x86, cpu: RDC doesn't have CPUID, which is what c_ident is
in [Kernel]
|
Prev: [PATCH 1/1] staging: hv: Fix missing functions for net_device_ops (with detailed comments)
Next: [PATCH] flex_array: add helpers to get and put to make pointers easy to use
From: Ingo Molnar on 3 Aug 2010 15:30 * tip-bot for H. Peter Anvin <hpa(a)linux.intel.com> wrote: > Commit-ID: c0366f5f5c1615c4f70b1bd646149ba9233830b9 > Gitweb: http://git.kernel.org/tip/c0366f5f5c1615c4f70b1bd646149ba9233830b9 > Author: H. Peter Anvin <hpa(a)linux.intel.com> > AuthorDate: Tue, 3 Aug 2010 10:09:56 -0700 > Committer: H. Peter Anvin <hpa(a)linux.intel.com> > CommitDate: Tue, 3 Aug 2010 10:09:56 -0700 > > x86, cpu: RDC doesn't have CPUID, which is what c_ident is > > c_ident is supposed to be a CPUID identifier, and RDC doesn't have > that. -tip testing found that these RDC patches cause a boot crash if a X86_RDC321X=y kernel is booted on a regular PC: [ 0.015999] Pid: 0, comm: swapper Not tainted 2.6.35-tip+ #23896 A8N-E/System Product Name [ 0.015999] EIP: 0060:[<c1267d14>] EFLAGS: 00010282 CPU: 0 [ 0.015999] EIP is at strcmp+0xf/0x22 [ 0.015999] EAX: c1a81541 EBX: 00000004 ECX: 0000000f EDX: 00000000 [ 0.015999] ESI: c1a81571 EDI: 00000000 EBP: c19ccf78 ESP: c19ccf70 [ 0.015999] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 [ 0.015999] Process swapper (pid: 0, ti=c19cc000 task=c1a07e60 task.ti=c19cc000) [ 0.015999] Stack: [ 0.015999] c1a81530 c1a81570 c19ccf8c c16d9b7f c1ac8538 c1a81530 c1a81570 c19ccfac [ 0.015999] <0> c16d9dcb c104dc9f c1a15210 c1a81548 c1ac8538 00099800 c19ce000 c19ccfb4 [ 0.015999] <0> c1a89a4f c19ccfc0 c1a89d2d c1a96147 c19ccfe0 c1a84953 431d9715 00000006 [ 0.015999] Call Trace: [ 0.015999] [<c16d9b7f>] ? get_cpu_vendor+0x22/0x8f [ 0.015999] [<c16d9dcb>] ? identify_cpu+0xc5/0x2e0 [ 0.015999] [<c104dc9f>] ? sub_preempt_count+0x9/0x39 [ 0.015999] [<c1a89a4f>] ? identify_boot_cpu+0xd/0x23 [ 0.015999] [<c1a89d2d>] ? check_bugs+0xb/0xaf [ 0.015999] [<c1a96147>] ? cpuset_init+0x65/0x77 [ 0.015999] [<c1a84953>] ? start_kernel+0x37e/0x392 [ 0.015999] [<c1a840c9>] ? i386_start_kernel+0xc9/0xd0 [ 0.015999] Code: c0 83 c9 ff f2 ae 4f 89 d1 49 78 06 ac aa 84 c0 75 f7 31 c0 aa 89 d8 5b 5e 5 [ 0.015999] EIP: [<c1267d14>] strcmp+0xf/0x22 SS:ESP 0068:c19ccf70 mercury:~> cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 15 model : 35 model name : AMD Athlon(tm) 64 X2 Dual Core Processor 3800+ stepping : 2 cpu MHz : 2010.301 cache size : 512 KB physical id : 0 siblings : 2 core id : 0 cpu cores : 2 fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt lm 3dnowext 3dnow rep_good pni lahf_lm cmp_legacy bogomips : 4021.85 TLB size : 1024 4K pages clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management: ts fid vid ttp config attached. Thanks, Ingo
From: Ingo Molnar on 4 Aug 2010 05:30 * bifferos <bifferos(a)yahoo.co.uk> wrote: > I couldn't reproduce this with 2.6.35 (without Peter's change), > I was trying on VirtualBox. If the detection code is causing a > problem we can just skip it if the CPU has cpuid. As i reported: > > -tip testing found that these RDC patches cause a boot > > crash if a X86_RDC321X=y kernel is booted on a regular PC: Those two patches caused the crash, with the config i sent, on a regular PC. Ingo -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: H. Peter Anvin on 4 Aug 2010 16:20 On 08/04/2010 02:12 AM, bifferos wrote: > > I couldn't reproduce this with 2.6.35 (without Peter's change), > I was trying on VirtualBox. If the detection code is causing a > problem we can just skip it if the CPU has cpuid. > > Incidentally if you need some hardware to test the RDC port on I can > arrange that. > Incidentally, please do note that there is no concrete benefit to this "support", since all it does is replicate information in /proc/cpuinfo that is already available through lspci. -hpa -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: H. Peter Anvin on 4 Aug 2010 19:20 On 08/04/2010 03:59 PM, bifferos wrote: > --- On Wed, 4/8/10, H. Peter Anvin <hpa(a)linux.intel.com> wrote: > >> >> Incidentally, please do note that there is no concrete >> benefit to this >> "support", since all it does is replicate information in >> /proc/cpuinfo >> that is already available through lspci. > > Without the patch RDC is misreported as Cyrix Cx486SLC. > And what problems does that cause? Presumably none, in which case you have to realize that the correctness threshold for this patch is enormous, as the benefit is extremely small compared with the potential (and actual!) harm. -hpa -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: H. Peter Anvin on 4 Aug 2010 22:10
On 08/04/2010 05:27 PM, bifferos wrote: > > --- On Thu, 5/8/10, H. Peter Anvin <hpa(a)linux.intel.com> wrote: >>> Without the patch RDC is misreported as Cyrix >> Cx486SLC. >>> >> >> And what problems does that cause? Presumably none, > > If I don't know the CPU how do I set the tick rate? > > http://android.git.kernel.org/?p=kernel/common.git;a=commitdiff_plain;h=e1b4d1143651fb3838be1117785b6e0386fa151f > That's a valid use, of course, but that wasn't sent with it. Even so, it's really a chipset property and not a CPU property and probably should simply be set via a PCI quirk. So, clean up the patch, add the usage, and then it can be put into the -tip tree for the next merge window. -hpa -- H. Peter Anvin, Intel Open Source Technology Center I work for Intel. I don't speak on their behalf. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |