| Summary: | DAHLIN-00249: floating point issues at startup | ||||
| Reporter: | Tzafrir Cohen (tzafrir) | Labels: | |||
| Date Opened: | 2011-08-11 03:52:50 | Date Closed: | 2012-04-05 00:38:40 | ||
| Priority: | Minor | Regression? | No | ||
| Status: | Closed/Complete | Components: | |||
| Versions: | 2.4.0 2.5.0 | Frequency of Occurrence | |||
| Related Issues: |
| ||||
| Environment: | Centos5 | Attachments: | ( 0) fp-test.c | ||
| Description: | The attached fp-test.c is a program that runs some floating-point calculations in a loop, and complains when they fail. They should not fail. On some systems when we run it at the time of loading or unloading DAHDI, we occasionally get errors. Specifically: a while after registering a span. Maybe also at the time of loading a module. Not sure exactly. No calls are running. The echo canceler selected is MG2. The system in which we see it has 4 cores, each: processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 23 model name : Intel(R) Core(TM)2 Quad CPU Q9550 @ 2.83GHz stepping : 10 cpu MHz : 2833.411 cache size : 6144 KB physical id : 0 siblings : 4 core id : 0 cpu cores : 4 apicid : 0 fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc pni monitor ds_cpl vmx smx est tm2 cx16 xtpr lahf_lm bogomips : 5666.82 We have not yet tested versions earlier than 2.4.0. | ||||
| Comments: | By: Shaun Ruffell (sruffell) 2011-08-11 12:01:57.843-0500 Is this with CONFIG_DAHDI_MMX and/or ECHO_CAN_FP set? And if so, are you running in 64-bit mode? By: Tzafrir Cohen (tzafrir) 2011-08-11 12:16:18.140-0500 I was completely sure it had nothing to do with those. But after trying to reproduce those again, it might as well be the remains of curreptions from a previously loaded kernel (I was sure I booted to a clean system before submitting this report). I'll leave this open for a while, and close it if I can't reproduce the issue. By: Shaun Ruffell (sruffell) 2012-04-04 13:03:22.333-0500 Any reason to keep this open? | ||||