- 11 Sep, 2013 40 commits
-
-
Ed Cashin authored
When the example udev rules in the documentation are used without modification, warnings like the one shown below appear in the system logs: /var/log/messages:Aug 22 11:09:11 kung udevd[445]: NAME="%k" \ is superfluous and breaks kernel supplied names, please remove \ it from /etc/udev/rules.d/60-aoe.rules:26 Removing the term does not cause any problems with the creation of the special character and block device nodes. Signed-off-by:Ed Cashin <ecashin@coraid.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Ed Cashin authored
If the system has trouble allocating memory for the creation of the aoe debugfs directory or of a file inside it, the debugfs member of an aoedev can be NULL. Do not treat a NULL debugfs pointer as a BUG on aoedev shutdown, avoiding the user impact of an unecessary panic. Signed-off-by:
-
Andy Shevchenko authored
This patch fixes following compiler warnings: drivers/block/aoe/aoecmd.c: In function `aoecmd_ata_rw': drivers/block/aoe/aoecmd.c:383:17: warning: variable `t' set but not used [-Wunused-but-set-variable] struct aoetgt *t; ^ drivers/block/aoe/aoecmd.c: In function `resend': drivers/block/aoe/aoecmd.c:488:21: warning: variable `ah' set but not used [-Wunused-but-set-variable] struct aoe_atahdr *ah; ^ Signed-off-by:Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Ed Cashin <ecashin@coraid.com> Signed-off-by:
-
Andy Shevchenko authored
In the kernel we have a nice helper that may be used here. This patch substitutes the custom implementation by the native function call. Signed-off-by:
-
Ed Cashin authored
Signed-off-by:
-
Ed Cashin authored
Signed-off-by:
-
Ed Cashin authored
This information is presented in a compact format that has evolved for easy routine scanning by expert humans, mostly developers and support technicians helping to troubleshoot or test AoE-based systems. Signed-off-by:
-
Ed Cashin authored
The place holder in the file contents is filled out in the following patch. Signed-off-by:
-
Ed Cashin authored
Signed-off-by:
-
Ed Cashin authored
This series adds the debugging information that the coraid.com-distributed aoe driver exports via sysfs, but instead of sysfs, it uses debugfs. With these patches applied, even without AoE targets on the network, KEDR reports new possible memory leaks, but these are from callers outside the aoe driver that have used aoe_devnode to get the name of the character devices through the aoe_class->devnode callback, and I believe they're responsible for freeing that memory. This patch: Create and destroy the debugfs directory. Signed-off-by:
-
Cody P Schafer authored
Signed-off-by:
Cody P Schafer <cody@linux.vnet.ibm.com> Reviewed-by:
Seth Jennings <sjenning@linux.vnet.ibm.com> Cc: David Woodhouse <David.Woodhouse@intel.com> Cc: Rik van Riel <riel@redhat.com> Cc: Michel Lespinasse <walken@google.com> Signed-off-by:
-
Cody P Schafer authored
No reason require rbtree test code to be a module, allow it to be builtin (streamlines my development process) Signed-off-by:
-
Cody P Schafer authored
Just check that we examine all nodes in the tree for the postorder iteration. Signed-off-by:
-
Cody P Schafer authored
Because deletion (of the entire tree) is a relatively common use of the rbtree_postorder iteration, and because doing it safely means fiddling with temporary storage, provide a helper to simplify postorder rbtree iteration. Signed-off-by:
-
Cody P Schafer authored
Postorder iteration yields all of a node's children prior to yielding the node itself, and this particular implementation also avoids examining the leaf links in a node after that node has been yielded. In what I expect will be its most common usage, postorder iteration allows the deletion of every node in an rbtree without modifying the rbtree nodes (no _requirement_ that they be nulled) while avoiding referencing child nodes after they have been "deleted" (most commonly, freed). I have only updated zswap to use this functionality at this point, but numerous bits of code (most notably in the filesystem drivers) use a hand rolled postorder iteration that NULLs child links as it traverses the tree. Each of those instances could be replaced with this common implementation. 1 & 2 add rbtree postorder iteration functions. 3 adds testing of the iteration to the rbtree runtime tests 4 allows building the rbtree runtime tests as builtins 5 updates zswap. This patch: Add postorder iteration functions for rbtree. These are useful for safely freeing an entire rbtree without modifying the tree at all. Signed-off-by:
-
Andrew Morton authored
Cc: Davidlohr Bueso <davidlohr@hp.com> Cc: Karel Zak <kzak@redhat.com> Cc: Matt Fleming <matt.fleming@intel.com> Signed-off-by:
-
Davidlohr Bueso authored
Trivial coding style cleanups - still plenty left. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
Davidlohr Bueso <davidlohr@hp.com> Reviewed-by:
Karel Zak <kzak@redhat.com> Acked-by:
Matt Fleming <matt.fleming@intel.com> Signed-off-by:
-
Davidlohr Bueso authored
I love emacs, but these settings for coding style are annoying when trying to open the efi.h file. More important, we already have checkpatch for that. Signed-off-by:
-
Davidlohr Bueso authored
When verifying GPT header integrity, make sure that first usable LBA is smaller than last usable LBA. Signed-off-by:
-
Davidlohr Bueso authored
The partition that has the 0xEE (GPT protective), must have the size in lba field set to the lesser of the size of the disk minus one or 0xFFFFFFFF for larger disks. Signed-off-by:
-
Davidlohr Bueso authored
One of the biggest problems with GPT is compatibility with older, non-GPT systems. The problem is addressed by creating hybrid mbrs, an extension, or variant, of the traditional protective mbr. This contains, apart from the 0xEE partition, up three additional primary partitions that point to the same space marked by up to three GPT partitions. The result is that legacy OSs can see the three required MBR partitions and at the same time ignore the GPT-aware partitions that protect the GPT structures. While hybrid MBRs are hacks, workarounds and simply not part of the GPT standard, they do exist and we have no way around them. For instance, by default, OSX creates a hybrid scheme when using multi-OS booting. In order for Linux to properly discover protective MBRs, it must be made aware of devices that have hybrid MBRs. No functionality is changed by this patch, just a debug message informing the user of the MBR scheme that is being used. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
-
Davidlohr Bueso authored
When detecting a valid protective MBR, the Linux kernel isn't picky about the partition (1-4) the 0xEE is at, but, unlike other operating systems, it does require it to begin at the second sector (sector 1). This check, apart from it not being enforced by UEFI, and causing Linux to potentially fail to detect any *valid* partitions on the disk, can present problems when dealing with hybrid MBRs[1]. For compatibility reasons, if the first partition is hybridized, the 0xEE partition must be small enough to ensure that it only protects the GPT data structures - as opposed to the the whole disk in a protective MBR. This problem is very well described by Rod Smith[1]: where MBR-only partitioning programs (such as older versions of fdisk) can see some of the disk space as unallocated, thus loosing the purpose of the 0xEE partition's protection of GPT data structures. By dropping this check, this patch enables Linux to be more flexible when probing for GPT disklabels. [1] http://www.rodsbooks.com/gdisk/hybrid.html#reactionsSigned-off-by:
-
Davidlohr Bueso authored
Per the UEFI Specs 2.4, June 2013, the starting lba of the partition that has the EFI GPT (0xEE) must be set to 0x00000001 - this is obviously the LBA of the GPT Partition Header. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
-
Davidlohr Bueso authored
The kernel's GPT implementation currently uses the generic 'struct partition' type for dealing with legacy MBR partition records. While this is is useful for disklabels that we designed for CHS addressing, such as msdos, it doesn't adapt well to newer standards that use LBA instead, such as GUID partition tables. Furthermore, these generic partition structures do not have all the required fields to properly follow the UEFI specs. While a CHS address can be translated to LBA, it's much simpler and cleaner to just replace the partition type. This patch adds a new 'gpt_record' type that is fully compliant with EFI and will allow, in the next patches, to add more checks to properly verify a protective MBR, which is paramount to probing a device that makes use of GPT. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
-
Michael Holzheu authored
Modify the s390 copy_oldmem_page() and remap_oldmem_pfn_range() function for zfcpdump to read from the HSA memory if memory below HSA_SIZE bytes is requested. Otherwise real memory is used. Signed-off-by:
Michael Holzheu <holzheu@linux.vnet.ibm.com> Cc: HATAYAMA Daisuke <d.hatayama@jp.fujitsu.com> Cc: Jan Willeke <willeke@de.ibm.com> Cc: Vivek Goyal <vgoyal@redhat.com> Signed-off-by:
-
Michael Holzheu authored
The patch "s390/vmcore: Implement remap_oldmem_pfn_range for s390" allows now to use mmap also on s390. So enable mmap for s390 again. Signed-off-by:
-
