- 10 Mar, 2005 40 commits
-
-
Stephen Rothwell authored
This patch pulls together the compat_sigevent structs. It also consolidates the copying of these structures into the kernel. The only part of the second union in sigevent that the kernel looks at currently is the _tid, so that is the only bit we copy. This patch depends on my previous two patches "add and use COMPAT_SIGEV_PAD_SIZE" and "Consolidate the last compat sigvals". Signed-off-by:
Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
Stephen Rothwell authored
This patch just consolidates the last of the (what should have been) compat_sigval_ts. It also fixes S390 that had a sigval_t in its struct compat_siginfo. Signed-off-by:
-
Stephen Rothwell authored
All the 32 bit architectures (effectively) define SIGEV_PAD_SIZE to be ((SIGEV_MAX_SIZE/sizeof(int)) - 3). So define COMPAT_SIGEV_PAD_SIZE to be this and replace SIGEV_PAD_SIZE32 where it is used. It also needs to be used in the definition of struct compat_sigevent as most of the architectures would have had it 4 bytes too small in the kernel (since we were using SIGEV_PAD_SIZE). Signed-off-by:
-
Stephen C. Tweedie authored
With Andrew Morton invalidate_inode_pages2_range() can livelock if all the pages in the pagevec are outside the end of teh range which we're trying to invalidate. Fix that up in a similar manner to the truncate_inode_pages_range() handling. Signed-off-by:
-
Karsten Keil authored
The Auvertech TurboPAM ISDN hardware was never produce and the company do not longer exist, so it makes no sense to support this any longer. Signed-off-by:
Karsten Keil <kkeil@suse.de> Acked-by:
Stelian Pop <stelian@popies.net> Signed-off-by:
-
Karsten Keil authored
This should resolve all the issues from the review and is a big cleanup for i4l-hfc-4s-and-hfc-8s-driver.patch. Signed-off-by:
-
Karsten Keil authored
Add a new driver for new ISDN hardware. Signed-off-by:
-
Karsten Keil authored
This is an update of the HFC USB ISDN driver, the current one has some bugs and support not all HW. It was already tested as standalone patch for some time, so including it should be safe. Signed-off-by:
-
Bodo Stroesser authored
fix_range_common and flush_tlb_kernel_range_common don't work correctly, if a PGD (or PUD or PMD) is not present and start_addr (resp. start) is not aligned to a PGD boundary (or PUD or PMD boundary). Signed-off-by:
Bodo Stroesser <bstroesser@fujitsu-siemens.com> Signed-off-by:
Jeff Dike <jdike@addtoit.com> Signed-off-by:
-
Jeff Dike authored
This patch merges now-identical page table walking and flushing code that had been duplicated in skas and tt modes. The differences had been the low-level address space updating operations, which are now abstracted away in the respective do_ops functions. Signed-off-by:
-
Jeff Dike authored
This patch optimizes tlb flushing in a couple of ways to reduce the number of system calls made to the host in order to update an address space. Operations are collected, and adjacent ones which can be merged, are. This includes consecutive munmaps, mprotects with the same permissions, and mmaps with the same backing file and permissions and linear in the file. Second, the munmaps that always preceded mmaps are now done instead of mmap if necessary. Signed-off-by:
-
Jeff Dike authored
To quote .config into config.c for building the result into the code, use sed instead of perl, as requested by one "embedded" UML user (which notes that perl is a big requirement, while busybox provides sed which is used in this patch). I've tested that there are only cosmethical differences in the produced config.c file, which don't change at all the result (i.e. "a" is replaced by "" "a" at the beginning, which is non-significant). Reported by, and initial patch provided by, Rob Landley. Signed-off-by:
Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it> Signed-off-by:
-
Jeff Dike authored
From: Esben Nielsen <simlo at phys au dk> One of the problems was use of direct architecture specific semaphores (which doesn't work under PREEMPT_REALTIME) and in places where a quick (maybe too quick) look at the code told me that completions ought to be used. Therefore I changed two semaphores to completions which compiled fine. I have tried the change on 2.6.11-rc2, and it seemed to work, but I have not tested it heavily. Signed-off-by:
-
Jeff Dike authored
The init function called by gcc when gcov is enabled is __gcov_init or __bb_init_func, depending on the gcc version. Anton is using 3.3.4 and seeing __gcov_init. I'm using 3.3.2 and seeing __bb_init_func, so we need to close that gap a bit. Signed-off-by:
-
Jeff Dike authored
This implements a hardware random number generator for UML which attaches itself to the host's /dev/random. Signed-off-by:
-
Jeff Dike authored
From: Bodo Stroesser, mostly: Gets rid of all inclusions of {sys,linux,asm}/ptrace.h since their effects and contents are somewhat distro- and architecture-dependent. arch/um/sysdep/ptrace_user is now responsible for providing the system's ptrace interfaces to UML. As such, it is a purely userspace header, and this exposed a couple of places where kernel files were including it. One of them needed MAX_REG_OFFSET, so this was provided by adding a new generated header, user-constants.h, which includes this. Also, changed PTRACE_{PEEK,POKE}_USER to _USR since that seems to be more standard. Signed-off-by: -
Jeff Dike authored
Fix a typo in the hostfs setgid code. Signed-off-by:
