ChangeSet@1.1554, 2005-02-07 17:22:57-02:00, marcelo@logos.cnet Linus Torvalds: Add extra debugging help for bad user accesses. ChangeSet@1.1553, 2005-02-07 17:20:08-02:00, marcelo@logos.cnet Linus Torvalds: backport 2.6 rw_verify_area() to check against file offset overflows - Make generic rw_verify_area check against file offset overflows. - Add 'f_maxcount' to allow filesystems to set a per-file maximum IO size. - Rename "locks_verify_area()" to "rw_verify_area()" and clean up the arguments. ChangeSet@1.1551, 2005-02-06 01:58:11-05:00, jgarzik@pobox.com [libata ahci] Add support for ULi M5288 ChangeSet@1.1550, 2005-02-06 00:30:11-05:00, bunk@stusta.de [PATCH] scsi/ahci.c: remove an unused function The patch below removes an unused function from drivers/scsi/ahci.c Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds ChangeSet@1.1549, 2005-02-06 00:06:58-05:00, mkrikis@yahoo.com [PATCH] libata: fix ata_piix on ICH6R in RAID mode Here is the cleaned up patch (as you suggested) that enables ata_piix to work in RAID mode on ICH6R. I tested it and it seems to behave correctly in all the modes---sees all 4 disks in IDE and RAID modes, doesn't see any in Compatibility mode (which is right, because only two are available and the regular IDE driver has picked them up already). Change description: only test for PCI IDE legacy mode, via the PCI programming interface register, if the PCI device is of class PCI_CLASS_STORAGE_IDE. Signed-off-by: Martins Krikis Signed-off-by: Jeff Garzik ChangeSet@1.1548, 2005-02-05 21:56:29-05:00, jpaana@s2.org [libata sata_promise] add PCI ID for new SATAII TX2 card ChangeSet@1.1547, 2005-02-05 21:29:38-05:00, syntax@pa.net [libata sata_sil] add another Seagate drive to blacklist ChangeSet@1.1546, 2005-02-05 21:24:09-05:00, zaitcev@redhat.com [libata] fix probe object allocation bugs The code previously alloced a 2-entry array, then freed each entry individually, for users of "combined mode". Convert code to use allocation function that only allocates one object at a time. Also, un-export ata_pci_init_legacy_mode(), as its only user is libata-core. ChangeSet@1.1545, 2005-02-05 20:31:23-05:00, russb@emc.com [libata scsi] verify cmd bug fixes/support Also, trim trailing whitespace. ChangeSet@1.1539.1.3, 2005-02-03 13:30:57-02:00, marcelo@logos.cnet Changed VERSION to 2.4.30-pre1 ChangeSet@1.1539.1.2, 2005-02-03 13:23:06-02:00, Mark.Haigh@spirentcom.com [PATCH] Remove redundant check and fix warning in pcibios_lookup_irq() In arch/i386/kernel/pci-irq.c:pcibios_enable_irq(), there is a redundant check: if (pin && !pcibios_lookup_irq(dev, 1) && !dev->irq) { /* ... */ if (pin) { We don't need the second 'if (pin)', as we already know it's nonzero from the first check. Also note that this fixes the following warning (which happens because gcc's isn't always perfect with determining whether a variable is used uninitialized): pci-irq.c: In function `pcibios_enable_irq': pci-irq.c:1128: warning: 'msg' might be used uninitialized in this function All the patch does is remove the duplicate check and shift everything else over. ChangeSet@1.1539.1.1, 2005-02-03 13:17:21-02:00, petrides@redhat.com [PATCH] fix for memory corruption from /proc/kcore access A fairly nasty memory corruption potential exists when /proc/kcore is accessed and there are at least 62 vmalloc'd areas. The problem is that get_kcore_size() does not properly account for the elf_prstatus, elf_prpsinfo, and task_struct structure sizes in the fabricated ELF header, and then elf_kcore_store_hdr() and its associated calls to storenote() will possibly overrun the "elf_buf" buffer allocated by read_kcore(). Because the requested buffer size is rounded up to a page multiple, only certain ranges of counts of