ChangeSet@1.1198, 2003-11-10 16:24:07-02:00, marcelo@logos.cnet Changed EXTRAVERSION to -rc1 TAG: v2.4.23-rc1 ChangeSet@1.1197, 2003-11-10 16:23:27-02:00, bunk@fs.tum.de [PATCH] fix SOUND_CMPCI Configure.help entry Hi, the issue below is only a minor documentation fix, but it has confused me when configuring a kernel for such a card. The Config.in already contains the correct number. Please apply Adrian ChangeSet@1.1196, 2003-11-10 15:43:54-02:00, herbert@gondor.apana.org.au [PATCH] Fix BUS_ISA name conflict Hi: This is a patch from Greg KH which fixes the name conflict between sysctl.h and input.h on BUS_ISA. It's been part of 2.5 for more than a year. Index: kernel-source-2.4/include/linux/sysctl.h =================================================================== RCS file: /home/gondolin/herbert/src/CVS/debian/kernel-source-2.4/include/linux/sysctl.h,v retrieving revision 1.6 ChangeSet@1.1195, 2003-11-10 15:17:28-02:00, marcel@holtmann.org [PATCH] Make firmware loading work builtin This patch fixes the broken exports of the firmware loader if it is compiled into kernel. ChangeSet@1.1194, 2003-11-10 08:44:37-02:00, dougg@torque.net [PATCH] Do not accept negative size's in SG_SET_RESERVED_SIZE. Marcelo, Thanks to work by Pat LaVarre one of sg's ioctls [SG_SET_RESERVED_SIZE] was found to allow negative sizes. This causes the kernel pain. This patch fixes that (, a similar patch was recently put into lk 2.6.0-test8). Doug Gilbert [ Part 2: "Attached Text" ] ChangeSet@1.1192, 2003-11-09 21:37:30-02:00, akpm@osdl.org [PATCH] Restore /proc/pid/maps formatting The seq_file conversion of /proc/pid/maps caused altered behaviour. Before the conversion, spaces and tabs in filenames were displayed verbatim. After the conversion they are escaped as \040, etc. Also, if the mmapped file has been unlinked the output appears as 40017000-40018000 rw-p 00000000 03:02 1425800 /home/akpm/foo\040(deleted) instead of 40017000-40018000 rw-p 00000000 03:02 1425800 /home/akpm/foo (deleted) This could break applications which parse /proc/pid/maps (one person has reported this). The patch restores the old 2.4 behaviour. fs/proc/array.c | 2 +- 1 files changed, 1 insertion(+), 1 deletion(-) ChangeSet@1.1179.1.2, 2003-11-08 12:30:03-05:00, ralf@linux-mips.org [netdrvr pcnet32] add missing pci_dma_sync_single a patch for the pcnet32.c driver which adds a missing call to pci_dma_sync_single. If a received packet is smaller than rx_copybreak the pcnet driver will recycle the receive buffer which requires calling pci_dma_sync_single. Patch is against 2.6 but I it's also needed in 2.4. Without that call the processor might still have old stale data in the data cache when the processor accesses the recycled buffer. ChangeSet@1.1179.1.1, 2003-11-08 12:29:06-05:00, pp@ee.oulu.fi [netdrvr b44] Fix irq enable/disable; fix oops due to lack of SET_NETDEV_DEV() call Also, add suspend/resume functions. ChangeSet@1.1190, 2003-11-06 14:19:37-02:00, jack@suse.cz [PATCH] Fix quota accounting bug > > > a friend of mine, made me aware of the following > > > imbalance, which looks like a minor accounting bug > > > to me, but might be a quota bug ... > > Sorry but the code seems correct to me - we get reference to dquot by > > get_dquot_ref() and than we put the reference by dqput(). dqput() is > > correct because something nasty might happen in the mean time and so we > > might be the last holders of the dquot. What do you think is wrong? > > dqput() does dqstats.drops++; > which isn't correct if this should be the same as > put_dquot_ref(), but maybe I'm just irritated by > strange statistics on some kernels showing more > drops than lookups+allocated after sync/quotaoff So I finally got to fixing the bug. The patch (also with a deletition of unnecessary variable) is attached. Marcelo please apply. Honza ChangeSet@1.1188.1.2, 2003-11-05 14:50:32-08:00, davem@nuts.ninka.net [TG3]: Bump driver version and release date. ChangeSet@1.1188.1.1, 2003-11-05 14:38:58-08:00, davem@nuts.ninka.net [TG3]: Fix bugs in ETHTOOL_SSET introduced by ethtool_ops conversion. - Missing spin_lock*() calls before tp->link_config twiddling. - Missing assignment to tp->link_config.autoneg ChangeSet@1.1189, 2003-11-05 09:49:29-02:00, johnstul@us.ibm.com [PATCH] Fix cyclone timer (x44x) Marcelo, All, In working to resolve an issue in SLES8 SP3, Andrea caught a bug in the cyclone counter's lost-tick compensation code. In the code, we use the calibrated loops_per_jiffies to detect the number of ticks that have passed, however the lost-ticks code begins running before loops_per_jiffies is calculated. This circular dependency that can cause loops_per_jiffies to be occasionally mis-calculated. The fix drops use of loops_per_jiffies and uses code from 2.5 to do the same thing. This patch only affects x440/x445 machines. Many thanks for Andrea for catching this. Any comments and feedback on the patch is welcome. thanks -john ChangeSet@1.1188, 2003-11-04 10:40:26-02:00, johnstul@us.ibm.com [PATCH] Fix x440+ACPI problem Marcelo, All, I noticed on x440s that when HT is disabled in the BIOS I was having problems properly booting 2.4 in ACPI mode. Further investigation found a subtle problem w/ smp_boot_cpus() when clustered_acpi_mode is set. During bootup, phys_cpu_present_map is initialized by ORing apicid_to_phys_cpu_present() for each cpu apicid(see MP_processor_info) On flat mode boxes this translates to "phys_cpu_present_map |= (1< dwmw2: I really don't like the semantics of ilookup() - or existing callers of that animal, for that matter. dwmw2: the thing is, it's just begging to be abused dwmw2: yes, in your case it is legitimate dwmw2: all right, to hell with that. dwmw2: feel free to backport it... Marcelo, please apply. This is the simplest way to fix a race condition in JFFS2, where we attempt to garbage-collect a physical node belonging to an inode which is being deleted and is gone from the icache by the time we look for it. Just using iget() and causing read_inode() to happen again for the deleted inode is the wrong thing to do. This is just the 2.4 backport of ilookup(). The JFFS2 change itself will be next. ===== fs/inode.c 1.40 vs edited ===== ChangeSet@1.1178.1.2, 2003-11-03 11:18:41-02:00, ak@muc.de [PATCH] Fix TSS limit on x86-64 The segment limit for the TSS segment on x86-64 was too big. The lead to the CPU reading random memory for the ioperm mask when ioperm() was not called, but iopl is >0. Index: linux/include/asm-x86_64/desc.h =================================================================== RCS file: /home/cvs/Repository/linux/include/asm-x86_64/desc.h,v retrieving revision 1.17 ChangeSet@1.1178.1.1, 2003-11-03 11:18:15-02:00, ak@suse.de [PATCH] Add missing nforce3s pci-id My bad. Please apply. From: "O.Sezer" Subject: missing nforce3s pci-id in current -bk [ Part 2.1: "Attached Text" ] Cset 1.1173.1.3 -> 1.1173.1.4 (ak@muc.de, K8 AGP driver updates) misses patching pci_ids.h for nforce3s. Attached. Özkan Sezer [ Part 2.2: "Attached Text" ] ChangeSet@1.1173.2.1, 2003-10-31 13:30:51-06:00, shaggy@shaggy.austin.ibm.com JFS: i_nlink should be checked while holding commit_sem In the 2.6 kernel, both jfs_link and jfs_unlink are called while holding i_sem. This is not true in the 2.4 kernel. There is still a window where a race between link and unlink can cause i_nlink to go to zero and then back to one. This fix closes that window. ChangeSet@1.1179, 2003-10-31 08:36:16-05:00, philipc@snapgear.com [netdrvr 8139cp] Fix NAPI race Cures reported lockups. ChangeSet@1.1178, 2003-10-30 19:51:09-02:00, ak@muc.de [PATCH] Fix Documentation.help for K8 AGP driver The config variable has been renamed, fix Documentation.help too. >From Adrian Bunk I think. ChangeSet@1.1177, 2003-10-30 19:50:40-02:00, ak@muc.de [PATCH] Make new driver i386 only The driver won't compile on x86-64, but only on i386. Express this is Config.in. ChangeSet@1.1173.1.4, 2003-10-30 18:53:23-02:00, ak@muc.de [PATCH] K8 AGP driver updates Updates for the K8 AGP driver. - Find the AGP bridge even when it isn't the first host bridge in the system (happens on some Tyan boards) - Support the NForce3S - Fix warning ChangeSet@1.1173.1.3, 2003-10-30 18:52:06-02:00, ak@muc.de [PATCH] x86-64 update Various bug fixes for x86-64 - Proper EOVERFLOW check in 32bit stat - Undo buggy irq.c change - Enlarge exception stack to 4K (Jim Paradis) - Work around K8 B stepping issue that allowed users to crash the kernel - Replace safe_smp_processor_id with reading CPUIDs - Export do_fork and bad_dma_address ChangeSet@1.1173.1.2, 2003-10-30 18:48:17-02:00, marcelo@logos.cnet MAINTAINERS update for HP ChangeSet@1.1166.1.7, 2003-10-30 10:14:12-06:00, shaggy@shaggy.austin.ibm.com JFS: Fix race between link() and unlink() JFS isn't happy it thinks a file has been removed, and link() increases its nlink count back from zero. ChangeSet@1.1175, 2003-10-30 10:44:04-05:00, scott.feldman@intel.com [e1000] sync with 2.6 updates Patch to sync e1000 in 2.4 with what's already been accepted into 2.6. Here's the change log: o Handle ethtool force flow control. o Correctly set flow control hi/low watermarks based on size of Rx FIFO area. The size can change if doing Jumbo Frames or, in the case of 82547, is smaller to start with. o Was not properly forcing flow control settings to fc_none if using strict IEEE flow control override. o Bug fix: SERDES devices might be connected to back-plane switch that doesn't support auto-neg, so add the capability to force 1000/full if autoneg fails. o Better propagation of error codes during probe/open paths. [Janice Girouard (janiceg@us.ibm.com)]. o Remove unused ASSERT macro. o Missed a free_netdev() in probe cleanup undo. ChangeSet@1.1174, 2003-10-30 10:44:00-05:00, scott.feldman@intel.com [e100] sync with 2.6 updates Patch to sync e100 in 2.4 with what's already been accepted into 2.6. Here's a change log summary: o PRO/10+ support wasn't being configured properly. o Trying to pci_alloc_consistent before pci_enable_device hurts on some archs. This gets things in the right order. o Driver was advertising NETIF_F_HW_CSUM, but hardware is only capable of NETIF_F_IP_CSUM. Breaks IPv6 Tx. ChangeSet@1.1173, 2003-10-30 08:49:40-02:00, marcelo@logos.cnet Changed EXTRAVERSION to -pre9 TAG: v2.4.23-pre9 ChangeSet@1.1166.3.1, 2003-10-29 12:47:04-08:00, ebrower@usa.net [SPARC]: Fix _IOC_SIZE() macro when direction is _IOC_NONE. ChangeSet@1.1166.2.9, 2003-10-29 15:43:27-05:00, akpm@osdl.org [netdrvr 3c527] add MODULE_LICENSE tag ChangeSet@1.1166.2.8, 2003-10-29 15:40:02-05:00, achirica@telefonica.net [PATCH] Fix wireless stats locking ChangeSet@1.1166.2.7, 2003-10-29 15:39:59-05:00, arjanv@redhat.com [PATCH] fix starfire 64-bit b0rkage Hi, >> 32 is undefined on a 32 bit integral variable in C; >>16 >> 16 *IS* (and gets optimized out to 0, while >> 32 gets optimized out to a nop). Fix for starfire below ChangeSet@1.1166.2.6, 2003-10-29 14:48:49-05:00, komujun@nifty.com [pcmcia fmvj18x_cs] share interrupts properly for TDK multifunction cards. ChangeSet@1.1166.2.5, 2003-10-29 14:41:26-05:00, achirica@telefonica.net [PATCH] Fix compatibily issue with some APs ChangeSet@1.1166.2.4, 2003-10-29 14:39:07-05:00, riel@surriel.com [netdrvr starfire] include asm/io.h Fixes build on some platforms. ChangeSet@1.1166.2.3, 2003-10-29 14:36:42-05:00, tsk@ibakou.com [netdrvr 8139too] add pci id ChangeSet@1.1166.2.2, 2003-10-29 14:31:38-05:00, arjanv@redhat.com [PATCH] r8169 module license tag ChangeSet@1.1166.2.1, 2003-10-29 14:26:10-05:00, akpm@osdl.org [PATCH] sis900 skb free fix This driver is freeing skb's from timer context, with local irq's disabled. It generates warnings from local_bh_enable() because local_bh_enable() reenables interrupts, exposing the machine to deadlocks. So use the deferred dev_kfree_skb_irq() instead. ChangeSet@1.1169, 2003-10-29 10:39:59-08:00, fubar@us.ibm.com [bonding] Restore compatibilty with old ifenslave The patch below restores compatibility that was removed during the propagation set. That set added the ability to set the bond's hardware address in the conventional way using SIOCSIFHWADDR but also removed support for the very old way using BOND_SETHWADDR, which, although has been removed from the ifenslave code for years, turns out to still be in use in the ifenslave binary of recent distributions. ChangeSet@1.1160.2.1, 2003-10-29 10:59:32-06:00, shaggy@shaggy.austin.ibm.com JFS: remove racy, redundant call to block_flushpage __invalidate_metapages references mp->page after after releasing the meta_lock spinlock, without increasing the use count. This is racy and unnecessary since setting the META_discard flag is sufficient. block_flushpage() will be called when the metapage is released. ChangeSet@1.1166.1.5, 2003-10-29 10:11:10-02:00, neilb@cse.unsw.edu.au [PATCH] Fix deadlock problem in lockd. nlmsvc_lock calls nlmsvc_create_block with file->f_sema held. nlmsvc_create_block calls nlmclnt_lookup_host which might call nlm_gc_hosts which might, eventually, try to claim file->f_sema for the same file -> deadlock. nlmsvc_create_block does not need any protection under any lock as lockd is single-threaded and _create_block only plays with internal data structures. So we release the f_sema before calling in, and make sure it gets claimed again afterwards. (This was fixed in 2.5 8 months ago) ChangeSet@1.1168, 2003-10-28 23:19:38-08:00, janitor@sternwelten.at [NETFILTER]: Add IPCHAINS to MAINTAINERS entry. ChangeSet@1.1063.44.37, 2003-10-29 01:08:40-05:00, len.brown@intel.com vsprintf needs PAGE_SIZE from page.h in 2.4 ChangeSet@1.1166.1.3, 2003-10-28 23:26:47-05:00, len.brown@intel.com Merge intel.com:/home/lenb/src/linux-acpi-test-2.4.22 into intel.com:/home/lenb/src/linux-acpi-test-2.4.23 ChangeSet@1.1063.44.36, 2003-10-28 23:19:54-05:00, len.brown@intel.com [ACPI] fix x86_64 ACPI build in 2.4.22 by backporting from 2.4.23 ChangeSet@1.1063.44.35, 2003-10-28 22:13:18-05:00, akpm@osdl.org [PATCH] make printk more robust with "null" pointers Expand printk's traditional handling of null pointers so that anything in the first page is considered a null pointer. This gives us better behaviour when someone (acpi..) accidentally prints a string which is embedded in a struct, the pointer to which is null. ChangeSet@1.1063.44.34, 2003-10-28 22:06:20-05:00, len.brown@intel.com [ACPI] REVERT ACPICA-20030918 CONFIG_ACPI_DEBUG printk that caused crash http://bugzilla.kernel.org/show_bug.cgi?id=1341 ChangeSet@1.1063.44.33, 2003-10-28 21:37:19-05:00, len.brown@intel.com [ACPI] REVERT acpi_ec_gpe_query(ec) T40 fix that crashed other boxes http://bugme.osdl.org/show_bug.cgi?id=1171 ChangeSet@1.1162.1.6, 2003-10-28 16:00:13-05:00, len.brown@intel.com Merge intel.com:/home/lenb/src/linux-acpi-test-2.4.22 into intel.com:/home/lenb/src/linux-acpi-test-2.4.23 ChangeSet@1.1063.44.32, 2003-10-28 15:56:16-05:00, len.brown@intel.com [ACPI] fix x86_64 build (Jeff Garzik) ChangeSet@1.1162.1.5, 2003-10-28 15:54:00-05:00, len.brown@intel.com [ACPI] fix x86_64 build (Jeff Garzik) ChangeSet@1.1167, 2003-10-28 08:44:03-08:00, kml@patheticgeek.net [TCP]: When SYN is set, the window is not scaled. ChangeSet@1.1166, 2003-10-27 11:35:52-02:00, marcelo@logos.cnet Cset exclude: 9a4gl@9a0tcp.ampr.org|ChangeSet|20031021055256|28265 ChangeSet@1.1162.1.4, 2003-10-27 11:21:04-02:00, bcollins@debian.org [PATCH] IEEE1394 fixes - Lots of cleanups using size_t for proper size handling. - Fix race in sbp2 where a packet was getting free'd while we were waiting on a semaphore in it. - Cleanups of various macros. - Fix memory leak in nodemgr. - Fix deadlock in nodemgr. ChangeSet@1.1162.1.3, 2003-10-27 08:50:57-02:00, stelian@popies.net [PATCH] compile mii when using usbnet Hi, This patch is necessary in order to select the compilation of mii.o when selecting usbnet.o. Marcelo, please apply. Thanks, Stelian. ===== drivers/net/Makefile 1.39 vs edited ===== ChangeSet@1.1162.1.2, 2003-10-27 08:45:59-02:00, stelian@popies.net [PATCH] sonypi driver update Hi, This is a small update to the sonypi driver which: * corrects the events for the Zoom and Help buttons * uses the irqreturn_t constructs backported from 2.5. This is not very intrusive and safe to be applied somewhat late in the -pre cycle. Marcelo, please apply. Thanks. Stelian. ===== drivers/char/sonypi.h 1.19 vs edited ===== ChangeSet@1.1162.1.1, 2003-10-27 08:32:05-02:00, zzz@anda.ru [PATCH] Fix aic7xxx compilation without PCI support I have these warnings when I'm compiling 2.4.22 for a 486 EISA system: aic7xxx_osm.c: In function `ahc_softc_comp': aic7xxx_osm.c:1560: warning: implicit declaration of function `ahc_get_pci_bus' aic7xxx_osm.c:1568: warning: implicit declaration of function `ahc_get_pci_slot' And then the make finishes with an error, because these functions really exist only if the PCI support is turned on. The patch below fixes this. And the same patch fits for the 2.6 kernels. Please, apply it. ChangeSet@1.1163, 2003-10-27 02:14:04-08:00, davem@nuts.ninka.net Merge nuts.ninka.net:/disk1/davem/BK/network-2.4 into nuts.ninka.net:/disk1/davem/BK/net-2.4 ChangeSet@1.1148.1.13, 2003-10-27 02:12:08-08:00, davem@nuts.ninka.net Cset exclude: kuznet@ms2.inr.ac.ru|ChangeSet|20031021053209|59468 ChangeSet@1.1148.1.12, 2003-10-27 00:09:52-08:00, bdschuym@pandora.be [NETFILTER]: Fix potential OOPS in ipt_REDIRECT. ChangeSet@1.1148.1.11, 2003-10-27 00:06:11-08:00, yoshfuji@linux-ipv6.org [IPV6]: Fix bogus semicolon typo in mcast.c ChangeSet@1.1160.1.1, 2003-10-25 18:03:58-02:00, chrisw@osdl.org [PATCH] sysctl core_setuid_ok fix The sysctl kern_table entry part of the core_setuid_ok patch has wrong ctl_name. Patch below is against current 2.4.23-pre8-bk. Seth Arnold pointed this problem out to me. thanks, -chris -- Linux Security Modules http://lsm.immunix.org http://lsm.bkbits.net ===== kernel/sysctl.c 1.28 vs edited ===== ChangeSet@1.1161, 2003-10-24 16:30:01-07:00, greg@kroah.com Merge ChangeSet@1.1148.7.20, 2003-10-24 14:00:57-07:00, david-b@pacbell.net [PATCH] USB: usb gadget Config.in updates This is a subset of work from Al Borchers which he sent as part of his "serial" gadget driver. The patch eliminates some of the mis-configurations that are possible with the 2.4 kconfig syntax. ChangeSet@1.1148.7.19, 2003-10-24 14:00:42-07:00, david-b@pacbell.net [PATCH] USB: updates This syncs this file with current declarations, in sync with 2.6 kernels. ChangeSet@1.1148.7.18, 2003-10-24 14:00:30-07:00, david-b@pacbell.net [PATCH] USB: usb "gadget zero" tweaks This updates this code to match the latest: builds on a few more 2.4 variants, handles "goku_udc". ChangeSet@1.1160, 2003-10-24 17:50:55-02:00, torvalds@osdl.org [PATCH] Add a quirk for the Intel ICH-[45] to add special ACPI regions Ok, I got my intel contacts to look up all the chipset quirks, and yup, that cardbus problem is explained (and removed) by adding the proper quirk entry for the newer ICH-[45] chips. Marcelo, this should go into 2.4.x too (it probably won't apply directly, but it might - and it should be trivial to fix up by hand). I'm still going to leave the default Cardbus IO address on 0x4000 on x86, since that has been successful on 2.4.x, and is thus the safest thing to do. But I have verified that the problem with starting allocations at 0x1000 is gone with this quirk fix. Linus ChangeSet@1.1148.2.68, 2003-10-24 10:25:23-02:00, paulus@samba.org [PATCH] Add load_addr arg to ELF_PLAT_INIT Marcelo, The patch below adds an argument to the ELF_PLAT_INIT so that the binfmt_elf code can pass the load address of the interpreter or binary to the architecture-specific code. We need this on PPC64. This code is already in 2.6.0-test. Please apply. Thanks, Paul. ChangeSet@1.1148.1.10, 2003-10-24 04:54:08-07:00, laurent.ml@linuxfr.org [IRDA]: Fix build fallout from gcc-3.3 changes. ChangeSet@1.1148.7.17, 2003-10-23 17:24:55-07:00, stern@rowland.harvard.edu [PATCH] USB: fix for earlier unusual_devs.h patch An earlier patch caused trouble because it effectively removed the US_FL_FIX_INQUIRY flag for devices with release number higher than 0x5009. This one might cause problems because it explicitly goes against the immediately preceding comment in unusual_devs.h. That comment says that these Casio digital cameras claim to use the CBI transport when in fact they only use CB. However, there have been two reports in the last few weeks from people getting the "unneeded SubClass and Protocol" log messages. One of them was using a device with release number 0x1000, right at the start of the range. The other had a device with release number 0x5010, just beyond the end of the current range. Maybe Casio is marketing two different devices with different behaviors but having the same Vendor, Product, and Release values -- I don't know. ChangeSet@1.1148.7.16, 2003-10-23 17:22:14-07:00, car.busse@gmx.de [PATCH] USB: one more digicam for unusual_devs.h its for the Jenoptik JD 5200 z3 Digicam, to enable it to work as a simple storage device more or less i took the values for the 0x0d96 vendor in the 2.6.0-test7 usb-storage and mixed them with my device id, which seems to work quite well tested with 2.4.22 kernel ChangeSet@1.1148.7.15, 2003-10-23 17:18:23-07:00, abbotti@mev.co.uk [PATCH] USB: ftdi_sio - version bump 1.3.5 Here is a small patch to bump the ftdi_sio driver's version number to 1.3.5 for the usb-2.4 tree. It would be nice if this made it into the final 2.4.23 kernel sources, then I could refer to that as version 1.3.5 of the driver. ChangeSet@1.1148.7.14, 2003-10-23 17:18:03-07:00, abbotti@mev.co.uk [PATCH] USB: ftdi_sio - Perle UltraPort new ids - 2 of 2 Patch by me (Ian Abbott). Removed aliases of FTDI_VID define for consistency. Renamed FTDI_PERLE_PID to FTDI_PERLE_ULTRAPORT_PID. Removed Matrix Orbital and Perle Systems devices from the 8U232AM device table (but left them in the FT232BM device table) as they are known to be FT232BM devices. ChangeSet@1.1148.7.13, 2003-10-23 17:17:54-07:00, abbotti@mev.co.uk [PATCH] USB: ftdi_sio - Perle UltraPort new ids - 1 of 2 Scott Allen's patch to add vid/pid support for Perle Systems UltraPort USB serial converter, merged up with minor whitespace changes by myself (Ian Abbott). ChangeSet@1.1148.7.12, 2003-10-23 15:04:27-07:00, greg@kroah.com [PATCH] USB: fix build bug with usbnet and older versions of gcc. ChangeSet@1.1148.2.67, 2003-10-23 18:10:56-02:00, riel@surriel.com [PATCH] silence warning in reiserfs_ioctl Gcc is afraid we might fall off the end of the function without returning. ChangeSet@1.1148.2.66, 2003-10-23 18:08:15-02:00, riel@surriel.com [PATCH] saa7110 typo fix One question mark should be enough for everybody. ChangeSet@1.1148.2.65, 2003-10-23 08:22:10-02:00, marcelo@logos.cnet Fix Makefile -pre8 typo ChangeSet@1.1148.2.64, 2003-10-22 20:43:41-02:00, marcelo@logos.cnet Jan Niehusmann: Make LBA48 work in pdc202xx_old.c. TAG: v2.4.23-pre8 ChangeSet@1.1148.2.63, 2003-10-22 20:39:47-02:00, marcelo@logos.cnet ide-disk.c: Limit disk size to 137GB if LBA48 is not available ChangeSet@1.1148.2.62, 2003-10-22 20:12:10-02:00, B.Zolnierkiewicz@elka.pw.edu.pl [PATCH] fix ServerWorks PIO auto-tuning No problems were reported with this patch, only success stories :). It is in 2.6-bk and here's version which applies cleanly to 2.4.23-pre7. [IDE] fix ServerWorks PIO auto-tuning If PIO mode should be auto-tuned xferspeed argument for svwks_tune_chipset() is equal to 255 (0xFF). It is then passed to ide_rate_filter() which matches desired mode with chipset capabilities. Since 255 is greater than any of the values used for transfer modes, ide_rate_filter() will always return the highest mode supported by both device/chipset (which sometimes should not be used ie. when host is a simplex one) and the wrong mode will be set. Thanks to Torben Mathiasen and Tomas Szepe for testing it. drivers/ide/pci/serverworks.c | 7 ++++++- 1 files changed, 6 insertions(+), 1 deletion(-) ChangeSet@1.1148.2.61, 2003-10-22 16:33:42-02:00, marcelo@logos.cnet Changed EXTRAVERSION to -pre8 ChangeSet@1.1148.2.60, 2003-10-22 09:40:23-02:00, daniel@deadlock.et.tudelft.nl [PATCH] atyfb ibook fix Fix atyfb on the ibook ChangeSet@1.1148.2.59, 2003-10-22 08:26:52-02:00, viro@parcelfarce.linux.theplanet.co.uk [PATCH] fix for do_tty_hangup() access of kfreed memory do_tty_hangup() does fput() on redirect struct file too early - it could've been the only holder of tty_struct we are working with and in that case we'll end up freeing it from fput() and then both reading and modifying kfreed memory. ChangeSet@1.1148.2.58, 2003-10-22 08:10:29-02:00, marcelo@logos.cnet Merge http://linux-acpi.bkbits.net/linux-acpi-release-2.4.23 into logos.cnet:/home/marcelo/bk/linux-2.4 ChangeSet@1.1148.1.9, 2003-10-22 02:57:40-07:00, davem@nuts.ninka.net [TG3]: Disable/enable timer in suspend/resume. ChangeSet@1.1148.21.11, 2003-10-21 07:59:16-02:00, xose@wanadoo.es [PATCH] change sym53c8xx.o to sym53c8xx_2.o in Configure.help -thanks- [ Part 2: "Attached Text" ] ChangeSet@1.1148.1.8, 2003-10-20 22:52:56-07:00, 9a4gl@9a0tcp.ampr.org [AX25]: Bug fixes and updates. Ralph and Steve are working on 2.6 ax25 stack, I fixed several issues in 2.4... the patch also add me to CREDITS file, add some new features - support for LinSV node which have some extensions to AX25, this layer3 node software will be soon on my homepage (by default this extension is disabled and can be enabled in kernel config) - software CRC check in dmascc driver (by default this feature is disabled and can be enabled in kernel config) - support for 1580 bytes MTU on AX25 interfaces (by default this is disabled and can be enabled in kernel config) ChangeSet@1.1148.1.7, 2003-10-20 22:32:09-07:00, kuznet@ms2.inr.ac.ru TCP: do not return -EINTR, when data are available for read() ChangeSet@1.1063.44.31, 2003-10-20 23:53:43-04:00, len.brown@intel.com [ACPI] Broken fan detection prevents booting (Shaohua David Li) http://bugme.osdl.org/show_bug.cgi?id=1185 ChangeSet@1.1148.2.56, 2003-10-20 23:20:22-04:00, len.brown@intel.com Merge intel.com:/home/lenb/bk/linux-acpi-test-2.4.22 into intel.com:/home/lenb/bk/linux-acpi-test-2.4.23 ChangeSet@1.1063.44.30, 2003-10-20 23:10:54-04:00, len.brown@intel.com [ACPI] fix use_acpi_pci !CONFIG_PCI build error per 2.6 http://bugzilla.kernel.org/show_bug.cgi?id=1392 ChangeSet@1.1063.44.29, 2003-10-20 18:10:12-04:00, len.brown@intel.com [ACPI] fix !CONFIG_PCI build use X86 ACPI specific version of eisa_set_level_irq() http://bugzilla.kernel.org/show_bug.cgi?id=1390 ChangeSet@1.1148.21.10, 2003-10-20 15:35:18-02:00, hugh@veritas.com [PATCH] tmpfs 5/5 writepage/truncate race If it's possible for a tmpfs page beyond i_size to remain in cache until shmem_truncate repeats truncate_inode_pages, then shmem_writepage's BUG_ON(index >= info->next_index) cannot be completely safe. But it's a useful check in a fragile area, so retain it when not in shmem_truncate. And there's a moment, when shmem_writepage's add_to_swap_cache races and fails, when the page would be absent from the inode lists searched by truncate_inode_pages: this case needs to set the pagein flag too. ChangeSet@1.1148.21.9, 2003-10-20 15:34:23-02:00, hugh@veritas.com [PATCH] tmpfs 4/5 getpage/truncate race Extend use of that SHMEM_PAGEIN flag to where shmem_getpage adds a page to the cache. It couldn't have caused a BUG_ON(inode->i_blocks), but if i_size is reduced (from another cpu) the instant after shmem_swp_alloc checks it, shmem_getpage could insert a page into the cache just after truncate_inode_pages has passed through cleaning it, leaving stale data (which may mysteriously reappear if the file is later extended). ChangeSet@1.1148.21.8, 2003-10-20 15:33:42-02:00, hugh@veritas.com [PATCH] tmpfs 3/5 swapoff/truncate race On 23 July, Sergey S. Kostyliov reported a tmpfs BUG_ON(inode->i_blocks) during swapoff: my last version of the fix to swapoff/truncate race was inadequate, since I_FREEING might get set or i_size be reduced (from another cpu) the instant after it's tested here. So revert to the previous version of the fix, shmem_truncate calling truncate_inode_pages again, if pages still left in cache; but avoid the recall in usual cases of partial truncation, by having a "pagein" flag to indicate when recall might be necessary. (Since those flags already use VM_ACCOUNT and VM_LOCKED, must redefine another VM_flag for this.) Sergey and 