November 2019 - kernelnewbies@kernelnewbies.org

Question regarding kernel?
by Strahinja Mitrovic Nov. 16, 2019

Nov. 16, 2019

Hi, I'm a new subscriber, and i have a question regarding the uart driver and kernel. How does uart driver reset the counter of input serial line interrupts? I'm currently using ioct with TIOCGICOUNT on linux for retrieving counts of input serial line interrupts. The counts are written to the serial_icounter_struct. Most data fields are 32-bit integers. What happens if a number of counts for a certain input line is greater then 32-bit integer value? Does a system refresh counter on daily basis automatically? Is there a way to reset a counter so that starts again from 0? Best Regards, Strahinja Mitrovic

2 1

Which package typically includes network controller drivers
by Jeffrey Walton Nov. 16, 2019

Nov. 16, 2019

Hi Everyone, My apologies if this one is off-topic. I recently encountered a bad apt update using Ubuntu. The 5.0.0-36 kernel was installed, but it was missing network card drivers. I tried to modprobe the network card drivers but they were missing. I was able to recover by standing up a VM, copying the *.deb files from /var/cache/apt/archive, sneaker netting to the wounded machine, and then manually re-installing the packages. My question is, which package typically includes the network card drivers? I believe the choices are linux-modules or linux-modules-extras. Thanks in advance $ apt-cache search 5.0.0-36 | grep generic | cut -f 1 -d ' ' linux-buildinfo-5.0.0-36-generic linux-cloud-tools-5.0.0-36-generic linux-headers-5.0.0-36-generic linux-image-5.0.0-36-generic linux-image-unsigned-5.0.0-36-generic linux-modules-5.0.0-36-generic linux-modules-extra-5.0.0-36-generic linux-tools-5.0.0-36-generic

2 3

NMI injection
by Muni Sekhar Nov. 15, 2019

Nov. 15, 2019

Hi All, Can someone clarify me what is “NMI injection” means? Does it helps when system is frozen? If so how to do this for Intel Atom processor machines? I configured my system to respond for SysRq key press, but I see when system is frozen then SysRq key press has no effect. I’ve read that when kernel is locked up it won’t respond for SysRq key press actions. What is the definition for “kernel locked up” means? And how to solve this? -- Thanks, Sekhar

1 0

Switch a fd between blocking and non-blocking mode?
by Jeffrey Walton Nov. 14, 2019

Nov. 14, 2019

Hi Everyone, I have one more question related to my problem of my program losing data. I want to ensure I'm not using an anti-pattern that's causing the problem. I have a worker thread that blocks on read(2). When the read() occurs, the fd is switch from blocking to non-blocking. Additional reads are performed until EAGAIN or EWOULDBLOCK. Then the fd is switched back to blocking mode. Does switching a socket between blocking and non-blocking mode cause the kernel to reset or delete queued data (that has not been read by the application yet)? Thanks in advance. Jeff ---------- Here are the functions that switch between blocking and non-blocking mode. There's not much to them. void make_blocking_fd(int fd) { const int old = fcntl(fd, F_GETFL, 0); fcntl(fd, F_SETFL, old & ~(int)O_NONBLOCK); } void make_nonblocking_fd(int fd) { const int old = fcntl(fd, F_GETFL, 0); fcntl(fd, F_SETFL, old | (int)O_NONBLOCK); }

2 2

[PATCH] hwmon: (dell-smm-hwmon) Disable BIOS fan control upon loading
by Giovanni Mascellani Nov. 13, 2019

