June 2016 - kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

where is __percpu_seg initialized?
by 落痕 March 28, 2018

March 28, 2018

Hi all: I knew the kernel implement the per-cpu variable via __percpu_seg. For example on x86 architecture, it's defined as: #ifdef CONFIG_X86_64 #define __percpu_seg gs #define __percpu_mov_op movq #else #define __percpu_seg fs #define __percpu_mov_op movl #endif But I'm curious where __percpu_seg is initialized?? I fail to find the relative code. Thanks.

2 2

How to handle Hotplug with UIO userspace driver
by Mandeep Sandhu Sept. 28, 2017

Sept. 28, 2017

Hi All, We're working on a custom PCIe based hardware, which can be hotplugged into the system. We're using the UIO approach of having a minimal kernel driver which basically handles the interrupt and provides a mmap of the device memory to userspace and userspace driver which mmap's the provided memory and manages the device. We're using systemd and udev rules to spawn off a user process whenever the hardware is plugged into the system. Now my question is, how can we gracefully inform the userspace process if someone pulls out the hardware (sitting on a PCI slot). Even with full hotplug support, there can always be a case of "surprise removal", where someone yanks the board without going through the whole "attention button" routine. While the kernel driver has no problem with hot-removal, I see that if I "unregister" our UIO driver in response to the removal, then the mapping for the hardware's physical memory would become "invalid" (done by UIO on unregsiter). This could possibly cause the user-space process to crash due to illegal memory access. Is there a "standard" way to handle this scenario, i.e for hotpluggable hardware using UIO? Basically, we'd want to "unregister" the UIO device only after the userspace process is informed of the removal so that it could unmap/close the device and exit gracefully. We could try doing it using some systemd trick, where we put a dependency on one of the UIO created files (/dev/uioX), so that when the file is destroyed, systemd could kill/shutdown the process. Although, there's no guarantee that the userspace process won't try to access the mmaped space during that interval, so this isn't a reliable solution either. Any hints, thoughts? Thanks., -mandeep

3 7

What are need to makes humidifier?
by 윤영석 April 1, 2017

April 1, 2017

Hi, I want to make own humidifier. What sensor need when making humidifier?? humidity indicator (detecting humidity), ultrasonic waves maker?? Please shares your wisdom to me. Thanks. Best regards.

2 1

Mapping of Filename to allocated blocks
by Manoj Rao July 21, 2016

July 21, 2016

Hi All, I'm looking for a way to get a filename associated with a given physical block (ideally what I'd like is a mapping of filename <=> all the allocated blocks for this file). Is there a recommended way to do this already in kernel? if not, then where should I start looking to add changes? Help appreciated. Thanks.

3 3

what is the minimal version of make supported for building the kernel?
by Robert P. J. Day July 6, 2016

July 6, 2016

another curiosity related to my current travels through the initramfs code, i noticed this git commit from back in 2013: $ git show 7ac1815683 commit 7ac181568342ec618d4da1ab03c2babcaa7ee84f Author: Jan Beulich <JBeulich(a)suse.com> Date: Wed Dec 18 17:08:57 2013 -0800 fix build with make 3.80 According to Documentation/Changes, make 3.80 is still being supported for building the kernel, hence make files must not make (unconditional) use of features introduced only in newer versions. Commit 1bf49dd4be0b ("./Makefile: export initial ramdisk compression config option") however introduced "else ifeq" constructs which make 3.80 doesn't understand. Replace the logic there with more conventional (in the kernel build infrastructure) list constructs (except that the list here is intentionally limited to exactly one element). Signed-off-by: Jan Beulich <jbeulich(a)suse.com> Cc: P J P <ppandit(a)redhat.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds(a)linux-foundation.org> diff --git a/Makefile b/Makefile index 858a147..89cee86 100644 --- a/Makefile +++ b/Makefile @@ -732,19 +732,13 @@ export mod_strip_cmd # Select initial ramdisk compression format, default is gzip(1). # This shall be used by the dracut(8) tool while creating an initramfs image. # -INITRD_COMPRESS=gzip -ifeq ($(CONFIG_RD_BZIP2), y) - INITRD_COMPRESS=bzip2 -else ifeq ($(CONFIG_RD_LZMA), y) - INITRD_COMPRESS=lzma -else ifeq ($(CONFIG_RD_XZ), y) - INITRD_COMPRESS=xz -else ifeq ($(CONFIG_RD_LZO), y) - INITRD_COMPRESS=lzo -else ifeq ($(CONFIG_RD_LZ4), y) - INITRD_COMPRESS=lz4 -endif -export INITRD_COMPRESS +INITRD_COMPRESS-y := gzip +INITRD_COMPRESS-$(CONFIG_RD_BZIP2) := bzip2 +INITRD_COMPRESS-$(CONFIG_RD_LZMA) := lzma +INITRD_COMPRESS-$(CONFIG_RD_XZ) := xz +INITRD_COMPRESS-$(CONFIG_RD_LZO) := lzo +INITRD_COMPRESS-$(CONFIG_RD_LZ4) := lz4 +export INITRD_COMPRESS := $(INITRD_COMPRESS-y) ... etc etc ... now, Documentation/Changes currently does indeed state: Current Minimal Requirements ============================ ... snip ... o GNU make 3.80 # make --version but i did a quick grep to find: $ grep -r "else ifeq" * lib/raid6/test/Makefile:else ifeq ($(HAS_NEON),yes) $ so would that single test not violate the rule just described above? is Documentation/Changes entirely up to date with respect to minimal requirements? rday -- ======================================================================== Robert P. J. Day Ottawa, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================

