- 23 Mar, 2006 40 commits
-
-
Eric Dumazet authored
1) Reduce the size of (struct fdtable) to exactly 64 bytes on 32bits platforms, lowering kmalloc() allocated space by 50%. 2) Reduce the size of (files_struct), using a special 32 bits (or 64bits) embedded_fd_set, instead of a 1024 bits fd_set for the close_on_exec_init and open_fds_init fields. This save some ram (248 bytes per task) as most tasks dont open more than 32 files. D-Cache footprint for such tasks is also reduced to the minimum. 3) Reduce size of allocated fdset. Currently two full pages are allocated, that is 32768 bits on x86 for example, and way too much. The minimum is now L1_CACHE_BYTES. UP and SMP should benefit from this patch, because most tasks will touch only one cache line when open()/close() stdin/stdout/stderr (0/1/2), (next_fd, close_on_exec_init, open_fds_init, fd_array[0 .. 2] being in the same cache line) Signed-off-by:
Eric Dumazet <dada1@cosmosbay.com> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
Andrew Morton authored
Linus points out that ext3_readdir's readahead only cuts in when ext3_readdir() is operating at the very start of the directory. So for large directories we end up performing no readahead at all and we suck. So take it all out and use the core VM's page_cache_readahead(). This means that ext3 directory reads will use all of readahead's dynamic sizing goop. Note that we're using the directory's filp->f_ra to hold the readahead state, but readahead is actually being performed against the underlying blockdev's address_space. Fortunately the readahead code is all set up to handle this. Tested with printk. It works. I was struggling to find a real workload which actually cared. (The patch also exports page_cache_readahead() to GPL modules) Cc: "Stephen C. Tweedie" <sct@redhat.com> Signed-off-by:
-
Jean Delvare authored
Use the standard BCD macros instead of redefining them. Signed-off-by:
Jean Delvare <khali@linux-fr.org> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Roman Zippel <zippel@linux-m68k.org> Signed-off-by:
-
Luca Tettamanti authored
Add the SNAPSHOT_S2RAM ioctl to the snapshot device. This ioctl allows a userland application to make the system (previously frozen with the SNAPSHOT_FREE ioctl) enter the S3 state without freezing processes and disabling nonboot CPUs for the second time. This will allow us to implement the suspend-to-disk-and-RAM (STDR) functionality in the userland suspend tools. Signed-off-by:
Luca Tettamanti <kronos.it@gmail.com> Signed-off-by:
Rafael J. Wysocki <rjw@sisk.pl> Cc: Pavel Machek <pavel@ucw.cz> Signed-off-by:
-
Rafael J. Wysocki authored
Remove the console-switching code from the suspend part of the swsusp userland interface and let the userland tools switch the console. Signed-off-by:
Pavel Machek <pavel@suse.cz> Signed-off-by:
-
Rafael J. Wysocki authored
It is unsafe to suspend devices if the hardware is controlled by X. Add an extra check to prevent this from happening. Signed-off-by:
-
Shaohua Li authored
Highmem could be in pcp list as well. Signed-off-by: Shaohua Li<shaohua.li@intel.com> Acked-by:
Pavel Machek <pavel@ucw.cz> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Signed-off-by:
-
Rafael J. Wysocki authored
This patch from Pavel moves userland freeze signals handling into more logical place. It now hits even with mysqld running. Signed-off-by:
-
Pavel Machek authored
Combination of printk/pr_debug led to <7> in the middle of the line, and we printed way too many dots. Signed-off-by:
-
Rafael J. Wysocki authored
Allow swsusp to freeze processes successfully under heavy load by freezing userspace processes before kernel threads. [Thanks to Nigel Cunningham <nigel@suspend2.net> for suggesting the way to go.] Signed-off-by:
-
Rafael J. Wysocki authored
