commit 669df1b478803f49a356528d290af7bf442eb3be Author: Linus Torvalds Date: Sun Dec 31 16:53:20 2006 -0800 Linux 2.6.20-rc3 ...because it's always a good idea to cut a release *before* you go out to party and get drunk. Remember kids: "Don't Drink and Release!" commit 241ceee0b442c69226fb882d61d9b9785743898f Author: Oleg Nesterov Date: Sun Dec 24 23:30:44 2006 +0300 [PATCH] restore ->pdeath_signal behaviour Commit b2b2cbc4b2a2f389442549399a993a8306420baf introduced a user- visible change: ->pdeath_signal is sent only when the entire thread group exits. While this change is imho good, it may break things. So restore the old behaviour for now. Signed-off-by: Oleg Nesterov To: Albert Cahalan Cc: Eric W. Biederman Cc: Andrew Morton Cc: Linus Torvalds Cc: Ingo Molnar Cc: Qi Yong Cc: Roland McGrath Signed-off-by: Linus Torvalds commit b836267aa79c1c5e23e00d9cec047b6870ae0db1 Author: Martin Willi Date: Thu Dec 28 21:27:48 2006 -0800 [XFRM]: Algorithm lookup using .compat name Installing an IPsec SA using old algorithm names (.compat) does not work if the algorithm is not already loaded. When not using the PF_KEY interface, algorithms are not preloaded in xfrm_probe_algs() and installing a IPsec SA fails. Signed-off-by: Martin Willi Acked-by: Herbert Xu Signed-off-by: David S. Miller commit cb4db4c22182163f66325053331a4e7096f663cd Author: David Woodhouse Date: Thu Dec 28 21:21:55 2006 -0800 [NET]: Don't export linux/random.h outside __KERNEL__. Don't add it there please; add it lower down inside the existing #ifdef __KERNEL__. You just made the _userspace_ net.h include random.h, which then fails to compile unless was already included. Signed-off-by: David Woodhouse Signed-off-by: David S. Miller commit e44c39bd8478bf9e41cbc611c74b5879a7b29f58 Author: David S. Miller Date: Thu Dec 28 21:14:52 2006 -0800 [NET]: Add memory barrrier to netif_poll_enable() When a driver writer calls this, they generally expect that all previous stores and modifications they've made will be visible before netif_poll_enable() executes, so ensure this. Noticed by Ben H. Signed-off-by: David S. Miller commit f4060c0dbbe9ad7b8c0aeefa142398c8d3468bac Author: David S. Miller Date: Thu Dec 28 21:43:51 2006 -0800 [SPARC64]: Handle ISA devices with no 'regs' property. And this points out that the return value from isa_dev_get_resource() and the 'pregs' arg to isa_dev_get_irq() are totally unused. Based upon a patch from Richard Mortimer Signed-off-by: David S. Miller commit 55d0bef587aa438fe5e8bed2ecce8611ebcd4c1a Author: David S. Miller Date: Thu Dec 28 21:16:40 2006 -0800 [SPARC64]: Update defconfig. Signed-off-by: David S. Miller commit e3a411a3dfc1d633504aa63efab32b7e00318454 Author: David S. Miller Date: Thu Dec 28 21:01:32 2006 -0800 [SPARC64]: Fix of_iounmap() region release. We need to pass in the resource otherwise we cannot release the region properly. We must know whether it is an I/O or MEM resource. Spotted by Eric Brower. Signed-off-by: David S. Miller commit 6fc5bae797a6632bbccdd49a1b6a96121368a4b9 Author: David S. Miller Date: Thu Dec 28 21:00:23 2006 -0800 [SPARC64]: Fix "mem=xxx" handling. We were not being careful enough. When we trim the physical memory areas, we have to make sure we don't remove the kernel image or initial ramdisk image ranges. Signed-off-by: David S. Miller commit 91f6e54b6e534edbf6208e1c750fff9e0491dfcc Author: Alexey Dobriyan Date: Fri Dec 29 16:50:08 2006 -0800 [PATCH] fuse: fix typo Signed-off-by: Thomas Hisch Signed-off-by: Alexey Dobriyan Cc: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit dbc1333a347745bd55844649c19eb7d28f383a84 Author: Jan Andersson Date: Fri Dec 29 16:50:04 2006 -0800 [PATCH] sparc32: add offset in pci_map_sg() Add sg->offset to sg->dvma_address in pci_map_sg() on sparc32. Without the offset, transfers to buffers that do not begin on a page boundary will not work as expected. Signed-off-by: Jan Andersson Cc: "David S. Miller" Cc: William Lee Irwin III Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 8018c27b26af56af18eb8b2dc600eba825792d8f Author: Ingo Molnar Date: Fri Dec 29 16:50:01 2006 -0800 [PATCH] kvm: fix GFP_KERNEL allocation in atomic section in kvm_dev_ioctl_create_vcpu() fix an GFP_KERNEL allocation in atomic section: kvm_dev_ioctl_create_vcpu() called kvm_mmu_init(), which calls alloc_pages(), while holding the vcpu. The fix is to set up the MMU state in two phases: kvm_mmu_create() and kvm_mmu_setup(). (NOTE: free_vcpus does an kvm_mmu_destroy() call so there's no need for any extra teardown branch on allocation/init failure here.) Signed-off-by: Ingo Molnar Cc: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 55a54f79e091d52c3c2835985c347ea4d2e1c21f Author: Avi Kivity Date: Fri Dec 29 16:49:58 2006 -0800 [PATCH] KVM: Fix oops on oom __free_page() doesn't like a NULL argument, so check before calling it. A NULL can only happen if memory is exhausted during allocation of a memory slot. Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit c68876fd2846d2aaf9770276db244b46f2bf52c7 Author: Nguyen Anh Quynh Date: Fri Dec 29 16:49:54 2006 -0800 [PATCH] KVM: Rename some msrs No need to append _MSR to msr names, a prefix should suffice. Signed-off-by: Nguyen Anh Quynh Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit a8d13ea28bcb63baf69aba84b61cad4c3ff7adb9 Author: Avi Kivity Date: Fri Dec 29 16:49:51 2006 -0800 [PATCH] KVM: More msr misery These msrs are referenced by benchmarking software when pretending to be an Intel cpu. Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 3bab1f5dda3443043cc8fe68c5ae75530339f28f Author: Avi Kivity Date: Fri Dec 29 16:49:48 2006 -0800 [PATCH] KVM: Move common msr handling to arch independent code Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 671d6564796e0c90398aab30f89b5e48fc5a3fbe Author: Avi Kivity Date: Fri Dec 29 16:49:45 2006 -0800 [PATCH] KVM: Implement a few system configuration msrs Resolves sourceforge bug 1622229 (guest crashes running benchmark software). Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 09db28b8a3765a7ec35eba80420c71a7973f5a88 Author: Yoshimi Ichiyanagi Date: Fri Dec 29 16:49:41 2006 -0800 [PATCH] KVM: Initialize kvm_arch_ops on unload The latest version of kvm doesn't initialize kvm_arch_ops in kvm_init(), which causes an error with the following sequence. 1. Load the supported arch's module. 2. Load the unsupported arch's module.$B!!(B(loading error) 3. Unload the unsupported arch's module. You'll get the following error message after step 3. "BUG: unable to handle to handle kernel paging request at virtual address xxxxxxxx" The problem here is that the unsupported arch's module overwrites kvm_arch_ops of the supported arch's module at step 2. This patch initializes kvm_arch_ops upon loading architecture specific kvm module, and prevents overwriting kvm_arch_ops when kvm_arch_ops is already set correctly. Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit a9058ecd3cd72634cf548588ce79b3f225c9ca32 Author: Avi Kivity Date: Fri Dec 29 16:49:37 2006 -0800 [PATCH] KVM: Simplify is_long_mode() Instead of doing tricky stuff with the arch dependent virtualization registers, take a peek at the guest's efer. This simlifies some code, and fixes some confusion in the mmu branch. Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 1e885461f02259d75e7480a70d291d2d8aaa938e Author: Avi Kivity Date: Fri Dec 29 16:49:34 2006 -0800 [PATCH] KVM: Use boot_cpu_data instead of current_cpu_data current_cpu_data invokes smp_processor_id(), which is inadvisable when preemption is enabled. Switch to boot_cpu_data instead. Resolves sourceforge bug 1621401. Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 2b7a52a459cb09864b736265bee1af5cef42af66 Author: Lennert Buytenhek Date: Fri Dec 29 16:49:30 2006 -0800 [PATCH] Update CREDITS and MAINTAINERS entries for Lennert Buytenhek Signed-off-by: Lennert Buytenhek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 44854add66811124a5667466245b6824f751143a Author: Sergei Shtylyov Date: Fri Dec 29 16:49:26 2006 -0800 [PATCH] PIIX/SLC90E66: PIO mode fallback fix The fallback to PIO mode in the hwif->dma_check() handler doesn't work in the Intel PIIX and SMsC SLC90E66 IDE drivers because: - config_drive_for_dma() calls the hwif->speedproc() handler with a wrong mode number (unbiased by XFER_PIO_0) in case of the PIO fallback; - hwif->tuneproc() handler doesn't really set the drive's own speed (this is not fixed as yet). Signed-off-by: Sergei Shtylyov Cc: Alan Cox Acked-by: Bartlomiej Zolnierkiewicz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 242ce41fc92214c3430495edd43783c3ba3dcd2b Author: Sergei Shtylyov Date: Fri Dec 29 16:49:18 2006 -0800 [PATCH] PIIX: remove check for broken MW DMA mode 0 There's no need to check in piix_config_drive_for_dma() for broken MW DMA mode 0 as this mode is not supported by the driver (it sets hwif->mwdma_mask to 0x6), and hence can't be selected by ide_dma_speed(). (Alan sayeth "Probably right but if not you've got a subtle corruptor. Should at least stick a BUG_ON mode 0 setting right close when the mode is set.") Signed-off-by: Sergei Shtylyov Cc: Alan Cox Acked-by: Bartlomiej Zolnierkiewicz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 755cd90029b61b96816b8ac0ab6a33a197f842d0 Author: Andrew Morton Date: Fri Dec 29 16:49:14 2006 -0800 [PATCH] lockdep: printk warning fix kernel/lockdep.c: In function `lookup_chain_cache': kernel/lockdep.c:1339: warning: long long unsigned int format, u64 arg (arg 2) kernel/lockdep.c:1344: warning: long long unsigned int format, u64 arg (arg 2) Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit e2a5d2f9b3e6011093a7587e0e18d653fb050a36 Author: Arnaud Patard (Rtp Date: Fri Dec 29 16:49:10 2006 -0800 [PATCH] spi_s3c24xx_gpio: use right header Russel King recently reminded us that one shouldn't use asm/arch/hardware.h but one should use asm/hardware.h. Unfortunately, the spi_s3c24xx_gpio driver is using the wrong header. This patch is fixing that. Signed-off-by: Arnaud Patard Signed-off-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 829ad751ab3e369ee85af83f1dee45640d3e2d29 Author: Mike Frysinger Date: Fri Dec 29 16:49:07 2006 -0800 [PATCH] respect srctree/objtree in Documentation/DocBook/Makefile The KERNELDOC and DOCPROC variables are relative to the $(srctree)/$(objtree) and expect to be run only from there ... attached patch adds proper srctree/objtree prefixes to both variables. Acked-by: "Randy.Dunlap" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 089e34b60033863549fbe561d31ac8c778a20e7f Author: Andrew Morton Date: Fri Dec 29 16:49:04 2006 -0800 [PATCH] cpuset procfs warning fix fs/proc/base.c:1869: warning: initialization discards qualifiers from pointer target type fs/proc/base.c:2150: warning: initialization discards qualifiers from pointer target type Cc: Paul Jackson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 918d3f90e8d5657491024f64427e9a5ea632d284 Author: Shantanu Goel Date: Fri Dec 29 16:48:59 2006 -0800 [PATCH] Buglet in vmscan.c Fix a rather obvious buglet. Noticed while instrumenting the VM using /proc/vmstat. Cc: Christoph Lameter Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 463cab36926a39d5e690664ebb5560a8b119fc71 Author: Kyungmin Park Date: Fri Dec 29 16:48:56 2006 -0800 [PATCH] ARM: OMAP: fix missing header on apollon board Fix apollon board compiler error Signed-off-by: Kyungmin Park Signed-off-by: Tony Lindgren Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 7f24516240c78760d0b19e6160dcab893ef81918 Author: Kyungmin Park Date: Fri Dec 29 16:48:51 2006 -0800 [PATCH] ARM: OMAP: fix GPMC compiler errors Fix GPMC compiler errors on OMAP2 Signed-off-by: Kyungmin Park Signed-off-by: Tony Lindgren Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit de4fa99266b2e1f6771644b9b50f33fe8f82d728 Author: David Brownell Date: Fri Dec 29 16:48:47 2006 -0800 [PATCH] SPI/MTD: mtd_dataflash oops prevention Return a fault code if the Dataflash driver runs into a "no device present" error when the MISO line has a pulldown (it currently expects a pullup), so that rmmod won't oops. Signed-off-by: David Brownell Cc: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 17c65d69906a6e2e6df4822b00a9e5e00e35ea9a Author: David Brownell Date: Fri Dec 29 16:48:44 2006 -0800 [PATCH] m25p80 build fixes (with MTD debug) Fix build issues that show up with the m25p80 SPI flash driver when building with MTD debug enabled. Signed-off-by: David Brownell Cc: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 4b1badf5d9ddfc46ad075ca5bfc465972c85cc7c Author: David Brownell Date: Fri Dec 29 16:48:39 2006 -0800 [PATCH] SPI: define null tx_buf to mean "shift out zeroes" Some issues were recently turned up with the current specification of what it means for spi_transfer.tx_buf to be null, as part of transfers which are (from the SPI protocol driver perspective) pure reads. Specifically, that it seems better to change the TX behaviour there from "undefined" to "will shift zeroes". This lets protocol drivers (like the ads7846 driver) depend on that behavior. It's what most controller drivers in the tree are already doing (with one exception and one case of driver wanting-to-oops), it's what Microwire hardware will necessarily be doing, and it removes an issue whereby certain security audits would need to define such a value anyway as part of removing covert channels. This patch changes the specification to require shifting zeroes, and updates all currently merged SPI controller drivers to do so. Signed-off-by: David Brownell Signed-off-by: Kumar Gala Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit d6e88e671ac12888df2d533dd4ddef705431a32a Author: Al Viro Date: Fri Dec 29 16:48:35 2006 -0800 [PATCH] page_mkclean_one(): fix call to set_pte_at() (akpm: macros are wonderful) Signed-off-by: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 0f5486ecf7d510595c2af630aae31ed5846bdbeb Author: Randy Dunlap Date: Fri Dec 29 16:48:31 2006 -0800 [PATCH] cciss: build with PROC_FS=n RAID_UNKNOWN is used even when PROC_FS=n, so move it outside of the CONFIG_PROC_FS block. drivers/block/cciss.c:1910: error: 'RAID_UNKNOWN' undeclared (first use in this function) Signed-off-by: Randy Dunlap Cc: Cc: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit bcb4ddb46a4c66d64d091e7ffa951b2aa1ba537f Author: Dimitri Gorokhovik Date: Fri Dec 29 16:48:28 2006 -0800 [PATCH] MM: SLOB is broken by recent cleanup of slab.h Recent cleanup of slab.h broke SLOB allocator: the routine kmem_cache_init has now the __init attribute for both slab.c and slob.c. This routine cannot be removed after init in the case of slob.c -- it serves as a timer callback. Provide a separate timer callback routine, call it once from kmem_cache_init, keep the __init attribute on the latter. Signed-off-by: Dimitri Gorokhovik Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 131612dfe7923bd0ce5f82d6ed8303a7ef96e574 Author: Dimitri Gorokhovik Date: Fri Dec 29 16:48:24 2006 -0800 [PATCH] ramfs breaks without CONFIG_BLOCK ramfs doesn't provide the .set_dirty_page a_op, and when the BLOCK layer is not configured in, 'set_page_dirty' makes a call via a NULL pointer. Signed-off-by: Dimitri Gorokhovik Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit a3f99f8ba8cbd2d4e231b767b3d6236a555da38c Author: Akinobu Mita Date: Fri Dec 29 16:48:20 2006 -0800 [PATCH] module: fix mod_sysfs_setup() return value mod_sysfs_setup() doesn't return error when kobject_add_dir() failed. Signed-off-by: Akinobu Mita Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 96ceeaf4ab66315a5149296bb58d8e1a9e6f2f6d