bitmap loading related reiserfs changes in 2.6.17-mm1 are broken

bitmap loading related reiserfs changes in 2.6.17-mm1 are broken

[Nick Orlov]

subj.

I've got a lot of BUG's during the boot and eventually box locked up.
SYS-RQ worked. Unfortunately none of these errors did make it to log files,
so I cannot provide the backtraces.

But reverting last 4 patches of reiserfs-changes series, namely

reiserfs-reorganize-bitmap-loading-functions.patch
reiserfs-on-demand-bitmap-loading.patch
reiserfs-on-demand-bitmap-loading-fix.patch
reiserfs-use-generic_file_open-for-open-checks.patch

fixed the problem for me.

I'm running 32 bit kernel on AMD64x2 w/ HIGHMEM,SMP,PREEMPT enabled

Hope you will find this report useful (sorry for the absence of the
backtrace though)
Please let me know if you want me to try some patches.

Thank you,
	Nick Orlov.

P.S. I'm on somewhat tough schedule during weekdays and I wont be able
to try the patches till tomorrow (Jun 24th) night.

-- 
With best wishes,
	Nick Orlov.

Re: bitmap loading related reiserfs changes in 2.6.17-mm1 are broken

[Andrew Morton]

On Wed, 21 Jun 2006 23:27:33 -0400
Nick Orlov <bugfixer@list.ru> wrote:

> subj.
> 
> I've got a lot of BUG's during the boot and eventually box locked up.
> SYS-RQ worked. Unfortunately none of these errors did make it to log files,
> so I cannot provide the backtraces.
> 
> But reverting last 4 patches of reiserfs-changes series, namely
> 
> reiserfs-reorganize-bitmap-loading-functions.patch
> reiserfs-on-demand-bitmap-loading.patch
> reiserfs-on-demand-bitmap-loading-fix.patch
> reiserfs-use-generic_file_open-for-open-checks.patch
> 
> fixed the problem for me.
> 

Yeah, sorry.  I've uploaded the below to the hot-fixes directory - it
should repair things.


Jeff, given its track record, I have to say that my confidence in this work
is nanoscopic.  Given that, and given the importance of reiserfs and given
the low rate of reiser3 development and given my ignorance of how reiserfs
works, I'm inclined to move very slowly on these patches.

It would really help if Chris or one of the namesys developers could take
the time to review and test these patches closely, please.



diff -puN fs/reiserfs/bitmap.c~reiserfs-reorganize-bitmap-loading-functions-fix fs/reiserfs/bitmap.c
--- a/fs/reiserfs/bitmap.c~reiserfs-reorganize-bitmap-loading-functions-fix
+++ a/fs/reiserfs/bitmap.c
@@ -1292,25 +1292,25 @@ void reiserfs_cache_bitmap_metadata(stru
                                     struct buffer_head *bh,
                                     struct reiserfs_bitmap_info *info)
 {
-	unsigned long *cur = (unsigned long *)bh->b_data;
-	int i;
+	unsigned long *cur = (unsigned long *)(bh->b_data + bh->b_size);
+
+	while (--cur >= (unsigned long *)bh->b_data) {
+		int base = ((char *)cur - bh->b_data) << 3;
 
-	for (i = sb->s_blocksize / sizeof (*cur); i > 0; i--, cur++) {
 		/* 0 and ~0 are special, we can optimize for them */
 		if (*cur == 0) {
-			info->first_zero_hint = i << 3;
-			info->free_count += sizeof (*cur) << 3;
+			info->first_zero_hint = base;
+			info->free_count += BITS_PER_LONG;
 		} else if (*cur != ~0L) {       /* A mix, investigate */
 			int b;
-			for (b = sizeof (*cur) << 3; b >= 0; b--) {
+			for (b = BITS_PER_LONG - 1; b >= 0; b--) {
 				if (!reiserfs_test_le_bit(b, cur)) {
-					info->first_zero_hint = (i << 3) + b;
+					info->first_zero_hint = base + b;
 					info->free_count++;
 				}
 			}
 		}
 	}
-
 	/* The first bit must ALWAYS be 1 */
 	BUG_ON(info->first_zero_hint == 0);
 }
_

