From: Janosch Frank <frankja@linux.ibm.com>
To: Thomas Huth <thuth@redhat.com>, kvm@vger.kernel.org
Cc: linux-s390@vger.kernel.org, imbrenda@linux.ibm.com,
pmorel@linux.ibm.com, david@redhat.com
Subject: Re: [kvm-unit-tests PATCH v2 4/8] s390x: Introduce and use CALL_INT_HANDLER macro
Date: Thu, 18 Feb 2021 10:16:03 +0100 [thread overview]
Message-ID: <1ff30077-ec0c-7bee-3770-d0ff0f4593a9@linux.ibm.com> (raw)
In-Reply-To: <bf4f49dd-5255-91c6-5a65-4d346d7d2701@redhat.com>
On 2/17/21 6:03 PM, Thomas Huth wrote:
> On 17/02/2021 17.22, Janosch Frank wrote:
>> On 2/17/21 4:55 PM, Thomas Huth wrote:
>>> On 17/02/2021 15.41, Janosch Frank wrote:
>>>> The ELF ABI dictates that we need to allocate 160 bytes of stack space
>>>> for the C functions we're calling. Since we would need to do that for
>>>> every interruption handler which, combined with the new stack argument
>>>> being saved in GR2, makes cstart64.S look a bit messy.
>>>>
>>>> So let's introduce the CALL_INT_HANDLER macro that handles all of
>>>> that, calls the C interrupt handler and handles cleanup afterwards.
>>>>
>>>> Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
>>>> ---
>>>> s390x/cstart64.S | 28 +++++-----------------------
>>>> s390x/macros.S | 17 +++++++++++++++++
>>>> 2 files changed, 22 insertions(+), 23 deletions(-)
>>>>
>>>> diff --git a/s390x/cstart64.S b/s390x/cstart64.S
>>>> index 35d20293..666a9567 100644
>>>> --- a/s390x/cstart64.S
>>>> +++ b/s390x/cstart64.S
>>>> @@ -92,37 +92,19 @@ memsetxc:
>>>>
>>>> .section .text
>>>> pgm_int:
>>>> - SAVE_REGS_STACK
>>>> - lgr %r2, %r15
>>>> - brasl %r14, handle_pgm_int
>>>> - RESTORE_REGS_STACK
>>>> - lpswe GEN_LC_PGM_OLD_PSW
>>>> + CALL_INT_HANDLER handle_pgm_int, GEN_LC_PGM_OLD_PSW
>>>>
>>>> ext_int:
>>>> - SAVE_REGS_STACK
>>>> - lgr %r2, %r15
>>>> - brasl %r14, handle_ext_int
>>>> - RESTORE_REGS_STACK
>>>> - lpswe GEN_LC_EXT_OLD_PSW
>>>> + CALL_INT_HANDLER handle_ext_int, GEN_LC_EXT_OLD_PSW
>>>>
>>>> mcck_int:
>>>> - SAVE_REGS_STACK
>>>> - brasl %r14, handle_mcck_int
>>>> - RESTORE_REGS_STACK
>>>> - lpswe GEN_LC_MCCK_OLD_PSW
>>>> + CALL_INT_HANDLER handle_mcck_int, GEN_LC_MCCK_OLD_PSW
>>>>
>>>> io_int:
>>>> - SAVE_REGS_STACK
>>>> - lgr %r2, %r15
>>>> - brasl %r14, handle_io_int
>>>> - RESTORE_REGS_STACK
>>>> - lpswe GEN_LC_IO_OLD_PSW
>>>> + CALL_INT_HANDLER handle_io_int, GEN_LC_IO_OLD_PSW
>>>>
>>>> svc_int:
>>>> - SAVE_REGS_STACK
>>>> - brasl %r14, handle_svc_int
>>>> - RESTORE_REGS_STACK
>>>> - lpswe GEN_LC_SVC_OLD_PSW
>>>> + CALL_INT_HANDLER handle_svc_int, GEN_LC_SVC_OLD_PSW
>>>>
>>>> .align 8
>>>> initial_psw:
>>>> diff --git a/s390x/macros.S b/s390x/macros.S
>>>> index a7d62c6f..212a3823 100644
>>>> --- a/s390x/macros.S
>>>> +++ b/s390x/macros.S
>>>> @@ -11,6 +11,23 @@
>>>> * David Hildenbrand <david@redhat.com>
>>>> */
>>>> #include <asm/asm-offsets.h>
>>>> +/*
>>>> + * Exception handler macro that saves registers on the stack,
>>>> + * allocates stack space and calls the C handler function. Afterwards
>>>> + * we re-load the registers and load the old PSW.