2.4-aa have run fine with this for a couple of months. ChangeSet@1.1148.21.7, 2003-10-20 15:32:31-02:00, hugh@veritas.com [PATCH] tmpfs 2/5 LTP S_ISGID dir LTP tests the filesystem on /tmp: many failures when tmpfs because it missed the way giddy directories hand down their gid. Also fix ramfs in the same way. ChangeSet@1.1148.21.6, 2003-10-20 15:32:09-02:00, hugh@veritas.com [PATCH] tmpfs 1/5 LTP ENAMETOOLONG Hi Marcelo, Here's first of 5 tmpfs patches, based on 2.4.23-pre7, total diffstat: fs/ramfs/inode.c | 9 ++++++- mm/shmem.c | 70 +++++++++++++++++++++++++++++++++++++++++++++---------- 2 files changed, 66 insertions(+), 13 deletions(-) These are 2.4 equivalents of (a subset of) the patches I sent Andrew and LKML last week, now in 2.6.0-test8-mm1: so I won't bother LKML with them again. I'd usually leave them longer in 2.6 before sending you for 2.4, but now you're looking towards 2.4.23-rc, better get them to you sooner than later: they're safe enough, and the issue in 3/5 was first found in 2.4 (-aa), with the fix verified there. Please apply, thanks! Hugh [PATCH] tmpfs 1/5 LTP ENAMETOOLONG LTP tests the filesystem on /tmp: many failures when tmpfs because shmem_lookup forgot to reject filenames longer than the NAME_MAX (255) tmpfs declares in its statfs. Also fix ramfs_lookup in the same way. ChangeSet@1.1148.21.5, 2003-10-20 11:13:31-02:00, neilb@cse.unsw.edu.au [PATCH] Remove un-necessary locking in lockd Two places in lockd readlock the nfsd export table before calling exp_getclient. This is not needed because the export table is already locked by code in svc.c. It is a problem because if a write_lock is attempted while the svc.c readlock is active, and before the secondary readlock is attempted, you get a deadlock. ----------- Diffstat output ------------ ./fs/lockd/svc4proc.c | 2 -- ./fs/lockd/svcproc.c | 2 -- 2 files changed, 4 deletions(-) ChangeSet@1.1148.21.4, 2003-10-20 08:37:58-02:00, marcelo@logos.cnet Shantanu Goel: Fix merge mistake in refill_inactive() ChangeSet@1.1148.21.3, 2003-10-19 19:01:52-02:00, mikpe@csd.uu.se [PATCH] APICBASE fix backport from 2.6 If the BIOS boots us with the local APIC enabled but at a non-standard address, and we're not parsing any MP/ACPI tables, then we will (a) map the wrong address, and (b) write the wrong address to APICBASE at resume from suspend. Apparently some HP machines are affected by this bug. This is a backport of the fix Linus put in 2.6.0-test7-bk8. /Mikael ChangeSet@1.1148.12.13, 2003-10-17 23:43:09-07:00, wensong@linux-vs.org [IPVS]: Fix ip_vs_ftp to use cp->vaddr because iph->daddr is already mangled. ChangeSet@1.1063.44.28, 2003-10-17 16:55:04-04:00, len.brown@intel.com [ACPI] correct parameter to acpi_ev_gpe_dispatch() take II (Bob Moore) ChangeSet@1.1148.2.51, 2003-10-17 16:21:24-04:00, len.brown@intel.com Merge intel.com:/home/lenb/bk/linux-acpi-test-2.4.22 into intel.com:/home/lenb/bk/linux-acpi-test-2.4.23 ChangeSet@1.1063.44.27, 2003-10-17 15:30:36-04:00, len.brown@intel.com [ACPI] correct parameter to acpi_ev_gpe_dispatch() (Shaohua David Li) ChangeSet@1.1063.44.26, 2003-10-17 15:25:40-04:00, len.brown@intel.com [ACPI] acpi_ec_gpe_query(ec) fix for T40 crash (Shaohua David Li) http://bugme.osdl.org/show_bug.cgi?id=1171 ChangeSet@1.1063.44.25, 2003-10-17 15:02:45-04:00, len.brown@intel.com [ACPI] fix object reference count bug for battery status (Shaohua David Li) http://bugme.osdl.org/show_bug.cgi?id=1038 ChangeSet@1.1063.44.24, 2003-10-17 14:08:09-04:00, len.brown@intel.com [ACPI] speed up reads from /proc/acpi/ (Shaohua David Li) http://bugme.osdl.org/show_bug.cgi?id=726 ChangeSet@1.1063.44.23, 2003-10-16 22:09:38-04:00, len.brown@intel.com [ACPI] delete descriptions for stale ACPI build options ChangeSet@1.1148.12.12, 2003-10-15 22:09:42-07:00, gorgo@thunderchild.debian.net [NET]: Fix get_random_bytes() call in sunhme.c:get_hme_mac_nonsparc(). ChangeSet@1.1148.7.11, 2003-10-14 16:48:33-07:00, david-b@pacbell.net [PATCH] USB: ehci-hcd, misc bugfixes This fixes some bugs observed in the EHCI code: - Byte-order confusion caused the wrong address to be set on big-endian hardware (reported last week on PPC and SPARC). That bug's been there for about a year, with no problem reports ... hmm. (a while less on 2.4) - Used the wrong bitmask to determine max packet size for interrupt transfers, so they were limited to 1023 bytes (not 1024 bytes) at high speed. - Because those two problems related to the masking, I sanity checked it and moved more of byteswapping to compile time. - Removes some oopsing in the (debug) periodic schedule dump, seen with patches that add more interesting behaviors (which folk are finally trying...). ChangeSet@1.1148.7.10, 2003-10-14 16:48:25-07:00, david-b@pacbell.net [PATCH] USB: usb ethernet gadget Please update the ethernet "gadget" driver with this patch: - fixes an newish oops in some cdc shutdown paths - works on more 2.4 kernel variants - synchronizes with more recent 2.6 code - supports another controller (tc86c001 "goku_udc") The oops basically came from recent changes that seemed to behave on PXA hardware, but not the more functional CDC-capable stuff (like net2280). ChangeSet@1.1148.7.9, 2003-10-14 16:48:18-07:00, bunk@fs.tum.de [PATCH] USB: add USB gadget Configure help entries In 2.4.23-pre7 USB gadget support was added but no Configure.help entries were added. The patch below adds these missing entries. the help texts were copied from 2.6, please check whether they are correct for 2.4, too. ChangeSet@1.1148.7.8, 2003-10-14 16:48:09-07:00, dax@gurulabs.com [PATCH] USB: Add Handspring Treo 600 ids I've tested this with my new Treo 600 and it works. Please apply. ChangeSet@1.1148.2.49, 2003-10-14 17:45:09-04:00, lethal@unusual.internal.linux-sh.org sh64: Add pcibios_scan_all_fns() definition. ChangeSet@1.1148.2.48, 2003-10-14 17:43:22-04:00, lethal@unusual.internal.linux-sh.org sh: Add BPS_230400 definition to sh-sci This fixes a compilation failure with the current sh-sci.c, due to BPS_230400 being undefinied. ChangeSet@1.1148.2.47, 2003-10-14 17:32:53-04:00, lethal@unusual.internal.linux-sh.org sh: irq_intc2 updates Patch from Alex Bennee: Merged in the INTC2 changes from ST's tree Should also fix broken ST40 PCI Interrupts. arch/sh/kernel/irq_intc2.c | 157 +++++++++++++++++++++++++++++++++++---------- arch/sh/kernel/irq_ipr.c | 6 + include/asm-sh/irq.h | 21 +++--- 3 files changed, 142 insertions(+), 42 deletions(-) ChangeSet@1.1148.2.46, 2003-10-14 17:22:06-04:00, lethal@unusual.internal.linux-sh.org sh: signal trampoline workaround for SH-4 core bug Patch from Kaz Kojima: The attached patch is mainly to avoid an SH-4 core problem described in http://ftp.america.renesas.com/netshare/capp01/Technical_Notices/Renesas/tnsh7456ae.pdf This patch is proposed by Sugioka-san and includes a forgotten part which set optional arguments for the signal handler in setup_frame(). arch/sh/kernel/signal.c | 35 ++++++++++++++++++++++++++--------- 1 files changed, 26 insertions(+), 9 deletions(-) ChangeSet@1.1148.12.11, 2003-10-14 14:02:23-07:00, chas@cmf.nrl.navy.mil [ATM]: setsockopt/getsockopt cleanup ChangeSet@1.1148.12.10, 2003-10-14 14:02:00-07:00, chas@cmf.nrl.navy.mil [ATM]: move vcc's to global sk-based linked list ChangeSet@1.1148.12.9, 2003-10-14 14:01:30-07:00, chas@cmf.nrl.navy.mil [ATM]: eliminate SOCKOPS_WRAPPED ChangeSet@1.1148.12.8, 2003-10-14 14:00:57-07:00, chas@cmf.nrl.navy.mil [ATM]: cleanup connect ChangeSet@1.1148.12.7, 2003-10-14 14:00:31-07:00, chas@cmf.nrl.navy.mil [ATM]: remove listenq and backlog_quota from struct atm_vcc ChangeSet@1.1148.12.6, 2003-10-14 13:59:37-07:00, chas@cmf.nrl.navy.mil [ATM]: rewrite recvmsg to use skb_copy_datagram_iovec ChangeSet@1.1148.18.2, 2003-10-14 13:16:13-07:00, davem@nuts.ninka.net [SPARC]: Audit inline asm. Check all inline asms for common bugs and things that prevent gcc-3.3 builds from working, in particular: 1) Missing memory or cc clobbers. 2) Do not clobber registers explicitly assigned to input variables. 3) extern __inline__ --> static inline. Also try to make the formatting more consistent so that future audits are a bit easier. Based upon work done by Keith W. and Meelis Roos. ChangeSet@1.1148.2.44, 2003-10-14 16:14:12-02:00, marcelo@logos.cnet Remove parcelfarce email from CREDITS ChangeSet@1.1148.17.9, 2003-10-14 14:40:13-04:00, pp@ee.oulu.fi [PATCH] b44 enable interrupts after tx timeout (2.4.23-pre6) Here's the 2.4.23-pre6 version, which syncs the driver fully with 2.6. Seems to work (tm). ChangeSet@1.1148.17.8, 2003-10-14 14:34:32-04:00, manfred@colorfullife.com [netdrvr natsemi] fix ring clean Too much copy&paste in a call to pci_unmap_single. ChangeSet@1.1148.17.7, 2003-10-14 14:33:10-04:00, jgarzik@redhat.com [netdrvr tulip] add pci id Contributed by Ken Zalewski. ChangeSet@1.1148.17.6, 2003-10-14 14:25:00-04:00, jgarzik@redhat.com [netdrvr 8139too] another new PCI ID Contributed by Josh Litherland, Donald Becker, and others. ChangeSet@1.1148.17.5, 2003-10-14 14:20:46-04:00, jgarzik@redhat.com [netdrvr 8139too] add pci id contributed by "JaReK" and Donald Becker. ChangeSet@1.1148.17.4, 2003-10-14 14:15:42-04:00, scott.feldman@intel.com [PATCH] hang on ZEROCOPY/TSO when hitting no-Tx-resources * Critical bug fix: under heavy Tx stress using ZEROCOPY or TSO, if we run out of Tx descriptors, we don't calculate for the context descriptor used as the first of the ZEROCOPY/TSO send, nor do we clean up the context descriptor bits in the case where the send isn't going to fit and we need to undo the mappings. This bug was introduced with the 5.2.16 patch set which included a workaround for a hang on 82544 over PCI-X. This workaround caused the check for no-Tx- resource logic to change, and this bug slipped in. ChangeSet@1.1148.17.3, 2003-10-14 13:52:30-04:00, scott.feldman@intel.com [PATCH] ethtool_ops eeprom stuff Finally got around to adding ethtool_ops to e100-3.0.x. I found a bug with get_eeprom() and it seems to work best if we add get_eeprom_len() to the ops list. Also moved check for offest + len < size into ethtool.c. I was able to test [GS]EEPROM, PHYS_ID, GSTATS, GSTRINGS, and TEST, and everything looks good. Should I send same for 2.4? ChangeSet@1.1148.17.2, 2003-10-14 13:34:34-04:00, mlindner@syskonnect.de [PATCH] sk98lin-2.4: Driver update to version 6.18 here is a new version of the sk98lin driver (v6.18) with some changes for the kernel 2.4.22-bk25. Patch 1/1 * Add: New Gigabit LOM vendors (Epox, Gigabyte, Abit, Iwill, ECS, Asus) * Add: Better Support for Yukon Plus chipset * Fix: IO-control return race condition * Fix: Interrupt moderation value check * Fix: PCI initialization (SkGeInitPCI) * Fix: TCP and UDP HW Checksum calculation ChangeSet@1.1148.17.1, 2003-10-14 13:25:15-04:00, jgarzik@redhat.com [netdrvr xircom_cb] backport 2.6 changes * use alloc_etherdev; fixes races in stats pointer setting (by Stephen Hemminger) * kill unnecessary whitespace and #includes * use C99-style initializers * speling and irqreturn_t changes ChangeSet@1.1155, 2003-10-14 11:17:08-05:00, shaggy@shaggy.austin.ibm.com JFS: Improved error handing This patch replaces many assert statements, which caused a BUG(), with improved code to mark the superblock dirty and then proceed as specified by the errors= mount flag (as ext2 and ext3 do). JFS's default for the errors option is "remount-ro" in order to prevent addition data corruption when a problem is found. These asserts are usually triggered by on-disk data corruption. By marking the superblock dirty, fsck will perform a complete check on the file system and correct the problems, rather than simply replaying the journal, inviting later trouble. Submitted by Karl Rister & Dave Kleikamp ChangeSet@1.1148.12.5, 2003-10-13 12:42:51-07:00, gandalf@netfilter.org [NETFILTER]: Remove unused destroy callback in ip6t_ipv6header.c, from Maciej Soltysiak ChangeSet@1.1148.12.4, 2003-10-13 12:40:52-07:00, kaber@trash.net [NETFILTER]: Add size check to ip_nat_mangle_udp_packet ChangeSet@1.1148.12.3, 2003-10-13 12:34:02-07:00, wensong@linux-vs.org [IPVS] Fix to set the statistics of dest zero when bound to a new service ChangeSet@1.1148.15.4, 2003-10-13 13:07:46+02:00, marcel@holtmann.org [Bluetooth] Add support for an old ALPS module This patch adds the specific vendor and product id's for an old ALPS module which don't uses the USB Bluetooth class id. ChangeSet@1.1148.15.3, 2003-10-13 13:06:48+02:00, marcel@holtmann.org [Bluetooth] Add support for the Digianswer USB devices The Digianswer USB Bluetooth devices uses a bRequestType of 0x40 for HCI commands, but the Bluetooth specification says that the correct value is 0x20. This patch sets the needed value according to the vendor and product id's of the Bluetooth device. ChangeSet@1.1148.15.2, 2003-10-13 13:05:41+02:00, marcel@holtmann.org [Bluetooth] Remove USB zero packet option This patch removes the USB zero packet option from the HCI USB Bluetooth driver, because it is useless. ChangeSet@1.1148.15.1, 2003-10-13 13:04:31+02:00, marcel@holtmann.org [Bluetooth] Always use two ISOC URB's This patch modifies the USB Bluetooth driver to use two ISOC URB's per RX and TX transfer paths. This is needed for in time transfer of SCO audio packets over HCI. ChangeSet@1.1148.2.41, 2003-10-12 15:09:20-02:00, marcelo@logos.cnet MIPS/MIPS64: Clear IRQ_INPROGRESS in setup_irq() ChangeSet@1.1148.2.40, 2003-10-12 15:03:04-02:00, olh@suse.de [PATCH] Fix NinjaSCSI compilation This patch is required to compile the NinjaSCSI-32Bi Cardbus driver: ChangeSet@1.1148.2.39, 2003-10-12 13:59:31-02:00, marcelo@logos.cnet x86: Clear IRQ_INPROGRESS in setup_irq() ChangeSet@1.1148.2.38, 2003-10-12 13:53:58-02:00, viro@parcelfarce.linux.theplanet.co.uk [PATCH] Alpha: clear IRQ_INPROGRESS in setup_irq() ChangeSet@1.1148.2.37, 2003-10-12 13:50:51-02:00, marcelo@logos.cnet Al Viro: Clear all flags in exec_usermodehelper ChangeSet@1.1148.14.6, 2003-10-11 17:16:44-04:00, trond.myklebust@fys.uio.no Make the client act correctly if the RPC server's asserts that it does not support a given program, version or procedure call. ChangeSet@1.1148.14.5, 2003-10-11 17:15:38-04:00, trond.myklebust@fys.uio.no Fix a stack overflow problem that was noticed by Jeff Garzik by removing some unused readdirplus cruft. ChangeSet@1.1148.14.4, 2003-10-11 17:13:44-04:00, trond.myklebust@fys.uio.no Increase the minimum RTO timer value to 1/10 second. This is more in line with what is done for TCP. ChangeSet@1.1148.14.3, 2003-10-11 17:12:56-04:00, trond.myklebust@fys.uio.no UDP round trip timer fix. Modify Karn's algorithm so that we inherit timeouts from previous requests. This means that we lengthen the window of time during which we accept updates to the RTO estimate if we see an update. Scheme proposed by Brian Mancuso, but it is standard for TCP congestion control implementations. ChangeSet@1.1148.14.2, 2003-10-11 17:12:08-04:00, trond.myklebust@fys.uio.no A request cannot be used as part of the RTO estimation if it gets resent since you don't know whether the server is replying to the first or the second transmission. However we're currently setting the cutoff point to be the timeout of the first transmission. This patch moves the cutoff point to the actual start of the retransmission. ChangeSet@1.1148.14.1, 2003-10-11 17:11:04-04:00, trond.myklebust@fys.uio.no Fix a deadlock in the NFS asynchronous write code. shrink_cache() does not have a reference to the inode when it calls down to writepage(), only a lock on the page to write. This may cause a deadlock if we have to wait on an existing request to complete since this may result in a call to iput() and hence to clear_inode(). Problem originally diagnosed by David Jeffery... ChangeSet@1.1148.13.2, 2003-10-11 13:58:39-07:00, davem@nuts.ninka.net [SPARC64]: Update defconfig. ChangeSet@1.1148.12.2, 2003-10-11 12:19:32-07:00, olh@suse.de [IRDA]: Fix build with gcc-3.4 ChangeSet@1.1148.12.1, 2003-10-11 12:02:47-07:00, ja@ssi.bg [IPV4]: ip_fragment must copy the nfcache field. ChangeSet@1.1148.2.36, 2003-10-11 14:29:33-03:00, Matt_Domsch@dell.com [PATCH] Fix megaraid2 compilation problems > I have the fix for the nsp driver but not for the megaraid2. ftp://ftp.lsil.com/pub/linux-megaraid/drivers/version-2.00.9/megaraid-2009-wo-hostlock.patch.gz ftp://ftp.lsil.com/pub/linux-megaraid/drivers/version-2.00.9/megaraid-2009-wo-hostlock.patch.gz.sig and appended. Done by Atul Mukker @ LSIL. -- Matt Domsch Sr. Software Engineer, Lead Engineer Dell Linux Solutions www.dell.com/linux Linux on Dell mailing lists @ http://lists.us.dell.com ChangeSet@1.1148.2.35, 2003-10-09 22:50:03-07:00, davem@nuts.ninka.net [VLAN]: kfree(skb) --> kfree_skb(skb). ChangeSet@1.1148.2.34, 2003-10-09 22:49:34-07:00, davem@nuts.ninka.net [NET]: linux/in.h needs linux/socket.h ChangeSet@1.1148.2.33, 2003-10-09 18:40:15-03:00, marcelo@logos.cnet nsp_cs.h: Remove irqreturn_t definition TAG: v2.4.23-pre7 ChangeSet@1.1148.2.32, 2003-10-09 18:29:47-03:00, marcelo@logos.cnet Cset exclude: bjorn.helgaas@hp.com|ChangeSet|20031007181555|25551 ChangeSet@1.1148.2.31, 2003-10-09 18:21:40-03:00, erik@harddisk-recovery.nl [PATCH] Change sym53c8xx_2 driver module name Anyway, here's a patch to rename sym53c8xx_2/sym53c8xx.o to sym53c8xx_2/sym53c8xx_2.o. It's against 2.4.22, but should apply to all 2.4 kernels. ChangeSet@1.1148.2.30, 2003-10-09 18:11:24-03:00, marcelo@logos.cnet Changed EXTRAVERSION to -pre7 ChangeSet@1.1148.2.29, 2003-10-09 17:54:32-03:00, marcelo@logos.cnet Fixup exec_mmap() race fix mess ChangeSet@1.1148.2.28, 2003-10-09 17:23:10-03:00, shaggy@austin.ibm.com [PATCH] BUG() in exec_mmap() Marcelo, A recent change to exec_mmap() removed the initialization of old_mm, leaving an uninitialized use of it. This patch would completely remove old_mm from the function. Is this what was intended? ===== fs/exec.c 1.30 vs edited ===== ChangeSet@1.1148.2.27, 2003-10-09 17:02:18-03:00, marcelo@logos.cnet Merge http://mdomsch.bkbits.net/linux-2.4-edd into logos.cnet:/home/marcelo/bk/linux-2.4 ChangeSet@1.1148.2.26, 2003-10-09 15:00:26-03:00, marcelo@logos.cnet From -aa tree: Fix potential PPPoE oops ChangeSet@1.1148.2.25, 2003-10-09 10:55:25-03:00, paulus@samba.org [PATCH] add Configure.help entries Marcelo, The patch below adds entries to Configure.help for config symbols used on PPC in various Config.in files in the kernel tree that don't currently have help. (I did a grep for symbols without help for which we did have help in the linuxppc-2.4 tree.) Please apply. Thanks, Paul. ChangeSet@1.1148.2.24, 2003-10-09 10:49:45-03:00, marcelo@logos.cnet Remove racy optimization from exec_mmap() ChangeSet@1.1148.1.2, 2003-10-08 22:59:44-07:00, maxk@qualcomm.com [Bluetooth] Credit based flow control (CFC) must be disabled by default for compatibility with 1.0b devices. Made CFC a session attribute, introduced CFC states and cleaned up CFC logic. ChangeSet@1.1148.7.7, 2003-10-08 12:04:27-07:00, luca@libero.it [PATCH] USB: add W996[87]CF driver. ChangeSet@1.1148.2.23, 2003-10-08 15:15:50-03:00, riel@redhat.com [PATCH] page->flags corruption fix documentation # This is a BitKeeper generated patch for the following project: # Project Name: Linux kernel tree # This patch format is intended for GNU patch command version 2.5 or higher. # This patch includes the following deltas: # ChangeSet 1.1193 -> 1.1194 # mm/filemap.c 1.89 -> 1.90 # # The following is the BitKeeper ChangeSet Log # -------------------------------------------- # 03/10/08 riel@cessna.boston.redhat.com 1.1194 # comment on why the atomic updates are needed, on request # -------------------------------------------- # ChangeSet@1.1148.2.22, 2003-10-08 14:36:59-03:00, marcelo@logos.cnet Merge bk://kernel.bkbits.net/davem/net-2.4 into logos.cnet:/home/marcelo/bk/linux-2.4 ChangeSet@1.1148.2.20, 2003-10-08 09:38:33-07:00, davem@nuts.ninka.net [NET]: Fix userland iproute2 build problems introduced by mcast changes. ChangeSet@1.1148.2.19, 2003-10-08 09:02:30-07:00, tommy.christensen@tpack.net [VLAN]: Do not modify the data of shared SKBs. ChangeSet@1.1148.2.18, 2003-10-08 08:37:22-07:00, davem@nuts.ninka.net [UDP/TCP]: Fix binding conflict tests wrt. SO_BINDTODEVICE. ChangeSet@1.1148.9.4, 2003-10-08 11:04:16-03:00, marcelo@logos.cnet Change my mail address in a few places ChangeSet@1.1148.9.3, 2003-10-08 10:53:39-03:00, marcelo@logos.cnet Fix typo in laptop mode patch ChangeSet@1.1148.9.2, 2003-10-08 10:49:45-03:00, axboe@suse.de [PATCH] laptop mode Hi Marcelo, Lots of people have been using this patch with great success, and it's been in the SuSE kernel for some months now too. It is also in the -benh ppc tree Basically, it introduces a write back mode of dirty and journal data that is more suitable for laptops. At the block layer end, it schedules write out of dirty data after the disk has been idle for 5 seconds. Laptop mode can be switched on and off with /proc/sys/vm/laptop_mode. There is also a block_dump sysctl, which if enabled will dump who dirties and writes out data. This is very helpful in pinning down who is causing unnecessary writes to the disk. # This is a BitKeeper generated patch for the following project: # Project Name: Linux kernel tree # This patch format is intended for GNU patch command version 2.5 or higher. # This patch includes the following deltas: # ChangeSet 1.1200 -> 1.1201 # include/linux/sysctl.h 1.33 -> 1.34 # kernel/sysctl.c 1.27 -> 1.28 # drivers/block/ll_rw_blk.c 1.50 -> 1.51 # fs/jbd/journal.c 1.12 -> 1.13 # include/linux/fs.h 1.89 -> 1.90 # mm/filemap.c 1.89 -> 1.90 # fs/buffer.c 1.96 -> 1.97 # (new) -> 1.1 Documentation/laptop-mode.txt # # The following is the BitKeeper ChangeSet Log # -------------------------------------------- # 03/10/08 axboe@smithers.home.kernel.dk 1.1201 # laptop mode # -------------------------------------------- # ChangeSet@1.1148.9.1, 2003-10-08 07:28:52-03:00, xose@wanadoo.es [PATCH] megaraid2 merge fixes Marcelo Tosatti wrote: > Fine. I agree on adding in to mainline 2.4.x. Here go some trivial fixes(add menu entry, list all compatibles boards in help and put tab instead spaces) for megaraid2. -thanks- -- If it wasn't for the presence of Lara Croft and Xena Warrior Princess, techies around the world would have posters of Torvalds on their walls. [ Part 2: "Attached Text" ] ChangeSet@1.1063.44.22, 2003-10-07 23:43:03-04:00, len.brown@intel.com [ACPI] fix acpi_asus module build (Stephen Hemminger) ChangeSet@1.1148.8.18, 2003-10-07 18:13:32-03:00, mike.miller@hp.com [PATCH] cciss update: support new controller The patch below adds support for a new intregrated cciss based controller. It also bumps the version to 2.4.50. It was built and tested against the 2.4.23-pre6 kernel tree. Please consider this patch for inclusion. Thanks, mikem ChangeSet@1.1148.8.17, 2003-10-07 18:00:31-03:00, jack@ucw.cz [PATCH] Fix quota counter overflow Hello, I'm sending you a patch which should fix a problem with used-bytes counter overflow in old quota format. The patch should apply well for both 2.4.22 and 2.6.0-test6 kernels. Please apply. ChangeSet@1.1148.8.16, 2003-10-07 16:02:15-03:00, andrea@suse.de [PATCH] do_proc_readlink failpath Hi Marcelo and Linus, this fix from spender@grsecurity.net applies to both 2.4 and 2.6. Not sure exactly when d_path can fail but it certainly looks reasonable (and if somebody can make it fail, he can leak lots of ram) Index: fs/proc//base.c =================================================================== RCS file: /home/andrea/crypto/cvs/linux-2.4/fs/proc/base.c,v retrieving revision 1.15 ChangeSet@1.1148.8.15, 2003-10-07 15:56:21-03:00, riel@redhat.com [PATCH] page->flags corruption fix In the "better safe than sorry" category. Thanks go out to Matt Domsch and Robert Hentosh. A similar fix went into the 2.6 kernel. Please apply. # This is a BitKeeper generated patch for the following project: # Project Name: Linux kernel tree # This patch format is intended for GNU patch command version 2.5 or higher. # This patch includes the following deltas: # ChangeSet 1.1192 -> 1.1193 # include/linux/mm.h 1.43 -> 1.44 # mm/page_alloc.c 1.63 -> 1.64 # mm/filemap.c 1.88 -> 1.89 # # The following is the BitKeeper ChangeSet Log # -------------------------------------------- # 03/10/07 riel@cessna.boston.redhat.com 1.1193 # fix page->flags corruption due to races between atomic and non-atomic # accesses, originally found and fixed by Robert Hentosh and Matt Domsch # -------------------------------------------- # ChangeSet@1.1148.8.14, 2003-10-07 15:45:54-03:00, benh@kernel.crashing.org [PATCH] kupdated: correctly dequeue SIGSTOP signals Here's the fix to the problem we have been discussing... ===== fs/buffer.c 1.80 vs edited ===== ChangeSet@1.1148.8.13, 2003-10-07 15:15:55-03:00, bjorn.helgaas@hp.com [PATCH] 2.4 IA64 DRM interface changes This patch adds a "dev" parameter to drm_ioremap(), drm_ioremap_nocache(), and drm_ioremapfree(). This change is already in 2.5, and is required to support DRM on ia64. The problem on ia64 is that CPU addresses don't get remapped by the GART, so we have to fiddle with the kernel page tables, and the ioremap interfaces need the dev pointer to do that. This part is strictly interface changes; the actual use of the new parameter is messier code that I'll send separately. The changes below should be straightforward enough to be fairly non-controversial. Bjorn ===== drivers/char/drm/drmP.h 1.10 vs edited ===== ChangeSet@1.1148.8.12, 2003-10-07 15:12:50-03:00, ink@jurassic.park.msu.ru [PATCH] Alpha: backport force_successful_syscall_return() ChangeSet@1.1148.8.11, 2003-10-07 15:02:04-03:00, marcelo@logos.cnet Cset exclude: 20030702202648|35018: i82092 update broke existing working setups ChangeSet@1.1148.2.17, 2003-10-07 08:36:10-07:00, ja@ssi.bg [IPV4/IPV6]: Do not modify skb->h.raw until skb is unshared. ChangeSet@1.1148.2.16, 2003-10-07 08:24:40-07:00, davem@nuts.ninka.net [NET]: Size hh_cache->hh_data[] properly. ChangeSet@1.1148.2.15, 2003-10-07 07:59:11-07:00, wensong@linux-vs.org [IPVS] add strict boundary check in parsing FTP commands ChangeSet@1.1148.2.14, 2003-10-07 07:58:44-07:00, wensong@linux-vs.org [IPVS] Fix ip_vs_tunnel_xmit to return NF_DROP when no memory available ChangeSet@1.1148.2.13, 2003-10-07 07:57:28-07:00, vandrove@vc.cvut.cz [IPV4]: Fix deadlock on ip_mc_list->lock ChangeSet@1.1148.8.10, 2003-10-07 11:49:54-03:00, kraxel@bytesex.org [PATCH] videodev update Hi, This patch is a minor update for the videodev.o module. It backports some of the recent 2.6.x changes to 2.4.x to simplify v4l driver maintainance. Gerd ChangeSet@1.1148.8.9, 2003-10-07 11:49:14-03:00, jes@wildopensource.com [PATCH] Major qla1280 update Hi Marcelo, About time from me but better late than never. This patch brings the qla1280 driver in 2.4.23pre6 upto the levels of whats in 2.6. It's a monster patch since it's pretty much a 85% rewrite or so. ChangeSet@1.1148.8.8, 2003-10-07 11:46:29-03:00, marcelo@logos.cnet Add Megaraid 2 driver Makefile, megaraid2.h, megaraid2.c, changelog.megaraid2, Configure.help: Import patch megaraid2 megaraid2.h, megaraid2.c, changelog.megaraid2: new file ChangeSet@1.1148.2.12, 2003-10-07 07:45:29-07:00, jan.oravec@6com.sk [IPV6]: Deactivate timers properly in ipv6_mc_destroy_dev(). ChangeSet@1.1148.8.7, 2003-10-07 11:43:43-03:00, perex@suse.cz [PATCH] PageReserved memory counting fix Hi, this fixes rss pages counting for drivers which returns a reserved page from the nopage callback (like ALSA). Thank you for applying to the 2.4 tree. ===== memory.c 1.56 vs edited ===== ChangeSet@1.1148.8.6, 2003-10-07 09:18:54-03:00, kraxel@bytesex.org [PATCH] Tuner update Hi, This is a update for the v4l tuner module. Also has the i2c changes for the new core code and a minor radio tuner fix. Gerd ChangeSet@1.1148.8.5, 2003-10-07 09:05:09-03:00, kraxel@bytesex.org [PATCH] bttv documentation update Hi, This patch updates the bttv card list, please apply, Gerd ChangeSet@1.1148.8.4, 2003-10-07 09:04:35-03:00, kraxel@bytesex.org [PATCH] bttv driver update Hi, This patch updates the bttv driver. Changes: * some card-specific updates. * fixes the firmware loader build problems. * i2c code got some ifdefs to build+work with both old and new i2c core code. Please apply, Gerd ChangeSet@1.1148.8.3, 2003-10-07 09:04:12-03:00, kraxel@bytesex.org [PATCH] v4l i2c modules update Hi, This patch updates the v4l i2c