[PATCH] Make drivers/media compile as modules
Hi Linus, The Makefile changes broke compiling drivers/media, such as bttv, as kernel modules. Below is the patch against test13-pre3 to fix it. Please apply. -Udo. --- /sources/linux/drivers/media/Makefile Thu Dec 21 08:17:17 2000 +++ /usr/src/linux/drivers/media/Makefile Thu Dec 21 08:15:55 2000 @@ -10,8 +10,10 @@ # subdir-y := video radio +subdir-m := video radio O_TARGET := media.o obj-y:= $(join $(subdir-y),$(subdir-y:%=/%.o)) +obj-m:= $(join $(subdir-m),$(subdir-m:%=/%.o)) include $(TOPDIR)/Rules.make - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
[PATCH] swap write clustering
Hi, Basically this new swap_writepage function looks for dirty swapcache pages which may be contiguous (reverse and forward searching wrt to the physical address of the page being passed to swap_writepage) and builds a page list which is written "at once". The patch is against test13pre3. Comments are welcome. (especially about the __find_page_nolock modification) diff -Nur --exclude-from=exclude linux.orig/include/linux/mm.h linux/include/linux/mm.h --- linux.orig/include/linux/mm.h Thu Dec 21 04:34:19 2000 +++ linux/include/linux/mm.hThu Dec 21 03:56:30 2000 @@ -451,6 +451,8 @@ extern int filemap_swapout(struct page *, struct file *); extern int filemap_sync(struct vm_area_struct *, unsigned long,size_t, unsigned int); extern struct page *filemap_nopage(struct vm_area_struct *, unsigned long, int); +extern inline struct page * ___find_page_nolock(struct address_space *, unsigned +long, struct page *); + /* * GFP bitmasks.. diff -Nur --exclude-from=exclude linux.orig/include/linux/swap.h linux/include/linux/swap.h --- linux.orig/include/linux/swap.h Thu Dec 21 04:34:19 2000 +++ linux/include/linux/swap.h Thu Dec 21 04:16:42 2000 @@ -166,6 +166,8 @@ extern unsigned long swap_cache_find_total; extern unsigned long swap_cache_find_success; #endif + +extern struct swap_info_struct swap_info[MAX_SWAPFILES]; /* * Work out if there are any other processes sharing this page, ignoring diff -Nur --exclude-from=exclude linux.orig/mm/filemap.c linux/mm/filemap.c --- linux.orig/mm/filemap.c Thu Dec 21 04:34:20 2000 +++ linux/mm/filemap.c Thu Dec 21 03:53:41 2000 @@ -242,7 +242,7 @@ spin_unlock(&pagecache_lock); } -static inline struct page * __find_page_nolock(struct address_space *mapping, unsigned long offset, struct page *page) +inline struct page * ___find_page_nolock(struct address_space *mapping, unsigned long +offset, struct page *page) { goto inside; @@ -250,12 +250,22 @@ page = page->next_hash; inside: if (!page) - goto not_found; + return NULL; if (page->mapping != mapping) continue; if (page->index == offset) break; } + return page; +} + +static inline struct page * __find_page_nolock(struct address_space *mapping, +unsigned long offset, struct page *page) +{ + page = ___find_page_nolock(mapping, offset, page); + + if(!page) + return NULL; + /* * Touching the page may move it to the active list. * If we end up with too few inactive pages, we wake @@ -264,7 +274,7 @@ age_page_up(page); if (inactive_shortage() > inactive_target / 2 && free_shortage()) wakeup_kswapd(0); -not_found: + return page; } diff -Nur --exclude-from=exclude linux.orig/mm/swap_state.c linux/mm/swap_state.c --- linux.orig/mm/swap_state.c Mon Dec 4 19:22:02 2000 +++ linux/mm/swap_state.c Thu Dec 21 04:23:47 2000 @@ -5,6 +5,8 @@ * Swap reorganised 29.12.95, Stephen Tweedie * * Rewritten to use page cache, (C) 1998 Stephen Tweedie + * + * 21/12/2000 Added swap write clustering. Marcelo Tosatti */ #include @@ -17,9 +19,95 @@ #include +static inline struct page * swap_page_dirty(unsigned long, unsigned long, struct +swap_info_struct *); + +#define SWAP_WRITE_CLUSTER (1 << page_cluster) + static int swap_writepage(struct page *page) { - rw_swap_page(WRITE, page, 0); + unsigned long page_offset, curr, offset, i, type; + struct swap_info_struct *swap; + swp_entry_t entry; + struct page *cpages[SWAP_WRITE_CLUSTER*2]; + int count, first; + + entry.val = page->index; + + type = SWP_TYPE(entry); + + swap = &swap_info[type]; + + /* If swap area is not a real device, do not try to write cluster. */ + if(!swap->swap_device) { + rw_swap_page(WRITE, page, 0); + return 0; + } + + page_offset = offset = SWP_OFFSET(entry); + cpages[SWAP_WRITE_CLUSTER] = page; + count = 1; + first = SWAP_WRITE_CLUSTER; + curr = 1; + + spin_lock(&pagecache_lock); + swap_device_lock(swap); + + /* +* Search for clusterable dirty swap pages. +*/ + + while (count < SWAP_WRITE_CLUSTER) { + struct page *p = NULL; + + if(offset <= 0) + break; + + offset = page_offset - curr; + p = swap_page_dirty(offset, type, swap); + + if(!p) + break; + + cpages[--first] = p; + + ClearPageDirty(p); + curr++; + count++; + } + + curr = 1; + + while (count < SWAP_WRITE_CLUSTER) { + struct page *p = NULL; + offset = page_offset + cur
Re: iptables: "stateful inspection?"
On Wed, 20 Dec 2000, Michael Rothwell wrote: >"Michael H. Warfield" wrote: >> I think that's more than a little overstatement on your >> part. It depends entirely on the application you intend to put >> it to. > >Fine. How do I make FTP work through it? How can I allow all outgoing >TCP connections without opening the network to inbound connections on >the ports of desired services? /etc/sysctl.conf: # Set local port range to be higher. net.ipv4.ip_local_port_range = 32768 33792 /etc/ftpaccess: passive ports 0.0.0.0/0 32768 36863 Firewall script: - STDPORT=32768:33792 IP=1.2.3.4/32 # Client FTP ipchains -A output -j ACCEPT -p tcp -s $IP $STDPORT -d 0.0.0.0/0 ftp-data -y -l ipchains -A output -j ACCEPT -p tcp -s $IP $STDPORT -d 0.0.0.0/0 ftp-data ipchains -A output -j ACCEPT -p tcp -s $IP $STDPORT -d 0.0.0.0/0 ftp -y -l ipchains -A output -j ACCEPT -p tcp -s $IP $STDPORT -d 0.0.0.0/0 ftp # Server FTP ipchains -A input -j ACCEPT -p tcp -s 0.0.0.0/0 ftp-data -d $IP $STDPORT # Needs SYN ipchains -A input -j ACCEPT -p tcp -s 0.0.0.0/0 ftp -d $IP $STDPORT ! -y [now deny all for all chains] Unfortunately, any FTP server that doesn't use port 20 for data streams won't work in Passive mode (oh well). So I just download elsewhere first and then get it locally for browsers that insist upon Passive. For allowing outgoing connections without inbound, you'd use: ipchains -A input -j DENY -p tcp -y or if that complains: ipchains -A input -j DENY -p tcp -s 0.0.0.0/0 -d $IP -y You'll notice above I used '! -y' on the Server FTP rule. If I missed a detail, it might be due to trying to condense everything I have into what you wanted. -George Greer (7,323 and 189 lines in my firewall rule script.) - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Strange warnings about .modinfo when compiling 2.2.18 on Alpha
On Tue, 19 Dec 2000 15:20:17 GMT, Jesper Juhl <[EMAIL PROTECTED]> wrote: >I just compiled 2.2.18 for my AlphaServer 400 4/233, and noticed a lot of >messages like the following during the compile, they all contain the >'Ignoring changed section attributes for .modinfo' part: The way .modinfo is created is a kludge to prevent the .modinfo section being loaded as part of the module. The initial reference to .modinfo creates it as non-allocated, later references try to allocate data in the section. Older versions of gcc silently ignored the mismatch, newer ones warn about the mismatch. modutils >= 2.3.19 makes sure that .modinfo is not loaded so the kernel kludge is no longer required. Alan Cox (quite rightly) will not force 2.2 users to upgrade modutils, but if you jump to modutils 2.3.23 and apply this patch against kernel 2.2.18 then the warnings will disappear. Index: 18.1/include/linux/module.h --- 18.1/include/linux/module.h Tue, 12 Sep 2000 13:37:17 +1100 kaos (linux-2.2/F/51_module.h 1.1.7.2 644) +++ 18.1(w)/include/linux/module.h Thu, 21 Dec 2000 17:55:23 +1100 kaos +(linux-2.2/F/51_module.h 1.1.7.2 644) @@ -190,11 +190,6 @@ const char __module_parm_desc_##var[] \ __attribute__((section(".modinfo"))) = \ "parm_desc_" __MODULE_STRING(var) "=" desc -/* The attributes of a section are set the first time the section is - seen; we want .modinfo to not be allocated. */ - -__asm__(".section .modinfo\n\t.previous"); - /* Define the module variable, and usage macros. */ extern struct module __this_module; - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: usb + smp + 2.4.0test = pci irq routing problem?
On Wed, Dec 20, 2000 at 02:06:12PM -0500, Pete Toscano wrote: > hello, > > i've been working with johannes erdfelt in fixing a problem with usb > on my machine. it's a dual pentium 3 system on a tyan tiger 133 mobo > (via apollo pro 133a chipset). basically, usb works when i don't > enable smp or when i disable apic on smp-enabled kernels. he believes > that we're seeing a pci irq routing problem and that i should contact > martin mares about this problem. (i've written him a couple times, > but have heard nothing, so i figure he's either away, busy, or whatnot > and i thought i'd try lkml for help.) I have this same exact motherboard (graciously donated by someone for me to try to help solve this problem) and the same exact problem. I don't have any shared interrupts, but the USB subsystem is not getting any interrupts through to it. Attached is my kernel startup log with DEBUG enabled in pci.c. This is for 2.4.0-test12 as I haven't seen any pci changes in the test13-pre series yet. 2.2.18 also has the same problem. If anyone needs any other information, or can suggest anything else, please let me know. thanks, greg k-h -- greg@(kroah|wirex).com Linux version 2.4.0-test12 (greg@duel) (gcc version egcs-2.91.66-StackGuard 19990314/Linux (egcs-1.1.2 release)) #6 SMP Wed Dec 20 16:20:46 EST 2000 BIOS-provided physical RAM map: BIOS-e820: 0009fc00 @ (usable) BIOS-e820: 0400 @ 0009fc00 (reserved) BIOS-e820: 0001 @ 000f (reserved) BIOS-e820: 1000 @ fec0 (reserved) BIOS-e820: 1000 @ fee0 (reserved) BIOS-e820: 0001 @ (reserved) BIOS-e820: 07ef @ 0010 (usable) BIOS-e820: d000 @ 07ff3000 (ACPI data) BIOS-e820: 3000 @ 07ff (ACPI NVS) Scan SMP from c000 for 1024 bytes. Scan SMP from c009fc00 for 1024 bytes. Scan SMP from c00f for 65536 bytes. found SMP MP-table at 000f5940 hm, page 000f5000 reserved twice. hm, page 000f6000 reserved twice. hm, page 000f1000 reserved twice. hm, page 000f2000 reserved twice. On node 0 totalpages: 32752 zone(0): 4096 pages. zone(1): 28656 pages. zone(2): 0 pages. Intel MultiProcessor Specification v1.1 Virtual Wire compatibility mode. OEM ID: OEM0 Product ID: PROD APIC at: 0xFEE0 Processor #0 Pentium(tm) Pro APIC version 17 Floating point unit present. Machine Exception supported. 64 bit compare & exchange supported. Internal APIC present. SEP present. MTRR present. PGE present. MCA present. CMOV present. Bootup CPU Processor #1 Pentium(tm) Pro APIC version 17 Floating point unit present. Machine Exception supported. 64 bit compare & exchange supported. Internal APIC present. SEP present. MTRR present. PGE present. MCA present. CMOV present. Bus #0 is PCI Bus #1 is PCI Bus #2 is ISA I/O APIC #2 Version 17 at 0xFEC0. Int: type 3, pol 0, trig 0, bus 2, IRQ 00, APIC ID 2, APIC INT 00 Int: type 0, pol 0, trig 0, bus 2, IRQ 01, APIC ID 2, APIC INT 01 Int: type 0, pol 0, trig 0, bus 2, IRQ 00, APIC ID 2, APIC INT 02 Int: type 0, pol 0, trig 0, bus 2, IRQ 03, APIC ID 2, APIC INT 03 Int: type 0, pol 0, trig 0, bus 2, IRQ 04, APIC ID 2, APIC INT 04 Int: type 0, pol 0, trig 0, bus 2, IRQ 06, APIC ID 2, APIC INT 06 Int: type 0, pol 0, trig 0, bus 2, IRQ 07, APIC ID 2, APIC INT 07 Int: type 0, pol 1, trig 1, bus 2, IRQ 08, APIC ID 2, APIC INT 08 Int: type 0, pol 0, trig 0, bus 2, IRQ 09, APIC ID 2, APIC INT 09 Int: type 0, pol 0, trig 0, bus 2, IRQ 0c, APIC ID 2, APIC INT 0c Int: type 0, pol 0, trig 0, bus 2, IRQ 0d, APIC ID 2, APIC INT 0d Int: type 0, pol 0, trig 0, bus 2, IRQ 0e, APIC ID 2, APIC INT 0e Int: type 0, pol 0, trig 0, bus 2, IRQ 0f, APIC ID 2, APIC INT 0f Int: type 0, pol 3, trig 3, bus 2, IRQ 0a, APIC ID 2, APIC INT 10 Int: type 0, pol 3, trig 3, bus 2, IRQ 0b, APIC ID 2, APIC INT 11 Int: type 0, pol 3, trig 3, bus 2, IRQ 05, APIC ID 2, APIC INT 13 Lint: type 3, pol 0, trig 0, bus 2, IRQ 00, APIC ID ff, APIC LINT 00 Lint: type 1, pol 0, trig 0, bus 2, IRQ 00, APIC ID ff, APIC LINT 01 Processors: 2 mapped APIC to e000 (fee0) mapped IOAPIC to d000 (fec0) Kernel command line: auto BOOT_IMAGE=greg ro root=305 BOOT_FILE=/boot/bzImage-2.4.0-test12 Initializing CPU#0 Detected 533.167 MHz processor. Console: colour dummy device 80x25 Calibrating delay loop... 1061.68 BogoMIPS Memory: 126560k/131008k available (1246k kernel code, 4060k reserved, 91k data, 216k init, 0k highmem) Dentry-cache hash table entries: 16384 (order: 5, 131072 bytes) Buffer-cache hash table entries: 4096 (order: 2, 16384 bytes) Page-cache hash table entries: 32768 (order: 5, 131072 bytes) Inode-cache hash table entries: 8192 (order: 4, 65536 bytes) CPU: Before vendor init, caps: 0387fbff , vendor = 0 CPU: L1 I cache: 16K, L1 D
Bug: 2.4.0-test12 w/ PCMCIA on ThinkPad: KERNEL: assertion(dev->ip_ptr==NULL)failed at dev.c(2422):netdev_finish_unregister
2.4.0-test12 compiled on an IBM ThinkPad 600 51U (Pentium II) with PCMCIA support. Same behavior with Linus PCMCIA and Hinds PCMCIA. I have a Xircom modem/ethernet card which works correctly using the serial_cs, xirc2ps_cs, ds, i82365 and pcmcia_core modules; however when I try to "cardctl eject" or "reboot" I get first, "KERNEL: assertion(dev->ip_ptr==NULL)failed at dev.c(2422):netdev_finish_unregister" (not exact since I had to copy it down on paper ... doesn't show up in the logs) then a perpetual series of: "unregister_netdevice: waiting for eth0 to become free. Usage count = -1" messages every five seconds or so. "ps -A" reveals that modprobe is running; it can't be killed even with "kill -9". The "ifconfig" command locks up. Thomas Hood Please cc: your replies to me at jdthood_AT_mail.com - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Extreme IDE slowdown with 2.2.18
On Thu, 21 Dec 2000, Alan Cox wrote: >Date: Thu, 21 Dec 2000 00:49:45 + (GMT) >From: Alan Cox <[EMAIL PROTECTED]> >To: Julian Anastasov <[EMAIL PROTECTED]> >Cc: Robert Högberg <[EMAIL PROTECTED]>, > linux-kernel <[EMAIL PROTECTED]> >Content-Type: text/plain; charset=us-ascii >Subject: Re: Extreme IDE slowdown with 2.2.18 > >> > known problem with the 2.2.18 kernel? >> >> Yes, 2.2.18 is not friendly to all MVP3 users. The autodma >> detection was disabled for the all *VP3 users in drivers/block/ide-pci.= >> c. > >Because it was causing disk corruption for some people. I wish I read this email 24 hours ago. ;o( >It took a lot of tracking down and I want the shipped kernel >safe. I now know I'm covering too many chip versions so 2.2.19 >I can get the later VP3's back okay Any info I can provide to help with my corruption problem enabling UDMA? 00:00.0 Host bridge: VIA Technologies, Inc. VT82C598 [Apollo MVP3] (rev 04) Flags: bus master, medium devsel, latency 16 Memory at d800 (32-bit, prefetchable) Capabilities: [a0] AGP version 1.0 00:01.0 PCI bridge: VIA Technologies, Inc. VT82C598 [Apollo MVP3 AGP] (prog-if 00 [Normal decode]) Flags: bus master, 66Mhz, medium devsel, latency 0 Bus: primary=00, secondary=01, subordinate=01, sec-latency=0 00:07.0 ISA bridge: VIA Technologies, Inc. VT82C596 ISA [Apollo PRO] (rev 23) Subsystem: VIA Technologies, Inc. VT82C596/A/B PCI to ISA Bridge Flags: bus master, stepping, medium devsel, latency 0 00:07.1 IDE interface: VIA Technologies, Inc. VT82C586 IDE [Apollo] (rev 10) (prog-if 8a [Master SecP PriP]) Flags: bus master, medium devsel, latency 32 I/O ports at e000 Capabilities: [c0] Power Management version 2 00:07.3 Host bridge: VIA Technologies, Inc.: Unknown device 3050 (rev 30) Flags: medium devsel Dunno if that helps... -- Mike A. Harris - Linux advocate - Open source advocate This message is copyright 2000, all rights reserved. Views expressed are my own, not necessarily shared by my employer. -- Red Hat FAQ tip: Having trouble upgrading RPM 3.0.x to RPM 4.0.x? Upgrade first to version 3.0.5, and then to 4.0.x. All packages are available on Red Hat's ftp sites: ftp://ftp.redhat.com ftp://rawhide.redhat.com - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Extreme IDE slowdown with 2.2.18
