[Mesa-dev] [Bug 87913] CPU cacheline size of 0 can be returned by CPUID leaf 0x80000006 in some virtual machines

2015-01-05 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=87913 Roland Scheidegger srol...@vmware.com changed: What|Removed |Added Status|NEW |RESOLVED

[Mesa-dev] [Bug 87913] CPU cacheline size of 0 can be returned by CPUID leaf 0x80000006 in some virtual machines

2015-01-01 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=87913 --- Comment #4 from Leonid Shatz leonid.sh...@ravellosystems.com --- In virtual environments you would never know which is the safest method. It depends on the implementation and correctness of the hypervisor. I would just add the suggested fix

[Mesa-dev] [Bug 87913] CPU cacheline size of 0 can be returned by CPUID leaf 0x80000006 in some virtual machines

2014-12-31 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=87913 Bug ID: 87913 Summary: CPU cacheline size of 0 can be returned by CPUID leaf 0x8006 in some virtual machines Product: Mesa Version: unspecified Hardware: x86-64

[Mesa-dev] [Bug 87913] CPU cacheline size of 0 can be returned by CPUID leaf 0x80000006 in some virtual machines

2014-12-31 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=87913 Leonid Shatz leonid.sh...@ravellosystems.com changed: What|Removed |Added CC|

[Mesa-dev] [Bug 87913] CPU cacheline size of 0 can be returned by CPUID leaf 0x80000006 in some virtual machines

2014-12-31 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=87913 --- Comment #2 from Leonid Shatz leonid.sh...@ravellosystems.com --- Sorry, the correct code would be something like: if (regs[0] = 0x8006) { cpuid(0x8006, regs2); + if ((regs2[2] 0xFF) 0)

[Mesa-dev] [Bug 87913] CPU cacheline size of 0 can be returned by CPUID leaf 0x80000006 in some virtual machines

2014-12-31 Thread bugzilla-daemon
https://bugs.freedesktop.org/show_bug.cgi?id=87913 --- Comment #3 from Roland Scheidegger srol...@vmware.com --- I'm wondering if we actually should use this detection method if the other one (which queries clflush size in fact) is also available. I guess though there's some (amd) cpus where the