|
View:
New views
13 Messages
—
Rating Filter:
Alert me
|
| < Prev | 1 - 2 | Next > |
|
|
|
Two replacements CONFIG_SPARSE_IRQ by CONFIG_GENERIC_HARDIRQSin 2.6.30 Andy's patch set allows to build working xenified 2.6.30.1 kernel been tested on Q9550,C2D E8400 boxes. Boris. --- On Fri, 7/10/09, Jan Beulich <JBeulich@...> wrote:
|
I made two replacements CONFIG_SPARSE_IRQ by CONFIG_GENERIC_HARDIRQS. **************************************************** In 60035_xen3-patch-2.6.29.patch1 line 11568 **************************************************** +#ifdef CONFIG_SPARSE_IRQ + irq_to_desc(irq)->kstat_irqs[cpu] = 0; +#else kstat_cpu(cpu).irqs[irq] = 0; **************************************************** In 60036_xen3-patch-2.6.30.patch1 line 18283 **************************************************** +#ifdef CONFIG_SPARSE_IRQ +int __init arch_probe_nr_irqs(void) +{ + int nr; + + if (nr_irqs > (NR_VECTORS * nr_cpu_ids)) + nr_irqs = NR_VECTORS * nr_cpu_ids; To fix this issue. But, i don't know how to rebuild Andy's patch set based on your commit. Boris. P.S. I haven't heard from Andy for a while. --- On Thu, 7/16/09, Kris Shannon <kris@...> wrote:
|
| Pasi, I understand it, but i cannot rebuild Andy's patch set. I am not aware of what the procedure is, but suspect it's not too much complicated. If commit was mentioned, it says to me only that some Git Repo is a source. Nothing else. If you can , please , do it. Boris --- On Fri, 7/17/09, Pasi Kärkkäinen <pasik@...> wrote:
|
| Sorry, i've got your point - just one more *.patch1 file. That's it. Thanks. Boris. --- On Fri, 7/17/09, Pasi Kärkkäinen <pasik@...> wrote:
|
| Fix patch set before applying. Boris. --- On Fri, 7/17/09, Pasi Kärkkäinen <pasik@...> wrote:
|
| Seems to work nice Solaris 10 U7 (64-bit) HVM DomU loads pretty fast with PV drivers at Xen 3.4.1-rc7 Ubuntu 9.04 Dom0 with 2.6.30.1 xenified via rebased patch set V.3 I believe V.2 should be considered as deprecated due to recent message from Jan Beulich :- >>> Boris Derzhavets <bderzhavets@...> 17.07.09 14:57 >>> >Fix patch set before applying. The second hunk is at best bogus: arch_probe_nr_irqs() only gets referenced from a code section that's inside CONFIG_SPARSE_IRQ, and when that option is selected, you'll have a duplicate symbol definition issue between the arch_probe_nr_irqs() in arch/x86/kernel/apic/io_apic.c and the one in drivers/xen/core/evtchn.c. Additionally, simply enabling the function in arch/x86/kernel/apic/io_apic.c does not do, as it may leave you with insufficient space for dynamic IRQs (which is why I cloned it into a properly working function in drivers/xen/core/evtchn.c). Jan --- On Fri, 7/17/09, Andrew Lyon <andrew.lyon@...> wrote:
|
| < Prev | 1 - 2 | Next > |
| Free embeddable forum powered by Nabble | Forum Help |