-
Domen Puncer authored
Unify the spinlock initialization as far as possible. Signed-off-by:
Amit Gud <gud@eth.net> Signed-off-by:
Domen Puncer <domen@coderock.org> Signed-off-by:
-
Jeff Dike authored
Noticed by Frank Sorenson - the methods in arch/um/drivers/tty.c should be static. It turns out that all the channels have the same problem, so these are all fixed. These files export only a structure of function pointers, so that structure should be the only externally visible symbol. Signed-off-by:
-
Jeff Dike authored
This patch adds some errors and warnings where there were none before: If someone typos "ubd" as "udb", that is caught and a warning is printed If a ubd file can't be opened, that now results in an error message If there are more telnet connections to port consoles than there are consoles, then a message will appear in the telnet session explaining why there is no login prompt. Signed-off-by: -
Jeff Dike authored
Remove a UML/x86_64 warning (and build failure if CONFIG_MODE_TT is disabled). Signed-off-by:
-
Jeff Dike authored
Spotted by Al Viro, there was some bogosity in the UML/x86_64 modify_ltd. Signed-off-by:
-
Jeff Dike authored
um_mount did nothing but turn around and call sys_mount with the same arguments. This makes it useless code, and it has been duly removed. Signed-off-by:
-
Jeff Dike authored
Prompted by Dave Mielke: This adds a comment explaining why we do not define _XOPEN_SOURCE, even though the pread man page says you should. Signed-off-by:
-
Jeff Dike authored
This patch makes some debug code in the system call path configurable. Signed-off-by:
-
Jeff Dike authored
This patch removes some useless code from the system call path, and removes some debug code, which will reappear in a configurable form in the syscall-debug patch. Signed-off-by:
-
Bodo Stroesser authored
Shifting register restore and userspace resume to the beginning of the userspace loop, we can remove the identical code done before the loop. Signed-off-by:
-
Jeff Dike authored
The network driver wasn't checking that the host side of an interface had been successfully opened before trying to close it at shuwtdown. Signed-off-by:
-
Jeff Dike authored
ubd_get_config wasn't using the standard device number parser, which caused it not to recognize letters. Signed-off-by:
-
Jeff Dike authored
This fixes a couple of copy-user problems spotted by Al Viro. copy_sc_from_user_tt was doing a copy_from_user to do an in-kernel assignment. I commented this, at the request of Chris Wedgewood. sys_ipc had a void *__user ptr which should have been void __user *ptr. Finally, there were a couple of bogus __user annotations on unsigned longs, which were never going to be passed into copy_user. Signed-off-by:
-
Jeff Dike authored
Get rid of the grepping for __st_ino in hostfs, since it doesn't work on x86_64 (the grep finds it, but it is ifdefed out), and Al says it's unnecessary anyway. Signed-off-by:
-
tzachar@cs.bgu.ac.il authored
Tell the netdevice code that a slirp device is not ethernet. Signed-off-by:
-
Jeff Dike authored
Signed-off-by:
-
Jeff Dike authored
This patch makes some small changes that parallel changes in 2.6.11: The csum buffers are now unsigned char. Got rid of an unused define from pgtable-2level.h. Signed-off-by:
-
Coywolf Qi Hunt authored
This patch removes the redundant compiler barrier. As Linus ever said "The mb() should make sure that gcc cannot move things around...". Signed-off-by:
Coywolf Qi Hunt <coywolf@gmail.com> Signed-off-by:
-
Michael S. Tsirkin authored
The AMD-8131 I/O APIC (device id 1022:7450/7451) does not support message signalled interrupts. Thus, if a device driver attempts to enable msi, it will suceed, but interrupts are not actually delivered to the cpu. The Nforce chipsets do not seem to have this limitation. AMD confirmed that MSI mode is unsupported with this APIC. The following patch adds a flag to pci quirks to detect this and disable msi. Signed-off-by:
Michael S. Tsirkin <mst@mellano.co.il> Signed-off-by:
-
Dave Hansen authored
Using the assumption that all addresses in the SRAT are ascending, the calculations can get a bit simpler, and remove the "been_here_before" variable. This also breaks that calculation out into its own function, which further simplifies the look of the code. Signed-off-by:
Dave Hansen <haveblue@us.ibm.com> Signed-off-by:
-
Dave Hansen authored
This patch is to allow the booting of a numa srat base i386 system without requiring memory to be in all of it's nodes. It breaks the assumption that all nodes have memory during bootup. Signed-off-by:
Keith Mannthey <kmannth@us.ibm.com> Signed-off-by:
-
Dave Hansen authored
memory_present() is how each arch/subarch will tell sparsemem and discontigmem where all of its memory is. This is what triggers sparse to go out and create its mappings for the memory, as well as allocate the mem_map[]. By: Andy Whitcroft <apw@shadowen.org> Signed-off-by:
-
Dave Hansen authored
Both the pgdats and the struct zonelist are zeroed unnecessarily. The zonelist is a member of the pgdat, so any time the pgdat is cleared, so is the zonelist. All of the architectures present a zeroed pgdat to the generic code, so it's not necessary to set it again. Not clearing it like this allows the functions to be reused by the memory hotplug code. The only architecture which has a dependence on these clears is i386. The previous patch in this series fixed that up. Signed-off-by:
-
