ChangeSet@1.1532.1.20, 2004-01-31 13:33:13-08:00, torvalds@home.osdl.org Make EHCI have a 20ms power-on to power-good timeout. This should fix some picky USB devices. ChangeSet@1.1532.1.19, 2004-01-31 12:22:51-08:00, akpm@osdl.org [PATCH] ppc32: MBX MAC address fix From Tom Rini On the MBX the kernel made an assumption about where the MAC address would be in the VPD (Vital Product Data). However, the documentation for the firmware (EPPC-Bug) describes the format of the VPD and makes no guarrantees about the location of any VPD record. It does however describe the format of each record type and the MAC address record will be of type '08' and of size 6. This changes the code so that instead of assuming a position, it searches for the record, which I assume is also what EPPC-Bug does. ChangeSet@1.1532.1.18, 2004-01-31 11:46:20-08:00, torvalds@home.osdl.org Upgrade x86 defconfig to something less ancient. ChangeSet@1.1532.1.17, 2004-01-31 10:06:29-08:00, akpm@osdl.org [PATCH] RAID-6: x86-64 crash workaround From: "H. Peter Anvin" Apparently, on x86-64, the stack isn't always aligned properly (16 bytes) in the kernel at the moment. This causes the RAID-6 code to crash the system. This patch is a workaround for that; the right thing is to muck with the assembly entrypoints to enforce proper stack alignment. However, that's not anything I feel comfortable doing in an evening, especially since I don't have a machine on which I can test the resulting kernels. ChangeSet@1.1532.1.16, 2004-01-31 10:05:21-08:00, akpm@osdl.org [PATCH] s390: sclp bug fixes. From: Martin Schwidefsky SCLP console/tty fixes: - Fix incorrect state change of SCLP_RUNNING flag in interrupt handler - Suppress emission of empty buffers to prevent stack overflow - Fix off by one error in sclp_write (used to return # of chars written + 1) - Prevent sclp_tty_write_string from waiting in interrupt (during flush) - Fix deadlock after TIOCSCLPSNL ioctl - Fix sclp_tty_chars_in_buffer calculation ChangeSet@1.1532.1.15, 2004-01-31 10:04:17-08:00, akpm@osdl.org [PATCH] s390: inline assembly constraints. From: Martin Schwidefsky s390 inline assembly bug-fixes: - Add memory barriers to spinlocks, atomic variable functions and to test_and_{set,clear,change}_bit. - Add "=m" and "m" contraints to tell gcc that the content of a variable is in fact used. - Replace "+m" constraints by "=m" on the output and "m" on the input list. - Use c-implemtation for ffz and __ffs. - Use generic c-implemtation for ffs and fls. ChangeSet@1.1532.1.14, 2004-01-31 10:03:17-08:00, akpm@osdl.org [PATCH] s390: general update. From: Martin Schwidefsky s390 general update: - Add archhelp for "make image". - 32-bit execve fix. - Export some symbols: sys_ioctl, put_files_struct & ccw_device_work. - Minor cleanups in arch/s390/mm/init.c & drivers/s390/net/ctctty.c. ChangeSet@1.1532.1.13, 2004-01-31 10:02:26-08:00, akpm@osdl.org [PATCH] Trivial cleanups to hugepage support From: David Gibson These are some trivial cleanups to the hugepage ppc64 support ChangeSet@1.1532.1.12, 2004-01-31 10:01:33-08:00, akpm@osdl.org [PATCH] ppc64: use CONFIG_MAGIC_SYSRQ around xmon sysrq code From: Anton Blanchard ChangeSet@1.1532.1.11, 2004-01-31 10:00:40-08:00, akpm@osdl.org [PATCH] ppc64: SLB rewrite From: Anton Blanchard The current SLB handling code has a number of problems: - We loop trying to find an empty SLB entry before deciding to cast one out. On large working sets this really hurts since the SLB is always full and we end up looping through all 64 entries unnecessarily. - During castout we currently invalidate the entry we are replacing. This is to avoid a nasty race where the entry is in the ERAT but not the SLB and another cpu does a tlbie that removes the ERAT at a critical point. If this race is fixed the SLB can be removed. - The SLB prefault code doesnt work properly The following patch addresses all the above concerns and adds some more optimisations: - feature nop out some segment table only code - slb invalidate the kernel segment on context switch (avoids us having to slb invalidate at each cast out) - optimise flush on context switch, the lazy tlb stuff avoids it being called when going from userspace to kernel thread, but it gets called when going to kernel thread to userspace. In many cases we are returning to the same userspace task, we now check for this and avoid the flush - use the optimised POWER4 mtcrf where possible ChangeSet@1.1532.1.10, 2004-01-31 09:59:52-08:00, akpm@osdl.org [PATCH] ppc64: fix && vs & bugs in lparcfg, from Julie DeWandel From: Anton Blanchard ChangeSet@1.1532.1.9, 2004-01-31 09:59:07-08:00, akpm@osdl.org [PATCH] ppc64: export memchr and csum_partial From: Anton Blanchard ChangeSet@1.1532.1.8, 2004-01-31 09:58:34-08:00, akpm@osdl.org [PATCH] ppc64: add/remove config.h From: Anton Blanchard ChangeSet@1.1532.1.7, 2004-01-31 09:57:58-08:00, akpm@osdl.org [PATCH] ppc64: Use preferred_console to select a reasonable default console From: Anton Blanchard Lets make a decent attempt to find out where your console is. The new preferred_console stuff is pretty nice. ChangeSet@1.1532.1.6, 2004-01-31 09:57:29-08:00, akpm@osdl.org [PATCH] ppc64: defconfig update From: Anton Blanchard ChangeSet@1.1532.1.5, 2004-01-31 09:57:06-08:00, akpm@osdl.org [PATCH] ppc64: missing set_fs(KERNEL_DS) in ppc32_timer_create, from Marcus Meissner From: Anton Blanchard Small obvious fix to ppc32_timer_create. Since sys_timer_create access structures we pass on the stack, we need set_fs(KERNEL_DS). ChangeSet@1.1532.1.4, 2004-01-31 09:56:46-08:00, akpm@osdl.org [PATCH] ppc64: fixes for compile with CONFIG_PROC_DEVICETREE=n, from Nathan Lynch From: Anton Blanchard - nop out proc_device_tree_add_node if CONFIG_PROC_DEVICETREE=n - stubs for procfs-related functions when CONFIG_PROC_DEVICETREE=n ChangeSet@1.1532.1.3, 2004-01-31 09:56:28-08:00, akpm@osdl.org [PATCH] ppc64: lparcfg_write From: Anton Blanchard Add lparcfg_write() for changing SPLPAR system parameters ChangeSet@1.1532.1.2, 2004-01-31 09:56:15-08:00, akpm@osdl.org [PATCH] ppc64: add missing include guards, from Nathan Lynch From: Anton Blanchard add missing include guards, from Nathan Lynch ChangeSet@1.1532.1.1, 2004-01-31 09:56:03-08:00, akpm@osdl.org [PATCH] ppc32: watchdog definition fixes From: Tom Rini Correct the defines for MC146818 RTCs, reg c is interrupts, reg d is not watchdog (it's valid time & nvram). ChangeSet@1.1533, 2004-01-31 01:11:03-05:00, mlindner@syskonnect.de [PATCH] sk98lin: Reset Xmac when stopping the port * Clear the Xmac fifo before stopping the port ChangeSet@1.1532, 2004-01-30 17:30:07-08:00, torvalds@home.osdl.org Linux 2.6.2-rc3 TAG: v2.6.2-rc3