Re: bitmap loading related reiserfs changes in 2.6.17-mm1 are broken

[Jeffrey Mahoney]

Andrew Morton wrote:
> On Wed, 21 Jun 2006 23:27:33 -0400
> Nick Orlov <bugfixer@list.ru> wrote:
> 
>> subj.
>>
>> I've got a lot of BUG's during the boot and eventually box locked up.
>> SYS-RQ worked. Unfortunately none of these errors did make it to log files,
>> so I cannot provide the backtraces.
>>
>> But reverting last 4 patches of reiserfs-changes series, namely
>>
>> reiserfs-reorganize-bitmap-loading-functions.patch
>> reiserfs-on-demand-bitmap-loading.patch
>> reiserfs-on-demand-bitmap-loading-fix.patch
>> reiserfs-use-generic_file_open-for-open-checks.patch
>>
>> fixed the problem for me.
>>
> 
> Yeah, sorry.  I've uploaded the below to the hot-fixes directory - it
> should repair things.
> 
> 
> Jeff, given its track record, I have to say that my confidence in this work
> is nanoscopic.  Given that, and given the importance of reiserfs and given
> the low rate of reiser3 development and given my ignorance of how reiserfs
> works, I'm inclined to move very slowly on these patches.
> 
> It would really help if Chris or one of the namesys developers could take
> the time to review and test these patches closely, please.

Understood. I'm more than a little embarrassed that these keep getting 
obvious errors that didn't appear in my own testing. I guess my only 
excuse is that these patches consistently get put on the back burner and 
get revisited when someone asks me about them. Unfortunately, sometimes 
it means I end up finding patches on my development nodes that have been 
obsoleted by versions that I've fixed already.

Ugh.

-Jeff

--
Jeff Mahoney
SUSE Labs

Re: bitmap loading related reiserfs changes in 2.6.17-mm1 are broken

[Barry K. Nathan]

On 6/21/06, Andrew Morton <akpm@osdl.org> wrote:
> Yeah, sorry.  I've uploaded the below to the hot-fixes directory - it
> should repair things.

FWIW, my main -mm testing system (with a reiserfs root filesystem)
blows up during boot with 2.6.17-mm1 + the hotfix. I haven't tried
plain 2.6.17-mm1 or 2.6.17-mm1 with the reiserfs patches reverted. The
last kernel this system ran was 2.6.17-rc6-mm2, with lockdep disabled,
with no problems whatsoever.

Unfortunately I still haven't gotten around to adding a serial console
to this box *and* I do not have access to my digital camera for
approx. 2 weeks. I guess I'll try to add a serial console ASAP. This
will probably take me a day or two however.
-- 
-Barry K. Nathan <barryn@pobox.com>

Re: bitmap loading related reiserfs changes in 2.6.17-mm1 are broken

[Barry K. Nathan]

[-- Attachment #1: Type: text/plain, Size: 422 bytes --]

On 6/21/06, Barry K. Nathan <barryn@pobox.com> wrote:
> Unfortunately I still haven't gotten around to adding a serial console
> to this box *and* I do not have access to my digital camera for
> approx. 2 weeks. I guess I'll try to add a serial console ASAP. This
> will probably take me a day or two however.

Never mind, it took a lot less time than that. Here's the backtrace...
-- 
-Barry K. Nathan <barryn@pobox.com>

[-- Attachment #2: capture2.txt --]
[-- Type: text/plain, Size: 15219 bytes --]

