Summary of changes from v2.5.12 to v2.5.13 ============================================ (02/04/05 1.369.119.4) dgrs net driver janitor fixes: o check return code of request_region function o let free irq, if something was wrong during driver initialization. o litle printk cleanup (02/04/05 1.369.120.1) Update old OSS ac97_codec driver to give AD1886 codecs a special set of initialization and control operations. Needed to get sound working on some Presario notebooks, at least. (02/04/05 1.369.119.5) Merge bug fixes and cleanups for sun3lance net driver from 2.4.x kernel series. (02/04/05 1.369.119.6) Small janitor fixes from net drivers rrunner and sgiseeq. (02/04/05 1.369.119.7) Merge apne net driver update from 2.4.x: * switch from MMIO to PIO as primary means of NIC access * remove now-unneeded offsets from some constants (02/04/05 1.369.119.8) Cosmetic cleanups to 8139cp net driver. (02/04/05 1.369.119.9) Add power management support to 3c509 net driver. (02/04/08 1.369.106.5) In math-emu/op-common.h:FP_FROM_INT, correct handling of normalizing too-big fraction parts. (02/04/08 1.369.106.6) Sparc64 math-emu fix: - UltraSPARC-III can generate unfinished FPops for F{I,X}TO{S,D} so handle it. (02/04/08 1.369.106.7) get_cycles() needs to be defined as volatile instruction set, else compiler code moves it out of loops. (02/04/09 1.448) ppc64: search module exception tables (02/04/09 1.369.106.8) Sparc64: Move cache+tlb flushes into new asm/{cache,tlb}flush.h (02/04/09 1.369.106.9) Sparc64 updates for recent preemption fixes. (02/04/09 1.369.106.10) Sparc64 fix: Move flush_tlb_pgtables to tlbflush.h (02/04/09 1.369.106.11) Sparc64 fixes: - Include asm/{cache,tlb}flush.h as needed. - Fix botched merge of Ultra-III mathemu fixes. (02/04/09 1.369.106.12) Sparc64: Fix typo in previous asm/{cache,tlb}flush.h changes. (02/04/09 1.369.106.13) Sparc64: Fix typo in previous asm/{cache,tlb}flush.h changes. (02/04/09 1.369.106.14) Sparc64: Include errno.h in arch/sparc64/math-emu/math.c (02/04/09 1.369.106.15) Sparc64: Fix typo in arch/sparc64/math-emu/math.c current_thread_info --> current_thread_info(). (02/04/09 1.369.106.16) Sparc64: Add dummy default_idle. (02/04/09 1.369.132.1) Ipv6 build fix: Use named struct initializers for ipip6_fb_tunnel_dev (02/04/09 1.369.106.17) Sparc64: Include linux/errno.h in arch/sparc64/solaris/signal.c (02/04/09 1.369.106.18) Sparc64: Add dummy flush_tlb_kernel_range for now (02/04/09 1.369.106.19) Sparc64: Real mans flush_tlb_kernel_range (02/04/10 1.369.132.2) Rearrange some int members of tcp_opt.ucopy to save space on 64-bit platforms. (02/04/10 1.369.132.3) CONFIG_APPLETALK and CONFIG_DEV_APPLETALK are meant for the same thing. Rename all uses to CONFIG_DEV_APPLETALK, use consistently inside drivers/net/appletalk/Config.in and update documentation to match. (02/04/10 1.369.132.4) Remove bogus networking stats from struct kernel_stat. They are unused and not visible in any way to userspace. (02/04/10 1.369.132.5) IPv6 netfilter fixes: - SMP TABLE_OFFSET fix - ordered CPUs - Set /proc entry owner - Fix ip6tables layer4 protocol header calculation - MAC match: register for FORWARD hook, add MODULE_LICENSE (02/04/10 1.369.132.6) Add tcp_read_sock which allows one to read directly, sendfile style, from a TCP socket. RPC layer patches to take advantage of this are pending to improve TCP NFS support. (02/04/10 1.369.132.7) Tigon3: Fix DRV_MODULE_{VERSION,RELDATA} (02/04/12 1.451) ppc64: remove ppc_md.post_irq, clean up do_IRQ (02/04/12 1.452) ppc64: add percpu.h (02/04/12 1.453) ppc64: formatting cleanups (02/04/12 1.454) ppc64: Paca -> paca, Naca -> naca from Dave Engebretsen (02/04/12 1.455) ppc64: EEH updates and FWNMI handler from Todd Inglett (02/04/16 1.457) ppc64: remove mem_map_check_hash(), it doesnt work with the radix pagecache changes. (02/04/16 1.458) ppc64: store trap into exception frame, make DSISR 32bit and create CCR field - from Dave Engebretsen (02/04/16 1.459) ppc64: signal fix from 2.4 branch (02/04/16 1.369.132.8) tcp.c: Fix tcp_read_sock() so that it advances tp->copied_seq. (02/04/16 1.369.106.20) Sparc64: Fix elf_gregset_t layout for native 64-bit binaries. (02/04/16 1.369.106.21) Sparc{,64}: On PTRACEd processes, do the syscall restart before we notify the parent. (02/04/16 1.369.106.22) Add asm-sparc{,64}/percpu.h (02/04/16 1.369.106.23) Sparc64: Fix flush_tlb_kernel_range to always flush using Nucleus context. (02/04/16 1.369.106.24) Sparc{,64}: seg/get affinity syscalls (02/04/16 1.369.106.25) arch/sparc64/kernel/sys_sparc32.c: Fix typo. (02/04/18 1.369.106.26) Sparc64: CONFIG_DEBUG_SPINLOCK typo in asm/system.h (02/04/18 1.369.132.9) Bridging fix: - Output netfilter hook does not care about input dev so do not even pass it in (02/04/18 1.369.106.27) Sparc64: - Propagate xattr syscall numbers from asm-sparc/unistd.h (02/04/23 1.463) ppc64: wrap iseries specific stuff in set_dec with ifdef (02/04/23 1.464) ppc64: put machine type into naca from Don Reed (02/04/23 1.465) ppc64: remove HDIO_[GS]ET_KEEPSETTINGS (02/04/23 1.466) ppc64: add missing memory barriers in cpu bringup code, from Colin Devilbiss (02/04/23 1.369.106.28) SunGEM bug fix: - Fix MAC address pointer calculation when using random MAC addresses (02/04/23 1.369.132.10) Net fixes: - Handle downing of multicast routes properly, since they refer potentially to multiple devices. - IPv6 raw had missing sk->filter handling and rawv6_rcv missing some checksum processing. - IPV6 udp needs to use skb_checksum not csum_partial on skb->h.raw - Ingress packet scheduler fixes from Jamal - Addrconf bug discoverd by Petr Baudis, ipv6_ifa_notify should pass NULL second arg to ipv6_chk_addr in RTM_DELADDR handling. (02/04/23 1.369.132.11) Bridging fix: - In br_handle_frame, dont forget to drop br->lock for handle_special_frame code path (02/04/24 1.369.106.29) Dont const objects put into initdata. (02/04/24 1.369.132.12) Kill in_ntoa from the tree. (02/04/24 1.369.106.30) Fix sparc printk levels in arch/sparc/mm/init.c From Alan (02/04/24 1.369.106.31) Add Sparc Voyager power management support. (02/04/24 1.369.106.32) Sparc: - Delete ID strings added files from Sparc power management changes This aint CVS no more. (02/04/24 1.369.132.13) Tigon3: Handle posted writes correctly. (02/04/24 1.369.132.14) In pfifo packet scheduler use qdisc->dev instead of skb->dev for consistency (02/04/24 1.369.132.15) Do not ask about ECONET suboptions if INET==n (02/04/25 1.470) ppc64: Remove WANT_PAGE_VIRTUAL now the compiler bug is fixed (02/04/25 1.471) ppc64: remove some old xmon debug code which broke with page->virtual removal (02/04/25 1.472) ppc64: merge in some iSeries fixes from 2.4 (02/04/25 1.473) ppc64: pci_dma updates - remove excess syncs and general cleanup, from Mike Corrigan (02/04/25 1.474) ppc64: pci write return code fix, from Don Reed (02/04/25 1.475) ppc64: error in set_dec introduced by me, fixed by Milton Miller (02/04/25 1.476) ppc64: change bitops to take unsigned long * (02/04/25 1.477) ppc64: fix compiler warnings (02/04/25 1.478) ppc64: fix compiler warnings (02/04/25 1.479) ppc64: dont kill init in do_page_fault and remove some old debug (02/04/25 1.369.106.33) SunGEM: - Call gem_init_pause_thresholds before gem_init_dma so that rx_pause_{on,off} are ready in time. (02/04/25 1.369.132.16) Avoid use of __FUNCTION__ cpp pasting. (02/04/26 1.480) ppc64: ioctl32 updates add missing dev_put in dev_ifname32 fix usbdevfs (from sparc64) add SG_IO (from sparc64) add tun, vlan and random device ioctls (02/04/25 1.369.106.34) Sparc64 ioctl32 updates: - Include linux/lp.h - Add COMPATIBLE_IOCTL lines for TIOCSTART, TIOCSLTC, SCSI_IOCTL_PROBE_HOST, SCSI_IOCTL_GET_PCI, and LPGETSTATUS. - Remove spurious semicolons. (02/04/26 1.481) ppc64: fixes for iseries (02/04/26 1.369.106.35) Sparc64: - sed s/__TIOC{START,STOP}/TIOC{START,STOP}/, kill SunOS specific comment - Make ioctl32 recognize TIOCSTOP too. (02/04/28 1.482) ppc64: extern inline -> static inline (02/04/28 1.447.52.3) Switch remaining users of autoirq_*() API to probe_irq_{on,off} autoirq_*() has been replaced by probe_irq_{on,off} a long time ago, so it's about time to have it go away. (02/04/29 1.447.58.1) Some how not all the files with the screen_base change got merged. Here are the rest, (02/04/29 1.447.58.2) Another change that didn't make it. Now fixed. (02/04/29 1.369.106.36) Sparc64 pagetable leak fix: - Defer changing personality until flush_old_exec has freed up all the pagetables. (02/04/30 1.447.58.3) A bunch of fixes needed. (02/04/30 1.447.61.4) ISDN: b1dma introduce b1dma_queue_tx() wrapper (02/04/30 1.447.61.5) ISDN: b1dma tx path cleanup (02/04/30 1.447.61.6) ISDN: B1 DMA drivers: Replace cli() by proper spin_lock. (02/04/30 1.447.61.7) ISDN: b1dma cleanup: Use readl/writel directly (02/04/30 1.369.106.37) Sparc64: Remove HDIO_{SET,GET}_KEEPSETTINGS references, as they no longer exist. (02/04/30 1.369.106.38) drivers/video/sbusfb.c (sbusfbcon_blank): Return zero if operation not supported. (02/04/30 1.447.55.18) Add new PC300 WAN driver (courtesy Cyclades), and my own changes: * patch added stuff to include/linux. move these three new headers to drivers/net/wan. * change the code to support these changes * slightly better Config.in entry. needs more work, though. (02/04/30 1.447.64.1) irda update 1/7: o [FEATURE] Reduce LAP Tx queue to 2 packets (from 10) Improve latency, reduce buffer usage o [FEATURE] LAP Tx queue not full notification (flow start) Poll higher layer to fill synchronously LAP window (7 packets) o [FEATURE] LMP LSAP scheduler Ensure Tx fairness between LSAPs (sockets, IrCOMM, IrNET...) (02/04/30 1.447.64.2) irda update 2/7: o [CORRECT] Fix race condition when starting todo timer o [CORRECT] Fix race condition when stopping higher layer Higher layer would think it is stopped and us it is started o [CORRECT] Give credit even if packets in Tx queue If Tx queue was stopped, could starve peer and deadlock o [CORRECT] Protect Rx credit update with spinlock o [CORRECT] Calculate properly self->avail_credit Didn't take into account queued Rx fragments Incremented even if Rx frame not delivered to higher layer -> would never stop the peer (i.e. not flow control) -> could become infinite o [CORRECT] Send credit when higher layer reenable receive Peer wouldn't restart Tx to us if flow stopped o [FEATURE] Implement LAP queue not full notification Lower latency, ... o [FEATURE] Reduce Tx queue to 8 packets (from 10) But make sure we can always send a full LAP window (7) o [FEATURE] Fix and optimise TTP flow control Make sure peer can always send a full LAP window (7) Minimise explicit credit updates (give_credit) o [FEATURE] Remove need for todo timer in Tx/Rx paths Less potential races, lower latency, lower context switches Could not use tasklet because broken API, better anyway ;-) (02/04/30 1.447.64.3) irda update 3/7: o [FEATURE] Replace interruptible_sleep_on() with wait_event(). Most races were taken care off, but cleaner anyway (02/04/30 1.447.64.4) irda update 4/7: o [FEATURE] Dongle driver for mcp2120/crystal hardware (02/04/30 1.447.64.5) irda update 5/7: o [CORRECT] Load dongle module with irq disabled in irtty (02/04/30 1.447.64.6) irda update 6/7: o [CORRECT] Cancel LSAP watchdog when putting socket back to listen o [CORRECT] Try to close LAP when closing LSAP still active o [CORRECT] Header fix for compile on Alpha architecture (02/04/30 1.447.64.7) irda update 7/7: o [CORRECT] Prevent dealock on simultaneous peer IrNET connections Only the primary peer will accept the IrNET connection (02/04/30 1.447.66.1) [PATCH] PATCH 2.5.10 -- pegasus ethtool support This resends the "usb_make_path()" update (my net-0408 patch) to pegasus, so it reports the same bus info the other usb network drivers now return, and fixes a couple other bugs in ethtool support I happened to notice: - driver info wasn't providing the "driver short name". - settings weren't correctly initting data (returned garbage) The first bug caused one text fields to overflow into the next in "ethtool -i eth1" output. The second caused a 10/100 device to report itself with "ethtool eth1" as having gigabit support ... :) (02/04/30 1.447.66.2) [PATCH] Re: ohci-hcd and that NCR implementation ... The attached patch makes the one SiS controller behave like the other... the key fix being wmb() to make sure that no writes are still in CPU caches before telling the HC it's OK to fetch the data. (And a minor tweak to access PCI less often through the hub driver.) But the OPTi problem seems to be different. Could you try this patch to see if it affects the problem your seeing with that NCR implementation? And in any case, please integrate it to Linus' latest. (Against 2.5.10, might be some fuzz on the pci tweak due to that warning fix.) Those missing wmb() statements have likely been the root cause of various usb-ohci flakes reported over time, too. The drivers are still almost identical in those areas. (Though something made ohci-hcd consistently lose that cache flush race; maybe it's a bit faster.) (02/04/30 1.447.66.3) [PATCH] usb ipaq driver update USB ipaq driver Added support for the Casio EM500. Completely untested. Thanks to info from Nathan (02/05/01 1.447.55.20) request_region janitor updates for sealevel wan driver (02/05/01 1.447.55.21) ifenslave sample program updates: * ifr2.ifr_flags was not initialized in the hwaddr_notset case, * SIOCGIFFLAGS now called before hwaddr_notset test (02/05/01 1.447.55.22) Fix typo in bonding network device documentation. (02/05/01 1.447.55.23) Add docs for 3c509 network driver. (02/05/01 1.447.55.24) AUI fixes for TLAN net driver. Also, mention new maintainer. (02/05/01 1.447.55.25) Comment out misplaced } in dscc4 wan driver. Spotted by Keith Owens. (02/05/01 1.447.55.26) s/strtok/strsep/ in gt96100 net driver (02/05/01 1.447.55.27) e100 net driver update: e100_phy_reset is called from non-init functions, so remove __devexit marker from that function. (02/05/01 1.447.55.28) Merge 2.4 update to ariadne net driver: remove unneeded asm/io.h include (02/05/01 1.447.55.29) net driver's Space.c was missing {'s. Weird. (02/05/01 1.447.55.30) request_region janitor update for sdla wan driver (02/05/01 1.447.55.31) baycom_ser_fdx hamradio driver request_region