From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Ingo Molnar <mingo@elte.hu>
Cc: LKML <linux-kernel@vger.kernel.org>, Ian Campbell <ijc@hellion.org.uk>
Subject: [PATCH 25 of 31] x86: only enable interrupts when kernel state has been set up
Date: Mon, 17 Mar 2008 16:37:16 -0700 [thread overview]
Message-ID: <c8c4e4f4c61c1ccc8326.1205797036@localhost> (raw)
In-Reply-To: <patchbomb.1205797011@localhost>
The sysenter path tries to enable interrupts immediately. Unfortunately
this doesn't work in a paravirt environment, because not enough kernel
state has been set up at that point (namely, pointing %fs to the kernel
percpu data segment). To fix this, defer ENABLE_INTERRUPTS until after
the kernel state has been set up.
Unfortunately this means that we're running with interrupts disabled
for a while without calling the IRQ tracing code, but that can't be
called without setting up %fs either.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
---
arch/x86/kernel/entry_32.S | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -290,10 +290,10 @@
movl TSS_sysenter_sp0(%esp),%esp
ENTRY(sysenter_past_esp)
/*
- * No need to follow this irqs on/off section: the syscall
- * disabled irqs and here we enable it straight after entry:
+ * Interrupts are disabled here, but we can't trace it until
+ * enough kernel state to call TRACE_IRQS_OFF can be called - but
+ * we immediately enable interrupts at that point anyway.
*/
- ENABLE_INTERRUPTS(CLBR_NONE)
pushl $(__USER_DS)
CFI_ADJUST_CFA_OFFSET 4
/*CFI_REL_OFFSET ss, 0*/
@@ -329,6 +329,7 @@
pushl %eax
CFI_ADJUST_CFA_OFFSET 4
SAVE_ALL
+ ENABLE_INTERRUPTS(CLBR_NONE)
GET_THREAD_INFO(%ebp)
/* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */
@@ -545,6 +546,7 @@
pushl %eax # save orig_eax
CFI_ADJUST_CFA_OFFSET 4
SAVE_ALL
+ ENABLE_INTERRUPTS(CLBR_NONE)
GET_THREAD_INFO(%ebp)
movl $-EFAULT,PT_EAX(%esp)
jmp resume_userspace
next prev parent reply other threads:[~2008-03-18 5:27 UTC|newest]
Thread overview: 58+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-03-17 23:36 [PATCH 00 of 31] x86: unification and xen updates Jeremy Fitzhardinge
2008-03-17 23:36 ` [PATCH 01 of 31] xen: fix RMW when unmasking events Jeremy Fitzhardinge
2008-03-17 23:36 ` [PATCH 02 of 31] xen: fix UP setup of shared_info Jeremy Fitzhardinge
2008-03-17 23:36 ` [PATCH 03 of 31] x86: convert pgalloc_64.h from macros to inlines Jeremy Fitzhardinge
2008-03-17 23:36 ` [PATCH 04 of 31] x86: add common mm/pgtable.c Jeremy Fitzhardinge
2008-03-17 23:36 ` [PATCH 05 of 31] x86: put paravirt stubs into common asm/pgalloc.h Jeremy Fitzhardinge
2008-03-17 23:36 ` [PATCH 06 of 31] x86: move pte functions " Jeremy Fitzhardinge
2008-03-17 23:36 ` [PATCH 07 of 31] x86: move pmd " Jeremy Fitzhardinge
2008-03-17 23:36 ` [PATCH 08 of 31] x86: move pgalloc pud and pgd operations into common place Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 09 of 31] x86: move all the pgd_list handling to one place Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 10 of 31] x86: rename paravirt_alloc_pt etc after the pagetable structure Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 11 of 31] x86: add pud_alloc for 4-level pagetables Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 12 of 31] x86/pgtable.h: demacro ptep_set_access_flags Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 13 of 31] x86/pgtable.h: demacro ptep_test_and_clear_young Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 14 of 31] x86/pgtable.h: demacro ptep_clear_flush_young Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 15 of 31] x86: demacro pgalloc paravirt stubs Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 16 of 31] xen: use appropriate pte types Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 17 of 31] xen: make use of pte_t union Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 18 of 31] xen: unify pte operations Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 19 of 31] xen: use phys_addr_t when referring to physical addresses Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 20 of 31] xen: unify pte operations on machine frames Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 21 of 31] xen: make sure iret faults are trapped Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 22 of 31] x86: unify KERNEL_PGD_PTRS Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 23 of 31] x86: unify pgd ctor/dtor Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 24 of 31] xen: add support for callbackops hypercall Jeremy Fitzhardinge
2008-03-17 23:37 ` Jeremy Fitzhardinge [this message]
2008-03-17 23:37 ` [PATCH 26 of 31] xen: support sysenter/sysexit if hypervisor does Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 27 of 31] xen: implement a debug-interrupt handler Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 28 of 31] xen: make sure retriggered events are set pending Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 29 of 31] xen: short-cut for recursive event handling Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 30 of 31] xen: no need for domU to worry about MCE/MCA Jeremy Fitzhardinge
2008-03-17 23:37 ` [PATCH 31 of 31] xen: jump to iret fixup Jeremy Fitzhardinge
2008-03-18 21:33 ` [PATCH 00 of 31] x86: unification and xen updates Zachary Amsden
2008-03-18 21:37 ` Jeremy Fitzhardinge
2008-03-19 9:22 ` Ian Campbell
2008-03-19 15:11 ` Jeremy Fitzhardinge
2008-03-19 16:50 ` Ian Campbell
2008-03-19 18:45 ` Jeremy Fitzhardinge
2008-03-19 19:07 ` Ingo Molnar
2008-03-19 19:19 ` Jeremy Fitzhardinge
2008-03-19 19:05 ` Ingo Molnar
2008-03-19 19:46 ` Ingo Molnar
2008-03-19 20:02 ` Jeremy Fitzhardinge
2008-03-19 20:09 ` Ingo Molnar
2008-03-19 21:12 ` Jeremy Fitzhardinge
2008-03-19 21:33 ` Ingo Molnar
2008-03-19 21:54 ` Jeremy Fitzhardinge
2008-03-19 22:00 ` Ingo Molnar
2008-03-19 21:58 ` Jeremy Fitzhardinge
2008-03-19 22:04 ` Ingo Molnar
2008-03-19 22:15 ` Jeremy Fitzhardinge
2008-03-19 23:52 ` Ingo Molnar
2008-03-20 20:24 ` Jeremy Fitzhardinge
2008-03-21 13:17 ` Ingo Molnar
2008-03-21 14:35 ` Ingo Molnar
2008-03-21 15:15 ` Jeremy Fitzhardinge
2008-03-21 15:25 ` Ingo Molnar
2008-03-22 2:34 ` Jeremy Fitzhardinge
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=c8c4e4f4c61c1ccc8326.1205797036@localhost \
--to=jeremy@goop.org \
--cc=ijc@hellion.org.uk \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
/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
Powered by JetHome