[    0.000000] Linux version 2.6.17-mm1nolockdep (barryn@nserv) (gcc version 4.1.1) #1 Thu Jun 22 00:36:48 PDT 2006
[    0.000000] BIOS-provided physical RAM map:
[    0.000000] sanitize start
[    0.000000] sanitize end
[    0.000000] copy_e820_map() start: 0000000000000000 size: 00000000000a0000 end: 00000000000a0000 type: 1
[    0.000000] copy_e820_map() type is E820_RAM
[    0.000000] add_memory_region(0000000000000000, 00000000000a0000, 1)
[    0.000000] copy_e820_map() start: 00000000000f0000 size: 0000000000010000 end: 0000000000100000 type: 2
[    0.000000] add_memory_region(00000000000f0000, 0000000000010000, 2)
[    0.000000] copy_e820_map() start: 0000000000100000 size: 000000000eef0000 end: 000000000eff0000 type: 1
[    0.000000] copy_e820_map() type is E820_RAM
[    0.000000] add_memory_region(0000000000100000, 000000000eef0000, 1)
[    0.000000] copy_e820_map() start: 000000000eff0000 size: 0000000000003000 end: 000000000eff3000 type: 4
[    0.000000] add_memory_region(000000000eff0000, 0000000000003000, 4)
[    0.000000] copy_e820_map() start: 000000000eff3000 size: 000000000000d000 end: 000000000f000000 type: 3
[    0.000000] add_memory_region(000000000eff3000, 000000000000d000, 3)
[    0.000000] copy_e820_map() start: 00000000ffff0000 size: 0000000000010000 end: 0000000100000000 type: 2
[    0.000000] add_memory_region(00000000ffff0000, 0000000000010000, 2)
[    0.000000]  BIOS-e820: 0000000000000000 - 00000000000a0000 (usable)
[    0.000000]  BIOS-e820: 00000000000f0000 - 0000000000100000 (reserved)
[    0.000000]  BIOS-e820: 0000000000100000 - 000000000eff0000 (usable)
[    0.000000]  BIOS-e820: 000000000eff0000 - 000000000eff3000 (ACPI NVS)
[    0.000000]  BIOS-e820: 000000000eff3000 - 000000000f000000 (ACPI data)
[    0.000000]  BIOS-e820: 00000000ffff0000 - 0000000100000000 (reserved)
[    0.000000] 239MB LOWMEM available.
[    0.000000] DMI 2.2 present.
[    0.000000] ACPI: PM-Timer IO Port: 0x408
[    0.000000] Allocating PCI resources starting at 10000000 (gap: 0f000000:f0ff0000)
[    0.000000] Detected 1399.443 MHz processor.
[   70.909638] Built 1 zonelists.  Total pages: 61424
[   70.909643] Kernel command line: BOOT_IMAGE=bzimage lapic root=/dev/sda1 vga=6 console=ttyS0,57600 console=tty0
[   70.910024] Local APIC disabled by BIOS -- reenabling.
[   70.910029] Found and enabled local APIC!
[   70.910038] Enabling fast FPU save and restore... done.
[   70.910041] Enabling unmasked SIMD FPU exception support... done.
[   70.910047] Initializing CPU#0
[   70.910114] CPU 0 irqstacks, hard=c03be000 soft=c03bd000
[   70.910120] PID hash table entries: 1024 (order: 10, 4096 bytes)
[   70.912324] Console: colour VGA+ 80x60
[   71.389932] Dentry cache hash table entries: 32768 (order: 5, 131072 bytes)
[   71.404154] Inode-cache hash table entries: 16384 (order: 4, 65536 bytes)
[   71.434455] Memory: 239984k/245696k available (1454k kernel code, 5160k reserved, 833k data, 156k init, 0k highmem)
[   71.455362] Checking if this processor honours the WP bit even in supervisor mode... Ok.
[   71.619372] Calibrating delay using timer specific routine.. 2800.22 BogoMIPS (lpj=14001119)
[   71.636399] Mount-cache hash table entries: 512
[   71.645671] CPU: L1 I cache: 16K, L1 D cache: 16K
[   71.655163] CPU: L2 cache: 512K
[   71.661498] Intel machine check architecture supported.
[   71.671980] Intel machine check reporting enabled on CPU#0.
[   71.683149] Compat vDSO mapped to ffffe000.
[   71.691551] CPU: Intel(R) Pentium(R) III CPU family      1400MHz stepping 01
[   71.705767] Checking 'hlt' instruction... OK.
[   71.750212] SMP alternatives: switching to UP code
[   71.759819] Freeing SMP alternatives: 0k freed
[   71.768730] ACPI: Core revision 20060608