janitor updates: 1) remove call to check_region 2) test result of request_region 3) call release_region in case of driver intialisation error later on (02/05/01 1.447.63.2) Fix typos in BK kernel howto. (02/05/01 1.447.63.3) request_region janitor updates for BusLogic SCSI driver. (02/05/01 1.447.63.4) request_region janitor updates for megaraid scsi driver: 1) removes calls to check_region 2) checks the result of request_region (02/05/01 1.447.63.5) request_region janitor updates for ultrastor scsi driver: 1) removes calls to check_region 2) checks the result of request_region calls 3) calls release_region where necessary in case of driver initialisation error (02/05/01 1.447.63.6) s/Efoo/-Efoo/ cleanup for drivers: etherh acorn net driver, mtdblock_ro mtd driver, and sym 2 scsi driver. (02/05/01 1.447.55.32) Add four new help texts to wan driver Config.help. (02/05/01 1.487) ppc64: sort exception table (02/05/01 1.488) ppc64: Add write memory barriers before atomics that return values (02/05/01 1.489) ppc64: Only implement thread priority macros on HMT or iSeries kernels Drop back to eieio in spinlocks for the moment due to performance issues of sync on power3 (02/05/01 1.447.55.33) Add support for SiS962 phy to sis900 net driver. Also introduce new maintainer. (02/05/01 1.447.55.34) Fix lance net driver build, after recent autoirq -> probe_irq_x changes (02/05/01 1.369.106.39) Sparc64: - Fix cache_decay_ticks calculation. - Mark some trap initialization funcs __init. (02/05/01 1.369.132.17) Netfilter trivial fixes: - net/core/netfilter.c (overlap): Overlap calculation error. - net/ipv6/netfilter/ip6_tables.c: Misplaced parentheses leads to pointer calculation error on SMP. (02/05/01 1.447.67.1) Small Rules.make cleanup Get rid of long obsolete (and unused) MOD_IN_SUBDIRS and unnecessary MOD_DIRS variable. (02/05/01 1.447.67.2) Documentation/kbuild/makefiles.txt update Remove description of long obsolete (pre-2.4) Makefiles and describe the current system. (02/05/01 1.447.67.3) ALSA: Fix link error If snd-pcm.o is selected built-in, snd-timer.o needs to be get built-in as well, since snd-pcm.o needs snd-timer.o's symbols. (02/05/01 1.447.69.1) [PATCH] (1/6) blksize_size[] removal - preliminary cleanups: make sure that swapoff restores original block size, kill set_blocksize() (and use of __bread()) in multipath.c, reorder opening device and finding its block size in mtdblock.c. (02/05/01 1.447.69.2) [PATCH] (2/6) blksize_size[] removal - switch set_blocksize() to struct block_device *. (02/05/01 1.447.69.3) [PATCH] (3/6) blksize_size[] removal - switch block_size() to struct block_device *. (02/05/01 1.447.69.4) [PATCH] (4/6) blksize_size[] removal - put block size in bdev->bd_block_size, make do_open() and check_partition() to set it (see above), switch set_blocksize() and block_size() to use of ->bd_block_size. Remove manipulations with blksize_size[] from drivers, remove blksize_size[] itself. (02/05/01 1.447.69.5) [PATCH] (5/6) blksize_size[] removal - kill bread()/getblk()/get_hash_table() (kdev_t-using wrappers; struct block_device * counterparts are obviously still alive). (02/05/01 1.447.69.6) [PATCH] (6/6) blksize_size[] removal - switch blk_get_ra_pages() to struct block_device *, move its use from bdget() to do_open(). (02/05/01 1.447.69.7) [PATCH] missing checks in exec_permission_light() Missing checks added... (02/05/01 1.447.69.8) [PATCH] scsi_error 2.5.12 The attachment is part of a patch from Justin Gibbs described in: http://marc.theaimsgroup.com/?l=linux-scsi&m=101200279101550&w=2 The