vmalloc'd areas will actually lead to a memory corruption. When it does happen, usually the end of the /proc/kcore reader's task_struct ends up being copied into a slab page (or sometimes into a data page) causing a kernel crash (or data corruption) at a later point in time. The 1st hunk of the patch below fixes this problem. The latter 3 hunks correct the "p_filesz" value for the note section (which is already initialized to 0 on line 232) as stored in the ELF header, but these hunks are not necessary to fix the corruption possiblity. The fix is already in 2.6. ChangeSet@1.1544, 2005-02-02 04:10:37-05:00, jgarzik@pobox.com [libata sata_promise] support Promise SATAII TX2/TX4 cards Originally from Peter Smith, with testing and additional contributions from Jens Axboe. ChangeSet@1.1543, 2005-02-02 04:10:25-05:00, albertcc@tw.ibm.com [libata] SCSI-to-ATA translation fixes Fixes lesser-used READ VERIFY and READ_16/WRITE_16 translations. Changes: #1. ata_scsi_rw_xlat(): Fix incorrect SCSI to ATA translation for 16 bytes SCSI Read/Write commands. #2. Fix ata_scsi_verify_xlat(): Fix incorrect tf->hob_lbax translation. Signed-off-by: Albert Lee ChangeSet@1.1542, 2005-02-02 04:10:03-05:00, jgarzik@pobox.com [libata] Remove CDROM drive from PATA DMA blacklist Mirrors changeset from Bart to drivers/ide/* ChangeSet@1.1541, 2005-02-02 04:09:43-05:00, jgarzik@pobox.com [libata] add DMA blacklist Blacklist devices that seem to have some problem with DMA. From drivers/ide/ide-dma.c blacklist. ChangeSet@1.1540, 2005-02-01 22:20:57-05:00, achew@nvidia.com [PATCH] sata_nv: enable generic class support for future NVIDIA SATA This patch adds a new entry in the pci_device_id table that filters by class code, in order to cause unlisted NVIDIA SATA controllers to be probed. In the probe function, we determine whether the device is a SATA or IDE controller by checking the device's bars (NVIDIA SATA controllers will always have 6 bars). Bar5 I/O mapped vs. memory mapped is now determined programmatically by looking at the resource flags of the bar. Signed-off-by: Andrew Chew Signed-off-by: Jeff Garzik ChangeSet@1.1521.19.2, 2005-02-01 12:26:07-08:00, davem@nuts.davemloft.net [SPARC64]: __atomic_{add,sub}() must sign-extend return value. Even though we declare these functions as returning a 32-bit signed integer, the sparc64 ABI states that such functions must properly sign-extend the return value to the full 64-bits. Due to this bug, parts of mm/rmap.c were misbehaving when compiled with gcc-3.4 on sparc64. gcc-3.4 was legally using a 64-bit comparison against zero with the return value of __atomic_add(). I would like to thank Hugh Daniels and others for helping to track down this peculiar bug. Signed-off-by: David S. Miller ChangeSet@1.1521.1.61, 2005-01-31 20:07:37-08:00, kaber@trash.net [NETFILTER]: Fix ip_fw_compat.c build after IP_DEFRAG_* changes. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller ChangeSet@1.1537, 2005-01-31 01:36:12-05:00, len.brown@intel.com Merge intel.com:/home/lenb/src/24-stable-dev into intel.com:/home/lenb/src/24-latest-dev ChangeSet@1.1521.22.1, 2005-01-31 00:38:19-05:00, len.brown@intel.com Merge intel.com:/home/lenb/bk/linux-2.4.29 into intel.com:/home/lenb/src/24-stable-dev ChangeSet@1.1536, 2005-01-31 00:13:49-05:00, len.brown@intel.com Merge intel.com:/home/lenb/src/24-stable-dev into intel.com:/home/lenb/src/24-latest-dev ChangeSet@1.1521.20.18, 2005-01-29 14:26:17-02:00, james4765@cwazy.co.uk [PATCH] lcd: fix memory leak in lcd_ioctl() This patch fixes a memory leak in the FLASH_Burn ioctl for the Cobalt LCD interface driver. Signed-off-by: James Nelson ChangeSet@1.1521.20.17, 2005-01-29 14:25:32-02:00, james4765@cwazy.co.uk [PATCH] lcd: Add checks to CAP_SYS_ADMIN