- 14 Sep, 2004 40 commits
-
-
Paolo \'Blaisorblade\' Giarrusso authored
Using CONFIG_UML_SMP and then making CONFIG_SMP = CONFIG_UML_SMP is useless (there was a reason in 2.4, to have different help texts, but not now). Signed-off-by:
Paolo 'Blaisorblade' Giarrusso <blaisorblade_spam@yahoo.it> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
Paolo \'Blaisorblade\' Giarrusso authored
From: Sonny Rao <sonny@burdell.org> Make the SMP code compile, at least, to make testing possible, and remove its dependency on CONFIG_BROKEN. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
Drop a config option which has disappeared from all archs. Btw, this shouldn't be in the UML-specific part, but since we cannot include generic Kconfigs to avoid problem with hardware-related configs, it's duplicated for now. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
Correct one Kconfig dependency, which should refer to CONFIG_USERMODE rather than to CONFIG_UM. We should also figure out how to make the config process work better for UML. We would like to make UML able to "source drivers/Kconfig" and have the right drivers selectable (i.e. LVM, ramdisk, and so on) and the ones for actual hardware excluded. I've been reading such a request even from Jeff Dike at the last Kernel Summit, (in the lwn.net coverage) but without any followup. Signed-off-by:
-
Jeff Dike authored
On reboot, all signals and signal sources are disabled so that late-arriving signals don't show up after the reboot exec, confusing the new image, which is not expecting signals yet. Signed-off-by:
-
Jeff Dike authored
This patch calculates section boundaries differently so as to not get tripped up by holes in the binary such as are introduced by exec-shield. Signed-off-by:
-
Jeff Dike authored
This fixes a use-after-free bug in the context switching. A process going out of context after exiting wakes up the next process and then kills itself. The problem is that when it gets around to killing itself is up to the host and can happen a long time later, including after the incoming process has freed its stack, and that memory is possibly being used for something else. The fix is to have the incoming process kill the exiting process just to make sure it can't be running at the point that its stack is freed. Signed-off-by:
-
Jeff Dike authored
This patch eliminates use of task.thread.kernel_stack. It was unnecessary, confusing, and was masking some kernel stack size assumptions. Signed-off-by:
-
Jeff Dike authored
Remove dependency on ghash.h. Remvoe ghash.h, too. Signed-off-by:
-
Ryan S. Arnold authored
Following the same advice you gave in a recent hvc_console patch I have modified HVCS to remove a while() { yield(); } from hvcs_close() which may cause problems where real time scheduling is concerned and replaced it with tty_wait_until_sent() which uses a real wait queue and is the proper method for blocking a tty operation while waiting for data to be sent. This patch has been tested to verify that all the paths of code that were changed were hit during the code run and performed as expected including hotplug remove of hvcs adapters and hangup of ttys. - Replaced yield() in hvcs_close() with tty_wait_until_sent() to prevent possible lockup with realtime scheduling. - Removed hvcs_final_close() and reordered cleanup operations to prevent discarding of pending data during an hvcs_close() call. - Removed spinlock protection of hvcs_struct data members in hvcs_write_room() and hvcs_chars_in_buffer() because they aren't needed. Signed-off-by:Ryan S. Arnold <rsa@us.ibm.com> Signed-off-by:
-
Ingo Molnar authored
Introduces two new /sys/block values: /sys/block/*/queue/max_hw_sectors_kb /sys/block/*/queue/max_sectors_kb max_hw_sectors_kb is the maximum that the driver can handle and is readonly. max_sectors_kb is the current max_sectors value and can be tuned by root. PAGE_SIZE granularity is enforced. It's all locking-safe and all affected layered drivers have been updated as well. The patch has been in testing for a couple of weeks already as part of the voluntary-preempt patches and it works just fine - people use it to reduce IDE IRQ handling latencies. Signed-off-by:
Ingo Molnar <mingo@elte.hu> Signed-off-by:
-
Andi Kleen authored
This patch adds a prctl to modify current->comm as shown in /proc. This feature was requested by KDE developers. In KDE most programs are started by forking from a kdeinit program that already has the libraries loaded and some other state. Problem is to give these forked programs the proper name. It already writes the command line in the environment (as seen in ps), but top uses a different field in /proc/pid/status that reports current->comm. And that was always "kdeinit" instead of the real command name. So you ended up with lots of kdeinits in your top listing, which was not very useful. This patch adds a new prctl PR_SET_NAME to allow a program to change its comm field. I considered the potential security issues of a program obscuring itself with this interface, but I don't think it matters much because a program can already obscure itself when the admin uses ps instead of top. In case of a KDE desktop calling everything kdeinit is much more obfuscation than the alternative. Signed-off-by:
-
Christoph Hellwig authored
Tons of exports in the new audit code, but not a single module that actually uses one of them. Signed-off-by:
-
Christoph Hellwig authored
Only used in dq_list_lock these days. Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
Christoph Hellwig authored
Signed-off-by:
-
Jesper Juhl authored
akpm: really, reads are supposed to return the number-of-bytes-read on faults, or -EFAULT of no bytes were read. This patch returns either zero or -EFAULT, ignoring any successfully transferred data. But the user interface (whcih is an ioctl()) was never set up to do that. Signed-off-by:
Jesper Juhl <juhl-lkml@dif.dk> Signed-off-by:
-
Anton Blanchard authored
We are now allocating twice as much memory as required. Signed-off-by:
Anton Blanchard <anton@samba.org> Signed-off-by:
-
Hugh Dickins authored
I _think_ shmem_file_setup is protected against negative loff_t size by the TASK_SIZE in each arch, but prefer the security of an explicit test. Wipe those parentheses off its return(file), and update our Copyright. Signed-off-by:
Hugh Dickins <hugh@veritas.com> Signed-off-by:
-
Hugh Dickins authored
Very minor adjustments to shmem_getpage return path: I now prefer it to return NULL and let do_shmem_file_read use ZERO_PAGE(0) in that case; and we don't need a local majmin variable, do_no_page initializes *type to VM_FAULT_MINOR already. Signed-off-by:
-
Hugh Dickins authored
If we're thinking about shmem scalability... isn't it silly that each shmem object is added to the shmem_inodes list on creation, and removed on deletion, yet the only use for that list is in swapoff (shmem_unuse)? Call it shmem_swaplist; shmem_writepage add inode to swaplist when first swap allocated (usually never); shmem_delete_inode remove inode from the list after truncating (if called before, inode could be re-added to it). Inode can remain on the swaplist after all its pages are swapped back in, just be lazy about it; but if shmem_unuse finds swapped count now 0, save itself time by then removing that inode from the swaplist. Signed-off-by:
-
Hugh Dickins authored
Some might want a tmpfs mount with the improved scalability afforded by omitting shmem superblock accounting; or some might just want to test it in an externally-visible tmpfs mount instance. Adopt the convention that mount option -o nr_blocks=0,nr_inodes=0 means without resource limits, and hence no shmem_sb_info. Not recommended for general use, but no worse than ramfs. Disallow remounting from unlimited to limited (no accounting has been done so far, so no idea whether it's permissible), and from limited to unlimited (because we'd need then to free the sbinfo, and visit each inode to reset its i_blocks to 0: why bother?). Signed-off-by:
-
Hugh Dickins authored
SGI investigations have shown a dramatic contrast in scalability between anonymous memory and shmem objects. Processes building distinct shmem objects in parallel hit heavy contention on shmem superblock stat_lock. Across 256 cpus an intensive test runs 300 times slower than anonymous. Jack Steiner has observed that all the shmem superblock free_blocks and free_inodes accounting is redundant in the case of the internal mount used for SysV shared memory and for shared writable /dev/zero objects (the cases which most concern them): it specifically declines to limit. Based upon Brent Casavant's SHMEM_NOSBINFO patch, this instead just removes the shmem_sb_info structure from the internal kernel mount, testing where necessary for null sbinfo pointer. shmem_set_size moved within CONFIG_TMPFS, its arg named "sbinfo" as elsewhere. This brings shmem object scalability up to that of anonymous memory, in the case where distinct processes are building (faulting to allocate) distinct objects. It significantly improves parallel building of a shared shmem object (that test runs 14 times faster across 256 cpus), but other issues remain in that case: to be addressed in later patches. Signed-off-by:
-
Hugh Dickins authored
Keith Mannthey's Bugzilla #3268 drew attention to how tmpfs inodes and dentries and long names and radix-tree nodes pin lowmem. Assuming about 1k of lowmem per inode, we need to lower the default nr_inodes limit on machines with significant highmem. Be conservative, but more generous than in the original patch to Keith: limit to number of lowmem pages, which works out around 200,000 on i386. Easily overridden by giving the nr_inodes= mount option: those who want to sail closer to the rocks should be allowed to do so. Notice how tmpfs dentries cannot be reclaimed in the way that disk-based dentries can: so even hard links need to be costed. They are cheaper than inodes, but easier all round to charge the same. This way, the limit for hard links is equally visible through "df -i": but expect occasional bugreports that tmpfs links are being treated like this. Would have been simpler just to move the free_inodes accounting from shmem_delete_inode to shmem_unlink; but that would lose the charge on unlinked but open files. Signed-off-by:
-
Hugh Dickins authored
Anton recently removed SLAB_HWCACHE_ALIGN from the fs inode caches, now do the same for tmpfs inode cache: fits 9 per page where 7 before. Was saying SLAB_RECLAIM_ACCOUNT too, but that's wrong: tmpfs inodes are not reclaimed under pressure; and hugetlbfs had copied that too. Rearrange shmem_inode_info fields so those most likely to be needed are most likely to be in the same cacheline as the spinlock guarding them. Signed-off-by:
-
Matthieu Castet authored
this patch fix a pnpbios problem with independant resource(http://bugzilla.kernel.org/show_bug.cgi?id=3295) : the old code assume that they are given at the beggining (before any SMALL_TAG_STARTDEP entry), but in some case there are found after SMALL_TAG_ENDDEP entry. tag : 6 SMALL_TAG_STARTDEP tag : 8 SMALL_TAG_PORT tag : 6 SMALL_TAG_STARTDEP tag : 8 SMALL_TAG_PORT tag : 7 SMALL_TAG_ENDDEP tag : 4 SMALL_TAG_IRQ <-- independant resource tag : f SMALL_TAG_END Signed-off-by:
-
Yuval Turgeman authored
I added the ability to search for parameters in make menuconfig (find a given parameter's location in the tree). You use '/' to invoke the feature. Regular expression searches are supported. Signed-off-by:
Yuval Turgeman <yuvalt@gmail.com> Signed-off-by:
-
Benjamin Herrenschmidt authored
This patch adds support for the 750CX based ibook2 600Mhz to the cpufreq powermac driver. Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Anton Blanchard authored
Remove -Wno-uninitialized on ppc32 too. Ive just found a number of real bugs on ppc64 by doing the same. Signed-off-by:
-
David Gibson authored
This patch has been tested both on SLB and segment table machines. This new approach is far from the final word in VSID/context allocation, but it's a noticeable improvement on the old method. Replace the VSID allocation algorithm. The new algorithm first generates a 36-bit "proto-VSID" (with 0xfffffffff reserved). For kernel addresses this is equal to the ESID (address >> 28), for user addresses it is: (context << 15) | (esid & 0x7fff) These are distinguishable from kernel proto-VSIDs because the top bit is clear. Proto-VSIDs with the top two bits equal to 0b10 are reserved for now. The proto-VSIDs are then scrambled into real VSIDs with the multiplicative hash: VSID = (proto-VSID * VSID_MULTIPLIER) % VSID_MODULUS where VSID_MULTIPLIER = 268435399 = 0xFFFFFC7 VSID_MODULUS = 2^36-1 = 0xFFFFFFFFF This scramble is 1:1, because VSID_MULTIPLIER and VSID_MODULUS are co-prime since VSID_MULTIPLIER is prime (the largest 28-bit prime, in fact). This scheme has a number of advantages over the old one: - We now have VSIDs for every kernel address (i.e. everything above 0xC000000000000000), except the very top segment. That simplifies a number of things. - We allow for 15 significant bits of ESID for user addresses with 20 bits of context. i.e. 8T (43 bits) of address space for up to 1M contexts, significantly more than the old method (although we will need changes in the hash path and context allocation to take advantage of this). - Because we use a real multiplicative hash function, we have better and more robust hash scattering with this VSID algorithm (at least based on some initial results). Because the MODULUS is 2^n-1 we can use a trick to compute it efficiently without a divide or extra multiply. This makes the new algorithm barely slower than the old one. Signed-off-by:
-
Anton Blanchard authored
Signed-off-by:
-
Anton Blanchard authored
Here are fixes for some false positives. Signed-off-by:
-
Anton Blanchard authored
The removal of -Wno-uninitialized on ppc64 revealed a number of real bugs. Signed-off-by:
-
Anton Blanchard authored
Signed-off-by:
-
Anton Blanchard authored
Clean up our idle loop code: - Remove a bunch of useless includes and make most functions static - There were places where we werent disabling interrupts before checking need_resched then calling the hypervisor to sleep our thread. We might race with an IPI and end up missing a reschedule. Disable interrupts around these regions to make them safe. - We forgot to turn off the polling flag when exiting the dedicated_idle idle loop. This could have resulted in all manner problems as other cpus would avoid sending IPIs to force reschedules. - Add a missing check for cpu_is_offline in the shared cpu idle loop. Signed-off-by:
-
Anton Blanchard authored
Now that we understand (and have fixed) the problem with using low power mode in the idle loop, lets enable it. It should save a fair amount of power. (The problem was that our exceptions were inheriting the low power mode and so were executing at a fraction of the normal cpu issue rate. We fixed it by always bumping our priority to medium at the start of every exception). Signed-off-by:
-
Anton Blanchard authored
Restore the smt-enabled=off kernel command line functionality: - Remove the SMT_DYNAMIC state now that smt_snooze_delay allows for the same thing. - Remove the early prom.c parsing for the option, put it into an early_param instead. - In setup_cpu_maps honour the smt-enabled setting Note to Nathan: In order to allow cpu hotplug add of secondary threads after booting with smt-enabled=off, I had to initialise cpu_present_map to cpu_online_map in smp_cpus_done. Im not sure how you want to handle this but it seems our present map currently does not allow cpus to be added into the partition that werent there at boot (but were in the possible map). Signed-off-by:
-
Anton Blanchard authored
Make use of Rusty's early_param code. Its good stuff. We appear to be the first user :) Move vpa_init and idle_setup later in boot, we dont have to do them right up front in setup_system. Signed-off-by:
-
Anton Blanchard authored
We have had reports of people attempting to disable EEH on POWER5 boxes. This is not supported, and the device will most likely not respond to config space reads/writes. Remove the IBM location matching code that was being used to disable devices as well as the global option. We already have the ability to ignore EEH erros via the panic_on_oops sysctl option, advanced users should make use of that instead. Signed-off-by:
-
Anton Blanchard authored
- Remove ppc64_calibrate_delay, no longer used - Formatting fixups Signed-off-by:
-