modules. It adds some #ifdefs to make them build+work with both old and new i2c core core. Not really needed, but nice-to-have for users which manually patch the kernel with recent i2c + lm_sensors patches. Gerd ChangeSet@1.1148.2.11, 2003-10-07 05:01:41-07:00, davem@nuts.ninka.net [NETLINK]: Set socket error on netlink_ack() allocation failure. ChangeSet@1.1148.8.2, 2003-10-07 07:54:35-03:00, marcelo@logos.cnet Cset exclude: ak@muc.de|ChangeSet|20031005214700|30577 ChangeSet@1.1148.8.1, 2003-10-07 07:43:23-03:00, andersen@codepoet.org [PATCH] fix 2.4.x incorrect argv[0] for init In 2.4.x when someone specifies "init=/bin/foo" to select an alternative binary to run instead of /sbin/init, argv[0] is not to the correct value. This is a problem for programs such as busybox that multiplex applications based on the value of argv[0]. For example, even if you specify init=/bin/sh" on the kernel command line, busybox will still receive "/sbin/init" as argv[0], and will therefore run init rather than /bin/sh... This problem was recently fixed in 2.6.x. http://linux.bkbits.net:8080/linux-2.5/diffs/init/main.c@1.108?nav=index.html|src/|src/init|hist/init/main.c Could you please also apply this patch to 2.4.x? -Erik -- Erik B. Andersen http://codepoet-consulting.com/ --This message was written using 73% post-consumer electrons-- ChangeSet@1.1148.4.35, 2003-10-07 01:16:18-07:00, ebrower@usa.net [SPARC64]: Fix kernel_thread() return value check in envctrl.c ChangeSet@1.1148.4.34, 2003-10-07 01:13:08-07:00, davem@nuts.ninka.net [SPARC64]: Implement force_successful_syscall_return(). ChangeSet@1.1148.4.32, 2003-10-06 21:56:00-03:00, viro@parcelfarce.linux.theplanet.co.uk [PATCH] attach_mnt() fix Please, apply. As it is, copy_tree() ends up with the reversed order of children in each node, which makes copy_namespace() do very interesting things... ChangeSet@1.1148.4.31, 2003-10-06 21:48:45-03:00, viro@parcelfarce.linux.theplanet.co.uk [PATCH] Cleanup /proc/ioports seqfile conversion Incremental to the posted patch: ChangeSet@1.1148.4.30, 2003-10-06 21:47:59-03:00, olof@austin.ibm.com [PATCH] Convert /proc/ioports to seqfile Marcelo, Al, Sorry for the delay in following up with this. I backported Al's 2.5 code as it was in our BK (which is synced to the bkbits one periodically), it had the patch from LKML in there. The init code ended up being a little different due to changes in infrastructure, but the functionality is the same. I gave it a few swings here, I don't have a machine right now that will result in >4k of data but I don't see why that would make much of a difference in the seq_file case. ===== fs/proc/proc_misc.c 1.23 vs edited ===== ChangeSet@1.1148.4.29, 2003-10-06 21:29:30-03:00, ak@suse.de [PATCH] Fix bug on ACPI sysrq poweroff Sylvain Pasche writes: > Hi, > > If I want to halt the system using sys-rq - o key, I get an oops instead > of a power down. > Inside pm.c:159, there is: > > if (in_interrupt()) > BUG(); > > But if we look at the trace, we are in the interrupt of the keyboard > handler. > One fix would be to comment out the BUG line, but there's certainly "a > better way to do it". Use this patch. ChangeSet@1.1148.4.28, 2003-10-06 21:24:23-03:00, marcelo@logos.cnet Fix ACPI config.in breakage ChangeSet@1.1148.4.27, 2003-10-06 21:23:13-03:00, marcelo@logos.cnet Fix dscc4 net/wan Config.in breakage ChangeSet@1.1148.7.6, 2003-10-06 15:56:00-07:00, greg@kroah.com [PATCH] USB: fix compiler error in sl811.c ChangeSet@1.1148.7.5, 2003-10-06 15:54:07-07:00, henry.ne@arcor.de [PATCH] USB: Update SL811, HC_SL811 driver Update and Bug fix for driver sl811 and hc_sl811 Diff build from kerneltree linux-2.4.23-pre6 (02.10.2003). Please tell me, if I should separate something of this patch in diffetent files. Changes by Henry.Ne@Arcor.de: o General cleanup driver to 2.4.23 o Moving hc_sl811 into directory drivers/usb/host o Moving hardware inline functions into include/asm o Enable driver in Configuration and Make o Help for configure, general help for SL811 o Add architecture x86 o Two and more instances of controllers (x86 only) o Some bugfixed for hc_sl811 o Remove some warnings ChangeSet@1.1148.7.4, 2003-10-06 14:45:21-07:00, henning@meier-geinitz.de [PATCH] USB: scanner driver: use static declarations (3/3) Use static declarations for usb_scanner_init/usb_scanner_exit (Daniele Bellucci). ChangeSet@1.1148.7.3, 2003-10-06 14:44:58-07:00, henning@meier-geinitz.de [PATCH] USB: scanner driver: added USB_CLASS_CDC_DATA (2/3) Added test for USB_CLASS_CDC_DATA which is used by some fingerprint scanners. ChangeSet@1.1148.7.2, 2003-10-06 14:44:36-07:00, henning@meier-geinitz.de [PATCH] USB: scanner driver: new device ids (1/3) The following 3 patches are the backports from those already applied to 2.6. Added vendor/product ids for Canon, HP, Microtek, Mustek, Siemens, UMAX, and Visioneer scanners. ChangeSet@1.1148.7.1, 2003-10-06 14:44:14-07:00, dan@reactivated.net [PATCH] USB brlvger: Debug code fixes When compiling this driver with WRITE_DEBUG defined (for extra debug output), gcc outputs warnings for three similar printf statements. The specifier in each printf statement is not totally correct. This patch removes these warnings. ChangeSet@1.1134.1.3, 2003-10-06 14:21:00-07:00, trini@kernel.crashing.org PPC: Change how we export some Openfirmware device nodes. From Ethan Benson . ChangeSet@1.1063.44.21, 2003-10-06 16:45:00-04:00, len.brown@intel.com [ACPI] Summary of changes for ACPICA version 20031002: Fixed a problem with Index Fields where the index was not incremented for fields that require multiple writes to the index/data registers (Fields that are wider than the data register.) Fixed a problem with all Field objects where a write could go beyond the end-of-field if the field was larger than the access granularity and therefore required multiple writes to complete the request. An extra write beyond the end of the field could happen inadvertently. Fixed a problem with Index Fields where a BUFFER_OVERFLOW error would incorrectly be returned if the width of the Data Register was larger than the specified field access width. Completed fixes for LoadTable() and Unload() and verified their operation. Implemented full support for the "DdbHandle" object throughout the ACPI CA subsystem. Implemented full support for the MADT and ECDT tables in the ACPI CA header files. Even though these tables are not directly consumed by ACPI CA, the header definitions are useful for ACPI device drivers. Integrated resource descriptor fixes posted to the Linux ACPI list. This included checks for minimum descriptor length, and support for trailing NULL strings within descriptors that have optional string elements. Fixed a problem where the SMI_CMD register could be written even if it was not supported on the platform (when FADT.SMI_CMD is zero) ChangeSet@1.1148.4.25, 2003-10-06 17:17:14-03:00, dhinds@sonic.net [PATCH] PCMCIA: cleanup i82365 driver Hi Marcelo, The following patch cleans up the i82365 socket driver, by stripping out a lot of dead #ifdef's. -- Dave ChangeSet@1.1134.1.2, 2003-10-06 08:50:10-07:00, trini@kernel.crashing.org PPC32: Always print the processor number in /proc/cpuinfo. Fix from Dan Burcaw . ChangeSet@1.1148.5.1, 2003-10-06 07:57:06-07:00, trini@kernel.crashing.org Merge bk://linux.bkbits.net/linux-2.4 into kernel.crashing.org:/home/trini/work/kernel/pristine/linux-2.4 ChangeSet@1.1148.4.24, 2003-10-06 10:24:26-03:00, hunold@linuxtv.org [PATCH] Remove bogus Philips SAA7146/DVD decoder info from ioctl-number.txt Hello Marcelo, please apply the following small patch against 2.4 which removes two bogus informations from "ioctl-number.txt": 1) The DVD decoder driver was never finished and never released. I recently removed these informations from our website -- most of the work that was freely available has gone into projects like libdvdread and libdvdnav. 2) There never was a "saa7146 driver" from Philips officially. Instead, in 2.6. my saa7146 driver was included which drives various analog TV cards and various DVB cards through Video4Linux-2. Thanks! Michael Hunold (LinuxTV.org CVS maintainer) [ Part 2: "Attached Text" ] ChangeSet@1.1148.4.23, 2003-10-06 10:19:58-03:00, romieu@fr.zoreil.com [PATCH] dscc4: Workaround for lack of true reset. Workaround for lack of true reset: - devices/ports are put in silent mode at ifconfig down time but some state is kept around to allow 'ifconfig up' issuing at a later time. Device specific structures are allocated when the pci asic is probed: dscc4_init_ring() moves from dscc4_open() to dscc4_found1(). dscc4_release_ring() disappears from dscc4_{open/close}. - try to use reset related board-specific feature at module removal time if available. Comments in the code explain the whole story. Not a complicated feature but it is unavailable on the cards I own and no user gave me feedback -> currently untested. It shouldn't harm anyway. Documentation/Configure.help | 20 ++++++ drivers/net/wan/Config.in | 10 +-- drivers/net/wan/dscc4.c | 141 ++++++++++++++++++++++++++++++++++--------- 3 files changed, 137 insertions(+), 34 deletions(-) ChangeSet@1.1148.4.22, 2003-10-06 10:11:22-03:00, romieu@fr.zoreil.com [PATCH] dscc4: Small fixes - informational message in dscc4_clock_setting() displays the line rate value before and after it changed instead of displaying the changed value twice; - more conditional debug messages. drivers/net/wan/dscc4.c | 15 ++++++++++++--- 1 files changed, 12 insertions(+), 3 deletions(-) ChangeSet@1.1148.4.21, 2003-10-06 09:43:03-03:00, romieu@fr.zoreil.com [PATCH] dscc4: Assorted fixes - Rx and Tx paths both use different macro instead of the ugly TO_STATE(); - small warning for people who would want to change the code. drivers/net/wan/dscc4.c | 22 +++++++++++++++------- 1 files changed, 15 insertions(+), 7 deletions(-) ChangeSet@1.1148.4.20, 2003-10-06 09:07:32-03:00, ak@muc.de [PATCH] Disable devfs for x86-64 As discussed recently: devfs corrupts memory madly on x86-64. As it's unmaintained and I'm not too motivated to debug it this patch just disables it for x86-64. -Andi ChangeSet@1.1148.4.19, 2003-10-06 08:28:11-03:00, yokota@netlab.is.tsukuba.ac.jp [PATCH] NinjaSCSI driver update Hello, Marcelo. Ninja SCSI driver is updated. Please apply this patch for Linux kernel 2.4.22 Detailed changelog in the patch ChangeSet@1.1148.4.18, 2003-10-06 08:21:05-03:00, geert@linux-m68k.org [PATCH] Amiga Zorro bus doc updates Amiga Zorro bus: Update the docs to match the current situation. ChangeSet@1.1148.4.17, 2003-10-06 08:19:34-03:00, geert@linux-m68k.org [PATCH] M68K: Sun-3 SBUS updates Sun-3 SBUS updates: Rename remaining occurrencies of struct linux_sbus_device to struct sbus_dev for source code compatibility with SPARC ChangeSet@1.1148.4.16, 2003-10-06 08:18:15-03:00, geert@linux-m68k.org [PATCH] M68K: Q40/Q60 interrupts Q40/Q60 updates (from Richard Zidlicky): - Must handle keyboard interrupts even before they are registered - Kill some warnings ChangeSet@1.1148.4.15, 2003-10-06 08:17:02-03:00, geert@linux-m68k.org [PATCH] M68k: Fix asm constraints in switch_to M68k: Fix asm constraints in switch_to() (from Andreas Schwab) ChangeSet@1.1148.4.14, 2003-10-06 08:16:05-03:00, geert@linux-m68k.org [PATCH] Minor q40fb fix q40fb: Relax checking of parameters that are ignored anyway (from Richard Zidlicky) ChangeSet@1.1148.4.13, 2003-10-05 18:47:00-03:00, ak@muc.de [PATCH] Fix overflow on unaligned checksum buffers The P6 checksum function can overflow its buffer when the buffer was unaligned to 1 byte. This can hit with sendfile when the application is just passing the right offsets and the buffer is mapped from highmem using kmap. This patch fixes it in a slow path exception handler. ChangeSet@1.1148.4.12, 2003-10-05 16:40:41-03:00, marcelo@logos.cnet Merge bk://kernel.bkbits.net/gregkh/linux/marcelo-2.4 into logos.cnet:/home/marcelo/bk/linux-2.4 ChangeSet@1.1148.4.11, 2003-10-05 16:35:15-03:00, marcelo@logos.cnet Add missing #endif to force_successful_syscall_return change ChangeSet@1.1148.4.10, 2003-10-05 11:12:36-03:00, romieu@fr.zoreil.com [PATCH] dscc4: misc fixes - more #define for specific bits; - more scc_patchl use; - just say no to bozo programming: + SCC core _really_ disabled at startup; + Interrupts Mask Register setup and SCC core activation are done as late as possible (i.e. in dscc4_open()); + they are reverted if dscc4_open() fails; + as well as unconditionnaly in dscc4_close(); - more or less paranoid quirk in Xpr handler. drivers/net/wan/dscc4.c | 53 ++++++++++++++++++++++++++++++++---------------- 1 files changed, 36 insertions(+), 17 deletions(-) ChangeSet@1.1148.4.9, 2003-10-05 11:09:34-03:00, romieu@fr.zoreil.com [PATCH] dscc4: Typo, tabs, unneeded include and misc things from 2.6 Typo, tabs, unneeded include and misc things from 2.6. drivers/net/wan/dscc4.c | 23 +++++++++++++---------- 1 files changed, 13 insertions(+), 10 deletions(-) ChangeSet@1.1148.4.8, 2003-10-05 11:06:58-03:00, romieu@fr.zoreil.com [PATCH] dscc4: More comments Clock modes on dscc4 are feature-rich enough to deserve some explanation. drivers/net/wan/dscc4.c | 51 ++++++++++++++++++++++++++++++++++++++++++++++-- 1 files changed, 49 insertions(+), 2 deletions(-) ChangeSet@1.1148.4.7, 2003-10-05 11:05:30-03:00, romieu@fr.zoreil.com [PATCH] dscc4: add comments Short comments so I can figure why things are done this way. drivers/net/wan/dscc4.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) ChangeSet@1.1148.4.6, 2003-10-05 11:03:56-03:00, romieu@fr.zoreil.com [PATCH] dscc4: dscc4_init_dummy_skb() returns a pointer. dscc4_init_dummy_skb() returns a pointer, not an integer. drivers/net/wan/dscc4.c | 2 +- 1 files changed, 1 insertion(+), 1 deletion(-) ChangeSet@1.1148.4.4, 2003-10-05 10:47:16-03:00, bjorn.helgaas@hp.com [PATCH] Backport force_successful_syscall() Here's a 2.4 backport of this change to 2.5: http://linux.bkbits.net:8080/linux-2.5/cset@1.1046.238.7?nav=index.html Alpha, ppc, and sparc64 define force_successful_syscall_return() in 2.5, but since it's not obvious to me how to do it correctly in 2.4, I left them unchanged. Bjorn ===== drivers/char/mem.c 1.17 vs edited ===== ChangeSet@1.1148.4.3, 2003-10-05 10:41:12-03:00, green@linuxhacker.ru [PATCH] fix LVM memleaks Hello! There are two error patchs in lvm code, that leads to leaking memory. One if creating too many volume gropups and one if incorrect buffer from userspace was passed. Fixes are trivial. Please apply. Found with help of smatch. ===== drivers/md/lvm.c 1.20 vs edited ===== ChangeSet@1.1148.4.2, 2003-10-05 10:36:09-03:00, bjorn.helgaas@hp.com [PATCH] Fix pci_generic_prep_mwi export breakage Looks to me like your tree still needs this change. Here's the patch again in case you need it: ===== drivers/pci/pci.c 1.45 vs edited ===== ChangeSet@1.1148.4.1, 2003-10-05 10:35:13-03:00, ak@muc.de [PATCH] x86-64 ACPI compilation fix [resend as requested] Make x86-64 compile again without ACPI. ChangeSet@1.1136.3.4, 2003-10-04 14:36:57-07:00, sheilds@msrl.com [SPARC64]: Fix typo in bbc_envctrl.c ChangeSet@1.1136.3.3, 2003-10-04 14:26:23-07:00, shields@msrl.com [SPARC64]: Fix watchdog on CP1500/Netra-t1. ChangeSet@1.1120.13.25, 2003-10-03 16:38:20-07:00, greg@kroah.com [PATCH] USB: fix up some non-GPL friendly license wording. Keyspan/InnoSys approved this change. ChangeSet@1.1120.13.24, 2003-10-03 16:22:30-07:00, lucy@innosys.com [PATCH] USB: Keyspan USB adapter patches Changes include: Add support for new USA19H(s) USB Serial Adapter Improve handing of config/control messages for all devices Note that the new adapter doesn't currently require a firmware download. We will add support for re-programming the eeprom with updated firmware when new firmware is required. ChangeSet@1.1148.3.1, 2003-10-03 09:22:43-05:00, Matt_Domsch@dell.com Merge dell.com:/home/mdomsch/bk/linux-2.4 into dell.com:/home/mdomsch/bk/edd/linux-2.4-edd ChangeSet@1.1148.2.10, 2003-10-02 23:38:58-07:00, laforge@netfilter.org [NETFILTER]: Fix SO_ORIGINAL_DST, broken by earlier endianness fixes. ChangeSet@1.1148.2.9, 2003-10-02 23:29:27-07:00, kadlec@blackhole.kfki.hu [NETFILTER]: Make conntrack timeouts become sysctls. This patch introduces the /proc/sys/net/ipv4/netfilter/ directory for the tuneable parameters of the conntrack engine of netfilter. The parameters which can currently be fine-tuned (if required at all) are the ip_conntrack_max, and the conntrack timeout parameters in seconds. The default timeout value for the TCP CLOSE_WAIT state is raised to 3 days, which should cause less problems with long-lived half-closed TCP sessions. For the sake of compatibility with 2.4, /proc/sys/net/ipv4/netfilter/ip_conntrack_max is duplicated as /proc/sys/net/ipv4/ip_conntrack_max as well. ChangeSet@1.1148.2.8, 2003-10-02 23:28:25-07:00, kaber@trash.net [NETFILTER]: Use pre-built table for TCP flag-check in ipt_unclean. ChangeSet@1.1148.2.7, 2003-10-02 23:25:29-07:00, rusty@rustcorp.com.au [NETFILTER]: LOCAL_OUT NAT fix, part 2. Balazs Scheidler pointed out that we can fail to call ip_nat_setup_info for local NAT, which causes the connection to be added twice. Boom. Also move the hash table insertion into ip_nat_setup_info. ChangeSet@1.1148.2.6, 2003-10-02 23:23:39-07:00, laforge@netfilter.org [NETFILTER]: LOCAL_OUT NAT fix. the 25_natcore-nohelper.patch (incorporated in 2.4.22) did actually make things not better. If a NAT rule was matched, num_manips can still be 0. However, some helpers (like PPTP) need to _always mangle the packets, even if there is no NAT mapping for a particular connection. On the other hand, as soon as CONFIG_IP_NF_LOCAL_NAT was disabled, we did end up half-mangling local (non-NAT'ed) connections. This patch removes the LOCAL_OUT hook from the NAT core unless CONFIG_IP_NF_NAT_LOCAL is enabled. ChangeSet@1.1148.2.5, 2003-10-02 23:22:54-07:00, mirage@kaotik.org [NETFILTER]: Add support for mIRC's 'server lookup' DCC address detection to ip_conntrack_irc.c ChangeSet@1.1148.2.4, 2003-10-02 23:04:10-07:00, laforge@netfilter.org [NETFILTER]: Fix UDP checksum in ip_nat_mangle_udp_packet, remove skb->csum hacks ChangeSet@1.1148.2.3, 2003-10-02 23:03:21-07:00, laforge@netfilter.org [NETFILTER]: Don't call ip_conntrack_put with ip_conntrack_lock held ChangeSet@1.1148.2.2, 2003-10-02 05:20:56-03:00, marcelo@dmt.cyclades Add TASK_SIZE check to do_brk() ChangeSet@1.1148.2.1, 2003-10-02 04:16:50-03:00, marcelo@dmt.cyclades Fix missing part of Centrino cache detection change ChangeSet@1.1149, 2003-10-01 15:59:25-05:00, shaggy@shaggy.austin.ibm.com JFS: Make sure journal records get flushed to disk A performance enhancement that reduced the number of partial-page journal writes resulted in some journal records staying in cache indefinately. We need to make sure that these records are written in a timely fashion. ChangeSet@1.1148, 2003-10-01 14:46:34-03:00, pfg@sgi.com [PATCH] Altix console driver update BitKeeper 2.4 patch for the Altix console driver - Most of this is code clean up-type things.... # This is a BitKeeper generated patch for the following project: # Project Name: Linux kernel tree # This patch format is intended for GNU patch command version 2.5 or higher. # This patch includes the following deltas: # ChangeSet 1.1121 -> 1.1122 # drivers/char/Config.in 1.57 -> 1.58 # drivers/char/sn_serial.c 1.2 -> 1.3 # # The following is the BitKeeper ChangeSet Log # -------------------------------------------- # 03/09/25 pfg@attica.americas.sgi.com 1.1122 # Mods for 2.4 serial console driver - mostly code cleanup # -------------------------------------------- # ChangeSet@1.1147, 2003-10-01 13:47:23-03:00, marcelo@dmt.cyclades Changed EXTRAVERSION to -pre6 TAG: v2.4.23-pre6 ChangeSet@1.1146, 2003-10-01 13:20:54-03:00, atulm@lsil.com [PATCH] Update megaraid driver to 1.18k I want to submit the changes to upgrade the megaraid driver to 1.18k in the stock kernel. Detailed changelog of the changes can be found in the patch itself ChangeSet@1.1145, 2003-10-01 13:17:48-03:00, ak@muc.de [PATCH] Remove IORR manipulation in agpgart nvidia drivers On Tue, Sep 30, 2003 at 11:07:42AM -0600, Bjorn Helgaas wrote: > Hi Marcelo & Andi, > > This change breaks all architectures except i386 and x86_64 (they're the > only ones with asm/msr.h). I believe this is the best fix for this. This just removes all IORR manipulation in the nvidia drivers. The Operating system should never touch the IORRs, they're only for the BIOS. If someone has a buggy BIOS that doesn't set that up they should fix it. Marcelo can you apply this? -Andi ChangeSet@1.1136.2.9, 2003-10-01 12:45:39-03:00, marcelo@dmt.cyclades Dave Jones: Fix cache size of Centrino CPU ChangeSet@1.1063.44.20, 2003-09-30 21:07:39-04:00, len.brown@intel.com [ACPI] deal with lack of acpi prt entries gracefully (Jesse Barnes) ChangeSet@1.1063.44.19, 2003-09-30 20:42:56-04:00, len.brown@intel.com [ACPI] build fix: remove 2nd __exit from asus_acpi.c ChangeSet@1.1120.13.23, 2003-09-30 17:05:39-07:00, bunk@fs.tum.de [PATCH] USB: USB_SERIAL_KEYSPAN_USA49WLC Configure.help entry the trivial patch below adds a Configure.help entry for CONFIG_USB_SERIAL_KEYSPAN_USA49WLC. ChangeSet@1.1120.13.22, 2003-09-30 16:57:43-07:00, baldrick@free.fr [PATCH] USB speedtouch: extra debug messages ChangeSet@1.1120.13.21, 2003-09-30 16:57:27-07:00, baldrick@free.fr [PATCH] USB speedtouch: reduce memory usage Currently, incoming packets are reassembled in a sk_buff which is then sent to the upper layers. The sk_buff needs to be big enough to hold the worst case packet since the size cannot be known in advance. This would not be so bad if the ATM layer paid attention to the mtu (usually 1500 bytes), but for some reason it blithely ignores it and typically passes 9188 bytes as the maximum size. This means that at best 5/6 of the space in every sk_buff is wasted. So instead let's just allocate an assembly buffer (sarb) which gets reused for every packet, and copy each assembled packet into a minimally sized sk_buff before sending. ChangeSet@1.1120.13.20, 2003-09-30 16:57:14-07:00, baldrick@free.fr [PATCH] USB speedtouch: biscuit for Greg 2.4 version. ChangeSet@1.1130.1.1, 2003-09-30 16:26:11-07:00, trini@kernel.crashing.org PPC32: Fixes to the MPC8xx uart driver, from Joakim Tjernlund . ChangeSet@1.1120.13.19, 2003-09-30 15:04:51-07:00, stern@rowland.harvard.edu [PATCH] USB: unusual_devs.h update ChangeSet@1.1140, 2003-09-30 16:59:22-04:00, len.brown@intel.com Merge intel.com:/home/lenb/bk/linux-acpi-test-2.4.22 into intel.com:/home/lenb/bk/linux-acpi-test-2.4.23 ChangeSet@1.1063.44.18, 2003-09-30 16:51:28-04:00, len.brown@intel.com [ACPI] acpi4asus-0.25-0.26 (Karol Kozimor) ChangeSet@1.1063.44.17, 2003-09-30 16:46:40-04:00, len.brown@intel.com [ACPI] acpi4asus-0.24a-0.25-2.4 (Karol Kozimor) ChangeSet@1.1136.3.2, 2003-09-30 05:16:06-07:00, davem@nuts.ninka.net [SPARC64]: Always use sethi+jmpl to reach VISenter{,half}. ChangeSet@1.1136.3.1, 2003-09-30 05:11:09-07:00, mroos@linux.ee [SPARC]: Use NR_CPUS for linux_cpus[]. ChangeSet@1.1136.2.8, 2003-09-30 05:04:49-07:00, dfages@arkoon.net [NET]: Fix HW_FLOWCONTROL on SMP. ChangeSet@1.1136.2.7, 2003-09-30 03:34:18-07:00, davem@nuts.ninka.net [IPV4]: In arp_rcv() do not inspect ARP header until packet length and linearity is verified. ChangeSet@1.1063.44.16, 2003-09-30 00:47:55-04:00, len.brown@intel.com [ACPI] acpi_pci_link_allocate() should stick with irq.active if set. (Andrew de Quincey) Fixes OSDL #1186 "broken USB" and others ChangeSet@1.1063.44.15, 2003-09-29 23:53:54-04:00, len.brown@intel.com [ACPI] CONFIG_ACPI is no longer necessary to enable HT if (CONFIG_ACPI || CONFIG_SMP) CONFIG_ACPI_BOOT=y ChangeSet@1.1063.44.14, 2003-09-29 23:13:17-04:00, len.brown@intel.com [ACPI] ACPI Component Architecture 20030918 (Bob Moore) Found and fixed a longstanding problem with the late execution of the various deferred AML opcodes (such as Operation Regions, Buffer Fields, Buffers, and Packages)... This fixes the "region size computed incorrectly" problem. Fixed several 64-bit issues with prototypes, casting and data types. Removed duplicate prototype from acdisasm.h ChangeSet@1.1136.2.5, 2003-09-29 15:12:41-03:00, ak@muc.de [PATCH] Support 32bit uids on x86-64 This fixes uid > 16bit support on x86-64. Please apply. -Andi Index: linux/arch/x86_64/config.in =================================================================== RCS file: /home/cvs/Repository/linux/arch/x86_64/config.in,v retrieving revision 1.50 ChangeSet@1.1136.2.4, 2003-09-29 15:12:17-03:00, ak@muc.de [PATCH] Use MTRR in vesafb by default on x86-64 With mtrr off vesafb scrolling is incredibly slow on x86-64. That's because it cannot call the video BIOS for hardware assisted scrolling and has to do it all in the CPU. Enable MTRR by default on x86-64. ChangeSet@1.1136.2.3, 2003-09-29 15:12:04-03:00, ak@muc.de [PATCH] Add 3GB personality for x86-64 - Add 3GB personality for x86-64 to personality.h ChangeSet@1.1136.2.2, 2003-09-29 15:11:50-03:00, ak@muc.de [PATCH] /proc/kcore fixes for x86-64 - Make the seek offsets fully 64bit clean. This is needed because x86-64 has negative 64bit addresses in kernel. - Check kern_addr_valid first before copying. -Andi ChangeSet@1.1136.2.1, 2003-09-29 14:58:39-03:00, moilanen@austin.ibm.com [PATCH] Workaround PPC64 PCI scan issue This patch is a port of some work that Anton Blanchard did for 2.6. There are some arch, like PPC64, that need to be able to scan all the PCI functions. The problem comes in on a logically partitioned system where function 0 on a PCI-PCI bridge is assigned to one partition and say function 2 is assiged to another partition. On the second partition, it would appear that function 0 does not exist, but function 2 does. If all the functions are not scanned, everything under function 2 would not be detected. Thanks, Jake ChangeSet@1.1136, 2003-09-29 06:22:57-03:00, marcelo@dhcp-103.cyclades.de Merge bk://kernel.bkbits.net/jgarzik/net-drivers-2.4 into dhcp-103.cyclades.de:/home/marcelo/bk/linux-2.4 ChangeSet@1.1129.1.1, 2003-09-27 14:10:43+10:00, paulus@samba.org Merge bk://source.mvista.com/linux-2.4-mdp into samba.org:/home/paulus/kernel/for-marcelo-ppc ChangeSet@1.1125.1.5, 2003-09-26 08:59:20-07:00, trini@kernel.crashing.org PPC32: Fix KGDB on MPC8xx targets with one serial port. Now console & kgdb can co-exist with something like kdmx. ChangeSet@1.1120.18.5, 2003-09-26 08:08:10-03:00, ak@muc.de [PATCH] AGP Updates for K8 Update the AMD K8 GART driver. With changes from me, Dave Jones, Allen Martin This brings it uptodate for all the new Hammer boards. - Rename it to CONFIG_AGP_AMD_K8 from AGP_AGP_AMD_8151. It is not 8151 specific - Minor fix to Configure.help - Add pci ids for nvidia and VIA chipsets - Cleanup of the core K8 AGP driver. Remove unnecessary code. - Fix bug in AGP probe table. It must be always ordered by vendor This fixes problem with try_unsupported with VIA and SIS chipsets - Add Nforce3 AGP support - Add more VIA AGP PCI IDs - Support try_unsupported command line option for K8 GART driver (command line option because it is normally compiled in on x86-64) - Support agp=off command line option for K8 GART driver too -Andi ChangeSet@1.1120.18.4, 2003-09-26 08:01:03-03:00, ak@muc.de [PATCH] x86-64 merge x86-64 update for 2.4.23pre5. Only touches x86-64 specific files. I also have some updates outside which I'm sending separately. - Minor fixes for boot-options (partly from Mikael P.) - Enable -funit-at-a-time when available - Fix assembly warning - Merge with i386 - Update defconfig - Make LVM 32bit emulation root only - Add compat handler for FIOQSIZE - Support NULL as alias for sys_ioctl in compat handlers (Arnd Bergmann) - Pass __SI_POLL information correctly to signal handlers - Implement emulation for module loading from 32bit userland (Arnd Bergmann, based on sparc64 code) - Add SEMTIMEDOP 32bit emulation - Fix warnings - Pad 32bit statvfs properly for glibc 2.3 compatibility - Scale sys32_siginfo information to avoid overflow (originally from ppc64) - Fix maxcpus=0, pci=noapic, noapic, apic parsing to be early enough. - More fixes for maxcpus=0. It turns off the apic properly now. - Fix iommu=memaper=... parsing - Don't try to parse APICss in MADT when not needed. - Support nolapic as alias to nolocalapic (i386 compatibility) - Print CPU for silent northbridge MCEs - Don't compile in ACPI