 10197d18:       0c 73 12 9c     stw  r19,e(sr0,r3)
> 10197d1c:       00 14 18 60     mtsm r20
> 10197d20:       85 e0 26 32     cmpib,=,n 0,r15,10198040 <.L1510+0x118>
> 10197d24:       00 01 0e 74     rsm 1,r20
> 10197d28:       0c 7c 10 93     ldw  e(sr0,r3),r19
> 10197d2c:       d6 7f 1f 5f     depwi -1,5,1,r19
> 10197d30:       0c 73 12 9c     stw  r19,e(sr0,r3)
> 10197d34:       00 14 18 60     mtsm r20
> 10197d38:       08 00 02 54     copy r0,r20
> 10197d3c:       08 00 02 55     copy r0,r21
> 10197d40:       37 c6 3e 51     ldo -d8(sp),r6
> 10197d44:       37 cd 3d c1     ldo -120(sp),r13
> 10197d48:       6b d4 3e 51     stw r20,-d8(sr0,sp)
> 10197d4c:       6b d5 3e 59     stw r21,-d4(sr0,sp)
> 10197d50:       6b d4 3e 61     stw r20,-d0(sr0,sp)
> 10197d54:       6b d5 3e 69     stw r21,-cc(sr0,sp)
> 10197d58:       08 0c 02 59     copy r12,r25
> 10197d5c:       6b c6 3e 01     stw r6,-100(sr0,sp)
> 10197d60:       6b cd 3f 99     stw r13,-34(sr0,sp)
> 10197d64:       08 08 02 58     copy r8,r24
> 10197d68:       08 05 02 5a     copy r5,r26
> 10197d6c:       eb ff b4 4d     b,l 10197798 <reiserfs_find_entry>,%r2
> 
> which makes the error somewhere around here in
> fs/reiserfs/namei.c, function reiserfs_add_entry, after call to
> padd_item, before call to reiserfs_find_entry:
> 
>     /* copy name */
>     memcpy ((char *)(deh + 1), name, namelen);
>     /* padd by 0s to the 4 byte boundary */
>     padd_item ((char *)(deh + 1), ROUND_UP (namelen), namelen);
> 
>     /* entry is ready to be pasted into tree, set 'visibility' and 'stat data in entry' attributes */
>     mark_de_without_sd (deh);
>     visible ? mark_de_visible (deh) : mark_de_hidden (deh);
> 
>     /* find the proper place for the new entry */
>     memset (bit_string, 0, sizeof (bit_string));
>     de.de_gen_number_bit_string = (char *)bit_string;
>     retval = reiserfs_find_entry (dir, name, namelen, &path, &de);
> 
> And there I'm stuck. I'm sending this to the parisc-list to document the
> failure, and to linux-kernel to reach some reiserfs-experts, perhaps.
> 
> I'm willing to test anything!
> 
> Good luck,
> Jurriaan
> --
> "I have seen her. That is enough, perhaps."
> Pug smiled. "You are a rare man."
>         Raymond E Feist - The King's Buccaneer
> GNU/Linux 2.4.9-ac5 SMP/ReiserFS 2x1402 bogomips load av: 0.04 0.01 0.06
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Re: documented Oops running big-endian reiserfs on parisc architecture

[thunder7]

On Mon, Sep 03, 2001 at 12:25:14AM +0100, Matthew Wilcox wrote:
> On Sun, Sep 02, 2001 at 04:04:41PM -0700, David S. Miller wrote:
> >    From: Matthew Wilcox <willy@debian.org>
> >    Date: Sun, 2 Sep 2001 15:00:23 +0100
> > 
> >    On Sun, Sep 02, 2001 at 10:55:38AM +0200, thunder7@xs4all.nl wrote:
> >    > ReiserFS version 3.6.25
> >    > bonnie[163]: Unaligned data reference 28
> >    
> >    As it says, an unaligned data reference.
> >    
> > BTW, you should not be OOPSing on this as unaligned references are
> > defined as completely normal, especially in the networking.
> > 
> > Is it impossible to handle unaligned access traps properly on
> > parisc?  If so, well you have some problems...
> 
> No, we just haven't bothered to implement it yet.  Not many people
> use IPX these days.
> 
OK, well this seems to have opened quite a can of worms. Programming the
handling of unaligned access traps on parisc is quite above my
possibilities.

