search for: __kernel_pda

Hi Andrew, This patch series adds to the end of the existing i386-gdt-cleanups patches: allow-per-cpu-variables-to-be-page-aligned.patch i386-gdt-cleanups-use-per-cpu-variables-for-gdt-pda.patch i386-gdt-cleanups-use-per-cpu-gdt-immediately-upon-boot.patch i386-gdt-cleanups-use-per-cpu-gdt-immediately-upon-boot-fix.patch i386-gdt-cleanups-clean-up-cpu_init.patch

Hi Andrew, This patch series adds to the end of the existing i386-gdt-cleanups patches: allow-per-cpu-variables-to-be-page-aligned.patch i386-gdt-cleanups-use-per-cpu-variables-for-gdt-pda.patch i386-gdt-cleanups-use-per-cpu-gdt-immediately-upon-boot.patch i386-gdt-cleanups-use-per-cpu-gdt-immediately-upon-boot-fix.patch i386-gdt-cleanups-clean-up-cpu_init.patch

Hi Andi, This is a series of patches based on your latest queue (as of the other day, at least). It includes: - the most recent patch to compute the appropriate amount of percpu space to allocate, using a separate reservation for modules where needed. - make the percpu sections page-aligned, so that percpu variables can be page aligned if needed (which is used by gdt_page) -

Hi Andi, This is a series of patches based on your latest queue (as of the other day, at least). It includes: - the most recent patch to compute the appropriate amount of percpu space to allocate, using a separate reservation for modules where needed. - make the percpu sections page-aligned, so that percpu variables can be page aligned if needed (which is used by gdt_page) -

...om> diff -r 2ac108843573 arch/i386/kernel/vmi.c --- a/arch/i386/kernel/vmi.c Thu Jan 04 20:01:52 2007 -0800 +++ b/arch/i386/kernel/vmi.c Thu Jan 04 20:02:42 2007 -0800 @@ -533,8 +533,8 @@ vmi_startup_ipi_hook(int phys_apicid, un ap.ds = __USER_DS; ap.es = __USER_DS; - ap.fs = 0; - ap.gs = __KERNEL_PDA; + ap.fs = __KERNEL_PDA; + ap.gs = 0; ap.eflags = 0;

...om> diff -r 2ac108843573 arch/i386/kernel/vmi.c --- a/arch/i386/kernel/vmi.c Thu Jan 04 20:01:52 2007 -0800 +++ b/arch/i386/kernel/vmi.c Thu Jan 04 20:02:42 2007 -0800 @@ -533,8 +533,8 @@ vmi_startup_ipi_hook(int phys_apicid, un ap.ds = __USER_DS; ap.es = __USER_DS; - ap.fs = 0; - ap.gs = __KERNEL_PDA; + ap.fs = __KERNEL_PDA; + ap.gs = 0; ap.eflags = 0;

Hi Jeremy, all, Sorry this took so long, spent last week in Japan at OSDL conf then netconf. After several false starts, I ended up with a very simple implementation, which clashes significantly with your work since then 8(. I've pushed the patches anyway, but it's going to be significant work for me to re-merge them, so I wanted your feedback first. The first patch simply changes

Hi Jeremy, all, Sorry this took so long, spent last week in Japan at OSDL conf then netconf. After several false starts, I ended up with a very simple implementation, which clashes significantly with your work since then 8(. I've pushed the patches anyway, but it's going to be significant work for me to re-merge them, so I wanted your feedback first. The first patch simply changes

Hi Andi, This is a set of updates for the firstfloor patch queue. Quick rundown: revert-mm-x86_64-mm-account-for-module-percpu-space-separately-from-kernel-percpu.patch separate-module-percpu-space.patch Update the module percpu accounting patch fix-ff-allow-percpu-variables-to-be-page-aligned.patch Make sure the percpu memory allocation is page-aligned

Hi Andi, This is a set of updates for the firstfloor patch queue. Quick rundown: revert-mm-x86_64-mm-account-for-module-percpu-space-separately-from-kernel-percpu.patch separate-module-percpu-space.patch Update the module percpu accounting patch fix-ff-allow-percpu-variables-to-be-page-aligned.patch Make sure the percpu memory allocation is page-aligned

