- 24 Aug, 2004 40 commits
-
-
Paolo \'Blaisorblade\' Giarrusso authored
Signed-off-by:
Paolo 'Blaisorblade' Giarrusso <blaisorblade_spam@yahoo.it> Cc: Jeff Dike <jdike@addtoit.com> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
Paolo \'Blaisorblade\' Giarrusso authored
From: Oleg Drokin <green@linuxhacker.ru>, Jeff Dike <jdike@addtoit.com>, and me If size > 128K, with this patch malloc will call vmalloc; free will detect whether to call vfree or kfree or __real_free(). The 2.4 version could forget free()ing something; this has been fixed. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
That code comes from the out_of_memory section; in 2.4 it was correct to put it for "default:", since it was called when handle_mm_fault() return value was != 0, 1, 2, i.e. it was 3, OOM (but the i386 code put it out of line, for better performance). Here, instead, the OOM case is handled on its own, so if handle_mm_fault() != from the listed cases we must BUG(). Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
From: Alex Züpke <azu@sysgo.de>, and me SKAS mode is like 4G/4G (here we have actually 3G/3G) for guest processes, so when checking for kernel stack overflow, we must first make sure we are checking a kernel-space address. Also, correctly test for stack overflows (i.e. check if there is less than 1k of stack left; see arch/i386/kernel/irq.c:do_IRQ()). And also, THREAD_SIZE != PAGE_SIZE * 2, in general (though this setting is almost never changed, so we didn't notice this1). Thanks to the good eye of Alex Züpke <azu@sysgo.de> for first seeing this bug, and providing a test program: /* * trigger.c - triggers panic("Kernel stack overflow") in UML * * 20040630, azu@sysgo.de */ #include <stdio.h> #include <setjmp.h> #include <fcntl.h> #include <unistd.h> #include <sys/types.h> #include <sys/stat.h> #include <sys/mman.h> #define LOW 0xa0000000 #define HIGH 0xb0000000 int main(int argc, char **argv) { unsigned long addr; int fd; fd = open("/dev/zero", O_RDWR); printf("This may take some time ... one more cup of coffee ...\n"); for(addr = LOW; addr < HIGH; addr += 0x1000) { pid_t p; if(mmap((void*)addr, 0x1000, PROT_READ, MAP_SHARED | MAP_FIXED, fd, 0) == MAP_FAILED) printf("mmap failed\n"); p = fork(); if(p == -1) printf("fork failed\n"); if(p == 0) { /* child context */ int *p = (int *)addr; volatile int x; x = *p; return 0; } /* father context */ waitpid(p, 0, 0); if(munmap((void*)addr, 0x1000) == -1) printf("munmap failed\n"); } close(fd); printf("done\n"); } Signed-off-by: -
Paolo \'Blaisorblade\' Giarrusso authored
Adds some exports Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
On various places (mostly waitpid() calls) this patch makes sure that if errno == EINTR on return, then the syscall is endlessly retried. It also defines a simple generic way to do this. Signed-off-by: <blaisorblade_spam@yahoo.it> Cc: Jeff Dike <jdike@addtoit.com> Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
- Correct some silly errors (dereferencing a pointer before checking if it's != NULL when creating /proc/sysemu, some error messages) - separate using_sysemu from sysemu_supported (so to refuse to activate sysemu if it is not supported, avoiding panics) - not probe sysemu if in tt mode. Signed-off-by: <blaisorblade_spam@yahoo.it> Cc: Jeff Dike <jdike@addtoit.com> Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
Adds /proc/sysemu to toggle SYSEMU usage. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
Adds the "nosysemu" command line parameter to disable SYSEMU Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
Turns off syscall emulation patch for ptrace (SYSEMU) on. SYSEMU is a performance-patch introduced by Laurent Vivier. It changes behaviour of ptrace() and helps reducing host context switch rate. To make it working, you need a kernel patch for your host, too. See http://perso.wanadoo.fr/laurent.vivier/UML/ for further information. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
Folds hostaudio_user.c into hostaudio_kern.c. A lot of code less. Also note that I no more update ppos(as I used to do in the 2.4 patch): I checked that OSS never changes ppos, so hostaudio did the right thing. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
[PATCH] uml: Fixes raw() and uses it in check_one_sigio; also fixes a silly panic (EINTR returned by call). Fixes raw() and uses it in check_one_sigio; also fixes a silly panic (EINTR returned by call). Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
Reduces code in *_user files, by moving it in _kern files if already possible. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
Avoids compile failure when host misses tkill(), by simply using kill() in that case. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
Fixes some little warnings about "Defined but not used ..." by #ifdef'ing things Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
