[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH v2 05/74] semihosting: Add target_strlen for softmmu-uaccess.
From: |
Peter Maydell |
Subject: |
Re: [PATCH v2 05/74] semihosting: Add target_strlen for softmmu-uaccess.h |
Date: |
Mon, 16 May 2022 16:11:03 +0100 |
On Tue, 3 May 2022 at 20:57, Richard Henderson
<richard.henderson@linaro.org> wrote:
>
> Mirror the user-only function of the same name.
>
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> ---
> include/semihosting/softmmu-uaccess.h | 3 +++
> semihosting/uaccess.c | 29 +++++++++++++++++++++++++++
> 2 files changed, 32 insertions(+)
>
> diff --git a/include/semihosting/softmmu-uaccess.h
> b/include/semihosting/softmmu-uaccess.h
> index 03300376d3..4f08dfc098 100644
> --- a/include/semihosting/softmmu-uaccess.h
> +++ b/include/semihosting/softmmu-uaccess.h
> @@ -53,4 +53,7 @@ void softmmu_unlock_user(CPUArchState *env, void *p,
> target_ulong addr, target_ulong len);
> #define unlock_user(s, args, len) softmmu_unlock_user(env, s, args, len)
>
> +ssize_t softmmu_strlen_user(CPUArchState *env, target_ulong addr);
> +#define target_strlen(p) softmmu_strlen_user(env, p)
> +
> #endif /* SEMIHOSTING_SOFTMMU_UACCESS_H */
> diff --git a/semihosting/uaccess.c b/semihosting/uaccess.c
> index 0d3b32b75d..3cd809122c 100644
> --- a/semihosting/uaccess.c
> +++ b/semihosting/uaccess.c
> @@ -23,6 +23,35 @@ void *softmmu_lock_user(CPUArchState *env, target_ulong
> addr,
> return p;
> }
>
> +ssize_t softmmu_strlen_user(CPUArchState *env, target_ulong addr)
> +{
> + char buf[256];
> + size_t len = 0;
> +
> + while (1) {
> + size_t chunk;
> + char *p;
> +
> + chunk = -(addr | TARGET_PAGE_MASK);
'chunk' is unsigned but we're assigning it a negative number here...
I assume this is doing some clever bit-twiddling trick but it isn't
very obvious.
In any case, rounding to a page boundary isn't sufficient to
avoid problems, because we don't mandate that RAM-to-device
MemoryRegion boundaries happen on page boundaries. I think you
either need to do this at a low enough level that you can
look at what MemoryRegion types you're reading from, or you need
to do it byte at a time.
> + chunk = MIN(chunk, sizeof(buf));
> +
> + if (cpu_memory_rw_debug(env_cpu(env), addr, buf, chunk, 0)) {
> + return -1;
> + }
> + p = memchr(buf, 0, chunk);
> + if (p) {
> + len += p - buf;
> + return len <= INT32_MAX ? (ssize_t)len : -1;
> + }
> +
> + len += chunk;
> + addr += chunk;
> + if (len > INT32_MAX) {
> + return -1;
> + }
> + }
> +}
thanks
-- PMM
- [PATCH v2 00/74] semihosting cleanup, Richard Henderson, 2022/05/03
- [PATCH v2 01/74] semihosting: Move exec/softmmu-semi.h to semihosting/softmmu-uaccess.h, Richard Henderson, 2022/05/03
- [PATCH v2 02/74] semihosting: Return failure from softmmu-uaccess.h functions, Richard Henderson, 2022/05/03
- [PATCH v2 04/74] semihosting: Move softmmu-uaccess.h functions out of line, Richard Henderson, 2022/05/03
- [PATCH v2 03/74] semihosting: Improve condition for config.c and console.c, Richard Henderson, 2022/05/03
- [PATCH v2 05/74] semihosting: Add target_strlen for softmmu-uaccess.h, Richard Henderson, 2022/05/03
- Re: [PATCH v2 05/74] semihosting: Add target_strlen for softmmu-uaccess.h,
Peter Maydell <=
- [PATCH v2 06/74] semihosting: Simplify softmmu_lock_user_string, Richard Henderson, 2022/05/03
- [PATCH v2 07/74] semihosting: Split out guestfd.c, Richard Henderson, 2022/05/03
- [PATCH v2 08/74] semihosting: Generalize GuestFDFeatureFile, Richard Henderson, 2022/05/03
- [PATCH v2 09/74] semihosting: Return void from do_common_semihosting, Richard Henderson, 2022/05/03