lkml.org 
[lkml]   [2011]   [Mar]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH v2 2.6.38-rc8-tip 6/20] 6: x86: analyze instruction and determine fixups.
On Mon, 14 Mar 2011, Srikar Dronamraju wrote:
> +/*
> + * TODO:
> + * - Where necessary, examine the modrm byte and allow only valid instructions
> + * in the different Groups and fpu instructions.
> + */
> +
> +static bool is_prefix_bad(struct insn *insn)
> +{
> + int i;
> +
> + for (i = 0; i < insn->prefixes.nbytes; i++) {
> + switch (insn->prefixes.bytes[i]) {
> + case 0x26: /*INAT_PFX_ES */
> + case 0x2E: /*INAT_PFX_CS */
> + case 0x36: /*INAT_PFX_DS */
> + case 0x3E: /*INAT_PFX_SS */
> + case 0xF0: /*INAT_PFX_LOCK */
> + return 1;

true

> + }
> + }
> + return 0;

false

> +}

> +static int validate_insn_32bits(struct uprobe *uprobe, struct insn *insn)
> +{
> + insn_init(insn, uprobe->insn, false);
> +
> + /* Skip good instruction prefixes; reject "bad" ones. */
> + insn_get_opcode(insn);
> + if (is_prefix_bad(insn)) {
> + report_bad_prefix();
> + return -EPERM;

-ENOTSUPP perhaps. That's not a permission problem

> + }

> +/**
> + * analyze_insn - instruction analysis including validity and fixups.
> + * @tsk: the probed task.
> + * @uprobe: the probepoint information.
> + * Return 0 on success or a -ve number on error.
> + */
> +int analyze_insn(struct task_struct *tsk, struct uprobe *uprobe)
> +{
> + int ret;
> + struct insn insn;
> +
> + uprobe->fixups = 0;
> +#ifdef CONFIG_X86_64
> + uprobe->arch_info.rip_rela_target_address = 0x0;
> +#endif

Please get rid of this #ifdef and use inlines (empty for 32bit)

> +
> + if (is_32bit_app(tsk))
> + ret = validate_insn_32bits(uprobe, &insn);
> + else
> + ret = validate_insn_64bits(uprobe, &insn);
> + if (ret != 0)
> + return ret;
> +#ifdef CONFIG_X86_64

Ditto

> + ret = handle_riprel_insn(uprobe, &insn);
> + if (ret == -1)
> + /* rip-relative; can't XOL */
> + return 0;

So we return -1 from handle_riprel_insn() and return success? Btw how
deals handle_riprel_insn() with 32bit user space ?

> +#endif
> + prepare_fixups(uprobe, &insn);
> + return 0;

Thanks,

tglx


\
 
 \ /
  Last update: 2011-03-15 15:41    [from the cache]
©2003-2011 Jasper Spaans