qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: David Hildenbrand <david@redhat.com>
To: Richard Henderson <richard.henderson@linaro.org>, qemu-devel@nongnu.org
Cc: thuth@redhat.com, qemu-s390x@nongnu.org, cohuck@redhat.com,
	laurent@vivier.eu
Subject: Re: [PATCH v2 01/15] linux-user/s390x: Fix sigframe types
Date: Thu, 29 Apr 2021 09:10:25 +0200	[thread overview]
Message-ID: <d5f5cf98-56e5-4000-62c0-7bbffe40b9a4@redhat.com> (raw)
In-Reply-To: <20210428193408.233706-2-richard.henderson@linaro.org>

On 28.04.21 21:33, Richard Henderson wrote:
> Noticed via gitlab clang-user job:
> 
>    TEST    signals on s390x
> ../linux-user/s390x/signal.c:258:9: runtime error: \
>    1.84467e+19 is outside the range of representable values of \
>    type 'unsigned long'
> 
> Which points to the fact that we were performing a double-to-uint64_t
> conversion while storing the fp registers, instead of just copying
> the data across.
> 
> Turns out there are several errors:
> 
> target_ulong is the size of the target register, whereas abi_ulong
> is the target 'unsigned long' type.  Not a big deal here, since we
> only support 64-bit s390x, but not correct either.
> 
> In target_sigcontext and target ucontext, we used a host pointer
> instead of a target pointer, aka abi_ulong.
> 
> Fixing this allows the removal of a cast to __put_user.
> 
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> ---
>   linux-user/s390x/signal.c | 26 +++++++++++++-------------
>   1 file changed, 13 insertions(+), 13 deletions(-)
> 
> diff --git a/linux-user/s390x/signal.c b/linux-user/s390x/signal.c
> index b68b44ae7e..707fb603d7 100644
> --- a/linux-user/s390x/signal.c
> +++ b/linux-user/s390x/signal.c
> @@ -37,13 +37,14 @@
>   
>   typedef struct {
>       target_psw_t psw;
> -    target_ulong gprs[__NUM_GPRS];
> -    unsigned int acrs[__NUM_ACRS];
> +    abi_ulong gprs[__NUM_GPRS];
> +    abi_uint acrs[__NUM_ACRS];
>   } target_s390_regs_common;
>   
>   typedef struct {
> -    unsigned int fpc;
> -    double   fprs[__NUM_FPRS];
> +    uint32_t fpc;
> +    uint32_t pad;
> +    uint64_t fprs[__NUM_FPRS];
>   } target_s390_fp_regs;
>   
>   typedef struct {
> @@ -51,22 +52,22 @@ typedef struct {
>       target_s390_fp_regs     fpregs;
>   } target_sigregs;
>   
> -struct target_sigcontext {
> -    target_ulong   oldmask[_SIGCONTEXT_NSIG_WORDS];
> -    target_sigregs *sregs;
> -};
> +typedef struct {
> +    abi_ulong oldmask[_SIGCONTEXT_NSIG_WORDS];
> +    abi_ulong sregs;
> +} target_sigcontext;
>   
>   typedef struct {
>       uint8_t callee_used_stack[__SIGNAL_FRAMESIZE];
> -    struct target_sigcontext sc;
> +    target_sigcontext sc;
>       target_sigregs sregs;
>       int signo;
>       uint8_t retcode[S390_SYSCALL_SIZE];
>   } sigframe;
>   
>   struct target_ucontext {
> -    target_ulong tuc_flags;
> -    struct target_ucontext *tuc_link;
> +    abi_ulong tuc_flags;
> +    abi_ulong tuc_link;
>       target_stack_t tuc_stack;
>       target_sigregs tuc_mcontext;
>       target_sigset_t tuc_sigmask;   /* mask last for extensibility */
> @@ -143,8 +144,7 @@ void setup_frame(int sig, struct target_sigaction *ka,
>   
>       save_sigregs(env, &frame->sregs);
>   
> -    __put_user((abi_ulong)(unsigned long)&frame->sregs,
> -               (abi_ulong *)&frame->sc.sregs);
> +    __put_user((abi_ulong)(unsigned long)&frame->sregs, &frame->sc.sregs);
>   
>       /* Set up to return from userspace.  If provided, use a stub
>          already in userspace.  */
> 

