[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <Y/CuL9CAf1tfAmhR@memverge.com>
Date: Sat, 18 Feb 2023 05:53:35 -0500
From: Gregory Price <gregory.price@...verge.com>
To: Oleg Nesterov <oleg@...hat.com>
Cc: Gregory Price <gourry.memverge@...il.com>,
linux-kernel@...r.kernel.org, linux-doc@...r.kernel.org,
avagin@...il.com, peterz@...radead.org, luto@...nel.org,
krisman@...labora.com, tglx@...utronix.de, corbet@....net,
shuah@...nel.org
Subject: Re: [PATCH 2/2] ptrace,syscall_user_dispatch: checkpoint/restore
support for SUD
On Thu, Feb 16, 2023 at 02:57:38PM +0100, Oleg Nesterov wrote:
> On 02/14, Gregory Price wrote:
> >
> > +struct compat_ptrace_sud_config {
> > + compat_ulong_t mode;
> > + compat_uptr_t selector;
> > + compat_ulong_t offset;
> > + compat_ulong_t len;
> > +};
>
> ...
>
> > +int syscall_user_dispatch_get_config(struct task_struct *task, unsigned long size,
> > + void __user *data)
> > +{
> > + struct syscall_user_dispatch *sd = &task->syscall_dispatch;
> > +#ifdef CONFIG_COMPAT
> > + if (unlikely(in_compat_syscall())) {
> > + struct compat_ptrace_sud_config cfg32;
> > +
> > + if (size != sizeof(struct compat_ptrace_sud_config))
> > + return -EINVAL;
> > +
>
> Horror ;) why?
>
> See my reply to v9, just make
>
> struct ptrace_sud_config {
> __u8 mode;
> __u64 selector;
> __u64 offset;
> __u64 len;
> };
>
> Oleg.
>
It was unclear to me what the prior note was asking an I followed the
pattern of other compat code in ptrace. For some reason i got it in my
head that u64 would compile down to u32 in compatibility mode and i went
full-stupid.
will back out this compat code here and fixup the struct.
Powered by blists - more mailing lists