[   71.782088] ACPI: setting ELCR to 0200 (from 1aa0)
[   71.981497] NET: Registered protocol family 16
[   71.990555] ACPI: bus type pci registered
[   72.001402] PCI: PCI BIOS revision 2.10 entry at 0xfb0c0, last bus=1
[   72.014133] Setting up standard PCI resources
[   72.033190] ACPI: Interpreter enabled
[   72.040558] ACPI: Using PIC for interrupt routing
[   72.050745] ACPI: PCI Root Bridge [PCI0] (0000:00)
[   72.069104] PCI quirk: region 0400-047f claimed by vt8235 PM
[   72.080471] PCI quirk: region 0500-050f claimed by vt8235 SMB
[   72.133255] ACPI: PCI Interrupt Link [LNKA] (IRQs 1 3 4 5 6 7 10 *11 12 14 15)
[   72.149049] ACPI: PCI Interrupt Link [LNKB] (IRQs 1 3 4 5 6 7 10 11 *12 14 15)
[   72.164810] ACPI: PCI Interrupt Link [LNKC] (IRQs 1 3 4 *5 6 7 10 11 12 14 15)
[   72.180582] ACPI: PCI Interrupt Link [LNKD] (IRQs 1 3 4 5 6 *7 10 11 12 14 15)
[   72.199156] Linux Plug and Play Support v0.97 (c) Adam Belay
[   72.210512] pnp: PnP ACPI init
[   72.221041] pnp: PnP ACPI: found 11 devices
[   72.229662] SCSI subsystem initialized
[   72.237245] PCI: Using ACPI for IRQ routing
[   72.245661] PCI: If a device doesn't work, try "pci=routeirq".  If it helps, post a report
[   72.311990] pnp: 00:02: ioport range 0x400-0x47f could not be reserved
[   72.325076] pnp: 00:02: ioport range 0x500-0x50f has been reserved
[   72.337833] PCI: Bridge: 0000:00:01.0
[   72.345189]   IO window: disabled.
[   72.352073]   MEM window: e4000000-e5ffffff
[   72.360505]   PREFETCH window: e0000000-e3ffffff
[   72.370031] NET: Registered protocol family 2
[   72.478540] IP route cache hash table entries: 2048 (order: 1, 8192 bytes)
[   72.492456] TCP established hash table entries: 8192 (order: 3, 32768 bytes)
[   72.506634] TCP bind hash table entries: 4096 (order: 2, 16384 bytes)
[   72.519571] TCP: Hash tables configured (established 8192 bind 4096)
[   72.532299] TCP reno registered
[   72.539661] Initializing Cryptographic API
[   72.547891] io scheduler noop registered
[   72.555832] io scheduler cfq registered (default)
[   72.565728] ACPI: Power Button (FF) [PWRF]
[   72.573965] ACPI: Power Button (CM) [PWRB]
[   72.585364] Linux agpgart interface v0.101 (c) Dave Jones
[   72.596315] agpgart: Detected VIA CLE266 chipset
[   72.607045] agpgart: AGP aperture is 4M @ 0xe7000000
[   72.617162] Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing disabled
[   72.633066] serial8250: ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
[   72.645512] Floppy drive(s): fd0 is 1.44M, fd1 is 1.44M
[   72.670162] FDC 0 is a post-1991 82077
[   72.679948] ACPI: PCI Interrupt Link [LNKA] enabled at IRQ 11
[   72.691474] ACPI: PCI Interrupt 0000:00:08.0[A] -> Link [LNKA] -> GSI 11 (level, low) -> IRQ 11
[   72.709030] sata_sil 0000:00:08.0: Applying R_ERR on DMA activate FIS errata fix
[   72.723965] ata1: SATA max UDMA/100 cmd 0xCF804080 ctl 0xCF80408A bmdma 0xCF804000 irq 11
[   72.740392] ata2: SATA max UDMA/100 cmd 0xCF8040C0 ctl 0xCF8040CA bmdma 0xCF804008 irq 11
[   72.756827] ata3: SATA max UDMA/100 cmd 0xCF804280 ctl 0xCF80428A bmdma 0xCF804200 irq 11
[   72.773263] ata4: SATA max UDMA/100 cmd 0xCF8042C0 ctl 0xCF8042CA bmdma 0xCF804208 irq 11
[   72.789666] scsi0 : sata_sil
[   73.277654] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
[   73.290920] ata1.00: ATA-7, max UDMA/133, 585940320 sectors: LBA48 
[   73.303477] ata1.00: applying bridge limits
[   73.312878] ata1.00: configured for UDMA/100
[   73.321453] scsi1 : sata_sil
[   73.807094] ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
[   73.820677] ata2.00: ATA-7, max UDMA/100, 781422768 sectors: LBA48 
[   73.833240] ata2.00: applying bridge limits
[   73.842975] ata2.00: configured for UDMA/100