On Wed, 20 Dec 2000, Julian Anastasov wrote: >> hda: QUANTUM FIREBALL ST6.4A, 6149MB w/81kB Cache, CHS=784/255/63 >> hdb: QUANTUM FIREBALL SE4.3A, 4110MB w/80kB Cache, CHS=524/255/63 >> hdc: IBM-DJNA-352030, 19470MB w/1966kB Cache, CHS=39560/16/63 >> >> When I performed the tests I used similiar .17 and .18 kernels with a >> minimum components included. No network, SCSI, sound and such things. >> .config files can be supplied if needed. >> >> Does anyone know what could be wrong? Have I forgot something? Is this a >> known problem with the 2.2.18 kernel? > > Yes, 2.2.18 is not friendly to all MVP3 users. The autodma >detection was disabled for the all *VP3 users in drivers/block/ide-pci.c. > > If you don't experience any problems with the DMA you can: > >1. Add append="ide0=dma ide1=dma" in lilo.conf > >2. Use ide patches: > >http://www.kernel.org/pub/linux/kernel/people/hedrick/ide-2.2.18/ide.2.2.18.1209.patch.bz2 Using an MVP3 board (DFI), 2.2.18 + the above patch, with the above mentioned config changes, DMA by default, and word93 invalidate enabled, I just enabled UDMA66 on my 2 drives and got disk corruption. Both drives are UDMA66 or better, and I'm using the 80 pin cable. 2 root@asdf:~# hdparm -i /dev/hd[ab] /dev/hda: Model=IBM-DTLA-307030, FwRev=TX4OA50C, SerialNo=YKDYKGF1437 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs } RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=40 BuffType=DualPortCache, BuffSize=1916kB, MaxMultSect=16, MultSect=16 CurCHS=16383/16/63, CurSects=-66060037, LBA=yes, LBAsects=60036480 IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: mdma0 mdma1 mdma2 udma0 udma1 *udma2 udma3 udma4 udma5 /dev/hdb: Model=QUANTUM FIREBALL EL7.6A, FwRev=A08.1100, SerialNo=347816714615 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs } RawCHS=15907/15/63, TrkSize=32256, SectSize=21298, ECCbytes=4 BuffType=DualPortCache, BuffSize=418kB, MaxMultSect=16, MultSect=16 CurCHS=15907/15/63, CurSects=1597178085, LBA=yes, LBAsects=15032115 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120} PIO modes: pio0 pio1 pio2 pio3 pio4 DMA modes: sdma0 sdma1 sdma2 mdma0 mdma1 mdma2 udma0 udma1 *udma2 Using "hdparm -d1X66" on these drives results in errors to syslog followed by disk corruption. With word93 thingie NOT built into the kernel, the corruption doesn't occur, but instead I get a message saying UDMA 3/4/5 is not supported. It also claims the MVP3 chipset is UDMA-33 only, whereas all relevant docs I can muster including the mobo manual state the board is UDMA-66 capable. Mental note to myself: Do not enable WORD93 invalidate. ;o) I've never seen UDMA66 work at all on any mobo/disk combo yet that I've tried. My belief has been that the mobo/chipsets are broken, and Andre's code just disables stuff known to be crap hardware. Forcing it as I did, resulted in corruption, so I'll tend to believe the driver next time and not push the issue. ;o) Andre, is MVP3 capable of UDMA66 in any way shape or form, or should I just drop the thought of ever getting it to work and buy an add-in board? If the latter, what recommendation of hardware would you give? I'm getting 11 - 12Mb/s out of my disks now with the IDE patches, which is a MAJOR improvement over the stock kernel. I'd like to push this up closer to the drive's rated capacities though. I'd also like to be able to use whatever kernel I want without using vendor supplied binary-only modules for IDE support. Is there a totally open-source solution for me? ;o) Would I get better results at all with 2.4.0testXX, with or without any patches, and what value of XX? TIA -- Mike A. Harris - Linux advocate - Open source advocate This message is copyright 2000, all rights reserved. Views expressed are my own, not necessarily shared by my employer. -- Red Hat FAQ tip: Having trouble upgrading RPM 3.0.x to RPM 4.0.x? Upgrade first to version 3.0.5, and then to 4.0.x. All packages are available on Red Hat's ftp sites: ftp://ftp.redhat.com ftp://rawhide.redhat.com - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: CPU attachent and detachment in a running Linux system
> That's a good point and it would probably work for attachment of cpus, but > it won't work for detachment because there are some data structures that > need to be updated if a cpu gets detached. For example it would be nice > to flush the per-cpu cache of the detached cpu in the slabcache. Then one > has to think of pending tasklets for the detached cpu which should be > moved to another cpu and then there are a lot of per-cpu data structures > in the networking part of the kernel.. most of them seem to be for > statistics only but I think these structures should be updated in any > case. > So at least for detaching it would make sense to register functions which > will be called whenever a cpu gets detached. I remember someone from SGI had a patch to merge all the per cpu structures together which would make this easier. It would also save bytes especially on machines like the e10k where we must have NR_CPUS = 64. Anton - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: [NFS] kNFSd maintenance in 2.2.19pre
Hi Neil, This sounds good. Any plans on implementing a backport of the nfs filesystem layer for handling inodes that you put together for 2.4. Ie.. the code that reiserfs uses in 2.4 to properly work with knfsd and inode issues. On Thu, 21 Dec 2000, Neil Brown wrote: > > Greeting all. > > Now that 2.2.18 is out with all the nfs (client and server) patches > that we were waiting for for so long, it is time to look at on-going > maintenance for knfsd. > > There are already a couple of issues that have come up and it is > quite possible that more will arise as the user-base grows. > Also, there are quite a few changes that have gone into 2.4 that > could usefully go into 2.2. > > I have discussed the issue of maintenance with Dave Higgen - the > maintainer of the knfsd patchset that finally went into 2.2.18, and > he is happy to leave knfsd for a while and let me run with it. > > So, I have started putting some patches together and they can be > found at > http://www.cse.unsw.edu.au/~neilb/patches/knfsd-2.2/ > > They are mostly back ports of bits from 2.4 with a couple of real bug > fixes, one thanks to Chip Salzenberg and one which allows Solaris > clients to access /dev/null over NFSv3 properly. > > I hope to feed these patches to Alan for inclusion in 2.2.19-preX > early in the new year after I (and you?) have done a bit of testing. > > Note: the patches aren't all quite as independant as they should be > just at the moment (e.g. I made a patch, started on another and then > found a bug in the first, so the fix for the first ended up in the > second). This will get sorted out next time I generate a patch set. > > NeilBrown > > ___ > NFS maillist - [EMAIL PROTECTED] > http://lists.sourceforge.net/mailman/listinfo/nfs > - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: fs corruption in 2.4.0-test11?
On Wed, Dec 20, 2000 at 04:47:42PM -0800, Larry McVoy wrote: > I just need a sanity check - do other pages/blocks sometimes show up in > recently created files in 2.4.0-test11? Mmmm. Yes. I think the final fixes for this went into v2.4.0-test12pre5, but since there's a test13-pre3 out that needs testing, go for that one directly... :^) > I have a (so far) non-reproducible case where the wrong data showed up in > a new file. The nice part is that it was when I was imploding a large > BitKeeper patch so I can run the test case over and over if that would > help find it. If you can reproduce it on test13-pre3, we have something to worry about, if not, feel happy; one bug less to worry about. /David Weinehall _ _ // David Weinehall <[EMAIL PROTECTED]> /> Northern lights wander \\ // Project MCA Linux hacker// Dance across the winter sky // \> http://www.acc.umu.se/~tao/http://www.tux.org/lkml/
Re: Laptop system clock slow after suspend to disk. (2.4.0-test9/hinote VP)
Ian Stirling <[EMAIL PROTECTED]> wrote: > I've not noticed this on earlier kernel versions, is there something > silly I'm missing that's making my DEC hinote VP (p100 laptop)s > system clock slow by a factor of five or so after resume? > Not the CPU or cmos clock, only the system clock. > Thoughts welcome. I saw something like this on my thinkpad (RH6.2) and it turned out to be connected to /etc/adjtime . It was cured by changing the large numbers in there to zeroes. Could someone explain the mechanism? Doug Gilbert - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: iptables: "stateful inspection?"
Alan Cox wrote: > There have been at least five holes found in pile that _could_ have been > [speech] > safe is the day you end up hurt. Your specific example of an executable (windows) attachment, not buffer overflows, etc. what what I was replying to. In general, you are correct. Now, how about including that procfs cleanup patch that I sent, and maybe the 64-bit printk patch? :) -M - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Laptop system clock slow after suspend to disk. (2.4.0-test9/hinote VP)
I've not noticed this on earlier kernel versions, is there something silly I'm missing that's making my DEC hinote VP (p100 laptop)s system clock slow by a factor of five or so after resume? Not the CPU or cmos clock, only the system clock. Thoughts welcome. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
kNFSd maintenance in 2.2.19pre
Greeting all. Now that 2.2.18 is out with all the nfs (client and server) patches that we were waiting for for so long, it is time to look at on-going maintenance for knfsd. There are already a couple of issues that have come up and it is quite possible that more will arise as the user-base grows. Also, there are quite a few changes that have gone into 2.4 that could usefully go into 2.2. I have discussed the issue of maintenance with Dave Higgen - the maintainer of the knfsd patchset that finally went into 2.2.18, and he is happy to leave knfsd for a while and let me run with it. So, I have started putting some patches together and they can be found at http://www.cse.unsw.edu.au/~neilb/patches/knfsd-2.2/ They are mostly back ports of bits from 2.4 with a couple of real bug fixes, one thanks to Chip Salzenberg and one which allows Solaris clients to access /dev/null over NFSv3 properly. I hope to feed these patches to Alan for inclusion in 2.2.19-preX early in the new year after I (and you?) have done a bit of testing. Note: the patches aren't all quite as independant as they should be just at the moment (e.g. I made a patch, started on another and then found a bug in the first, so the fix for the first ended up in the second). This will get sorted out next time I generate a patch set. NeilBrown - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
PANIC: reproducable with nfs, lynx and kernel 2.4.0-test12
-BEGIN PGP SIGNED MESSAGE- I'm not on the list, send private for more info I got a kernel panic with 2.4.0-test12 on a p90 with 24 MB RAM. It's a newly installed debian potato. What I do to trigger the panic is: mount otherbox:/export /mnt cd /mnt lynx www.pgpi.com [ i click to download the latest pgp from norway over http ] [ it downloads and asks where to save it, I just click enter for default ] *crash* A lot of stuff goes by that looks like (this is the last line of this kind): [] [] [] [] Followed by: Code: 89 42 04 89 10 b8 01 00 00 00 c7 43 04 00 00 00 00 c7 03 00 Aiee, killing interrupt handler Kernel panic: attempted to kill the idle task! In interrupt handler - not syncing NOTE that I just compiled the entire kernel source over that same nfs mount, without problems, which leads me to think that it's not a hw issue. More information availible by request. - typedef struct me_s { char name[] = { "Thomas Habets" }; char email[] = { "[EMAIL PROTECTED]" }; char kernel[]= { "Linux 2.2" }; char *pgpKey[] = { "finger -m [EMAIL PROTECTED]" }; char pgpfinger[] = { "6517 2898 6AED EA2C 1015 DCF0 8E53 B69F 524B B541" }; char coolcmd[] = { "echo '. ./_&. ./_'>_;. ./_" }; } me_t; -BEGIN PGP SIGNATURE- Version: PGPfreeware 5.0i for non-commercial use MessageID: F/1sCKH/HYdhVYGAp9oLQgVrxJAoT9GU iQA/AwUBOkFZgyhq6QqtSOhUEQLkvACfTEODuoCPF/Ve3EA1F8xIuT0ClL4AoPtw MKFh2IhXngI87G4BGhRWKVuY =CSfy -END PGP SIGNATURE- - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: CPRM copy protection for ATA drives
> Does anyone have any details on this? I presume that the drive > firmware is capable of identifying copy-protected data during > a write. I also presume that nobody on lkml would condone It seems to be very similar to the DVD stuff, including ideas for play once only blocks and the like. Pay per read hard disk... > such a terrible idea. I imagine that this system is pretty > easy to defeat if you can modify the filesystem. Perhaps even Its probably very hard to defeat. It also in its current form means you can throw disk defragmenting tools out. Dead, gone. Welcome to the United Police State Of America. > The consequences of being able to corrupt other people's backups > by introducing "copy-protected" data are intriguing... I'm just waiting for a few class action law suits against drive manufacturers when people's backup tools cannot cope - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
fs corruption in 2.4.0-test11?
I just need a sanity check - do other pages/blocks sometimes show up in recently created files in 2.4.0-test11? I have a (so far) non-reproducible case where the wrong data showed up in a new file. The nice part is that it was when I was imploding a large BitKeeper patch so I can run the test case over and over if that would help find it. -- --- Larry McVoy lm at bitmover.com http://www.bitmover.com/lm - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Extreme IDE slowdown with 2.2.18
> > known problem with the 2.2.18 kernel? > > Yes, 2.2.18 is not friendly to all MVP3 users. The autodma > detection was disabled for the all *VP3 users in drivers/block/ide-pci.= > c. Because it was causing disk corruption for some people. It took a lot of tracking down and I want the shipped kernel safe. I now know I'm covering too many chip versions so 2.2.19 I can get the later VP3's back okay - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: iptables: "stateful inspection?"