This patch introduces a user space interface for swsusp. The interface is based on a special character device, called the snapshot device, that allows user space processes to perform suspend and resume-related operations with the help of some ioctls and the read()/write() functions. Additionally it allows these processes to allocate free swap pages from a selected swap partition, called the resume partition, so that they know which sectors of the resume partition are available to them. The interface uses the same low-level system memory snapshot-handling functions that are used by the built-it swap-writing/reading code of swsusp. The interface documentation is included in the patch. The patch assumes that the major and minor numbers of the snapshot device will be 10 (ie. misc device) and 231, the registration of which has already been requested. Signed-off-by:
-
Pavel Machek authored
Update suspend-to-RAM documentation with new machines, and makes message when processes can't be stopped little clearer. (In one case, waiting longer actually did help). From: "Rafael J. Wysocki" <rjw@sisk.pl> Warn in the documentation that data may be lost if there are some filesystems mounted from USB devices before suspend. [Thanks to Alan Stern for providing the answer to the question in the Q:-A: part.] Signed-off-by:
-
Randy Dunlap authored
Move externs from C source files to header files. Signed-off-by:
Randy Dunlap <rdunlap@xenotime.net> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Pavel Machek <pavel@ucw.cz> Signed-off-by:
-
Rafael J. Wysocki authored
Move the swap-writing/reading code of swsusp to a separate file. Signed-off-by:
-
Rafael J. Wysocki authored
Introduce the low level interface that can be used for handling the snapshot of the system memory by the in-kernel swap-writing/reading code of swsusp and the userland interface code (to be introduced shortly). Also change the way in which swsusp records the allocated swap pages and, consequently, simplifies the in-kernel swap-writing/reading code (this is necessary for the userland interface too). To this end, it introduces two helper functions in mm/swapfile.c, so that the swsusp code does not refer directly to the swap internals. Signed-off-by:
Adrian Bunk <bunk@stusta.de> Signed-off-by:
-
Andrew Morton authored
This was a temporary thing for 2.6.16. Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Pavel Machek <pavel@ucw.cz> Signed-off-by:
-
Ashok Raj authored
Don't create "online" control file for BSP (i386/x86_64) since its not removable. We originally added this to support ppc64 if the kernel has support but BIOS indicated no offline support, we just didnt create online files for them. We used the same method in ia64 as well, if we have a cpu taking platform interrupts but cannot be removed if those interrupts cannot be re-targeted to another cpu. Signed-off-by:
Ashok Raj <ashok.raj@intel.com> Signed-off-by:
-
Markus Gutschke authored
Gcc reserves %ebx when compiling position-independent-code on i386. This means, the _syscallX() macros in include/asm-i386/unistd.h will not compile. This patch is changes the existing macros to take special care to preserve %ebx. The bug can be tracked at http://bugzilla.kernel.org/show_bug.cgi?id=6204Signed-off-by:
Markus Gutschke <markus@google.com> Signed-off-by:
-
Adrian Bunk authored
You must always ensure to fulfill the dependencies of what you are select'ing. Signed-off-by:
-
Chuck Ebbert authored
_raw_spin_lock_flags() is entered with interrupts disabled. If it cannot obtain a spinlock, it checks the flags that were passed and re-enables interrupts before spinning if that's how the flags are set. When the spinlock might be available, it disables interrupts (even if they are already disabled) before trying to get the lock. Change that so interrupts are only disabled if they have been enabled. This costs nine bytes of duplicated spinloop code. Fastpath before patch: jle <keep looping> not-taken conditional jump cli disable interrupts jmp <try for lock> unconditional jump Fastpath after patch, if interrupts were not enabled: jg <try for lock> taken conditional branch Signed-off-by:Chuck Ebbert <76306.1226@compuserve.com> Acked-by:
Ingo Molnar <mingo@elte.hu> Signed-off-by:
-
Shaohua Li authored
Checking APIC version instead of CPU family to determine XAPIC. Family 6 CPU could have xapic as well. Signed-off-by: Shaohua Li<shaohua.li@intel.com> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: "Seth, Rohit" <rohit.seth@intel.com> Cc: Andi Kleen <ak@muc.de> Signed-off-by:
-
Shaohua Li authored
Detecting cache line using cpuid.4, cpuid level 4 is enough. Signed-off-by: Shaohua Li<shaohua.li@intel.com> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: "Seth, Rohit" <rohit.seth@intel.com> Cc: Andi Kleen <ak@muc.de> Signed-off-by:
-
Chuck Ebbert authored
i386 has a small bug in the stack dump code where it prints an extra log level code. Remove that and fix the alignment of normal stack dump printout. Also remove some unnecessary printk() calls. Signed-off-by:
-
Jan Beulich authored
With cpu_gdt_descr having been converted to per-CPU data, the old object (in head.S) no longer needs to reserve space for each CPU's instance. With cpu_gdt_table not being used for CPU 0 anymore, it doesn't seem to need page alignment (or if in fact there is a need for it to retain that alignment, the whole object should go into .data.page_align). Signed-off-by:
Jan Beulich <jbeulich@novell.com> Acked-by:
Zachary Amsden <zach@vmware.com> Signed-off-by:
-
Jesper Juhl authored
arch/i386/kernel/cpu/centaur.c: In function `centaur_mcr_insert': arch/i386/kernel/cpu/centaur.c:33: warning: implicit declaration of function `mtrr_centaur_report_mcr' Signed-off-by:
Jesper Juhl <jesper.juhl@gmail.com> Signed-off-by:
-
Jesper Juhl authored
arch/i386/kernel/apic.c:840: warning: implicit declaration of function `GET_APIC_ID' Signed-off-by:
-
Chuck Ebbert authored
Document a limitation of vsyscall-sysenter, since patches to fix it have been rejected. Signed-off-by:
-
Chuck Ebbert authored
Using PTRACE_SINGLESTEP on a child that does an int80 syscall misses the SIGTRAP that should be delivered upon syscall exit. Fix that by setting TIF_SINGLESTEP when entering the kernel via int80 with TF set. /* Test whether singlestep through an int80 syscall works. */ #define _GNU_SOURCE #include <stdio.h> #include <unistd.h> #include <fcntl.h> #include <sys/ptrace.h> #include <sys/wait.h> #include <sys/mman.h> #include <asm/user.h> static int child, status; static struct user_regs_struct regs; static void do_child() { ptrace(PTRACE_TRACEME, 0, 0, 0); kill(getpid(), SIGUSR1); asm ("int $0x80" : : "a" (20)); /* getpid */ } static void do_parent() { unsigned long eip, expected = 0; again: waitpid(child, &status, 0); if (WIFEXITED(status) || WIFSIGNALED(status)) return; if (WIFSTOPPED(status)) { ptrace(PTRACE_GETREGS, child, 0, ®s); eip = regs.eip; if (expected) fprintf(stderr, "child stop @ %08x, expected %08x %s\n", eip, expected, eip == expected ? "" : " <== ERROR"); if (*(unsigned short *)eip == 0x80cd) { fprintf(stderr, "int 0x80 at %08x\n", (unsigned int)eip); expected = eip + 2; } else expected = 0; ptrace(PTRACE_SINGLESTEP, child, NULL, NULL); } goto again; } int main(int argc, char * const argv[]) { child = fork(); if (child) do_parent(); else do_child(); return 0; } Signed-off-by: -
Jan Beulich authored
>commit 76381fee >Author: Vincent Hanquez <vincent.hanquez@cl.cam.ac.uk> >Date: Thu Jun 23 00:08:46 2005 -0700 > > [PATCH] xen: x86_64: use more usermode macro > > Make use of the user_mode macro where it's possible. This is useful for Xen > because it will need only to redefine only the macro to a hypervisor call. I am of the opinion that the above changeset is incomplete, i.e. it missed converting some previous uses of user_mode to user_mode_vm. While most of them could be considered just cosmetical, at least the one in die_nmi doesn't appear to be. Signed-off-by:
-
Jan Beulich authored