Reviewed-by: David Hildenbrand <david@redhat.com>

-- 
Thanks,

David / dhildenb



  reply	other threads:[~2021-04-29  7:12 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-28 19:33 [PATCH v2 00/15] linux-user/s390x: some signal fixes Richard Henderson
2021-04-28 19:33 ` [PATCH v2 01/15] linux-user/s390x: Fix sigframe types Richard Henderson
2021-04-29  7:10   ` David Hildenbrand [this message]
2021-04-28 19:33 ` [PATCH v2 02/15] linux-user/s390x: Use uint16_t for signal retcode Richard Henderson
2021-04-29  7:10   ` David Hildenbrand
2021-04-28 19:33 ` [PATCH v2 03/15] linux-user/s390x: Remove PSW_ADDR_AMODE Richard Henderson
2021-04-29  7:11   ` David Hildenbrand
2021-04-28 19:33 ` [PATCH v2 04/15] linux-user/s390x: Remove restore_sigregs return value Richard Henderson
2021-04-29  7:11   ` David Hildenbrand
2021-04-28 19:33 ` [PATCH v2 05/15] linux-user/s390x: Fix trace in restore_regs Richard Henderson
2021-04-29  7:12   ` David Hildenbrand
2021-04-28 19:33 ` [PATCH v2 06/15] linux-user/s390x: Fix sigcontext sregs value Richard Henderson
2021-04-29  7:13   ` David Hildenbrand
2021-04-28 19:34 ` [PATCH v2 07/15] linux-user/s390x: Use tswap_sigset in setup_rt_frame Richard Henderson
2021-04-29  7:14   ` David Hildenbrand
2021-04-28 19:34 ` [PATCH v2 08/15] linux-user/s390x: Tidy save_sigregs Richard Henderson
2021-04-29  7:14   ` David Hildenbrand
2021-04-28 19:34 ` [PATCH v2 09/15] linux-user/s390x: Clean up single-use gotos in signal.c Richard Henderson
2021-04-29  7:15   ` David Hildenbrand
2021-04-28 19:34 ` [PATCH v2 10/15] linux-user/s390x: Set psw.mask properly for the signal handler Richard Henderson
2021-04-29  7:20   ` David Hildenbrand
2021-04-28 19:34 ` [PATCH v2 11/15] linux-user/s390x: Add stub sigframe argument for last_break Richard Henderson
2021-04-29  7:21   ` David Hildenbrand
2021-04-28 19:34 ` [PATCH v2 12/15] linux-user/s390x: Fix frame_addr corruption in setup_frame Richard Henderson
2021-04-29  7:21   ` David Hildenbrand
2021-04-28 19:34 ` [PATCH v2 13/15] linux-user/s390x: Add build asserts for sigset sizes Richard Henderson
2021-04-29  7:21   ` David Hildenbrand
2021-04-28 19:34 ` [PATCH v2 14/15] linux-user/s390x: Clean up signal.c Richard Henderson
2021-04-29  7:29   ` David Hildenbrand
2021-04-28 19:34 ` [PATCH v2 15/15] linux-user/s390x: Handle vector regs in signal stack Richard Henderson
2021-04-29  7:32   ` David Hildenbrand
2021-04-28 19:40 ` [PATCH v2 00/15] linux-user/s390x: some signal fixes Richard Henderson
2021-04-29  7:33 ` David Hildenbrand
2021-05-06 11:54 ` Cornelia Huck
2021-05-06 13:31   ` Laurent Vivier
2021-05-15 19:44 ` Laurent Vivier

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=d5f5cf98-56e5-4000-62c0-7bbffe40b9a4@redhat.com \
    --to=david@redhat.com \
    --cc=cohuck@redhat.com \
    --cc=laurent@vivier.eu \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-s390x@nongnu.org \
    --cc=richard.henderson@linaro.org \
    --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).