> Alan Cox wrote: > > It does SYN checking. If you are running 'serious' security you wouldnt be > > allowing outgoing connections anyway. One windows christmascard.exe virus that > > connects back to an irc server to take input and you are hosed. > > Thankfully, pine and mutt are, to date, immune to that kind of thing. :) There have been at least five holes found in pile that _could_ have been exploited, and even one in all xterms pre X11R6 where ascii+escape codes was all you needed. Mutt has had minor things fixed for security reasons too. It's harder. But you ignore two things - once someone does it anyone can repeat it - and more importantly almost all exploits rely on user error. Linux users are not always brighter than windows ones and there isnt a lot you can do to make them smarter Think of computer security like powertools. The day you think you are totally safe is the day you end up hurt. Alan - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
emu10k1 and 8139oo with 2.4.0test13pre3
Hello, I'm using the emu10k1 module with my SB Live. This works quite fine, but I cannot switch the recording channel. This worked with 2.2.17. Now volume works, but selecting the input channel not anymore. is anyone else experiencing this problem , or don't i just get the right setting? second, i habe 2 nics in my K6-2 system, both with rtl8139 chip and compiled 8139oo into the kernel. the cards work fine, but dmesg says: eth0: Abnormal interrupt, status 2002 and eth0: Abnormal interrupt, status 0020 endless times. Usually about 2-3 entries per minute. The cards work fine, so how can I get rid of the message, other then uncommenting the line in the source code? This also happens if I compile the driver as module. And happened sind 2.4.0-test12 (the first 2.4.0er I installed). This is my first post on this list. If you need additional information, I'd like to provide it. So far, Jan - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: [Patch] performance enhancement for simple_strtoul
Steve Grubb wrote: > It seems gcc creates much better code with the variables set to register > types. Curious. GCC should be generating the same code regardless; ah well. Is strtoul actually used in the kernel other than for the occasional (rare) write to /proc/sys and parsing boot options? > But this is the kernel and there are people that would reject my patch > purely on the basis that it adds precious bytes to the kernel. Perhaps I am mistaken but I'd expect it to be called what, ten times at boot time, and a couple of times when X loads the MTRRs? Sounds like the neatest trick would be reducing bytes used here... -- Jamie - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: tighter compression for x86 kernels
- Original Message - From: "Frank v Waveren" <[EMAIL PROTECTED]> To: "Adrian Bunk" <[EMAIL PROTECTED]> Cc: "John Reiser" <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]> Sent: Thursday, December 21, 2000 12:22 AM Subject: Re: tighter compression for x86 kernels > Seems GPL2 to me. I haven't read all of the rest of the page, but > that'd either be dual licensing stuff, or further restrictions, which > would be in contradiction with the GPL. > Seems to be kind of dual licensing: "The stub which is imbedded in each UPX compressed program is part of UPX and UCL, and contains code that is under our copyright. The terms of the GNU General Public License still apply as compressing a program is a special form of linking with our stub. As a special exception we grant the free usage of UPX for all executables, including commercial programs. See below for details and restrictions." It extends the scope of the license to _linking_ with commercial software. Jens - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
CPRM copy protection for ATA drives
I read this article on theregister today: http://www.theregister.co.uk/content/2/15620.html Does anyone have any details on this? I presume that the drive firmware is capable of identifying copy-protected data during a write. I also presume that nobody on lkml would condone such a terrible idea. I imagine that this system is pretty easy to defeat if you can modify the filesystem. Perhaps even a ROT13 modification to ext2 would be sufficient? The consequences of being able to corrupt other people's backups by introducing "copy-protected" data are intriguing... Paul - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: tighter compression for x86 kernels
On Thu, Dec 21, 2000 at 12:15:13AM +0100, Adrian Bunk wrote: > > Both source (GPLv2) and pre-compiled binary for x86 are available. >^ > That's not true. Read > http://wildsau.idv.uni-linz.ac.at/mfx/upx-license.html >From that page: UPX and the UCL library are free software; you can redistribute them and/or modify them under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. Seems GPL2 to me. I haven't read all of the rest of the page, but that'd either be dual licensing stuff, or further restrictions, which would be in contradiction with the GPL. -- Frank v Waveren Fingerprint: 0EDB 8787 fvw@[var.cx|dse.nl|stack.nl|chello.nl] ICQ#10074100 09B9 6EF5 6425 B855 Public key: http:[EMAIL PROTECTED] 7179 3036 E136 B85D - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: tighter compression for x86 kernels
On Wed, 20 Dec 2000, John Reiser wrote: > Beta release v1.11 of the UPX executable compressor http://upx.tsx.org > offers new, tighter re-compression of compressed Linux kernels for x86. > Additional space savings of about 15% have been seen using > "upx --best vmlinuz" (example: 617431 ==> 525099, saving 92332 bytes). > Both source (GPLv2) and pre-compiled binary for x86 are available. ^ That's not true. Read http://wildsau.idv.uni-linz.ac.at/mfx/upx-license.html > [I'm not subscribed to this mailing list, so CC: or mail me if appropriate.] cu, Adrian -- A "No" uttered from deepest conviction is better and greater than a "Yes" merely uttered to please, or what is worse, to avoid trouble. -- Mahatma Ghandi - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
tighter compression for x86 kernels
Beta release v1.11 of the UPX executable compressor http://upx.tsx.org offers new, tighter re-compression of compressed Linux kernels for x86. Additional space savings of about 15% have been seen using "upx --best vmlinuz" (example: 617431 ==> 525099, saving 92332 bytes). Both source (GPLv2) and pre-compiled binary for x86 are available. [I'm not subscribed to this mailing list, so CC: or mail me if appropriate.] -- John Reiser, [EMAIL PROTECTED] - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: iptables: "stateful inspection?"
Michael Rothwell said once upon a time (Wed, 20 Dec 2000): > Alan Cox wrote: > > > It does SYN checking. If you are running 'serious' security you wouldnt be > > allowing outgoing connections anyway. One windows christmascard.exe virus that > > connects back to an irc server to take input and you are hosed. > > Thankfully, pine and mutt are, to date, immune to that kind of thing. :) Try again. Pine less than 4.30 has a buffer overflow builtin. A properly formated "From" header (or something) can hose you. No need for any attachment. Dax - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: PCMCIA modem (v.90 X2) not working with 2.4.0-test12 PCMCIA services
On Wed, Dec 20, 2000 at 04:33:04AM -0800, David Hinds wrote: > On Wed, Dec 20, 2000 at 12:10:41PM -0700, Jeff V. Merkey wrote: > > On Tue, Dec 19, 2000 at 05:05:16PM -0700, Jeff V. Merkey wrote: > > > > Do you think there's a solution for this problem. Sorry for bothering > > you again. I'm available if you need some help retesting and fixes. > > I do not have a solution. I have a few reports of tx timeout problems > that I have so far been unable to reproduce. It is a sufficiently > nonspecific outcome that I don't have any good ideas for how to track > down the problem; all my attempts so far have come up blank. > > -- Dave If you have a mailing address, I can overnight the laptop computer to you provided you agree to return it after you run down the problem. :-) Jeff - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: 2.4.0-test12-pre7 shutdowns and eepro100 woes
On Tue, Dec 12, 2000, Andrey Savochkin <[EMAIL PROTECTED]> wrote: ; To answer your question in short, yet, we hope to fix the problem sooner or ; later. I added the print out of the message to see in what state was the card being left after it was wedged. The card seems to be locking up with undefined opcodes, atleast according to my specs. The command doesn't necessarly come from the driver, I'we done some experimenting and it seems that sending the card an undefined opcode will lock up the card with a different value in the command register. I am still waiting for latest specs from intel, I wonder if the new specs will define those values. -- I knew I was alone, I was scared, it was getting dark and it was a hardware problem. -Dragan - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH] fix emu10k1 init breakage in 2.2.18
Hello, The patch indeed solves the problem with EMU10K. It now works well except from the fact that the trebble and bass controls still have been vanished. Thanks for the patch. Kees BTW could it be something simular for es1371?. This also fails with 2.2.18 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Announce: modutils 2.3.23 is available
On Wed, 20 Dec 2000 10:31:12 +0100, Christian Gennerat <[EMAIL PROTECTED]> wrote: >About Standard aliases: >> modprobe -c >... >alias ppp-compress-21 bsd_comp >... > >Why bsd_comp is the standard alias? You should also have alias ppp-compress-24 ppp_deflate alias ppp-compress-26 ppp_deflate The number is the CPP option that was requested by pppd, which compression option is used is controlled by userspaqce, not the kernel. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Extreme IDE slowdown with 2.2.18
Hello, On Wed, 20 Dec 2000, Robert HÃgberg wrote: > hda: QUANTUM FIREBALL ST6.4A, 6149MB w/81kB Cache, CHS=784/255/63 > hdb: QUANTUM FIREBALL SE4.3A, 4110MB w/80kB Cache, CHS=524/255/63 > hdc: IBM-DJNA-352030, 19470MB w/1966kB Cache, CHS=39560/16/63 > > When I performed the tests I used similiar .17 and .18 kernels with a > minimum components included. No network, SCSI, sound and such things. > .config files can be supplied if needed. > > Does anyone know what could be wrong? Have I forgot something? Is this a > known problem with the 2.2.18 kernel? Yes, 2.2.18 is not friendly to all MVP3 users. The autodma detection was disabled for the all *VP3 users in drivers/block/ide-pci.c. If you don't experience any problems with the DMA you can: 1. Add append="ide0=dma ide1=dma" in lilo.conf 2. Use ide patches: http://www.kernel.org/pub/linux/kernel/people/hedrick/ide-2.2.18/ide.2.2.18.1209.patch.bz2 Regards -- Julian Anastasov <[EMAIL PROTECTED]> - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: [2.2.18] VM: do_try_to_free_pages failed
On Wednesday, December 20, 2000 13:03:00 +0100 Matthias Andree <[EMAIL PROTECTED]> wrote: > Last night, one of your production machines got wedged, I caught a lot > of kernel: VM: do_try_to_free_pages failed for ... for a whole range of > processes, among them ypbind, klogd, syslogd, xntpd, cron, nscd, X, > master (Postfix super daemon), pvmd3, K applications and so on, I was > unable to log in via ssh, someone on-site has finally reset that machine > this noon to bring it back online. > > How can I get rid of those do_try_to_free_pages lockups? That box > exports root file systems for some SparcStation 2 that are used as X > terminals, so it's pretty important I keep that box running. > > Should I try the most recent 2.2.19-pre? > > The machine is a pentium-MMX with 64 MB RAM with a kernel 2.2.18 that > has these patches/updated drivers (none VM related AFAICS): > > IDE 2.2.18.1209 > I²C 2.5.4 > LM_Sensors 2.5.4 > DC390 2.0e7 > ReiserFS 3.5.28 > If you still see the problem with Andrea's VM global patch (you can get just that one patch from ftp.kernel.org/pub/people/andrea), try cutting JOURNAL_MAX_BATCH in half. This will lower the amount of memory reiserfs is willing to pin in one transaction... -chris - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: iptables: "stateful inspection?"
Alan Cox wrote: > It does SYN checking. If you are running 'serious' security you wouldnt be > allowing outgoing connections anyway. One windows christmascard.exe virus that > connects back to an irc server to take input and you are hosed. Thankfully, pine and mutt are, to date, immune to that kind of thing. :) -M - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: iptables: "stateful inspection?"
> "Michael H. Warfield" wrote: > > I think that's more than a little overstatement on your > > part. It depends entirely on the application you intend to put > > it to. > > Fine. How do I make FTP work through it? How can I allow all outgoing Passive mode or a proxy. > TCP connections without opening the network to inbound connections on > the ports of desired services? It does SYN checking. If you are running 'serious' security you wouldnt be allowing outgoing connections anyway. One windows christmascard.exe virus that connects back to an irc server to take input and you are hosed. So its perfectly adequate for basic security, but if you want serious security and you don't have passwords on outgoing connections think again. If you are using ftp then be sure to also use other methods to verify a third party didnt change the file you up/downloaded too. Alan - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: 2.2.18: Thread problem with smbfs
Hi, Urban Widmark wrote: > I don't really know how signal delivery works within the kernel, but > smb_trans2_request tries to disable some signals. That does not work > (completely?) so either it needs fixing or the -512 errno needs to be > handled. > > Why so bad in gdb? perhaps it causes more signals. > Why does one thread end up in D state? don't know. > > > > Kernel 2.2.18, smbfs as a module. I can provide more info if necessary. > > A small testprogram that causes this would be nice. The -512 is easy to > reproduce but I haven't seen the 'D' before. > > If someone is interested the relevant code is fs/smbfs/sock.c > (smb_trans2_request, ..., _recvfrom) Here is a test program to reproduce this. Don't worry about missing error checks and so on, it's just a quick hack. Create the required files file1..file5 on a SMB share and edit the #define accordingly. File sizes of 1-2 MB should suffice. Then run the program. It should copy the files to the current directory. Then run it under gdb. It should hang until you kill gdb. I tested only with a NT 4 server (sp 5 or 6). Regards, hjb #include #include #include #include #include #include #include /* Size of the blocks we read from a file. */ static const int ChunkSize = 8192; /* Path on the mounted SMB share from which we copy files */ #define SourcePath "/mnt/net/test" struct CopyThreadInfo { char*src; char*dst; }; /* returns 1 on success */ int CopyFile(char* src, char* dst) { charbuffer[ChunkSize]; int f, g; ssize_t nRet; int nError; if ((f = open(src, O_RDONLY)) < 0) return 0; g = open(dst, O_WRONLY | O_CREAT | O_TRUNC, 0666); if (g < 0) { close(f); return 0; } do { nRet = read(f, buffer, sizeof(buffer)); if (nRet < 0 && errno == EINTR) nRet = 0; if (nRet < 0) { return 0; } if (nRet > 0) nRet = write(g, buffer, nRet); } while (nRet > 0); close(g); close(f); if (nRet < 0) return 0; return 1; } void* Copy(struct CopyThreadInfo *info) { CopyFile(info->src, info->dst); return NULL; } void Fetch(char* name) { char src[4096]; char dst[4096]; pthread_attr_t attr; pthread_t pid; struct CopyThreadInfo* pCopy = (struct CopyThreadInfo *) malloc(sizeof(struct CopyThreadInfo)); strcpy(src, SourcePath); strcat(src, name); strcpy(dst, name); pCopy->src = strdup(src); pCopy->dst = strdup(dst); pthread_attr_init(&attr); pthread_attr_setdetachstate(&attr, PTHREAD_CREATE_DETACHED); pthread_create(&pid, &attr, Copy, pCopy); } int main() { Fetch("file1"); Fetch("file2"); Fetch("file3"); Fetch("file4"); Fetch("file5"); while(1) ; return 0; } -- http://www.pro-linux.de/ - Germany's largest volunteer Linux support site - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
2.4.0-test13-pre3 drivers/char/Makefile and CONFIG_DRM_MGA=m
Hello, When I use 'make menuconfig' on 2.4.0-test13-pre3 to select DRM and the Matrox DRM driver as a module, I get a .config with CONFIG_DRM=y and CONFIG_DRM_MGA=m. However, this causes drivers/char/Makefile to skip the drm subdirectory when compiling modules: its only reference to CONFIG_DRM is the line 'subdir-$(CONFIG_DRM) += drm' and 'make menuconfig' does not allow me to select 'm' for CONFIG_DRM. I don't know enough about the kernel Makefiles to figure out the proper solution, but I was able to get the modules compiled by adding the line 'mod-subdirs += drm' to drivers/char/Makefile. This is not the proper solution, I expect, as now 'make modules' will enter the driver/char/drm subdirectory even if CONFIG_DRM=n. Cheers, Wayne --- drivers/char/Makefile~ Wed Dec 20 11:16:59 2000 +++ drivers/char/Makefile Wed Dec 20 11:21:54 2000 @@ -154,6 +154,7 @@ subdir-$(CONFIG_FTAPE) += ftape subdir-$(CONFIG_DRM) += drm +mod-subdirs += drm subdir-$(CONFIG_PCMCIA) += pcmcia subdir-$(CONFIG_AGP) += agp - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Extreme IDE slowdown with 2.2.18
On Wed, 20 Dec 2000, Robert Högberg wrote: > I'm having problems with the performance of my harddrives after I > upgraded my kernel from 2.2.17 to 2.2.18. > The performancedrop is noticable on every IDE drive. [snip] > Does anyone know what could be wrong? Have I forgot something? Is this a > known problem with the 2.2.18 kernel? Is DMA enabled on the hard drives? Did you turn on "Use DMA by default" in the kernel configuration? Did you compile in DMA support (if needed)? What is the output of "hdparm /dev/hda /dev/hdb /dev/hdc" ? Rob Adamson. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Startup IPI (was: Re: test13-pre3)
On 20 Dec 00 at 19:52, Maciej W. Rozycki wrote: > > it kills machine; only problem is that 0x1300 wr-rd cycles to VGA apperture > > take 3.48ms, and this does not correspond with needed 200us udelay. > > Hmm, how do you calculate the time? Assuming AGP4x runs at 133MHz and a > read or write cycle lasts for a single clock tick (I don't know exact AGP > specs -- please correct me if I'm wrong), I find 0x1300 cycles to finish > in about 73usecs. The loop execution overhead may double the result and > it will still fit within 300usecs. It is easy: int mfd; volatile unsigned long* memory; int i; mfd = open("/dev/mem", O_RDWR); memory = mmap(0, 4096, PROT_READ|PROT_WRITE, MAP_SHARED, mfd, 0x000B8000); close(mfd); for (i = 0; i < 0x1300 * 1000; i++) { *memory = i; *memory; } munmap(memory, 4096); /usr/bin/time says that program runs for 3.40 - 3.56secs, so after dividing by 1000 I get 3.4ms... Maybe I should complain to VIA or to Matrox that it is piece of crap ? > > Without VIA datasheet I cannot try to disable some PCI features to find > > which one is culprit, so I'm sorry. > > But you may complain to the manufacturer and/or change hardware. I'm > still uncertain the delay should stay in... My order was simple: no rambus memory, dual PIII at least on 800MHz and UDMA66. Yes, maybe I should buy ServerWorks instead of VIA, but I hoped... Best regards, Petr Vandrovec [EMAIL PROTECTED] - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
usb + smp + 2.4.0test = pci irq routing problem?
hello, i've been working with johannes erdfelt in fixing a problem with usb on my machine. it's a dual pentium 3 system on a tyan tiger 133 mobo (via apollo pro 133a chipset). basically, usb works when i don't enable smp or when i disable apic on smp-enabled kernels. he believes that we're seeing a pci irq routing problem and that i should contact martin mares about this problem. (i've written him a couple times, but have heard nothing, so i figure he's either away, busy, or whatnot and i thought i'd try lkml for help.) i have an ethernet card on my system and it shares an irq with usb-uhci. in this state, i see interrupts for the irq eth0 and usb-uhci share. when i remove the ethernet card, i get this in /proc/interrupts: CPU0 CPU1 0: 37124 19379IO-APIC-edge timer 1:146 84IO-APIC-edge keyboard 2: 0 0 XT-PIC cascade 8: 1 0IO-APIC-edge rtc 14: 1640 1910IO-APIC-edge ide0 15: 1 1IO-APIC-edge ide1 16: 29 28 IO-APIC-level ide2 18: 26 27 IO-APIC-level aic7xxx 19: 0 0 IO-APIC-level usb-uhci NMI: 56419 56419 LOC: 56392 56403 ERR: 0 from this, je thought that this was a pci irq routing problem and not a usb problem. because of this, running an smp-enabled kernel with apic enabled yields the "device not accepting new address" error on startup (usb is compiled into my kernel, so i'm not sure what part is actually triggering the error) and none of the usb devices work. (yes, i've checked the mps and tried both 1.1 and 1.4.) if i disable apic or don't use an smp-enabled kernel, everything works fine. this has been happening for quite a while, at least since 2.4.0test9, right up to test13-pre3. i really don't know what kind of information would be useful for debugging this problem. i don't know much about kernel programming, but i am more than willing to try any kind of patch or give any information about my system that could help squash this bug. it's a problem that quite a few people on the linux-usb list are complaining about (all, it seems, have this via chipset). please let me know if there's any more info i can provide, i'm more than happy to help. thanks, pete -- Pete Toscanop:[EMAIL PROTECTED] w:[EMAIL PROTECTED] GPG fingerprint: D8F5 A087 9A4C 56BB 8F78 B29C 1FF0 1BA7 9008 2736 PGP signature
memmove() in 2.4.0-test12, alpha platform
Hello ! New (since test12) optimized memmove function seems to be broken on alpha platform. If dest and src arguments are misaligned, new memmove does wrong things. example: static char p[] = "abcdefghijklmnopkrstuvwxyz01234567890"; memmove(p + 2, p + 13, 17); printk ("DEBUG: memmove test: %s\n", p); produces: DEBUG: memmove test: abyz0123tuvwxyz0123tuvwxyz01234567890 Old memmove variant didn't have this problem. Thanks, Alex. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Startup IPI (was: Re: test13-pre3)
On Tue, 19 Dec 2000, Petr Vandrovec wrote: > I did... So it uses 'xchg %eax,APIC_ICR' instead of 'movl %eax,APIC_ICR', > yes (as verified in generated code...)? No change, still dies, as expected > (do not forget that before it dies, it can do ~0x1300 write-read cycles I've forgotten indeed... > from videomemory (AGP4x), so secondary CPU just does some thinking before This might be the time needed to deliver the IPI. Remember that the inter-APIC bus is serial and not that fast. > it kills machine; only problem is that 0x1300 wr-rd cycles to VGA apperture > take 3.48ms, and this does not correspond with needed 200us udelay. Hmm, how do you calculate the time? Assuming AGP4x runs at 133MHz and a read or write cycle lasts for a single clock tick (I don't know exact AGP specs -- please correct me if I'm wrong), I find 0x1300 cycles to finish in about 73usecs. The loop execution overhead may double the result and it will still fit within 300usecs. > Maybe chipset decides to do something when second CPU cannot obtain > bus access in 10 pci cycles?). I guess a certain initial cycle from the AP confuses the chipset somehow. > Do you (or anyone else) have code which can dump MTRR registers of each > of CPU before mtrr driver takes over them? At least first CPU does not have > any problem... A brief look at arch/i386/kernel/mtrr.c reveals the bootstrap CPU's settings do not get changed. As a result they may always be fetched from the /proc filesystem. For APs you probably need to tweak sources. > I even placed 'wbinvd' and 'wbinvd; cpuid' before sending startup IPI, > but it does not matter. Secondary CPU just does not finish even first > instruction when first CPU reads from videoram again and again. Well, the CPU obeys the writeback and the invalidation, but does the chipset? > Without VIA datasheet I cannot try to disable some PCI features to find > which one is culprit, so I'm sorry. But you may complain to the manufacturer and/or change hardware. I'm still uncertain the delay should stay in... Maciej -- + Maciej W. Rozycki, Technical University of Gdansk, Poland + +--+ +e-mail: [EMAIL PROTECTED], PGP key available+ - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Problems with ToPIC97 reappear between test10 and test13pre2
I sent a similar message earlier, but it has not shown up on linux-kernel so I guess something went wrong... I am starting to have problems again with the Cardbus controller somewhere inbetween 2.4.0-test10 (works) and 2.4.0-test13-pre2 (fails). The problem shows as follows: When I boot the kernel (I have PCMCIA compiled in) only one of my cards gets detected due to the following error: cs: socket c5feb800 timed out during reset. Try increasing setup_delay. cs: socket c5fb5000 voltage interrogation timed out I did not have see this error before and increasing setup_delay manually in the source did not help. Now when I do "cardctl eject; cardctl insert" things work out fine. I am not sure if this is somehow related to the problems that were fixed with test5pre6, but the sumptoms are kind of similar (one slot works - other does not). If you want me to try something please let me know. Jens ps: attached please find the dmesg output. -- Jens Taprogge Linux version 2.4.0-test13-pre2 (root@al) (gcc version 2.95.2 2220 (Debian GNU/Linux)) #9 Sun Dec 17 15:21:56 CET 2000 BIOS-provided physical RAM map: BIOS-e820: 0009fc00 @ (usable) BIOS-e820: 0400 @ 0009fc00 (reserved) BIOS-e820: 4000 @ 000e8000 (reserved) BIOS-e820: 0001 @ 000f (reserved) BIOS-e820: 05ef @ 0010 (usable) BIOS-e820: 0001 @ 05ff (ACPI data) BIOS-e820: 00016e00 @ 100a (reserved) BIOS-e820: 0200 @ 100b6e00 (ACPI NVS) BIOS-e820: 00049000 @ 100b7000 (reserved) BIOS-e820: 0008 @ fff8 (reserved) Scan SMP from c000 for 1024 bytes. Scan SMP from c009fc00 for 1024 bytes. Scan SMP from c00f for 65536 bytes. Scan SMP from c009fc00 for 4096 bytes. On node 0 totalpages: 24576 zone(0): 4096 pages. zone(1): 20480 pages. zone(2): 0 pages. mapped APIC to e000 (0119a000) Kernel command line: root=/dev/hda2 video=vesa vga=771 mem=96M parport=0x378,auto pci=biosirq opl3sa2=0x538,5,1,0,0x530,0x388 Initializing CPU#0 Detected 233.294 MHz processor. Console: colour dummy device 80x25 Calibrating delay loop... 465.31 BogoMIPS Memory: 93888k/98304k available (1795k kernel code, 4028k reserved, 123k data, 204k init, 0k highmem) Dentry-cache hash table entries: 16384 (order: 5, 131072 bytes) Buffer-cache hash table entries: 4096 (order: 2, 16384 bytes) Page-cache hash table entries: 32768 (order: 5, 131072 bytes) Inode-cache hash table entries: 8192 (order: 4, 65536 bytes) CPU: Before vendor init, caps: 0183f9ff , vendor = 0 CPU: L1 I cache: 16K, L1 D cache: 16K CPU: L2 cache: 512K Intel machine check architecture supported. Intel machine check reporting enabled on CPU#0. CPU: After vendor init, caps: 0183f9ff CPU: After generic, caps: 0183f9ff CPU: Common caps: 0183f9ff CPU: Intel Pentium II (Deschutes) stepping 00 Checking 'hlt' instruction... OK. POSIX conformance testing by UNIFIX mtrr: v1.37 (20001109) Richard Gooch ([EMAIL PROTECTED]) mtrr: detected mtrr type: Intel PCI: PCI BIOS revision 2.10 entry at 0xf1927, last bus=21 PCI: Using configuration type 1 PCI: Probing PCI hardware PCI: Using IRQ router PIIX [8086/7110] at 00:07.0 PCI: Found IRQ 11 for device 00:02.0 PCI: Found IRQ 11 for device 00:02.1 PCI: Cannot allocate resource region 4 of device 00:07.1 got res[1000:1fff] for resource 0 of Toshiba America Info Systems ToPIC97 got res[10001000:10001fff] for resource 0 of Toshiba America Info Systems ToPIC97 (#2) got res[1000:100f] for resource 4 of Intel Corporation 82371AB PIIX4 IDE Limiting direct PCI/PCI transfers. Linux NET4.0 for Linux 2.4 Based upon Swansea University Computer Society NET3.039 IA-32 Microcode Update Driver: v1.08 <[EMAIL PROTECTED]> apm: BIOS version 1.2 Flags 0x02 (Driver version 1.14) Starting kswapd v1.8 0x378: FIFO is 16 bytes 0x378: writeIntrThreshold is 8 0x378: readIntrThreshold is 8 0x378: PWord is 8 bits 0x378: Interrupts are ISA-Pulses 0x378: ECP port cfgA=0x10 cfgB=0x4b 0x378: ECP settings irq=7 dma=3 parport0: PC-style at 0x378 (0x778), irq 7, using FIFO [PCSPP,TRISTATE,COMPAT,ECP] parport0: cpp_daisy: aa5500ff(38) parport0: assign_addrs: aa5500ff(38) parport0: cpp_daisy: aa5500ff(38) parport0: assign_addrs: aa5500ff(38) vesafb: framebuffer at 0xfe00, mapped to 0xc680, size 2048k vesafb: mode is 800x600x8, linelength=800, pages=0 vesafb: protected mode interface info at c000:96e0 vesafb: scrolling: redraw Console: switching to colour frame buffer device 100x37 fb0: VESA VGA frame buffer device pty: 256 Unix98 ptys configured lp0: using parport0 (interrupt-driven). Uniform Multi-Platform E-IDE driver Revision: 6.31 ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx PIIX4: IDE controller on PCI bus 00 dev 39 PIIX4: chipset revision
Re: [Patch] performance enhancement for simple_strtoul
Hello, I continued experimenting with the Test Case and found a further speed improvement & I am re-submiting the patch. It is the same as the first one with the two local variables changed to register storage types. On a K6-2, I now see: Base 10 - 28% speedup Base 16 - 24% speedup Base 8 - 30% speedup On a P3 system, I now see: Base 10 - 25% speedup Base 16 - 17% speedup Base 8 - 20% speedup It seems gcc creates much better code with the variables set to register types. Please apply the following patch. It should apply to any recent 2.2.x without problems. In 2.4 the function starts 2 lines later. Cheers, Steve Grubb -- --- lib/vsprintf.orig Fri Dec 1 08:58:02 2000 +++ lib/vsprintf.c Wed Dec 20 13:14:13 2000 @@ -14,10 +14,13 @@ #include #include +/* +* This function converts base 8, 10, or 16 only - Steve Grubb +*/ unsigned long simple_strtoul(const char *cp,char **endp,unsigned int base) { - unsigned long result = 0,value; - + register unsigned char c; + register unsigned long result = 0; if (!base) { base = 10; if (*cp == '0') { @@ -29,11 +32,36 @@ } } } - while (isxdigit(*cp) && (value = isdigit(*cp) ? *cp-'0' : (islower(*cp) - ? toupper(*cp) : *cp)-'A'+10) < base) { - result = result*base + value; - cp++; - } + c = *cp; +switch (base) { +case 10: +while (isdigit(c)) { +result = (result*10) + (c & 0x0f); +c = *(++cp); +} +break; +case 16: +while (isxdigit(c)) { +result = result<<4; +if (c&0x40) + result += (c & 0x07) + 9; +else +result += c & 0x0f; +c = *(++cp); +} +break; +case 8: +while (isdigit(c)) { +if ((c&0x37) == c) +result = (result<<3) + (c & 0x07); +else +break; +c = *(++cp); +} +break; +default: /* Is anything else used by the kernel? */ +break; +} if (endp) *endp = (char *)cp; return result; - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: iptables: "stateful inspection?"
Hello all! On Wed, Dec 20, 2000 at 01:08:07PM -0500, Michael H. Warfield wrote: > On Wed, Dec 20, 2000 at 12:52:27PM -0500, Michael Rothwell wrote: > > "Michael H. Warfield" wrote: > > > You can use spf to add some stateful inspection for PORT mode > > > ftp. Personally, I like the masquerading option better, though. > > Can you give an example of using MASQ selectively? I have real addresses > > on both sides of the firewall, but want things like FTP to work > > correctly. I think the IPChains HOWTOs are just a little terse. :) Michael Rothwell kindly pointed out to me in private mail that I SCREWED UP (he didn't say that, I did) the copy-and-past on one of the command lines and left out a "little detail"... > modprobe ip_masq_ftp > ipchains -A forward -p tcp -s {Source Addresses} -d 0/0 21 This should have been: modprobe ip_masq_ftp ipchains -A forward -p tcp -s {Source Addresses} -d 0/0 21 -j MASQ DOH! Sorry! > Seems to work for me (mine includes a "tag" and a policy route > rule to send it out my cable modem that I've left off here)... > If you don't load the ip_masq_ftp module, you WILL get illegal > port errors on the PORT commands. > > Thanks! Mike -- Michael H. Warfield| (770) 985-6132 | [EMAIL PROTECTED] (The Mad Wizard) | (678) 463-0932 | http://www.wittsend.com/mhw/ NIC whois: MHW9 | An optimist believes we live in the best of all PGP Key: 0xDF1DD471| possible worlds. A pessimist is sure of it! - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: PCMCIA modem (v.90 X2) not working with 2.4.0-test12 PCMCIA services
On Wed, Dec 20, 2000 at 12:10:41PM -0700, Jeff V. Merkey wrote: > On Tue, Dec 19, 2000 at 05:05:16PM -0700, Jeff V. Merkey wrote: > > Do you think there's a solution for this problem. Sorry for bothering > you again. I'm available if you need some help retesting and fixes. I do not have a solution. I have a few reports of tx timeout problems that I have so far been unable to reproduce. It is a sufficiently nonspecific outcome that I don't have any good ideas for how to track down the problem; all my attempts so far have come up blank. -- Dave - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Linux 2.2.19pre2
On Wed, Dec 20, 2000 at 03:48:06PM -0200, Rik van Riel wrote: > On Wed, 20 Dec 2000, Andrea Arcangeli wrote: > > On Thu, Dec 21, 2000 at 01:57:15AM +1100, Andrew Morton wrote: > > > If a task is on two waitqueues at the same time it becomes a bug: > > > if the outer waitqueue is non-exclusive and the inner is exclusive, > > > > Your 2.2.x won't allow that either. You set the > > `current->task_exclusive = 1' and so you will get an exclusive > > wakeups in both waitqueues. You simply cannot tell register in > > two waitqueue and expect a non-exlusive wakeup in one and an > > exclusive wakeup in the other one. > > Why not? Having a wake-all wakeup on one event and a > wake-one wakeup on another kind of event seems perfectly > reasonable to me at a first glance. Is there something > I've overlooked ? I think you overlooked, never mind. I only said kernel 2.2.19pre2 won't allow that either. I'm not saying it's impossible to implement or unrasonable. > > > Anyway, it's academic. davem would prefer that we do it properly > > > and move the `exclusive' flag into the waitqueue head to avoid the > > > task-on-two-waitqueues problem, as was done in 2.4. I think he's > > > > The fact you could mix non-exclusive and exlusive wakeups in the > > same waitqueue was a feature not a misfeature. Then of course > > you cannot register in two waitqueues one with wake-one and one > > with wake-all but who does that anyways? > > The "who does that anyways" argument could also be said about > mixing exclusive and non-exclusive wakeups in the same waitqueue. > Doing this seems rather confusing ... would you know any application > which needs this ? wake-one accept(2) in 2.2.x unless you want to rewrite part of the TCP stack to still get select wake-all right (the reason they was mixed in whole 2.3.x was the same reason we _need_ to somehow mix non-excusive and exlusive waiters in the same waitqueue in 2.2.x to provide wake-one in accept). And in 2.2.x the "who does that anyways" is much stronger, since _only_ accept is using the exclusive wakeup. > I think it would be good to have the same semantics in 2.2 as > we have in 2.4. [..] 2.3.0 born for allowing O(1) inserction in the waitqueue because only that change generated and huge amount of details that was not possible to handle in 2.2.x. > [..] If there is a good reason to go with the > semantics Andrea proposed [..] NOTE: I'm only talking about 2.2.19pre2, not 2.4.x. I didn't suggested anything for 2.4.x and I'm totally fine with two different waitqueues. I even wanted to differentiate them too in mid 2.3.x to fix accept that was FIFO but still allowing insertion in the waitqueue in O(1), but didn't had the time to rework the TCP stack and the rest of wake-one users. Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: PCMCIA modem (v.90 X2) not working with 2.4.0-test12 PCMCIA services
On Tue, Dec 19, 2000 at 05:05:16PM -0700, Jeff V. Merkey wrote: David, Do you think there's a solution for this problem. Sorry for bothering you again. I'm available if you need some help retesting and fixes. :-) Jeff > On Tue, Dec 19, 2000 at 01:51:14PM -0800, David Hinds wrote: > > On Tue, Dec 19, 2000 at 03:41:29PM -0700, Jeff V. Merkey wrote: > > > > > > On a related topic, the 3c575_cb driver on an IBM Thinkpad 765D is getting > > > tx errors on the 2.2.18 kernel with PCMCIA services 3.1.22. > > > > > > Card is a 3Com 3CCFE575BT Cyclone Cardbus Adapter. > > > > > > Error is: > > > > > > eth0: transmit timed out, tx_status 00 status e000. > > > diagnostics net 0cc2 media a800 dma 00a0 > > > > What host bridge is in the 765D? Is it perhaps a TI 1131 rev 1, or > > something else? Also, try adding: > > > > /proc/bus/pccard/00/info reports TI 1130 chipset. > > > module "3c575_cb" opts "down_poll_rate=0" > > Adding this does not fix the problem, but does cause a little more > error info to get printed. Now in addition to the original message, > I am also seeing: > > eth0: Tx ring full, refusing to send buffer. > > Looks like some type of interrupt problem. I am available to assist > you in debugging this problem. > > Jeff > > > > > to /etc/pcmcia/config.opts and see if that makes any difference. > > > > -- Dave > > - > > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > > the body of a message to [EMAIL PROTECTED] > > Please read the FAQ at http://www.tux.org/lkml/ > - > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to [EMAIL PROTECTED] > Please read the FAQ at http://www.tux.org/lkml/ - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: iptables: "stateful inspection?"