wakeup PML4 - Merge ACPI fixes from i386 (tolerant SMI parsing) - Add kludge for nforce3/Via K8 chipsets: when Nvidia or VIA pci bridges are detected disable the IO-APIC by default unless "apic" is specified. Note they still need Andrew Quincey's PIC fallback changes in the generic ACPI code, without this they only boot with acpi=off. - Fix memory corruption in sys_ioperm - Support panic on nmi watchdog timeout (nmi_watchdog=panic) - Fix pci=noacpi path - Fix cpuinfo for L2 cache associability (Richard Brunner) - Check nmi watchdog on SMP too - Fix non vsyscall sys_time: must write a long to user space, not an int - Print less stack on oopses - Cleanup oops printing: unify code between page_fault and other exceptions, sti before calling do_exit, add executive summary at the end - Export touch_nmi_watchdog - Make int_ret_from_sys_call export NOVERS - Remove unnecessary prefetches - Disable fix alignment code in copy_user_generic - Don't use NTI stores in clear_user - Prefetch errata workaround for Opteron (based on code by Richard Brunner) - Make NUMA topology discovery less noisy - Fix RBP define in ptrace.h/calling.h (Bryan O'Sullivan) - Support 3GB personality for 32bit programs -Andi ChangeSet@1.1120.18.3, 2003-09-26 07:49:17-03:00, mike.miller@hp.com [PATCH] cciss support more than 8 controllers The following patch enables support for more than 8 controllers in the cciss driver. It was tested against the 2.4.23-pre4 kernel tree. Changes: Enables support for more than 8 controllers by passing 0 as a major number and allowing the OS the to dynamically assign a major number. Primary reasons for this are to support the 6402 & 6404 controllers and to avoid preallocating any more major numbers than we already have. Please consider this for inclusion. Thanks, mikem ChangeSet@1.1120.18.2, 2003-09-26 05:45:41-03:00, johnstul@us.ibm.com [PATCH] Fix boot code overflow with more CPUs than CONFIG_NR_CPUS On Tue, 2003-09-23 at 17:59, john stultz wrote: > Marcelo, all, > This is a backported patch from 2.5 that fixes array overflows and > memory corruption when booting on boxes with more processors then > CONFIG_NR_CPUS. > Marcelo, All, James Cleverdon pointed out that I had a printk formatting typo in the last revision of this patch ( "0x%d" - doh!). Anyway, this revision corrects the error, changing 0x%d -> 0x%x. thanks -john ChangeSet@1.1120.12.3, 2003-09-26 05:39:38-03:00, axboe@suse.de [PATCH] cdrom memory leak Marcelo, Please apply this one to fix a memory leak in the cdrom uniform layer. Patch from Szombathelyi György. ===== drivers/cdrom/cdrom.c 1.13 vs edited ===== ChangeSet@1.1120.17.1, 2003-09-26 05:03:32-03:00, ink@jurassic.park.msu.ru [PATCH] Alpha update - New build target: compressed bootp image (Jay Estabrook) - tweaked LEGACY_START_ADDRESS config option to reduce the size of the generic kernel and solve some other problems (Jay Estabrook); - allow more pte pages to map large firmware images on newer systems (Jeff Wiedemeier); - fix multi-line asm in xor.h that causes compile failure with gcc-3.3 (from 2.6; originally by Thorsten Kranzkowski); - correct memory freeing after PCI init on Nautilus; - fix UAC shift in per-thread flags. Ivan. ChangeSet@1.1120.15.4, 2003-09-25 22:47:30-05:00, galak@blarg.somerset.sps.mot.com PPC32: Simplified handling of big/little endian pci indirect access ChangeSet@1.1120.11.16, 2003-09-25 19:59:25-04:00, mlindner@syskonnect.de [PATCH] sk98lin-2.4: Readme version update here is a new version of the sk98lin driver (v6.18) with some changes for kernel 2.4.22. This is a diff between the latest patched kernel version with sk98lin driver v6.17. Patch 2/3 * Update the version number in the readme file ChangeSet@1.1120.11.15, 2003-09-25 19:59:22-04:00, mlindner@syskonnect.de [PATCH] sk98lin-2.4: Remove useless configure options here is a new version of the sk98lin driver (v6.18) with some changes for kernel 2.4.22. This is a diff between the latest patched kernel version with sk98lin driver v6.17. Patch 1/3 * Remove bogus config stuff ChangeSet@1.1120.11.14, 2003-09-25 19:59:18-04:00, mporter@kernel.crashing.org [PATCH] 2.4 IBM EMAC updates This update has a couple bug fixes, cleanup, and partial support for a new revision of EMAC controller. # -------------------------------------------- # 03/09/24 mporter@kernel.crashing.org 1.1107 # Cleanup EMAC initialization sequence. # Add EMAC4/440GX support. # Fix bug in RCBS initialization. # Fix bug where TX processing is enabled before driver is opened. # -------------------------------------------- ChangeSet@1.1120.13.18, 2003-09-25 16:00:03-07:00, bunk@fs.tum.de [PATCH] USB: fix USB_MOUSE help text The USB_MOUSE help text was obviously copied from the USB_KBD help text. The patch below does a s/keyboard/mouse/g ChangeSet@1.1120.13.17, 2003-09-25 15:59:50-07:00, amn3s1a@ono.com [PATCH] USB: New unusual_devs.h entry (Minolta DiMAGE E223 Digital Camera) ChangeSet@1.1120.15.3, 2003-09-25 17:54:29-05:00, galak@blarg.somerset.sps.mot.com PPC32: Added big-endian cfg_addr access ChangeSet@1.1063.44.13, 2003-09-25 16:16:51-04:00, len.brown@intel.com [ACPI] For ThinkPad -- carry on in face of ECDT probe failure (Andi Kleen) ChangeSet@1.1125.1.4, 2003-09-25 11:27:11-07:00, trini@kernel.crashing.org PPC32: Update MPC8xx code so that it uses consistent_alloc. From Dan Malek. ChangeSet@1.1120.15.2, 2003-09-25 01:43:16-05:00, galak@blarg.somerset.sps.mot.com PPC32: Added "user64" versions of the user acess functions that allow modification of 64-bit data. ChangeSet@1.1120.14.2, 2003-09-24 17:01:21-07:00, mporter@kernel.crashing.org PPC32: Use CONFIG_PTE_64BIT instead of CONFIG_44x where appropriate ChangeSet@1.1120.14.1, 2003-09-24 16:28:07-07:00, mporter@kernel.crashing.org PPC32: Fix 44x _PMD_PRESENT bug ChangeSet@1.1120.13.16, 2003-09-24 14:32:19-07:00, green@linuxhacker.ru [PATCH] USB: devio.c memleak on unexpected disconnect There is a memleak in devio.c (User space communication with USB devices) recently added, it forgets to free the buffer if device was disconnected. The patch is trivial, please apply. Found with help of smatch. ChangeSet@1.1120.13.15, 2003-09-24 14:27:13-07:00, abbotti@mev.co.uk [PATCH] USB: ftdi_sio - new vid/pid for OCT US101 USB to RS-232 converter Here are a couple of patches against a copy of Greg's usb-2.4 and usb-2.5 bk trees to add a VID/PID for Omnidirectional Control Technology's US101 USB to RS-232 converter. This has also been rebadged by Dick Smith Electronics (New Zealand) as a XH6361 USB to serial converter. Thanks to Donald Gordon for the info, which I have verified by checking the Windows INF files. ChangeSet@1.1120.13.14, 2003-09-24 13:15:45-07:00, greg@kroah.com [PATCH] USB: fix up two locking issues in mdc800 and vicam drivers These were introduced by my previous copy_to_user audit and noticed by Sergey Vlasov ChangeSet@1.1120.13.13, 2003-09-24 13:12:02-07:00, greg@kroah.com [PATCH] USB: port ipaq fix to 2.4 ChangeSet@1.1120.13.12, 2003-09-24 13:09:10-07:00, dhollis@davehollis.com [PATCH] USB: ax8817x support for usbnet and ethtool_ops support This is a cleaned up and functioning patch from earlier that adds ASIX ax8817x support to the 2.4 usbnet.c driver. It also adds ethtool_ops support to usbnet so that specific drivers can easily extend or override ethtool options as necessary. This addition essentially obsoletes the current ax8817x.c driver. ChangeSet@1.1120.13.11, 2003-09-24 13:04:22-07:00, stern@rowland.harvard.edu [PATCH] USB: Pad UFI commands to 12 bytes with zeros This patch from Alan Stern zero-pads commands using the UFI protocol. This patch exists in 2.5, but never made it into 2.4 Not all devices require this padding, but some do. ChangeSet@1.1120.13.10, 2003-09-24 13:04:15-07:00, baldrick@free.fr [PATCH] USB speedtouch: bump the version number ChangeSet@1.1120.13.9, 2003-09-24 13:04:07-07:00, baldrick@free.fr [PATCH] USB: New email address ChangeSet@1.1120.13.8, 2003-09-24 13:03:57-07:00, baldrick@free.fr [PATCH] USB speedtouch: neater sanity check Neater sanity check. ChangeSet@1.1120.13.6, 2003-09-24 11:56:01-07:00, david-b@pacbell.net [PATCH] USB: usb gadget support for 2.4 (5/5): kbuild/kconf This last part just adds kbuild support for the three drivers sent in the earlier patches. (There's no "mid-layer".) ChangeSet@1.1120.13.5, 2003-09-24 11:55:54-07:00, david-b@pacbell.net [PATCH] USB: usb gadget support for 2.4 (4/5): ether This gadget driver provides Ethernet-style networking over USB. When used with more functional hardware (such as net2280), it implements the CDC Ethernet class, which is widely supported by non-Microsoft operating systems. (And since some cable modems talk CDC Ethernet, it can also talk to Windows; just not without installing a driver.) With hardware that has functional restrictions (like pxa2xx), it implements a "good parts" subset of the CDC spec, but doesn't advertise itself as conformant (subsets aren't). ChangeSet@1.1120.13.4, 2003-09-24 11:55:46-07:00, david-b@pacbell.net [PATCH] USB: usb gadget support for 2.4 (3/5): zero "Gadget Zero" is useful for testing the usb stack in a given device, by giving control and bulk/interrupt transfers a good workout. If "g_zero" passes stress tests, then the driver for the underlying hardware is probably in good enough shape to handle more challenging tasks. Oh -- this also includes "usbstring.c", which is a handy utility for turning C strings (assumed to be ISO 8859-1) into the Unicode descriptors specified by USB. That's used by other drivers too. ChangeSet@1.1120.13.3, 2003-09-24 11:55:38-07:00, david-b@pacbell.net [PATCH] USB: usb gadget support for 2.4 (2/5): net2280 This implements the "gadget" API for use with the "NET2280 PCI USB 2.0 High Speed Peripheral Controller". A USB "gadget" embedding Linux will have a driver for the hardware (like this one) plus a "gadget driver" using it to implement the chosen functionality. ChangeSet@1.1120.13.2, 2003-09-24 11:55:28-07:00, david-b@pacbell.net [PATCH] USB: usb gadget support for 2.4 (1/5): api This is the two header files used in the API. Note that on 2.4 drivers can only #include one of and . On 2.6 the APIs can be used together. ChangeSet@1.1125.1.3, 2003-09-24 11:16:01-07:00, trini@kernel.crashing.org PPC32: Add a potential bugfix to the MPC8xx uart driver, by way of Dan Malek. ChangeSet@1.1125.1.2, 2003-09-24 11:10:26-07:00, trini@kernel.crashing.org PPC32: UART support and configuration updates from Gary Thomas. ChangeSet@1.1120.13.1, 2003-09-24 11:00:44-07:00, greg@kroah.com [PATCH] USB: add a new pl2303 device id. Thanks to Stefan Verkoyen for the information. ChangeSet@1.1120.11.13, 2003-09-24 12:23:10-04:00, mpm@selenic.com [PATCH] netif_carrier_* support for tlan Add proper netif_carrier_* to tlan driver l-mpm/drivers/net/tlan.c | 7 ++++++- 1 files changed, 6 insertions(+), 1 deletion(-) ChangeSet@1.1125.1.1, 2003-09-24 09:14:18-07:00, trini@kernel.crashing.org PPC32: Fix a multiple definition problem in the bootwrapper. ChangeSet@1.1120.9.4, 2003-09-23 17:56:21-07:00, davem@nuts.ninka.net [NET]: Unlink qdiscs in qdisc_destroy even when CONFIG_NET_SCHED is not enabled. ChangeSet@1.1120.12.1, 2003-09-23 16:53:17-07:00, lm@bitmover.com Add pre-apply.paranoid trigger from the Linux 2.5 tree. ChangeSet@1.1120.11.12, 2003-09-23 19:14:32-04:00, jgarzik@redhat.com [wireless airo] Fix build Fix suggested someone on lkml whose name I forgot (I apologize) ChangeSet@1.1120.11.11, 2003-09-23 19:01:10-04:00, scott.feldman@intel.com [e1000] 82544 PCI-X hang fix + TSO updates * Bug fix: 82544 hang with PCI-X: if outgoing Tx buffers terminate within evenly-aligned dwords, and the device is sharing the bus segment with another PCI-X device, 82544 can hang the bus on a split-completion transaction. Fix is to split buffer into two buffers with the first one not terminating within evenly-aligned dword address, and the second on being 4-bytes, which goes as a non-split-completion PCI-X transaction. * 8254x controller that support TSO do an internal calculation to make sure there is enough FIFO space to handle the overhead of each TSO segment before DMA'ing TSO data from host memory. The internal calculation is dependent on the mss of the TSO (defines the number of segments), but the reserved space is a constant, so we need to adjust the maximum size of each buffer queued to the hardware to hold the equation and not overrun the FIFO. This is per TSO because the mss can change from one send to the next. ChangeSet@1.1120.11.10, 2003-09-23 19:01:06-04:00, scott.feldman@intel.com [e1000] misc whitespace cleanup, changelog * misc whitespace cleanup, changelog ChangeSet@1.1120.11.9, 2003-09-23 19:01:03-04:00, scott.feldman@intel.com [e1000] new 82541/5/6/7 hardware support * Added 82545 (rev3), 82546 (rev3), and 82541/7 (rev 2) support - new device IDs - internal SERDES support for 82545/6 (rev 3) - don't apply MMRBC workaround for 82545/6 (rev 3) - don't use IO mapping for reset for 82545/6 (rev 3) ChangeSet@1.1120.11.8, 2003-09-23 19:00:59-04:00, scott.feldman@intel.com [e1000] read correct bit from EEPROM for getting WoL settings * Bug fix: read the correct bit from the EEPROM that controls the initial setting for WoL after a reset. ChangeSet@1.1120.11.7, 2003-09-23 19:00:55-04:00, scott.feldman@intel.com [e1000] sync w/ 2.6 e1000 driver * Sync from 2.6 driver for APIs that have moved back to 2.4. * Sync from 2.6 driver ethtool SG/CSUM/TSO support. ChangeSet@1.1120.11.6, 2003-09-23 19:00:52-04:00, scott.feldman@intel.com [e1000] make function out of setting media type * Consolidate code and make function out of setting media type. ChangeSet@1.1120.11.5, 2003-09-23 19:00:48-04:00, scott.feldman@intel.com [e1000] Turn off ASF support on Fiber nics * Turn off ASF support on fiber nics. Wasn't tested and isn't known to work, so disable before someone hurts themselves. ChangeSet@1.1120.11.4, 2003-09-23 19:00:45-04:00, scott.feldman@intel.com [e1000] move static to table from .h to .c * Move static table from hw.h to hw.c to avoid creating a copy of table everytime hw.h is included in a .c. ChangeSet@1.1120.11.3, 2003-09-23 19:00:41-04:00, scott.feldman@intel.com [e1000] add ethtool flow control support * Add ethtool flow control support. ChangeSet@1.1120.11.2, 2003-09-23 19:00:38-04:00, scott.feldman@intel.com [e1000] Add PHY master/slave #define override * Add PHY master/slave #define override to address link issues with 82541/7 (rev2) against some low-end switches. Forcing master will improve the time-to-link against these switches. ChangeSet@1.1120.11.1, 2003-09-23 19:00:34-04:00, scott.feldman@intel.com [e1000] cleanup error return codes * Cleanup error return code propagation and eliminate redundant DEBUGOUT statements. ChangeSet@1.1120.9.3, 2003-09-23 12:54:04-07:00, laforge@netfilter.org [NETFILTER]: In ipt_REJECT handle various hooks correctly in route_reverse(). ChangeSet@1.1120.10.1, 2003-09-23 14:52:55-05:00, Matt_Domsch@dell.com EDD: BIOS Enhanced Disk Drive Services 3.0 support This code performs BIOS EDD 3.0 calls during kernel real mode initialization, saves away the results, and exports the results through /proc/bios/edd/. This code has been in 2.4.x-ac since January 2003 with no ill reports. ChangeSet@1.1120.9.2, 2003-09-23 04:08:25-07:00, chas@cmf.nrl.navy.mil [ATM]: Update link in documentation. ChangeSet@1.1120.9.1, 2003-09-23 04:06:23-07:00, chas@cmf.nrl.navy.mil [ATM]: [he] Possibly using corrupted structure (from felipewd@terra.com.br) ChangeSet@1.1127, 2003-09-23 09:58:43+10:00, paulus@samba.org Merge ChangeSet@1.1120.1.13, 2003-09-22 12:05:23-07:00, trini@kernel.crashing.org PPC32: Fix dependancies on uImage. ChangeSet@1.1120.1.12, 2003-09-22 11:03:48-07:00, trini@kernel.crashing.org PPC32: Add a 'uImage' target for U-Boot. ChangeSet@1.1125, 2003-09-22 10:05:49-07:00, trini@kernel.crashing.org PPC32: Misc changes for the D-Box2. From Andreas Oberritter . ChangeSet@1.1124, 2003-09-22 09:57:50-07:00, trini@kernel.crashing.org PPC32: Add the D-Box2 MPC8xx board. From Andreas Oberritter . ChangeSet@1.1123, 2003-09-22 09:35:29-07:00, trini@kernel.crashing.org PPC32: Make ISA support a question on CONFIG_ALL_PPC. ChangeSet@1.1120.1.10, 2003-09-22 06:34:48-03:00, viro@parcelfarce.linux.theplanet.co.uk [PATCH] Convert /proc//maps to seqfile Patch below is a backport of 2.6 implementation of /prc//maps. It fixes the crap with lseek() behaviour and it's cleaner than current code. Basically, it's a switch to seq_file. ChangeSet@1.1120.1.8, 2003-09-22 06:12:00-03:00, marcelo@logos.cnet Removed unused label in page_alloc.c ChangeSet@1.1120.5.12, 2003-09-21 18:26:03-07:00, laforge@netfilter.org [NETFILTER]: Fix ipt_REJECT when used in OUTPUT. ChangeSet@1.1120.5.11, 2003-09-21 18:25:10-07:00, willy@debian.org [NETFILTER]: Use net/checksum.h instead of asm/checksum.h ChangeSet@1.1120.5.10, 2003-09-21 18:00:36-07:00, davem@nuts.ninka.net [SCTP]: Include linux/crypto.h as needed. ChangeSet@1.1120.1.7, 2003-09-21 15:45:56-03:00, marcelo@logos.cnet Merge bk://kernel.bkbits.net/jgarzik/net-drivers-2.4 into logos.cnet:/home/marcelo/bk/linux-2.4 TAG: v2.4.23-pre5 ChangeSet@1.1120.1.5, 2003-09-21 15:43:08-03:00, marcelo@logos.cnet Fix Andrea VM merge error ChangeSet@1.1120.1.4, 2003-09-21 14:18:14-03:00, mkp@mkp.net [PATCH] forte sound driver update 2003-09-18 Martin K. Petersen * forte.c: Use C99 initializers. (forte_channel_init): Set up frag size before calling reset. (forte_dsp_ioctl): Proper copy_to_user() return values. (forte_channel_free): Fix mem_map_unreserve() bug. ChangeSet@1.1120.7.1, 2003-09-20 20:43:29-04:00, jgarzik@redhat.com [sound i810_audio] sync with 2.5 * s/spaces/tabs/ * speling fixes * remove need for wrapper.h * clean up includes * C99 initializers * remove __initdata (incorrect, and 2.5 prefers to __devinitdata) * irqreturn_t compat * cpu_relax() in one loop * no need to check pci_present(), PCI layer does this for us anyway ChangeSet@1.1120.6.5, 2003-09-20 14:47:26-04:00, amir.noam@intel.com [PATCH] [bonding] Convert /proc to seq_file This patch converts /proc/net/bondX/info into /proc/net/bonding/bondX using the seq_file interface. This is based on Stephen's recent patch, but slightly modified to work with the propagation patch set and with some locking changes to make it simpler. The patch applies both on 2.4 (after the sync set from earlier today) and on 2.6 (after the propagation set from 2003/11/9). Amir ChangeSet@1.1120.6.4, 2003-09-20 14:19:45-04:00, amir.noam@intel.com [PATCH] [bonding 2.4] Backport PDE() ChangeSet@1.1120.6.3, 2003-09-20 14:19:42-04:00, amir.noam@intel.com [PATCH] [bonding 2.4] Backport free_netdev() ChangeSet@1.1120.6.2, 2003-09-20 14:19:38-04:00, amir.noam@intel.com [PATCH] [bonding 2.4] Get rid of MOD_*_USE_COUNT, and use C99 initializers ChangeSet@1.1120.6.1, 2003-09-20 13:43:15-04:00, achirica@telefonica.net [wireless airo] Fix MIC support using CryptoAPI ChangeSet@1.1120.5.9, 2003-09-20 02:58:25-07:00, davem@nuts.ninka.net [SCTP]: Do not redefine SMTP stat inc macros. ChangeSet@1.1120.5.8, 2003-09-20 01:27:55-07:00, krkumar@us.ibm.com [IPV6]: Export ipv6_devconf via netlink. ChangeSet@1.1120.5.7, 2003-09-20 01:24:22-07:00, chas@cmf.nrl.navy.mil [ATM]: Implement br2684_ioctl_hook for br2684. ChangeSet@1.1120.5.6, 2003-09-20 01:23:49-07:00, chas@cmf.nrl.navy.mil [ATM]: Implement pppoatm_ioctl_hook for pppoatm. ChangeSet@1.1120.5.5, 2003-09-20 01:23:13-07:00, chas@cmf.nrl.navy.mil [ATM]: Cleanup atm_dev_ioctl a bit (from mitch@sfgoth.com). ChangeSet@1.1120.5.4, 2003-09-20 01:22:30-07:00, chas@cmf.nrl.navy.mil [ATM]: Split atm_ioctl into vcc_ioctl and atm_dev_ioctl. ChangeSet@1.1120.5.3, 2003-09-20 01:09:34-07:00, davem@nuts.ninka.net [IPV4]: Fix route leak in igmp.c ChangeSet@1.1120.5.1, 2003-09-20 00:39:15-07:00, davem@nuts.ninka.net Merge http://linux-lksctp.bkbits.net:8080/lksctp-2.4 into nuts.ninka.net:/disk1/davem/BK/net-2.4 ChangeSet@1.1120.4.1, 2003-09-19 17:29:43-05:00, galak@blarg.somerset.sps.mot.com Added "user64" versions of the user access functions that allow modification of 64-bit data. ChangeSet@1.1120.3.12, 2003-09-19 15:21:38-04:00, lethal@unusual.internal.linux-sh.org sh64: Fix CONFIG_SH64_USER_MISALIGNED_FIXUP compile error. Patch from Richard Curnow. arch/sh64/kernel/traps.c | 6 ++++-- 1 files changed, 4 insertions(+), 2 deletions(-) ChangeSet@1.1120.3.11, 2003-09-19 15:17:21-04:00, lethal@unusual.internal.linux-sh.org sh: Interim cache coherency fix for 2-way caches. ChangeSet@1.1120.3.10, 2003-09-19 15:02:43-04:00, lethal@unusual.internal.linux-sh.org sh64: Fixup unaligned accesses Add support for fixing up loads and stores to misaligned addresses occurring within the kernel, and optionally in user mode as well. Patch from Richard Curnow. arch/sh64/config.in | 2 arch/sh64/defconfig | 1 arch/sh64/kernel/fpu.c | 145 ++++----- arch/sh64/kernel/traps.c | 684 ++++++++++++++++++++++++++++++++++++++++++- include/asm-sh64/processor.h | 8 include/asm-sh64/registers.h | 1 6 files changed, 758 insertions(+), 83 deletions(-) ChangeSet@1.1120.3.9, 2003-09-19 14:29:36-04:00, lethal@unusual.internal.linux-sh.org sh: cleanup microdev merge. ChangeSet@1.1120.3.8, 2003-09-19 14:05:55-04:00, lethal@unusual.internal.linux-sh.org Merge ChangeSet@1.1120.3.7, 2003-09-19 13:56:26-04:00, lethal@unusual.internal.linux-sh.org [sh64] Add a new configure option + support code to provide a /proc/asids file. Patch from Richard Curnow: arch/sh64/config.in | 1 + arch/sh64/defconfig | 1 + arch/sh64/kernel/process.c | 45 +++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 47 insertions(+) ChangeSet@1.1120.1.3, 2003-09-19 10:23:46-07:00, len.brown@intel.com Merge intel.com:/home/lenb/bk/linux-acpi-test-2.4.22 into intel.com:/home/lenb/bk/linux-acpi-test-2.4.23 ChangeSet@1.1063.44.12, 2003-09-19 10:21:02-07:00, len.brown@intel.com [ACPI] fix IO-APIC mode SCI storm due to sharing with PCI device (David Shaohua Li) http://bugzilla.kernel.org/show_bug.cgi?id=1165 ChangeSet@1.1120.3.4, 2003-09-19 13:00:01-04:00, lethal@unusual.internal.linux-sh.org sh: sh-sci updates This includes several updates for the sh-sci driver. From TAKANO Nobuyuki: The patch bellow is given by TAKANO nobuyuki on linux-sh-ja list with the problem reports. The former hunk fixes wrong masks for SH3's SCPCR and the latter two removes a harmful code which handles TTY_HW_COOK_IN/OUT flags. TTY_HW_COOK_IN/OUT flags are for the intelligent serial processor but not for a chip like SCI. Without the latter part of patch, the behavior of sh-sci driver is worng when using the canonical input mode. and from David McCullough: * 230400 support for platforms where it is possible * breaks cause interrupt lockup where the driver never removed an RX interrupt condition * if the kernel was too slow and the flip buffer filled it would also lockup for the same reason as above * added DCD/DTR support for SnapGear platforms drivers/char/sh-sci.c | 226 ++++++++++++++++++++++++++++++++++++++------------ drivers/char/sh-sci.h | 25 ----- 2 files changed, 174 insertions(+), 77 deletions(-) ChangeSet@1.1120.3.3, 2003-09-19 12:50:11-04:00, lethal@unusual.internal.linux-sh.org sh: Add bzImage support Patch from SUGIOKA Toshinobu to add bzImage support for sh (backport from 2.6) arch/sh/Makefile | 4 +-- arch/sh/boot/compressed/Makefile | 4 +-- arch/sh/boot/compressed/head.S | 50 ++++++++++++++++++++++++++++++++++++++- arch/sh/config.in | 3 ++ 4 files changed, 56 insertions(+), 5 deletions(-) ChangeSet@1.1120.3.2, 2003-09-19 12:46:08-04:00, lethal@unusual.internal.linux-sh.org sh: div64 backport and random cleanups Backport div64 changes from the 2.6 tree, also clean up some other random things. Patches from SUGIOKA Toshinobu. arch/sh/kernel/sh_ksyms.c | 6 ++++++ arch/sh/lib/Makefile | 2 +- arch/sh/lib/div64.S | 46 ++++++++++++++++++++++++++++++++++++++++++++++ arch/sh/lib/udivdi3.c | 16 ++++++++++++++++ include/asm-sh/checksum.h | 2 -- include/asm-sh/delay.h | 1 + include/asm-sh/div64.h | 16 +++++++++++++--- 7 files changed, 83 insertions(+), 6 deletions(-) ChangeSet@1.1120.3.1, 2003-09-19 12:19:02-04:00, lethal@unusual.internal.linux-sh.org sh: shmse updates arch/sh/kernel/io_shmse.c: Add external irq setups for SH-Mobile Solution Engine arch/sh/kernel/setup_shmse.c: Add support for MSTLANEX01. Patches from YOSHII Takashi. ChangeSet@1.1109.15.3, 2003-09-19 13:20:42+02:00, damien.morange@hp.com [SCTP] LKSCTP 0.6.9 backport on kernel 2.4 patch #3 This patch consists only in new files. It adds the SCTP protocol engine and the SCTP sockets layers to the IP stack. This patchs requires patches #1 and #2 to be applied. ChangeSet@1.1109.15.2, 2003-09-19 13:07:32+02:00, damien.morange@hp.com [SCTP] LKSCTP 0.6.9 backport on kernel 2.4 patch #2 This patch modifies the kernel existing files to add the SCTP functionality. It cannot work without the 3rd patch of this serie, but it is issued to better highlight the required changes. ChangeSet@1.1109.15.1, 2003-09-19 13:01:11+02:00, damien.morange@hp.com [SCTP] LKSCTP 0.6.9 backport on kernel 2.4 patch #1 Adding the necessary changes to the kernel IP stack so that SCTP can perform Path MTU discovery. No other SCTP-specific changes. The resulting kernel is functionally unchanged. ChangeSet@1.1109.1.12, 2003-09-18 20:16:59-07:00, len.brown@intel.com Merge intel.com:/home/lenb/bk/linux-acpi-test-2.4.22 into intel.com:/home/lenb/bk/linux-acpi-test-2.4.23 ChangeSet@1.1063.44.11, 2003-09-18 20:14:07-07:00, len.brown@intel.com [ACPI] acpi_disabled is used after __initdata is freed. fixes random battery module loading problem -- SuSE bug #30477 (Andi Kleen) ChangeSet@1.1109.14.3, 2003-09-18 14:32:00-07:00, greg@kroah.com [PATCH] USB: unusual device fixup for the Y-E floppy drive. ChangeSet@1.1109.1.10, 2003-09-18 10:18:37-07:00, len.brown@intel.com Merge intel.com:/home/lenb/bk/linux-acpi-test-2.4.22 into intel.com:/home/lenb/bk/linux-acpi-test-2.4.23 ChangeSet@1.1063.44.10, 2003-09-18 10:14:52-07:00, len.brown@intel.com [ACPI] Fix SCI storm on out of spec boards like Tyan http://bugzilla.kernel.org/show_bug.cgi?id=774 ChangeSet@1.1109.14.2, 2003-09-18 09:56:26-03:00, andrea@suse.de [PATCH] Remove unused code from balance_classzone() (this had more prio than the waterkmark fixes, they will came next) According to the kernel CVS you didn't merge this yet, so please merge the below too, it will remove a not necessary branch that also generates a gcc false positive (all harmless of course but it's more correct to remove it): ChangeSet@1.1109.14.1, 2003-09-18 09:55:39-03:00, andrea@suse.de [PATCH] Fix nr_free_buffer_pages() Hi Marcelo, this patch corrects two bugs in nr_free_buffer_pages (they weren't fatal but it's better to get them right of course ;) http://www.us.kernel.org/pub/linux/kernel/people/andrea/patches/v2.4/2.4.23pre4/nr_free_buffer_pages-fixes-1 the first bug is the zonep++ before dereferencing it, the second bug is the free variable that has to be unsigned to handle empty zones correctly (thanks to Bjorn Helgaas for noticing the latter sign issue). ChangeSet@1.1119, 2003-09-18 15:17:58+10:00, paulus@samba.org PPC32: Change the ucontext to move the sigmask back where it was. Previous changes to the struct ucontext added space for the registers and also moved the sigmask field. This moves the sigmask field back to where it was before and thus restores binary compatibility. Userspace should continue to access the register values through the uc_regs pointer as before. ChangeSet@1.1109.13.2, 2003-09-17 17:46:46-04:00, len.brown@intel.com [ACPI] delete acpitable.[ch], which used to be just for CONFIG_ACPI_HT_ONLY. Now CONFIG_ACPI_HT_ONLY uses the analogous sub-set of the full ACPI driver. ChangeSet@1.1063.44.9, 2003-09-17 17:17:14-04:00, len.brown@intel.com exclude acpitable.[ch] from the CONFIG_ACPI_HT_ONLY build ChangeSet@1.1109.1.8, 2003-09-17 17:29:34-03:00, marcelo@logos.cnet Changed EXTRAVERSION to -pre5 ChangeSet@1.1109.8.12, 2003-09-17 16:59:37-03:00, bjorn.helgaas@hp.com [PATCH] IA64 ZX1 gart support This updates the GART support for the IA64 HP ZX1 chipset. As with the 460GX, there's more infrastructure that this depends on that will remain in the ia64 patch. ChangeSet@1.1109.8.11, 2003-09-17 16:57:53-03:00, bjorn.helgaas@hp.com [PATCH] IA64 460GX gart support This patch adds the bulk of the GART support for the IA64 460GX chipset. This isn't everything needed; there's some infrastructure stuff that probably won't be acceptable in 2.4, so I'll keep that in the ia64 patch for now. # This is a BitKeeper generated patch for the following project: # Project Name: Linux kernel tree # This patch format is intended for GNU patch command version 2.5 or higher. # This patch includes the following deltas: # ChangeSet 1.1110 -> 1.1111 # drivers/char/agp/agpgart_be.c 1.46 -> 1.47 # drivers/char/Config.in 1.55 -> 1.56 # drivers/char/agp/agp.h 1.22 -> 1.23 # include/linux/agp_backend.h 1.19 -> 1.20 # drivers/char/drm-4.0/agpsupport.c 1.5 -> 1.6 # # The following is the BitKeeper ChangeSet Log # -------------------------------------------- # 03/09/10 bjorn.helgaas@hp.com 1.1111 # 460GX gart support. # -------------------------------------------- # ChangeSet@1.1109.8.10, 2003-09-17 16:57:00-03:00, bjorn.helgaas@hp.com [PATCH] IA64 config help update This updates some IA64-specific config help text to match what's in 2.5. ===== Documentation/Configure.help 1.192 vs edited ===== ChangeSet@1.1109.12.1, 2003-09-17 12:24:05-06:00, bjorn.helgaas@hp.com ia64: Add arch/ia64/drivers subdir, so ski drivers can be under arch/ia64 while still getting their module_inits called late (i.e., simscsi module_init needs to happen after init_scsi). ChangeSet@1.1109.8.9, 2003-09-17 15:10:50-03:00, marcelo@logos.cnet Fix ide-scsi initialization lockup (kudos to Alan) ChangeSet@1.1109.8.8, 2003-09-17 14:43:31-03:00, Stuart_Hayes@dell.com [PATCH] Fix ide-tape lock up Resending patch... the text got mangled on the last message. Thanks Stuart Stuart_Hayes@Dell.com ChangeSet@1.1063.44.8, 2003-09-17 13:00:53-04:00, len.brown@intel.com from 2.6 acpi_pci_link_get_irq() returns 0 on error, not -ENODEV. (Christophe Saout) ChangeSet@1.1063.44.7, 2003-09-17 11:48:55-04:00, len.brown@intel.com tables.c.diff ChangeSet@1.1118, 2003-09-17 08:11:14-07:00, trini@kernel.crashing.org PPC32: Fix a rounding error in the bootwrapper udelay. From Milton Miller. ChangeSet@1.1117, 2003-09-17 21:02:17+10:00, paulus@samba.org PPC32: Use nap mode in the idle loop on the PPC970. ChangeSet@1.1109.1.6, 2003-09-16 11:15:54-07:00, len.brown@intel.com Merge intel.com:/home/lenb/bk/linux-acpi-test-2.4.23 into intel.com:/home/lenb/bk/linux-acpi-release-2.4.23 ChangeSet@1.1063.44.6, 2003-09-16 10:26:26-07:00, len.brown@intel.com ACPI_CA_VERSION 0x20030916 ChangeSet@1.1109.9.2, 2003-09-16 09:52:15-07:00, len.brown@intel.com Merge intel.com:/home/lenb/bk/linux-2.4.23 into intel.com:/home/lenb/bk/linux-acpi-release-2.4.23 ChangeSet@1.1109.1.4, 2003-09-16 09:51:24-07:00, len.brown@intel.com Merge intel.com:/home/lenb/bk/linux-2.4.23 into intel.com:/home/lenb/bk/linux-acpi-test-2.4.23 ChangeSet@1.1109.8.7, 2003-09-16 12:04:47-03:00, daniel@deadlock.et.tudelft.nl [PATCH] Patch: Fix reported Atyfb problem on Sparc Hello, The patch below fixes the problems that were reported on the Sparc architecture with the Atyfb. Greetings, Daniël ChangeSet@1.1109.8.6, 2003-09-16 10:28:56-03:00, marcelo@logos.cnet Liam Girdwood: eliminates pop noises when doing a PM suspend/resume with the WM9712 AC97 codec. ChangeSet@1.1116, 2003-09-16 14:54:05+10:00, paulus@samba.org PPC32: Fix compile for "Walnut" board. Patch from David Gibson. ChangeSet@1.1109.8.5, 2003-09-15 22:01:42-03:00, marcelo@logos.cnet Merge bk://namesys.com/bk/reiser3-linux-2.4-help into logos.cnet:/home/marcelo/bk/linux-2.4 ChangeSet@1.1109.11.3, 2003-09-15 16:30:16-07:00, chas@cmf.nrl.navy.mil [ATM]: [clip] Fix race between modifying entry->vccs and clip_start_xmit(). ChangeSet@1.1109.11.2, 2003-09-15 16:24:02-07:00, hadi@cyberus.ca [NET]: Make pfifo_fast actually report statistics. ChangeSet@1.1109.11.1, 2003-09-15 16:12:31-07:00, davem@nuts.ninka.net [NET]: Increase ethernet tx_queue_len to 1000. ChangeSet@1.1109.10.10, 2003-09-15 18:36:13-03:00, paulus@samba.org [PATCH] Fix IDE compile on PPC in 2.4.23-pre4 Marcelo & Bart, Currently, the IDE code in 2.4.23-pre4 does not compile on PPC because of a missed symbol name change in drivers/ide/ide-probe.c. This instance got missed because it is in a #ifdef CONFIG_PPC section. The patch below fixes it. Please apply. Thanks, Paul. ChangeSet@1.1109.10.9, 2003-09-15 17:30:41-03:00, liam.girdwood@wolfsonmicro.com [PATCH] 2.4.23-pre3 WM97xx touchscreen documentation Hi Marcelo, 3rd attempt at sending a patch, this is getting embarrassing ;). Apparently gnome-terminal is mangling it during a cut and paste. Please find attached a small patch thats adds some documentation for the Wolfson WM97xx touchscreen driver to the configure system. I'm having to resend this because Evolution mangled the patch when I sent it to lkml on last Friday. ChangeSet@1.1109.10.8, 2003-09-15 16:10:21-03:00, axboe@suse.de [PATCH] ide-cd capacity "bug" >From Daniel Pittman Using the lra (last recorded address) isn't always reliable on some drives, so use the regular TOC if it returns 0 capacity. ===== drivers/ide/ide-cd.c 1.23 vs edited ===== ChangeSet@1.1109.10.7, 2003-09-15 14:53:05-03:00, m.c.p@wolk-project.de [PATCH] Fwd: [PATCH 2.4.23-pre1] Menu fixes Hi Marcelo, 2.4.22 has some bogus config.in stuff. Let's fix this up. Patch info: ChangeSet@1.1109.10.6, 2003-09-15 14:50:33-03:00, m.c.p@wolk-project.de [PATCH] Missing 'Hermes in TMD7160/NCP130 based PCI adaptor support' config option Hi David, Marcelo, is there any reason why the config.in option is missing for TMD7160/NCP130? If there is no reason and it's only a mistake, please apply the attached patch. It adds a Config.in option and a Configure.help entry. ChangeSet@1.1109.10.5, 2003-09-15 14:48:24-03:00, m.c.p@wolk-project.de [PATCH] Fix 'head.S:116: warning: extra tokens at end of #endif directive' subject says it all. Please apply. Thank you. ChangeSet@1.1109.10.3, 2003-09-15 14:29:02-03:00, m.c.p@wolk-project.de [PATCH] Fix wrong slash/backslash in ACPI very simple fix. This is bogus: printk(KERN_ERR "ERROR: Unable to locate IOAPIC for IRQ %d/n", irq); This is right: printk(KERN_ERR "ERROR: Unable to locate IOAPIC for IRQ %d\n", irq); ChangeSet@1.1109.10.2, 2003-09-15 14:27:11-03:00, bunk@fs.tum.de [PATCH] add CONFIG_AGP_ATI Configure.help entry The patch below against 2.4.23-pre4 adds the missing Configure.help entry for CONFIG_AGP_ATI. The help text is stolen from 2.6. Please apply Adrian ChangeSet@1.1063.2.106, 2003-09-15 08:46:40-07:00, trini@kernel.crashing.org PPC32: Fix another incorrect asm statement. From Scott Anderson . ChangeSet@1.1063.2.105, 2003-09-15 08:46:01-07:00, trini@kernel.crashing.org PPC32: Fix udelay in the PPC boot code for non-16.6 MHz timebases. Patch from Scott Anderson originally. ChangeSet@1.1109.10.1, 2003-09-15 08:39:19-03:00, jgarzik@pobox.com [PATCH] fix ifdown+ifup see $attachment [ Part 2: "Attached Text" ] ChangeSet@1.1109.8.4, 2003-09-15 15:26:20+04:00, green@angband.namesys.com Update reiserfs configure help ChangeSet@1.1109.9.1, 2003-09-15 02:17:23-04:00, lenb@dhcppc11. Merge dhcppc11.:/home/lenb/bk/linux-2.4.23 into dhcppc11.:/home/lenb/bk/linux-acpi-release-2.4.23 ChangeSet@1.1109.1.3, 2003-09-15 02:16:23-04:00, lenb@dhcppc11. Merge dhcppc11.:/home/lenb/bk/linux-2.4.23 into dhcppc11.:/home/lenb/bk/linux-acpi-test-2.4.23 ChangeSet@1.1109.8.3, 2003-09-14 19:51:21-03:00, neilb@cse.unsw.edu.au [PATCH] Revert broken knfsd change On Sunday September 14, willy@w.ods.org wrote: > Hi Marcelo, Neil, > > I've tested -pre4 on my alpha, and noticed that knsfd doesn't work anymore : > the client sticks in D state forever. It has been working flawlessly for > weeks with 2.4.22-rc2. What's strange is that 23-pre4 is OK on my athlon with > the same nfs-utils (1.0.5). > > I have the following NFSD options on both kernels : > CONFIG_NFSD=m > CONFIG_NFSD_V3=y > CONFIG_NFSD_TCP=y > > My alpha kernels were build with GCC 3.2.3, while the athlon one is done with > 2.95.3. > > If I have some time, I'll try intermediate kernels to find which one brought > the problem. I noticed that there were knfsd changes in 2.4.23-pre3, perhaps > they're related. If you want me to try a patch, please ask. I know what broke it. Please try this patch and let me know. NeilBrown ======================================================================= Fix cmsg setup for sock_sendmsg in svc_sendto From: Trond Myklebust ... see the code in ip_sockglue.c + the macros in socket.h.... AFAICS the control messages have wierd alignment requirements. ----------- Diffstat output ------------ ./net/sunrpc/svcsock.c | 20 ++++++++++---------- 1 files changed, 10 insertions(+), 10 deletions(-) ChangeSet@1.1109.8.2, 2003-09-13 16:16:13-03:00, geert@linux-m68k.org [PATCH] Fixup atyfb changes in -pre4. > atyfb_base.c: In function `aty_set_crtc': > atyfb_base.c:501: warning: passing arg 2 of `aty_st_lcd' makes integer > from pointer without a cast > atyfb_base.c:501: too few arguments to function `aty_st_lcd' > atyfb_base.c:504: warning: passing arg 2 of `aty_st_lcd' makes integer > from pointer without a cast > atyfb_base.c:504: too few arguments to function `aty_st_lcd' > make[3]: *** [atyfb_base.o] Error 1 > make[3]: Leaving directory > `/data2/usr/local/src/linux-2.4-pre/drivers/video/aty' > > I now disabled CONFIG_FB_ATY_GENERIC_LCD and it builds. Apparently Daniël didn't sent the latest version to Marcelo? Here are some fixes: ChangeSet@1.1109.8.1, 2003-09-12 21:06:32-07:00, laforge@netfilter.org [NETFILTER]: Use u16 for port numbers. ChangeSet@1.1109.2.19, 2003-09-12 17:59:19-03:00, mikpe@csd.uu.se [PATCH] repair mpparse for default MP systems Marcelo, This patch fixes a memory corruption bug which causes SMP kernels to malfunction on some older dual-Pentium systems. Tested. Please apply. 2.4.21-pre2 changed a number of static arrays used by the I/O-APIC code to be pointers instead, the intent being to handle recent "large" machines. smp_read_mpc() was changed to allocate memory and initialise these pointers. However, that function is never called on older systems where the MP table uses a "default configuration". The result is that on those systems the pointers remain uninitialised, with random failures as the result. The patch adds allocation and initialisation of the pointers to construct_default_ISA_mptable() which is what these systems call instead of smp_read_mpc(). The resulting array layouts are exactly like they were before the 2.4.21-pre2 changes. /Mikael ChangeSet@1.1109.2.18, 2003-09-12 17:29:08-03:00, ak@muc.de [PATCH] Fix x86-64 compatilation for pre4 Just make x86-64 compile again. -Andi ChangeSet@1.1109.2.17, 2003-09-12 15:34:59-03:00, marcelo@logos.cnet Cset exclude: root@macp.eti.br|ChangeSet|20030912113656|10550 TAG: v2.4.23-pre4 ChangeSet@1.1109.2.16, 2003-09-12 15:00:27-03:00, marcelo@logos.cnet Changed EXTRAVERSION to -pre4 ChangeSet@1.1109.2.15, 2003-09-12 14:50:34-03:00, marcelo@logos.cnet -aa VM merge: Fixup page reclaiming changes patch ChangeSet@1.1109.2.14, 2003-09-12 14:49:13-03:00, marcelo@logos.cnet -aa VM merge: Kill PF_MEMDIE ChangeSet@1.1109.2.13, 2003-09-12 14:48:17-03:00, marcelo@logos.cnet -aa VM merge: tunables ChangeSet@1.1109.7.1, 2003-09-12 10:25:42-07:00, trini@kernel.crashing.org PPC32: Add CONFIG_ADVNACED_OPTIONS to make the kernel more flexible. This is from Matt Porter , and allows for the kernel to be better configured for specific boards. ChangeSet@1.1109.2.12, 2003-09-12 14:13:55-03:00, marcelo@logos.cnet -aa VM merge: Page accounting helpers changes ChangeSet@1.1109.2.11, 2003-09-12 14:12:38-03:00, marcelo@logos.cnet -aa VM merge: page reclaiming logic changes: Kills oom killer that achieves multiple things. It avoids oom deadlocks by not wasting time in the pagetable walking anymore after we failed once, it protects init from being killed, and most important it avoids failed oom kills if a task has been killed under us (or if plenty of ram has been freed under us for whatever else reason). See the check_classzone_need_balance checks. Then it gives classzone awareness to refill_inactive so we make sure to make progress for non highmem allocs too and to shrink stuff properly, the lists are global. Plus it checkpoints the point in the active list where it stopped the last time. It also changes the shrink_cache function to shrink the vfs lists internally if needed. The max_scan etc.. in shrink_cache are as well classzone aware, since the lists are global but we skip over the non interesting pages (like in refill_inactive). ChangeSet@1.1109.2.10, 2003-09-12 14:11:05-03:00, marcelo@logos.cnet -aa VM merge: Per-zone watermark changes, add lower_zone_reserve_ratio ChangeSet@1.1109.2.9, 2003-09-12 12:06:20-03:00, bjorn.helgaas@hp.com [PATCH] 2.4 HCDP early printk support This adds config and HCDP support for early printk. This is currently used only by IA64, and the equivalent code is already in 2.5. ===== Documentation/Configure.help 1.192 vs edited ===== ChangeSet@1.1109.5.3, 2003-09-12 08:36:56-03:00, root@macp.eti.br [PATCH] Support for Netmos 6 port 16550a serial card Follows a trivial patch to add support of a popular serial card in Brazil that contains 6 standard 16550A uarts with a Netmos PCI glue. Only adds the apropriate entries to the proper pci and serial ids. Please apply. ChangeSet@1.1109.5.2, 2003-09-12 08:25:03-03:00, green@linuxhacker.ru [PATCH] [2.4] Rocketport driver compile fix Hello! Rocketport driver does not compile in latest 2.4 bk, it seems to assume that tty->count is of atomic type which is no longer true. Seems that this patch is needed now. ===== drivers/char/rocket.c 1.9 vs edited ===== ChangeSet@1.1109.5.1, 2003-09-12 08:05:51-03:00, axboe@suse.de [PATCH] Add NEC iStorage to SCSI blacklist Marcelo, Please add this patch as requested by NEC. Subject: Additioal Line for SCSI Blacklist According to our recent conversation via e-mail, please add following one line into "scsi_scan.c device_list". Both NEC and our customer tested the line repeatedly and confirmed proper operation. {"NEC", "iStorage", "*", BLIST_SPARSELUN | BLIST_LARGELUN | BLIST_FORCELUN}, ===== drivers/scsi/scsi_scan.c 1.29 vs edited ===== ChangeSet@1.1063.44.5, 2003-09-12 04:15:36-04:00, len.brown@intel.com fix off-by-one error in ioremap() fixes kernel crash in acpi mode: http://bugzilla.kernel.org/show_bug.cgi?id=1085 ChangeSet@1.1109.4.1, 2003-09-11 18:29:12-07:00, wesolows@foobazco.org [SPARC32]: Ignore btfixups in .text.exit ChangeSet@1.1109.2.5, 2003-09-11 18:24:37-07:00, chas@cmf.nrl.navy.mil [ATM]: Backport lane/mpoa module locking cleanup from 2.6.x ChangeSet@1.1109.2.4, 2003-09-11 18:10:34-07:00, shemminger@osdl.org [BRIDGE]: Clear hw checksum flags when bridging. ChangeSet@1.1109.3.1, 2003-09-11 20:55:31-03:00, alan@lxorguk.ukuu.org.uk [PATCH] Fix ymfpci oops Sorry for the delay ChangeSet@1.1109.2.3, 2003-09-11 16:46:03-07:00, erlend-a@us.his.no [CRYPTO]: Add alg. type to /proc/crypto output. ChangeSet@1.1109.2.1, 2003-09-11 17:39:57-03:00, eric@lammerts.org [PATCH] fix current->user->processes leak in reparent_to_init() On Tue, 28 Jan 2003, Brian Sullivan wrote: /> The problem I am experiencing is that after a certain number of mounts I / /> get the error message "fork: Resource temporarily unavailable" on the / /> command line. / /> After much trouble shooting I realized that number of mounts/umount / /> sequences I am limited to is the max number of processes for my user id. I / /> confirmed this by using the "ulimit -u" command to lower my process limit. / /> It appears the mount command is leaving someting in some sort of process / /> table in kernel memory (nothing shows in ps or top or in /proc/#### as / /> being left behind). Has anybody any sort of experience with this at all? / /> Any suggestions? / It's a kernel bug (and easy to reproduce). Every time you do a loop mount, a kernel thread is started (those processes are called "loop0", "loop1", etc.). The problem is that when it starts, it's counted as one of your processes *). Then, it's changed to be a root-owned process without correcting that count **). Patch below fixes the problem. It moves the bookkeeping of changing current->user to a new function switch_uid() (which is now also used by exec_usermodehelper() in kmod.c). The patch is tested. Eric *) "atomic_inc(&p->user->processes);" in do_fork(). **) "this_task->user = INIT_USER;" in reparent_to_init(). ChangeSet@1.1108, 2003-09-09 13:44:04-06:00, bjorn.helgaas@hp.com Merge hp.com:/home/helgaas/linux/linux-2.4 into hp.com:/home/helgaas/linux/to-marcelo-2.4 ChangeSet@1.1063.45.2, 2003-09-09 14:22:33-03:00, paulkf@microgate.com [PATCH] synclink drivers fixup Marcelo: These patches replace a check for NULL context passed to the driver that was incorrectly removed in the last patches I sent you (now applied in 2.4.22-BK13) drivers/char/synclink.c drivers/char/synclinkmp.c drivers/char/pcmcia/synclink.c ChangeSet@1.1063.45.1, 2003-09-09 09:24:38-03:00, ralf@linux-mips.org [PATCH] PATCH: avoid glibc conflict This is needed to avoid a conflict with glibc on bigendian platforms when -O or higher is specified. It's already in 2.6, and I'm not sure why it hasn't been seen in 2.4. The symptom is that this program will not compile with -O2: #include #include int main () { } ChangeSet@1.1063.43.2, 2003-09-08 15:40:21-04:00, len.brown@intel.com Merge intel.com:/home/lenb/src/marcelo-2.4 into intel.com:/home/lenb/src/to-linux-acpi-2.4.23 ChangeSet@1.1063.44.4, 2003-09-08 15:30:57-04:00, len.brown@intel.com remove ASUS A7V BIOS version 1011 from blacklist (Eric Valette) ChangeSet@1.1063.44.3, 2003-09-08 15:30:41-04:00, len.brown@intel.com support non ACPI compliant SCI over-ride specs (Jun Nakajima) ChangeSet@1.1063.44.2, 2003-09-08 15:29:32-04:00, len.brown@intel.com Fix ACPI oops on ThinkPad T32/T40 (Shaohua David Li) ChangeSet@1.1063.44.1, 2003-09-08 15:10:38-04:00, len.brown@intel.com Extended IRQ resource type for nForce (Andrew de Quincey) Handle BIOS with _CRS that fails (Jun Nakajima) ChangeSet@1.1063.2.103, 2003-09-06 22:12:34+10:00, paulus@samba.org PPC32: Fix a bug where TLB entries didn't get execute permission on 40x. ChangeSet@1.1063.2.102, 2003-09-06 20:14:17+10:00, paulus@samba.org PPC32: Fix a typo in the PPC 440GP support. ChangeSet@1.1063.2.101, 2003-09-06 20:13:33+10:00, paulus@samba.org PPC32: Simplify VMALLOC_START, make it just a variable. Previously VMALLOC_START was a complex expression which was reevaluated each time it was used, and it had multiple ifdefs around it. Now VMALLOC_START is defined to be vmalloc_start, which is a variable computed once at mem_init time. ChangeSet@1.1063.2.100, 2003-09-06 13:33:16+10:00, paulus@samba.org PPC32: Fix for highmem on machines with 64-bit PTEs (e.g. PPC440) ChangeSet@1.1063.40.11, 2003-09-05 16:09:03-07:00, joris@struyve.be [PATCH] unusual_devs.h entry Hereby you may find my patch for a Medion digital camera along with the /proc/bus/usb/devices output. ChangeSet@1.1063.40.10, 2003-09-05 15:47:40-07:00, greg@kroah.com [PATCH] USB: fix copy_to_user calls in vicam driver. ChangeSet@1.1063.40.9, 2003-09-05 15:47:32-07:00, greg@kroah.com [PATCH] USB: remove duplicated copy_from_user call in stv680 driver ChangeSet@1.1063.40.8, 2003-09-05 15:47:25-07:00, greg@kroah.com [PATCH] USB: fix copy_to_user call in mdc800 driver ChangeSet@1.1063.40.7, 2003-09-05 15:47:17-07:00, greg@kroah.com [PATCH] USB: fix copy_to_user call in uhci-debug.h ChangeSet@1.1063.40.6, 2003-09-05 15:47:10-07:00, greg@kroah.com [PATCH] USB: fix copy_from_user call in aiptek.c ChangeSet@1.1063.40.5, 2003-09-05 15:47:03-07:00, greg@kroah.com [PATCH] USB: fix copy_from_user call in acm.c ChangeSet@1.1063.40.4, 2003-09-05 15:46:55-07:00, greg@kroah.com [PATCH] USB: fix oops when yanking a usb-serial device from the system with the port still opened. Thanks to Ian Abbott for help finding this bug. ChangeSet@1.1063.2.97, 2003-09-05 12:11:29-07:00, trini@kernel.crashing.org PPC32: Minor bootwrapper fixups. ChangeSet@1.1063.2.96, 2003-09-05 11:04:21-03:00, daniel@deadlock.et.tudelft.nl [PATCH] Implement LCD display support in atyfb driver + LCD display detection code added + LCD mode calculation code added + Enable interlacing when requested instead of abort with error + Enable double scan when requested instead of abort with error * Clock programming code rewritten to support asynchrone clocks, necessary to clock all Rage chips at the right frequencies * Clock table checked and updated with information from ATi * Display fifo code rewritten to take into account lcd mode calculations and eliminate display artifacts at very low resolution video modes + Added possibility to disable clock programming code in case of problems by modprobing with 'default_xclk=-1 default_mclk=-1' ChangeSet@1.1063.42.1, 2003-09-05 22:38:16+10:00, paulus@samba.org PPC32: Handle single-stepped emulated instructions correctly. When we emulate an instruction in the kernel (e.g. for an alignment exception, or if we are doing FP emulation), we need to check if we are single-stepping after we successfully emulate the instruction. If so we deliver a SIGTRAP to the process. ChangeSet@1.1063.2.94, 2003-09-04 20:04:58-03:00, marcelo@logos.cnet Mehmet Ceyran/Alan Cox: Longer i810_audio.c retries ChangeSet@1.1107, 2003-09-04 14:39:26-06:00, erikj@subway.americas.sgi.com ia64: 9/3/2003 SGI update. ChangeSet@1.1063.2.93, 2003-09-04 16:29:31-03:00, paulkf@microgate.com [PATCH] n_hdlc update * make global vars static * fix memory leak please apply -- Paul Fulghum, paulkf@microgate.com Microgate Corporation, http://www.microgate.com ChangeSet@1.1063.2.92, 2003-09-04 16:26:30-03:00, paulkf@microgate.com [PATCH] synclink_cs update * fix arbitration between net open and tty open * add MODULE_LICENSE() macro * add dosyncppp module parameter Please apply -- Paul Fulghum, paulkf@microgate.com Microgate Corporation, http://www.microgate.com ChangeSet@1.1063.2.91, 2003-09-04 16:25:54-03:00, paulkf@microgate.com [PATCH] synclinkmp update * fix arbitration between net open and tty open * add MODULE_LICENSE() macro * use time_after() macro ChangeSet@1.1063.2.90, 2003-09-04 16:25:14-03:00, paulkf@microgate.com [PATCH] synclink update * Add support for version 2 of synclink PCI adapter * Fix arbitration between net open and tty open * Use time_after() macro * Use mod_timer() macro * Fix RCC underrun handling Please apply. -- Paul Fulghum, paulkf@microgate.com Microgate Corporation, http://www.microgate.com ChangeSet@1.1063.2.89, 2003-09-04 15:56:16-03:00, m.c.p@wolk-project.de [PATCH] Fixup 'make xconfig' problem caused by fetchop Config.in change On Thursday 04 September 2003 05:18, Supphachoke Suntiwichaya wrote: Hi Mrchoke, > I can't make xconfig :: > [root@potter linux-2.4.23-pre3]# make xconfig > rm -f include/asm > ( cd include ; ln -sf asm-i386 asm) > make -C scripts kconfig.tk > make[1]: Entering directory `/usr/src/linux-2.4.23-pre3/scripts' > cat header.tk >> ./kconfig.tk > ./tkparse < ../arch/i386/config.in >> kconfig.tk > drivers/char/Config.in: 270: bad if condition > make[1]: *** [kconfig.tk] Error 1 > make[1]: Leaving directory `/usr/src/linux-2.4.23-pre3/scripts' > make: *** [xconfig] Error 2 I can ... with the attached fix :p Marcelo, please apply this too. Thank you. ChangeSet@1.1063.2.88, 2003-09-04 15:52:52-03:00, neilb@cse.unsw.edu.au [PATCH] knfsd: Make sure nfsd replies from the address the request was sent to. This is important on multi-homes hosts. ChangeSet@1.1063.2.87, 2003-09-04 15:52:13-03:00, neilb@cse.unsw.edu.au [PATCH] knfsd: Change name of a #define in nfsd to match 2.6 ### Comments for ChangeSet The meaning is "nohide" not "crossmnt", so we should spell it that way. ----------- Diffstat output ------------ ./fs/nfsd/export.c | 2 +- ./fs/nfsd/vfs.c | 6 +++--- ./include/linux/nfsd/export.h | 4 ++-- 3 files changed, 6 insertions(+), 6 deletions(-) ChangeSet@1.1063.2.86, 2003-09-04 15:51:01-03:00, neilb@cse.unsw.edu.au [PATCH] knfsd: Make sure nfs/tcp socket only gets closed It is possible for svc_delete_socket to be called multiple times, which can cause problems. This patch uses SK_DEAD to make sure it only has any effect the first time, and uses SK_DEAD in a few other places to make sure nothing is done on a deleted socket. ChangeSet@1.1063.2.85, 2003-09-04 14:27:35-03:00, arjanv@redhat.com [PATCH] LSB compliance fix in mprotect Hi, The following fix is needed for the mprotect code in order to make the 2.4 kernel pass the LSB testsuite. Posix/SUS specify, per http://www.opengroup.org/onlinepubs/007904975/functions/mprotect.html that mprotect is allowed to return -EINVAL only in the case of the address not being page aligned. Any other address and address range errors are supposed to return -ENOMEM. Greetings, Arjan van de Ven ChangeSet@1.1106, 2003-09-04 11:06:30-06:00, eranian@hpl.hp.com ia64: Fix perfmon usage of rum/srsm and sum/ssm - fixes a problem with the rum/sum instructions being inoperative even at priv level zero when psr.sp=0. Switching to ssm/rsm fixes the potential problem. ChangeSet@1.1063.40.3, 2003-09-04 08:52:26-07:00, greg@kroah.com [PATCH] USB: backport some pl2303 B0 fixes. Also add proper locking for the port structure. This comes from the 2.6 kernel tree too. ChangeSet@1.1063.2.84, 2003-09-04 09:54:28-03:00, neilb@cse.unsw.edu.au [PATCH] Fix typo in umem.c This would only cause a problem if initialising a device fails part-way through, so it probably hasn't hurt anyone. ChangeSet@1.1063.2.83, 2003-09-04 09:48:45-03:00, neilb@cse.unsw.edu.au [PATCH] knfsd: lockd fails to purge blocked NLM_LOCKs From: Mark Hemment When stopping server-side NFS, or deleting a client (NFSCTL_DELCLIENT), lockd fails to purge blocked lockd requests. The call path is; exp_freeclient() nfsd_lockd_unexport() nlmsvc_invalidate_client() Where nlmsvc_invalidate_client(), looks up the host with; host = nlm_lookup_host(clnt, NULL, 0, 0) ie. the host-lookup keys off the client's "svc_client" (memory) address from NFSD's "clients" list. This "host" value is passed through; nlmsvc_free_host_resources() nlm_traverse_files() nlm_inspect_file() and into nlmsvc_traverse_blocks(), where blocked locks are tested for with; nlmsvc_traverse_blocks(host, ....) { ... if (host == NULL || host == block->b_host) nlmsvc_delete_block(block, 1); ... } However, where the block entries are created in nlmsvc_create_block(); host = nlmclnt_lookup_host(&rqstp->rq_addr, rqstp->rq_prot, rqstp->rq_vers); ... block->b_host = host; this host-lookup (nlmclnt_lookup_host()) is keyed off the client's network address. So, back in nlmsvc_traverse_blocks() the test; host == block->b_host is checking different values; "host" found via "svc_client"s memory address and "b_host" found/created via the client's network address. The bug/confusion probably originates as "nlm_hosts[]" is overload with two different types of entries. nlmsvc_create_block() still needs to create an entry in nlm_hosts[] for the sending of the 'NLM_GRANTED', but it already stores a reference to that entry in "call->a_host". Solution is to use the value in "call->a_host" for the nlm_rebind_host(), and to set "b_host" with the host value found via an nlmsvc_lookup_host(). ChangeSet@1.1063.2.82, 2003-09-04 09:47:47-03:00, neilb@cse.unsw.edu.au [PATCH] knfsd: Set d_op when creating a parent directory during nfsd fh->dentry conversion ### Comments for ChangeSet While it is not necessarily safe to assume the parent has the same d_op as the child, it is safe than assuming the parent has a d_op of NULL. ----------- Diffstat output ------------ ./fs/nfsd/nfsfh.c | 1 + 1 files changed, 1 insertion(+) ChangeSet@1.1063.2.81, 2003-09-04 09:44:58-03:00, neilb@cse.unsw.edu.au [PATCH] knfsd: Lock client list while detaching locks From: Mark Hemment The 'client' list should always be protected by the export table lock, but it isn't here! ChangeSet@1.1063.41.5, 2003-09-04 04:33:00-07:00, karlis@mt.lv [BRIDGE]: kfree --> kfree_skb. ChangeSet@1.1063.41.4, 2003-09-04 04:11:37-07:00, laforge@netfilter.org [NETFILTER]: NAT range calculation fix. This patch fixes a logic bug in NAT range calculations, which also causes a large slowdown when ICMP floods go through NAT. Author: Karlis Piesenieks ChangeSet@1.1063.41.3, 2003-09-03 20:36:17-07:00, chas@cmf.nrl.navy.mil [ATM]: #define'ing pci_pool_create() breaks CONFIG_MODVERSION. ChangeSet@1.1063.41.2, 2003-09-03 20:35:35-07:00, chas@cmf.nrl.navy.mil [ATM]: If clip isn't a module don't __MOD_DEC_USE_COUNT(). ChangeSet@1.1063.41.1, 2003-09-03 20:31:21-07:00, chas@nrl.navy.mil [ATM]: In atm_getaddr() do not copy_to_user() with locks held. ChangeSet@1.1063.2.80, 2003-09-03 18:55:03-03:00, jt@bougret.hpl.hp.com [PATCH] wireless extension update: 802.11a/802.11g fixes Hi Marcelo, Could you add this in 2.4.23 ? This patch update Wireless Extensions to v16. This is the same patch that went into kernel 2.5.70, but rediffed and tested with 2.4.23-pre2. This update fixes a few details for 802.11a/802.11g support, like increasing the number of frequencies and bitrates. It also features an improved iwspy support. Note that airo.c included in 2.4.23-pre2 already include WE-16 support, so this patch is needed to enable this support. Of course, I tested that airo.c works properly after the upgrade. Thanks... Jean ChangeSet@1.1063.2.79, 2003-09-03 18:45:43-03:00, marcelo@logos.cnet Changed EXTRAVERSION to -pre3 TAG: v2.4.23-pre3 ChangeSet@1.1063.2.78, 2003-09-03 17:23:43-03:00, marcel@holtmann.org [PATCH] PCI quirk for SMBus bridge on Asus P4 boards Hi Marcelo, this has been in 2.5 for some time now and I think it can be added to 2.4 too. ChangeSet@1.1131, 2003-09-03 08:42:53+02:00, marcel@holtmann.org [PATCH] PCI quirk for SMBus bridge on Asus P4 boards Asus hides the SMBus PCI bridge within the ICH2 or ICH4 southbridge on Asus P4 mainboards. The attached patch adds a quirk to re-enable the SMBus PCI bridge for P4B533, P4T533, P4PE and P4G8X Deluxe mainboards. Original patch for 2.5 from Dominik Brodowski drivers/pci/quirks.c | 66 ++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/pci_ids.h | 4 ++ 2 files changed, 70 insertions(+) TAG: v2.4.23-pre3 ChangeSet@1.1063.40.2, 2003-09-03 11:56:08-07:00, greg@kroah.com [PATCH] USB: update usb-serial.h with spelling fixes and get and set functions Based off of the current 2.6 version, makes keeping the usb-serial drivers more in sync between the two trees. ChangeSet@1.1063.40.1, 2003-09-03 11:54:59-07:00, greg@kroah.com [PATCH] USB: fix data toggle problem for pl2303 driver. ChangeSet@1.1063.38.15, 2003-09-03 15:26:05-03:00, m.c.p@wolk-project.de [PATCH] Add missing IRQ_NONE clash fix hunk > This change added an (empty) IRQ_NONE #define to interrupt.h. > Several scsi drivers are already using an IRQ_NONE. Rename that to > SCSI_IRQ_NONE (a similar change was done in 2.5 by Andrew Morton several > months ago). right, but you forgot one header :-) ... Attached is a patch. ChangeSet@1.1063.38.14, 2003-09-03 14:57:40-03:00, marcelo@logos.cnet Adrian & Chantal: Unused variable in ip2main.c ChangeSet@1.1063.2.76, 2003-09-03 13:47:45-04:00, javier@tudela.mad.ttd.net [wireless airo] MIC support using CryptoAPI ChangeSet@1.1063.2.75, 2003-09-03 13:47:06-04:00, javier@tudela.mad.ttd.net [wireless airo] PCI detection code fixes ChangeSet@1.1063.2.74, 2003-09-03 13:46:36-04:00, javier@tudela.mad.ttd.net [wireless airo] Build fixes when MIC disabled ChangeSet@1.1063.2.73, 2003-09-03 13:45:42-04:00, jgarzik@redhat.com Cset exclude: jgarzik@redhat.com|ChangeSet|20030826234629|07076 ChangeSet@1.1063.2.72, 2003-09-03 13:15:13-04:00, mlindner@syskonnect.de [netdrvr sk98lin] update to driver version 6.17 ChangeSet@1.1063.2.71, 2003-09-03 12:46:44-04:00, jgarzik@redhat.com [netdrvr 3c515] fix non-modular build ChangeSet@1.1063.2.70, 2003-09-03 12:45:35-04:00, jgarzik@redhat.com [netdrvr] add MV-64340 gigabit ethernet driver (MIPS only) Contributed via Marcelo via Ralf, the MIPS maintainer. ChangeSet@1.1063.2.69, 2003-09-03 12:35:33-04:00, xose@wanadoo.es [netdrvr 3c59x] update pci ids In 2.4.23-pre2 two pci.ids are missing and there is one duplicate. 