...guest_init(void) { @@ -466,10 +465,8 @@ __init void lguest_init(void) /* We use top of mem for initial pagetables. */ init_pg_tables_end = __pa(pg0); - /* set up PDA descriptor */ -// setup_pda(); - load_gdt(&early_gdt_descr); -// asm volatile ("mov %0, %%fs" : : "r" (__KERNEL_PDA) : "memory"); + /* Set up per-cpu descriptor (pointing to master area) */ + asm volatile ("mov %0, %%fs" : : "r" (__KERNEL_DS) : "memory"); reserve_top_address(lguest_data.reserve_mem);

...guest_init(void) { @@ -466,10 +465,8 @@ __init void lguest_init(void) /* We use top of mem for initial pagetables. */ init_pg_tables_end = __pa(pg0); - /* set up PDA descriptor */ -// setup_pda(); - load_gdt(&early_gdt_descr); -// asm volatile ("mov %0, %%fs" : : "r" (__KERNEL_PDA) : "memory"); + /* Set up per-cpu descriptor (pointing to master area) */ + asm volatile ("mov %0, %%fs" : : "r" (__KERNEL_DS) : "memory"); reserve_top_address(lguest_data.reserve_mem);

Seems to work OK for native and Xen. I had to play a bit with the paravirt-sched-clock patch to deal with the VMI changes. Zach, can you check that it still works? Thanks, J

Seems to work OK for native and Xen. I had to play a bit with the paravirt-sched-clock patch to deal with the VMI changes. Zach, can you check that it still works? Thanks, J

Hi, Anyone has this working? Looks like there is a chicken-and-egg issue with pda setup: * xen_load_gdt() uses multicalls, thus depends on cpu-specific variables (per-cpu mc buffer) which in turn requires pda being setup already. * pda setup can't be done before xen_load_gdt() ... Next question while looking at xen_load_gdt(): why does it use multicalls in the first place?

Hi, Anyone has this working? Looks like there is a chicken-and-egg issue with pda setup: * xen_load_gdt() uses multicalls, thus depends on cpu-specific variables (per-cpu mc buffer) which in turn requires pda being setup already. * pda setup can't be done before xen_load_gdt() ... Next question while looking at xen_load_gdt(): why does it use multicalls in the first place?

As promised to Andrew, and with much thanks to Andi Kleen for feedback, this is the new series of lguest patches. Main change is the move to drivers/lguest (for future non-i386 expansion), but lots of cleanups driven by Andi's feedback and the documentation effort (which made me examine every line of code). It's not perfect, but it's definitely useful. Cheers, Rusty. List of

As promised to Andrew, and with much thanks to Andi Kleen for feedback, this is the new series of lguest patches. Main change is the move to drivers/lguest (for future non-i386 expansion), but lots of cleanups driven by Andi's feedback and the documentation effort (which made me examine every line of code). It's not perfect, but it's definitely useful. Cheers, Rusty. List of

...d); + + ap.idtr_limit = IDT_ENTRIES * 8 - 1; + ap.idtr_base = (unsigned long) idt_table; + + ap.ldtr = 0; + + ap.cs = __KERNEL_CS; + ap.eip = (unsigned long) start_eip; + ap.ss = __KERNEL_DS; + ap.esp = (unsigned long) start_esp; + + ap.ds = __USER_DS; + ap.es = __USER_DS; + ap.fs = 0; + ap.gs = __KERNEL_PDA; + + ap.eflags = 0; + + setup_pda(); + +#ifdef CONFIG_X86_PAE + /* efer should match BSP efer. */ + if (cpu_has_nx) { + unsigned l, h; + rdmsr(MSR_EFER, l, h); + ap.efer = (unsigned long long) h << 32 | l; + } +#endif + + ap.cr3 = __pa(swapper_pg_dir); + /* Protected mode, paging, AM, WP...

...d); + + ap.idtr_limit = IDT_ENTRIES * 8 - 1; + ap.idtr_base = (unsigned long) idt_table; + + ap.ldtr = 0; + + ap.cs = __KERNEL_CS; + ap.eip = (unsigned long) start_eip; + ap.ss = __KERNEL_DS; + ap.esp = (unsigned long) start_esp; + + ap.ds = __USER_DS; + ap.es = __USER_DS; + ap.fs = 0; + ap.gs = __KERNEL_PDA; + + ap.eflags = 0; + + setup_pda(); + +#ifdef CONFIG_X86_PAE + /* efer should match BSP efer. */ + if (cpu_has_nx) { + unsigned l, h; + rdmsr(MSR_EFER, l, h); + ap.efer = (unsigned long long) h << 32 | l; + } +#endif + + ap.cr3 = __pa(swapper_pg_dir); + /* Protected mode, paging, AM, WP...