| From: |
| Peter Zijlstra <peterz-AT-infradead.org> |
| To: |
| x86-AT-kernel.org |
| Subject: |
| [RFC][PATCH 0/9] Variable size jump_label support |
| Date: |
| Mon, 07 Oct 2019 10:44:43 +0200 |
| Message-ID: |
| <20191007084443.79370128.1@infradead.org> |
| Cc: |
| peterz-AT-infradead.org, linux-kernel-AT-vger.kernel.org, rostedt-AT-goodmis.org, mhiramat-AT-kernel.org, bristot-AT-redhat.com, jbaron-AT-akamai.com, torvalds-AT-linux-foundation.org, tglx-AT-linutronix.de, mingo-AT-kernel.org, namit-AT-vmware.com, hpa-AT-zytor.com, luto-AT-kernel.org, ard.biesheuvel-AT-linaro.org, jpoimboe-AT-redhat.com, hjl.tools-AT-gmail.com |
| Archive-link: |
| Article |
These here patches are something I've been poking at for a while, enabling
jump_label to use 2 byte jumps/nops.
It _almost_ works :-/
That is, you can build some kernels with it (x86_64-defconfig for example works
just fine).
The problem comes when GCC generates a branch into another section, mostly
.text.unlikely. At that point GAS just gives up and throws a fit (more details
in the last patch).
Aside from anyone coming up with a really clever GAS trick, I don't see how we
can do this other than:
- get binutils/gas 'fixed', for this there's two possible approaches:
* add some extra operators such that something like:
.set disp %[l_yes] ~ 1b ? (%[l_yes] - (1b + 2)) : 128
works, where the new '~' infix operator would indicate left and right hand
operands are from the same section, and the ?: conditional operator would be
added to make it all work.
* add a 'fake' mnemonic for x86 that directly generates the right NOP:
nojmp{.d8,d32} %target
which would completely mirror how the existing 'jmp{.d8,d32} %target'
works, except it would emit single instruction NOP2/NOP5.
- use 'jmp' and get objtool to rewrite the text. Steven has earlier proposed
something like that (using recordmcount) and Linus hated that.