2 1

IOMMU
by Max Ruttenberg July 4, 2016

July 4, 2016

What does code in drivers/iommu do? What is it responsible for and how would it relate to a custom kernel module say for a PCI driver? -- Max Ruttenberg, Member of the Technical Staff Emu *Technology* 1400 E Angela Blvd, Unit 101 South Bend, IN 46617

2 1

no device creation by udev
by anno79＠mailbox.org July 1, 2016

July 1, 2016

I have written the following lines of code to get a char device working. ( lines with # indicate that I have tried versions of this code snippet without them) class_hello = class_create(THIS_MODULE, device_name); # class_hello->devnode = &hello_devnode; retval = alloc_chrdev_region( &dev_file, 0, 1, device_name ); cdev_init( &device, &file_ops ); # device.owner = THIS_MODULE; # kobject_set_name(&device.kobj, device_name); cdev_add( &device, MKDEV(dev_file, 0), 1 ) device = device_create( class_hello, NULL, MKDEV(dev_file, 0), NULL, device_name ); # kobj_device->devt = MKDEV(dev_file, 0); Make it open / close / read / write works fine - but look at that $ ls /sys/class/hello/hello power subsystem uevent The "dev" entry is missing - and udev does not react. Anybody knows what is else needed to make udev react?

2 1

Init task 8192 byte alignment query.
by Shiva Shankar June 30, 2016

June 30, 2016

Hi I was going through arm vmlinux script file. and i found below statements /* * first, the init task union, aligned * to an 8192 byte boundary. */ INIT_TASK_DATA(THREAD_SIZE) #define INIT_TASK_DATA(align) \ . = ALIGN(align); \ *(.data..init_task) As i know we align data structures on a n byte boundary to save access time and for speed execution but i did not understand why do we need to align this particular data structure on a 8192 byte boundary. sometimes it will be on 1024 or 2048 byte boundary. Any inputs would be appreciated!!! Regards

2 1

Mount jffs2 as rootfs
by Chen, NoamX June 27, 2016

June 27, 2016

Hi all, I've encountered a problem. I'm trying to mount jffs2 rootfs. When the fs is mounted the following warning is displayed: Jffs2: Node totlen on flash (0xffffffff) != totlen from node ref (0x00000044) While the node ref varies, the node totlen on flash does not. This warning did not occur when used ramfs as rootfs, and jffs2 rootfs was mounted manually. This warning occurs when jffs2 fs tries to mark a node as obsolete. In any other sense, the fs appears operable, and I'm able to create files and delete files. Regardless, every so often the warning is displayed and aside from being annoyed by it, I fear it will eventually lead to the corruption of the fs. I use buildroot to build the rootfs image. Buildroot configuration: BR2_TARGET_ROOTFS_JFFS2=y BR2_TARGET_ROOTFS_JFFS2_FLASH_64=y BR2_TARGET_ROOTFS_JFFS2_EBSIZE=0x10000 BR2_TARGET_ROOTFS_JFFS2_PAD=y BR2_TARGET_ROOTFS_JFFS2_PADSIZE=0x00B00000 BR2_TARGET_ROOTFS_JFFS2_LE=y Linux configuration: CONFIG_JFFS2_FS=y CONFIG_JFFS2_FS_DEBUG=0 CONFIG_JFFS2_FS_WRITEBUFFER=y CONFIG_JFFS2_ZLIB=y CONFIG_JFFS2_RTIME=y The rootfs.jffs2 image is loaded to flash at offset 0x00500000, corresponding to /dev/mtdblock3 partition: > cat /proc/mtd dev: size erasesize name mtd0: 00100000 00010000 "u-boot" mtd1: 003f0000 00010000 "kernel" mtd2: 00010000 00010000 "dtb" mtd3: 00b00000 00010000 "rootfs" I pass bootargs = "root=/dev/mtdblock3 rw rootfstype=jffs2, earlycon=uart8250,mmio32,0xf4200000,115200n8 console=tty0 console=ttyS0,115200n8 consoleblank=0"; > cat /etc/fstab # <file system> <mount pt> <type> <options> <dump> <pass> /dev/root / jffs2 rw,noauto 0 1 I found that when I decreased the partition size to 0x450000 this warning was abolished, however, there was not enough space for /etc/dropbear, and I wasn't able to find a "sweet-spot" where there is enough space and no warning. I tried various padding sizes, and removing padding altogether, but to no avail. I tried padding the flash partition itself with 0xff during the construction of the flash binary (the content of the simulation). I've built some rootfs.jffs2 images with varying configurations, but none solved the issue. I looked at hexdumps of the rootfs.jffs2 image, an empty jffs2 partition and the created /dev/mtdblock3 inside the simulation-but have found nothing out of the ordinary. Any help or a lead will be greatly appreciated, Noam --------------------------------------------------------------------- Intel Israel (74) Limited This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies.

2 1

Easiest way to follow the execution process of a driver
by Teoh Choon Zone June 27, 2016

June 27, 2016

What is the easiest way to follow the execution process of a driver, on an embedded ARM board? I'm specifically interested in the stmicro ethernet driver.

2 1