On Wed, Dec 20, 2000 at 12:52:27PM -0500, Michael Rothwell wrote: > "Michael H. Warfield" wrote: > > You can use spf to add some stateful inspection for PORT mode > > ftp. Personally, I like the masquerading option better, though. > Can you give an example of using MASQ selectively? I have real addresses > on both sides of the firewall, but want things like FTP to work > correctly. I think the IPChains HOWTOs are just a little terse. :) modprobe ip_masq_ftp ipchains -A forward -p tcp -s {Source Addresses} -d 0/0 21 Seems to work for me (mine includes a "tag" and a policy route rule to send it out my cable modem that I've left off here)... If you don't load the ip_masq_ftp module, you WILL get illegal port errors on the PORT commands. > Thanks! Mike -- Michael H. Warfield| (770) 985-6132 | [EMAIL PROTECTED] (The Mad Wizard) | (678) 463-0932 | http://www.wittsend.com/mhw/ NIC whois: MHW9 | An optimist believes we live in the best of all PGP Key: 0xDF1DD471| possible worlds. A pessimist is sure of it! - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: /dev/random: really secure?
Bernd Eckenfels wrote: > In article <[EMAIL PROTECTED]> you wrote: > > A potential weakness. The entropy estimator can be manipulated by > > feeding data which looks random to the estimator, but which is in fact > > not random at all. > > That's why feeding randomness is a priveledgedoperation. I was referring to randomness influenced externally, e.g. network packet timing, hard disk timing by choice of which http requests, etc. -- Jamie - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: 2.2.19/2.4.0-test and usbdevfs
> modprobe: modprobe: Can't locate module usbdevfs > mount: fs type usbdevfs not supported by kernel Add: alias usbdevfs usbcore to /etc/modules.conf Tom - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
sm56 modem driver installation help
IMP:please reply me my mail address and not to the vger.kernel.org I used the follwing to setup the modem driver from motorola, afterinstalling rpm thru Kpackagedepmod -aln -sf /dev/ttyEo /dev/modemmknod /dev/motomem c 28 0The query in kppp says, "Sorry,the didn't respond" . I see sm56.o in/2.2.14/misc directory and clearly can see the package in Kpackage .Please help me in making my SM56PCI imternal modem work...Regards,Madan A SThese are the files after setupfile:conf.modules--alias char-major-24 sm56options sm56 country=1alias sound cmpcifile:/proc/devices---Character devices: 1 mem 2 pty 3 ttyp 4 ttyS 5 cua 7 vcs 10 misc 14 sound 29 fb 36 netlink128 ptm136 pts162 rawBlock devices: 1 ramdisk 2 fd 3 ide0 9 md 22 ide1file:/proc/modules---lockd 31176 1 (autoclean)sunrpc 52964 1 (autoclean) [lockd]ppp 20236 0 (autoclean) (unused)slhc 4504 0 (autoclean) [ppp]nls_iso8859-1 2240 3 (autoclean)nls_cp437 3748 3 (autoclean)vfat 9372 3 (autoclean)fat 30720 3 (autoclean) [vfat]cmpci 20060 0soundcore 2596 4 [cmpci]file:/var/log/messagesDec 20 14:35:51 localhost syslogd 1.3-3: restart.Dec 20 14:35:51 localhost syslog: syslogd startup succeededDec 20 14:35:51 localhost syslog: klogd startup succeededDec 20 14:35:51 localhost kernel: klogd 1.3-3, log source = /proc/kmsgstarted.Dec 20 14:35:51 localhost kernel: Inspecting /boot/System.map-2.2.14-12Dec 20 14:35:51 localhost kernel: Loaded 7330 symbols from/boot/System.map-2.2.14-12.Dec 20 14:35:51 localhost kernel: Symbols match kernel version 2.2.14.Dec 20 14:35:51 localhost kernel: Loaded 187 symbols from 10 modules.Dec 20 14:35:51 localhost kernel: Linux version 2.2.14-12([EMAIL PROTECTED]) (gcc version egcs-2.91.66 19990314/Linux(egcs-1.1.2 release)) #1 Tue Apr 25 13:04:07 EDT 2000Dec 20 14:35:51 localhost kernel: Detected 400906147 Hz processor.Dec 20 14:35:51 localhost kernel: Console: colour VGA+ 80x25Dec 20 14:35:51 localhost kernel: Calibrating delay loop... 399.77 BogoMIPSDec 20 14:35:51 localhost kernel: Memory: 22356k/24512k available (1068kkernel code, 412k reserved, 612k data, 64k init, 0k bigmem)Dec 20 14:35:51 localhost kernel: Dentry hash table entries: 262144 (order9, 2048k)Dec 20 14:35:51 localhost kernel: Buffer cache hash table entries: 32768(order 5, 128k)Dec 20 14:35:51 localhost kernel: Page cache hash table entries: 8192 (order3, 32k)Dec 20 14:35:51 localhost kernel: VFS: Diskquotas version dquot_6.4.0initializedDec 20 14:35:51 localhost kernel: CPU: Intel Celeron (Mendocino) stepping 05Dec 20 14:35:51 localhost kernel: Checking 386/387 coupling... OK, FPU usingexception 16 error reporting.Dec 20 14:35:51 localhost kernel: Checking 'hlt' instruction... OK.Dec 20 14:35:51 localhost kernel: POSIX conformance testing by UNIFIXDec 20 14:35:51 localhost kernel: mtrr: v1.35a (19990819) Richard Gooch([EMAIL PROTECTED])Dec 20 14:35:51 localhost kernel: PCI: PCI BIOS revision 2.10 entry at0xfb300Dec 20 14:35:51 localhost kernel: PCI: Using configuration type 1Dec 20 14:35:51 localhost kernel: PCI: Probing PCI hardwareDec 20 14:35:51 localhost kernel: Linux NET4.0 for Linux 2.2Dec 20 14:35:51 localhost kernel: Based upon Swansea University ComputerSociety NET3.039Dec 20 14:35:51 localhost kernel: NET4: Unix domain sockets 1.0 for LinuxNET4.0.Dec 20 14:35:51 localhost kernel: NET4: Linux TCP/IP 1.0 for NET4.0Dec 20 14:35:51 localhost kernel: IP Protocols: ICMP, UDP, TCP, IGMPDec 20 14:35:51 localhost kernel: TCP: Hash tables configured (ehash 32768bhash 32768)Dec 20 14:35:51 localhost kernel: Initializing RT netlink socketDec 20 14:35:51 localhost kernel: Starting kswapd v 1.5Dec 20 14:35:51 localhost kernel: Detected PS/2 Mouse Port.Dec 20 14:35:51 localhost kernel: Serial driver version 4.27 with MANY_PORTSMULTIPORT SHARE_IRQ enabledDec 20 14:35:51 localhost kernel: ttyS00 at 0x03f8 (irq = 4) is a 16550ADec 20 14:35:51 localhost kernel: ttyS01 at 0x02f8 (irq = 3) is a 16550ADec 20 14:35:51 localhost kernel: pty: 256 Unix98 ptys configuredDec 20 14:35:51 localhost kernel: apm: BIOS version 1.2 Flags 0x07 (Driverversion 1.9)Dec 20 14:35:51 localhost kernel: Real Time Clock Driver v1.09Dec 20 14:35:51 localhost kernel: RAM disk driver initialized: 16 RAM disksof 4096K sizeDec 20 14:35:51 localhost kernel: SIS5513: IDE controller on PCI bus 00 dev01Dec 20 14:35:51 localhost kernel: SIS5513: not 100% native mode: will probeirqs laterDec 20 14:35:51 localhost kernel: ide0: BM-DMA at 0x4000-0x4007, BIOSsettings: hda:DMA, hdb:DMADec 20 14:35:51 localhost kernel: ide1: BM-DMA at 0x4008-0x400f, BIOSsettings: hdc:DMA, hdd:DMADec 20 14:35:51 localhost kernel: hda: S
Re: Linux 2.2.19pre2
On Wed, 20 Dec 2000, Andrea Arcangeli wrote: > On Thu, Dec 21, 2000 at 01:57:15AM +1100, Andrew Morton wrote: > > If a task is on two waitqueues at the same time it becomes a bug: > > if the outer waitqueue is non-exclusive and the inner is exclusive, > > Your 2.2.x won't allow that either. You set the > `current->task_exclusive = 1' and so you will get an exclusive > wakeups in both waitqueues. You simply cannot tell register in > two waitqueue and expect a non-exlusive wakeup in one and an > exclusive wakeup in the other one. Why not? Having a wake-all wakeup on one event and a wake-one wakeup on another kind of event seems perfectly reasonable to me at a first glance. Is there something I've overlooked ? > > Anyway, it's academic. davem would prefer that we do it properly > > and move the `exclusive' flag into the waitqueue head to avoid the > > task-on-two-waitqueues problem, as was done in 2.4. I think he's > > The fact you could mix non-exclusive and exlusive wakeups in the > same waitqueue was a feature not a misfeature. Then of course > you cannot register in two waitqueues one with wake-one and one > with wake-all but who does that anyways? The "who does that anyways" argument could also be said about mixing exclusive and non-exclusive wakeups in the same waitqueue. Doing this seems rather confusing ... would you know any application which needs this ? > I think the real reason for spearating the two things as davem > proposed is because otherwise we cannot register for a LIFO > wake-one in O(1) as we needed for accept. Do you have any reason to assume Davem and Linus lied about why they changed the semantics? ;) I'm pretty sure the reason why Linus and Davem changed the semantics is the same reason they mailed to this list ... ;) I think it would be good to have the same semantics in 2.2 as we have in 2.4. Using different semantics will probably only lead to confusion. If there is a good reason to go with the semantics Andrea proposed over the semantics Linus and Davem have in 2.4, we should probably either change 2.4 or use the 2.4 semantics in 2.2 anyway just to avoid the confusion... regards, Rik -- Hollywood goes for world dumbination, Trailer at 11. http://www.surriel.com/ http://www.conectiva.com/ http://distro.conectiva.com.br/ - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: iptables: "stateful inspection?"
"Michael H. Warfield" wrote: > You can use spf to add some stateful inspection for PORT mode > ftp. Personally, I like the masquerading option better, though. Can you give an example of using MASQ selectively? I have real addresses on both sides of the firewall, but want things like FTP to work correctly. I think the IPChains HOWTOs are just a little terse. :) Thanks! - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: 2.2.18 question (fh_lock_parent)
Apparently in the init scripts, sendmail starts before mounting /dev/sda1 .. but it never happened before changing kernels.. that's why it was using nfs instead of the scsi disk. (I'm smrt.) Thanks for the help though. :) ,.;:: : Michael J. Dikkema | Systems / Network Admin - Internet Solutions, Inc. | http://www.moot.ca Work: (204) 982-1060 ; [EMAIL PROTECTED] ',. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: getsockopt() with IP_PKTINFO not working?
On Wed, Dec 20, 2000 at 05:44:34PM +0100, Cord Seele wrote: > I am trying to get the destination address of an incoming udp packet > with getsockopt(). > According to the man pages flag IP_PKTINFO should do that. But it > doesn't work: > > struct in_pktinfo pktinfo; > socklen_t optlen; > struct in_addr local_addr; > > optlen=(socklen_t)sizeof(pktinfo); > syslog (LOG_ERR, "ERR %d", > getsockopt(fd, SOL_IP, IP_PKTINFO, &pktinfo, &optlen)); You're misreading the manpage. IP_PKTINFO just enables recvmsg() to pass ancillary messages that contain pktinfo structures. getsockopt IP_PKTINFO returns the state of the the IP_PKTINFO flag on the socket. So use setsockopt IP_PKTINFO to set the flag to true and then receive them using recvmsg(). -Andi - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: iptables: "stateful inspection?"
On Wed, Dec 20, 2000 at 08:51:34AM -0800, David Lang wrote: > On Wed, 20 Dec 2000, Michael Rothwell wrote: > > Date: Wed, 20 Dec 2000 11:30:15 -0500 > > From: Michael Rothwell <[EMAIL PROTECTED]> > > To: Michael H. Warfield <[EMAIL PROTECTED]> > > Cc: [EMAIL PROTECTED] > > Subject: Re: iptables: "stateful inspection?" > > "Michael H. Warfield" wrote: > > > I think that's more than a little overstatement on your > > > part. It depends entirely on the application you intend to put > > > it to. > > Fine. How do I make FTP work through it? How can I allow all outgoing > > TCP connections without opening the network to inbound connections on > > the ports of desired services? > > > for the issue of outbound TCP connections you can set ipchains filters > based on the Syn flag to prevent inbound connections. > for FTP I don't know off the top of my head how to do it when not > masquerading, when NAT is turned on load the FTP masq helper module and it > will allow you to do ftp out with no problems. You can use spf to add some stateful inspection for PORT mode ftp. Personally, I like the masquerading option better, though. > the real point that you need the stateful filtering is on UDP ports. for > that again I don't know any way when not doing NAT, but when NAT is > enabled it does do basic stateful filtering (but watch out for timeouts) Stateful filter also helps block FIN scans and other stealth scans, as well as some other esoteric attacks (fragmentation attacks, Ping'O Death, etc...). There are other ways to deal with those attacks as well, but stateful filtering helps. You also need it if you want to take advantage of some ICMP as well. Big thing for me about NetFilter over IPChains, in addition to statefull inspection, is the fact that we finally have an IPv6 aware firewall now. I've been chomping at the bit to get on IPv6 but couldn't till I had working firewall code for that. > David Lang > > > Yes it does. It's clearly stated in all the documentation > > > on netfilter and in it's design. Read the fine manual (or web site) > > > and you would have uncovered this (or been run over by it) for yourself. > > > > > > http://netfilter.filewatcher.org/ > > > > Thanks. > > > > -M Mike -- Michael H. Warfield| (770) 985-6132 | [EMAIL PROTECTED] (The Mad Wizard) | (678) 463-0932 | http://www.wittsend.com/mhw/ NIC whois: MHW9 | An optimist believes we live in the best of all PGP Key: 0xDF1DD471| possible worlds. A pessimist is sure of it! - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: iptables: "stateful inspection?"
On Wed, Dec 20, 2000 at 11:30:15AM -0500, Michael Rothwell wrote: > "Michael H. Warfield" wrote: > > I think that's more than a little overstatement on your > > part. It depends entirely on the application you intend to put > > it to. > Fine. How do I make FTP work through it? How can I allow all outgoing > TCP connections without opening the network to inbound connections on > the ports of desired services? Passive mode ftp works great for me. You can also tack spf on top of IPChains and get port mode working if that's really part of your requirements. If you really want to get sexy, you can use the MASQ code to masquarade and handle the FTP for you. Personally, I like the MASQ trick better than using spf and enabling PORT mode. Policy routing helps out there as well where you want to masquarade some services and let others pass untampered. (Actually you only REALLY need policy routing if you are also playing tricks with the routing when you masquarade.) I use policy routing anyways, so I can route outbound ftp and http out a big fat unreliable broadband pipe while protecting my static addresses through my nice reliable ISDN channels. Your second question doesn't even seem to make sense to me. Doesn't make sense as in either I don't understand your question or the answer is so obvious if I do. You allow outbound "SYN" packets and block all (or only allow appropriate) inbound "SYN" packets (-y option on the ipchains rules). Or did I misunderstand your question? In my case, inappropriate inbound SYN packets get portforwarded up to Abacus PortSentry on the firewall to deal with port scanners. Yes, that setup still does allow people to do "FIN" scans and other stealthy scans, but with Abacus PortSentry running in front of everything and shutting down rogue sites that try to scan me that's not a real great threat. The IDS behind the firewall also fires off if anyone tricky enough tries to stealth scan me WITHOUT an initial SYN half scan or full scan (which would cut them off). Snort, behind the firewall, deals with the next layer of ankle bitters that are just a little cut above the common riff raff that try to port scan me. Snort makes for yet another good adjunct to both IPChains or NetFilter and PortSentry. The combination is awesome for frontend filtering and detection. Anyone getting through that without tripping an alarm is NOT an amateur and is worthy of my full, undivided, PERSONAL attention (and I have custom detectors and surprises for that level of "talent" as well). :-)=) BTW... Before anyone raises the customary remark about "What about denial of service attacks by spoofing Abacus PortSentry"... No one has documented an effective DoS attack against PortSentry in the field. It's just too difficult to do and too easy to protect against. My "evil twin" David LeBlanc (when he was still working with me at Internet Security Systems a couple of years ago) tried it against my PortSentry protected workstation. He failed. He knew everything I had on that system including the PortSentry configuration and never once managed to spoof so much as a single DoS attack that was effective. If he couldn't do it with his level of talent and his knowledge of my systems, it's going to take a world class talent who already knows my entire setup to make that happen. At that point, I have bigger problems than worrying about PortSentry (and it's also a tip-off from PortSentry that I need to be worried). It would take a lot of effort and a lot of incentive and a lot of access to make a real one happen. If you have all three of those, there are easier DoS attacks than attacking PortSentry. Lots of them that are LOTS easier. > > Yes it does. It's clearly stated in all the documentation > > on netfilter and in it's design. Read the fine manual (or web site) > > and you would have uncovered this (or been run over by it) for yourself. > > http://netfilter.filewatcher.org/ > Thanks. No problem. > -M Mike -- Michael H. Warfield| (770) 985-6132 | [EMAIL PROTECTED] (The Mad Wizard) | (678) 463-0932 | http://www.wittsend.com/mhw/ NIC whois: MHW9 | An optimist believes we live in the best of all PGP Key: 0xDF1DD471| possible worlds. A pessimist is sure of it! - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
panic with squid's pinger
Hi! Reproducible panic when squid gets the first request. Always at the same place in the pinger process. test12, test13-pre3 fail, but test12 runs fine on another machine with nearly the same config (netcard and disk drivers differ, and the failing machine has devfs). Hardware: Compaq proliant dl360 with a quad starfire card, UP (Serverworks chipset), cpqarray. Unable to handle kernel NULL pointer dereference at virtual address 003c c01a20de *pde = Oops: CPU:0 EIP:0010:[] Using defaults from ksymoops -t elf32-i386 -a i386 EFLAGS: 00010246 eax: ebx: ecx: c7b5e9e0 edx: c7b5e9e0 esi: 1fb1 edi: c7b09c00 ebp: 1df0 esp: c6e4bc40 ds: 0018 es: 0018 ss: 0018 Process pinger (pid: 289, stackpage=c6e4b000) Stack: c7b5e9e0 8b5e 017f 0014 c01a2453 c7b5e9e0 c7b09c00 c6e4e500 c7b09c00 c01a4db8 c6e4bd44 11e4bc84 c01c5770 c7b09c00 c6e4bd34 c029d998 c01c5089 c7b09c00 c6e4bd34 c029d998 c01a4db8 Call Trace: [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] [] Code: 8b 40 3c 89 41 3c 8b 47 5c c7 47 18 00 00 00 00 01 41 18 8b >>EIP; c01a20de<= Trace; c01a2453 Trace; c01a4db8 Trace; c01c5770 Trace; c01c5089 Trace; c01a4db8 Trace; c0108dc8 Trace; c01c44e0 Trace; c01a4db8 Trace; c018ea5c Trace; c01a4db8 Trace; c01a4db8 Trace; c018ec73 Trace; c01a4db8 Trace; c01a440b Trace; c01a4db8 Trace; c01b82cc Trace; c01a450e Trace; c01b82cc Trace; c01b8725 Trace; c01b82cc Trace; c018815b Trace; c0188db0 Trace; c01b7aec Trace; c01bd5d6 Trace; c01857d5 Trace; c018641c Trace; c0128a37 <__free_pages+13/14> Trace; c0128a72 Trace; c013a6da Trace; c013a9d8 Trace; c018645a Trace; c0186bb1 Trace; c0108d1f Code; c01a20de <_EIP>: Code; c01a20de<= 0: 8b 40 3c mov0x3c(%eax),%eax <= Code; c01a20e1 3: 89 41 3c mov%eax,0x3c(%ecx) Code; c01a20e4 6: 8b 47 5c mov0x5c(%edi),%eax Code; c01a20e7 9: c7 47 18 00 00 00 00 movl $0x0,0x18(%edi) Code; c01a20ee 10: 01 41 18 add%eax,0x18(%ecx) Code; c01a20f1 13: 8b 00 mov(%eax),%eax Kernel panic: Aiee, killing interrupt handler! Does anyone know, what this can be? If any other information would be needed, please tell me! Thanks. -- Dani ...and Linux for all. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: iptables: "stateful inspection?"
