July 2020 - kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

Are there some potentially serious problems that I should be aware of if I totally disable the CONFIG_ACPI option on the X86_64 platform?
by 孙世龙 sunshilong July 19, 2020

July 19, 2020

Hi, list Are there some potentially serious problems that I should be aware of if I totally disable the CONFIG_ACPI option on the X86_64 platform? Would it do harm to the hardware? Thank you for your attention to this matter.

3 9

Old Driver Code Using Streams
by Steve Boley July 19, 2020

July 19, 2020

I have old source for a proprietary card that used to dev for both unix and linux and uses the old sysV streams code for talking to the native operating system. The streams is long out of development and trying to port it to working CentOS 6 2.6.32 is proving less than fun. I'd like to know what we need to use within the native linux kernel to ditch the streams package altogether and just rewrite both ends. Thanks -- Steve Boley Senior Linux Administrator IT Manager Carrier Management Systems Inc. Abilene TX 79601 USA ~gnuu. O uGNUGNu o @g~nu~~gNu G/@||@) N| @,----.GN| GN \__/ gNU gNU gNUu gNU gNUu gNU gNUu GNU GNUU GnU GNUU __| \. |\gNugNu | `. | `/ \Gn _) \.___.,| .' \____ )GNUGNU| .' `-' `--'

2 1

Help troubleshooting memory leak when highres=on
by William Tambe July 17, 2020

July 17, 2020

I am doing porting work; I am having a memory leak issue in a kernel build with CONFIG_HIGH_RES_TIMERS=y. When highres=on, load average as returned by uptime are always null, and KernelStack from /proc/meminfo steadily grows until the system runs out of memory triggering oom-killer and eventually the whole system hangs. Above memory leak and load average issues are not seen when booting the kernel using highres=off. Any idea where I should start debugging these issues ?

1 0

How to comprehend this code snippet: __asm__ __volatile__("rdtsc" : "=A"(t))?
by 孙世龙 sunshilong July 16, 2020

July 16, 2020

Hi, list Here is the code snippet: #define ipipe_read_tsc(t) \ __asm__ __volatile__("rdtsc" : "=A"(t)) I found that the rdtsc (Read Time-Stamp Counter) instruction is used to determine how many CPU ticks took place since the processor was reset. But what does "=A"(t) mean? Thank you for your attention to this matter.

2 3

Why the trigered counts of NMIs is equivalent to PMI's?
by 孙世龙 sunshilong July 16, 2020

July 16, 2020

As the subject, here is the statics got from my PC at a different time: $cat /proc/interrupts | grep -E 'NMI|PMI' NMI: 649 374 347 424 Non-maskable interrupts PMI: 649 374 347 424 Performance monitoring interrupts $ cat /proc/interrupts | grep -Ei 'PMI|NMI' NMI: 247 274 238 289 Non-maskable interrupts PMI: 247 274 238 289 Performance monitoring interrupts

1 0

When returning from schedule_tail() ...
by William Tambe July 16, 2020

July 16, 2020

When returning from schedule_tail(), is it still necessary to check for (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME) before resuming task ?

2 2

Re: [PATCH] mmc: cqhci: Fix a print format for the task descriptor
by Joe Perches July 15, 2020

July 15, 2020

On Wed, 2020-07-15 at 15:33 +0800, Ben Chuang wrote: > The format string of the task descriptor should be "%016llx". There are a lot of these: $ git grep -P -n '\b0x\d+%' arch/m68k/include/asm/mmu_context.h:166: pr_info("ksp load failed: mm=0x%p ksp=0x08%lx\n", mm, mmuar); arch/powerpc/kvm/book3s_xive.c:1706: pr_devel(" val=0x016%llx (server=0x%x, guest_prio=%d)\n", drivers/atm/eni.c:590: EVENT("bad PDU (descr=0x08%lx,length=%ld)\n",descr, drivers/dma/ppc4xx/adma.c:4366: "uses 0x1%02x polynomial.\n", reg); drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c:3023: DRM_INFO("Using rlc debug ucode. mmRLC_GPM_GENERAL_6 ==0x08%x / fw_ver == %i \n", drivers/gpu/drm/i915/intel_uncore.c:495: drm_dbg(&uncore->i915->drm, "GTFIFODBG = 0x08%x\n", fifodbg); drivers/media/usb/gspca/ov519.c:2760: gspca_err(gspca_dev, "Unknown sensor: 0x76%02x\n", drivers/mmc/host/cqhci.c:425: pr_debug("%s: cqhci: tag %d task descriptor 0x016%llx\n", drivers/mmc/host/s3cmci.c:192: "#%u%s op:%i arg:0x%08x flags:0x08%x retries:%u", drivers/net/can/flexcan.c:1517: "gpr %s req_gpr=0x02%x req_bit=%u ack_gpr=0x02%x ack_bit=%u\n", drivers/net/ethernet/intel/i40e/i40e_debugfs.c:444: " info: cmd_reserved[] = 0x%02x 0x%02x 0x%02x 0x0%02x 0x%02x 0x%02x 0x%02x 0x0%02x\n", drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c:1397: "i = %d buffer_info->dma = 0x08%llx buffer_info->length = 0x%x\n", drivers/net/wireless/intel/iwlwifi/pcie/trans.c:2298: "Queue %d is %sactive on fifo %d and stuck for %u ms. SW [%d, %d] HW [%d, %d] FH TRB=0x0%x\n", drivers/net/wireless/intersil/hostap/hostap_hw.c:779: "(%d,0x04%x,%d); reg=0x%04x\n", drivers/net/wireless/realtek/rtw88/rtw8822c.c:429: rtw_dbg(rtwdev, RTW_DBG_RFK, "[DACK] ADCK 0x%08x=0x08%x\n", drivers/scsi/mpt3sas/mpt3sas_scsih.c:6161: ioc_err(ioc, "device is not present handle(0x04%x)!!!\n", drivers/scsi/mpt3sas/mpt3sas_scsih.c:6923: ioc_err(ioc, "device is not present handle(0x04%x)!!!\n", drivers/staging/rtl8723bs/hal/hal_com_phycfg.c:328: DBG_871X("Invalid RegAddr 0x3%x in PHY_GetRateSectionIndexOfTxPowerByRate()", RegAddr); drivers/usb/serial/spcp8x5.c:225: dev_dbg(&port->dev, "0xc0:0x22:0:6 %d - 0x02%x\n", ret, *buf); drivers/video/fbdev/s1d13xxxfb.c:500: dbg_blit("(copyarea) bpp=%d=0x0%d, mem_offset1=%d, mem_offset2=%d\n", bpp, (bpp >> 1), sound/pci/es1938.c:306: "snd_es1938_write_cmd timeout (0x02%x/0x02%x)\n", cmd, v); sound/pci/es1938.c:319: dev_err(chip->card->dev, "get_byte timeout: status 0x02%x\n", v);

1 0

What's tha KOBJ_MAX action used for?
by Garrit Franke July 14, 2020

July 14, 2020

Hi all, I was wondering, what the KOBJ_MAX action in the kobject.h file does. It is self explainatory what KOBJ_ADD, KOBJ_REMOVE etc. are used for, but KOBJ_MAX puzzles me. On top of that, it is never referenced somewhere in the sourcecode (https://elixir.bootlin.com/linux/v5.7.8/C/ident/KOBJ_MAX) Can it maybe even be removed? Regards, Garrit Franke

3 4

Re: Turn private hugetlb mapping into VM_SHARED
by heep July 14, 2020

July 14, 2020

> Hello, list, > > I am working on a tool to introspect live memory of KVM machines, and mmapping its memory, instead of calling process_vm_readv on every read, would increase its performance significantly. Understandably, almost all virtual machine monitors (like QEMU) create private mappings for the memory. So, trying to understand the memory managment subsystem I am hoping to write a kernel module to do the remappings, and conversion from private, to shared mappings. > > My current focus is on hugetlb pages. At first, I remapped the pages using remap_pfn_range, but there is a memory safety risk when the VM shuts off, and that memory gets used elsewhere. Now, I am trying to turn the underlying mapping into a shared one, so that the kernel can keep track of the memory used. What would be the best way to do so? > > I noticed there is resv_map that stores represented mappings, and while on private mappings it's on vma, on the shared ones - it is in the inode, but making the inode point to the VMAs map leads to a bus error in userspace. I must be missing some key pieces. > > Does anyone know how to best achieve what I am trying to? Or is it something fundamentally wrong to do? > > Many thanks, > Auri B. It appears I have been mistaken when it came to remap_pfn_range being memory unsafe, I just haven't read get_user_pages_remote documentation attentively enough. Sorry for that, this whole unshared -> shared mapping change is very much not the right thing to do.

1 0

How to understand the comment of macro *compute_indices*?
by carver July 14, 2020

July 14, 2020

Hello, Macro *compute_indices* in arch/arm64/kernel/head.S have a comment as shown below: mov \istart, \ptrs mul \istart, \istart, \count add \iend, \iend, \istart // iend += (count - 1) * ptrs I am wondering whether this comment should be change to "iend += count * ptrs". Or am I misunderstanding this code? Thanks -Lio

2 1