Jan Willeke authored
Introduce the s390 specific way to map pages from oldmem. The memory area below OLDMEM_SIZE is mapped with offset OLDMEM_BASE. The other old memory is mapped directly. Signed-off-by:
Jan Willeke <willeke@de.ibm.com> Signed-off-by:
-
Michael Holzheu authored
For zfcpdump we can't map the HSA storage because it is only available via a read interface. Therefore, for the new vmcore mmap feature we have introduce a new mechanism to create mappings on demand. This patch introduces a new architecture function remap_oldmem_pfn_range() that should be used to create mappings with remap_pfn_range() for oldmem areas that can be directly mapped. For zfcpdump this is everything besides of the HSA memory. For the areas that are not mapped by remap_oldmem_pfn_range() a generic vmcore a new generic vmcore fault handler mmap_vmcore_fault() is called. This handler works as follows: * Get already available or new page from page cache (find_or_create_page) * Check if /proc/vmcore page is filled with data (PageUptodate) * If yes: Return that page * If no: Fill page using __vmcore_read(), set PageUptodate, and return page Signed-off-by:
Vivek Goyal <vgoyal@redhat.com> Cc: HATAYAMA Daisuke <d.hatayama@jp.fujitsu.com> Cc: Jan Willeke <willeke@de.ibm.com> Signed-off-by:
-
Michael Holzheu authored
Exchange the old relocate mechanism with the new arch function call override mechanism that allows to create the ELF core header in the 2nd kernel. Signed-off-by:
-
Michael Holzheu authored
For s390 we want to use /proc/vmcore for our SCSI stand-alone dump (zfcpdump). We have support where the first HSA_SIZE bytes are saved into a hypervisor owned memory area (HSA) before the kdump kernel is booted. When the kdump kernel starts, it is restricted to use only HSA_SIZE bytes. The advantages of this mechanism are: * No crashkernel memory has to be defined in the old kernel. * Early boot problems (before kexec_load has been done) can be dumped * Non-Linux systems can be dumped. We modify the s390 copy_oldmem_page() function to read from the HSA memory if memory below HSA_SIZE bytes is requested. Since we cannot use the kexec tool to load the kernel in this scenario, we have to build the ELF header in the 2nd (kdump/new) kernel. So with the following patch set we would like to introduce the new function that the ELF header for /proc/vmcore can be created in the 2nd kernel memory. The following steps are done during zfcpdump execution: 1. Production system crashes 2. User boots a SCSI disk that has been prepared with the zfcpdump tool 3. Hypervisor saves CPU state of boot CPU and HSA_SIZE bytes of memory into HSA 4. Boot loader loads kernel into low memory area 5. Kernel boots and uses only HSA_SIZE bytes of memory 6. Kernel saves registers of non-boot CPUs 7. Kernel does memory detection for dump memory map 8. Kernel creates ELF header for /proc/vmcore 9. /proc/vmcore uses this header for initialization 10. The zfcpdump user space reads /proc/vmcore to write dump to SCSI disk - copy_oldmem_page() copies from HSA for memory below HSA_SIZE - copy_oldmem_page() copies from real memory for memory above HSA_SIZE Currently for s390 we create the ELF core header in the 2nd kernel with a small trick. We relocate the addresses in the ELF header in a way that for the /proc/vmcore code it seems to be in the 1st kernel (old) memory and the read_from_oldmem() returns the correct data. This allows the /proc/vmcore code to use the ELF header in the 2nd kernel. This patch: Exchange the old mechanism with the new and much cleaner function call override feature that now offcially allows to create the ELF core header in the 2nd kernel. To use the new feature the following function have to be defined by the architecture backend code to read from new memory: * elfcorehdr_alloc: Allocate ELF header * elfcorehdr_free: Free the memory of the ELF header * elfcorehdr_read: Read from ELF header * elfcorehdr_read_notes: Read from ELF notes Signed-off-by: -
Xishi Qiu authored
Code can not run here forever, so remove the unnecessary return. Signed-off-by:
Xishi Qiu <qiuxishi@huawei.com> Suggested-by:
Zhang Yanfei <zhangyanfei@cn.fujitsu.com> Reviewed-by:
Simon Horman <horms@verge.net.au> Reviewed-by:
-
Oleg Nesterov authored
The error hanling and ret-from-loop look confusing and inconsistent. - "retval >= 0" simply returns - "!bprm->file" returns too but with read_unlock() because binfmt_lock was already re-acquired - "retval != -ENOEXEC || bprm->mm == NULL" does "break" and relies on the same check after the main loop Consolidate these checks into a single if/return statement. need_retry still checks "retval == -ENOEXEC", but this and -ENOENT before the main loop are not needed. This is only for pathological and impossible list_empty(&formats) case. It is not clear why do we check "bprm->mm == NULL", probably this should be removed. Signed-off-by:
Oleg Nesterov <oleg@redhat.com> Acked-by:
Kees Cook <keescook@chromium.org> Cc: Al Viro <viro@ZenIV.linux.org.uk> Cc: Evgeniy Polyakov <zbr@ioremap.net> Cc: Zach Levis <zml@linux.vnet.ibm.com> Signed-off-by:
-
Oleg Nesterov authored
A separate one-liner for better documentation. It doesn't make sense to retry if request_module() fails to exec /sbin/modprobe, add the additional "request_module() < 0" check. However, this logic still doesn't look exactly right: 1. It would be better to check "request_module() != 0", the user space modprobe process should report the correct exit code. But I didn't dare to add the user-visible change. 2. The whole ENOEXEC logic looks suboptimal. Suppose that we try to exec a "#!path-to-unsupported-binary" script. In this case request_module() + "retry" will be done twice: first by the "depth == 1" code, and then again by the "depth == 0" caller which doesn't make sense. 3. And note that in the case above bprm->buf was already changed by load_script()->prepare_binprm(), so this looks even more ugly. Signed-off-by:
-
Oleg Nesterov authored
search_binary_handler() uses "for (try=0; try<2; try++)" to avoid "goto" but the code looks too complicated and horrible imho. We still need to check "try == 0" before request_module() and add the additional "break" for !CONFIG_MODULES case. Kill this loop and use a simple "bool need_retry" + "goto retry". The code looks much simpler and we do not even need ifdef's, gcc can optimize out the "if (need_retry)" block if !IS_ENABLED(). Signed-off-by:
-
Oleg Nesterov authored
search_binary_handler() checks ->load_binary != NULL for no reason, this method should be always defined. Turn this check into WARN_ON() and move it into __register_binfmt(). Also, kill the function pointer. The current code looks confusing, as if ->load_binary can go away after read_unlock(&binfmt_lock). But we rely on module_get(fmt->module), this fmt can't be changed or unregistered, otherwise this code is buggy anyway. Signed-off-by:
-
Oleg Nesterov authored
When search_binary_handler() succeeds it does allow_write_access() and fput(), then it clears bprm->file to ensure the caller will not do the same. We can simply move this code to exec_binprm() which is called only once. In fact we could move this to free_bprm() and remove the same code in do_execve_common's error path. Signed-off-by:
-
Oleg Nesterov authored
A separate one-liner with the minor fix. PROC_EVENT_EXEC reports the "exec" event, but this message is sent at least twice if search_binary_handler() is called by ->load_binary() recursively, say, load_script(). Move it to exec_binprm(), this is "depth == 0" code too. Signed-off-by:
-
Oleg Nesterov authored
Nobody except search_binary_handler() should touch ->recursion_depth, "int depth" buys nothing but complicates the code, kill it. Probably we should also kill "fn" and the !NULL check, ->load_binary should be always defined. And it can not go away after read_unlock() or this code is buggy anyway. Signed-off-by:
-
Oleg Nesterov authored
task_pid_nr_ns() and trace/ptrace code in the middle of the recursive search_binary_handler() looks confusing and imho annoying. We only need this code if "depth == 0", lets add a simple helper which calls search_binary_handler() and does trace_sched_process_exec() + ptrace_event(). The patch also moves the setting of task->did_exec, we need to do this only once. Note: we can kill either task->did_exec or PF_FORKNOEXEC. Signed-off-by:
-
Oleg Nesterov authored
proc_fd_permission() says "process can still access /proc/self/fd after it has executed a setuid()", but the "task_pid() = proc_pid() check only helps if the task is group leader, /proc/self points to /proc/<leader-pid>. Change this check to use task_tgid() so that the whole thread group can access its /proc/self/fd or /proc/<tid-of-sub-thread>/fd. Notes: - CLONE_THREAD does not require CLONE_FILES so task->files can differ, but I don't think this can lead to any security problem. And this matches same_thread_group() in __ptrace_may_access(). - /proc/self should probably point to /proc/<thread-tid>, but it is too late to change the rules. Perhaps it makes sense to add /proc/thread though. Test-case: void *tfunc(void *arg) { assert(opendir("/proc/self/fd")); return NULL; } int main(void) { pthread_t t; pthread_create(&t, NULL, tfunc, NULL); pthread_join(t, NULL); return 0; } fails if, say, this executable is not readable and suid_dumpable = 0. Signed-off-by:
-