You probably saw that if you change one config option, even if linux/autoconf.h (which is included by everything) changes, the kernel is smart enough not to recompile everything. But with UML this no more holds. Why? Because, as you see in this patch, fixdep avoids making anything depend onto linux/autoconf.h *explicitly*, but nobody taught him to do the same for arch/um/include/uml-config.h. So apply this patch. Do not say "I don't want to change the generic Kbuild for one arch": this cannot hurt. It's a bugfix for us, a no-op for others. Note: with this patch, fixdep will still add a dependency from a file containing UML_CONFIG_BYE onto CONFIG_BYE. Since someone could think that fixdep should grep for [^A-Z_]CONFIG_ rather than simply for CONFIG_, I've added a comment that ask *not to fix* this "bug". Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
Makes "make help ARCH=um" work. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
The second adds the LEGACY_PTY config option. Without it, with late 2.6 kernels /dev/ptyxx won't work. In fact, with those kernels, root_fs_toms does not work, because it's "unable to allocate TTY pair". And removes the dead option "UNIX98_PTY_COUNT" (just commented out for now). Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
In detail, on 2.4 we used force_delete() to make sure inode were not cached, and we then close the host file when the inode is cleared; when porting to 2.6 the "force_delete" thing was dropped, and this patch adds a fix for this (by setting drop_inode = generic_delete_inode). Search for drop_inode in the 2.6 Documentation/filesystems/vfs.txt for info about this. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
Avoid that gcc breaks UML with "unit at a time" compilation mode. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
Just for now and just for UML; it will go away. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
In the -mm tree (in this moment) and not in 2.6.7 there is another console_device in include/linux/console.h; so I renamed the UML one (it's static). Cc: Jeff Dike <jdike@addtoit.com> Signed-off-by:
-
Andrew Morton authored
Update UML for CPU scheduler changes Cc: Jeff Dike <jdike@addtoit.com> Signed-off-by:
-
Jeff Dike authored
The patch below brings UML up to date with some changes in the rest of the kernel: an updated defconfig checksum.h includes in6.h to get a definition of in6_addr added a missing cpu_{set,clear} change removed include/asm-um/module.h since it's really a link Signed-off-by: -
Jeff Dike authored
The code is still there but it's not built. Below is a patch which removes it totally. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
The main part of UML; it is the last distributed patch for 2.6.7 Removes skas support from the main UML patch; apply or get conflicts. Signed-off-by:
-
Anton Blanchard authored
From: <arjanv@redhat.com> Implement the new address space layout for 32-bit apps running on ppc64. Signed-off-by:
-
Arjan van de Ven authored
Below is a patch from Pete Zaitcev (zaitcev@redhat.com) to also use the flex mmap infrastructure for s390(x). The IBM Domino guys *really* seem to want this. Signed-off-by:
-
Arjan van de Ven authored
Create /proc/sys/vm/legacy_va_layout. If this is non-zero, the kernel will use the old mmap layout for all tasks. it presently defaults to zero (the new layout). From: William Lee Irwin III <wli@holomorphy.com> hugetlb CONFIG_SYSCTL=n fix Signed-off-by:
Ingo Molnar <mingo@elte.hu> Signed-off-by:
-
Arjan van de Ven authored
Utz Lehmann <u.lehmann@de.tecosim.com> found a problem with the flexmmap patches on x86-64, what he is seeing is that the 32 bit personality isn't set at the first point of setting the allocator strategy. The solution is simple, in binfmt_elf the personality is set so put the pick-layout function there. Please consider, Signed-off-by:
Arjan van de Ven <arjanv@redhat.com> Signed-off-by:
-
Ingo Molnar authored
Rework the i386 mm layout to allow applications to allocate more virtual memory, and larger contiguous chunks. - the patch is compatible with existing architectures that either make use of HAVE_ARCH_UNMAPPED_AREA or use the default mmap() allocator - there is no change in behavior. - 64-bit architectures can use the same mechanism to clean up 32-bit compatibility layouts: by defining HAVE_ARCH_PICK_MMAP_LAYOUT and providing a arch_pick_mmap_layout() function - which can then decide between various mmap() layout functions. - I also introduced a new personality bit (ADDR_COMPAT_LAYOUT) to signal older binaries that dont have PT_GNU_STACK. x86 uses this to revert back to the stock layout. I also changed x86 to not clear the personality bits upon exec(), like x86-64 already does. - once every architecture that uses HAVE_ARCH_UNMAPPED_AREA has defined its arch_pick_mmap_layout() function, we can get rid of HAVE_ARCH_UNMAPPED_AREA altogether, as a final cleanup. the new layout generation function (__get_unmapped_area()) got significant testing in FC1/2, so i'm pretty confident it's robust. Compiles & boots fine on an 'old' and on a 'new' x86 distro as well. The two known breakages were: http://www.redhatconfig.com/msg/67248.html [ 'cyzload' third-party utility broke. ] http://www.zipworld.com/au/~akpm/dde.tar.gz [ your editor broke :-) ] both were caused by application bugs that did: int ret = malloc(); if (ret <= 0) failure; such bugs are easy to spot if they happen, and if it happens it's possible to work it around immediately without having to change the binary, via the setarch patch. No other application has been found to be affected, and this particular change got pretty wide coverage already over RHEL3 and exec-shield, it's in use for more than a year. The setarch utility can be used to trigger the compatibility layout on x86, the following version has been patched to take the `-L' option: http://people.redhat.com/mingo/flexible-mmap/setarch-1.4-2.tar.gz "setarch -L i386 <command>" will run the command with the old layout. From: Hugh Dickins <hugh@veritas.com> The problem is in the flexible mmap patch: arch_get_unmapped_area_topdown is liable to give your mmap vm_start above TASK_SIZE with vm_end wrapped; which is confusing, and ends up as that BUG_ON(mm->map_count). The patch below stops that behaviour, but it's not the full solution: wilson_mmap_test -s 1000 then simply cannot allocate memory for the large mmap, whereas it works fine non-top-down. I think it's wrong to interpret a large or rlim_infinite stack rlimit as an inviolable request to reserve that much for the stack: it makes much less VM available than bottom up, not what was intended. Perhaps top down should go bottom up (instead of belly up) when it fails - but I'd probably better leave that to Ingo. Or perhaps the default should place stack below text (as WLI suggested and ELF intended, with its text defaulting to 0x08048000, small progs sharing page table between stack and text and data); with a further personality for those needing bigger stack. From: Ingo Molnar <mingo@elte.hu> - fall back to the bottom-up layout if the stack can grow unlimited (if the stack ulimit has been set to RLIM_INFINITY) - try the bottom-up allocator if the top-down allocator fails - this can utilize the hole between the true bottom of the stack and its ulimit, as a last-resort effort. Signed-off-by: -
Ingo Molnar authored
while looking at HT scheduler bugreports and boot failures i discovered a bad assumption in most of the HT scheduling code: that resched_task() can be called without holding the task's runqueue. This is most definitely not valid - doing it without locking can lead to the task on that CPU exiting, and this CPU corrupting the (ex-) task_info struct. It can also lead to HT-wakeup races with task switching on that other CPU. (this_CPU marking the wrong task on that_CPU as need_resched - resulting in e.g. idle wakeups not working.) The attached patch against fixes it all up. Changes: - resched_task() needs to touch the task so the runqueue lock of that CPU must be held: resched_task() now enforces this rule. - wake_priority_sleeper() was called without holding the runqueue lock. - wake_sleeping_dependent() needs to hold the runqueue locks of all siblings (2 typically). Effects of this ripples back to schedule() as well - in the non-SMT case it gets compiled out so it's fine. - dependent_sleeper() needs the runqueue locks too - and it's slightly harder because it wants to know the 'next task' info which might change during the lock-drop/reacquire. Ripple effect on schedule() => compiled out on non-SMT so fine. - resched_task() was disabling preemption for no good reason - all paths that called this function had either a spinlock held or irqs disabled. Compiled & booted on x86 SMP and UP, with and without SMT. Booted the SMT kernel on a real SMP+HT box as well. (Unpatched kernel wouldn't even boot with the resched_task() assert in place.) Signed-off-by:
-
Ingo Molnar authored
disable preemption in the self-reap codepath, as such tasks may not be on the tasklist anymore and CPU-hotplug relies on the tasklist to migrate tasks. Signed-off-by:
-
Ingo Molnar authored
release_task() calls proc_pid_flush() call dput(), which can sleep. But that's a late-in-exit no-preempt path with CONFIG_PREEMPT. Signed-off-by:
-
Ingo Molnar authored
Rusty noticed that we update the parent ->avg_sleep without holding the runqueue lock. Also the code needed cleanups. Signed-off-by:
-
Ingo Molnar authored
* Nick Piggin <nickpiggin@yahoo.com.au> wrote: > Increasing priority (negative nice) doesn't have much impact. -20 CPU > hog only gets about double the CPU of a 0 priority CPU hog and only > about 120% the CPU time of a nice -10 hog. this is a property of the base scheduler as well. We can do a nonlinear timeslice distribution trivially - the attached patch implements the following timeslice distribution ontop of 2.6.8-rc3-mm1: [ -20 ... 0 ... 19 ] => [800ms ... 100ms ... 5ms] the nice-20/nice+19 ratio is now 1:160 - sufficient for all aspects. Signed-off-by:
-
Ingo Molnar authored
- whitespace and style cleanups Signed-off-by:
-
Andrew Morton authored
SMP fix -- for_each_domain() is not defined if not CONFIG_SMP, so show_schedstat needed a couple of extra ifdefs. Signed-off-by:Rick Lindsley <ricklind@us.ibm.com> Signed-off-by:
-
William Lee Irwin III authored
Fix up sparc32 properly. Signed-off-by:
-
William Lee Irwin III authored
It appears that init_idle() and fork_by_hand() could be combined into a single method that calls init_idle() on behalf of the caller. Signed-off-by:
-