Author: Andreas Schwab Date: Fri Dec 29 16:48:16 2006 -0800 [PATCH] Fix compilation of via-pmu-backlight Signed-off-by: Andreas Schwab Cc: "Yu, Luming" Cc: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 9414232fa0cc28e2f51b8c76d260f2748f7953fc Author: Ingo Molnar Date: Fri Dec 29 16:48:13 2006 -0800 [PATCH] sched: fix cond_resched_softirq() offset Remove the __resched_legal() check: it is conceptually broken. The biggest problem it had is that it can mask buggy cond_resched() calls. A cond_resched() call is only legal if we are not in an atomic context, with two narrow exceptions: - if the system is booting - a reacquire_kernel_lock() down() done while PREEMPT_ACTIVE is set But __resched_legal() hid this and just silently returned whenever these primitives were called from invalid contexts. (Same goes for cond_resched_locked() and cond_resched_softirq()). Furthermore, the __legal_resched(0) call was buggy in that it caused unnecessarily long softirq latencies via cond_resched_softirq(). (which is only called from softirq-off sections, hence the code did nothing.) The fix is to resurrect the efficiency of the might_sleep checks and to only allow the narrow exceptions. Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit d449db98d5d7d90f29f9f6e091b0e1d996184df1 Author: Mikael Pettersson Date: Fri Dec 29 16:48:09 2006 -0800 [PATCH] fix mrproper incompleteness include/linux/utsrelease.h and include/linux/version.h aren't removed any more by mrproper in kernel 2.6.20-rc2. The patch below fixes this. The definition of MRPROPER_FILES looks weird: generated-headers looks like a misspelling of generated_headers, but that one is a Makefile target, not a variable or a file, so I don't see how including it in MRPROPER_FILES could have any effect. Signed-off-by: Mikael Pettersson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 1f59c140f8387d32b8fe137eb02da40630b618e8 Author: Tilman Schmidt Date: Fri Dec 29 16:48:03 2006 -0800 [PATCH] Update to Documentation/tty.txt on line disciplines While trying to develop a line discipline I found a couple of things worth mentioning in Documentation/tty.txt which weren't, so I decided to add them. It would be nice if someone more knowledgeable than me in that area would look over them, in case I got something wrong. Signed-off-by: Tilman Schmidt Cc: Alan Cox Cc: Theodore Ts'o Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit cfe7c09ac2be2a89aa46bb23d480d9d908e8c041 Author: Jiri Slaby Date: Fri Dec 29 16:48:00 2006 -0800 [PATCH] Char: isicom, eliminate spinlock recursion Many spinlock recursion was in the isicom driver. Eliminate it. Signed-off-by: Jiri Slaby Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 10f549fa1538849548787879d96bbb3450f06117 Author: Soeren Sonnenburg Date: Fri Dec 29 16:47:46 2006 -0800 [PATCH] make fn_keys work again on power/macbooks The apple fn keys don't work anymore with 2.6.20-rc1. The reason is that USB_HID_POWERBOOK appears in several files although USB_HIDINPUT_POWERBOOK is the thing to be used. The patch fixes this. Cc: Greg KH Cc: Dmitry Torokhov Cc: Benjamin Herrenschmidt Cc: Jiri Kosina Cc: Marcel Holtmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit d8a82d7b0a22495023e90856e58639412a5ee3f7 Author: Eric Moore Date: Fri Dec 29 16:47:43 2006 -0800 [PATCH] MAINTAINERS: email addr change for Eric Moore Update to maintainers list. My employer has changed the domain from lsil to lsi. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 9d79f1b4677cfe503f721000529f1462ca7b6f6d Author: Michael S. Tsirkin Date: Fri Dec 29 16:47:37 2006 -0800 [PATCH] IB/mthca: Fix FMR breakage caused by kmemdup() conversion Commit bed8bdfd ("IB: kmemdup() cleanup") introduced one bad conversion to kmemdup() in mthca_alloc_fmr(), where the structure allocated and the structure copied are not the same size. Revert this back to the original kmalloc()/memcpy() code. Reported-by: Dotan Barak . Signed-off-by: Michael S. Tsirkin Signed-off-by: Roland Dreier Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 31f87cf48d54f5930e6e8e827989c18b0069c94a Author: Jiri Slaby Date: Fri Dec 29 16:47:34 2006 -0800 [PATCH] Char: mxser, fix oops when removing opened tty_driver->owner is not set, so if somebody remove mxser_module, it might oops (and doesn't tell the user: no way, it's in use). Set the .owner value. Cc: Signed-off-by: Jiri Slaby Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 76e6a1d6607e5b58003bb87e3b3220d5fdb7568d Author: Randy Dunlap Date: Fri Dec 29 16:47:29 2006 -0800 [PATCH] pci/probe: fix macro that confuses kernel-doc Don't have macros between a function's kernel-doc block and the function definition. This is not valid for kernel-doc. Warning(/var/linsrc/linux-2.6.20-rc1-git8//drivers/pci/probe.c:653): No description found for parameter 'IORESOURCE_PCI_FIXED' Signed-off-by: Randy Dunlap Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 29263fbf9ecfcf294efdd8a51aa9191f423515ed Author: Thomas Meyer Date: Fri Dec 29 16:47:25 2006 -0800 [PATCH] Add .gitignore file for relocs in arch/i386 Due to the changes to make the kernel relocateable a new file is created during the build process. [jirislaby@gmail.com: The .gitigonre was intended to be in arch/ subtree] Signed-off-by: Thomas Meyer Signed-off-by: Jiri Slaby Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 96ac5913f4e45c6a1b98350f2c0a8bb3abe2646a Author: KAMEZAWA Hiroyuki Date: Fri Dec 29 16:47:20 2006 -0800 [PATCH] fix oom killer kills current every time if there is memory-less-node take2 constrained_alloc(), which is called to detect where oom is from, checks passed zone_list(). If zone_list doesn't include all nodes, it thinks oom is from mempolicy. But there is memory-less-node. memory-less-node's zones are never included in zonelist[]. contstrained_alloc() should get memory_less_node into count. Otherwise, it always thinks 'oom is from mempolicy'. This means that current process dies at any time. This patch fix it. Signed-off-by: KAMEZAWA Hiroyuki Cc: Paul Jackson Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit e4e6bdbb426d1ecd9e4587f22115f8d0d426d21f Author: Ingo Molnar Date: Fri Dec 29 16:47:14 2006 -0800 [PATCH] rcu: rcutorture suspend fix Fix suspend hang: rcutorture threads need to be nofreeze. Signed-off-by: Ingo Molnar Acked-by: Paul E. McKenney Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 52e88f5d4a6b06f3a945728dd3bc403632afe069 Author: Ingo Molnar Date: Fri Dec 29 16:47:10 2006 -0800 [PATCH] change WARN_ON back to "BUG: at ..." WARN_ON() ever triggering is a kernel bug. Do not try to paper over this fact by suggesting to the user that this is 'only' a warning, as the following recent commit does: commit 30e25b71e725b150585e17888b130e3324f8cf7c Author: Jeremy Fitzhardinge Date: Fri Dec 8 02:36:24 2006 -0800 [PATCH] Fix generic WARN_ON message A warning is a warning, not a BUG. ( it might make sense to rename BUG() to CRASH() and BUG_ON() to CRASH_ON(), but that does not change the fact that WARN_ON() signals a kernel bug. ) i and others objected to this change during lkml review: http://marc.theaimsgroup.com/?l=linux-kernel&m=116115160710533&w=2 still the change slipped upstream - grumble :) Also, use the standard "BUG: " format to make it easier to grep logs and to make it easier to google for kernel bugs. Signed-off-by: Ingo Molnar Cc: Jeremy Fitzhardinge Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 76598ebe081b04132f272440183e0aeb6507a079 Author: Judith Lebzelter Date: Fri Dec 29 16:47:06 2006 -0800 [PATCH] powerpc iseries link error in allmodconfig Choose rpa_vscsi.c over iseries_vscsi.c when building both pseries and iseries. This fixes a link error. Signed-off-by: Judith Lebzelter Acked-by: Stephen Rothwell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 1ebb1101c556b1915ff041655e629a072e64dcda Author: Zach Brown Date: Fri Dec 29 16:47:02 2006 -0800 [PATCH] Fix lock inversion aio_kick_handler() lockdep found a AB BC CA lock inversion in retry-based AIO: 1) The task struct's alloc_lock (A) is acquired in process context with interrupts enabled. An interrupt might arrive and call wake_up() which grabs the wait queue's q->lock (B). 2) When performing retry-based AIO the AIO core registers aio_wake_function() as the wake funtion for iocb->ki_wait. It is called with the wait queue's q->lock (B) held and then tries to add the iocb to the run list after acquiring the ctx_lock (C). 3) aio_kick_handler() holds the ctx_lock (C) while acquiring the alloc_lock (A) via lock_task() and unuse_mm(). Lockdep emits a warning saying that we're trying to connect the irq-safe q->lock to the irq-unsafe alloc_lock via ctx_lock. This fixes the inversion by calling unuse_mm() in the AIO kick handing path after we've released the ctx_lock. As Ben LaHaise pointed out __put_ioctx could set ctx->mm to NULL, so we must only access ctx->mm while we have the lock. Signed-off-by: Zach Brown Signed-off-by: Suparna Bhattacharya Acked-by: Benjamin LaHaise Cc: "Chen, Kenneth W" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 43cdff92ad47e0ca024c8a07d29f9bb6119e759c Author: Sebastien Dugué Date: Fri Dec 29 16:46:53 2006 -0800 [PATCH] Fix IPMI watchdog set_param_str() using kstrdup set_param_str() cannot use kstrdup() to duplicate the parameter. That's fine when the driver is compiled as a module but it sure is not when built into the kernel as the kernel parameters are parsed before the kmalloc slabs are setup. Signed-off-by: Sebastien Dugué Acked-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds commit 97d552e35d9404df3254e1157df3340e4e2eaedc Author: Stefan Richter Date: Fri Dec 29 23:47:04 2006 +0100 ieee1394: sbp2: fix bogus dma mapping Need to use a PCI device, not a FireWire host device. Problem found by Andreas Schwab, mistake pointed out by Benjamin Herrenschmidt. http://ozlabs.org/pipermail/linuxppc-dev/2006-December/029595.html Signed-off-by: Stefan Richter Tested-by: Andreas Schwab commit b2bb550c4a10c44e99fe469cfaee81e2e3109994 Author: Stefan Richter Date: Thu Dec 28 19:57:49 2006 +0100 ieee1394: sbp2: pass REQUEST_SENSE through to the target Delete some incorrect code, left over from the initial driver submission in March 2001. SBP-2 targets should provide sense data via the SBP-2 status block (autosense). We have to pass the REQUEST_SENSE command through to targets which don't implement autosense, if there are any, and to accomodate application clients which use this command. Signed-off-by: Stefan Richter commit 7658cc289288b8ae7dd2c2224549a048431222b3 Author: Linus Torvalds Date: Fri Dec 29 10:00:58 2006 -0800 VM: Fix nasty and subtle race in shared mmap'ed page writeback The VM layer (on the face of it, fairly reasonably) expected that when it does a ->writepage() call to the filesystem, it would write out the full page at that point in time. Especially since it had earlier marked the whole page dirty with "set_page_dirty()". But that isn't actually the case: ->writepage() does not actually write a page, it writes the parts of the page that have been explicitly marked dirty before, *and* that had not got written out for other reasons since the last time we told it they were dirty. That last caveat is the important one. Which _most_ of the time ends up being the whole page (since we had called "set_page_dirty()" on the page earlier), but if the filesystem had done any dirty flushing of its own (for example, to honor some internal write ordering guarantees), it might end up doing only a partial page IO (or none at all) when ->writepage() is actually called. That is the correct thing in general (since we actually often _want_ only the known-dirty parts of the page to be written out), but the shared dirty page handling had implicitly forgotten about these details, and had a number of cases where it was doing just the "->writepage()" part, without telling the low-level filesystem that the whole page might have been re-dirtied as part of being mapped writably into user space. Since most of the time the FS did actually write out the full page, we didn't notice this for a loong time, and this needed some really odd patterns to trigger. But it caused occasional corruption with rtorrent and with the Debian "apt" database, because both use shared mmaps to update the end result. This fixes it. Finally. After way too much hair-pulling. Acked-by: Nick Piggin Acked-by: Martin J. Bligh Acked-by: Martin Michlmayr Acked-by: Martin Johansson Acked-by: Ingo Molnar Acked-by: Andrei Popa Cc: High Dickins Cc: Andrew Morton , Cc: Peter Zijlstra Cc: Segher Boessenkool Cc: David Miller Cc: Arjan van de Ven Cc: Gordon Farquharson Cc: Guillaume Chazarain Cc: Theodore Tso Cc: Kenneth Cheng Cc: Tobias Diedrich Signed-off-by: Linus Torvalds commit 92efc15241ceebc23451691971897020e8563a70 Author: Zhen Wei Date: Fri Dec 8 00:48:17 2006 -0700 ocfs2: export heartbeat thread pid via configfs The patch allows the ocfs2 heartbeat thread to prioritize I/O which may help cut down on spurious fencing. Most of this will be in the tools - we can have a pid configfs attribute and let userspace (ocfs2_hb_ctl) calls the ioprio_set syscall after starting heartbeat, but only cfq scheduler supports I/O priorities now. Signed-off-by: Zhen Wei Signed-off-by: Mark Fasheh commit 7f4a2a97e324e8c826d1d983bc8efb5c59194f02 Author: Mark Fasheh Date: Mon Dec 11 11:06:36 2006 -0800 ocfs2: always unmap in ocfs2_data_convert_worker() Mmap-heavy clustered workloads were sometimes finding stale data on mmap reads. The solution is to call unmap_mapping_range() on any down convert of a data lock. Signed-off-by: Mark Fasheh commit 6c2aad0567e693f9588d0a0683f96ed872fb4641 Author: Mark Fasheh Date: Tue Dec 19 15:25:52 2006 -0800 ocfs2: ignore NULL vfsmnt in ocfs2_should_update_atime() This can come from NFSD. Signed-off-by: Mark Fasheh commit 564f8a3228879d6962edb3432d01bcd7499a67ec Author: Mark Fasheh Date: Thu Dec 14 13:01:05 2006 -0800 ocfs2: Allow direct I/O read past end of file ocfs2_direct_IO_get_blocks() was incorrectly returning -EIO for a direct I/O read whose start block was past the end of the file allocation tree. Fix things so that we return a hole instead. do_direct_IO() will then notice that the range start is past eof and return a short read. While there, remove the unused vbo_max variable. Signed-off-by: Mark Fasheh commit 0333394bff439c3fb09264303de42e7038b3e709 Author: Mark Fasheh Date: Thu Dec 14 15:29:25 2006 -0800 ocfs2: don't print error in ocfs2_permission() Errors from generic_permission() can happen in valid cases and shouldn't be reported. Signed-off-by: Mark Fasheh commit aa77015c4e94cb1d30680646c163d7ae1f93f941 Author: Michael Holzheu Date: Thu Dec 28 00:35:36 2006 +0100 [S390] cio: fix stsch_reset. Copy inline assembly of stsch and add "memory" to clobber list in order to prevent gcc from optimizing away the checking of the global variable "pgm_check_occured". Signed-off-by: Michael Holzheu Signed-off-by: Martin Schwidefsky commit 524a237e4512038d6e07ad3b2e44e70902b76738 Author: Melissa Howland Date: Thu Dec 28 00:35:25 2006 +0100 [S390] Change max. buffer size for monwriter device. Reduce the max. buffer size for the monwriter device to prevent a possible problem with the z/VM monitor service. Signed-off-by: Melissa Howland Signed-off-by: Martin Schwidefsky commit 69f7e75a9d45e5eaca16917a8d0dedf76149f13f Author: Jean Delvare Date: Sat Dec 23 14:11:03 2006 -0300 V4L/DVB (5010): Cx88: Fix leadtek_eeprom tagging reference to .init.text: from .text between 'cx88_card_setup' (at offset 0x68c) and 'cx88_risc_field' Caused by leadtek_eeprom() being declared __devinit and called from a non-devinit context. Signed-off-by: Jean Delvare Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab commit 70bdd9c83a6e757e5fb9cc3195f27d2c1d0a4c39 Author: Alexey Dobriyan Date: Tue Dec 26 07:43:25 2006 -0300 V4L/DVB (5012): Usbvision fix: It was using "&&" instead "&" Signed-off-by: Alexey Dobriyan Signed-off-by: Mauro Carvalho Chehab commit 0b778a56e342e0e1d274298ad47d9af22a9de633 Author: Mauro Carvalho Chehab Date: Wed Dec 27 14:04:09 2006 -0200 V4L/DVB (5001): Add two required headers on kernel 2.6.20-rc1 include/media/ir-common.h:78: error: field 'work' has incomplete type drivers/media/common/ir-functions.c: In function 'ir_rc5_timer_end': drivers/media/common/ir-functions.c:301: error: 'jiffies' undeclared (first use in this function) drivers/media/common/ir-functions.c:301: error: (Each undeclared identifier is reported only once) drivers/media/common/ir-functions.c:301: error: for each function it appears in.) drivers/media/common/ir-functions.c:347: error: 'HZ' undeclared (first use in this function) Signed-off-by: Mauro Carvalho Chehab commit 9a2816c1c4ca7f5d02de4339589913701251dd5a Author: David Brownell Date: Wed Dec 27 09:53:44 2006 -0300 V4L/DVB (5014): Allyesconfig build fixes on some non x86 arch - CAFE_CCIC needs to depend on PCI, else "allyesconfig" breaks on systems without PCI - em28xx-video can't udelay(2500) else "allyesconfig" breaks on systems that refuse to spin that long (I saw it on ARM) Signed-off-by: David Brownell Signed-off-by: Mauro Carvalho Chehab commit 5f1693fe82b499a3306993020ef14011ed0420d0 Author: Akinobu Mita Date: Wed Dec 20 10:08:56 2006 -0300 V4L/DVB (4997): Bttv: delete duplicated ioremap() ioremap() is called twice to same resource. The returen value of first one is not error-checked. second one is complely ignored. Signed-off-by: Akinobu Mita Signed-off-by: Mauro Carvalho Chehab commit 25821400359d2b3daacfb93303944211b8411ac3 Author: Akinobu Mita Date: Wed Dec 20 10:07:30 2006 -0300 V4L/DVB (4996): Msp3400: fix kthread_run error check The return value of kthread_run() should be checked by IS_ERR(). Signed-off-by: Akinobu Mita Signed-off-by: Mauro Carvalho Chehab commit 054afee4736677a00982f6414a983dec8ee511db Author: Akinobu Mita Date: Wed Dec 20 10:04:00 2006 -0300 V4L/DVB (4995): Vivi: fix kthread_run() error check The return value of kthread_run() should be checked by IS_ERR(). Signed-off-by: Akinobu Mita Signed-off-by: Mauro Carvalho Chehab commit 72f678c30185df18ac38ef14c9045d139c2263bc Author: Akinobu Mita Date: Wed Dec 20 10:03:53 2006 -0300 V4L/DVB (4994): Vivi: fix use after free in list_for_each() Freeing data including list_head in list_for_each() is not safe. Signed-off-by: Akinobu Mita Signed-off-by: Mauro Carvalho Chehab commit b331daa03cd949b1b0a4b761b4b4a92d71f7d626 Author: Stephan Berberig Date: Wed Dec 20 09:37:05 2006 -0300 V4L/DVB (4992): Fix typo in saa7134-dvb.c Fix a typo (use_frontent -> use_frontend) in saa7134-dvb.c. Signed-off-by: Stephan Berberig Signed-off-by: Mauro Carvalho Chehab commit d4f60baf7827c6a46b7ee2ed0a2a32a373febfcb Author: Adrian Bunk Date: Wed Dec 20 09:34:32 2006 -0300 V4L/DVB (4991): Cafe_ccic.c: fix NULL dereference We shouldn't dereference "cam" when we already know it's NULL. Spotted by the Coverity checker. Signed-off-by: Adrian Bunk Signed-off-by: Mauro Carvalho Chehab commit d82d418a3af4d7d07151f9d45ca20f2ce61289a0 Author: Amit Choudhary Date: Wed Dec 20 09:30:45 2006 -0300 V4L/DVB (4990): Cpia2/cpia2_usb.c: fix error-path leak Free previously allocated memory (in array elements) if kmalloc() returns NULL in submit_urbs(). Signed-off-by: Amit Choudhary Signed-off-by: Andrew Morton Signed-off-by: Mauro Carvalho Chehab commit cb2c7b4927c8f376b7ba9557978d8c59ed472664 Author: Hans Verkuil Date: Wed Dec 20 06:51:37 2006 -0300 V4L/DVB (4988): Cx2341x audio_properties is an u16, not u8 This bug broke the MPEG audio mode controls. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab commit 83aaf13c5ba5e1e244339807d8a752808e763cdd Author: Hans Verkuil Date: Mon Dec 18 13:40:23 2006 -0300 V4L/DVB (4984): LOG_STATUS should show the real temporal filter value. The temporal filter is forced off when scaling. The VIDIOC_LOG_STATUS handler still showed the old temporal filter. It is now consistent with the real temporal filter value. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab commit 12b896e436a3b1bc83856962388ae1ab154742a6 Author: Hans Verkuil Date: Mon Dec 18 13:37:50 2006 -0300 V4L/DVB (4983): Force temporal filter to 0 when scaling to prevent ghosting. Change the code to unconditionally turn off the temporal filter when scaling. If the window is not full screen the filter will introduce a nasty ghosting effect. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab commit b331def2d382d7a51c379f336fe80ef87d6674e3 Author: Hans Verkuil Date: Mon Dec 18 13:18:28 2006 -0300 V4L/DVB (4982): Fix broken audio mode handling for line-in in msp3400. The wrong matrix was used when an external input was selected instead of the tuner input. The rxsubchans field was also not initialized to STEREO for an external input. And finally the msp34xxg_detect_stereo() should not try to detect stereo for an external input, that code is for the tuner input only. Together these bugs made it hit 'n miss whether you ever got stereo out of the msp3400 for an external input. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab commit 315eb962d2e9438bc10da2488b604f04a1c0006f Author: Mauro Carvalho Chehab Date: Sun Dec 17 23:30:47 2006 -0300 V4L/DVB (4980): Fixes bug 7267: PAL/60 is not working On cx88 driver, sampling rate should be at chroma subcarrier freq (FSC). However, driver were programming wrong values for PAL/60, PAL/Nc and NTSC 4.43. This patch do the proper calculation. It also calculates htotal, hdelay and hactive constants, according with the sampling rate. It is tested with PAL/60 by Piotr Maksymuk and Olivier. Also tested with the already-supported standards. Test is still required for PAL/Nc. Signed-off-by: Mauro Carvalho Chehab commit ed00b41dc8bc286682d31ad64060ccc70513e90b Author: Dwaine Garden Date: Wed Dec 27 10:23:28 2006 -0200 V4L/DVB (4979): Fixes compilation when CONFIG_V4L1_COMPAT is not selected - SYSFS: Replaced all to_video_device(cd), video_device_create_file, video_device_remove_file and add the proper checks at create_file - Converted old norm values to V4L2 ones. - Robustness on sysfs hue/contrast/saturation queries. Additional check in order to return 0 if the driver is not opened. - Whitespace cleanups in usbvision-cards.c This patch merges two fixes by Thierry MERLE and Mauro Chehab, and adds additional checks. Signed-off-by: Dwaine Garden Signed-off-by: Mauro Carvalho Chehab commit 5ef35be42dfaa155e02718a9bd34cc341f327883 Author: Michael Krufky Date: Thu Dec 14 13:53:33 2006 -0300 V4L/DVB (4973): Dvb-core: fix printk type warning dvb_net.c: In function 'dvb_net_ule': dvb_net.c:628: warning: format '%#lx' expects type 'long unsigned int', but argument 3 has type 'u32' dvb_net.c:628: warning: format '%#lx' expects type 'long unsigned int', but argument 4 has type 'u32' Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab commit dedcefb085fe98a1feaf63590fe2fc7e0ecb1987 Author: Ang Way Chuang Date: Thu Dec 14 13:51:44 2006 -0300 V4L/DVB (4972): Dvb-core: fix bug in CRC-32 checking on 64-bit systems CRC-32 checking during ULE decapsulation always failed on x86_64 systems due to the size of a variable used to store CRC. This bug was discovered on Fedora Core 6 with kernel-2.6.18-1.2849. The i386 counterpart has no such problem. This patch has been tested on 64-bit system as well as 32-bit system. Signed-off-by: Ang Way Chuang Signed-off-by: Michael Krufky Signed-off-by: Mauro Carvalho Chehab commit 38284ba361d69eca34a3bfc553ebfac81fea2698 Author: Thierry MERLE Date: Fri Dec 15 16:46:53 2006 -0300 V4L/DVB (4970): Usbvision memory fixes - fix decompression buffer allocation not done at first driver open - simplification of USB sbuf allocation (use of usb_buffer_alloc) - replaced vmalloc by vmalloc_32 (for homogeneity) - add of saa7111 (i2cAddr=0x48) detection printout in attach_inform Signed-off-by: Thierry MERLE Signed-off-by: Mauro Carvalho Chehab commit 3a4456a073150c8b0d790daa007d2aab9ebdecb5 Author: Hans Verkuil Date: Tue Dec 12 16:27:28 2006 -0300 V4L/DVB (4968): Add PAL-60 support for cx2584x. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab commit e71ced1a5dfebfde57c0ed2c9bd3c7b4596071c1 Author: Hans Verkuil Date: Mon Dec 11 15:51:43 2006 -0300 V4L/DVB (4967): Add missing tuner module option pal=60 for PAL-60 support. Signed-off-by: Hans Verkuil Signed-off-by: Mauro Carvalho Chehab commit 2485eb0a553f2ac8c73267ce48fcdee8a9728d43 Author: audetto@tiscali.it Date: Tue Dec 12 10:35:57 2006 -0300 V4L/DVB (4964): VIDEO_PALETTE_YUYV and VIDEO_PALETTE_YUV422 are the same palette Consistent handling of VIDEO_PALETTE_YUYV and VIDEO_PALETTE_YUV422 Signed-off-by: Andrea A Odetti Signed-off-by: Mauro Carvalho Chehab commit c6268461473e1c62e31afde657069c050ee8dea7 Author: Mauro Carvalho Chehab Date: Mon Dec 11 17:18:15 2006 -0300 V4L/DVB (4960): Removal of unused code from usbvision-i2c.c i2c_adap is almost not used. This patch removes it, cleaning the i2c support, and improving driver understanding. Thanks to Thierry Merle for testing it. Signed-off-by: Mauro Carvalho Chehab commit 0a0ceadebd4551c26110f5daf0b38d9054674c9e Author: Adrian Bunk Date: Mon Dec 11 15:51:36 2006 -0300 V4L/DVB (4959): Usbvision: possible cleanups This patch contains the following possible cleanups: - make needlessly global functions static - remove the unused EXPORT_SYMBOL's Signed-off-by: Adrian Bunk Signed-off-by: Mauro Carvalho Chehab commit bee8a44ecb37c1d80afd2af932f60e786f0eabc2 Author: Ralf Baechle Date: Mon Dec 11 11:14:54 2006 -0300 V4L/DVB (4958): Fix namespace conflict between w9968cf.c on MIPS Both use __SC. Since __* is sort of private namespace I've choosen to fix this in the driver. For consistency I decieded to also change __UNSC to UNSC. Signed-off-by: Ralf Baechle Signed-off-by: Andrew Morton Signed-off-by: Mauro Carvalho Chehab commit 6ccd60172e271d156dba8a4002568d3bf1912e01 Author: Mario Rossi Date: Mon Dec 11 05:42:36 2006 -0300 V4L/DVB (4956): [NOVA-T-USB2] Put remote-debugging in the right place This patch removes unnecessary (and misleading) debug output (it printed the values of the keys in the table up to the value of the key pressed). Signed-off-by: Mario Rossi Signed-off-by: Patrick Boettcher Signed-off-by: Mauro Carvalho Chehab commit a821e990e4db77cd87bec488fb425d8d4ced79bd Author: Mario Rossi Date: Mon Dec 11 05:37:31 2006 -0300 V4L/DVB (4955): Fix autosearch index After rewriting the driver the wrong autosearch index was used when COFDM-parameter needed to be detected. Thanks to Mario Rossi who found it. Signed-off-by: Mario Rossi Signed-off-by: Patrick Boettcher Signed-off-by: Mauro Carvalho Chehab commit 81f4e6c190a0fa016fd7eecaf76a5f95d121afc2 Author: Francois Romieu Date: Mon Dec 18 00:00:55 2006 +0100 r8169: extraneous Cmd{Tx/Rx}Enb write Checked in Realtek's driver, this one has no business being there. The driver still works but there is a noticeable performance drop. Signed-off-by: Francois Romieu commit 87046e5062c483820f20c1fa6ed0826bb38fd4f0 Author: Ayaz Abdulla Date: Tue Dec 19 23:33:32 2006 -0500 forcedeth: modified comment header This patch removes comment that forcedeth is not supported by NVIDIA. Signed-Off-By: Ayaz Abdulla Signed-off-by: Jeff Garzik commit bd56c6b15e809d0fcb43bc8d505ca4d15debac62 Author: Amit S. Kale Date: Mon Dec 18 05:54:36 2006 -0800 NetXen: Reducing ring sizes for IOMMU issue. Signed-off-by: Amit S. Kale netxen_nic.h | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-) Signed-off-by: Jeff Garzik commit ed33ebe464d973ab656cfc9ff0ca40a4c1e38238 Author: Amit S. Kale Date: Mon Dec 18 05:53:59 2006 -0800 NetXen: Fix for PPC machines. Signed-off-by: Amit S. Kale netxen_nic.h | 2 +- netxen_nic_init.c | 12 ++++++------ netxen_nic_main.c | 4 ++-- 3 files changed, 9 insertions(+), 9 deletions(-) Signed-off-by: Jeff Garzik commit 6c5866441669ab71f6297556ea281a6bcbfd13db Author: Amit S. Kale Date: Mon Dec 18 05:53:36 2006 -0800 NetXen: work queue fixes. Signed-off-by: Amit S. Kale netxen_nic.h | 3 +-- netxen_nic_init.c | 2 +- netxen_nic_main.c | 15 +++++++-------- 3 files changed, 9 insertions(+), 11 deletions(-) Signed-off-by: Jeff Garzik commit d2378e89720f364ccf3a57b26f1827923d3464f2 Author: Amit S. Kale Date: Mon Dec 18 05:53:12 2006 -0800 NetXen: Link status message correction for quad port cards. Signed-off-by: Amit S. Kale netxen_nic_isr.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletion(-) Signed-off-by: Jeff Garzik commit a379cb3c136526617c7aed555a249cb03e951e60 Author: Amit S. Kale Date: Mon Dec 18 05:52:39 2006 -0800 NetXen: Multiple adapter fix. Signed-off-by: Amit S. Kale netxen_nic.h | 3 +-- netxen_nic_main.c | 12 ------------ 2 files changed, 1 insertion(+), 14 deletions(-) Signed-off-by: Jeff Garzik commit c75e86b47f97831732fc450fcc5a4230b00071fb Author: Amit S. Kale Date: Mon Dec 18 05:51:58 2006 -0800 NetXen: Using correct CHECKSUM flag. Signed-off-by: Amit S. Kale netxen_nic_hw.c | 2 +- 1 files changed, 1 insertion(+), 1 deletion(-) Signed-off-by: Jeff Garzik commit b0d541dbb940fe8899e3020fcc9ee5a1e2be0306 Author: Amit S. Kale Date: Mon Dec 18 05:51:29 2006 -0800 NetXen: driver reload fix for newer firmware. Signed-off-by: Amit S. Kale netxen_nic_main.c | 7 +++++++ 1 files changed, 7 insertions(+) Signed-off-by: Jeff Garzik commit 0d39073484d6e01d325e6d94a3967ce0e98161a7 Author: Amit S. Kale Date: Mon Dec 18 05:50:59 2006 -0800 NetXen: Adding new device ids. Signed-off-by: Amit S. Kale netxen_nic_main.c | 2 ++ 1 files changed, 2 insertions(+) Signed-off-by: Jeff Garzik commit 3e65bb94a9de542c464bf00035a08b0e0a76993d Author: Hynek Petrak Date: Tue Dec 19 13:08:49 2006 -0800 PHY probe not working properly for ibm_emac (PPC4xx) I have a system with AMCC PowerPC 405EP and PHY Intel LXT971A. Linux 2.6.18.3 is not able to detect the PHY ID correctly. The PHY ID detected is 0, but should be 0x1d. This is because phy_read() (__emac_mdio_read() resp.) from drivers/net/ibm_emac/ibm_emac_core.c might return -ETIMEDOUT or -EREMOTEIO on error. This is ignored inside the int mii_phy_probe(struct mii_phy *phy, int address) from drivers/net/ibm_emac/ibm_emac_phy.c as the return value is assigned to an u32 variable. Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik commit ebf5112ca7a714a3961a322459747059cbf41d7b Author: Yan Burman Date: Tue Dec 19 13:08:48 2006 -0800 ep93xx: some minor cleanups to the ep93xx eth driver Small cleanup in the Cirrus Logic EP93xx ethernet driver: Check for NULL pointer before dereferencing it instead of after. Remove unreferenced variable. Signed-off-by: Yan Burman Cc: Jeff Garzik Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik commit aed2cec45c426c941c80d84ec89ce34f2f2607cc Author: Stephen Hemminger Date: Wed Dec 20 13:06:35 2006 -0800 sky2: phy power down needs PCI config write enabled In order to change PCI registers (via the iomap'd window), it needs to be enabled; this wasn't being done in sky2_phy_power the function that turns on/off power to the PHY. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik commit 44ade178249fe53d055fd92113eaa271e06acddd Author: Stephen Hemminger Date: Wed Dec 20 13:06:34 2006 -0800 sky2: power management/MSI workaround MSI doesn't work properly on resume on many platforms because the BIOS goes and changes it back to INTx mode after the sky2 driver has restored in resume. It is really a bug in the base power management resume code, and this workaround is temporary until the change to PM code works it's way through the release process. The PM fix is non-trivial since it needs to change when non-boot CPU's are enabled. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik commit 25d82d7a2fc20cc4be414d738139594dc0adf179 Author: Stephen Hemminger Date: Wed Dec 20 13:06:33 2006 -0800 sky2: dual port NAPI problem Shutting down port 0 disables the NAPI poll used by both ports. The long term fix will be to separate NAPI object from net device until then just reenable if needed. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik commit ce9f7fe3c3f5b9ed1d3e91db78c45d0df5672816 Author: Randy Dunlap Date: Mon Dec 18 21:21:10 2006 -0800 via-velocity uses INET interfaces via-velocity doesn't build when CONFIG_INET=n: drivers/built-in.o: In function `velocity_unregister_notifier': via-velocity.c:(.text+0xe9b46): undefined reference to `unregister_inetaddr_notifier' drivers/built-in.o: In function `velocity_init_module': via-velocity.c:(.init.text+0xa027): undefined reference to `register_inetaddr_notifier' I wanted to make this change in drivers/net/Kconfig, but this isn't legal kconfig language: config VIA_VELOCITY tristate "VIA Velocity support" depends on NET_PCI && PCI + depends on INET if PM select CRC32 select CRC_CCITT select MII so fix it in via-velocity.c instead. Builds with all 4 combinations of CONFIG_NET & CONFIG_PM. Signed-off-by: Randy Dunlap Signed-off-by: Jeff Garzik commit 683a2aa339f607c8a422835161ceab68b2a5a18a Author: Herbert Xu Date: Sat Dec 16 12:04:33 2006 +1100 e1000: Do not truncate TSO TCP header with 82544 workaround The e1000 driver has a workaround for 82544 on PCI-X where if the terminating byte of a buffer is at addresses 0-3 mod 8, then 4 bytes are shaved off it and defered to a new segment. This is due to an erratum that could otherwise cause TX hangs. Unfortunately this breaks TSO because it may cause the TCP header to be split over two segments which itself causes TX hangs. The solution is to pull 4 bytes of data up from the next segment rather than pushing 4 bytes off. This ensures the TCP header remains in one piece and works around the PCI-X hang. This patch is based on one from Jesse Brandeburg. This bug has been trigered by both CONFIG_DEBUG_SLAB as well as Xen. Note that the only reason we don't see this normally is because the TCP stack starts writing from the end, i.e., it writes the TCP header first then slaps on the IP header, etc. So the end of the TCP header (skb->tail - 1 here) is always aligned correctly. Had we made the start of the IP header (e.g., IPv6) 8-byte aligned instead, this would happen for normal TCP traffic as well. Signed-off-by: Herbert Xu Acked-by: Jesse Brandeburg -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt -- Signed-off-by: Jeff Garzik commit 1a63e846a4099e6fbff86a3b112064378515f254 Author: Brice Goglin Date: Mon Dec 18 11:52:34 2006 +0100 myri10ge: handle failures in suspend and resume On suspend, handle pci_set_power_state errors, and on resume handle failures in pci_resume_state(). Signed-off-by: Brice Goglin Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik commit 83f6e1524586119df84f70b7127688c57349664a Author: Brice Goglin Date: Mon Dec 18 11:52:02 2006 +0100 myri10ge: no need to save MSI and PCIe state in the driver The PCI MSI and express state are already saved and restored by the current versions of pci_save_state/pci_restore_state. Therefore it is no longer necessary for the driver to do it. Signed-off-by: Brice Goglin Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik commit 3621cec5b5847348248e749c647219f409550f2f Author: Brice Goglin Date: Mon Dec 18 11:51:22 2006 +0100 myri10ge: make msi configurable at runtime through sysfs Now that IRQ are requested is called on open() and freed on close(), we can safely switch from/to MSI without unloading the module. We are guaranteed to correctly free IRQ even if the sysfs file got written in the meantime since the MSI initialization is stored in mgp->msi_enabled. Signed-off-by: Brice Goglin Signed-off-by: Jeff Garzik commit df30a740e4647b035e44a94e6977ad42a94cb850 Author: Brice Goglin Date: Mon Dec 18 11:50:40 2006 +0100 myri10ge: move request_irq to myri10ge_open Request IRQ in myri10ge_open() and free in close() instead of probe() and remove() to eliminate potential race between the watchdog and the interrupt handler. Additionaly, the interrupt handler won't get called on shared irq anymore when the interface is down. Signed-off-by: Brice Goglin Signed-off-by: Jeff Garzik commit 7adda30c82f403c948b016a40fc68a1ef427dbca Author: Brice Goglin Date: Mon Dec 18 11:50:00 2006 +0100 myri10ge: match number of save_state and restore Since pci_save_state() pushes MSI and PCIe states on a kind of stack, myri10ge saving the state in advance for parity recovery will push the state again on the stack on suspend. This leads to some memory leak. We add a couple additional calls to save_state and restore_state so that we don't leak anymore. For the future, we are thinking of a better way to recover from parity error without using pci_save_state(). Signed-off-by: Brice Goglin Signed-off-by: Jeff Garzik commit a27993f3d9daca0dffa26577a83822db99c952e2 Author: Francois Romieu Date: Mon Dec 18 00:04:19 2006 +0100 r8169: use the broken_parity_status field in pci_dev The former option is removed and platform code can now specify the expected behavior. Signed-off-by: Lennert Buytenhek Signed-off-by: Francois Romieu Signed-off-by: Jeff Garzik commit d15e9c4d9a75702b30e00cdf95c71c88e3f3f51e Author: Francois Romieu Date: Sun Dec 17 23:03:15 2006 +0100 netpoll: drivers must not enable IRQ unconditionally in their NAPI handler net/core/netpoll.c::netpoll_send_skb() calls the poll handler when it is available. As netconsole can be used from almost any context, IRQ must not be enabled blindly in the NAPI handler of a driver which supports netpoll. b57bd06655a028aba7b92e1c19c2093e7fcfb341 fixed the issue for the 8139too.c driver. Signed-off-by: Francois Romieu Signed-off-by: Jeff Garzik commit 92104e94806f47e090e3bb40f6ae45f695351a5e Author: Stephen Hemminger Date: Fri Dec 15 07:57:08 2006 -0800 netxen: remove private ioctl The netxen driver includes a private ioctl that provides access to functionality that is already available in other ways. The PCI layer has application access hooks (see setpci), and the statistics are available in ethtool/netstats. Signed-off-by: Stephen Hemminger Signed-off-by: Jeff Garzik commit 79f3d3996f06ee339c6f173e573826eccd3914ab Author: Jesse Brandeburg Date: Fri Dec 15 10:42:34 2006 +0100 [PATCH] e1000: No-delay link detection at interface up Currently after an interface up, the link state is detected 2 seconds later when the first watchdog timer runs. This patch changes that by triggering the hardware to generate a link-change interrupt from the up() function instead. This has the result that the link state gets detected immediately and without races. This has the potential to speed up booting since a normal distribution boot process waits for a link before DHCP is attempted. Signed-off-by: Jesse Brandeburg Signed-off-by: Auke Kok Signed-off-by: Arjan van de Ven Signed-off-by: Jeff Garzik commit 15e376b4eed2575b323d575403569ec1b5f63fda Author: Jeff Garzik Date: Fri Dec 15 11:16:33 2006 -0500 e1000: 3 new driver stats for managability testing Add 3 extra packet redirect counters for tracking purposes to make sure we can test that all packets arrive properly. Originally from Jesse Brandeburg , rewritten to use feature flags by me. Signed-off-by: Jeff Garzik commit 1f753861d272e648f3234e4f5fdea9434ae7bee4 Author: Jesse