IPMI intel LPC driver
Hi, everyone I'm trying to search for Intel Host IPMI driver for LPC bus. I saw the IPMI interfaces as the following : - DMI - ACPI (PNP, SPMI) - PCI - I2C - hardcoded Among these interface, is there interface for LPC ipmi driver? Can any one inform me of the Intel IPMI interface driver for LPC? Thanks in advance, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
PCIe bus enumeration
Hi, Is there any method for assigning PCIe bus fixed? For example, my device pcie bus number is changed when another pcie device is inserted. I have a 4 same pcie device and i want to identify each device according to the pci bus number (according to pcie slot they are inserted into). When i change kernel boot parameter for kernel to assign the pci bus number, the result was the same. Is there any method for that? Thanks in advance. Regards, J.Hwan Kim. ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
pmbus driver for x86
Hi, My server (S2600CP) has BMC module and power supply with pmbus function. When I try to control and monitor with pmbus driver in linux, it fails. I connected oscilloscope to pmbus connector (SCL, SDA) pins, there was no signal from pmbus driver. But when the system went down, there are pmbus signals which may be generated by BMC. In x86 server, is pmbus connected only to BMC(ipmi)? Is it not connected to CPU i2c bus? is it impossible to monitor power supply with pmbus drvier? Thanks in advance. Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Non-maskable interrupt
Hi, everyone In my system, NMI intrrupt counter is increased continuously, which I found in /proc/interrupts. How can I find the reasons of NMI interrupts? Best Regards Kim. ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Re: Non-maskable interrupt
2014년 10월 23일 02:37에 valdis.kletni...@vt.edu 이(가) 쓴 글: On Thu, 23 Oct 2014 00:26:32 +0900, J.Hwan Kim said: Hi, everyone In my system, NMI intrrupt counter is increased continuously, which I found in /proc/interrupts. How can I find the reasons of NMI interrupts? How fast are they increasing? A few an hour? A few a minute? once per second... and I found the nmi couter is same with PMI. What does 'grep -i nmi your-config-here' return? My kernel config about NMI is as follows : CONFIG_DEBUG_NMI_TIMEOUT=5 CONFIG_HAVE_FTRACE_NMI_ENTER=y CONFIG_FTRACE_NMI_ENTER=y Thanks ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Re: Non-maskable interrupt
2014년 10월 23일 11:06에 valdis.kletni...@vt.edu 이(가) 쓴 글: On Thu, 23 Oct 2014 10:47:57 +0900, J.Hwan Kim said: How fast are they increasing? A few an hour? A few a minute? once per second... I bet 'cat /proc/sys/kernel/nmi_watchdog' returns '1'. See Documentation/lockup-watchdogs.txt In my kernel, there is no kernel paramter of nmi_watchdog and I cannot find the /proc/sys/kernel/nmi_watchdog file in my system. ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
The start physical and virtual address of NUMA memory node
Hi, everyone I have 2 memory nodes in my NUMA system. I want to know the start (physical and virtual) address of each memory node. How can I do? Please give me any hint... Thanks in advance J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Re: The start physical and virtual address of NUMA memory node
2014년 09월 02일 00:57, Oussama Jabbari 쓴 글: I have 2 memory nodes in my NUMA system. I want to know the start (physical and virtual) address of each memory node. How can I do? Please give me any hint... Hi, there is some infos when you type : dmesg | grep -i numa My embedded system says no numa configuration information through kernel message. Is there other methods? If I trace the data structure pg_data_t *pgdat = NODE_DATA(nid), can I find out the information of start address in each nodes? Is there simple way rather than searching the kernel MM source code? Thanks ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
How to diasble interrupt of current core in multicore cpu
Hi, everyone How to disable interrupt for current core in multicore CPU? Thanks in advance Kim. ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
module usage count
Hi, everyone When I use /proc file system in communicating between kernel and user space, is the usage count of moudule automatically incremented or decremented? I register proc file via create_proc_entry() and use .open, .close and .mmap not using .read and .write function. Please give me any hint.. Thanks in advance Best Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
How to prevent other processes from being scheduled
Hi, everyone How can I prevent other processes than my specific application from being scheduled? I thougt that when my program being started, it would invokes the kernel to do local_irq_disable() and when it ended, it would call local_irq_enable() in order that scheduler not working when my specific program runs. But i seems do not work as I likes. Please give me some hints what i should do... Thanks in advance. J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
How to view processor schedule history
Hi, everyone Can I find the schedule history of CPU in system, for example /proc Thanks in advance J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
MAX_ORDER in mmzone.h
Hi, everyone I wish to expand the limit for allocating contiguous memory in kernel. So, if I change the value CONFIG_FORCE_MAX_ZONEORDER or MAX_ORDER value in /include/linux/mmzone.h, will it cause any problem or side effect in x86_64 sysem? Thanks in advance. Best Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
How to allocate hugepage in kernel module?
Hi, everyone How can I allocate physically contiguous huge page in kernel module ? The routine of _get_fee_pages() fails whenever there are much free memory in system. I found the procedures for set hugepages with sysctl or /proc/sys/vm/nr_hugepages but I've not found the api for allocate the huge page in kernel. Thanks in advnace. Best Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Re: How to allocate hugepage in kernel module?