Nov. 13, 2019

Hello, I would like to submit the following patch for inclusion in Linux. Since it is my first Linux patch, I would like to have a preliminary review here, if possible. I have a question in particular: I would like the functions i8k_disable_bios() and i8k_enable_bios() to be called when the computer is about to enter or is recovering suspension to RAM, or is recovering from hibernation to disk. How can I do that? The original commit message follows. Thanks, Giovanni. On some Dell laptops the BIOS directly controls fan speed, ignoring SET_FAN commands. Also, the BIOS controller often happens to be buggy, failing to increase fan speed when the CPU is under heavy load and setting fan at full speed even when the CPU is idle and relatively cool. Disable BIOS fan control on such laptops at module loading, and re-enable it at module unloading, so that a userspace controller like i8kmon can take control of the fan. Signed-off-by: Giovanni Mascellani <gio(a)debian.org> --- drivers/hwmon/dell-smm-hwmon.c | 61 ++++++++++++++++++++++++++++++++++ 1 file changed, 61 insertions(+) diff --git a/drivers/hwmon/dell-smm-hwmon.c b/drivers/hwmon/dell-smm-hwmon.c index 4212d022d253..6110558c8609 100644 --- a/drivers/hwmon/dell-smm-hwmon.c +++ b/drivers/hwmon/dell-smm-hwmon.c @@ -32,6 +32,12 @@ #include <linux/i8k.h> +/* + * The code for enabling and disabling BIOS fan control were found by + * trial and error with the program at + * https://github.com/clopez/dellfan. + */ + #define I8K_SMM_FN_STATUS 0x0025 #define I8K_SMM_POWER_STATUS 0x0069 #define I8K_SMM_SET_FAN 0x01a3 @@ -43,6 +49,8 @@ #define I8K_SMM_GET_TEMP_TYPE 0x11a3 #define I8K_SMM_GET_DELL_SIG1 0xfea3 #define I8K_SMM_GET_DELL_SIG2 0xffa3 +#define I8K_SMM_DISABLE_BIOS 0x30a3 +#define I8K_SMM_ENABLE_BIOS 0x31a3 #define I8K_FAN_MULT 30 #define I8K_FAN_MAX_RPM 30000 @@ -68,6 +76,7 @@ static uint i8k_pwm_mult; static uint i8k_fan_max = I8K_FAN_HIGH; static bool disallow_fan_type_call; static bool disallow_fan_support; +static bool disable_bios; #define I8K_HWMON_HAVE_TEMP1 (1 << 0) #define I8K_HWMON_HAVE_TEMP2 (1 << 1) @@ -419,6 +428,26 @@ static int i8k_get_power_status(void) return (regs.eax & 0xff) == I8K_POWER_AC ? I8K_AC : I8K_BATTERY; } +/* + * Disable BIOS fan control. + */ +static int i8k_disable_bios(void) +{ + struct smm_regs regs = { .eax = I8K_SMM_DISABLE_BIOS, .ebx = 0 }; + + return i8k_smm(&regs) ? : regs.eax & 0xff; +} + +/* + * Enable BIOS fan control. + */ +static int i8k_enable_bios(void) +{ + struct smm_regs regs = { .eax = I8K_SMM_ENABLE_BIOS, .ebx = 0 }; + + return i8k_smm(&regs) ? : regs.eax & 0xff; +} + /* * Procfs interface */ @@ -1135,6 +1164,22 @@ static struct dmi_system_id i8k_blacklist_fan_support_dmi_table[] __initdata = { { } }; +/* + * On some machines the BIOS disregards all SET_FAN commands unless it + * is explicitly disabled. + * See bug: https://bugzilla.kernel.org/show_bug.cgi?id=200949 + */ +static struct dmi_system_id i8k_disable_bios_dmi_table[] __initdata = { + { + .ident = "Dell Precision 5530", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "Precision 5530"), + }, + }, + { } +}; + /* * Probe for the presence of a supported laptop. */ @@ -1169,6 +1214,11 @@ static int __init i8k_probe(void) disallow_fan_type_call = true; } + if (dmi_check_system(i8k_disable_bios_dmi_table)) { + pr_warn("broken Dell BIOS detected, disabling BIOS fan control\n"); + disable_bios = true; + } + strlcpy(bios_version, i8k_get_dmi_data(DMI_BIOS_VERSION), sizeof(bios_version)); strlcpy(bios_machineid, i8k_get_dmi_data(DMI_PRODUCT_SERIAL), @@ -1218,6 +1268,11 @@ static int __init i8k_probe(void) i8k_fan_mult = fan_mult; } + if (disable_bios) { + if (i8k_disable_bios()) + pr_warn("could not disable BIOS fan control\n"); + } + return 0; } @@ -1241,6 +1296,12 @@ static void __exit i8k_exit(void) { hwmon_device_unregister(i8k_hwmon_dev); i8k_exit_procfs(); + + if (disable_bios) { + pr_warn("re-enabling BIOS fan control\n"); + if (i8k_enable_bios()) + pr_warn("could not re-enable BIOS fan control\n"); + } } module_init(i8k_init); -- 2.24.0

3 4

How to log USB modem data for a specific device?
by Jeffrey Walton Nov. 12, 2019

Nov. 12, 2019

Hi Everyone, I'm having trouble with a C application that opens a USB modem. The application is not receiving some data from a USR modem. Conexant modems are OK. I'm working through https://www.kernel.org/doc/Documentation/usb/usbmon.txt . I found the modem at Bus=01 Dev=24: T: Bus=01 Lev=03 Prnt=07 Port=00 Cnt=01 Dev#= 24 Spd=480 MxCh= 0 D: Ver= 2.00 Cls=02(comm.) Sub=00 Prot=00 MxPS=64 #Cfgs= 1 P: Vendor=0baf ProdID=0303 Rev= 2.00 S: Manufacturer=U.S.Robotics S: Product=USB Modem This is where my confusion lies. It is not clear to me how to determine which socket to use for step #3 in usbmon.txt. Step #3 says: # cat /sys/kernel/debug/usb/usbmon/3u > /tmp/1.mon.out How do I determine which socket to use for Bus=01 Dev=24? Thanks in advance.

2 7

Netlink socket returns NULL in vmx.c kernel file
by Irfan Ullah (울라 이르판) Nov. 8, 2019

Nov. 8, 2019

Dear All, I am posting again the same problem that is "Netlink socket creation returns NULL" when I try to create it in handle_ept_violation() in the vmx.c file. Actually, I am trying to use netlink to send information (i.e., guest physical address) from kernel space to user space. The netlink socket creation is not working in the context of vmx.c, but it works perfectly when I test my code in a kernel module and load it. Code is in the attachement. I have tested many things such as shared memory, kernel thread, sleep function, etc. but nothing worked out. Can you please what's the problem? *What's the best too/way that can provide full duplex (two-way) communication between kernel and user spaces in any context (from any kernel source file)*. Thank you. -- *Best Regards,* *Mr. Irfan Ullah* PhD Candidate Data and Knowledge Engineering(DKE) Lab Department of Computer Science and Engineering Kyung Hee University, South Korea. +82-010-591-51651 <+82%2010-3877-8867> sahibzada.iu(a)gmail.com sahibzada_irfanullah

3 6

Kernel Module with multiple source files not initializing
by Irfan Ullah (울라 이르판) Nov. 7, 2019

Nov. 7, 2019

Dear All, I have developed a kernel module consists of one source file that sends and receives message to the user space. I have spitted the source code in two files, and now I am trying to develop kernel module from these source files. After compiling and linking without any problems, I insmod the module but I didn’t see any of the printk() I wrote, in fact, the module can be inserted and removed, but it does nothing. Code is in the attached file zipped file. I tried many things , e.g., link <https://paguilar.org/?p=7>1, link2 <https://www.linuxquestions.org/questions/linux-kernel-70/kernel-module-with…>, and link3 <https://android.googlesource.com/kernel/msm/+/android-msm-dory-3.10-kitkat-…> etc. but nothings worked out. I also used *nm* to inspect the module, but, as expected, there was no signs of "__init and __exit" functions can in the output. U __alloc_skb 00000000000001c0 T create_socket 0000000000000030 T data_update U __fentry__ U fortify_panic 0000000000000000 r __func__.5058 U init_net 0000000000000000 t kernel_space_receiver 0000000000000080 T kernel_space_sender U memcpy 0000000000000038 r __module_depends 0000000000000008 D msg_size U msleep U __netlink_kernel_create U netlink_kernel_release U netlink_unicast 0000000000000078 B nlh U __nlmsg_put 0000000000000080 B nl_sk 0000000000000000 r _note_6 0000000000000004 D pid U printk 0000000000000000 D res 0000000000000000 B s_data 0000000000000070 B skb_out U __stack_chk_fail U strlen 0000000000000000 D __this_module 0000000000000000 r __UNIQUE_ID_license381 000000000000004d r __UNIQUE_ID_name240 0000000000000041 r __UNIQUE_ID_retpoline241 0000000000000010 r __UNIQUE_ID_srcversion242 0000000000000068 r __UNIQUE_ID_vermagic239 -------------------- *Following is the output of some useful commands:* ------------------------------------------------- *ssd@ssd:~/Desktop/Kernel_User_comm$ ls* Makefile netlink_kernel_module.c netlink_kernel_space.c netlink_kernel_space.h netlink_user_module.c --------------------------------------------------- *ssd@ssd:~/Kernel_User_comm$ make --version* GNU Make 4.1 Built for x86_64-pc-linux-gnu Copyright (C) 1988-2014 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html > This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. ------------------------------------------------------- *ssd@ssd:~/Kernel_User_comm$ uname -r* 5.2.0-rc6 -------------------------------------------------------- Can you please help me: what's the problem/mistake I am doing? Thank you very much. -- *Best Regards,* *Mr. Irfan Ullah* PhD Candidate Data and Knowledge Engineering(DKE) Lab Department of Computer Science and Engineering Kyung Hee University, South Korea. +82-010-591-51651 <+82%2010-3877-8867> sahibzada.iu(a)gmail.com sahibzada_irfanullah

4 8

Linux stable: 4.19 vs 4.14
by Avinash Patil Nov. 5, 2019

Nov. 5, 2019

Hi Greg, I am looking at Linux kernel page: https://www.kernel.org/category/releases.html I am curious as to why Linux4.19 which was released later has earlier EOL than 4.14? If we have to choose one version over another for BSP, which one is preferred? Thanks, Avinash

2 2

[PATCH 1/1] syscalls: Fix references to filenames containing syscall defs
by Mohammad Nasirifar Nov. 5, 2019

Nov. 5, 2019

Fix stale references to files containing syscall definitions in 'include/linux/syscalls.h' and 'include/uapi/asm-generic/unistd.h', pointing to 'kernel/itimer.c', 'kernel/hrtimer.c', and 'kernel/time.c'. They are now under 'kernel/time'. Also definitions of 'getpid', 'getppid', 'getuid', 'geteuid', 'getgid', 'getegid', 'gettid', and 'sysinfo' are now in 'kernel/sys.c'. Signed-off-by: Mohammad Nasirifar <farnasirim(a)gmail.com> --- include/linux/syscalls.h | 8 ++++---- include/uapi/asm-generic/unistd.h | 8 ++++---- 2 files changed, 8 insertions(+), 8 deletions(-) diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index f7c561c4dcdd..1dbefa5e00e5 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -584,13 +584,13 @@ asmlinkage long sys_get_robust_list(int pid, asmlinkage long sys_set_robust_list(struct robust_list_head __user *head, size_t len); -/* kernel/hrtimer.c */ +/* kernel/time/hrtimer.c */ asmlinkage long sys_nanosleep(struct __kernel_timespec __user *rqtp, struct __kernel_timespec __user *rmtp); asmlinkage long sys_nanosleep_time32(struct old_timespec32 __user *rqtp, struct old_timespec32 __user *rmtp); -/* kernel/itimer.c */ +/* kernel/time/itimer.c */ asmlinkage long sys_getitimer(int which, struct itimerval __user *value); asmlinkage long sys_setitimer(int which, struct itimerval __user *value, @@ -731,7 +731,7 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, unsigned long arg4, unsigned long arg5); asmlinkage long sys_getcpu(unsigned __user *cpu, unsigned __user *node, struct getcpu_cache __user *cache); -/* kernel/time.c */ +/* kernel/time/time.c */ asmlinkage long sys_gettimeofday(struct timeval __user *tv, struct timezone __user *tz); asmlinkage long sys_settimeofday(struct timeval __user *tv, @@ -739,7 +739,7 @@ asmlinkage long sys_settimeofday(struct timeval __user *tv, asmlinkage long sys_adjtimex(struct __kernel_timex __user *txc_p); asmlinkage long sys_adjtimex_time32(struct old_timex32 __user *txc_p); -/* kernel/timer.c */ +/* kernel/sys.c */ asmlinkage long sys_getpid(void); asmlinkage long sys_getppid(void); asmlinkage long sys_getuid(void); diff --git a/include/uapi/asm-generic/unistd.h b/include/uapi/asm-generic/unistd.h index 1fc8faa6e973..6423f5fa4b8e 100644 --- a/include/uapi/asm-generic/unistd.h +++ b/include/uapi/asm-generic/unistd.h @@ -328,13 +328,13 @@ __SC_COMP(__NR_set_robust_list, sys_set_robust_list, \ __SC_COMP(__NR_get_robust_list, sys_get_robust_list, \ compat_sys_get_robust_list) -/* kernel/hrtimer.c */ +/* kernel/time/hrtimer.c */ #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32 #define __NR_nanosleep 101 __SC_3264(__NR_nanosleep, sys_nanosleep_time32, sys_nanosleep) #endif -/* kernel/itimer.c */ +/* kernel/time/itimer.c */ #define __NR_getitimer 102 __SC_COMP(__NR_getitimer, sys_getitimer, compat_sys_getitimer) #define __NR_setitimer 103 @@ -507,7 +507,7 @@ __SYSCALL(__NR_prctl, sys_prctl) #define __NR_getcpu 168 __SYSCALL(__NR_getcpu, sys_getcpu) -/* kernel/time.c */ +/* kernel/time/time.c */ #if defined(__ARCH_WANT_TIME32_SYSCALLS) || __BITS_PER_LONG != 32 #define __NR_gettimeofday 169 __SC_COMP(__NR_gettimeofday, sys_gettimeofday, compat_sys_gettimeofday) @@ -517,7 +517,7 @@ __SC_COMP(__NR_settimeofday, sys_settimeofday, compat_sys_settimeofday) __SC_3264(__NR_adjtimex, sys_adjtimex_time32, sys_adjtimex) #endif -/* kernel/timer.c */ +/* kernel/sys.c */ #define __NR_getpid 172 __SYSCALL(__NR_getpid, sys_getpid) #define __NR_getppid 173 -- 2.23.0

2 1