[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAMzpN2i2DUD5i0ui7HE_F7+U80hXs2Fa+RgZa_DOCsJ+Eae52Q@mail.gmail.com>
Date: Wed, 26 Feb 2020 19:04:47 -0500
From: Brian Gerst <brgerst@...il.com>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: LKML <linux-kernel@...r.kernel.org>,
"the arch/x86 maintainers" <x86@...nel.org>,
Steven Rostedt <rostedt@...dmis.org>,
Juergen Gross <jgross@...e.com>,
Paolo Bonzini <pbonzini@...hat.com>,
Arnd Bergmann <arnd@...db.de>
Subject: Re: [patch 01/15] x86/irq: Convey vector as argument and not in ptregs
On Wed, Feb 26, 2020 at 6:43 PM Thomas Gleixner <tglx@...utronix.de> wrote:
>
> Brian Gerst <brgerst@...il.com> writes:
> > On Wed, Feb 26, 2020 at 3:13 PM Thomas Gleixner <tglx@...utronix.de> wrote:
> >> Brian Gerst <brgerst@...il.com> writes:
> >> Now the question is whether we care about the packed stubs or just make
> >> them larger by using alignment to get rid of this silly +0x80 and
> >> ~vector fixup later on. The straight forward thing clearly has its charm
> >> and I doubt it matters in measurable ways.
> >
> > I think we can get rid of the inversion. That was done so orig_ax had
> > a negative number (signifying it's not a syscall), but if you replace
> > it with -1 that isn't necessary. A simple -0x80 offset should be
> > sufficient.
> >
> > I think it's a worthy optimization to keep. There are 240 of these
> > stubs, so increasing the allocation to 16 bytes would add 1920 bytes
> > to the kernel text.
>
> I rather pay the 2k text size for readable and straight forward
> code. Can you remind me why we are actually worrying at that level about
> 32bit x86 instead of making it depend on CONFIG_OBSCURE?
Because this also applies to the 64-bit kernel?
--
Brian Gerst
Powered by blists - more mailing lists