Registering a callback handler through register_die_notifier() is obviously primarily intended for use by modules. However, the way these currently get called it is basically impossible for them to actually be used by modules, as there is, on non-PAE configurationes, a good chance (the larger the module, the better) for the system to crash as a result. This is because the callback gets invoked (a) in the page fault path before the top level page table propagation gets carried out (hence a fault to propagate the top level page table entry/entries mapping to module's code/data would nest infinitly) and (b) in the NMI path, where nested faults must absolutely not happen, since otherwise the IRET from the nested fault re-enables NMIs, potentially resulting in nested NMI occurences. Besides the modular aspect, similar problems would even arise for in- kernel consumers of the API if they touched ioremap()ed or vmalloc()ed memory inside their handlers. Signed-off-by: -
Stas Sergeev authored
Register the boot-cpu in the cpu maps earlier to allow the early printk to work, and to fix an obscure deadlock at boot. Signed-off-by:
Stas Sergeev <stsp@aknet.ru> Signed-off-by:
-
Andrew Morton authored
Expand out these fairly pointless macros. Cc: Chuck Ebbert <76306.1226@compuserve.com> Cc: Stas Sergeev <stsp@aknet.ru> Cc: Andi Kleen <ak@muc.de> Signed-off-by:
-
Chuck Ebbert authored
Use boot info to start early_printk() at the current row on VGA console, as left by the boot loader. Signed-off-by:
-
Stas Sergeev authored
The history is that -mm kernels do not work for me for a few months already. The things started from crashing somewhere after starting init, and for the last month - no boot at all, just "Uncompressing... OK, booting kernel", and silence. Early console didn't work too. With the latest releases this degraded into an infinite stream of the "Unknown interrupt or fault" messages. So today my patience ran out and I started to think how can I collect at least some info for the bug-report. Attached is the patch that allows to gather some valueable debug info on the problem by making an early console more useable. I can't properly test the patch, as the kernel still doesn't boot, so I'll explain it in details in a hope someone else can justify the intrusive changes. arch_hooks.h: added prototypes for setup_early_printk() and early_printk(). setup.c: killed wrong setup_early_printk() prototype. Moved setup_early_printk() a bit earlier, as it was not "early enough" to cover the bug I was fighting with. early_printk.c: made it to start printing from the bottom of the screen, otherwise the messages interfere with the ones of the boot-loader, so you can't read them. Signed-off-by:
-
Chuck Ebbert authored
Allow signal handlers to set the RF bit in EFLAGS. This lets a simple debugger using SIGTRAP skip one instruction after returning from a signal. Signed-off-by:
-
Chuck Ebbert authored
There's no good reason for allowing ptrace to set the NT bit in EFLAGS, so mask it off. Signed-off-by:
-
Jean Delvare authored
Merge a few printk calls in i386 traps. Signed-off-by:
-
Chris Wright authored
mp_bus_id_to_pci_bus is declared identically twice. Signed-off-by:
Chris Wright <chrisw@sous-sol.org> Signed-off-by:
-
Natalie.Protasevich@unisys.com authored
ES7000 platform code clean up for compilation errors and a warning. Ifdef'd the ACPI related parts in the ES7000 platform code. They were causing compile errors in certain configuration (without ACPI defined). I think this approach would be best (as opposed to Kconfig changes) since it only touches the subarch... Signed-off-by: <Natalie.Protasevich@unisys.com> Signed-off-by:
-
Chuck Ebbert authored
When vendor-specific i386 initialization code is unavailable the kernel falls back to a default CPU model name. Make that model name reflect the CPU family instead of an internal vendor index. Tested on Pentium II (family 6 model 5). /proc/cpuinfo before: model name : ff/05 after: model name : 06/05 Signed-off-by:"Seth, Rohit" <rohit.seth@intel.com> Signed-off-by:
-