2.6.0-preX driver is ok. ChangeSet@1.1063.38.13, 2003-09-03 13:35:30-03:00, marcelo@logos.cnet Cset exclude: agruen@suse.de|ChangeSet|20030902115108|61891 ChangeSet@1.1063.2.68, 2003-09-03 12:34:08-04:00, bunk@fs.tum.de [wireless airo] fix build with gcc 2.95 ChangeSet@1.1063.2.67, 2003-09-03 12:10:33-04:00, jgarzik@redhat.com Merge redhat.com:/spare/repo/linux-2.4 into redhat.com:/spare/repo/net-drivers-2.4 ChangeSet@1.1105, 2003-09-02 19:04:21-06:00, steiner@sgi.com ia64: remove some SN1 remnants, add a bit more SN2 support. linux/arch/ia64/config.in - delete some remnents of support for the SGI SN1 platform - add config option for SN2. linux_base/arch/ia64/kernel/Makefile - fix compile error (SN2 needs iosapic.o) linux/arch/ia64/acpi.c - update acpi.c to recognize the SN2 platform linux/arch/ia64/kernel/efi_stub.S linux/arch/ia64/kernel/head.S linux/arch/ia64/kernel/pal.S - backport of 2.6 "ia64_switch_mode" changes for non-identity mapped kernels ChangeSet@1.1063.39.1, 2003-09-02 14:36:07-07:00, davem@nuts.ninka.net [IPV4]: Do not BUG() on icmp_xmit_lock() contention, just drop. ChangeSet@1.1063.38.11, 2003-09-02 18:11:13-03:00, bunk@fs.tum.de [PATCH] Fix IRQ_NONE clash in SCSI drivers On Wed, Aug 27, 2003 at 02:52:45PM -0300, Marcelo Tosatti wrote: >... > Summary of changes from v2.4.22 to v2.4.23-pre1 > ============================================ >... > Arnaldo Carvalho de Melo: > o irqreturn_t compatibility with 2.6 >... This change added an (empty) IRQ_NONE #define to interrupt.h. Several scsi drivers are already using an IRQ_NONE. Rename that to SCSI_IRQ_NONE (a similar change was done in 2.5 by Andrew Morton several months ago). I've tested the compilation with 2.4.23-pre2. Please apply Adrian ChangeSet@1.1063.38.10, 2003-09-02 10:21:42-03:00, khc@pm.waw.pl [PATCH] generic HDLC update Marcelo Tosatti writes: > Krzysztof, please send me a mail with "generic HDLC update" subject and > detailed changelog in the body, followed by the patch. (thats easier for > me to apply :)) As you wish: Looks like I lied - the DCD (carrier detect) isn't here as it needs more testing - it's in 2.6 only. Changes: * Documentation and spelling fixes * user can query PVC device for type and DLCI * private Frame Relay helper routines moved out of hdlc.h * added hdlc_type_trans() to simplify packet receive path and make tcpdump etc. happy * missing rtnl_lock()/rtnl_unload() bug in unregister_hdlc_device fixed * N2, C101: interrupt handler now works under high IRQ load from other devices (with previous versions, the IRQ processing for the card could sometimes stop after reaching "work limit") * Added Ethernet device emulation for raw HDLC. VLAN and bridging compatible, tested with RAD ChipBridge (ChipBridge doesn't work with full size VLAN frames, though). * Added Ethernet device emulation for bridged Ethernet frames on Frame-Relay. * no more "protocol 0008 is buggy" while using tcpdump, at last. * Frame-Relay DCE (network) side now sorts DLCI list in PVC FULL STATUS. Some FR DTE had problems with unsorted list. * Small changes to LMI logic with Frame-Relay. * raw HDLC now defaults to ITU-T 16-bit CRC. * fix for SCA-II based cards (drivers not in standard kernel) interrupt setting when using 2 ports. * there is now one detect_ram routine for all SCA-based cards. * hw drivers have now smaller TX packet rings to decrease TX latency for high priority packets (LMI etc) * fixed subtle transmit bug in c101.c which could lead to transmitter hangs and duplicated frames with SCA HD64570 working in 8-bit mode. * kmalloc(GFP_KERNEL) in interrupt context has been fixed. * Some __init etc fixes. * other misc fixes -- Krzysztof Halasa, B*FH ChangeSet@1.1063.38.9, 2003-09-02 08:51:08-03:00, agruen@suse.de [PATCH] fix setxattr const-ness Hello Marcelo, this is a fix I've already sent you earlier. It fixes the declaration of the value parameter of the setxattr inode operation, which is supposed to be const---it's a trivial bug. Please apply. Index: linux-2.4.23-pre2/fs/jfs/jfs_xattr.h =================================================================== ChangeSet@1.1063.38.8, 2003-09-01 20:57:39-04:00, jgarzik@redhat.com [PCI, ia32] don't assume "c->x86 > 6" applies to non-Intel CPUs when programming PCI cache line size. Just to be on the safe side. Suggested by Ingo Oeser. ChangeSet@1.1063.2.66, 2003-09-01 20:26:32-04:00, jgarzik@redhat.com [NET] move ethtool_op_set_tx_csum from 8139cp drvr to net/core/ethtool.c, where it belongs. ChangeSet@1.1063.2.65, 2003-09-01 17:23:07-04:00, jgarzik@redhat.com [netdrvr] sync with 2.5: epic100, fealnx, via-rhine, winbond-840 Minor cleanups and fixes: * whitespace * remove __devinitdata * irqreturn_t and pci_name() changes * set_bit removal (a fix) * C99 initializers ChangeSet@1.1063.2.64, 2003-09-01 17:12:15-04:00, jgarzik@redhat.com [netdrvr] ethtool_ops for epic100, fealnx, winbond-840, via-rhine ChangeSet@1.1063.38.6, 2003-09-01 17:08:23-03:00, m.c.p@wolk-project.de [PATCH] Disable alpha S3 Savage/VIACLE266 DRM support On Thursday 28 August 2003 00:21, Alan Cox wrote: Hi Alan, > My preference would be to leave the VIA and S3 modules out for the > moment and wait until the XFree DRI folks have the stuff working nicely. > Its in -ac because I'm hacking on it not because it works, is ready. > Or I guess you could just zap the config.in for them ? ok with me. Attached a patch ontop of the previous sent update. ChangeSet@1.1063.38.5, 2003-09-01 17:05:39-03:00, m.c.p@wolk-project.de [PATCH] Update DRI/DRM so XFree v4.3.0 and above works Hi Marcelo, Patch information: ------------------ - Update XFree DRM code to recent code from 2.4.22-rc2-ac3 - DRI now works with XFree v4.3.0 - ATI IGP chipset support added incl. pci ids, AGP support - Initial VIA CLE266 and S3 Savage DRM modules merges from VIA (These are marked up with some warnings and need a chunk of clean up work) - enabled 3dlabs GMX 2000 config option All compiles cleanly. r128 (ati rage 128) at least works perfectly with XFree v4.3.0. I have such a card :) Please apply for 2.4.23-pre2. Imho Alan said, it needs some cleanups but I think, Mikael too, that the cleanup should be made once it's in. ChangeSet@1.1063.38.4, 2003-09-01 16:11:03-03:00, cw@sgi.com [PATCH] SGI fetchop driver Marcelo, As discussed here is an updated version of the SGI SN fetchop (AMO) driver for inclusion: Config.in | 4 Makefile | 3 fetchop.c | 516 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 522 insertions(+), 1 deletion(-) # This is a BitKeeper generated patch for the following project: # Project Name: Linux kernel tree # This patch format is intended for GNU patch command version 2.5 or higher. # This patch includes the following deltas: # ChangeSet 1.1087 -> 1.1088 # drivers/char/Config.in 1.51 -> 1.52 # drivers/char/Makefile 1.36 -> 1.37 # (new) -> 1.1 drivers/char/fetchop.c # # The following is the BitKeeper ChangeSet Log # -------------------------------------------- # 03/08/27 cw@tomahawk.engr.sgi.com 1.1088 # Add SGI fetchop driver # -------------------------------------------- # ChangeSet@1.1063.2.63, 2003-09-01 15:03:22-04:00, jgarzik@redhat.com [NET] s/blog_dev/backlog_dev/ in process_backlog, net/core/dev.c This 100% cosmetic change reduces the diff between 2.4 and 2.5. ChangeSet@1.1063.2.61, 2003-09-01 14:56:29-04:00, jgarzik@redhat.com [NET] move netif_* helpers from tg3 driver to linux/netdevice.h Helpers moved: netif_poll_{enable,disable}, __netif_rx_complete, netif_tx_disable Use the helpers in net/core/dev.c. ChangeSet@1.1063.38.3, 2003-09-01 15:11:53-03:00, adam@os.inf.tu-dresden.de [PATCH] Add kmap_types.h to include/asm-alpha A kmap_types.h is needed to compile crypto on alpha. ChangeSet@1.1063.38.2, 2003-09-01 14:23:36-03:00, m.c.p@wolk-project.de [PATCH] aty128fb: find the video bios on a Latitude C600 (M3) Inspiron 8000 (M4) Hi Marcelo, this patch makes aty128fb find the video bios on a Latitude C600 (M3) and Inspiron 8000 (M4). Patch from Andreas Oberritter. Anyway, works. Please apply for 2.4.23-pre2. Thanks. ChangeSet@1.1063.2.59, 2003-09-01 13:03:50-04:00, jgarzik@redhat.com [netdrvr ne2k-pci] ethtool_ops support ChangeSet@1.1063.37.5, 2003-09-01 13:59:20-03:00, stelian@popies.net [PATCH] meye driver update Hi, This is a sync with the 2.5 meye driver. Nothing really important here, just makes it easier for me to maintain both branches. Marcelo, please apply. Thanks, Stelian. ===== drivers/media/video/meye.h 1.8 vs edited ===== ChangeSet@1.1063.37.4, 2003-09-01 13:58:56-03:00, stelian@popies.net [PATCH] reenable CAPTURE button in sonypi This attached patch reenables the CAPTURE button events in the sonypi driver, which were lost with the latest patch... The same patch should apply cleanly on both 2.4 and 2.6 kernels. Marcelo, Linus, please apply. Stelian. ===== drivers/char/sonypi.h 1.18 vs edited ===== ChangeSet@1.1063.37.3, 2003-09-01 13:57:49-03:00, javaman@katamail.com [PATCH] small config fix for ISDN Hi, trying to install modules in 2.4.22 I get this error: depmod: *** Unresolved symbols in /lib/modules/2.4.22/kernel/drivers/isdn/isdn.o depmod: sk_run_filter depmod: sk_chk_filter this is the fix for drivers/isdn/Config.in, please apply: ChangeSet@1.1063.2.58, 2003-09-01 12:56:54-04:00, jgarzik@redhat.com [netdrvr ne2k-pci] sync with 2.5 (100% minor cleanups) ChangeSet@1.1063.2.57, 2003-09-01 12:51:51-04:00, jgarzik@redhat.com [netdrvr ne2k-pci] allocate netdev+8390 struct using new alloc_ei_netdev() Also, call pci_disable_device() in PCI ->remove handler, to match pci_enable_device() in PCI ->probe handler. ChangeSet@1.1063.2.56, 2003-09-01 12:48:28-04:00, jgarzik@redhat.com [netdrvr 8390] new function alloc_ei_netdev() (preferred over alloc_etherdev + 8390-specific ethdev_init) ChangeSet@1.1063.37.2, 2003-09-01 10:30:31-03:00, marcelo@logos.cnet Merge http://linux-mh.bkbits.net/fw-loader-2.4 into logos.cnet:/home/marcelo/bk/linux-2.4 ChangeSet@1.1063.37.1, 2003-09-01 09:14:13-03:00, cw@sgi.com [PATCH] Remove kdb hooks from SGI Altix Console driver Marcelo, Keith does not want KDB distributed outside the main KDB patches. Please apply this incremental update to the SN serial/console driver. Thanks! --cw ChangeSet@1.1063.36.4, 2003-09-01 02:44:55-07:00, davem@nuts.ninka.net [IPV6]: Do not BUG() on icmp6 socket contention, just drop. ChangeSet@1.1063.36.3, 2003-09-01 01:12:11-07:00, xose@wanadoo.es [TG3]: ICH2 needs MBOX write reorder bug workaround too. ChangeSet@1.1063.2.55, 2003-08-31 20:09:19-04:00, jgarzik@redhat.com [netdrvr 8139cp] PCI MWI cleanup; remove unneeded workaround * The PCI layer now handles incorrect cacheline size settings, as it should. Remove our own workarounds. * Move pci_set_mwi up much earlier in the probe process, and check its return value. * Call pci_clear_mwi() in ->probe error handling * Call pci_clear_mwi() in ->remove ChangeSet@1.1063.2.53, 2003-08-31 15:38:40-04:00, hirofumi@mail.parknet.co.jp [netdrvr 8139too] don't start thread when it's not needed The thread for was unneeded on chips other than CH_8139_K/8129. So, this patch doesn't create the thread on chips other than CH_8139_K/8129. ChangeSet@1.1063.2.52, 2003-08-31 15:34:30-04:00, hirofumi@mail.parknet.co.jp [netdrvr 8139too] remove driver-based poisoning of net_device Harmless in 2.4, but causes oopses on rmmod in 2.6. slab poisoning can take care of this for us, anyway. ChangeSet@1.1063.35.1, 2003-08-31 12:22:46-03:00, marcelo@logos.cnet Change contact information, again. ChangeSet@1.1063.2.51, 2003-08-31 11:01:26-04:00, jgarzik@redhat.com [netdrvr 8139cp] must call NAPI-specific vlan hook ChangeSet@1.1063.31.5, 2003-08-31 09:53:21-04:00, jgarzik@redhat.com [BK] ignore auto-generated files lib/{crc32table.h,gen_crc32table} ChangeSet@1.1063.31.4, 2003-08-31 09:52:16-04:00, jgarzik@redhat.com [PCI] fix export of pdev_set_mwi/pci_generic_prep_mwi Missed in recent PCI MWI change. ChangeSet@1.1063.31.3, 2003-08-31 09:51:32-04:00, jgarzik@redhat.com [ia32] add PCI id for VIA irq router Noticed by Sebastian Reichelt. ChangeSet@1.1063.2.50, 2003-08-31 09:33:38-04:00, purna@jcom.home.ne.jp [netdrvr] fix skb_padto bugs introduced when skb_padto was introduced It seems that skb_padto security fixes in 2.4 and 2.5 trying to fix "CAN-2003-0001:Multiple ethernet NID device drivers do not pad frames with null bytes", do not put the skb_padto blocks in proper places in the 3c527, eth16i, fmv18x, seeq8005, yellowfin device drivers. In case a driver calls skb_padto(), it is possible that the space available in the original skb buffer tailroom is less than the space to pad. In this case, in short, the skb_padto() will create a new skb buffer, copy data from the original skb buffer to a new skb buffer, free the original buffer, and finally return the new buffer. If this happens to the aforementioned device drivers, they come to point to wrong data. And, for 3c527 and yellowfin, the drivers can unexpectedly double free the original skb buffers since they still point to the original skb buffers. The attached patch against 2.4.23pre1 fixes these issues. ChangeSet@1.1063.2.49, 2003-08-31 08:48:07-04:00, jgarzik@redhat.com [netdrvr 8139too] remove useless board names The only thing that differentiated most of the entries in the board_info[] table and the board_t type was the vendor branding string for the board. This table is a pain to maintain, so we prefer to simply use "RTL8129" or "RTL8139". ChangeSet@1.1063.2.48, 2003-08-31 08:40:01-04:00, lethal@linux-sh.org [netdrvr 8139too] fix and pci ids needed for SH platform a.k.a. Sega Broadband Adapter. ChangeSet@1.1063.2.47, 2003-08-31 08:17:34-04:00, jgarzik@redhat.com [netdrvr pcmcia] support SIOC[GS]MII{PHY,REG} ioctls Updated drivers; 3c574_cs, axnet_cs, pcnet_cs, xirc2ps_cs Thanks to Komuro for pointing this out. ChangeSet@1.1063.34.1, 2003-08-31 01:20:34-07:00, davem@nuts.ninka.net Merge nuts.ninka.net:/home/davem/src/BK/tg3-2.4 into nuts.ninka.net:/disk1/davem/BK/tg3-2.4 ChangeSet@1.1063.2.46, 2003-08-31 02:06:24-04:00, jgarzik@redhat.com [netdrvr 8139too] make features more persistent; fix PCI DAC mode * only set PCIDAC (64-bit PCI) bit in hardware if sizeof(dma_addr_t) > 32. Need a better test for whether 64-bit mode is _really_ needed. * cache chip command register in private struct. this allows the setting of rx-vlan, rx-csum, and other features to be persistent across the entire lifetime of the net device. * remove dead private struct members frag_skb, dropping_frag, and pci_using_dac. ChangeSet@1.1063.33.3, 2003-08-30 22:39:31-07:00, davem@nuts.ninka.net [SPARC64]: In sysv IPC translation, mask out IPC_64 as appropriate. Based upon a patch from Keith M Wesolowski ChangeSet@1.1063.33.2, 2003-08-30 22:21:31-07:00, davem@nuts.ninka.net [SPARC]: Fix uniprocessor build. ChangeSet@1.1063.2.45, 2003-08-31 00:48:20-04:00, jgarzik@redhat.com [netdrvr 8139cp] stats improvements and fixes * make sure rx_frags is still accounted * query RxMissed register, and clear, upon each get-stats func call ChangeSet@1.1063.32.1, 2003-08-30 20:45:22-07:00, laforge@netfilter.org [NETFILTER]: Fix routing key in ipt_MASQUERADE.c ChangeSet@1.1063.2.44, 2003-08-30 23:34:29-04:00, jgarzik@redhat.com [netdrvr 8139cp] bump version ChangeSet@1.1063.2.43, 2003-08-30 23:23:08-04:00, jgarzik@redhat.com [netdrvr 8139cp] fix NAPI bug; remove board_type distinction, not needed ChangeSet@1.1063.31.1, 2003-08-30 22:53:25-04:00, ink@jurassic.park.msu.ru [PCI] update Memory-Write-Invalidate (MWI) transaction support Sync up PCI MWI code with 2.5: - no changes required for drivers or architectures with HAVE_ARCH_PCI_MWI; - do respect BIOS settings: if the cacheline size is multiple of that we have expected, assume that this is on purpose; - assume cacheline size of 32 bytes for all x86s except K7/K8 and P4. Actually it's good for 386/486s as quite a few PCI devices do not support smaller values. ChangeSet@1.1063.2.42, 2003-08-30 22:44:26-04:00, jgarzik@redhat.com [netdrvr 8139cp] small cleanups * remove netif_queue_stopped test, netif_wake_queue already does that * move vlan stuff to top of file * remove __dev markers * update todo list at top of file * remove pci_set_dma_mask argument casts; ULL suffixes preferred. ChangeSet@1.1063.2.41, 2003-08-30 22:17:19-04:00, jgarzik@redhat.com [netdrvr 8139cp] remove mentions of RTL8169 (now handled by "r8169") ChangeSet@1.1063.2.40, 2003-08-30 22:14:04-04:00, jgarzik@redhat.com [netdrvr 8139cp] update todo list in header ChangeSet@1.1063.2.39, 2003-08-30 22:09:37-04:00, jgarzik@redhat.com [netdrvr 8139cp] support NAPI on RX path; Ditch RX frag handling. NAPI is turned on unconditionally for the RX path. The hardware supports interrupt mitigation, so that should be investigated too. RX fragment handling removed. We simply ensure that we alloc buffers large enough to hold incoming packets. Any stray RX frags that occur (shouldn't be any) will be dropped. ChangeSet@1.1063.2.38, 2003-08-30 21:01:59-04:00, jgarzik@redhat.com [netdrvr 8139cp] build TX checksumming code, but default OFF (previously it was ifdef'd) Also, bump version to 1.0. ChangeSet@1.1063.2.37, 2003-08-30 19:56:01-03:00, johnstul@us.ibm.com [PATCH] Convert /proc/interrupts to use seq_file Marcelo, All, This patch converts /proc/interrupts to use the seq_file interface on i386 (thanks to Randy for adding the seq "single" code). This prevents memory corruption caused by /proc/interrupts overflowing its page sized buffer on systems with high cpu and interrupt counts. Please consider for inclusion. thanks -john ChangeSet@1.1063.2.36, 2003-08-30 18:53:48-03:00, marcelo@logos.cnet Updated my contact information ChangeSet@1.1063.2.35, 2003-08-30 16:20:20-03:00, akpm@osdl.org [PATCH] inodes_stat.nr_inodes race fix Small fix to dispose_list() to ensure that inodes_stat.nr_inodes is protected by inode_lock. fs/inode.c | 19 +++++++++++-------- 1 files changed, 11 insertions(+), 8 deletions(-) ChangeSet@1.1063.26.32, 2003-08-30 14:23:17-03:00, ntfs@flatcap.org [PATCH] Fix NTFS build warnings Hi Marcelo, This patch: Fixes all NTFS build warnings Fixes some broken macros (debug build only) Gerardo Exequiel Pozzi sent you a patch to fix the NTFS build warnings. I have tested the patch and fixed a couple of broken macros he missed. Anton Altaparmakov has given the patch his seal of approval :-) Please can you apply the patch to 2.4.22-rc4 Cheers, FlatCap - Richard Russon ntfs@flatcap.org ChangeSet@1.1063.26.31, 2003-08-30 11:53:58-03:00, marcelo@logos.cnet Changed EXTRAVERSION to -pre2 TAG: v2.4.23-pre2 ChangeSet@1.1063.26.30, 2003-08-30 11:49:43-03:00, elenstev@mesatop.com [PATCH] Add six more Configure.help texts from the -ac tree Hi Marcelo, Here are another six Configure.help texts which have been sitting in Alan's tree for a while. Like the others, these texts originated in the development tree. The order in the -ac tree has be retained for easier re-merging. These six texts appear to be the last which are in the -ac tree and which are currently needed by mainline 2.4. Please apply. This patch was made against your current (this morning) bk tree. Steven CONFIG_BLK_DEV_TRIFLEX CONFIG_BLK_DEV_SC1200 CONFIG_BLK_DEV_PDC202XX_NEW CONFIG_MTD_CONCAT CONFIG_HP_SDC CONFIG_PDC_CONSOLE ChangeSet@1.1063.26.29, 2003-08-30 11:45:36-03:00, m.c.p@wolk-project.de [PATCH] DRM menu the right fix thanks for merging the DRM menu cleanup patch. But now looking at it, let's do it a different and the most right way. Delete the extra DRM-Config.in file and move it back to were it was again before the cleanup but with a mainmenu_option. This looks better and has the same effect, but w/o introducing another new file. ChangeSet@1.1063.26.27, 2003-08-30 11:37:30-03:00, lethal@linux-sh.org [PATCH] Add Paul Mundt to CREDITS Here's a quick patch to add myself to CREDITS, please apply. ChangeSet@1.1063.26.26, 2003-08-30 11:37:02-03:00, rddunlap@osdl.org [PATCH] add seq_file "single" interfaces Hi, This patch adds the seq_file "single" interfaces from 2.6.0-test4 to 2.4.22++. This will enable larger /proc/interrupts and /proc/mdstat, which currently have some oopsing problems with large outputs. Please apply. -- ~Randy patch_name: seq_single_2423p1.patch patch_version: 2003-08-25.09:34:59 author: Randy.Dunlap description: add seq_file "single" interfaces from 2.6.0-test4 product: Linux product_versions: 2422 diffstat: = fs/seq_file.c | 90 +++++++++++++++++++++++++++++++++++++++++++++-- include/linux/seq_file.h | 13 ++++++ 2 files changed, 100 insertions(+), 3 deletions(-) ChangeSet@1.1063.26.25, 2003-08-30 11:35:21-03:00, akpm@osdl.org [PATCH] tty oops fix Ancient bug, reported by Hiroshi Inoue : 1. login to tty2 (not tty1) 2. start kon (Kanji cONsole emulator, console which support Japanese characters) 3. exit kon 4. logout It oopses in the debugging function check_tty_count(), walking a list_head which has been list_del()'d. Call trace is: #0 check_tty_count (tty=0x10d42000, routine=0xc817b00 ".paths") at include/asm/processor.h:583 #1 0x022c6c00 in do_tty_hangup (data=0x10d42000) at drivers/char/tty_io.c:426 #2 0x022c6f60 in tty_vhangup (tty=0xc817b00) at drivers/char/tty_io.c:536 #3 0x022c6fcc in disassociate_ctty (on_exit=1) at drivers/char/tty_io.c:574 #4 0x02127aee in do_exit (code=0) at kernel/exit.c:718 #5 0x02127caa in do_group_exit (exit_code=0) at kernel/exit.c:796 #6 0x02127cbc in sys_exit_group (error_code=0) at kernel/exit.c:807 The tty refcount is zero, so everything seems consistent and sensible. The fix just uses list_del_init() on that list_head. drivers/char/tty_io.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) ChangeSet@1.1063.26.24, 2003-08-30 11:34:42-03:00, akpm@osdl.org [PATCH] fix possible busywait in rtc_read() If two processes are waiting in rtc_read(), only one will get the data. The other will madly spin around the loop in state TASK_RUNNING until another interrupt happens. Fix it by resetting TASK_INTERRUPTIBLE _inside_ the retry loop. drivers/char/rtc.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) ChangeSet@1.1063.26.23, 2003-08-30 11:32:03-03:00, willy@w.ods.org [PATCH] Fix log buffer length issues On Fri, Aug 29, 2003 at 01:14:19PM -0300, Marcelo Tosatti wrote: > I'm waiting for your selectable log buffer lengh fixes :) OK, Marcelo, here comes my proposed fix. Tom, does it look better for you ? BTW, Tom, I found the exact reason for the line which irritated you : #if !defined(CONFIG_LOG_BUF_SHIFT) || (CONFIG_LOG_BUF_SHIFT - 0 == 0) it was because because I was worried about CONFIG_LOG_BUG_SHIFT being defined to the empty string (a simple #define CONFIG_LOG_BUG_SHIFT is enough), as I remember it happened to me when working on it. But there's no reason here for this to happen, otherwise we would have far worse problems with values extracted from .config, so I fixed it. I added the config line to all archs. I used copy paste, so either I broke them all, either none. I'll start a compilation on my alpha while I'm on the way home, to check. I you have questions, please do ask. Cheers, Willy ChangeSet@1.1063.26.22, 2003-08-30 11:22:54-03:00, geert@linux-m68k.org [PATCH] Sonic Ethernet unsafe interrupt Sonic Ethernet: Use the interrupt safe version of dev_kfree_skb{_irq} within the interrupt handler (from Michael Müller) ChangeSet@1.1063.26.21, 2003-08-30 11:22:15-03:00, geert@linux-m68k.org [PATCH] lmc_proto.c includes lmc_proto.c includes , causing build failures on UMP-only architectures ChangeSet@1.1063.26.20, 2003-08-30 11:21:33-03:00, mike.miller@hp.com [PATCH] cciss multi-path failover in md Marcelo, Here's an updated failover patch created from your BK tree. I downloaded the tree this morning. I'm sorry it took me so long to get back to you. I've been working some HP related issues over the last few days. Please consider this for inclusion. Patch description: Enables multi-path failover support in md environments. Changes: 1. Sends a no-op command down to the controller and expects it to return in a very short time. 2. Adds the ability to extend the time in which the command can return. 