to potentially dangerous ioctl's This patch adds CAP_SYS_ADMIN checks to the potentially dangerous ioctls FLASH_Erase and FLASH_Burn in the Cobalt LCD interface driver. Signed-off-by: James Nelson ChangeSet@1.1521.20.16, 2005-01-29 14:23:05-02:00, khali@linux-fr.org [PATCH] I2C updates: Document that the "id" member of the i2c_client structure was discarded in Linux 2.6. Original discussion: http://archives.andrew.net.au/lm-sensors/msg29038.html http://archives.andrew.net.au/lm-sensors/msg29041.html Bottom line: The "id" member of the i2c_client structure was discarded in Linux 2.6 due to a lack of consistent use and overall need. While we of course won't backport this to Linux 2.4 so as to not break compatibility with existing 2.4 drivers, it still seems wise to update the documentation so that new drivers written for Linux 2.4 today (*sigh*) do not rely on it. ChangeSet@1.1521.20.15, 2005-01-29 14:21:15-02:00, khali@linux-fr.org [PATCH] I2C updates: small header cleanups While editing i2c-algo-bit.c and i2c-algo-pcf.c (see 1/3 of this patch set), I noticed that both files include header files they do not need at all (namely linux/ioport.h and asm/uaccess.h). The following patch cleans this, additionally bringing the files in line with what we have in i2c CVS (thus the blank lines changes). ChangeSet@1.1521.20.14, 2005-01-29 13:27:33-02:00, khali@linux-fr.org [PATCH] I2C updates: The "bit" and "pcf" i2c algorithms should declare themselves fully I2C capable Bottom line: The "bit" and "pcf" i2c algorithms should declare themselves fully I2C capable, but do not. Credits go to Ian Campbell for noticing and proposing patches. This is a backport from Linux 2.6.11-rc1. ChangeSet@1.1521.18.4, 2005-01-28 22:30:08-05:00, mkrikis@yahoo.com [PATCH] fix an oops in ata_to_sense_error Signed-off-by: Jeff Garzik ChangeSet@1.1521.20.13, 2005-01-28 15:18:14-02:00, kronos@kronoz.cjb.net [PATCH] Fix MSF overflow in ide-cd with multisession DVDs This a backport of my patch that went into 2.6.10. cdrom_read_toc (ide-cd.c) always reads the TOC using MSF format. If the last session of the disk starts beyond block 1152000 (LBA) there's an overflow in the MSF format and kernel complains: Unable to identify CD-ROM format. So read the multi-session TOC in LBA format in order to avoid an overflow in MSF format with multisession DVDs. Signed-off-by: Luca Tettamanti ChangeSet@1.1521.20.12, 2005-01-28 07:46:20-02:00, npollitt@mvista.com [PATCH] Configure mangles hex values When doing a make oldconfig, the hex function strips the leading '0x' from hex values. The '0x' is needed in the final autoconf.h, and its absence causes the following problem. If I start with a hex value in my config file like this: CONFIG_LOWMEM_SIZE=0x40000000 When I run make oldconfig, it strips out the '0x' leaving this: CONFIG_LOWMEM_SIZE=40000000 Then if I run make xconfig, this is not considered a valid hex value, so it replaces my value with the default: CONFIG_LOWMEM_SIZE=0x20000000 The following patch removes the lines that strip the 0x from the hex value. It also checks the result for the leading 0x and inserts it if necessary. ChangeSet@1.1521.20.11, 2005-01-28 07:43:18-02:00, marcelo@logos.cnet Hugh Dickins: remove rlim_rss and this RLIMIT_RSS code from madvise. Presumably the code crept in by mistake. Isn't the right fix just to remove rlim_rss and this RLIMIT_RSS code from here? It's pretty silly for madvise alone to be taking notice of it. Presumably the code crept in by mistake from some tree which was using an RLIMIT_RSS patch on 2.4. ChangeSet@1.1521.20.10, 2005-01-28 07:35:27-02:00, marcelo@logos.cnet Cset exclude: dan@embeddedalley.com|ChangeSet|20050128083257|00819 ChangeSet@1.1521.20.9, 2005-01-28 