original patch was targeted at lk 2.4 and Dave forwarded ported it into 2.5. Other bits (e.g. sr) have already found there way into your tree. One bit in the sd driver will be included in my following patch. (02/05/01 1.447.70.1) [PATCH] scsi disk (sd) driver 2.5.12 This patch has the last bit of Justin Gibb's patch described in: http://marc.theaimsgroup.com/?l=linux-scsi&m=101200279101550&w=2 [this bit for the sd driver] There is also a major code cleanup of the sd driver with documentation headers added and a few obvious bugs fixed described in: http://marc.theaimsgroup.com/?l=linux-scsi&m=101798201714399&w=2 I did this cleanup. Justin's patch has been in Dave's tree for several months while my code cleanup patch has been there since 2.5.9-dj1 . (02/05/01 1.447.69.10) [PATCH] Export path_lookup() path_lookup() is no longer an inline function, and needs to be exported for jffs2, nfsd and af_unix to load as modules. (02/05/01 1.447.69.11) [PATCH] Unresolved symbol block_flushpage block_flushpage() used to be a macro which pointed at the exported discard_bh_page(). I turned block_flushpage() into a real function but forgot the export. (02/05/01 1.447.55.38) [PATCH] four compile fixes for 2.5.12 Pretty boring stuff. Four compile fixes for 2.5.12. (02/05/01 1.493) [PATCH] aha152x driver Patch to the aha152x driver, which makes it work under 2.5. (02/05/01 1.494) [PATCH] bad BUG_ON from acct.c Remove bogus BUG_ON(). On UP kernels spinlocks become NOPs so it is not possible to test the state of a lock. (02/05/01 1.495) Missed one-liner in the manual merge between Al and Douglas (02/05/01 1.447.66.4) [PATCH] sddr09 write capability USB sddr09 The main purpose of the patch is to add write capability. (02/05/01 1.496) [PATCH] Tweak VM dirty thresholds The below patch on 2.5.12 takes the 'dbench 32' throughput on a 1024 megabyte machine from 35 megs/sec up to 125 megs/sec. That's on the second run. The first run is much slower because the ext2 bitmaps aren't in cache. Presumably, s/60/80/ will take it even higher. (02/05/01 1.497) [PATCH] ATTR_MODE breakage (originally caught by Paul Larson ) Patch fixes a couple of idiotic bugs in notify_change() (my fault). Unary operations have higher priority than binary ones, so if (!valid & ATTR_MODE) doesn't do the right thing. I plead temporary braindamage... (02/05/01 1.447.1.102) NTFS: Minor cleanups, contact details updates, remove superfluous set_blocksize. (02/05/01 1.499) [PATCH] UFS sanity checks The real rule is not "block size is at most 32Kb", it's "block size is at most 8 times the fragment size". (02/05/01 1.500) [PATCH] Fix the set_bit()'s warnings Fix some set_bit() warnings. (02/05/01 1.501) Manual merge from Andrew Morton: don't assume read-ahead for md (02/05/01 1.447.65.2) ISDN: AVM drivers: Get rid of ->interrupt Interrupt handlers won't be reentered anyway. (02/05/01 1.447.65.3) ISDN: cleanup AVM drivers Remove unnecessary wrappers around irq handlers. (02/05/01 1.447.65.4) ISDN: cleanup AVM drivers Allocate and free the avmcard structure using a shared function. (02/05/01 1.447.71.1) fs/autofs/inode.c:parse_options - Fix bug in strsep/strchr changes, dereference *this_char not *value at top of while loop. This matches how the same code in fs/autofs4/inode.c looks right now. Withtout this autofs loading causes an OOPS as the first time through the loop *value is dereferencing a NULL pointer. (02/05/01 1.447.71.2) fs/exportfs/expfs.c:dprintk - Fix macro varargs usage, you need to specify a variable name before the triple dot or else most current compilers