Linux 4.14: Reported regressions as of Sunday, 2017-10-15

Hi! Find below my third regression report for Linux 4.14. It lists 9 regressions I'm currently aware of. Two regressions got fixed since last weeks report. As always: Are you aware of any other regressions? Then please let me know by mail (a simple bounce or forward in my direction is enough!).

[PATCH 2/2] powerpc/perf/imc: use NUMA_NO_NODE for alloc_pages_node

alloc_pages_node() when passed NUMA_NO_NODE for the node_id, could get memory from closest node. Cleanup core imc and thread imc memory init functions to use NUMA_NO_NODE. Signed-off-by: Madhavan Srinivasan --- arch/powerpc/perf/imc-pmu.c | 8 +++- 1 file changed,

[PATCH 1/2] powerpc/perf/imc: use cpu_to_node instead topology_physical_package_id

... [1069001518.00] [c03f95b3f770] [c00b2574] init_imc_pmu+0x1f4/0xc40 [1069005374.00] [c03f95b3f850] [c008fec8] opal_imc_counters_probe+0x2e8/0x3e0 [1069009426.00] [c03f95b3f950] [c06153a4] platform_drv_probe+0x44/0x90 [1069012818.00]

Re: [PATCH 1/2] powerpc/strict_rwx: quirks for STRICT_RWX patches

On Wed, 04 Oct 2017 20:04:52 +1100 Michael Ellerman wrote: > Hi Balbir, > > Mainly I think we need to add a check in mark_initmem_nx() too don't we? > Or should we put it somewhere common to both? > > But seeing as I'm replying here are some more comments. > > > Subject:

Re: Linux 4.14: Reported regressions as of Sunday, 2017-10-15

On Sun, Oct 15, 2017 at 02:55:24PM +0200, Thorsten Leemhuis wrote: > == Fixed since last report == > > "hangs when building e.g. perf" & "Random insta-reboots on AMD Phenom II" > Status: Fixed by https://git.kernel.org/torvalds/c/67bb8e999e0a That should be: b956575bed91 ("x86/mm: Flush more

Re: [PATCH 2/2] powerpc/strict_rwx: fixup region splitting

On Wed, 04 Oct 2017 22:14:17 +1100 Michael Ellerman wrote: > Balbir Singh writes: > > > We were aggressive with splitting regions and missed the case > > when _stext and __init_begin completely overlap addr and addr+mapping. > > > > This patch fixes

Re: [PATCH] powerpc/powernv: Enable reset_devices parameter to issue a PHB reset

On 10/14/2017 06:13 AM, Benjamin Herrenschmidt wrote: > No, he's saying this is useful for the developers when debugging the > kernel driver (or for asking users to "test" something as part of > debugging a driver problem). > > It's common to have various command line options affecting PCIe >

Re: [PATCH 2/2] powerpc/perf/imc: use NUMA_NO_NODE for alloc_pages_node

On Mon, 16 Oct 2017 00:13:42 +0530 Madhavan Srinivasan wrote: > alloc_pages_node() when passed NUMA_NO_NODE for the > node_id, could get memory from closest node. Cleanup > core imc and thread imc memory init functions to use > NUMA_NO_NODE. The changelog is not

Re: [PATCH v3] powerpc/xmon: Support dumping software pagetables

On Thu, 12 Oct 2017 18:20:39 +0800 kbuild test robot wrote: > Hi Balbir, > > [auto build test ERROR on powerpc/next] > [also build test ERROR on v4.14-rc4 next-20171009] > [if your patch is applied to the wrong git tree, please drop us a note to > help improve the system] > >

Re: [PATCH v3] powerpc/xmon: Support dumping software pagetables

Michael Ellerman writes: > From: Balbir Singh > > It would be nice to be able to dump page tables in a particular > context. > > eg: dumping vmalloc space: > > 0:mon> dv 0xd00037f0 > pgd @ 0xc17c > pgdp @ 0xc17c00d8

[PATCH v2] powerpc/64s: Fix crashes on Power9 DD1 with radix MMU and STRICT_RWX

When using the radix MMU on Power9 DD1, to work around a hardware problem, radix__pte_update() is required to do a two stage update of the PTE. First we write a zero value into the PTE, then we flush the TLB, and then we write the new PTE value. In the normal case that works OK, but it does not

Re: [PATCH v2] powerpc/xmon: Support dumping software pagetables

On Mon, Oct 16, 2017 at 2:33 PM, Balbir Singh wrote: > It would be nice to be able to dump page tables in a particular > context. > Should be v4 and not v2.. resending Balbir

[PATCH] selftests/powerpc: Add alignment handler selftest

Add a selftest to exercise the powerpc alignment fault handler. Signed-off-by: Michael Neuling Signed-off-by: Andrew Donnellan --- tools/testing/selftests/powerpc/alignment/Makefile | 3 +- .../powerpc/alignment/alignment_handler.c |

Re: [PATCH v4 2/5] powerpc/mce: align the print of physical address better

Balbir Singh writes: > Use the same alignment as Effective address > and rename phyiscal address to Page Frame Number You didn't do the 2nd part AFAICS? Will fix it up. cheers > diff --git a/arch/powerpc/kernel/mce.c b/arch/powerpc/kernel/mce.c > index e254399..fef1408

Re: [PATCH v3 3/3] powerpc:selftest update memcmp_64 selftest for VMX implementation

On Fri, 2017-10-13 at 12:30 +0800, wei.guo.si...@gmail.com wrote: > From: Simon Guo > > This patch adjust selftest memcmp_64 so that memcmp selftest can be > compiled successfully. > Do they not compile at the moment? > It also adds testcases for: > - memcmp over 4K

[PATCH v2] powerpc/xmon: Support dumping software pagetables

It would be nice to be able to dump page tables in a particular context. eg: dumping vmalloc space: 0:mon> dv 0xd00037f0 pgd @ 0xc17c pgdp @ 0xc17c00d8 = 0xf10b1000 pudp @ 0xc000f10b13f8 = 0xf10d pmdp @ 0xc000f10d1ff8 =

Re: [PATCH v4 1/5] powerpc/mce.c: Remove unused function get_mce_fault_addr()

On Mon, Oct 16, 2017 at 4:13 PM, Michael Ellerman wrote: > Balbir Singh writes: > >> There are no users of get_mce_fault_addr() >> >> Fixes: b63a0ff ("powerpc/powernv: Machine check exception handling.") > > That fixes line is wrong,

Re: [PATCH v4 5/5] powerpc/mce: hookup memory_failure for UE errors

Balbir Singh writes: > If we are in user space and hit a UE error, we now have the > basic infrastructure to walk the page tables and find out > the effective address that was accessed, since the DAR > is not valid. > > We use a work_queue content to hookup the bad pfn,

Re: [PATCH v4 3/5] powerpc/mce: Hookup derror (load/store) UE errors

On Mon, Oct 16, 2017 at 4:36 PM, Michael Ellerman wrote: > Balbir Singh writes: > >> diff --git a/arch/powerpc/kernel/mce_power.c >> b/arch/powerpc/kernel/mce_power.c >> index b76ca19..0e584d5 100644 >> --- a/arch/powerpc/kernel/mce_power.c >> +++

[PATCH kernel] RFC: prom_init: Fetch flatten device tree from the system firmware

At the moment, on 256CPU + 256 PCI devices guest, it takes the guest about 8.5sec to read the entire device tree. Some explanation can be found here: https://patchwork.ozlabs.org/patch/826124/ but mostly it is because the kernel traverses the tree twice and it calls "getprop" for each properly

[PATCH v6] powerpc/vdso64: Add support for CLOCK_{REALTIME/MONOTONIC}_COARSE

Current vDSO64 implementation does not have support for coarse clocks (CLOCK_MONOTONIC_COARSE, CLOCK_REALTIME_COARSE), for which it falls back to system call, increasing the response time, vDSO implementation reduces the cycle time. Below is a benchmark of the difference in execution times.

[PATCH v4] powerpc/xmon: Support dumping software pagetables

It would be nice to be able to dump page tables in a particular context. eg: dumping vmalloc space: 0:mon> dv 0xd00037f0 pgd @ 0xc17c pgdp @ 0xc17c00d8 = 0xf10b1000 pudp @ 0xc000f10b13f8 = 0xf10d pmdp @ 0xc000f10d1ff8 =

Re: [PATCH v3] powerpc/xmon: Support dumping software pagetables

On 10/16/2017 10:43 AM, Balbir Singh wrote: On Mon, Oct 16, 2017 at 2:34 PM, Aneesh Kumar K.V wrote: Michael Ellerman writes: + +#ifdef CONFIG_HUGETLB_PAGE + if (pud_huge(*pudp)) { + format_pte(pudp, pud_val(*pudp)); +

Re: [PATCH v4 5/5] powerpc/mce: hookup memory_failure for UE errors

On Mon, Oct 16, 2017 at 4:38 PM, Michael Ellerman wrote: > Balbir Singh writes: > >> If we are in user space and hit a UE error, we now have the >> basic infrastructure to walk the page tables and find out >> the effective address that was accessed,

Re: [PATCH v3] powerpc/xmon: Support dumping software pagetables

On Mon, Oct 16, 2017 at 2:34 PM, Aneesh Kumar K.V wrote: > Michael Ellerman writes: > >> From: Balbir Singh >> >> It would be nice to be able to dump page tables in a particular >> context. >> >> eg: dumping vmalloc

Re: [PATCH v4 1/5] powerpc/mce.c: Remove unused function get_mce_fault_addr()

Balbir Singh writes: > There are no users of get_mce_fault_addr() > > Fixes: b63a0ff ("powerpc/powernv: Machine check exception handling.") That fixes line is wrong, get_mce_fault_addr() was used in that commit. The last usage was removed in: 1363875bdb63

Re: [PATCH v4 2/5] powerpc/mce: align the print of physical address better

On Mon, Oct 16, 2017 at 4:18 PM, Michael Ellerman wrote: > Balbir Singh writes: > >> Use the same alignment as Effective address >> and rename phyiscal address to Page Frame Number > > You didn't do the 2nd part AFAICS? > > Will fix it up. Sorry the

Re: [PATCH v4 3/5] powerpc/mce: Hookup derror (load/store) UE errors

Balbir Singh writes: > diff --git a/arch/powerpc/kernel/mce_power.c b/arch/powerpc/kernel/mce_power.c > index b76ca19..0e584d5 100644 > --- a/arch/powerpc/kernel/mce_power.c > +++ b/arch/powerpc/kernel/mce_power.c > @@ -27,6 +27,36 @@ > #include > #include > #include