On 24 November 2017 at 16:42, Greg KH greg@kroah.com wrote:

On Fri, Nov 24, 2017 at 12:11:41PM +0100, Greg KH wrote:

...

On Fri, Nov 24, 2017 at 04:32:42PM +0530, Naresh Kamboju wrote:

...

Hi,

We have found "no MPX support" bug while running kselftests on SuperServer 5019S-ML - x86_64 device. Please provide your comments to fix this bug.

LKFT: 4.4-rc 4.9-rc 4.13-rc 4.14-rc: x86: kselftest mpx-mini-test_64 - no MPX support - failed - 3869 Aborted (core dumped) (edit) https://bugs.linaro.org/show_bug.cgi?id=3497

NOTE: Please add your self to Bugzilla to add comments.

Ugh, what? Why not discuss it here, and provide the full log of what is happening and what you think should be done to fix it. Don't bury stuff in some random bugzilla...

Perhaps a simple abort once you determine that MPX is not supported?

Wait, that is what the test is doing, so it's working properly. Why is this an issue?

seems to be that the hardware change in LAB brought up this failures Here you could notice difference in test out on old and new machines.

old machine: ------------ processor lacks XSAVE, can not run MPX tests selftests: mpx-mini-test_64 [PASS]

New machine: ------------- XSAVE is supported by HW & OS XSAVE processor supported state mask: 0x1f XSAVE OS supported state mask: 0x1f BNDREGS: size: 64 user: 1 supervisor: 0 aligned: 0 BNDCSR: size: 64 user: 1 supervisor: 0 aligned: 0 no MPX support ./run_kselftest.sh: line 214: 3869 Aborted (core dumped) ./mpx-mini-test_64 selftests: mpx-mini-test_64 [FAIL]

code snippet: -------------- bool process_specific_init(void) { <> if (prctl(43, 0, 0, 0, 0)) { printf("no MPX support\n"); abort(); return false; } return true; }

bool process_specific_finish(void) { if (prctl(44)) { printf("no MPX support\n"); return false; } return true; }

Source code: https://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest.git/tr...

Additional information,

1) The processor have Memory Protection Extensions (MPX) capability. boot log, <> [ 0.000000] x86/fpu: Supporting XSAVE feature 0x001: 'x87 floating point registers' [ 0.000000] x86/fpu: Supporting XSAVE feature 0x002: 'SSE registers' [ 0.000000] x86/fpu: Supporting XSAVE feature 0x004: 'AVX registers' [ 0.000000] x86/fpu: Supporting XSAVE feature 0x008: 'MPX bounds registers' [ 0.000000] x86/fpu: Supporting XSAVE feature 0x010: 'MPX CSR' <>

2) The processor flags shows MPX capability # cat /proc/cpuinfo | grep ' mpx '

flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb intel_pt tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx rdseed adx smap clflushopt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp

3) For MPX to work, changes are required in the kernel, binutils and compiler. No source changes are required for applications, just a recompile.

4) GCC with -fmpx: * enables annotation of code with MPX instructions and prefixes

Ref: https://www.kernel.org/doc/Documentation/x86/intel_mpx.txt

- Naresh

greg k-h

-- To unsubscribe from this list: send the line "unsubscribe linux-kselftest" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

+ x86 + tglx

On 24 November 2017 at 17:16, Greg KH greg@kroah.com wrote:

On Fri, Nov 24, 2017 at 04:52:22PM +0530, Naresh Kamboju wrote:

...

On 24 November 2017 at 16:42, Greg KH greg@kroah.com wrote:

...

On Fri, Nov 24, 2017 at 12:11:41PM +0100, Greg KH wrote:

...

On Fri, Nov 24, 2017 at 04:32:42PM +0530, Naresh Kamboju wrote:

...

Hi,

We have found "no MPX support" bug while running kselftests on SuperServer 5019S-ML - x86_64 device. Please provide your comments to fix this bug.

LKFT: 4.4-rc 4.9-rc 4.13-rc 4.14-rc: x86: kselftest mpx-mini-test_64 - no MPX support - failed - 3869 Aborted (core dumped) (edit) https://bugs.linaro.org/show_bug.cgi?id=3497

NOTE: Please add your self to Bugzilla to add comments.

Ugh, what? Why not discuss it here, and provide the full log of what is happening and what you think should be done to fix it. Don't bury stuff in some random bugzilla...

Perhaps a simple abort once you determine that MPX is not supported?

Wait, that is what the test is doing, so it's working properly. Why is this an issue?

seems to be that the hardware change in LAB brought up this failures Here you could notice difference in test out on old and new machines.

old machine:

processor lacks XSAVE, can not run MPX tests selftests: mpx-mini-test_64 [PASS]

New machine:

XSAVE is supported by HW & OS XSAVE processor supported state mask: 0x1f XSAVE OS supported state mask: 0x1f BNDREGS: size: 64 user: 1 supervisor: 0 aligned: 0 BNDCSR: size: 64 user: 1 supervisor: 0 aligned: 0 no MPX support ./run_kselftest.sh: line 214: 3869 Aborted (core dumped) ./mpx-mini-test_64 selftests: mpx-mini-test_64 [FAIL]

Yes, the tests runs differently on different hardware.

So abort() should not happen if there is no MPX support, care to make up a patch and test it on your system?

In the first case, The test case is detecting hardware has MPX feature supported or not. and if it not exit gracefully.

In the second case, Hardware have MPX support. but test got Aborted. The reason for abort to be investigated.

I am suspecting. As a pre-requirement, (For MPX to work, changes are required in the kernel, binutils and compiler.) (GCC with -fmpx: * enables annotation of code with MPX instructions and prefixes) Do the kernel to be re-compiled with -fmpx ?

Test source: tools/testing/selftests/x86/mpx-mini-test.c

Ref: Bugzilla link: https://bugs.linaro.org/show_bug.cgi?id=3497 https://www.kernel.org/doc/Documentation/x86/intel_mpx.txt

Please suggest a way forward to debug and resolve this issue. Thanks in advance.

- Naresh

thanks,

greg k-h

-- To unsubscribe from this list: send the line "unsubscribe linux-kselftest" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html