I did notice that mark_de_without_sd() in reiserfs_fs.h just calls the
ext2_{set|clear|test}_bit routines. If the host architecture should
handle that with unaligned addresses, obviously the exception for s390
should also go. The current source would make reiserfs slow on sparc as
well, if I understand correctly.

To solve this for parisc, I see two possibilities:

- rewrite asm/parisc/bitops.h to have the ext2_* routines handle
  unaligned addresses. This would possibly be slowing down all aligned
  access. Not so nice.

- rewrite reiserfs_fs.h to use it's own test/set/clear bit routines.
  This would lose all the optimizations all specific architectures have
  in their asm/*/bitops.h. Also not nice.

Keeping in mind that I don't know any parisc assembly, is there any way
I can help resolve this in an elegant manner?

Greetings,
Jurriaan
-- 
If all else fails, immortality can always be assured by spectacular error.
        John Kenneth Galbraith
GNU/Linux 2.4.9-ac5 SMP/ReiserFS 2x1402 bogomips load av: 0.45 0.12 0.04

Re: [parisc-linux] Re: documented Oops running big-endian reiserfs on parisc architecture

[Matthew Wilcox]

On Mon, Sep 03, 2001 at 02:25:39PM +0400, Hans Reiser wrote:
> 
> 
> 
> Well, I should have read the next email before saying these patches seemed to be
> stable.
> 
> Jeff?

you misread the archive, Hans.  This is the original bugreport; once the
alignment was fixed it worked fine.

-- 
Revolutions do not require corporate support.

Re: [SOLVED + PATCH]: documented Oops running big-endian reiserfs on parisc architecture

[Jeff Mahoney]

On Mon, Sep 03, 2001 at 12:34:37AM +0200, Richard Zidlicky wrote:
> On Sun, Sep 02, 2001 at 07:57:17PM +0200, thunder7@xs4all.nl wrote:
> > 
> > --- linux/include/linux/reiserfs_fs.h   Sun Sep  2 21:54:25 2001
> > +++ linux-new/include/linux/reiserfs_fs.h       Sun Sep  2 20:47:27 2001
> > @@ -924,7 +924,7 @@
> >  #define DEH_Visible 2
> > 
> >  /* 64 bit systems (and the S/390) need to be aligned explicitly -jdm */
> > -#if BITS_PER_LONG == 64 || defined(__s390__)
> > +#if BITS_PER_LONG == 64 || defined(__s390__) || defined(__hppa__)
> >  #   define ADDR_UNALIGNED_BITS  (3)
> >  #endif
> 
> couldn't reiserfs use asm/unaligned.h like anyone else?
> Seems at least sparc and mips may need the same treatment.

    I'll be the first to admit that having the #if followed by arch defines
    really sucks, and that asm/unaligned.h would be much cleaner if for no
    other reason than avoiding an unnecessary gotcha for someone porting a new
    arch.

    Unfortunately, this doesn't appear to work in all cases. I know from
    testing that the S/390 case will fail using the {get,put)_unaligned macros,
    since they are identical to a normal, unassisted assignment.

    I did kick around the idea of making those macros the default accessors for
    the deh_state member (which is the only place they're used), but it unfairly
    penalizes arches that don't need them.

    I'm open to suggestions.

    -Jeff

-- 
Jeff Mahoney           |   "Bill Gates is a monocle and a Persian cat away
jeffm@suse.com         |    from being the villain in a James Bond movie."
jeffm@csh.rit.edu      |                   -- Dennis Miller