I set the __GFP_COMP flag to __get_free_pages(), but the result was same. 2012년 08월 09일 17:59, Denis Kirjanov 쓴 글: Forgot to CC kernelnewbies: Did you pass the __GFP_COMP flag to __get_free_pages? On 8/9/12, J.Hwan Kimfrog1...@gmail.com wrote: Hi, everyone How can I allocate physically contiguous huge page in kernel module ? The routine of _get_fee_pages() fails whenever there are much free memory in system. I found the procedures for set hugepages with sysctl or /proc/sys/vm/nr_hugepages but I've not found the api for allocate the huge page in kernel. Thanks in advnace. Best Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
is memcpy a system call?
Hi, everyone Is memcpy a system call? Thanks in advnace, Best Regards, J.Hwna Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
PCI system start point
Hi, everyone In kernel source 2.6.25, the start point of PCIe system was pci_init(). Now in version 3.x.x, the function name of pci_init() disappeared. Where is the start point of pci system in current version? Thanks in advance. Best Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
SRIOV memory allocation
Hi, everyone I'd like to know what routine of the PCIe allocates SR-IOV BAR memory. When I put into the network card supporting SR-IOV in my system, it returns error indicating MMIO resource is insufficient when sriov is enabled. The routine is in iov.c for (i = 0; i PCI_SRIOV_NUM_BARS/*6*/; i++) { res = dev-resource + PCI_IOV_RESOURCES + i; if (res-parent) nres++; } if (nres != iov-nres) { dev_err(dev-dev, not enough MMIO resources for SR-IOV\n); return -ENOMEM; } So, I hope to know where the resources of IOV is assigned the parent pointer. Thanks in advance. Best Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
kernel memory allocation
Hi, everyone How can I allocated contiguous kernel memory over 128MB ? When I use _get_free_pages() function, it returns error. I guess the memory size is greater than the amount which the function can allocate. Thanks in advance. Best Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
The core in which NAPI is using
Hi, everyone My ethernet card always interrupts at Core 0 . The driver of the device is using NAPI. When NAPI is appled, does the core in which NAPI is processing vary? Is the NAPI triggered by software interrupt, isn't it? My system monitor indicates the CPU loading varies randomly among 8 cores. Can you give an information if the core which NAPI is using varies? Thanks in advance. Best Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
How to distribute IRQs to all cores
Hi, everyone The interrupts of my ixgbevf driver occurs only Core 0 although the user space irqbalance serivce is working. How can I distribute the interrupt of RX in ixgbevf to all cores? Is there any kernel compile option for it ? cat /proc/interrupts | grep isv 97: 8 0 0 0 0 0 0 0 PCI-MSI-edge isv0-rx-0 99: 7 0 0 0 0 0 0 0 PCI-MSI-edge isv0:lsc 103: 2059 0 0 0 0 0 0 0 PCI-MSI-edge isv2-rx-0 104: 140 0 0 0 0 0 0 PCI-MSI-edge isv2-tx-0 105: 1 0 0 0 0 0 0 0 PCI-MSI-edge isv2:mbx isv is netdevice name of my ixgbevf. Thanks in advance. Best Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Re: How to distribute IRQs to all cores
On 2011? 08? 26? 03:19, Akash wrote: Try programming the CPU-irq affinity by writing to /proc/irq/irq-number/smp_affinity For example masks for irqs could be : irq 97 - 0x1 99 - 0x2 103 - 0x4 104 - 0x8 That way, you are hardcoding on which cpu the particular irqs should land. HTH Akash On Thu, Aug 25, 2011 at 4:08 AM, J.Hwan Kim frog1...@gmail.com mailto:frog1...@gmail.com wrote: Hi, everyone The interrupts of my ixgbevf driver occurs only Core 0 although the user space irqbalance serivce is working. How can I distribute the interrupt of RX in ixgbevf to all cores? Is there any kernel compile option for it ? cat /proc/interrupts | grep isv 97: 8 0 0 0 0 0 0 0 PCI-MSI-edge isv0-rx-0 99: 7 0 0 0 0 0 0 0 PCI-MSI-edge isv0:lsc 103: 2059 0 0 0 0 0 0 0 PCI-MSI-edge isv2-rx-0 104: 140 0 0 0 0 0 0 PCI-MSI-edge isv2-tx-0 105: 1 0 0 0 0 0 0 0 PCI-MSI-edge isv2:mbx isv is netdevice name of my ixgbevf. Thank you for reply. I want to distribute, for example, IRQ 103 to all cores as follows : 103 : 111 2344 1334 3233 2233 2233 362262 isv2-rx-0 That cannot be achieved by setting /proc/irq/irq-num/smp_affinity. ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Soft interrupt in multi-core environment
Hi, everyone When network card interrupts occur in multi-core system, what core is used by soft irq for network interrupts? That is, for example, if the IRQ happens in core 0, does the soft irq run on the same core ? Is it possible for soft irq to use other core different from hard IRQ? Thanks in advance. Best Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
PCI IO memory resource allocation
Hi, everyone In kernel source, where is the PCI IO memory resource allocation is done? (memory mapped io) And can you explain the IO memory resource allocation mechanism? Thanks in advance Best Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
MSI-edge in /proc/interrupts
Hi, everyone When I read /proc/interrupts, it displays interrupt type of my ethernet card with MSI-edge. Does it mean MSI-X interrupt or MSI interrupt? I set to my ethernet driver operates with MSI-X interrupt. Thanks in advance. Best Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
How to get current core id in kernel source
Hi, everyone I wish to get the current core id in which the code is being processed, in device driver code. What function or global variable can I use ? Thanks in advance. Best Regards, J.Hwan Kim ___ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies