- kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

HID Monitor Driver Advice
by mark＠bolhuis.dev April 16, 2021

April 16, 2021

Hi all My name is Mark and I'm currently writing a device driver for Eizo EV FlexScan monitors, so that I can control them over usb. This driver is built by reverse engineering the hid reports sent by the propriety and windows only software. It can be found here: https://github.com/markbolhuis/openeizo I have some questions about this that hopefully somebody can answer. 1) What would be the recommended way to interface this with userspace? My plan is to expose a device file for each unique setting that the monitor has such as brightness control, input selection and so on... However I'm not sure this is a suitable way of doing this since the monitors have many options that are very similar. For example contiguous numerical values only differ by their usage and value min and max, meaning that if I have a file per setting there will be a lot of repetition. There are already 40+ unique attributes/settings that may need a file. 2) To properly initialize the driver I will have to fetch a second report descriptor from the monitor. This isn't a standard hid report descriptor. It seems that Eizo are simply using the format instead of inventing their own. Is there a way for the hid subsystem to parse arbitrary report descriptors and expose them as a structure for generic use? I couldn't find much that would help with this. hid_parse_report @ inlcude/linux/hid.h:895 seems to be the closest to what I need but it's tied to a device. 3) I contacted Eizo and politely asked if they could provide an hid specification for the vendor defined usages however they weren't able to help. This forced me to reverse engineer. Is there anything that I should be aware of if I want to submit code that based on reverse engineering? I didn't reverse engineer the windows driver itself, just sniffed the packets with usbmon. 4) How would I find people who have compatible monitors to test it out and/or help reverse engineer? I only have one monitor so I'm writing against that, with no guarantee that it will work for others. Thank you for reading Mark Bolhuis

3 4

Why does dump_stack need asmlinkage?
by Lv Ying April 14, 2021

April 14, 2021

Hi all: asmlinkage is only defined in arch ia64 and x86_32: 1. arch/x86/include/asm/linkage.h: #ifdef CONFIG_X86_32 #define asmlinkage CPP_ASMLINKAGE __attribute__((regparm(0))) #endif /* CONFIG_X86_32 */ 2. arch/ia64/include/asm/linkage.h: #define asmlinkage CPP_ASMLINKAGE __attribute__((syscall_linkage)) 3. arch independent include/linux/linkage.h: #ifdef __cplusplus #define CPP_ASMLINKAGE extern "C" #else #define CPP_ASMLINKAGE #endif #ifndef asmlinkage #define asmlinkage CPP_ASMLINKAGE #endif As kernelnewbies FAQ(https://kernelnewbies.org/FAQ/asmlinkage) said, in short, asmlinkage tells your compiler to look on the CPU stack for the function parameters, instead of registers. So, there are two questions: 1. asmlinkage is only defined in x86_32 and ia64, what about other archs? How do they achieve the same effect as mentioned above? 2. "asmlinkage __visible void dump_stack(void)", dump_stack does not have any parameter, so why do it need asmlinkage? Thanks! Lv Ying

1 0

Linux kernel BSP
by Pankaj Vinodrao Joshi April 14, 2021

April 14, 2021

Hi, i am newbie to linux BSP and i want to get some hand on with BSP. Can someone kindly suggest what all the components of BSP also i need to port BSP from one arch to other what and all things i should consider to get it ported ?? Thanks

3 8

Notify special task kill using wait* functions
by John Wood April 11, 2021

April 11, 2021

Hi, I'm working in a LSM whose goal is to detect and mitigate fork brute force attacks against vulnerable userspace applications. The detection and mitigation works as expected by I'm stuck at this point. The mitigation method used is to kill all the offending tasks involved in the attack. To do so, I kill the tasks using: do_send_sig_info(SIGKILL, SEND_SIG_PRIV, p, PIDTYPE_PID); The question is: How can I notify to wait* functions that the task has been killed by the "Brute" LSM. For example, in the function wait_task_zombie, in the "out_info:" label, the code is the following: out_info: infop = wo->wo_info; if (infop) { if ((status & 0x7f) == 0) { infop->cause = CLD_EXITED; infop->status = status >> 8; } else { infop->cause = (status & 0x80) ? CLD_DUMPED : CLD_KILLED; infop->status = status & 0x7f; } infop->pid = pid; infop->uid = uid; } I think I need to modify this code to achieve the commented goal. But I don't know how to proceed. Is it possible to kill a task in a way that this code can detect ? Now, with this code, we know that a task has been killed and with what signal. But is it possible to know that the task has been killed by the "Brute" LSM. Using for example the 8 upper bits in the status that I think are unused when a process is killed or dumped. It is possible to use the do_send_sig_info passing to the struct kernel_siginfo the necessary information to be able to detect this scenario in the wait* functions? If yes, what info I need to pass? The final purpose of all of this is that using the waitid function from userspace (or waitpid function) we can know that a child task has been killed by the "Brute" LSM or not. I try to inform to userspace that a task has been killed due to the "Brute" mitigation. Sorry, but I'm stuck at this point. Any help would be greatly appreciated. Regards, John Wood

3 17

Controlling IOVA addresses generated by IOMMU
by Sadanand Warrier April 7, 2021

April 7, 2021

Folks Is there any kernel API or control mechanism for an IO-MMU that one could use to specify that a buffer earmarked for DMA would always have an IOVA greater than 4GB? Basically if I used pci_map_sg and passed a scatter list I would like the dma_address field in the scatterlist structure to be always greater than 4GB when it returned. Thanks S

1 0

How to free a page table page
by Wonkyo Choe April 6, 2021

April 6, 2021

Hello there, I'm trying to manage a custom page table list for my toy project. A basic process of the custom list is to allocate few pages (struct page) to the list from the buddy allocator and the list will give some pages when a user process needs page table pages. In contrast, when this process terminates, page table pages will return to the list. Finally, when the list is not needed, all pages will be delivered to the buddy allocator. This relationship can be described as: process - custom list - buddy allocator My problem is that a user process' page table pages (pud, pmd, pte) seem to free their page to the buddy allocator instead of to the custom list. (allocation works. I've checked it) I put my custom functions in release_pages / free_unref_page / free_unref_page_list, but somehow the functions do not work as I intended. (Actually the functions are not called so I may choose the wrong functions.) So, I was wondering whether I intercepted the right functions (three functions above). Also, I have a question about freeing a page-table page. 1. In x86, are all page-table pages released at the end of termination by using free_pgtables()? 2. In x86, a page-table page can be freed when there is no entry? In other words, does the Linux kernel release a page-table page on runtime? If so, what function would do this kind of task? I'm trying to find freeing functions, but I can't find them and I'm not sure when this kind of function is called. (I'm pretty sure that pte_free / pmd_free / pud_free functions are not for this case.) Any help would be greatly appreciated. Regards, Wonkyo

3 3

pid start time and new display field in proc pid stat
by Navin P April 3, 2021

April 3, 2021

Hi, As of 5.11 kernel (pid,pid_start_time) is not unique /monotonic even though the underlying counters are . I chose start_boottime because i wanted the counter to increase during suspend as well. 1. Is there any case where task->start_boottime or ktime_get_boottime_ns doesn't become monotonic i.e increasing ? 2. If start_boottime is not monotonic which counter to use ? 3. If i create a new field in task_struct , then i can use a atomic_add_return(1,&v) to fill in the task->new_field. Will this also work ? By doing this <pid,pid_start_time> becomes unique. In linux/fs/proc/array.c at line 566 we have /* apply timens offset for boottime and convert nsec -> ticks */ start_time = nsec_to_clock_t(timens_add_boottime_ns(task->start_boottime)); task->start_boottime is a monotonic increasing counter fetched from ktime_get_boottime_ns in fork.c nsec_to_clock_t contains div_u64 due to which we loose some lower bits/digits on divison and is not unique unless the divisor is 1. if CONFIG_HZ = 250 and nsec_to_clock_t x=4000001 , then #if (NSEC_PER_SEC % USER_HZ) == 0 return div_u64(x, NSEC_PER_SEC / USER_HZ); becomes div_u64(x, 4000000) then return value is 1 when x=4000002, return value is 1 until x=8000000 which returns 2. The value shown in /proc/[pid]/stat is actually the truncated value. Hence i'm planning to display a counter at the end of /proc/[pid]/stat as the 53rd field. I've prepared a patch as inlined. >From a2c6b5d6435394f015d38700008ff74f16dfa5fd Mon Sep 17 00:00:00 2001 From: Navin P <navinp0304(a)gmail.com> Date: Thu, 25 Mar 2021 15:27:30 +0530 Subject: [PATCH] Display task->start_boottime as 53rd field in /proc/[pid]/stat. The 22nd field start_time currently shown in /proc/[pid]/stat as start_time is truncated by division.Hence it is not unique . Signed-off-by: Navin P <navinp0304(a)gmail.com> --- Documentation/filesystems/proc.rst | 1 + fs/proc/array.c | 1 + 2 files changed, 2 insertions(+) diff --git a/Documentation/filesystems/proc.rst b/Documentation/filesystems/proc.rst index 48fbfc336ebf..3b7a1543b2c0 100644 --- a/Documentation/filesystems/proc.rst +++ b/Documentation/filesystems/proc.rst @@ -381,6 +381,7 @@ It's slow but very precise. env_end address below which program environment is placed exit_code the thread's exit_code in the form reported by the waitpid system call + start_boottime the process start time in nanoseconds since boot ============= =============================================================== The /proc/PID/maps file contains the currently mapped memory regions and diff --git a/fs/proc/array.c b/fs/proc/array.c index bb87e4d89cd8..74389aaefa9c 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -645,6 +645,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, else seq_puts(m, " 0"); + seq_put_decimal_ull(m, " ", task->start_boottime); seq_putc(m, '\n'); if (mm) mmput(mm); -- 2.25.1

3 5

Unable to boot to kernel built from stable 5.11.y
by Andrew Adamson April 2, 2021

April 2, 2021

Hi Everyone, I tried to build the kernel myself for the first a few days ago. While the build was successful, I am unable to boot to it. After grub I get a blinking cursor for 15 to 30 minutes, and then a message that says "Please reboot and select proper boot device." Since ubuntu ships with a 5.4 kernel, I copied the config from /boot and used make olddefconfig. I also checked the menu entries in grub and my 5.11.y entry looks identical to my 5.4.y entry. This feels like a config issue with how ubuntu configures the boot process, but I am not sure how to investigate. I checked the ubuntu wiki for advice on kernel building, and noticed they recommend using make deb-pkg instead of make all. I cannot get that make recipe to work, it throws errors from dpkg-source that I'm not sure how to resolve """ dpkg-source: error: unrepresentable changes to source dpkg-buildpackage: error: dpkg-source -i.git -b . subprocess returned exit status 1 """ Does anyone have any recommendations? Thanks in advance -Andrew

3 2

memory order seq cst for atomics
by Navin P April 2, 2021

April 2, 2021

Hi, C11 and c++11 have memory order like memory_order_seq_cst .What is the equivalent in Linux kernel ? atomic64_inc(&var) ; __SMP_mb(); Regards, Navin

2 1

Status of a kernel.org hosted project, mmsd?
by Chris Talbot April 1, 2021

April 1, 2021

Hello, I am reaching out here in the hopes of finding out the status of a kernel.org hosted project, mmsd. Bottom line, I think it is abandoned. I think this based on on the fact that despite attempting to contact the maintainers several times in the past few months over several forms of communication, I never got a reply. Additionally, it seems to have not been worked on in quite some time. I am trying to find out how I can submit patches for it, and if it makes sense to find a new maintainer for it. Upstream is here: https://git.kernel.org/pub/scm/network/ofono/mmsd.git I have attempted to contact the ofono maintainers via IRC in December 2020, January 2021, and February 2021 in regards to submitting patches. I have yet to recieve a reply. I also tried to submit patches for mmsd on the Ofono Mailing List starting in February of 2021: https://lists.ofono.org/hyperkitty/list/ofono@ofono.org/thread/HFGZCER3I6G5… https://lists.ofono.org/hyperkitty/list/ofono@ofono.org/thread/CVOWCDC7H4G4… https://lists.ofono.org/hyperkitty/list/ofono@ofono.org/thread/C47Q5UY2NVHQ… None of my emails got a reply. At the advice of some other devs, I have additionally tried emailling the Linux-netdev group mailing list, the authors listed on mmsd, and the folks who have contributed in the past, but that also did not recieve a reply. If you are curious to see the status of my work, it is here: https://source.puri.sm/kop316/mmsd/ Thank you for your time! Respectfully, Chris Talbot

2 1