[   73.851561] scsi2 : sata_sil
[   74.346527] ata3: SATA link up 1.5 Gbps (SStatus 113 SControl 310)
[   74.359798] ata3.00: ATA-6, max UDMA/100, 625142448 sectors: LBA48 
[   74.372350] ata3.00: applying bridge limits
[   74.381753] ata3.00: configured for UDMA/100
[   74.390326] scsi3 : sata_sil
[   74.716118] ata4: SATA link down (SStatus 0 SControl 310)
[   74.727111]   Vendor: ATA       Model: Maxtor 4A300J0    Rev: RAMB
[   74.741105]   Type:   Direct-Access                      ANSI SCSI revision: 05
[   74.756053]   Vendor: ATA       Model: ST3400832A        Rev: 3.01
[   74.770033]   Type:   Direct-Access                      ANSI SCSI revision: 05
[   74.784952]   Vendor: ATA       Model: WDC WD3200JB-00K  Rev: 08.0
[   74.798927]   Type:   Direct-Access                      ANSI SCSI revision: 05
[   74.814010] ACPI: PCI Interrupt 0000:00:11.1[A] -> Link [LNKA] -> GSI 11 (level, low) -> IRQ 11
[   74.831559] PCI: VIA IRQ fixup for 0000:00:11.1, from 255 to 11
[   74.843537] ata5: PATA max UDMA/133 cmd 0x1F0 ctl 0x3F6 bmdma 0xE400 irq 14
[   74.857501] scsi4 : pata_via
[   75.016563] ata5.00: ATA-7, max UDMA/133, 490234752 sectors: LBA48 
[   75.030066] ata5.00: configured for UDMA/133
[   75.038732]   Vendor: ATA       Model: Maxtor 6Y250P0    Rev: YAR4
[   75.052702]   Type:   Direct-Access                      ANSI SCSI revision: 05
[   75.067593] ata6: PATA max UDMA/133 cmd 0x170 ctl 0x376 bmdma 0xE408 irq 15
[   75.081548] scsi5 : pata_via
[   75.246438] ata6.00: ATA-6, max UDMA/100, 625142448 sectors: LBA48 
[   75.260036] ata6.00: configured for UDMA/100
[   75.268710]   Vendor: ATA       Model: WDC WD3200JB-00K  Rev: 08.0
[   75.282674]   Type:   Direct-Access                      ANSI SCSI revision: 05
[   75.297678] SCSI device sda: 585940320 512-byte hdwr sectors (300001 MB)
[   75.311128] sda: Write Protect is off
[   75.318516] SCSI device sda: drive cache: write back
[   75.328545] SCSI device sda: 585940320 512-byte hdwr sectors (300001 MB)
[   75.341991] sda: Write Protect is off
[   75.349380] SCSI device sda: drive cache: write back
[   75.359340]  sda: sda1
[   75.367974] sd 0:0:0:0: Attached scsi disk sda
[   75.376982] SCSI device sdb: 781422768 512-byte hdwr sectors (400088 MB)
[   75.390413] sdb: Write Protect is off
[   75.397805] SCSI device sdb: drive cache: write back
[   75.407812] SCSI device sdb: 781422768 512-byte hdwr sectors (400088 MB)
[   75.421246] sdb: Write Protect is off
[   75.428635] SCSI device sdb: drive cache: write back
[   75.438594]  sdb: unknown partition table
[   75.464089] sd 1:0:0:0: Attached scsi disk sdb
[   75.473084] SCSI device sdc: 625142448 512-byte hdwr sectors (320073 MB)
[   75.486529] sdc: Write Protect is off
[   75.493910] SCSI device sdc: drive cache: write back
[   75.503925] SCSI device sdc: 625142448 512-byte hdwr sectors (320073 MB)
[   75.517360] sdc: Write Protect is off
[   75.524744] SCSI device sdc: drive cache: write back
[   75.534712]  sdc: sdc1 sdc2
[   75.561786] sd 2:0:0:0: Attached scsi disk sdc
[   75.570779] SCSI device sdd: 490234752 512-byte hdwr sectors (251000 MB)
[   75.584220] sdd: Write Protect is off
[   75.591614] SCSI device sdd: drive cache: write back
[   75.601616] SCSI device sdd: 490234752 512-byte hdwr sectors (251000 MB)
[   75.615108] sdd: Write Protect is off
[   75.622496] SCSI device sdd: drive cache: write back
[   75.632453]  sdd: unknown partition table
[   75.658873] sd 4:0:0:0: Attached scsi disk sdd
[   75.667861] SCSI device sde: 625142448 512-byte hdwr sectors (320073 MB)
[   75.681304] sde: Write Protect is off
[   75.688698] SCSI device sde: drive cache: write back
[   75.698713] SCSI device sde: 625142448 512-byte hdwr sectors (320073 MB)