complain. See include/linux/ext2_fs.h:ext2_debug for another example where it is done correctly. (02/05/02 1.447.1.104) NTFS: 2.0.5 release. Major buffer overflow fix in extent inode handling. Upgrade is imperative! (02/05/02 1.447.66.5) [PATCH] tiny race with devfs in printer USB printer bugfix looking through printer.c in preparation for shifting devfs support to usbcore I noticed that printer advertises a device through devfs before it can be opened. As devfs, or more precisely devfsd can be used to trigger actions this matters and is wrong. (02/05/02 1.447.66.6) [PATCH] pl2303.c: do not reset termios settings in each open() USB pl2303 driver This is a patch which avoids resetting the termios settings to default values (9600 Baud etc.) in each call to pl2303_open (). It does this only on the first call to pl2303_set_termios. After that it sets the termios to the last stored values. This way commands like stty -F /dev/ttyUSB0 115200 work the same way as with other serial ttys. (02/05/02 1.447.66.7) USB io_edgeport driver added Black Box OEM device ids to the io_edgeport driver. (02/05/02 1.447.66.8) USB cleaned up checking the return value of usb_register_dev() (02/05/02 1.447.66.9) [PATCH] 2.5.12 - make usbvideo.c use USBVIDEO_NUMFRAMES correctly This patch makes usbvideo.c use the correct value of USBVIDEO_NUMFRAMES. A few places in the code assumed it was 2. This patch was tested with USBVIDEO_NUMFRAMES = 2,3,4,8,11,32 (02/05/02 1.447.66.10) [PATCH] 2.5.12 - make usbvideo_register take a usb_device_id table This patch makes usbvideo_register take a usb_device_id argument which it can then pass to usb_register via the struct usb_device. Currently it passes NULL. (02/05/02 1.503) [PATCH] bio tagged command support Add support for request tagging of the block queue. (02/05/02 1.504) [PATCH] 2.5.12 IDE 49 - Fix compilation of piix.c - Remove the PADAM_ prefix from sleep hwgroup member. - Fix Pacific Digital host chip driver API. - Fix Tekram host chip driver API. - Fold hwif_unregister() directly in to channel code. (02/05/02 1.505) [PATCH] unneeded readpage in block_symlink The new consistency checks in buffer.c are complaining about block_symlink's ->readpage against an already uptodate page. (02/05/02 1.447.72.1) [PATCH] VM dirty page balancing - The balance_dirty_pages() logic is simply wrong. It goes: if (value > threshold) go_and_write(value - threshold); which is just fine for a single process writing data. But for many processes, they *all* go and bring things back into balance, and too much data gets written out. - The go_and_write(this much) logic is inoperative, because I turned off the ->writeback_mapping() function in ext2. So a call to writeback_unlocked_inodes(this_much) doesn't actually decrement and test *this_much. It will walk every inode, all the time. Silly. So quickly fixing the above things, the amount of dirty+writeback memory in the machine nicely stabilises at 500 megabytes across the run. (02/05/02 1.447.72.2) Remove left-over debug printk from Andrew (02/05/02 1.507) Update kernel version (02/05/02 1.508) [PATCH] scsi_merge.c 2.5.12 Fix oopses on ISA scsi adapters. [ This is a transfer from Dave's tree ] (02/05/02 1.509) [PATCH] scsi tape (st) 2.5.12 Documentation enhancement from Kai.Makisara@metla.fi. [ This is a transfer from Dave's tree ] (02/05/02 1.510) [PATCH] cpqfc compaq FC controller 2.5.12 Minor cleanups to a scsi fibre channel HBA driver that was broken by the introduction of SCSI_REPORT_LUNS to scsi.h in a patch that went into 2.5.11 . [ Another transfer from Dave's tree ]