3. Monitors the heartbeat register on the controller to cover the special case of a PCI Master Abort. Thanks, mikem ChangeSet@1.1063.26.19, 2003-08-30 11:20:44-03:00, geert@linux-m68k.org [PATCH] dmasound core fixes Dmasound core fixes from Christoph Hellwig in 2.6.0: - Some exported symbols are declared __init - in the modular case this is freed before the other modules can call it.. ChangeSet@1.1063.26.18, 2003-08-30 11:04:23-03:00, geert@linux-m68k.org [PATCH] M68k FPU emulator M68k FPU emulator: Add fgetman, fgetexp and fsqrt (from Michael Müller and Roman Zippel) ChangeSet@1.1063.26.17, 2003-08-30 11:03:50-03:00, geert@linux-m68k.org [PATCH] dmasound SOUND_PCM_READ_RATE dmasound: Implement missing SOUND_PCM_READ_RATE ioctl (from Richard Zidlicky) ChangeSet@1.1063.26.16, 2003-08-30 11:02:51-03:00, geert@linux-m68k.org [PATCH] M68k cpu_relax() M68k: Make cpu_relax() compatible with 2.6. ChangeSet@1.1063.26.15, 2003-08-30 11:01:56-03:00, geert@linux-m68k.org [PATCH] Dmasound invalid vs. illegal Dmasound: Use `invalid' instead of `illegal' (from 2.5.x) ChangeSet@1.1063.26.14, 2003-08-30 11:00:51-03:00, geert@linux-m68k.org [PATCH] M68k invalid vs. illegal M68k: Use `invalid' instead of `illegal' (from Steven Cole in 2.5.x) ChangeSet@1.1063.26.13, 2003-08-30 10:59:56-03:00, geert@linux-m68k.org [PATCH] M68k free_io_area() M68k: Take the gap into account in free_io_area() (from Michael Müller) ChangeSet@1.1063.26.12, 2003-08-30 10:58:54-03:00, geert@linux-m68k.org [PATCH] M68k mm cleanup M68k: Kill superfluous includes and obsolete commented-out code in mm code. ChangeSet@1.1063.26.11, 2003-08-30 10:58:36-03:00, geert@linux-m68k.org [PATCH] Rename ariadne2 to zorro8390 Rename ariadne2 to zorro8390, since this driver supports several NS8390-based Zorro Ethernet cards. Kill all never used occurrencies of CONFIG_NE2K_ZORRO. ChangeSet@1.1063.26.10, 2003-08-30 10:56:49-03:00, geert@linux-m68k.org [PATCH] M68k RTC updates M68k: Use genrtc on all m68k platforms ChangeSet@1.1063.26.9, 2003-08-30 10:55:16-03:00, geert@linux-m68k.org [PATCH] fb_cmap and transparency If a colormap contains no transparency information, fb_set_cmap() calls fb_setcolreg() with trans = 0. This causes all CLUT entries to be fully transparent on hardware that does have transparency information in the CLUT registers. The following patch solves this problem by changing the default transparency from 0 (full transparent) to 0xffff (full opaque). ChangeSet@1.1063.26.8, 2003-08-30 10:52:24-03:00, geert@linux-m68k.org [PATCH] Isapnp warning Isapnp: Kill warning if CONFIG_PCI is not set, like it's done in 2.6.0. ChangeSet@1.1063.26.7, 2003-08-30 10:51:06-03:00, geert@linux-m68k.org [PATCH] M68k ptrace M68k: Use ptrace_check_attach instead of doing the checks ourselves (from Andreas Schwab). ChangeSet@1.1063.26.5, 2003-08-30 10:40:43-03:00, thomas@winischhofer.net [PATCH] sisfb update Marcelo, attached is an update for the sis framebuffer driver. As usual, heavily tested. Reduces source size by about 5000 lines... Please apply. Thomas ChangeSet@1.1063.26.4, 2003-08-30 10:39:17-03:00, andrea@suse.de [PATCH] vmalloc allocations in ipc needs smp initialized Hi Marcelo, I submit you this patch for 2.4.23 merging. In short if you change SEMMNI to 8192 the kernel will crash at boot, beause it tries to call vmalloc before the smp is initialized. The reason is that vmalloc calls into the pte alloc code, and the fast pte alloc is tried first, but that reads into the pte_quicklist, that requires the cpu_data to be initialized (and that happens in smp_init()). the patch is obviously safe, since no piece of kernel (especially the code in the check_bugs and smp_init paths ;) calls into the ipc subsystem. The reason this started to trigger wasn't really that we increased SEMMNI, but what happend is that some IPC data structure grown, and for some reason the corruption due the uninitalized pte_quicklist triggers only for smp boxes with less than 1G (not very common anymore ;). So it wasn't immediatly reproducible on all setups. 2.6 doesn't suffer from the same problem, simply because 2.6 isn't using the quicklist anymore, but I think it would be much more correct to make the same change in 2.6 too, since whatever cond_resched() in the vm paths (and they're definitely allowed to call it), will lead to a crash since the init task isn't initialized and the scheduler can't be invoked yet. (and 2.6 already has the bigger data structures that should trigger the vmalloc all the time on all setups) ChangeSet@1.1063.30.40, 2003-08-30 10:19:18-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: resend - mm checks have precedence bugs ChangeSet@1.1063.30.39, 2003-08-30 10:09:02-03:00, marcelo@logos.cnet add sysctl bits for setuid core ChangeSet@1.1063.30.38, 2003-08-30 09:55:31-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: Fix the file sharing/initrd bug ChangeSet@1.1063.30.37, 2003-08-30 09:55:07-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: Add headers for wolfson codecs ChangeSet@1.1063.30.36, 2003-08-30 09:54:44-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: add sysctl number for setuid core ChangeSet@1.1063.30.35, 2003-08-30 09:50:13-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: allow setuid core dumps In some situations you want to get dumps of setuid apps, either because you administer the system or because you have a secure area to collect them all (remember core path is configurable too). If the admin enables them this patch dumps setuid core files as root owned files in the nominated location. The default behaviour remains that users dont get core of setuid files. This has been in -ac for a while and is from RH work done at the request of various large users who collect all their cores in a dump dir ChangeSet@1.1063.30.34, 2003-08-30 09:47:38-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: tdfxfb updates for 24/32 and big endian ChangeSet@1.1063.30.33, 2003-08-30 09:46:49-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: VGA also works on IA64 ChangeSet@1.1063.30.32, 2003-08-30 09:46:20-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: USB audio fixes for OSS API compliance ChangeSet@1.1063.30.31, 2003-08-30 09:45:26-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: makefile/config update for sound changes ChangeSet@1.1063.30.30, 2003-08-30 09:45:18-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: fix i810 audio leak ChangeSet@1.1063.30.29, 2003-08-30 09:42:16-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: cmpci update ChangeSet@1.1063.30.28, 2003-08-30 09:41:54-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: ALi5455 update ChangeSet@1.1063.30.27, 2003-08-30 09:41:20-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: ad1889 error handling fixes ChangeSet@1.1063.30.26, 2003-08-30 09:40:37-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: wolfson ac97 touchscreen driver ChangeSet@1.1063.30.25, 2003-08-30 09:39:35-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: aacraid update ChangeSet@1.1063.30.24, 2003-08-30 09:38:40-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: update cpia driver to fix warnings ChangeSet@1.1063.30.23, 2003-08-30 09:37:46-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: Add disk hotplug to the IDE core IFF your hardware supports it you can now do hdparm -b0 /dev/hda remove device replace with different device hdparm -b1 /dev/hda This has been tested a fair bit with thinkpad laptops but not extensively hammered with other controllers supporting the facility (some HPT/SI cards). Also note that in general even the hardware supporting it does not support hotplug in a master/slave setup - that is a hardware limit. ChangeSet@1.1063.30.22, 2003-08-30 09:30:31-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: split ide probe code up Lots of probe internals are needed for hotplug too so we split it into a set of meaningful routines and export the right bits ChangeSet@1.1063.30.21, 2003-08-30 09:29:51-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: fix cable detect issue with sata ChangeSet@1.1063.30.20, 2003-08-30 09:13:21-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: update ide headers for hotplug ChangeSet@1.1063.30.19, 2003-08-30 09:10:29-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: update ide raid for info pointer changes ChangeSet@1.1063.30.18, 2003-08-30 08:58:15-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: siimage: set a sata flag on the hwif so we can do cable det ChangeSet@1.1063.30.17, 2003-08-30 08:57:01-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: Add Intel ICH3 hotplug support This covers quite a few laptops ChangeSet@1.1063.30.16, 2003-08-30 08:55:38-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: move sibyte driver into the right dir ChangeSet@1.1063.30.15, 2003-08-30 08:52:47-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: add open/close methods to ide-default for hotplug ChangeSet@1.1063.30.14, 2003-08-30 08:51:50-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: fix missing formatting info in ide-cd ChangeSet@1.1063.30.13, 2003-08-30 08:50:45-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: fix nowayout handling on softdog ChangeSet@1.1063.30.12, 2003-08-30 08:49:35-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: warning fix ChangeSet@1.1063.30.11, 2003-08-30 08:48:46-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: fix a missing rocket card ChangeSet@1.1063.30.10, 2003-08-30 00:38:23-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: improved extra key bounce fix ChangeSet@1.1063.30.9, 2003-08-30 00:36:46-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: update i8xx watchdog ChangeSet@1.1063.30.8, 2003-08-30 00:32:59-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: amd watchdog update ChangeSet@1.1063.30.7, 2003-08-30 00:30:41-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: wolfson touchscreen docs #ra1 ChangeSet@1.1063.30.6, 2003-08-30 00:29:59-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: update INDEX for docs ChangeSet@1.1063.30.5, 2003-08-30 00:29:40-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: replace the pci router logic with working code This is more verbose but actually uses less memory as its all __init unlike the big table. The table also fails because there are controllers that have dependancies for routing on either additional components (440GX) or version (SiS) which it could not handle. ChangeSet@1.1063.30.4, 2003-08-30 00:10:18-03:00, alan@lxorguk.ukuu.org.uk [PATCH] PATCH: remove all the 440gx broken bios stuff There is a better way to fix this and the SiS and other bits in the PCI layer. That fix means this bit can go ChangeSet@1.1063.30.3, 2003-08-29 23:18:59-03:00, cw@sgi.com [PATCH] SGI SN Serial/Console Driver Marcelo, As discussed here is the serial & console driver required for Altix systems. Documentation/Configure.help | 11 drivers/char/Config.in | 6 drivers/char/Makefile | 1 drivers/char/sn_serial.c | 1190 +++++++++++++++++++++++++++++++++ drivers/char/tty_io.c | 7 5 files changed, 1215 insertions(+) # This is a BitKeeper generated patch for the following project: # Project Name: Linux kernel tree # This patch format is intended for GNU patch command version 2.5 or higher. # This patch includes the following deltas: # ChangeSet 1.1086 -> 1.1087 # drivers/char/Config.in 1.50 -> 1.51 # drivers/char/tty_io.c 1.26 -> 1.27 # Documentation/Configure.help 1.182 -> 1.183 # drivers/char/Makefile 1.35 -> 1.36 # (new) -> 1.2 drivers/char/sn_serial.c # # The following is the BitKeeper ChangeSet Log # -------------------------------------------- # 03/08/27 cw@tomahawk.engr.sgi.com 1.1087 # Add SGI l1 Serial/Console Driver # -------------------------------------------- # ChangeSet@1.1104, 2003-08-29 13:29:44-06:00, bjorn.helgaas@hp.com ia64: Remove AIC7XXX driver from ski defconfig. ChangeSet@1.1063.30.2, 2003-08-29 13:27:42-03:00, marcelo@logos.cnet Delete unused drivers/scsi/aic79xx (now aic7xxx supports it) .del-scsi_message.h~702d04df39cec421: Delete: drivers/scsi/aic79xx/scsi_message.h .del-scsi_iu.h~19dfb28b403d5076: Delete: drivers/scsi/aic79xx/scsi_iu.h .del-queue.h~22333ebc4752ba27: Delete: drivers/scsi/aic79xx/queue.h .del-cam.h~142be79aa2e7b6e4: Delete: drivers/scsi/aic79xx/cam.h .del-aiclib.h~91a202de5ec07b45: Delete: drivers/scsi/aic79xx/aiclib.h .del-aiclib.c~4f9c5e58fe29475f: Delete: drivers/scsi/aic79xx/aiclib.c .del-aic79xx_seq.h~f087a47d39ec35d1: Delete: drivers/scsi/aic79xx/aic79xx_seq.h .del-aic79xx_reg_print.c~4602f821bb30aa98: Delete: drivers/scsi/aic79xx/aic79xx_reg_print.c .del-aic79xx_reg.h~2a5d1aeafe35d774: Delete: drivers/scsi/aic79xx/aic79xx_reg.h .del-aic79xx_proc.c~65ba7c0cbcae9e17: Delete: drivers/scsi/aic79xx/aic79xx_proc.c .del-aic79xx_pci.c~f3be49004874d74d: Delete: drivers/scsi/aic79xx/aic79xx_pci.c .del-aic79xx_osm_pci.c~ce497673678cc8cb: Delete: drivers/scsi/aic79xx/aic79xx_osm_pci.c .del-aic79xx_osm.h~a9993aeb34f68385: Delete: drivers/scsi/aic79xx/aic79xx_osm.h .del-aic79xx_osm.c~c85edf4627f2d3b8: Delete: drivers/scsi/aic79xx/aic79xx_osm.c .del-aic79xx_inline.h~4fe599bddc047da0: Delete: drivers/scsi/aic79xx/aic79xx_inline.h .del-aic79xx_core.c~3acdfe0b216b89b2: Delete: drivers/scsi/aic79xx/aic79xx_core.c .del-aic79xx.seq~1fc373c5c245fa53: Delete: drivers/scsi/aic79xx/aic79xx.seq .del-aic79xx.reg~c7d1afd63bee29d: Delete: drivers/scsi/aic79xx/aic79xx.reg .del-aic79xx.h~f52414d5ceae1d0c: Delete: drivers/scsi/aic79xx/aic79xx.h .del-Makefile~170d5d2f4fc1242d: Delete: drivers/scsi/aic79xx/Makefile .del-Config.in~5417c5049523b240: Delete: drivers/scsi/aic79xx/Config.in .del-CHANGELOG~d01c3638ccfa2574: Delete: drivers/scsi/aic79xx/CHANGELOG ChangeSet@1.1063.30.1, 2003-08-29 13:17:06-03:00, masanari.iida@hp.com [PATCH] SCSI blacklist HP Va7140 As Stefan Schindewolf reported on 25/Aug 2003 on Kernel mailing list, scsi_scan.c has missing entory and also typo. Apply this patch on 2.4.23-preX. Regards, masanari ChangeSet@1.1063.29.6, 2003-08-29 21:11:09+10:00, paulus@samba.org PPC32: Implement semtimedop system call. ChangeSet@1.1063.29.5, 2003-08-29 20:56:54+10:00, paulus@samba.org PPC32: Restructure signal code, new ucontext structure, add swapcontext syscall. The ucontext struct now corresponds with glibc's new ucontext_t. The old ucontext didn't actually have room for all the registers, and it had to be changed to allow setcontext et al. to be implemented. This means that the stack layout for real-time signal delivery has changed slightly. Existing applications using the uc_mcontext.regs pointer will still work since the new uc_regs field is in the same position and points to the start of the mcontext, where the registers are saved. We now save the vector registers on the stack on signal delivery if the process has ever used any altivec instructions (since the last exec). Swapcontext is implemented as a system call because it isn't possible to set all three of LR, CTR and the instruction pointer to arbitrary values at user level, and because the kernel knows whether the FP and vector registers are owned by this process and can optimize in the case that they aren't. With the swapcontext system call we can implement setcontext, getcontext and swapcontext. ChangeSet@1.1063.29.4, 2003-08-29 20:37:52+10:00, paulus@samba.org PPC32: Minor cleanups and fixes for 4xx/BookE systems ChangeSet@1.1063.29.3, 2003-08-29 19:08:02+10:00, paulus@samba.org PPC32: Add support for the PPC970 processor. ChangeSet@1.1063.29.2, 2003-08-29 17:22:05+10:00, paulus@samba.org PPC32: Use CONFIG_IBM_OPENBIOS instead of CONFIG_TREEBOOT ChangeSet@1.1063.24.5, 2003-08-29 14:31:33+10:00, mporter@kernel.crashing.org PPC32: Add support for DMA controllers on PPC 4xx processors. ChangeSet@1.1063.28.1, 2003-08-28 22:20:32-04:00, jgarzik@redhat.com [scsi] add SCSI opcodes and SAM status codes to scsi/scsi.h Added: report luns, read/write 16, read capacity 16, and the SAM_STAT_xxx set of constants. ChangeSet@1.1103, 2003-08-28 18:55:36-06:00, tony.luck@intel.com ia64: Trim granules correctly in efi_memmap_walk() This was a fun one to track down. I was trying to provide a fake SRAT table so that I could start working on some more ccNUMA issues on a non-NUMA tiger box, but my kernel kept dying while running scripts out of /etc/init.d/* I found I was getting an MCA while executing an lfetch.excl in clear_page_tables(), the address that was being fetched was 0xe0000000000a00e8, which is non-cacheable VGA memory on Tiger. Root cause was the "trim" code in efi_memmap_walk() had failed to trim away all the blocks of memory in the bottom granule when it had seen the memory hole there, so the kernel had managed to allocate a page at 0x9c000 as a page table, and when the process using it ended, the cleanup code prefetched off the end of the page, and into the VGA memory, causing an MCA. The problem is that efi_memmap_walk() repeatedly trims the same entry, instead of trimming each of the entries in the granule. ChangeSet@1.1102, 2003-08-28 18:47:46-06:00, bjorn.helgaas@hp.com [PATCH] ia64: fix SAVE_RESET so OS INIT handler works again The syscall optimization patches broke the OS INIT handler because SAVE_RESET was addressing relative to r12, which contains the virtual address of the stack pointer. Fixed by addressing relative to r2/r3 instead. ChangeSet@1.1101, 2003-08-28 18:00:17-06:00, bjorn.helgaas@hp.com ia64: Fix minstate comments. ChangeSet@1.1100, 2003-08-28 17:08:38-06:00, bjorn.helgaas@hp.com ia64: minstate.h: whitespace changes to reduce diffs with 2.5. ChangeSet@1.1099, 2003-08-28 15:01:10-06:00, bjorn.helgaas@hp.com ia64: TRIVIAL: Remove extraneous '`'. ChangeSet@1.1063.21.9, 2003-08-28 13:50:03-07:00, russell_d_cagle@mindspring.com [PATCH] USB: add Garmin iQue support to visor driver ChangeSet@1.1063.25.4, 2003-08-28 13:09:28-07:00, greg@kroah.com [PATCH] PCI: add PCI_DEVICE_CLASS() macro to match PCI_DEVICE() macro. ChangeSet@1.1063.25.3, 2003-08-28 13:09:20-07:00, greg@kroah.com [PATCH] PCI: add PCI_DEVICE() macro to make pci_device_id tables easier to read. ChangeSet@1.1063.2.32, 2003-08-28 16:14:21-03:00, cw@sgi.com [PATCH] Add SGI IOC4 IDE Driver Marcelo, As discussed here is patch against you BK tree to add the SGI IOC4 IDE driver will all complaints addressed! Documentation/Configure.help | 6 MAINTAINERS | 5 drivers/ide/Config.in | 3 drivers/ide/pci/Makefile | 1 drivers/ide/pci/sgiioc4.c | 891 +++++++++++++++++++++++++++++++++ drivers/ide/pci/sgiioc4.h | 177 ++++++ include/linux/pci_ids.h | 1 7 files changed, 1084 insertions(+) *Please* apply this :) # This is a BitKeeper generated patch for the following project: # Project Name: Linux kernel tree # This patch format is intended for GNU patch command version 2.5 or higher. # This patch includes the following deltas: # ChangeSet 1.1085 -> 1.1086 # drivers/ide/pci/Makefile 1.4 -> 1.5 # include/linux/pci_ids.h 1.69 -> 1.70 # MAINTAINERS 1.105 -> 1.106 # drivers/ide/Config.in 1.35 -> 1.36 # Documentation/Configure.help 1.181 -> 1.182 # (new) -> 1.1 drivers/ide/pci/sgiioc4.c # (new) -> 1.1 drivers/ide/pci/sgiioc4.h # # The following is the BitKeeper ChangeSet Log # -------------------------------------------- # 03/08/27 cw@tomahawk.engr.sgi.com 1.1086 # Add IOC4 IDE driver for SGI Altix Systems # -------------------------------------------- # ChangeSet@1.1063.2.31, 2003-08-28 16:07:55-03:00, marcelo@logos.cnet Fix compilation warning in panic.c ChangeSet@1.1063.2.30, 2003-08-28 15:59:11-03:00, marcelo@logos.cnet Changed hch contact information ChangeSet@1.1063.2.29, 2003-08-28 15:57:38-03:00, elenstev@mesatop.com [PATCH] Add 39 Configure.help texts from -ac tree. Hi Marcelo, This patch cleans up some old information at the top of the Configure.help file, and adds help tests for the following 39 options. The help texts originally came from the development kernel, and were ported to 2.4 form by myself and others. A few help texts containing copy and paste errors in 2.6 were corrected. These help texts were accepted by Alan into his -ac tree. I retained their position from Alan's Configure.help file to minimize difficulty in merging for Bernhard R. in maintaining the (p)ac tree when you merge these into your tree (I hope). Please apply this. The patch is made against your current 2.4-bk tree. There are a few more of these, but this should be enough for now. Thanks, Steven CONFIG_IP6_NF_MATCH_AHESP CONFIG_IP6_NF_MATCH_RT CONFIG_IP6_NF_MATCH_OPTS CONFIG_IP6_NF_MATCH_FRAG CONFIG_IP6_NF_MATCH_HL CONFIG_IP6_NF_MATCH_IPV6HEADER CONFIG_FUSION_MAX_SGE CONFIG_QDIO CONFIG_QDIO_PERF_STATS CONFIG_IEEE1394_OUI_DB CONFIG_MTD_REDWOOD CONFIG_MTD_NAND_AUTCPU12 CONFIG_MTD_NAND_EDB7312 CONFIG_I2C_KEYWEST CONFIG_DRM_I830 CONFIG_HISAX_ENTERNOW_PCI CONFIG_CPU_ICACHE_DISABLE CONFIG_CPU_DCACHE_DISABLE CONFIG_CPU_DCACHE_WRITETHROUGH CONFIG_CPU_CACHE_ROUND_ROBIN CONFIG_CPU_BPREDICT_DISABLE CONFIG_ZBOOT_ROM CONFIG_ZBOOT_ROM_TEXT CONFIG_ZBOOT_ROM_BSS CONFIG_USB_BRLVGER CONFIG_USB_KBTAB CONFIG_USB_SERIAL_EDGEPORT_TI CONFIG_USB_SERIAL_KEYSPAN_MPR CONFIG_CHECKING CONFIG_CHASSIS_LCD_LED CONFIG_GSC CONFIG_GSC_WAX CONFIG_GSC_DINO CONFIG_HPET_TIMER CONFIG_GART_IOMMU CONFIG_INIT_DEBUG CONFIG_IOMMU_DEBUG CONFIG_IOMMU_LEAK CONFIG_HVC_CONSOLE ChangeSet@1.1063.25.2, 2003-08-28 11:27:38-07:00, vmlinuz386@yahoo.com.ar [PATCH] PCI Hotplug: fix __FUNCTION__ warnings this patch fix the warning: concatenation of string literals with __FUNCTION__ is deprecated ChangeSet@1.1063.25.1, 2003-08-28 11:23:37-07:00, greg@kroah.com [PATCH] PCI hotplug: fix up a bunch of copyrights that were incorrectly declared. It needs to be "Copyright (C)" not "Copyright (c)" according to the lawyers who know these things... ChangeSet@1.1063.21.8, 2003-08-28 11:14:28-07:00, greg@kroah.com [PATCH] USB: fix up a bunch of copyrights that were incorrectly declared. It needs to be "Copyright (C)" not "Copyright (c)" according to the lawyers who know these things... ChangeSet@1.1063.21.7, 2003-08-28 11:05:57-07:00, Stefan.Becker@nokia.com [PATCH] USB: acm.c update for new devices This support is already in 2.6 ChangeSet@1.1063.21.6, 2003-08-28 11:05:50-07:00, ddstreet@ieee.org [PATCH] USB: backport usbfs 'disconnect' here is a backport of the usbfs 'disconnect' functionality. The only difference I know of from the 2.5 code is this doesn't use the kernel lock. I'm not sure if that is really needed? Anyway it would be great to finally get this into 2.4... ChangeSet@1.1063.21.5, 2003-08-28 11:05:43-07:00, abbotti@mev.co.uk [PATCH] USB: ftdi_sio - tidy up write bulk callback On Tuesday 19 Aug 2003 21:57, Greg KH wrote: > On Tue, Aug 19, 2003 at 05:14:28PM +0100, Ian Abbott wrote: > > The attached patch (usb-2.4-ftdi_sio-wbcbfix.patch) reinstates > > the port->open_count check in the ftdi_write_bulk_callback > > function, and should be applied after my earlier patch > > 'usb-2.4-ftdi_sio-wbcleanup.patch'. > > So you're just really getting rid of the "serial" pointer check, > right? That's fine, but can you send me just one patch that does > this instead? The attached patch combines the changes, but apart from a tiny reduction in code size, it doesn't achieve anything, but doesn't seem to cause any harm either. I'll leave it to your discretion whether or not to apply it. As I mentioned previously, leaving the port->open_count check in (to only schedule a soft interrupt when the port is open) seems to prevent an oops, but I suppose it just makes the oops less likely to happen rather than eliminate it entirely. ChangeSet@1.1098, 2003-08-28 11:54:35-06:00, bjorn.helgaas@hp.com ia64: Clarify ACPI available_cpus handling. ChangeSet@1.1063.21.4, 2003-08-28 10:52:44-07:00, stern@rowland.harvard.edu [PATCH] USB: Another unusual_devs.h entry update This information was provided by Anthony Arkles . Please apply to both 2.4 and 2.6. ChangeSet@1.1063.24.4, 2003-08-28 22:40:33+10:00, paulus@samba.org PPC32: Fix typo in arch/ppc/Makefile ChangeSet@1.1063.24.3, 2003-08-28 22:31:07+10:00, mporter@kernel.crashing.org PPC32: Add support for the IBM PPC 440 family of processors. This adds support for the 440GP and 440GX, which are the first "Book E" processors from IBM. This also adds support for the "Ebony" (440GP) and "Ocotea" (440GX) reference boards. ChangeSet@1.1063.24.2, 2003-08-28 21:35:03+10:00, paulus@samba.org PPC32: Add the infrastructure to allow for 64-bit PTEs. The IBM PPC440 family allow a 36-bit physical address and this will let us support them. ChangeSet@1.1063.24.1, 2003-08-28 20:51:53+10:00, paulus@samba.org PPC32: Add support for DMA mapping on non-cache-coherent machines. ChangeSet@1.1063.12.50, 2003-08-28 00:23:40-07:00, xose@wanadoo.es [TG3]: More missing PCI ids. ChangeSet@1.1063.23.3, 2003-08-28 00:12:38-07:00, rob@osinvestor.com [SPARC]: gcc-3.3 compile fixes, part 3. ChangeSet@1.1063.23.2, 2003-08-28 00:11:25-07:00, rob@osinvestor.com [SPARC]: gcc-3.3 compile fixes, part 2. ChangeSet@1.1063.23.1, 2003-08-28 00:11:05-07:00, rob@osinvestor.com [SPARC]: gcc-3.3 compile fixes, part 1. ChangeSet@1.1063.12.49, 2003-08-27 23:37:21-07:00, jgarzik@redhat.com [TG3]: Remove pci-set-dma-mask casts. ChangeSet@1.1097, 2003-08-27 18:37:31-06:00, mort@wildopensource.com [PATCH] ia64: max user stack size of main thread configurable via RLIMIT_STACK Make the size of the user stack based on the stack rlimit. The stack hard stack size now defaults to 2GB, but can be increased with ulimit up to 1/2 of the max mappable space in a region. For 16k pages, this makes the max stack size 8TB. ChangeSet@1.1096, 2003-08-27 18:15:05-06:00, steiner@sgi.com ia64: add support for non-identity mapped kernels Attached is a patch that makes arch/ia64/kernel/head.S work on the SGI SN system. The patch consists mostly of a backport of portions of the same file from 2.6. I tried to minimized differences between the new file & 2.6 as opposed to minimizing differences between the new file & 2.4.21. This makes the patch somewhat larger than needed but (I think) is correct for the long term. The changes add support for non-identity mapped kernels. This patch does NOT change the kernel load address - it just adds support for the case where it was loaded into non-identity mapped memory. ChangeSet@1.1063.22.16, 2003-08-27 18:03:18-04:00, willy@debian.org [netdrvr 8139too] ethtool_ops support ChangeSet@1.1063.22.15, 2003-08-27 17:55:34-04:00, willy@debian.org [ethtool] fix ethtool_get_strings counting bug ChangeSet@1.1063.22.14, 2003-08-27 17:43:01-04:00, shmulik.hen@intel.com [netdrvr bonding] Enhance netdev notification handling. Also, add comment block and bump version. ChangeSet@1.1063.22.13, 2003-08-27 17:42:26-04:00, shmulik.hen@intel.com [netdrvr bonding] Consolidate /proc code, add CHANGENAME handler ChangeSet@1.1063.22.12, 2003-08-27 17:41:56-04:00, shmulik.hen@intel.com [netdrvr bonding] support for changing MAC addr, MTU in ALB/TLB modes ChangeSet@1.1063.22.11, 2003-08-27 17:41:18-04:00, shmulik.hen@intel.com [netdrvr bonding] support for changing HW address and MTU ChangeSet@1.1063.22.10, 2003-08-27 17:40:43-04:00, shmulik.hen@intel.com [netdrvr bonding] Decouple promiscuous handling from the multicast mode setting. ChangeSet@1.1063.22.9, 2003-08-27 17:40:13-04:00, shmulik.hen@intel.com [netdrvr bonding] Modes that don't use primary don't use the new prop. code ChangeSet@1.1063.22.8, 2003-08-27 17:39:31-04:00, shmulik.hen@intel.com [netdrvr bonding] Change monitoring function to use new slave setting propagation ChangeSet@1.1063.22.7, 2003-08-27 17:38:40-04:00, shmulik.hen@intel.com [netdrvr bonding] update slave setting propagation Distinguish between modes that use a primary slave from those that don't, and propagate settings accordingly; Consolidate change_active opeartions and add reselect_active and find_best opeartions. ChangeSet@1.1063.22.6, 2003-08-27 17:31:35-04:00, shmulik.hen@intel.com [netdrvr bonding] add another ifenslave.c include ChangeSet@1.1063.22.5, 2003-08-27 17:31:05-04:00, shmulik.hen@intel.com [netdrvr bonding] update credits/version ChangeSet@1.1063.22.4, 2003-08-27 17:30:18-04:00, shmulik.hen@intel.com [netdrvr bonding] use linked list to handle multiple bond devices ChangeSet@1.1063.22.3, 2003-08-27 17:28:40-04:00, shmulik.hen@intel.com [netdrvr bonding] fix /proc read function ChangeSet@1.1063.22.2, 2003-08-27 17:26:39-04:00, shmulik.hen@intel.com [list] backport list_for_each_entry_safe macro from 2.6 ChangeSet@1.1095, 2003-08-27 12:06:45-06:00, matthewc@cse.unsw.edu.au smpboot.c, acpi.c: ia64: NR_CPUS and number of CPUs While building a kernel for our 4-way Lion box, I made the mistake of setting NR_CPUS to 4. Little did I know that the Lion ACPI tables always list 8 CPUs (with only the first N enabled), and so the resulting kernel overflowed the smp_boot_data.cpu_phys_id array, crashed and burned. (Backported from 2.5 by Bjorn Helgaas). ChangeSet@1.1063.21.2, 2003-08-27 11:36:30-03:00, marcelo@logos.cnet Changed EXTRAVERSION to -pre1 TAG: v2.4.23-pre1 ChangeSet@1.1063.21.1, 2003-08-27 11:07:03-03:00, marcelo@logos.cnet Fix possible IRQ handling SMP race: Kudos to TeJun Huh ChangeSet@1.1063.20.1, 2003-08-27 05:20:22-07:00, davem@kernel.bkbits.net