>>>> + */
>>>> + .macro CALL_INT_HANDLER c_func, old_psw
>>>> + SAVE_REGS_STACK
>>>> + /* Save the stack address in GR2 which is the first function argument */
>>>> + lgr %r2, %r15
>>>> + /* Allocate stack pace for called C function, as specified in s390 ELF ABI */
>>>> + slgfi %r15, 160
>>>
>>> By the way, don't you have to store a back chain pointer at the bottom of
>>> that area, too, if you want to use -mbackchoin in the next patch?
>>
>> Don't I already do that in #2?
>
> You do it in the SAVE_REGS_STACK patch, yes. But not on the bottom of the
> new 160 bytes stack frame that you've added here. But I guess it doesn't
> really matter for your back traces, since you load %r2 with %r15 before
> decrementing the stack by 160, so this new stack frame simply gets ignored
> anyway.
>
> Thomas
>
Right, I'm currently fixing that up for the next version.
diff --git i/s390x/macros.S w/s390x/macros.S
index b4275c77..13cff299 100644
--- i/s390x/macros.S
+++ w/s390x/macros.S
@@ -22,6 +22,11 @@
lgr %r2, %r15
/* Allocate stack space for called C function, as specified in
s390 ELF ABI */
slgfi %r15, 160
+ /*
+ * Save the address of the interrupt stack into the back chain
+ * of the called function.
+ */
+ stg %r2, STACK_FRAME_INT_BACKCHAIN(%r15)
brasl %r14, \c_func
algfi %r15, 160
RESTORE_REGS_STACK
next prev parent reply other threads:[~2021-02-18 10:24 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-17 14:41 [kvm-unit-tests PATCH v2 0/8] s390x: Cleanup exception register save/restore and implement backtrace Janosch Frank
2021-02-17 14:41 ` [kvm-unit-tests PATCH v2 1/8] s390x: Fix fpc store address in RESTORE_REGS_STACK Janosch Frank
2021-02-17 14:41 ` [kvm-unit-tests PATCH v2 2/8] s390x: Fully commit to stack save area for exceptions Janosch Frank
2021-02-17 15:35 ` Thomas Huth
2021-02-17 16:54 ` Janosch Frank
2021-02-17 14:41 ` [kvm-unit-tests PATCH v2 3/8] RFC: s390x: Define STACK_FRAME_INT_SIZE macro Janosch Frank
2021-02-17 15:38 ` Thomas Huth
2021-02-17 16:08 ` Janosch Frank
2021-02-17 16:10 ` Thomas Huth
2021-02-17 14:41 ` [kvm-unit-tests PATCH v2 4/8] s390x: Introduce and use CALL_INT_HANDLER macro Janosch Frank
2021-02-17 15:49 ` Thomas Huth
2021-02-17 15:55 ` Thomas Huth
2021-02-17 16:22 ` Janosch Frank
2021-02-17 17:03 ` Thomas Huth
2021-02-18 9:16 ` Janosch Frank [this message]
2021-02-17 14:41 ` [kvm-unit-tests PATCH v2 5/8] s390x: Provide preliminary backtrace support Janosch Frank
2021-02-17 16:01 ` Thomas Huth
2021-02-17 16:12 ` Janosch Frank
2021-02-17 14:41 ` [kvm-unit-tests PATCH v2 6/8] s390x: Print more information on program exceptions Janosch Frank
2021-02-17 14:41 ` [kvm-unit-tests PATCH v2 7/8] s390x: Move diag308_load_reset to stack saving Janosch Frank
2021-02-17 14:41 ` [kvm-unit-tests PATCH v2 8/8] s390x: Remove SAVE/RESTORE_stack Janosch Frank
2021-02-17 16:18 ` Thomas Huth
2021-02-17 16:46 ` Janosch Frank
2021-02-17 16:50 ` Thomas Huth
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=1ff30077-ec0c-7bee-3770-d0ff0f4593a9@linux.ibm.com \
--to=frankja@linux.ibm.com \
--cc=david@redhat.com \
--cc=imbrenda@linux.ibm.com \
--cc=kvm@vger.kernel.org \
--cc=linux-s390@vger.kernel.org \
--cc=pmorel@linux.ibm.com \
--cc=thuth@redhat.com \
/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 NNTP newsgroup(s).