search for: pvop_call_arg1

...hat makes it easier to provide a better > optimized assembly version of that function. > > For Xen, vcpu_is_preempted(long) calls xen_vcpu_stolen(int), the > downcast from long to int is not a problem as vCPU number won't exceed > 32 bits. > Note that because of the cast in PVOP_CALL_ARG1() this patch is pointless. Then again, it doesn't seem to affect code generation, so why not. Takes away the reliance on that weird cast.

...hat makes it easier to provide a better > optimized assembly version of that function. > > For Xen, vcpu_is_preempted(long) calls xen_vcpu_stolen(int), the > downcast from long to int is not a problem as vCPU number won't exceed > 32 bits. > Note that because of the cast in PVOP_CALL_ARG1() this patch is pointless. Then again, it doesn't seem to affect code generation, so why not. Takes away the reliance on that weird cast.

...ESAVE(rettype, native, op) #define PVOP_VCALLEE0(op) \ __PVOP_VCALLEESAVE(op, "", "") +#define PVOP_ALT_VCALLEE0(native, op) \ + __PVOP_ALT_VCALLEESAVE(native, op) #define PVOP_CALL1(rettype, op, arg1) \ __PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1)) #define PVOP_VCALL1(op, arg1) \ __PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1)) +#define PVOP_ALT_VCALL1(native, op, arg1) \ + __PVOP_ALT_VCALL(native, op, PVOP_CALL_ARG1(arg1)) #define PVOP_CALLEE1(rettype, op, arg1) \ __PVOP_CALLEESAVE(rettype, op, &q...

...avirt_patch_unlock(void *insnbuf, unsigned len); unsigned paravirt_patch_ignore(unsigned len); unsigned paravirt_patch_call(void *insnbuf, const void *target, u16 tgt_clobbers, @@ -620,6 +622,12 @@ int paravirt_disable_iospace(void); __PVOP_CALL(rettype, op, "", "", PVOP_CALL_ARG1(arg1)) #define PVOP_VCALL1(op, arg1) \ __PVOP_VCALL(op, "", "", PVOP_CALL_ARG1(arg1)) +/* + * Using LOCK_PREFIX_HERE works because the lock prefix or it's replacement + * is checked to be present before being replaced. + */ +#define PVOP_VCALL1_LOCK(op, arg1) \ +...

...vide a better >> optimized assembly version of that function. >> >> For Xen, vcpu_is_preempted(long) calls xen_vcpu_stolen(int), the >> downcast from long to int is not a problem as vCPU number won't exceed >> 32 bits. >> > Note that because of the cast in PVOP_CALL_ARG1() this patch is > pointless. > > Then again, it doesn't seem to affect code generation, so why not. Takes > away the reliance on that weird cast. I add this patch because I am a bit uneasy about clearing the upper 32 bits of rdi and assuming that the compiler won't have a previ...

Paravirtualized spinlocks produce some overhead even if the kernel is running on bare metal. The main reason are the more complex locking and unlocking functions. Especially unlocking is no longer just one instruction but so complex that it is no longer inlined. This patch series addresses this issue by adding two more pvops functions to reduce the size of the inlined spinlock functions. When

Paravirtualized spinlocks produce some overhead even if the kernel is running on bare metal. The main reason are the more complex locking and unlocking functions. Especially unlocking is no longer just one instruction but so complex that it is no longer inlined. This patch series addresses this issue by adding two more pvops functions to reduce the size of the inlined spinlock functions. When

On Mon, Feb 13, 2017 at 05:34:01PM -0500, Waiman Long wrote: > It is the address of &steal_time that will exceed the 32-bit limit. That seems extremely unlikely. That would mean we have more than 4G worth of per-cpu variables declared in the kernel.

On Mon, Feb 13, 2017 at 05:34:01PM -0500, Waiman Long wrote: > It is the address of &steal_time that will exceed the 32-bit limit. That seems extremely unlikely. That would mean we have more than 4G worth of per-cpu variables declared in the kernel.

This changes the pv ops code generation to more closely match reality. For example, instead of: callq *0xffffffff81e3a400 (pv_irq_ops.save_fl) vmlinux will now show: pushfq pop %rax nop nop nop nop nop which is what the runtime version of the code will show in most cases. This idea was suggested by Andy Lutomirski. The benefits are: - For the most common runtime cases

This changes the pv ops code generation to more closely match reality. For example, instead of: callq *0xffffffff81e3a400 (pv_irq_ops.save_fl) vmlinux will now show: pushfq pop %rax nop nop nop nop nop which is what the runtime version of the code will show in most cases. This idea was suggested by Andy Lutomirski. The benefits are: - For the most common runtime cases

v3->v4: - Fix x86-32 build error. v2->v3: - Provide an optimized __raw_callee_save___kvm_vcpu_is_preempted() in assembly as suggested by PeterZ. - Add a new patch to change vcpu_is_preempted() argument type to long to ease the writing of the assembly code. v1->v2: - Rerun the fio test on a different system on both bare-metal and a KVM guest. Both sockets were

v3->v4: - Fix x86-32 build error. v2->v3: - Provide an optimized __raw_callee_save___kvm_vcpu_is_preempted() in assembly as suggested by PeterZ. - Add a new patch to change vcpu_is_preempted() argument type to long to ease the writing of the assembly code. v1->v2: - Rerun the fio test on a different system on both bare-metal and a KVM guest. Both sockets were