From: Florian Bezdeka <florian.bezdeka@siemens.com>
To: rpm@xenomai.org
Cc: xenomai@lists.linux.dev, jan.kiszka@siemens.com,
Florian Bezdeka <florian.bezdeka@siemens.com>
Subject: [PATCH dovetail 6.8, 6.6, 6.1] x86: dovetail: Fix compat syscall handling, account for new IDT entry
Date: Tue, 16 Apr 2024 12:24:36 +0200 [thread overview]
Message-ID: <20240416102436.550147-1-florian.bezdeka@siemens.com> (raw)
The following upstream commit introduces a new IDTENTRY for handling
compat syscall. As that commit is now backported into stable trees it
broke compat syscall handling for dovetail.
be5341eb0d43 ("x86/entry: Convert INT 0x80 emulation to IDTENTRY")
We no longer end up in do_int80_syscall_32() which has the dovetail
specific bits. We need the same handling in asm_int80_emulation().
The problematic part in asm_int80_emulation() is the call to
do_syscall_32_irqs_on() which basically overwrites regs->ax
unconditionally. Linux does not know about the co-kernel syscalls, so
we get -ENOSYS for each syscall result.
Signed-off-by: Florian Bezdeka <florian.bezdeka@siemens.com>
---
arch/x86/entry/common.c | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c
index 0220ca163a32..42e8cb036fff 100644
--- a/arch/x86/entry/common.c
+++ b/arch/x86/entry/common.c
@@ -212,9 +212,21 @@ DEFINE_IDTENTRY_RAW(int80_emulation)
local_irq_enable();
nr = syscall_enter_from_user_mode_work(regs, nr);
+
+ if (dovetailing()) {
+ if (nr == EXIT_SYSCALL_OOB) {
+ hard_local_irq_disable();
+ return;
+ }
+ if (nr == EXIT_SYSCALL_TAIL)
+ goto done;
+ }
+
do_syscall_32_irqs_on(regs, nr);
instrumentation_end();
+
+done:
syscall_exit_to_user_mode(regs);
}
#else /* CONFIG_IA32_EMULATION */
--
2.39.2
next reply other threads:[~2024-04-16 10:24 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-04-16 10:24 Florian Bezdeka [this message]
2024-04-23 6:26 ` [PATCH dovetail 6.8, 6.6, 6.1] x86: dovetail: Fix compat syscall handling, account for new IDT entry Philippe Gerum
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=20240416102436.550147-1-florian.bezdeka@siemens.com \
--to=florian.bezdeka@siemens.com \
--cc=jan.kiszka@siemens.com \
--cc=rpm@xenomai.org \
--cc=xenomai@lists.linux.dev \
/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).