On Wed, 20 Dec 2000, Michael Rothwell wrote: > Date: Wed, 20 Dec 2000 11:30:15 -0500 > From: Michael Rothwell <[EMAIL PROTECTED]> > To: Michael H. Warfield <[EMAIL PROTECTED]> > Cc: [EMAIL PROTECTED] > Subject: Re: iptables: "stateful inspection?" > > "Michael H. Warfield" wrote: > > I think that's more than a little overstatement on your > > part. It depends entirely on the application you intend to put > > it to. > > Fine. How do I make FTP work through it? How can I allow all outgoing > TCP connections without opening the network to inbound connections on > the ports of desired services? > for the issue of outbound TCP connections you can set ipchains filters based on the Syn flag to prevent inbound connections. for FTP I don't know off the top of my head how to do it when not masquerading, when NAT is turned on load the FTP masq helper module and it will allow you to do ftp out with no problems. the real point that you need the stateful filtering is on UDP ports. for that again I don't know any way when not doing NAT, but when NAT is enabled it does do basic stateful filtering (but watch out for timeouts) David Lang > > Yes it does. It's clearly stated in all the documentation > > on netfilter and in it's design. Read the fine manual (or web site) > > and you would have uncovered this (or been run over by it) for yourself. > > > > http://netfilter.filewatcher.org/ > > Thanks. > > -M > - > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to [EMAIL PROTECTED] > Please read the FAQ at http://www.tux.org/lkml/ > - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
getsockopt() with IP_PKTINFO not working?
I am trying to get the destination address of an incoming udp packet with getsockopt(). According to the man pages flag IP_PKTINFO should do that. But it doesn't work: struct in_pktinfo pktinfo; socklen_t optlen; struct in_addr local_addr; optlen=(socklen_t)sizeof(pktinfo); syslog (LOG_ERR, "ERR %d", getsockopt(fd, SOL_IP, IP_PKTINFO, &pktinfo, &optlen)); syslog (LOG_ERR, "LENGTH %d %d", (int)optlen, sizeof(pktinfo)); local_addr=pktinfo.ipi_addr; syslog (LOG_ERR,"ADDR %s",inet_ntoa(local_addr)); results in /var/log/messages: Dec 19 19:27:49 coda tftpd[20081]: ERR 0 Dec 19 19:27:49 coda tftpd[20081]: LENGTH 4 12 Dec 19 19:27:49 coda tftpd[20081]: ADDR 232.252.255.191 While getsockopt() returns no error, the resulting length is too short and the addr is definitely invalid. I would expect either getsockopt() to return -1, it this is not implemented or return at least 12 valid bytes. (I am running a 2.2.16 kernel with glibc 2.1.3.) I even tried the 'hard way' using recvmsg() but the resulting msg_controllen == 0. Background: If a machine has more than one address in a single network, i.e. eth0 192.168.0.10 eth0:1 192.168.0.20 a call to bind() normally assigns the primary ip address (.10) to the socket. If the server was addressed on his second address (.20) the request is not answered and fails. I have this problem with tftpd. Or is there a better way to get the destination address of an incoming udp packet? Thanks for any help. Cord Seele PS: Please CC me since I am not an the list. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Extreme IDE slowdown with 2.2.18
Hello, I'm having problems with the performance of my harddrives after I upgraded my kernel from 2.2.17 to 2.2.18. The performancedrop is noticable on every IDE drive. Here are some numbers to show what I mean: 2.2.17: /dev/hdc: Timing buffered disk reads: 64 MB in 4.32 seconds =14.81 MB/sec 2.2.18: /dev/hdc: Timing buffered disk reads: 64 MB in 10.49 seconds = 6.10 MB/sec These are hdparm -t outputs and the performance drop is pretty noticable :-/ I also copied a 600Mb file from my HDD to /dev/null and the results were: 2.2.17: 1 minute 9 seconds 2.2.18: 1 minute 38 seconds My system consists of: FIC VA-503+ motherboard with the MVP3 chipset K6-2 500MHz 128Mb SDRAM 3 IDE disks (see below) Slackware 7.0 dmesg output for the IDE system (no differences between .17 and .18): VP_IDE: IDE controller on PCI bus 00 dev 39 VP_IDE: not 100% native mode: will probe irqs later ide0: BM-DMA at 0xe400-0xe407, BIOS settings: hda:DMA, hdb:DMA ide1: BM-DMA at 0xe408-0xe40f, BIOS settings: hdc:DMA, hdd:DMA hda: QUANTUM FIREBALL ST6.4A, ATA DISK drive hdb: QUANTUM FIREBALL SE4.3A, ATA DISK drive hdc: IBM-DJNA-352030, ATA DISK drive ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 ide1 at 0x170-0x177,0x376 on irq 15 hda: QUANTUM FIREBALL ST6.4A, 6149MB w/81kB Cache, CHS=784/255/63 hdb: QUANTUM FIREBALL SE4.3A, 4110MB w/80kB Cache, CHS=524/255/63 hdc: IBM-DJNA-352030, 19470MB w/1966kB Cache, CHS=39560/16/63 When I performed the tests I used similiar .17 and .18 kernels with a minimum components included. No network, SCSI, sound and such things. .config files can be supplied if needed. Does anyone know what could be wrong? Have I forgot something? Is this a known problem with the 2.2.18 kernel? Thanks in advance! Robert - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: [Patch] performance enhancement for simple_strtoul
Hello, I thought about that. This would be my recommendation for glibc where the general public may be doing scientific applications. But this is the kernel and there are people that would reject my patch purely on the basis that it adds precious bytes to the kernel. But since the kernel is "controllable" & printf() and its variants only support 8, 10, & 16, perhaps a better solution might be to trap the odd case and write something for it if its that important, or simply don't allow it. The base guessing part at the beginning of the function only supports base 8, 10, & 16. Therefore, the only way to require another base is to specify it in the function call (param - unsigned int base). A quick scan of the current linux source shows no one using something odd. So... If the maintainers of vsprintf.c want support for all number bases, that's fine with me. Just say the word & I'll gen up another patch...but it will be more bytes. Cheers, Steve Grubb - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: iptables: "stateful inspection?"
"Michael H. Warfield" wrote: > I think that's more than a little overstatement on your > part. It depends entirely on the application you intend to put > it to. Fine. How do I make FTP work through it? How can I allow all outgoing TCP connections without opening the network to inbound connections on the ports of desired services? > Yes it does. It's clearly stated in all the documentation > on netfilter and in it's design. Read the fine manual (or web site) > and you would have uncovered this (or been run over by it) for yourself. > > http://netfilter.filewatcher.org/ Thanks. -M - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: iptables: "stateful inspection?"
On Wed, Dec 20, 2000 at 11:18:10AM -0500, Michael Rothwell wrote: > IPChains is essentially useless as a firewall due to its lack of I think that's more than a little overstatement on your part. It depends entirely on the application you intend to put it to. It may be entirely useless TO YOU and your applications, but your statement is far to broad to be accurate. > stateful packet filering. Will the IPTables code in 2.4 maintain > connection state? Yes it does. It's clearly stated in all the documentation on netfilter and in it's design. Read the fine manual (or web site) and you would have uncovered this (or been run over by it) for yourself. http://netfilter.filewatcher.org/ > -M Mike -- Michael H. Warfield| (770) 985-6132 | [EMAIL PROTECTED] (The Mad Wizard) | (678) 463-0932 | http://www.wittsend.com/mhw/ NIC whois: MHW9 | An optimist believes we live in the best of all PGP Key: 0xDF1DD471| possible worlds. A pessimist is sure of it! - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
iptables: "stateful inspection?"
IPChains is essentially useless as a firewall due to its lack of stateful packet filering. Will the IPTables code in 2.4 maintain connection state? -M - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: [Patch] performance enhancement for simple_strtoul
On Wed, Dec 20, 2000 at 09:09:03AM -0500, Steve Grubb wrote: > Hello, > > The following patch is a faster implementation of the simple_strtoul > function. [snip] Why not preserve the existing code for bases other than 8, 10, and 16? Admittedly, the only other case that is likely to be used would be base 2, but surely there's only a penalty of a few dozen bytes for the following code.. > - while (isxdigit(*cp) && (value = isdigit(*cp) ? *cp-'0' : (islower(*cp) > - ? toupper(*cp) : *cp)-'A'+10) < base) { > - result = result*base + value; > - cp++; > - } Jeff - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
[PATCH] ne2k-pci
[Sorry if this message is dup. I sent it to the list but did not see it, so I think this was a problem with my mailer or ISP] This is a port of the new 1.02 features of the ne2k-pci driver to kernel 2.2.18. Reviewed by Donald Becker, and tested on 2.2.18, 19-pre1 and pre2. New features: - module info - full duplex support in RealTek and Holtek cards. Must be activated on module load with full_duplex=1, (up to 6 cards). The patch does not use the pci stuff that D.Becker included in the scyld drivers ===> Question: is there any similar pci stuff in 2.2.18 and a driver to look at its use ? Thanks.ñ patch-ne2k-pci --- linux/drivers/net/ne2k-pci.c.orgSun Dec 17 01:51:04 2000 +++ linux/drivers/net/ne2k-pci.cTue Dec 19 01:48:00 2000 @@ -12,19 +12,31 @@ This software may be used and distributed according to the terms of the GNU Public License, incorporated herein by reference. - The author may be reached as [EMAIL PROTECTED], or C/O - Center of Excellence in Space Data and Information Sciences - Code 930.5, Goddard Space Flight Center, Greenbelt MD 20771 + Drivers based on or derived from this code fall under the GPL and must + retain the authorship, copyright and license notice. This file is not + a complete program and may only be used when the entire operating + system is licensed under the GPL. + + The author may be reached as [EMAIL PROTECTED], or C/O + Scyld Computing Corporation + 410 Severn Ave., Suite 210 + Annapolis MD 21403 + Issues remaining: People are making PCI ne2000 clones! Oh the horror, the horror... + Limited full-duplex support. - Issues remaining: - No full-duplex support. + ChangeLog: + + 12/15/2000 Merged Scyld v1.02 into 2.2.18 + J.A. Magallon +<[EMAIL PROTECTED]> */ -/* Our copyright info must remain in the binary. */ -static const char *version = -"ne2k-pci.c:vpre-1.00e 5/27/99 D. Becker/P. Gortmaker http://cesdis.gsfc.nasa.gov/linux/drivers/ne2k-pci.html\n"; +/* These identify the driver base version and may not be removed. */ +static const char version1[] = +"ne2k-pci.c:v1.02 10/19/2000 D. Becker/P. Gortmaker\n"; +static const char version2[] = +" http://www.scyld.com/network/ne2k-pci.html\n"; #include #include @@ -49,7 +61,18 @@ #endif /* Set statically or when loading the driver module. */ -static int debug = 1; +static int debug = 1; /* 1 normal messages, 0 quiet .. 7 verbose. */ +/* More are supported, limit only on options */ +#define MAX_UNITS 6 +/* Used to pass the full-duplex flag, etc. */ +static int full_duplex[MAX_UNITS] = {0, }; +static int options[MAX_UNITS] = {0, }; + +MODULE_AUTHOR("Donald Becker / Paul Gortmaker"); +MODULE_DESCRIPTION("PCI NE2000 clone driver"); +MODULE_PARM(debug, "i"); +MODULE_PARM(options, "1-" __MODULE_STRING(MAX_UNITS) "i"); +MODULE_PARM(full_duplex, "1-" __MODULE_STRING(MAX_UNITS) "i"); /* Some defines that people can play with if so inclined. */ @@ -62,12 +85,13 @@ /* Do we have a non std. amount of memory? (in units of 256 byte pages) */ /* #define PACKETBUF_MEMSIZE 0x40 */ -#define ne2k_flags reg0/* Rename an existing field to store flags! */ - -/* Only the low 8 bits are usable for non-init-time flags! */ +/* Flags. We rename an existing ei_status field to store flags! */ +/* Thus only the low 8 bits are usable for non-init-time flags. */ +#define ne2k_flags reg0 enum { - HOLTEK_FDX=1, /* Full duplex -> set 0x80 at offset 0x20. */ - ONLY_16BIT_IO=2, ONLY_32BIT_IO=4, /* Chip can do only 16/32-bit xfers. */ + ONLY_16BIT_IO=8, ONLY_32BIT_IO=4, /* Chip can do only 16/32-bit xfers. */ + FORCE_FDX=0x20, /* User override. */ + REALTEK_FDX=0x40, HOLTEK_FDX=0x80, STOP_PG_0x60=0x100, }; @@ -79,18 +103,17 @@ int flags; } pci_clone_list[] __initdata = { - {0x10ec, 0x8029, "RealTek RTL-8029", 0}, - {0x1050, 0x0940, "Winbond 89C940", 0}, - {0x11f6, 0x1401, "Compex RL2000", 0}, - {0x8e2e, 0x3000, "KTI ET32P2", 0}, - {0x4a14, 0x5000, "NetVin NV5000SC", 0}, - {0x1106, 0x0926, "Via 86C926", ONLY_16BIT_IO}, - {0x10bd, 0x0e34, "SureCom NE34", 0}, - {0x1050, 0x5a5a, "Winbond", 0}, - {0x12c3, 0x0058, "Holtek HT80232", ONLY_16BIT_IO | HOLTEK_FDX}, - {0x12c3, 0x5598, "Holtek HT80229", -ONLY_32BIT_IO | HOLTEK_FDX | STOP_PG_0x60 }, - {0,} +{0x10ec, 0x8029, "RealTek RTL-8029", REALTEK_FDX}, +{0x1050, 0x0940, "Winbond 89C940", 0}, +{0x1050, 0x5a5a, "Winbond w89c940", 0}, +{0x8e2e, 0x3000, "KTI ET32P2", 0}, +{0x4a14, 0x5000, "NetVin NV5000SC", 0}, +{0x1106, 0x0926, "Via 86C926", ONLY_16BIT_IO}, +{0x10bd, 0x0e34, "SureCom NE34", 0}, +{0x12c3, 0x0058, "Holtek HT80232", ONLY_16BIT_IO|HOLTEK_FDX}, +{0x12c3, 0x5598, "Holtek HT8
Anyone have an OXSEMI PCI parallel port card?