Merge davem@nuts.ninka.net:/home/davem/src/BK/tg3-2.4 into kernel.bkbits.net:/home/davem/tg3-2.4 ChangeSet@1.1063.2.27, 2003-08-26 19:46:29-04:00, jgarzik@redhat.com [wireless airo] build fixes ChangeSet@1.1063.2.26, 2003-08-26 19:40:44-04:00, javier@tudela.mad.ttd.net [wireless airo] add support for MIC and latest firmwares ChangeSet@1.1063.2.25, 2003-08-26 19:37:39-04:00, greg@kroah.com [netdrvr sis900] don't call pci_find_device from irq context I realized that I've had this patch in my tree for a while, and forgot to send it to you and lkml. The patch below fixes bug number 923: http://bugme.osdl.org/show_bug.cgi?id=923 (basically keeps us from calling pci_find_device from interrupt context.) It's been tested by a few people with this device, and they say it works just fine for them. Please forward it on up the food chain. ChangeSet@1.1063.2.24, 2003-08-26 19:20:47-04:00, hirofumi@mail.parknet.co.jp [netdrvr 8139too] add more h/w revision ids ChangeSet@1.1063.2.23, 2003-08-26 19:18:51-04:00, hirofumi@mail.parknet.co.jp [netdrvr 8139too] remove unused RxConfigMask ChangeSet@1.1063.2.22, 2003-08-26 19:16:58-04:00, hirofumi@mail.parknet.co.jp [netdrvr 8139too] lwake unlock fix ChangeSet@1.1063.2.21, 2003-08-26 19:11:37-04:00, srompf@isg.de [netdrvr 8139too] use mii_check_media lib function, instead of homebrew MII bitbanging. ChangeSet@1.1063.19.14, 2003-08-26 18:33:48-04:00, jgarzik@redhat.com [netdrvr 8139too] minor bits from 2.6 Keeping the driver in sync to ease maintenance headaches. ChangeSet@1.1094, 2003-08-26 16:20:31-06:00, alex.williamson@hp.com ia64: Correct NR_CPUS/cpu_online test order in CMC/CPE polling Oops, I must have only tested the UP polling on a 2.4 kernel. On 2.6, I hit the BUG_ON calling cpu_online(1). Simply swapping the order of the test fixes it. This should be applied for 2.6 and 2.4. ChangeSet@1.1093, 2003-08-26 16:14:35-06:00, arun.sharma@intel.com ia64: fix memory leak in sys32_execve path The attached patch by Tony Luck fixes a memory leak in the ia32 execve code path. Please apply to both 2.4 and 2.5. ChangeSet@1.1092, 2003-08-26 15:55:12-06:00, adsharma@unix-os.sc.intel.com [PATCH] ia64: IA-32 compatibility patch: FP denormal handling The following patch makes ia64 compatible with i386 with respect to siginfo.si_code on FP denormal operands. This is necessary for the correct emulation of manycat /r/napali/tmp/p7 |bk-applypatch adsharma unix-os.sc.intel.com ia64: ChangeSet@1.1091, 2003-08-26 15:47:44-06:00, bjorn.helgaas@hp.com ia64: Use $(CC), not $(AS), when checking for "hint @pause" support in binutils. ChangeSet@1.1063.19.13, 2003-08-26 17:41:18-04:00, jgarzik@redhat.com [netdrvr pcmcia] ethtool_ops support for several more pcmcia drivers Drivers updated: fmvj18x_cs, ibmtr_cs, netwave_cs, nmclan_cs, pcnet_cs, ray_cs, wavelan_cs, xirc2ps_cs, xircom_cb. ChangeSet@1.1063.19.12, 2003-08-26 17:26:13-04:00, jgarzik@redhat.com [netdrvr pcmcia] use SET_ETHTOOL_OPS in 3c574, 3c589, aironet4500, and axnet ChangeSet@1.1063.19.11, 2003-08-26 17:15:08-04:00, jgarzik@redhat.com [NET] add SET_ETHTOOL_OPS back-compat hook ChangeSet@1.1063.19.10, 2003-08-26 17:11:53-04:00, jgarzik@redhat.com [netdrvr pcmcia] ethtool_ops for 3c574, 3c589, aironet4500, axnet ChangeSet@1.1063.19.9, 2003-08-26 16:39:48-04:00, jgarzik@redhat.com [netdrvr] ethtool_ops support for 3c515, 3c523, 3c527, and dmfe ChangeSet@1.1063.19.8, 2003-08-26 16:28:32-04:00, jgarzik@redhat.com [netdrvr] ethtool_ops support in 3c503, 3c505, 3c507 ChangeSet@1.1063.19.7, 2003-08-26 16:18:55-04:00, jgarzik@redhat.com [netdrvr 3c501] ethtool_ops support ChangeSet@1.1063.19.6, 2003-08-26 16:12:29-04:00, jgarzik@redhat.com [netdrvr 3c59x] add a piece missed in previous ethtool_ops patch Contributed by Matthew Wilcox. ChangeSet@1.1063.19.5, 2003-08-26 15:38:11-04:00, jgarzik@redhat.com [netdrvr 8139cp] ethtool_ops support ChangeSet@1.1063.2.19, 2003-08-26 16:05:07-03:00, johnstul@us.ibm.com [PATCH] Handle clustered XAPIC in set_ioapic_affinity() Marcelo, All, In set_ioapic_affinity() we take a cpu mask which we normally pass into the apic. However, on xapics in clustered physical mode (ie: x440s) we need to convert this mask into a physical apicid before sending it to the apic. If clustered_apic_mode is XAPIC, this patch selects the lowest cpu in the mask and routes the interrupts there. This avoids timedoubling and other interrupt misdirection seen on x440s after playing with /proc/irq/x/smp_affinity. Please apply. thanks -john ChangeSet@1.1063.2.18, 2003-08-26 16:01:48-03:00, stelian@popies.net [PATCH] meye driver updates The attached patch updates the meye driver with some tiny changes: * use SetPageReserved instead of mem_map_reserve, and remove the wrapper.h dependency * document the supported hardware. ===== Documentation/video4linux/meye.txt 1.6 vs edited ===== ChangeSet@1.1063.2.17, 2003-08-26 16:01:08-03:00, stelian@popies.net [PATCH] sonypi driver update Hi, Now that we are starting a new -pre cycle, here are the latest sonypi updates: * document the fact that FX501/FX702 laptops are not supported * add battery insert/remove events (thanks to Daniel K.) * improve the event detection using a different port offset on 'type2' models (thanks to Daniel K.) * now that ACPI is finally in the kernel, simplify the ACPI tests. Marcelo, please apply. Stelian. ===== Documentation/sonypi.txt 1.12 vs edited ===== ChangeSet@1.1063.19.4, 2003-08-26 14:47:50-04:00, jgarzik@redhat.com [netdrvr 8139cp] minor bits from 2.6 ChangeSet@1.1063.19.3, 2003-08-26 14:45:14-04:00, jgarzik@redhat.com [netdrvr sis900] minor bits from 2.6 ChangeSet@1.1063.19.2, 2003-08-26 14:42:50-04:00, jgarzik@redhat.com [netdrvr sis900] ethtool_ops support ChangeSet@1.1063.2.16, 2003-08-26 15:37:53-03:00, hch@lst.de [PATCH] reserve a sysctl number for XFS (pagebuf) Reserve the 2.5 number for VM_PAGEBUF in 2.4 aswell. ChangeSet@1.1090, 2003-08-26 12:33:19-06:00, bjorn.helgaas@hp.com ia64: initialize bootmem early for acpi_table_init(). ChangeSet@1.1063.19.1, 2003-08-26 14:32:28-04:00, willy@debian.org [netdrvr 3c59x] ethtool_ops support ChangeSet@1.1063.2.15, 2003-08-26 15:29:58-03:00, johnstul@us.ibm.com [PATCH] Do not clear SMI pin at bootup Marcelo, All, Some of our more recent hardware requires that SMIs are routed through the IOAPIC, thus when we clear_IO_APIC() at boot time, we clear the BIOS initialized SMI pin (which is unfortunately not described in the MPS or ACPI tables). This basically clobbers the SMI, which can cause problems with console redirection as well keeping us from being able to transition into full ACPI mode. This patch (back ported from 2.5) simply reads the apic entry in clear_IO_APIC to make sure the delivery_mode isn't dest_SMI. If it is, we leave the apic entry alone and return. With this patch, booting with full ACPI works and SMIs function properly. Please consider for acceptance. thanks -john ChangeSet@1.1063.2.14, 2003-08-26 15:28:13-03:00, willy@w.ods.org [PATCH] make log buffer length selectable On Mon, Aug 25, 2003 at 04:48:30AM -0700, Marcelo Tosatti wrote: > final: > > - 2.4.22-rc4 was released as 2.4.22 with no changes. Hi Marcelo, as you requested, here is the log_buf_len patch for inclusion in 23-pre. Cheers, Willy ChangeSet@1.1063.2.13, 2003-08-26 11:52:36-03:00, hch@lst.de [PATCH] use list_add_tail in buffer_insert_list This is a really old patch from the XFS tree. We need to call list_add_tail in buffer_insert_list to preserve buffer ordering. This essential for a good extent layout with XFS's delayed allocation and a while ago the reiserfs group requested the same change. ACKed by sct. ChangeSet@1.1063.2.12, 2003-08-26 10:38:45-03:00, marcelo@logos.cnet Merge bk://kernel.bkbits.net/davem/net-2.4 into logos.cnet:/home/marcelo/bk/linux-2.4 ChangeSet@1.1063.2.11, 2003-08-26 09:51:54-03:00, arjanv@redhat.com [PATCH] Fix asm constraint bug in arch/i386/kernel/pci-pc.c Hi, Reported by HP, found by Jakub. The patch below fixes a missing asm constraint. The asm constraint of this function needs to depend on (opt) so that gcc knows that the initialisation of the opt structure needs to happen before the asm code; in the old code only the address of opt was in the constraints which isn't enough. Newer gcc will just miscompile this for p4. Greetings, Arjan van de Ven ChangeSet@1.1063.2.10, 2003-08-26 09:51:29-03:00, ak@muc.de [PATCH] Compile fix for ACPI in 2.4.22/x86-64 Marcelo unfortunately added an last minute ACPI update that changed ACPI interfaces and broke x86-64 compilation. I didn't catch it in time, so 2.4.22 does not compile out of the box for AMD64. This patch fixes it. You'll have to apply it when compiling 2.4.22 for x86-64 -Andi ChangeSet@1.1063.12.48, 2003-08-25 16:47:00-07:00, davem@nuts.ninka.net [TG3]: More fixes and enhancements. - Use ethtool_op_{get,set}_tso(). - Avoid partial byte enables on DMA writes, this upsets several non-x86 PCI controllers. ChangeSet@1.1063.12.47, 2003-08-25 16:45:11-07:00, davem@nuts.ninka.net [TG3]: Bump version/reldate. ChangeSet@1.1063.12.46, 2003-08-25 16:44:38-07:00, davem@nuts.ninka.net [TG3]: Fix tg3_phy_reset_5703_4_5 chip rev test. ChangeSet@1.1063.12.45, 2003-08-25 16:44:03-07:00, davem@nuts.ninka.net [TG3]: Bump version/reldate. ChangeSet@1.1063.12.44, 2003-08-25 16:42:24-07:00, davem@nuts.ninka.net [TG3]: Add {get,set}_tso ethtool_ops support. Also, include TSO support code when NETIF_F_TSO is available but do not enable TSO by default even on capable cards. User can turn it on via ethtool. ChangeSet@1.1063.12.43, 2003-08-25 16:40:28-07:00, davem@nuts.ninka.net [TG3]: Differentiate between TSO capable and TSO enabled. ChangeSet@1.1063.12.42, 2003-08-25 16:39:41-07:00, davem@nuts.ninka.net [TG3]: Fix 5788/5901, update TSO code. - Do not set RDMAC_MODE_FIFO_LONG_BURST on 5788 - Do not set WDMAC_MODE_RX_ACCEL on 5788 - Note that 5788 cannot use tagged irq status. - 5788 cannot do TSO - 5788 cannot do NETIF_F_HIGHDMA. - 5901 is 10/100 only. - Update TSO firmware, add 5705 specific TSO firmware. - Update TSO packet handling in ->hard_start_xmit() to match updated TSO firmware. - TSO is still off by default until more perf analysis is done. ChangeSet@1.1063.12.41, 2003-08-25 16:38:37-07:00, davem@nuts.ninka.net [TG3]: Reset PHY more reliably on 570{3,4,5} chips. ChangeSet@1.1063.12.40, 2003-08-25 16:38:00-07:00, davem@nuts.ninka.net [TG3]: More missing PCI IDs. ChangeSet@1.1063.12.39, 2003-08-25 16:34:06-07:00, davem@nuts.ninka.net [TG3]: Do not reset the RX_MAC unless PHY is Serdes. Bug discovered by Michael Chan @ Broadcom. ChangeSet@1.1063.12.38, 2003-08-25 16:33:23-07:00, davem@nuts.ninka.net [TG3]: Fix statistics on 5705. ChangeSet@1.1063.12.37, 2003-08-25 16:32:45-07:00, jgarzik@redhat.com [TG3]: Add 5782 pci id. ChangeSet@1.1063.12.36, 2003-08-25 16:31:24-07:00, jgarzik@redhat.com [TG3]: Only write the on-nic sram addr on non-5705. ChangeSet@1.1063.12.35, 2003-08-25 16:30:42-07:00, jgarzik@redhat.com [TG3]: Another 5705 fix: enable eeprom write prot as needed. ChangeSet@1.1063.12.34, 2003-08-25 16:30:06-07:00, jgarzik@redhat.com [TG3]: More 5705 fixes. - call set-wire-speed phy function from tg3_phy_reset, not just _phy_probe - disable set-wire-speed usage on certain chips, per bcm5700 driver ChangeSet@1.1063.12.33, 2003-08-25 16:29:15-07:00, jgarzik@redhat.com [TG3]: More 5705 updates. - do not zero stats/status block on 5705, they don't exist (in controller memory, at least) - define TG3_MAX_MTU macro such that it takes a (tp) pointer arg, because the value varies between 5705 and non-5705 - use the macro to simply tg3_change_mtu ChangeSet@1.1063.12.32, 2003-08-25 16:27:11-07:00, jgarzik@redhat.com [TG3]: Bug fixes for 5705 support. - (major) status block DMA address wasn't being set, for 5705. whoops. :) - (minor) pass 5705-specific RX_STD_MAX_SIZE - (minor) don't program RCVBDI_JUMBO_THRESH reg on 5705 - (very minor) re-order *THRESH, HOSTCC* regs per bcm570x driver order ChangeSet@1.1063.12.31, 2003-08-25 16:26:01-07:00, davem@nuts.ninka.net [TG3]: Initial implementation of 5705 support. ChangeSet@1.1063.12.30, 2003-08-25 16:21:53-07:00, davem@nuts.ninka.net [TG3]: Merge comment typo fixes from 2.5.x ChangeSet@1.1063.12.29, 2003-08-25 16:20:18-07:00, davem@nuts.ninka.net [TG3]: Sync TSO changes from base 2.5.x ChangeSet@1.1063.12.28, 2003-08-25 16:15:43-07:00, jgarzik@redhat.com [TG3]: Convert to using ethtool_ops. ChangeSet@1.1063.12.27, 2003-08-25 16:11:30-07:00, greg@kroah.com [TG3]: pci_device_id can not be marked __devinitdata. ChangeSet@1.1063.12.26, 2003-08-25 16:10:41-07:00, jgarzik@redhat.com [TG3]: Detect shared (and screaming) interrupts. ChangeSet@1.1063.12.25, 2003-08-25 16:10:02-07:00, davem@nuts.ninka.net [TG3]: Update to irqreturn_t. ChangeSet@1.1063.12.24, 2003-08-25 16:09:15-07:00, jmorris@intercode.com.au [TG3]: skb_headlen() cleanup. ChangeSet@1.1063.12.23, 2003-08-25 16:04:29-07:00, davem@nuts.ninka.net Merge nuts.ninka.net:/home/davem/src/BK/sparc-2.4 into nuts.ninka.net:/home/davem/src/BK/tg3-2.4 ChangeSet@1.1063.18.1, 2003-08-25 18:56:42-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/marcelo-2.4 into redhat.com:/garz/repo/net-drivers-2.4 ChangeSet@1.1063.17.10, 2003-08-25 19:28:53-03:00, mikpe@csd.uu.se [PATCH] 2.4.22 local APIC updates 3/3: disable APIC_BASE on reboot Marcelo, This patch ensures that we properly disable the local APIC before reboot. This fixes BIOS problems reported by a few people. disable_local_APIC() now checks if detect_init_APIC() enabled the local APIC via the APIC_BASE MSR, and if so it now disables APIC_BASE. Previously we would leave APIC_BASE enabled, and that made some BIOSen malfunction. The SMP reboot code calls disable_local_APIC(). On SMP HW there is no change since detect_init_APIC() isn't called and APIC_BASE isn't enabled by us. An SMP kernel on UP HW behaves just like an UP_APIC kernel, so it disables APIC_BASE if we enabled it at boot. The UP_APIC suspend code is simplified since the existing code to disable APIC_BASE is moved into disable_local_APIC(). (Felix Kühling originally reported the BIOS reboot problem. This is a fixed-up version of his preliminary patch.) Backport from 2.6.0-test4. Tested in 2.4.22-rc. Please apply. /Mikael ChangeSet@1.1063.17.9, 2003-08-25 19:28:16-03:00, mikpe@csd.uu.se [PATCH] 2.4.22 local APIC updates 2/3: add lapic/nolapic options Marcelo, This patch adds the lapic and nolapic kernel options, which give users greater control of the local APIC enable process on x86: - "nolapic" prevents the kernel from enabling or using the local APIC. This is stronger than listing a machine in the DMI scan blacklist, since it also works for machines that boot with the local APIC already enabled. "nolapic" can also be seen as a simple-to-deploy alternative to keeping the DMI blacklist rules up-to-date. - "lapic" tells the kernel to force-enable the P4 local APIC if the BIOS disabled it. This is needed on some machines. The default (don't enable if BIOS didn't) can't be changed since there is an even larger number of P4s where ACPI breaks if the local APIC is enabled. Backport from 2.6.0-test4. Tested in 2.4.22-rc. Please apply. This patch renames the dont_enable_local_apic variable, so if you apply it you also need to apply "local APIC updates 1/3: remove incorrect blacklist rules". /Mikael ChangeSet@1.1063.17.8, 2003-08-25 19:26:55-03:00, mikpe@csd.uu.se [PATCH] 2.4.22 local APIC updates 1/3: remove incorrect blacklist rules Marcelo, This patch removes the x86 local APIC blacklist rules for the Microstar 6163 and Intel AL440LX mainboards, as these boards _do_ work with local APIC enabled. The AL440LX blacklist rule was the only place where the dont_enable_local_apic_timer flag could be set, so the patch removes that variable and the code which checks it. The hangs these boards used to have has since been traced to a combination of using specific graphics cards with APM's DISPLAY_BLANK option. Backport from 2.5.72. Also tested in 2.4 for months. Please apply. /Mikael ChangeSet@1.1063.2.7, 2003-08-25 18:26:15-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/marcelo-2.4 into redhat.com:/garz/repo/hwrandom-2.4 ChangeSet@1.1063.17.7, 2003-08-25 19:21:59-03:00, m.c.p@wolk-project.de [PATCH] Proper APIC with HyperThreading Hi Marcelo, the boot issue with DL760 G2 issue with Hyper-Threading (16-way logical) was root-caused to be coming from hardcoding of apic-version during the acpi boot-up code. The attached patch fixes the issue, by reading in the actual apic version. This apic-version is used later in the boot process, to identify whether the processor supports xAPIC or not. And that in turn was resulting in boot failure. Refer following lkml threads for detailed the problem description: Subject: SUMMARY: DL760 G2 issue with Hyper-Threading: Subject: PROBLEM: DL760 G2 issue on 2.4.21 with Hyper-Threading In -ac and -wolk for some time. Patch by "Venkatesh Pallipadi" ChangeSet@1.1063.17.6, 2003-08-25 19:18:30-03:00, m.c.p@wolk-project.de [PATCH] Avoid potentially leaking pagetables into the per-cpu queues Hi Marcelo, $subject says it all. Patch from Andrea Arcangeli! It's in -aa since 2.4.17rc2aa1, in -wolk for some time. ChangeSet@1.1063.17.5, 2003-08-25 19:11:07-03:00, marcelo@logos.cnet Cset exclude: m.c.p@wolk-project.de|ChangeSet|20030825194257|34486 ChangeSet@1.1063.12.22, 2003-08-25 14:51:42-07:00, davem@nuts.ninka.net [NET]: Include asm/uaccess.h in net/core/ethtool.c ChangeSet@1.1063.2.6, 2003-08-25 17:38:58-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/marcelo-2.4 into redhat.com:/garz/repo/hwrandom-2.4 ChangeSet@1.1063.2.5, 2003-08-25 17:37:39-04:00, jgarzik@redhat.com [ia32] mention that X86_VENDOR_ID is tied to NCAPINTS, in a comment in arch/i386/kernel/head.S. ChangeSet@1.1063.17.4, 2003-08-25 18:24:46-03:00, m.c.p@wolk-project.de [PATCH] CONFIG_NR_CPUS Hi Marcelo, this patch has been around for ages. 2.6 has the same change/cleanup. Now, w/o the patch, the kernel will be compiled with 32 CPUs, allthough you might only have 2 or 4 or so. This is purely to save memory - each supported CPU adds approximately eight kilobytes to the kernel image. ChangeSet@1.1063.12.21, 2003-08-25 14:24:34-07:00, jgarzik@redhat.com [NET]: Backport ethtool_ops from 2.6.x ChangeSet@1.1063.17.3, 2003-08-25 18:18:51-03:00, m.c.p@wolk-project.de [PATCH] LVM Update v1.0.5 to v1.0.7 On Thursday 31 July 2003 03:04, Marcelo Tosatti wrote: Hi Marcelo, > > would you consider updating LVM from 1.05 to 1.07 in -pre10? It's out > > there for many months now, many vendor tree's are using it, my tree uses > > it and prolly many others and I am not aware of any problems. > > Should I send a patch? Please let me know. > It depends on the type of changes. I prefer including it in -pre10 only if > its obvioux bugfixes only. > Please send me it together with a detailed changelog. ok, here we go, 5 small but important bugfixes: Changes from v1.0.5 -> v1.0.7 ----------------------------- o fixed lvrename bug leaving wrong filesystem entries around in /proc/lvm/ o fixed lvextend/lvreduce bug on snapshots preventing correct reload of the excpetion table possibly giving wrong snapshot content. o fixed snapshot extend/reduce memory leak o fixed wrong VG PE counter o some whitespace cleanup ChangeSet@1.1063.17.1, 2003-08-25 18:06:46-03:00, m.c.p@wolk-project.de [PATCH] Handle get_block errors correctly in block_read_full_page() Hi Marcelo, quoting Andrea: > Anton found get_block errors weren't handled correctly by > block_read_full_page(). This is my proposed fix for the problem, > implementation and patching is from Anton Altaparmakov. It's in -aa and -wolk for some time. Please apply for 2.4.23-pre1. Thank you :) ChangeSet@1.1063.12.20, 2003-08-25 14:00:11-07:00, davem@nuts.ninka.net [TG3]: Fix AC1001 typo in pci_ids.h ChangeSet@1.1063.16.1, 2003-08-25 13:41:49-07:00, davem@nuts.ninka.net Merge nuts.ninka.net:/home/davem/src/BK/sparcwork-2.4 into nuts.ninka.net:/home/davem/src/BK/sparc-2.4 ChangeSet@1.1063.12.19, 2003-08-25 13:38:25-07:00, davem@nuts.ninka.net Merge nuts.ninka.net:/home/davem/src/BK/network-2.4 into nuts.ninka.net:/home/davem/src/BK/net-2.4 ChangeSet@1.1063.14.2, 2003-08-25 13:29:23-07:00, trini@kernel.crashing.org PPC32: Fix KGDB and userland GDB interactions. ChangeSet@1.1063.12.18, 2003-08-25 16:44:39-03:00, m.c.p@wolk-project.de [PATCH] Unblank console if panic() Hi Marcelo, if a kernel panic()'s, and a console blank timer is running, it is possible that the screen goes blank and he/she isn't able to read the panic. Bad. The attached patch disables the console blank timer when panic(). Patch from Daniel Phillips! It's in -wolk for a long time. NOTE: You have to apply "2.4-panic-console-switch-1.patch" first, otherwise this will fail with 1 hunk. ChangeSet@1.1063.12.17, 2003-08-25 16:42:57-03:00, m.c.p@wolk-project.de [PATCH] Allow sysrq() via /proc/sys/kernel/magickey Hi Marcelo, sysrq() is a good thing to debug things, but unfortunately you need to have physical access to the keyboard. My company for instance maintains tons of remote machines and sometimes we need to do sysrq() too, but it's not possible to do so the remote way. Attached patch enables emulation of the Magic SysRq key (Alt-SysRq-key) via the /proc interface. Just echo the desired character into the file and there you go. Patch from Randy Dunlap! It's in -wolk for a long time and also in some other kernel tree forks. 2.5/2.6 has almost the same (/proc/sysrq-trigger) ChangeSet@1.1063.12.16, 2003-08-25 16:30:09-03:00, acme@conectiva.com.br [PATCH] irqreturn_t compatibility with 2.6 Hi Marcelo, As I described to you in another message this patch is needed so that drivers don't have to be adding private copies of irqreturn_t and friends, like ATM, aic7xxx already do in 2.4.22-pre10, please apply. It is still against 2.4.22-pre10, but I think it'll apply cleanly, holler if not. - Arnaldo ChangeSet@1.1063.12.15, 2003-08-25 16:25:44-03:00, m.c.p@wolk-project.de [PATCH] Allow console switching after kernel panic() Hi Marcelo, if a kernel panic()'s, there's no chance to switch to another console for maybe important informations either about the panic or some important email the user was writing or such stuff. Attached patch allows _only_ console switching if a panic occured. Patch from Andi Kleen! It's in -wolk for a long time. ChangeSet@1.1063.12.14, 2003-08-25 15:37:31-03:00, marcelo@logos.cnet Cset exclude: m.c.p@wolk-project.de|ChangeSet|20030825183254|28555 ChangeSet@1.1063.12.13, 2003-08-25 15:32:54-03:00, m.c.p@wolk-project.de [PATCH] disable READA is an SMP only thing. Make it so. [ The following text is in the "iso-8859-15" character set. ] [ Your display is set for the "ISO-8859-1" character set. ] [ Some characters may be displayed incorrectly. ] Hi Marcelo, $subject says it all. It's in -aa, -wolk for a long time. Reference: http://lists.insecure.org/lists/linux-kernel/2002/Jul/6615.html ChangeSet@1.1063.13.4, 2003-08-25 13:19:37-05:00, shaggy@shaggy.austin.ibm.com JFS: add nointegrity mount option (Karl Rister) This option allows a performance boost by not writing to the journal at the expense of loss of data integrity if the system crashes or is powered off. The intended use of this option is to allow faster restores from backup media. ChangeSet@1.1063.12.12, 2003-08-25 15:15:54-03:00, m.c.p@wolk-project.de [PATCH] Replace bogus and obsolete "#if __SMP__" -> CONFIG_SMP Hi Marcelo, there are some places in the current kernel where __SMP__ appears. This is from old 2.2 times and is never ever defined nor used in current 2.4. Attached patch replaces __SMP__ with CONFIG_SMP. It also cleanes up 3 bogus #ifdef __SMP__ in fs/jbd code which is never used and not needed. Patch from Andrea Arcangeli! It's in -aa since 2.4.21-aa1, in -wolk for some time. Same cleanup was made to 2.5.x. Please apply for 2.4.23-pre1. Thank you :) With this applied, _all_ bogus __SMP__ defines are gone from 2.4. This patch is absolutely safe and just a cleanup! ChangeSet@1.1063.14.1, 2003-08-25 11:12:16-07:00, trini@kernel.crashing.org PPC32: Change the default behavior of a kernel with KGDB. We now don't default to an initial breakpoint, as this is how KGDB on i386 works. ChangeSet@1.1063.12.11, 2003-08-25 15:11:30-03:00, m.c.p@wolk-project.de [PATCH] Cleanup DRM submenu Hi Marcelo, attached patch cleans up the DRM submenu. ChangeSet@1.1063.12.10, 2003-08-25 15:04:58-03:00, m.c.p@wolk-project.de [PATCH] Fix initrd with netboot From "Martin Persson" we've encountered a problem when trying to netboot 2.4.21 and above. After /linuxrc has been executed and the kernel tries to remount the root, it panics with the all too well known message "Unable to mount root fs on ...". The kernel bugs out in mount_block_root in the file init/do_mounts.c, to be more precise in the for-loop. What happens is that it tries to mount the file system as type ext2 (which happens to be first in the list in our case), but instead of returning -EINVAL it returns -EBUSY, the loop exits instead of trying the next (correct) fs-type and the kernel panics. In -wolk for some time and maybe others. ChangeSet@1.1063.12.9, 2003-08-25 15:03:11-03:00, m.c.p@wolk-project.de [PATCH] Cleanup kmem_cache_reap() From "fbl@netbank.com.br" # This patch includes the following deltas: # ChangeSet 1.1193 -> 1.1194 # mm/slab.c 1.22 -> 1.23 # # The following is the BitKeeper ChangeSet Log # -------------------------------------------- # 03/06/03 fbl@netbank.com.br 1.1194 # [PATCH] Cleanup kmem_cache_reap() # # There is no need to keep the list_entry() # out of DEBUG condition. # -------------------------------------------- In 2.5 and -wolk for some time. # This is a BitKeeper generated patch for the following project: # Project Name: Linux kernel tree # This patch format is intended for GNU patch command version 2.5 or higher. # This patch includes the following deltas: # ChangeSet 1.1193 -> 1.1194 # mm/slab.c 1.22 -> 1.23 # # The following is the BitKeeper ChangeSet Log # -------------------------------------------- # 03/06/03 fbl@netbank.com.br 1.1194 # [PATCH] Cleanup kmem_cache_reap() # # There is no need to keep the list_entry() # out of DEBUG condition. # -------------------------------------------- # ChangeSet@1.1063.12.8, 2003-08-25 14:46:28-03:00, hch@lst.de [PATCH] fix copy_namespace() Originally by aeb, in 2.5 for ages. Fix far too small kmalloc in copy_namespace and promote errors properly. ChangeSet@1.1063.12.7, 2003-08-25 14:37:52-03:00, Jack_Hammer@adaptec.com [PATCH] ServeRAID 6.10 Driver Update With the completion of months of testing at both Adaptec and IBM, the newest version of the ServeRAID Linux device driver( ips Version 6.10.24 ) iscompleted and ready to ship. Attached is the full source ( ips.c & ips.h )and patches for the 2.4 Linux kernel. Both 32-bit andia64 architectures are supported. Please update all 2.4.x kernels with this latest driver as soon as possible. Updates for the 2.5 kernel are being handled separately. Let me know if you have any problems. Jack Hammer ChangeSet@1.1063.12.6, 2003-08-25 14:21:29-03:00, marcelo@logos.cnet pcwd.c: fix oops on unload From Arkadiusz Miskiewicz This patch is from Alan Cox and fixes problems when pcwd driver is loaded while there is no pcwd hardware installed. No idea why it was not feeded to you. Problem is described here http://www.ussg.iu.edu/hypermail/linux/kernel/0308.1/0178.html and tracked down in private conversation with Cox. ChangeSet@1.1063.12.5, 2003-08-25 14:16:14-03:00, mmagallo@debian.org [PATCH] AGPGART support for Intel 7x05 chipsets (backported from 2.6) Hi Marcelo, can you please apply the attached patch before releasing 2.4.22? It adds AGPGART support for Intel's 7205 and 7505 chipsets, which are becoming increasingly common on the street. I have checked on several machines that the patch works, even with AGP 8x. The only hardware I have access to is from NVIDIA, which means I haven't been able to test this with a non-tainted kernel, but I don't expect this to be a problem. ChangeSet@1.1063.12.4, 2003-08-25 13:58:04-03:00, willy@w.ods.org [PATCH] Fix amd67x_pm.c crash with no chipsets / CONFIG_HOTPLUG amd76x_pm.c will crash if no chipsets are found and CONFIG_HOTPLUG is turned on. This patch makes it return with a failure instead. ChangeSet@1.1063.12.3, 2003-08-25 13:57:10-03:00, marcelo@logos.cnet Changed Makefile to 2.4.22 TAG: v2.4.22