[   75.712156] sde: Write Protect is off
[   75.719546] SCSI device sde: drive cache: write back
[   75.729504]  sde: unknown partition table
[   75.755625] sd 5:0:0:0: Attached scsi disk sde
[   75.764637] PNP: PS/2 Controller [PNP0303:PS2K] at 0x60,0x64 irq 1
[   75.777027] PNP: PS/2 controller doesn't have AUX irq; using default 12
[   75.790438] serio: i8042 AUX port at 0x60,0x64 irq 12
[   75.800581] serio: i8042 KBD port at 0x60,0x64 irq 1
[   75.810774] mice: PS/2 mouse device common for all mice
[   75.821331] NET: Registered protocol family 1
[   75.830101] Testing NMI watchdog ... OK.
[   75.938071] Using IPI Shortcut mode
[   75.945152] ACPI: (supports S0 S1 S3 S4 S5)
[   75.954083] Freeing unused kernel memory: 156k freed
[   75.964047] Time: tsc clocksource has been installed.
[   75.974213] Write protecting the kernel read-only data: 298k
[   75.992335] ReiserFS: sda1: found reiserfs format "3.6" with standard journal
[   76.006670] ReiserFS: sda1: using ordered data mode
[   76.018727] ReiserFS: sda1: journal params: device sda1, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30
[   76.050463] ReiserFS: sda1: checking transaction log (sda1)
[   76.109979] input: AT Translated Set 2 keyboard as /class/input/input0
[   76.162999] ReiserFS: sda1: Using r5 hash to sort names
[   99.661389] ------------[ cut here ]------------
[   99.670669] kernel BUG at fs/reiserfs/bitmap.c:1307!
[   99.680622] invalid opcode: 0000 [#1]
[   99.687980] 4K_STACKS 
[   99.692793] last sysfs file: /block/sda/sda1/dev
[   99.702055] Modules linked in:
[   99.708239] CPU:    0
[   99.708241] EIP:    0060:[<c017ab78>]    Not tainted VLI
[   99.708243] EFLAGS: 00010202   (2.6.17-mm1nolockdep #1) 
[   99.734127] EIP is at reiserfs_cache_bitmap_metadata+0x78/0x85
[   99.745819] eax: 00000001   ebx: 00000000   ecx: c02ca4f4   edx: 00000001
[   99.759409] esi: cee10d24   edi: 00000000   ebp: ce808d70   esp: ce808d64
[   99.773000] ds: 007b   es: 007b   ss: 0068
[   99.781222] Process mount (pid: 338, ti=ce808000 task=cef9d560 task.ti=ce808000)
[   99.795677] Stack: cee10d24 cef6ea00 00000001 ce808d88 c017ac76 cf812000 cf812004 cef6ea00 
[   99.814530]        ceecf1c4 ce808d9c c017b211 00008180 fffffff4 ceecf210 ce808e84 c0185172 
[   99.831737]        ceecf210 ce808ea8 c017f465 cef6ea00 00c1ceff 00000001 00000000 00000000 
[   99.848912] Call Trace:
[   99.854631]  [<c017ac76>] reiserfs_read_bitmap_block+0xf1/0xfb
[   99.866663]  [<c017b211>] reiserfs_choose_packing+0x4d/0x7c
[   99.878177]  [<c0185172>] reiserfs_new_inode+0x90/0x609
[   99.889023]  [<c017ff3a>] reiserfs_create+0x71/0x12b
[   99.899334]  [<c01564ea>] vfs_create+0x67/0xad
[   99.908539]  [<c0158d86>] open_namei+0x181/0x636
[   99.918083]  [<c0149608>] do_filp_open+0x1f/0x35
[   99.927598]  [<c0149676>] do_sys_open+0x58/0xe3
[   99.936931]  [<c014972d>] sys_open+0x16/0x18
[   99.945741]  [<c026adf7>] syscall_call+0x7/0xb
[   99.954713] Code: 66 89 41 02 4b 83 fb ff 75 e2 8b 46 14 83 ea 04 39 c2 73 af 31 d2 b8 f4 a4 2c c0 66 83 39 00 0f 94 c2 e8 b4 62 03 00 85 c0 74 08 <0f> 0b 1b 05 f8 60 28 c0 5b 5e 5f 5d c3 55 89 e5 57 89 d7 56 89 
[   99.995769] EIP: [<c017ab78>] reiserfs_cache_bitmap_metadata+0x78/0x85 SS:ESP 0068:ce808d64
[  100.012596]  BUG: warning at kernel/exit.c:855/do_exit()
[  100.023494]  [<c0103079>] show_trace_log_lvl+0x54/0xfd
[  100.033885]  [<c010363d>] show_trace+0xd/0x10
[  100.042708]  [<c01036da>] dump_stack+0x19/0x1b
[  100.051694]  [<c01178a8>] do_exit+0x5f/0x8ab
[  100.060409]  [<c0103566>] die+0x218/0x220
[  100.068533]  [<c01035ea>] do_trap+0x7c/0x96
[  100.076997]  [<c0103cbb>] do_invalid_op+0x89/0x93
[  100.086501]  [<c0102c9d>] error_code+0x39/0x40