Does anyone have an OXSEMI PCI parallel port PCI card with device id 0x9513 (i.e. an OXSEMI 16PCI954)? I'd like to get them to try out something.. Thanks, Tim. */ PGP signature
Re: Oops with 2.4.0-test13pre3 - swapoff
> Zdenek Kabelac wrote: > > This is oops I've got when rebooting after some heavy disk activity on > > my SMP system: > > > > Written by hand: > > > > kernel BUG swap_state.c:78! > [snip] > > Same here during a halt of a RH 6.2 based K6-2 500 MHz > UP machine running lk240t13p3. The machine had been on > for a while and had built a kernel amongst other things. > I'll just append that my machine has been up for just several minutes (maybe 10) but has been doing heavy copying - several 600MB files between some partitions. So maybe the problem with memory thrashing is still not fully fixed ??? -- There are three types of people in the world: those who can count, and those who can't. Zdenek Kabelac http://i.am/kabi/ [EMAIL PROTECTED] {debian.org; fi.muni.cz} - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: 2.2.18 question (fh_lock_parent)
On Wed, 20 Dec 2000, Neil Brown wrote: > On Tuesday December 19, [EMAIL PROTECTED] wrote: > > > > I've been getting tonnes of these since I installed 2.2.18. Is this a > > problem? Should I even worry about this? If I don't need to worry about > > it, is there a way to stop displaying this message? > > > > fh_lock_parent: mqueue/xfBAA14279 parent changed or child unhashed > > fh_lock_parent: mqueue/xfBAA16413 parent changed or child unhashed > > You are running sendmail on an NFS client with /var/spool mounted off > the NFS server which is giving these message - right? > > These messages tend to indicate a race between two different NFS > requests that try to do something to the one file - probably unlink > it, though possibly rename it. That's what you'd think. All these machines have a drive mounted on /var. [root@www /root]# df Filesystem 1k-blocks Used Available Use% Mounted on 10.0.0.10:/nfsroot34020868 24061800 8230876 75% / /dev/sda1 8744304 1473776 6826336 18% /var I've checked to see if mtab was lying to me, and it's not.. I can unmount and remount the drive on /var.. The system runs with one nfs server, and 4 children, all using nfs root. They all mount /dev/sda1 on /var. Yet for some reason I get these messages. I went from 2.2.16 -> 2.2.18 without changing anything but the kernel. Has something changed? This doesn't look right to me at all. ,.;:: : Michael J. Dikkema | Systems / Network Admin - Internet Solutions, Inc. | http://www.moot.ca Work: (204) 982-1060 ; [EMAIL PROTECTED] ',. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Linux 2.2.19pre2
On Thu, Dec 21, 2000 at 01:57:15AM +1100, Andrew Morton wrote: > If a task is on two waitqueues at the same time it becomes a bug: > if the outer waitqueue is non-exclusive and the inner is exclusive, Your 2.2.x won't allow that either. You set the `current->task_exclusive = 1' and so you will get an exclusive wakeups in both waitqueues. You simply cannot tell register in two waitqueue and expect a non-exlusive wakeup in one and an exclusive wakeup in the other one. The only design difference (non implementation difference) between my patch and your patch is that you have to clear task_exlusive explicitly. You moved the EXCLUSIVE bitflag out of current->state field. That gives no advantages and it looks ugiler to me. The robusteness point doesn't hold IMHO: as soon as current->state is been changed by somebody else you don't care anymore if it was exclusive wakeup or not. About the fact I mask out the exlusive bit in schedule that's zero cost compared to a cacheline miss, but it also depends if you have more wakeups or schedules (with accept(2) usage there are going to be more schedule than wakeups, but on other usages that could not be the case) but ok, the performance point was nearly irrelevant ;). > Anyway, it's academic. davem would prefer that we do it properly > and move the `exclusive' flag into the waitqueue head to avoid the > task-on-two-waitqueues problem, as was done in 2.4. I think he's The fact you could mix non-exclusive and exlusive wakeups in the same waitqueue was a feature not a misfeature. Then of course you cannot register in two waitqueues one with wake-one and one with wake-all but who does that anyways? Definitely not an issue for 2.2.x. I think the real reason for spearating the two things as davem proposed is because otherwise we cannot register for a LIFO wake-one in O(1) as we needed for accept. Other thing about your patch, adding TASK_EXCLUSIVE to wake_up/wake_up_interruptible is useless. You kind of mixed the two things at the source level. In your patch TASK_EXCLUSIVE should not be defined. Last thing the wmb() in accept wasn't necessary. At that point you don't care at all what the wakeup can see. Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
[Test Case] performance enhancement for simple_strtoul
Hello, Here's the test case for the suggested simple_strtoul function. I just finished testing on a P3 where it seems to show a 16-20% speed improvement over the current algorithm. compile it as: gcc /usr/src/linux/lib/ctype.c strtoul_test.c -o strtoul_test You can change the numeric base value with this define to 8, 10, or 16 to see the speed change for each numeric representation: #define BASE 10 Have fun, Steve Grubb --strtoul_test.c-- #include #include #include #include #include #include struct timeval last_stopwatch_time; void stopwatch() { struct timeval now; int delta; gettimeofday(&now, 0); delta = (now.tv_sec - last_stopwatch_time.tv_sec) * 100 + (now.tv_usec - last_stopwatch_time.tv_usec); printf ("Stopwatch: elapsed time %d.%03d seconds\n\n", delta / 100, (delta / 1000) % 1000); last_stopwatch_time = now; } unsigned long old_simple_strtoul(const char *cp,char **endp,unsigned int base) { unsigned long result = 0,value; if (!base) { base = 10; if (*cp == '0') { base = 8; cp++; if ((*cp == 'x') && isxdigit(cp[1])) { cp++; base = 16; } } } while (isxdigit(*cp) && (value = isdigit(*cp) ? *cp-'0' : (islower(*cp) ? toupper(*cp) : *cp)-'A'+10) < base) { result = result*base + value; cp++; } if (endp) *endp = (char *)cp; return result; } unsigned long new_simple_strtoul2(const char *cp,char **endp,unsigned int base) { unsigned char c; unsigned long result = 0; if (!base) { base = 10; if (*cp == '0') { base = 8; cp++; if ((*cp == 'x') && isxdigit(cp[1])) { cp++; base = 16; } } } c = *cp; switch (base) { case 10: while (isdigit(c)) { result = (result*10) + (c & 0x0f); c = *(++cp); } break; case 16: while (isxdigit(c)) { result = (result<<4); if (c&0x40) result += (c & 0x07) + 9; else result += (c & 0x0f); c = *(++cp); } break; case 8: while (isdigit(c)) { if ((c&0x37) == c) result = (result<<3) + (c & 0x07); else break; c = *(++cp); } } if (endp) *endp = (char *)cp; return result; } #define NUMBER_TO_TEST 32768 #define BASE 10 char f[3][3] = { "%d", "%X", "%o"}; char str[NUMBER_TO_TEST][32]; unsigned long r[NUMBER_TO_TEST]; int main() { int rn, i, j, iterations = 1000; time_t tm; time(&tm); srand((unsigned) tm); rn = rand(); // do setup here for (i=0; ihttp://www.tux.org/lkml/
Re: [2.2.18] VM: do_try_to_free_pages failed
On Wed, 20 Dec 2000, Alan Cox wrote: > > How can I get rid of those do_try_to_free_pages lockups? That box > > exports root file systems for some SparcStation 2 that are used as X > > terminals, so it's pretty important I keep that box running. > > > > Should I try the most recent 2.2.19-pre? > > 2.2.19pre2 should resolve that problem I'll give that a try. Thanks to you and Ville Herva for replying. -- Matthias Andree - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Weird vmstat reports in 2.2.18
I'm getting some strange reports with vmstat on a dual iPPro running 2.2.18, it doesnt happen very frequently, but i see it a lot when compiling something (kernel and mysql specially, not when compiling small stuff), though it doesnt look like a high-load issue. When the machine is idle (ie. most of the time at the moment) it doesnt show up. The report is like this: #vmstat 1 60 | awk '{ print $16 }' id 0 0 20452224 1 20452224 0 1 20452224 1 0 0 I wasnt able to trigger it in a predictable way, it just pops up... BUT if i open two vmstats in different consoles.. the number doesnt show up in both, just in one of them... so i'm not sure at all if this is a kernel bug, or just another (vmstat?) feature =) I also found a reference to something similar in http://www.uwsg.iu.edu/hypermail/linux/kernel/0009.3/0273.html Alberto - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Linux 2.2.19pre2
Andrea Arcangeli wrote: > > > o wake_one semantics for accept() (Andrew Morton) > > I dislike the implementation. I stick with my faster and nicer implementation > that was just included in aa kernels for some time (2.2.18aa2 included it for > example). Andrew, I guess you didn't noticed I just implemented the wakeone for > accept. (I just ported it on top of 2.2.19pre2 after backing out the wakeone in > pre2) Yes, I noticed your implementation a few weeks back. 'fraid I never liked the use of the TASK_EXCLUSIVE bit in task_struct.state. It's an enumerated state, not an aggregation of flags. Most of the kernel treats it as an enumerated state and so will happily clear the TASK_EXCLUSIVE bit without masking it off. Fragile. If a task is on two waitqueues at the same time it becomes a bug: if the outer waitqueue is non-exclusive and the inner is exclusive, the task suddenly becomes exclusive on the outer one and converts it from wake-all to wake-some! Is it faster? Not sure. You've saved a cacheline read in __wake_up (which was in fact a preload, if you look at what comes later) at the cost of having to mask out the bit in current->state every time we schedule(). Anyway, it's academic. davem would prefer that we do it properly and move the `exclusive' flag into the waitqueue head to avoid the task-on-two-waitqueues problem, as was done in 2.4. I think he's right. Do you? - - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Oops with 2.4.0-test13pre3 - swapoff
Zdenek Kabelac wrote: > This is oops I've got when rebooting after some heavy disk activity on > my SMP system: > > Written by hand: > > kernel BUG swap_state.c:78! [snip] Same here during a halt of a RH 6.2 based K6-2 500 MHz UP machine running lk240t13p3. The machine had been on for a while and had built a kernel amongst other things. Lead up was: $ halt . Sending all processes the KILL signal[OK] Turning off swap VM: __lru_cache_del, found unknown page ?! kernel BUG at swap_state.c:78 Doug Gilbert - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
[Patch] performance enhancement for simple_strtoul
Hello, The following patch is a faster implementation of the simple_strtoul function. This function differs from the original in that it reduces the multiplies to shifts and logical operations wherever possible. My testing shows that it adds around 100 bytes, but is about 6% faster on a K6-2. (It is 40% faster that glibc's strtoul, but that's a different story.) My guess is that the performance gain will be higher on platforms with slower multiplication instructions. I have tested it for numerical accuracy so I think this is safe to apply. If anyone is interested, I can also supply a test application that demonstrates the performance gain. This patch was generated against 2.2.16, but should apply to 2.2.19 cleanly. In 2.4.0-test9, simple_strtoul starts on line 19 rather than 17, hopefully that's not a problem. Cheers, Steve Grubb - --- lib/vsprintf.orig Fri Dec 1 08:58:02 2000 +++ lib/vsprintf.c Wed Dec 20 08:42:26 2000 @@ -14,10 +14,13 @@ #include #include +/* +* This function converts base 8, 10, or 16 only - Steve Grubb +*/ unsigned long simple_strtoul(const char *cp,char **endp,unsigned int base) { - unsigned long result = 0,value; - + unsigned char c; + unsigned long result = 0; if (!base) { base = 10; if (*cp == '0') { @@ -29,11 +32,36 @@ } } } - while (isxdigit(*cp) && (value = isdigit(*cp) ? *cp-'0' : (islower(*cp) - ? toupper(*cp) : *cp)-'A'+10) < base) { - result = result*base + value; - cp++; - } + c = *cp; +switch (base) { +case 10: +while (isdigit(c)) { +result = (result*10) + (c & 0x0f); +c = *(++cp); +} +break; +case 16: +while (isxdigit(c)) { +result = result<<4; +if (c&0x40) + result += (c & 0x07) + 9; +else +result += c & 0x0f; +c = *(++cp); +} +break; +case 8: +while (isdigit(c)) { +if ((c&0x37) == c) +result = (result<<3) + (c & 0x07); +else +break; +c = *(++cp); +} +break; +default: /* Is anything else used by the kernel? */ +break; +} if (endp) *endp = (char *)cp; return result; - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Linux 2.2.19pre2
On Sat, Dec 16, 2000 at 07:11:47PM +, Alan Cox wrote: > o E820 memory detect backport from 2.4(Michael Chen) It's broken, it will crash machines: for (i = 0; i < e820.nr_map; i++) { unsigned long start, end; /* RAM? */ if (e820.map[i].type != E820_RAM) continue; start = PFN_UP(e820.map[i].addr); end = PFN_DOWN(e820.map[i].addr + e820.map[i].size); if (start >= end) continue; if (end > max_pfn) max_pfn = end; } memory_end = (max_pfn << PAGE_SHIFT); this will threat non-RAM holes as RAM. On 2.4.x we do a different things, that is we collect the max_pfn but then we don't assume that there are no holes between 1M and max_pfn ;), we instead fill the bootmem allocator _only_ with E820_RAM segments. I was in the process of fixing this (I also just backported the thinkpad %edx clobber fix), but if somebody is going to work on this please let me know so we stay in sync. > o wake_one semantics for accept() (Andrew Morton) I dislike the implementation. I stick with my faster and nicer implementation that was just included in aa kernels for some time (2.2.18aa2 included it for example). Andrew, I guess you didn't noticed I just implemented the wakeone for accept. (I just ported it on top of 2.2.19pre2 after backing out the wakeone in pre2) Andrea - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: [2.2.18] VM: do_try_to_free_pages failed
> How can I get rid of those do_try_to_free_pages lockups? That box > exports root file systems for some SparcStation 2 that are used as X > terminals, so it's pretty important I keep that box running. > > Should I try the most recent 2.2.19-pre? 2.2.19pre2 should resolve that problem - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
2.2.19/2.4.0-test and usbdevfs
Hi ! I use the usb bus. Mostly it is for a mouse and today, I have installed a scanner. I've read the documentation (yes!) in /usr/src/linux/Documentation/usb/scanner.txt There is a mention about the usbdevfs and an advice to add a line in /etc/fstab. I have added the line "none /proc/bus/usb usbdevfs defaults 0 0", I expected to have access to /proc/bus/usb/devices without any extra manipulation. # /etc/fstab: static file system information. # # /dev/hda2 / ext2 defaults,errors=remount-ro 0 1 /dev/hdc1 none swap sw 0 0 proc/proc proc defaults 0 0 none/dev/shm shmdefaults 0 0 none/proc/bus/usb usbdevfs defaults 0 0 At boot time, I get the messages : Mounting local file systems... modprobe: modprobe: Can't locate module shm mount: fs type shm not supported by kernel modprobe: modprobe: Can't locate module usbdevfs mount: fs type usbdevfs not supported by kernel Running dns-clean. It is ok fr shm : I have the same fstab for both 2.2.19pre and 2.4.0-test and I boot 2.2.19pre... And the I've nothing in /proc/bus/usb If I enter the command 'by hand' : mount -t usbdevfs /proc/bus/usb /proc/bus/usb The system accepts the command and I can do a cat /proc/bus/usb/devices : [root@debian-f5ibh] ~ # mount -t usbdevfs /proc/bus/usb /proc/bus/usb [root@debian-f5ibh] ~ # cat /proc/bus/usb/devices T: Bus=01 Lev=00 Prnt=00 Port=00 Cnt=00 Dev#= 1 Spd=12 MxCh= 2 B: Alloc= 0/900 us ( 0%), #Int= 0, #Iso= 0 D: Ver= 1.00 Cls=09(hub ) Sub=00 Prot=00 MxPS= 8 #Cfgs= 1 P: Vendor= ProdID= Rev= 0.00 S: Product=USB UHCI Root Hub S: SerialNumber=6100 C:* #Ifs= 1 Cfg#= 1 Atr=40 MxPwr= 0mA I: If#= 0 Alt= 0 #EPs= 1 Cls=09(hub ) Sub=00 Prot=00 Driver=hub E: Ad=81(I) Atr=03(Int.) MxPS= 8 Ivl=255ms T: Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 2 Spd=12 MxCh= 0 D: Ver= 1.10 Cls=ff(vend.) Sub=ff Prot=ff MxPS=64 #Cfgs= 1 P: Vendor=04b8 ProdID=010a Rev= 1.04 S: Manufacturer=EPSON S: Product=Perfection1640 C:* #Ifs= 1 Cfg#= 1 Atr=40 MxPwr= 2mA I: If#= 0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none) E: Ad=81(I) Atr=02(Bulk) MxPS= 64 Ivl= 0ms E: Ad=02(O) Atr=02(Bulk) MxPS= 64 Ivl= 0ms T: Bus=01 Lev=01 Prnt=01 Port=01 Cnt=02 Dev#= 3 Spd=1.5 MxCh= 0 D: Ver= 1.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS= 8 #Cfgs= 1 P: Vendor=04b4 ProdID=0001 Rev= 0.00 S: Manufacturer=Cypress Sem. S: Product=Cypress USB Mouse C:* #Ifs= 1 Cfg#= 1 Atr=a0 MxPwr=100mA I: If#= 0 Alt= 0 #EPs= 1 Cls=03(HID ) Sub=01 Prot=02 Driver=hid E: Ad=81(I) Atr=03(Int.) MxPS= 3 Ivl= 10ms [root@debian-f5ibh] ~ # cat /proc/bus/usb/drivers hid hub usbdevfs Do I missed something in my usb configuration or in my modules managements ? Remark : The problem. is the same with 2.4.0-test13 System is : --- Pentium 200MMX /64 Mb with : -- Versions installed: (if some fields are empty or look -- unusual then possibly you have very old versions) Linux debian-f5ibh 2.2.19pre1 #1 sam déc 16 10:30:50 CET 2000 i586 unknown Kernel modules 2.3.23 Gnu C 2.95.2 Binutils 2.9.5.0.41 Linux C Library2.1.3 Dynamic linker ldd: version 1.9.11 Procps 2.0.6 Mount 2.10o Net-tools 2.05 Console-tools 0.2.3 Sh-utils 2.0 Modules Loaded af_packet scc ax25 parport_pc lp parport mousedev usb-uhci hid usbcore input autofs lockd sunrpc unix serial .config file is (many lines removed) : --- # CONFIG_EXPERIMENTAL=y CONFIG_M586TSC=y CONFIG_X86_WP_WORKS_OK=y CONFIG_X86_INVLPG=y CONFIG_X86_BSWAP=y CONFIG_X86_POPAD_OK=y CONFIG_X86_TSC=y CONFIG_MODULES=y CONFIG_KMOD=y CONFIG_NET=y CONFIG_PCI=y CONFIG_PCI_GOANY=y CONFIG_PCI_BIOS=y CONFIG_PCI_DIRECT=y CONFIG_PCI_QUIRKS=y CONFIG_PCI_OPTIMIZE=y CONFIG_SYSVIPC=y CONFIG_SYSCTL=y CONFIG_BINFMT_AOUT=m CONFIG_BINFMT_ELF=y CONFIG_BINFMT_MISC=m CONFIG_PARPORT=m CONFIG_PARPORT_PC=m # # USB support # CONFIG_USB=m CONFIG_USB_DEVICEFS=y CONFIG_HOTPLUG=y CONFIG_USB_UHCI=m CONFIG_USB_SCANNER=m CONFIG_USB_HID=m CONFIG_INPUT_MOUSEDEV=m CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024 -- Regards Jean-Luc - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