Brandeburg Date: Fri Dec 15 10:40:39 2006 +0100 [PATCH] e1000: Make the copybreak value a module parameter Allow the user to vary the size that copybreak works. Currently cb is enabled for packets < 256 bytes, but various tests indicate that this should be configurable for specific use cases. In addition, this parameter allows us to force never/always during testing to get full and predictable coverage of both code paths. Signed-off-by: Jesse Brandeburg Signed-off-by: Auke Kok Signed-off-by: Arjan van de Ven Signed-off-by: Jeff Garzik commit 018ea44ef1eade417296c4a57afe3cd963268433 Author: Bruce Allan Date: Fri Dec 15 10:39:45 2006 +0100 [PATCH] e1000: Fix PBA allocation calculations Assign the PBA to be large enough to contain at least 2 jumbo frames on all adapters. This dramatically increases performance on several adapters and fixes TX performance degradation issues where the PBA was misallocated in the old algorithm. Signed-off-by: Bruce Allan Signed-off-by: Auke Kok Signed-off-by: Arjan van de Ven Signed-off-by: Jeff Garzik commit d89b6c6750e7d7527603b573ec60ba787f5c04a6 Author: Jesse Brandeburg Date: Fri Dec 15 10:38:32 2006 +0100 [PATCH] e1000: narrow down the scope of the tipg timer tweak the driver has (ancient) code for messing with TIPG from the 82542 days. Unfortunately this code was running on our current adapters and setting TIPG for fiber to be +1 over the copper value. This caused 1.45Mpps to be sent instead of 1.487Mpps. Signed-off-by: Jesse Brandeburg Signed-off-by: Auke Kok Signed-off-by: Arjan van de Ven Signed-off-by: Jeff Garzik commit c3813ae6615107cc93c79200f477ffd2a870c8ab Author: Jeff Kirsher Date: Fri Dec 15 10:37:32 2006 +0100 [PATCH] e1000: fix ethtool reported bus type for older adapters For older adapters we know that they are of the PCI bus type, so we can just set this. Signed-off-by: Jeff Kirsher Signed-off-by: Auke Kok Signed-off-by: Arjan van de Ven Signed-off-by: Jeff Garzik commit 83cd827977d8d20ac8c4fe56c88b53a70ac54af3 Author: Bruce Allan Date: Fri Dec 15 10:36:35 2006 +0100 [PATCH] e1000: fix to set the new max frame size before resetting the adapter This bugfix makes sure that the driver data reflects the full new situation before the adapter is reinitialized. Signed-off-by: Bruce Allan Signed-off-by: Auke Kok Signed-off-by: Arjan van de Ven Signed-off-by: Jeff Garzik commit bb8e3311ef9de8e72f45f910e4a977c313c7009c Author: Jeff Garzik Date: Fri Dec 15 11:06:17 2006 -0500 e1000: workaround for the ESB2 NIC RX unit issue In rare occasions, ESB2 systems would end up started without the RX unit being turned on. Add a check that runs post-init to work around this issue. Originally from Jesse Brandeburg , rewritten to use feature flags by me. Signed-off-by: Jeff Garzik commit 72f3ab7462f4e153d1e8ac78e379716ad71d6923 Author: Jesse Brandeburg Date: Fri Dec 15 10:34:46 2006 +0100 [PATCH] e1000: disable TSO on the 82544 with slab debugging CONFIG_DEBUG_SLAB changes alignments of the data structures the slab allocators return. These break certain workarounds for TSO on the 82544. Since DEBUG_SLAB is relatively rare and not used for performance sensitive cases, the simplest fix is to disable TSO in this special situation. Signed-off-by: Jesse Brandeburg Signed-off-by: Auke Kok Signed-off-by: Arjan van de Ven Signed-off-by: Jeff Garzik commit 3d5460a0ba17437cf278ee46886d557340d028da Author: Jesse Brandeburg Date: Fri Dec 15 10:33:46 2006 +0100 [PATCH] e1000: Fix Wake-on-Lan with forced gigabit speed If the user has forced gigabit speed, phy power management must be disabled; otherwise the NIC would try to negotiate to a linkspeed of 10/100 mbit on shutdown, which would lead to a total loss of link. This loss of link breaks Wake-on-Lan and IPMI. Signed-off-by: Jesse Brandeburg Signed-off-by: Auke Kok Signed-off-by: Arjan van de Ven Signed-off-by: Jeff Garzik commit 0fccd0e9e3802748855d967940c16f8c59d4e2b6 Author: Jeff Garzik Date: Fri Dec 15 10:56:10 2006 -0500 e1000: consolidate managability enabling/disabling Several bugs existed in how we handle manageability issues all over the driver. This patch consolidates all the managability release and init code in two single functions and call them from appropriate locations. This fixes several BMC packet redirect issues and powerup/down hiccups. Originally from Jesse Brandeburg , rewritten to use feature flags by me. Signed-off-by: Jeff Garzik commit 167fb2841633edd2812e385af4b593d870abd15a Author: Jeff Garzik Date: Fri Dec 15 10:41:15 2006 -0500 e1000: omit stats for broken counter in 82543 The 82543 chip does not count tx_carrier_errors properly in FD mode; report zeros instead of garbage. Originally from Jesse Brandeburg , rewritten to use feature flags by me. Signed-off-by: Jeff Garzik commit bd2371ebcc71d0a276ae341d735326a8beab6627 Author: Jeff Garzik Date: Fri Dec 15 10:31:40 2006 -0500 e1000: For sanity, reformat e1000_set_mac_type(), struct e1000_hw[_stats] Makes future changes a bit more readable. Signed-off-by: Jeff Garzik commit 2b65326e67f89899e8bcaed1989d8cfb0ed01f55 Author: Jesse Brandeburg Date: Fri Dec 15 10:30:44 2006 +0100 [PATCH] e1000: dynamic itr: take TSO and jumbo into account The dynamic interrupt rate control patches omitted proper counting for jumbo's and TSO resulting in suboptimal interrupt mitigation strategies. Signed-off-by: Jesse Brandeburg Signed-off-by: Auke Kok Signed-off-by: Arjan van de Ven Signed-off-by: Jeff Garzik commit 7d16e65ba57f181732ec52626736b27904198edf Author: Jesse Brandeburg Date: Fri Dec 15 10:29:31 2006 +0100 [PATCH] e1000: The user-supplied itr setting needs the lower 2 bits masked off The lower 2 bits of a user-supplied itr setting (via ethtool) need to be masked off: These lower two bits are used as control bits. Signed-off-by: Jesse Brandeburg Signed-off-by: Auke Kok Signed-off-by: Arjan van de Ven Signed-off-by: Jeff Garzik commit 0c234ae655a45ac3ee53a25b2e56e9bb6c27d71d Author: Ulrich Kunitz Date: Sun Dec 10 18:27:01 2006 +0100 [PATCH] ieee80211softmac: Fix mutex_lock at exit of ieee80211_softmac_get_genie ieee80211softmac_wx_get_genie locks the associnfo mutex at function exit. This patch fixes it. The patch is against Linus' tree (commit af1713e0). Signed-off-by: Ulrich Kunitz Signed-off-by: Michael Buesch Acked-by: Johannes Berg Signed-off-by: Larry Finger Signed-off-by: Andrew Morton Signed-off-by: John W. Linville commit 6f07a8a3eb2adbfacd1322862368da4eb9dd3b45 Author: Ulrich Kunitz Date: Sun Dec 10 18:39:08 2006 +0100 [PATCH] ieee80211softmac: Fix errors related to the work_struct changes The signature of work functions changed recently from a context pointer to the work structure pointer. This caused a problem in the ieee80211softmac code, because the ieee80211softmac_assox_work function has been called directly with a parameter explicitly casted to (void*). This compiled correctly but resulted in a softlock, because mutex_lock was called with the wrong memory address. The patch fixes the problem. Another issue was a wrong call of the schedule_work function. Softmac works again and this fixes the problem I mentioned earlier in the zd1211rw rx tasklet patch. The patch is against Linus' tree (commit af1713e0). Signed-off-by: Ulrich Kunitz Acked-by: Michael Buesch Signed-off-by: Larry Finger Signed-off-by: Andrew Morton Signed-off-by: John W. Linville commit 4d1feabcbf41f875447a392015acd0796f57baf6 Author: Ulrich Kunitz Date: Sun Dec 10 11:13:12 2006 -0800 [PATCH] zd1211rw: Call ieee80211_rx in tasklet The driver called ieee80211_rx in hardware interrupt context. This has been against the intention of the ieee80211_rx function. It caused a bug in the crypto routines used by WPA. This patch calls ieee80211_rx in a tasklet. Signed-off-by: Ulrich Kunitz Signed-off-by: Andrew Morton Signed-off-by: John W. Linville