From: Samuel Thibault <samuel.thibault@gnu.org>
To: Sergey Bugaev <bugaevc@gmail.com>
Cc: bug-hurd@gnu.org, libc-alpha@sourceware.org,
"Flávio Cruz" <flaviocruz@gmail.com>
Subject: Re: [RFC PATCH glibc 11/12] hurd, htl: Add some x86_64-specific code
Date: Sun, 12 Feb 2023 18:02:10 +0100 [thread overview]
Message-ID: <20230212170210.rcr3ru7jpvma5euj@begin> (raw)
In-Reply-To: <CAN9u=HfQdsbKp92K3SgyQt4qs_j=jS9+r1QxHEtVoq717J4U6Q@mail.gmail.com>
Sergey Bugaev, le dim. 12 févr. 2023 19:51:33 +0300, a ecrit:
> On Sun, Feb 12, 2023 at 7:36 PM Samuel Thibault <samuel.thibault@gnu.org> wrote:
> >
> > Sergey Bugaev, le dim. 12 févr. 2023 19:25:11 +0300, a ecrit:
> > > On Sun, Feb 12, 2023 at 7:11 PM Samuel Thibault <samuel.thibault@gnu.org> wrote:
> > > > Sergey Bugaev, le dim. 12 févr. 2023 14:10:42 +0300, a ecrit:
> > > > > We should not need a getter routine, because one can simply inspect the target
> > > > > thread's state (unless, again, I misunderstand things horribly).
> > > >
> > > > For 16bit fs/gs values we could read them from userland yes. But for
> > > > fs/gs base, the FSGSBASE instruction is not available on all 64bit
> > > > processors. And ATM in THREAD_TCB we want to be able to get the base of
> > > > another thread.
> > >
> > > What I've meant is:
> > >
> > > __thread_get_state (whatever_thread, &state);
> > > uintptr_t its_fs_base = state->fs_base;
> > >
> > > You can't really do the same to *write* [fg]s_base, because doing
> > > thread_set_state on your own thread is bound to end badly.
> >
> > ? Well, sure, just like setting fs/gs through thread state was not done
> > for i386.
> >
> > I don't see where you're aiming. Getting fs/gs from __thread_get_state
> > won't actually give you the base, you'll just read something like 0.
>
> It is my understanding that the actual values of fs/gs (i.e. the index
> of a descriptor) are not useful on x86_64. But fs_base and gs_base are
> now things that you have to store in the thread state and save/restore
> on every context switch. fs_base and gs_base are like registers in
> their own right (well, MSRs are registers). Thus, it should be easy to
> read them from the state structure exposed by the kernel.
Ah, so you mean adding the fs/gs bases to the thread_state content.
I'd rather make it a separate state content, like we have a separate
i386_DEBUG_STATE content.
(And thus no need for a new RPC).
Samuel
next prev parent reply other threads:[~2023-02-12 17:02 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-02-12 11:10 [RFC PATCH 0/12] Towards glibc on x86_64-gnu Sergey Bugaev
2023-02-12 11:10 ` [RFC PATCH glibc 1/12] hurd: Refactor readlinkat() Sergey Bugaev
2023-02-12 14:49 ` Samuel Thibault
2023-02-12 11:10 ` [RFC PATCH glibc 2/12] hurd: Use mach_msg_type_number_t where appropriate Sergey Bugaev
2023-02-12 14:52 ` Samuel Thibault
2023-02-12 11:10 ` [RFC PATCH glibc 3/12] mach, hurd: Cast through uintptr_t Sergey Bugaev
2023-02-12 14:55 ` Samuel Thibault
2023-02-12 11:10 ` [RFC PATCH glibc 4/12] hurd: Fix xattr error value Sergey Bugaev
2023-02-12 14:56 ` Samuel Thibault
2023-02-12 11:10 ` [RFC PATCH glibc 5/12] htl: Fix semaphore reference Sergey Bugaev
2023-02-12 14:57 ` Samuel Thibault
2023-02-12 11:10 ` [RFC PATCH hurd 6/12] hurd: Fix modes_t and speeds_t types on 64-bit Sergey Bugaev
2023-02-12 15:00 ` Samuel Thibault
2023-02-12 15:15 ` Sergey Bugaev
2023-02-12 15:22 ` Samuel Thibault
2023-02-12 16:13 ` Sergey Bugaev
2023-02-12 16:30 ` Samuel Thibault
2023-02-12 19:03 ` Samuel Thibault
2023-02-12 11:10 ` [RFC PATCH mig 7/12] Drop -undef -ansi from cpp flags Sergey Bugaev
2023-02-12 15:01 ` Samuel Thibault
2023-02-12 18:43 ` Flávio Cruz
2023-02-12 18:44 ` Samuel Thibault
2023-02-12 11:10 ` [RFC PATCH mig 8/12] Set max type alignment to sizeof(long) Sergey Bugaev
2023-02-12 11:10 ` [RFC PATCH glibc 9/12] mach: Look for mach_i386.defs on x86_64 too Sergey Bugaev
2023-02-12 15:07 ` Samuel Thibault
2023-02-12 15:38 ` Sergey Bugaev
2023-02-12 15:46 ` Samuel Thibault
2023-02-12 16:01 ` Sergey Bugaev
2023-02-16 20:22 ` Joseph Myers
2023-02-12 11:10 ` [RFC PATCH glibc 10/12] hurd: Set up the basic tree for x86_64-gnu Sergey Bugaev
2023-02-12 15:15 ` Samuel Thibault
2023-02-12 11:10 ` [RFC PATCH glibc 11/12] hurd, htl: Add some x86_64-specific code Sergey Bugaev
2023-02-12 16:11 ` Samuel Thibault
2023-02-12 16:25 ` Sergey Bugaev
2023-02-12 16:36 ` Samuel Thibault
2023-02-12 16:40 ` Florian Weimer
2023-02-12 16:46 ` Samuel Thibault
2023-02-12 19:29 ` Florian Weimer
2023-02-12 19:37 ` Samuel Thibault
2023-02-12 16:51 ` Sergey Bugaev
2023-02-12 17:02 ` Samuel Thibault [this message]
2023-02-12 17:06 ` Sergey Bugaev
2023-02-12 11:10 ` [RFC PATCH glibc 12/12] C11 thrd: Downgrade the default alignment of mtx_t Sergey Bugaev
2023-02-12 15:18 ` Samuel Thibault
2023-02-12 15:52 ` Sergey Bugaev
2023-02-12 16:29 ` Samuel Thibault
2023-02-12 16:12 ` [RFC PATCH 0/12] Towards glibc on x86_64-gnu Samuel Thibault
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230212170210.rcr3ru7jpvma5euj@begin \
--to=samuel.thibault@gnu.org \
--cc=bug-hurd@gnu.org \
--cc=bugaevc@gmail.com \
--cc=flaviocruz@gmail.com \
--cc=libc-alpha@sourceware.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).