possible pty DoS
hello all I am not subscribed to linux-kernel*; please CC any follow-ups to [EMAIL PROTECTED] (I probably won't reply from 2000-12-22 to 2001-01-07) I am running 2.4.0-test12-pre2 This snippet can prevent progress of any other processes that tries to do a write to a pty: #include #include int main() { int ptm; ptm = open("/dev/ptmx", O_WRONLY); while (1) write(ptm, "hello, world!\n", 14); } With this running, and no process eating up the greetings, I can telnet to my machine; the banner and login prompt appear. ps -alx at this point reveals in.telnetd is in do_select(). As soon as I type even one char of username, another ps reveals in.telnetd now stuck in __down_interruptible() Lucky I usually leave a few logged in consoles lying around; xterm also uses pty's! Some observations: 2.2.12 behaves fine (telnet logins work fine) 2.2.12-2.4.0 diffs between tty_io.c show changes involving up/down/etc. in do_tty_write(). Bernd Jendrissek P.S. apologies to all for my void * arithmetic a few months ago; it was a moment of eager-beaver weakness. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: [2.2.18] VM: do_try_to_free_pages failed
On Wed, Dec 20, 2000 at 01:03:00PM +0100, you [Matthias Andree] claimed: > Last night, one of your production machines got wedged, I caught a lot > of kernel: VM: do_try_to_free_pages failed for ... for a whole range of > processes, among them ypbind, klogd, syslogd, xntpd, cron, nscd, X, > How can I get rid of those do_try_to_free_pages lockups? That box Almost everybody (including me) who have seen that problem seem to have had it fixed with Andrea Arcangeli's VM-global-7 patch (ftp://ftp.kernel.org/pub/linux/kernel/people/andrea...). > Should I try the most recent 2.2.19-pre? Yes, Andrea's VM-global-7 is included in pre2. -- v -- [EMAIL PROTECTED] - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
[2.2.18] VM: do_try_to_free_pages failed
Last night, one of your production machines got wedged, I caught a lot of kernel: VM: do_try_to_free_pages failed for ... for a whole range of processes, among them ypbind, klogd, syslogd, xntpd, cron, nscd, X, master (Postfix super daemon), pvmd3, K applications and so on, I was unable to log in via ssh, someone on-site has finally reset that machine this noon to bring it back online. How can I get rid of those do_try_to_free_pages lockups? That box exports root file systems for some SparcStation 2 that are used as X terminals, so it's pretty important I keep that box running. Should I try the most recent 2.2.19-pre? The machine is a pentium-MMX with 64 MB RAM with a kernel 2.2.18 that has these patches/updated drivers (none VM related AFAICS): IDE 2.2.18.1209 I²C 2.5.4 LM_Sensors 2.5.4 DC390 2.0e7 ReiserFS 3.5.28 -- Matthias Andree - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: TCP/IP kernel modification
> i.e after the kernel calls ip_route_output() and > ip_route_output_slow() and fails to find a match, i > need the kernel to somehow "hook-up" with a > process/daemon(routing protocol) and access a user > route cache there. You may try this: http://sites.inka.de/~bigred/sw/rrouted.txt> Olaf - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Linux 2.2.19pre2
On Wed, Dec 20, 2000 at 12:32:54PM +0200, Petri Kaukasoina wrote: > OK, I booted 2.4.0-test12 which even prints that list: > > BIOS-provided physical RAM map: > BIOS-e820: 0009fc00 @ (usable) > BIOS-e820: 0400 @ 0009fc00 (reserved) > BIOS-e820: 0348 @ 0010 (usable) > BIOS-e820: 0010 @ fec0 (reserved) > BIOS-e820: 0010 @ fee0 (reserved) > BIOS-e820: 0001 @ (reserved) > Memory: 52232k/54784k available (831k kernel code, 2164k reserved, 62k data, 168k >init, 0k highmem) > > The last three reserved lines correspond to the missing 2.5 Megs. What are > they? Data reserved by your system for whatever purpose. Most probably ACPI data or similar. > 2.2.18 sees all 56 Megs and works ok and after adding mem=56M on the > kernel command line even 2.2.19pre2 works ok with all the 56 Megs. No > crashes. If you would have ACPI events, you would probably run into trouble. Apart from this, chances are that the reserved data is not needed by Linux and never accessed by the BIOS, so you may get away with using the reserved memory. The safe way is to respect the BIOS' RAM map. Regards, -- Kurt Garloff <[EMAIL PROTECTED]> Eindhoven, NL GPG key: See mail header, key servers Linux kernel development SuSE GmbH, Nuernberg, FRG SCSI, Security PGP signature
Re: Linux 2.2.19pre2
On Sun, Dec 17, 2000 at 04:38:02PM +0100, Kurt Garloff wrote: > On Sun, Dec 17, 2000 at 12:56:56PM +0200, Petri Kaukasoina wrote: > > I guess the new memory detect does not work correctly with my old work > > horse. It is a 100 MHz pentium with 56 Megs RAM. AMIBIOS dated 10/10/94 with > > a version number of 51-000-0001169_0011-101094-SIS550X-H. > > > > 2.2.18 reports: > > Memory: 55536k/57344k available (624k kernel code, 412k reserved, 732k data, 40k >init) > > > > 2.2.19pre2 reports: > > Memory: 53000k/54784k available (628k kernel code, 408k reserved, 708k data, 40k >init) > > > > 57344k is 56 Megs which is correct. > > 54784k is only 53.5 Megs. > > It's this patch that changes things for you: > o E820 memory detect backport from 2.4(Michael Chen) > > The E820 memory detection parses a list from the BIOS, which specifies > the amount of memory, holes, reserved regions, ... > Apparently, your BIOS does not do it completely correctly; otherwise you > should have had crashes before ... OK, I booted 2.4.0-test12 which even prints that list: BIOS-provided physical RAM map: BIOS-e820: 0009fc00 @ (usable) BIOS-e820: 0400 @ 0009fc00 (reserved) BIOS-e820: 0348 @ 0010 (usable) BIOS-e820: 0010 @ fec0 (reserved) BIOS-e820: 0010 @ fee0 (reserved) BIOS-e820: 0001 @ (reserved) Memory: 52232k/54784k available (831k kernel code, 2164k reserved, 62k data, 168k init, 0k highmem) The last three reserved lines correspond to the missing 2.5 Megs. What are they? 2.2.18 sees all 56 Megs and works ok and after adding mem=56M on the kernel command line even 2.2.19pre2 works ok with all the 56 Megs. No crashes. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Purging the Buffer Cache
Al Peat wrote: > > Is there any way to completely purge the buffer > cache -- not just the write requests (ala 'sync' or > 'update'), but the whole thing? Can I just call > invalidate_buffers() or destroy_buffers()? > > I know, why in the world would a person do such a > thing? Research. It'd be easier for me to write a > little program or add it to a module than wait for a > reboot each time I need a clean buffer cache. What about the ioctl BLKFLSBUF ? If you are running a SuSE distrib there is already a tool called flushb that does what you want. If not, you can download the simple tool from http://innominate.org/~juri/flushb.tar.gz Juri -- [EMAIL PROTECTED] system engineer innominate AG clustering & securitythe linux architects tel: +49-30-308806-45 fax: -77http://www.innominate.com - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH] fix emu10k1 init breakage in 2.2.18
"Andreas M. Kirchwitz" wrote: > > Mikael Pettersson wrote: > > > 2.2.18 broke the emu10k1 driver when compiled into the kernel. > > The problem is that 2.2.18 now implements 2.4-style module_init, > > so emu10k1 ended up being initialised twice when built non-modular, > > which rendered it dysfunctional. The fix is to remove the now > > obsolete explicit init calls. Patch below. Please apply. > > Is there also a fix available to make the bass and treble settings > work again in mixer applications (for example, Gnome mix 1.2.0)? > This is (now, was) one of the biggest advantages of this card to have > control over bass and treble settings. > > It worked for the early 2.2.18pre patches, but stopped working in > the latest ones (including final 2.2.18). Yes, put something like "EXTRA_CFLAGS += -DTONE_CONTROL" into the Makefile in drivers/sound/emu10k1/ Juri -- [EMAIL PROTECTED] system engineer innominate AG clustering & securitythe linux architects tel: +49-30-308806-45 fax: -77http://www.innominate.com - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Announce: modutils 2.3.23 is available
Christian Gennerat wrote: > > About Standard aliases: > > modprobe -c > ... > alias ppp-compress-21 bsd_comp > ... > > Why bsd_comp is the standard alias? > /src/linux/Configure.help says that > > The PPP Deflate compression method ("PPP Deflate compression", > above) is preferable to BSD-Compress, because it compresses better > and is patent-free. > ppp-compress-21 refers to PPP compression method 21, which happens to be BSD Compress. Deflate is 26 (and also 24, because it was assigned that value in the draft RFC). Aliasing ppp-compress-21 to anything other than bsd_comp would break PPP. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: Announce: modutils 2.3.23 is available
About Standard aliases: > modprobe -c ... alias ppp-compress-21 bsd_comp ... Why bsd_comp is the standard alias? /src/linux/Configure.help says that The PPP Deflate compression method ("PPP Deflate compression", above) is preferable to BSD-Compress, because it compresses better and is patent-free. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: [Acpi] 2.4.0-test13pre3 acpi circular dependency
The following fixes a circular depency problem between drivers/acpi/ and arch/{i386,ia64}/kernel/acpi.c. I think the problem only occurs if you manually tweak the build to make acpi.o as a module, but it still should be fixed. This patch also fixes the Makefiles in drivers/acpi so that they do not blow up if you try to build drivers/acpi as a module (these are corrections to some variable names, not a new functional addition to the Makefiles). I have deliberately not included the patch to change CONFIG_ACPI into a tristate because I wonder if there is some problem with acpi.o as a module that I am not aware of that is the reason that CONFIG_ACPI in the stock kernels is configured as a boolean, even though there is module initialization code in drivers/acpi, that seems to work just fine, at least for my purposes of deactivating the power after a shutdown. Does anybody know if there some known problem with acpi.o as a module? I have attached my kernel patch below. If this meets with no obections, can somebody bless this and "send" it to Linus for integration? On Tue, Dec 19, 2000 at 06:00:15PM -0800, Grover, Andrew wrote: > I'm thinking arch/i386/kernel/acpi.c should just go away, yes? > > Its purpose is probably better served by an ifdef, like you mentioned. [...] > > From: Adam J. Richter [mailto:[EMAIL PROTECTED]] > > > > Although the stock linux-2.4.0-test13pre3 does not allow > > one to build the acpi interpreter as a loadable module, I had > > tweaked the Makefiles in previous kernels to do this (the supporting > > code is there and it seemed to work, at least for shutting off the > > power after a shutdown). Unfortunately, in 2.4.0-test13pre3, this > > is no harder to do, because there is a circular dependency: > > > > drivers/acpi/ references acpi_get_rsdp_ptr in arch/i386/kernel/acpi.c, > > and > > arch/i386/kernel/acpi.c references acp_find_root_pointer in > > drivers/acpi/. > > > > > > I would like to recommend that the contents of > > arch/i{386,a64}/kernel/acpi.c be merged back somewhere in > > drivers/acpi/, > > and just selected with Makefile options, ifdefs, or perhaps runtime > > options (if the ia64 code is potentionally useable to an i386 kernel > > that find itself running on an ia64 CPU, which will probably > > be the case > > with most Linux distributions initially installed on ia64 hardware). > > > > If need be, I would be willing to at least write a quick and > > dirty #ifdef-based version of this proposed change. -- Adam J. Richter __ __ 4880 Stevens Creek Blvd, Suite 104 [EMAIL PROTECTED] \ / San Jose, California 95129-1034 +1 408 261-6630 | g g d r a s i l United States of America fax +1 408 261-6631 "Free Software For The Rest Of Us." diff --new-file -r -u linux-2.4.0-test13-pre3/drivers/acpi/Makefile linux/drivers/acpi/Makefile --- linux-2.4.0-test13-pre3/drivers/acpi/Makefile Wed Dec 20 00:49:37 2000 +++ linux/drivers/acpi/Makefile Wed Dec 20 00:03:27 2000 @@ -3,6 +3,7 @@ # O_TARGET := acpi.o +obj-m := $(O_TARGET) export-objs := ksyms.o @@ -25,13 +26,23 @@ subdir-$(CONFIG_ACPI) += $(acpi-subdirs) -obj-$(CONFIG_ACPI) := $(patsubst %,%.o,$(acpi-subdirs)) -obj-$(CONFIG_ACPI) += os.o ksyms.o +obj-y := $(patsubst %,%.o,$(acpi-subdirs)) +obj-y += os.o ksyms.o + +$(patsubst %,%.o,$(acpi-subdirs)): + $(MAKE) $(MFLAGS) -C $$(basename $@ .o) ../$@ ifdef CONFIG_ACPI_KERNEL_CONFIG - obj-$(CONFIG_ACPI) += acpiconf.o osconf.o + obj-y += acpiconf.o osconf.o else - obj-$(CONFIG_ACPI) += driver.o cmbatt.o cpu.o ec.o ksyms.o sys.o table.o + obj-y += driver.o cmbatt.o cpu.o ec.o ksyms.o sys.o table.o +endif + +ifdef CONFIG_X86 + obj-y += rsdp_x86.o +endif +ifdef CONFIG_IA64 + obj-y += rsdp_ia64.o endif include $(TOPDIR)/Rules.make diff --new-file -r -u linux-2.4.0-test13-pre3/drivers/acpi/common/Makefile linux/drivers/acpi/common/Makefile --- linux-2.4.0-test13-pre3/drivers/acpi/common/MakefileWed Dec 20 00:49:37 2000 +++ linux/drivers/acpi/common/Makefile Tue Dec 19 08:58:42 2000 @@ -4,7 +4,7 @@ O_TARGET := ../$(shell basename `pwd`).o -obj-$(CONFIG_ACPI) := $(patsubst %.c,%.o,$(wildcard *.c)) +obj-y := $(patsubst %.c,%.o,$(wildcard *.c)) EXTRA_CFLAGS += -I../include diff --new-file -r -u linux-2.4.0-test13-pre3/drivers/acpi/dispatcher/Makefile linux/drivers/acpi/dispatcher/Makefile --- linux-2.4.0-test13-pre3/drivers/acpi/dispatcher/MakefileWed Dec 20 00:49:37 2000 +++ linux/drivers/acpi/dispatcher/Makefile Tue Dec 19 08:58:42 2000 @@ -4,7 +4,7 @@ O_TARGET := ../$(shell basename `pwd`).o -obj-$(CONFIG_ACPI) := $(patsubst %.c,%.o,$(wildcard *.c)) +obj-y := $(patsubst %.c,%.o,$(wildcard *.c)) EXTRA_CFLAGS += -I../include diff --new-file -r -u linux-2.4.0-test13-pre3/drivers/acpi/events/Makefile linux/drivers/acpi/events/Makefile --- linux-2.4.0-test13-pre3/drivers/acpi/eve
Re: PROBLEM: mounting affs over loop hangs in syscall (x86 only?)
Hi, On Mon, 18 Dec 2000, Bernardo Innocenti wrote: > [1.] One line summary of the problem: > mounting affs over loop hangs in syscall (x86 only?) affs plays some games with the suberblock lock, I have a patch that plays even worse games, but it works. I hope to finish a major cleanup of affs over christmas. bye, Roman - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/
Re: kapm-idled : is this a bug?
Hi! > > How about adding a flag to FLAGS, or a new letter in STATE in > > /proc/pid/stat, to mean "this is an idle task"? > > > > ps & top could easily by taught to recognise the flag. > > What's the problem with using PID 0 as the idle task ? That's 'standard' > with OS'ses that display the idle task. Linux has already another thread with pid 0, called "swapper" which is in fact idle. kidle-apmd is different beast. > It's also the 'right' thing to do, and should directly work with top / ps Yes, we should make pid 0 visible to userlnad, agreed. Pavel -- The best software in life is free (not shareware)! Pavel GCM d? s-: !g p?:+ au- a--@ w+ v- C++@ UL+++ L++ N++ E++ W--- M- Y- R+ - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] Please read the FAQ at http://www.tux.org/lkml/