Re: bitmap loading related reiserfs changes in 2.6.17-mm1 are broken

[Hans Reiser]

Andrew Morton wrote:

>It would really help if Chris or one of the namesys developers could take
>the time to review and test these patches closely, please.
>  
>
Chris's code also has QA problems.  Because I don't control their
salaries, they both just basically ignore the QA process that the rest
of Namesys uses, and have exactly the results one would expect.  I no
longer encourage non-Namesys companies to pay our developers directly 
--- because of this experience.

That QA process is: everyone must get their patch reviewed and tested by
a second developer before sending it in.  Elena is not qualified to
review the code, but she is able to run a standard suite of tests (one
of which is gcc....) in addition to whatever special test is required
for the patch.   I review the design aspects of the patches, but leave
to others to find most of the coding errors.

Everyone once in a while someone strays from it, and I complain
privately about it.  The percentage of time that it is a mistake to
stray from it is remarkably high.....

Unfortunately, there is both real user demand for the
on-demand-bitmap-loading, and a proper qa of it is non-trivial.  I
propose that vs look at it next week.  vs, please ack.  Chris, if you
are available, please take a look as well.

Jeff has done a lot of good work in optimizing V3 bitmap related code,
and I would like to thank him for that.  It did a lot for our
performance.  Chris's work on the journaling code was also very
important and I am very grateful for it.  A better QA methodology could
have dramatically reduced the number of bugs in that code though.

Hans

Re: bitmap loading related reiserfs changes in 2.6.17-mm1 are broken

[Hans Reiser]

One of the things about my getting older is that I stopped thinking that
I would ever be anything but a fool, and started looking for ways to
cheat so that being a fool doesn't hurt so much. 

1) record all experiments/benchmarks in some form so that I can find
them later (because 6 months later when I am  thinking I should have
checked XYZ, I won't remember them well enough to be sure of them anymore).

2) have someone review all code and ideas before I give them to a large
audience

3) if some benchmark result doesn't make sense, I don't let go of it
until it does, because chances are high it is the only clue I will ever
get that something bigger than the obscure measurement is wrong in the
code and in my understanding.

There is a remarkable tendency that I have noticed, that the best
scientists in most fields are more quick to assume themselves to need
careful methodology than most others in the field.

I encourage you to figure these things out at a younger age than I
did.;-)  

Hans