06:32:57-02:00, dan@embeddedalley.com [PATCH] add AMD Geode processor support Here is a patch for 2.4 that adds the basic AMD Geode GX2/GX3 and GX1/SC1200 support. This patch updates configuration scripts, defconfig, and setup files. Signed-off-by: Dan Malek ChangeSet@1.1521.18.3, 2005-01-27 16:53:51-05:00, jason.d.gaston@intel.com [PATCH] SATA AHCI support for Intel ICH7R This patch adds the Intel ICH7R DID's to the ahci.c SATA AHCI driver for ICH7R SATA support. Signed-off-by:  Jason Gaston Signed-off-by: Jeff Garzik ChangeSet@1.1521.18.2, 2005-01-27 15:38:14-05:00, haroldo.gamal@infolink.com.br [libata sata_sil] add another Seagate driver to blacklist ChangeSet@1.1521.20.8, 2005-01-27 08:06:04-02:00, marcelo.tosatti@cyclades.com [PATCH] Ake Sandgren: Fix RLIMIT_RSS madvise calculation bug On Wed, Jan 26, 2005 at 12:07:50PM +0100, Ake wrote: > Use of rlim[RLIMIT_RSS] in mm/filemap.c is wrong. > It is passed down to kernel as a number of bytes but is being used as a > number of pages. > > There is also a misinformative comment in fs/proc/array.c > in proc_pid_stat where it says > mm ? mm->rss : 0, /* you might want to shift this left 3 */ > the number 3 should probably be PAGE_SHIFT-10. Amazing that this has never been noticed before - I bet not many people use RSS limits with madvise(). This transform the rlimit in pages before the comparison, can you please test it. ChangeSet@1.1521.20.7, 2005-01-27 07:28:56-02:00, zaitcev@redhat.com [PATCH] USB: Workarounds for Genesys Logics Disk enclosures with Genesys Logics chipset require additional delays, or commands are not processed. Also, their maximum transfer size is limited. Patch by Martin Strigl. ChangeSet@1.1521.20.6, 2005-01-27 07:28:42-02:00, zaitcev@redhat.com [PATCH] USB: Prevent hiddev from looping In the loop, schedule() returns with the current state TASK_RUNNING, so at the next revolution it returns immediately, and the task sits there burning CPU. The fix is to add the forgotten state assignment. Patch from David Micon. ChangeSet@1.1521.1.59, 2005-01-26 22:04:53-08:00, kaber@trash.net [IPV4]: Keep fragment queues private to each user. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller ChangeSet@1.1521.1.58, 2005-01-26 15:10:46-08:00, davem@nuts.davemloft.net [TG3]: Update driver version and reldate. Signed-off-by: David S. Miller ChangeSet@1.1521.1.57, 2005-01-26 15:10:13-08:00, mchan@broadcom.com [TG3]: 5750 fixes Some misc. fixes for 5750. The first 2 fixes below are for correctness. - Do not set extended MAC addresses for 5750 - Do not set NIC ring addresses in send ring and receive return ring RCBs for 5750 - Check eeprom write protect bit for 5750 Signed-off-by: Michael Chan Signed-off-by: David S. Miller ChangeSet@1.1521.20.5, 2005-01-26 10:10:40-02:00, mauelshagen@redhat.com [PATCH] fix panics while backing up LVM snapshots this patch fixes lvm-snap.c in order to avoid a list update on the snapshot exception hash happening while only holding a read lock as documented in Red Hat bugzilla #135266. ChangeSet@1.1499.5.2, 2005-01-26 03:49:55-05:00, len.brown@intel.com [ACPI] via interrupt quirk fix from 2.6 http://bugzilla.kernel.org/show_bug.cgi?id=3319 Signed-off-by: David Shaohua Li Signed-off-by: Len Brown Signed-off-by: David S. Miller ChangeSet@1.1521.1.55, 2005-01-25 21:32:20-08:00, mostrows@watson.ibm.com [MAINTAINERS]: Fix my email address in PPPOE entry. Signed-off-by: David S. Miller ChangeSet@1.1521.20.4, 2005-01-25 06:56:30-02:00, marcelo@logos.cnet Cset exclude: temnota@kmv.ru|ChangeSet|20050119161632|63236 ChangeSet@1.1521.1.54, 2005-01-24 17:41:05-08:00, davem@nuts.davemloft.net [TG3]: Update driver version and reldate. Signed-off-by: David S. Miller ChangeSet@1.1521.1.53, 2005-01-24 17:40:28-08:00, mchan@broadcom.com [TG3]: Fix TSO for 5750 - Fix TSO for 5750 chips by setting tcp checksum field to 0 for TSO packets - Add TG3_FLG2_HW_TSO flag for 5750 and newer chips that use the same TSO scheme Signed-off-by: Michael Chan Signed-off-by: David S. Miller ChangeSet@1.1521.20.3, 2005-01-24 21:16:35-02:00, marcelo@logos.cnet Merge bk://linux@linux.bkbits.net/linux-2.4 into logos.cnet:/home/marcelo/bk/linux-2.4 ChangeSet@1.1521.21.2, 2005-01-24 16:59:07-02:00, paul.clements@steeleye.com [PATCH] Rogier Wolff: fix nbd ioctl permissions Here's a patch for nbd that Rogier recently sent me. It allows non-root to do BLKGETSIZE, et al. on nbd devices, which he needs for his data recovery applications. From: Rogier Wolff Signed-Off-By: Paul Clements Description: We shouldn't need CAP_SYS_ADMIN to ask for disk capacity and such. ChangeSet@1.1521.21.1, 2005-01-24 16:51:44-02:00, marcelo@logos.cnet [PATCH] Karsten Keil: Eicon Diva PCI 2.02 bugfix This bugfix solve a race between card initialization and S0 bus events, if the S0 bus is already active on driver load time. Signed-off-by: Karsten Keil ChangeSet@1.1521.20.2, 2005-01-24 12:48:08-02:00, paul.clements@steeleye.com [PATCH] nbd: fix ioctl permissions Here's a patch for nbd that Rogier recently sent me. It allows non-root to do BLKGETSIZE, et al. on nbd devices, which he needs for his data recovery applications. ChangeSet@1.1521.20.1, 2005-01-24 07:36:05-02:00, khali@linux-fr.org [PATCH] PCI: Kill duplicate definition of INTEL_82801DB_10 I noticed that PCI_DEVICE_ID_INTEL_82801DB_10 is currently defined twice in linux-2.4.29/include/linux/pci_ids.h. The trivial patch below kills the second definition. Signed-off-by: Jean Delvare ChangeSet@1.1521.1.52, 2005-01-22 19:47:23-08:00, davem@nuts.davemloft.net [TG3]: Update driver version and reldate. Signed-off-by: David S. Miller ChangeSet@1.1521.19.1, 2005-01-21 13:59:28-08:00, krzysztof.h1@wp.pl [SPARC]: Fix asm constraints in muldiv.c Signed-off-by: David S. Miller ChangeSet@1.1521.1.51, 2005-01-20 16:31:46-08:00, mchan@broadcom.com [TG3]: add tg3_set_eeprom() - Add nvram size detection - Add appropriate byte swapping to tg3_get_eeprom so that the same byte stream is read in all systems - Fix tg3_get_eeprom to read both eeprom and flash - Add tg3_set_eeprom to write eeprom and flash - Change tg3_nvram_init to detect all supported nvram devices - Change tg3_nvram_read to properly detect Atmel flash that requires address translation - Increase nvram polling delay to account for slower eeprom devices - Remove some of the flushing read that is not required for the production 5750 devices Signed-off-by: Michael Chan Signed-off-by: David S. Miller ChangeSet@1.1521.1.49, 2005-01-19 14:57:17-02:00, trini@kernel.crashing.org [PATCH] ppc32: Fix a problem with the TLB Miss handler There is a problem in the TLB Miss (and Error, as they jump to the Miss handler) handlers. The problem is that when an app spans more than one L1 entry, we don't have all of the correct information, and do_page_fault() things a protection fault happened, when it didn't really. The fix for this is to modify the handlers slightly to force a TLB Error in this case. Lots of suggestions from: Joakim Tjernlund Signed-off-by: Tom Rini ChangeSet@1.1521.1.48, 2005-01-19 14:16:32-02:00, temnota@kmv.ru [PATCH] Prevent NMI oopser from triggering when megaraid2 waits for abort/reset cmd completion > We should backport msleep() in 2.4.29-pre1. Ok, msleep() backported, but driver isn't fixed. This patch acceptable? Prevent NMI oopser kill kernel thread when megaraid2 driver waiting abort or reset command completion. Signed-off-by: Andrey Melnikov ChangeSet@1.1521.1.47, 2005-01-19 14:09:30-02:00, marcelo@logos.cnet Change VERSION to v2.4.29 TAG: v2.4.29