Re: [kvm-unit-tests PATCH 01/16] x86: make irq_enable avoid the interrupt shadow

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, 2022-10-27 at 15:50 +0000, Sean Christopherson wrote:
> On Thu, Oct 27, 2022, Maxim Levitsky wrote:
> > On Mon, 2022-10-24 at 22:49 +0000, Sean Christopherson wrote:
> > > On Mon, Oct 24, 2022, Maxim Levitsky wrote:
> > > > I usually use just "\n", but the safest is "\n\t".
> > > 
> > > I'm pretty sure we can ignore GCC's warning here and maximize readability.  There
> > > are already plenty of asm blobs that use a semicolon.
> > 
> > IMHO this is corner cutting and you yourself said that this is wrong.
> > 
> > The other instances which use semicolon should be fixed IMHO.
> 
> The kernel itself has multiple instances of "sti; ..." alone, I'm quite confident
> this we can prioritize making the code easy to read without risking future breakage.
> 
> $ git grep -E "\"sti\;"
> arch/x86/include/asm/irqflags.h:        asm volatile("sti; hlt": : :"memory");
> arch/x86/include/asm/mwait.h:   asm volatile("sti; .byte 0x0f, 0x01, 0xc9;"
> arch/x86/include/asm/paravirt.h:        PVOP_ALT_VCALLEE0(irq.irq_enable, "sti;", ALT_NOT(X86_FEATURE_XENPV));
> tools/testing/selftests/kvm/x86_64/xapic_ipi_test.c:            asm volatile("sti; hlt; cli");
> tools/testing/selftests/x86/iopl.c:             asm volatile("sti; pushf; pop %[flags]"
> 

All right, let it be, but then lets also replace of '\n\t' with just '\n', just so that we don't pretend that
we follow the gcc advice, to at least be consistent.

Best regards,
	Maxim Levitsky




[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux