* [PATCH] linux-user: fix handling of cpu mask in riscv_hwprobe syscall
@ 2025-03-04 16:52 Andreas Schwab
2025-03-10 23:10 ` Alistair Francis
0 siblings, 1 reply; 3+ messages in thread
From: Andreas Schwab @ 2025-03-04 16:52 UTC (permalink / raw)
To: Laurent Vivier; +Cc: Palmer Dabbelt, Robbin Ehn, Alistair Francis, qemu-devel
The third argument of the riscv_hwprobe syscall contains the size of the
cpu mask in bytes, not bits.
Signed-off-by: Andreas Schwab <schwab@suse.de>
---
linux-user/syscall.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/linux-user/syscall.c b/linux-user/syscall.c
index a22a5df8cc..4cc1a31d0d 100644
--- a/linux-user/syscall.c
+++ b/linux-user/syscall.c
@@ -9120,16 +9120,16 @@ static void risc_hwprobe_fill_pairs(CPURISCVState *env,
static int cpu_set_valid(abi_long arg3, abi_long arg4)
{
- int ret, i, tmp;
+ int ret, i;
size_t host_mask_size, target_mask_size;
unsigned long *host_mask;
/*
* cpu_set_t represent CPU masks as bit masks of type unsigned long *.
- * arg3 contains the cpu count.
+ * arg3 contains the size of the cpu mask.
*/
- tmp = (8 * sizeof(abi_ulong));
- target_mask_size = ((arg3 + tmp - 1) / tmp) * sizeof(abi_ulong);
+ target_mask_size = (arg3 + (sizeof(abi_ulong) - 1)) &
+ ~(sizeof(abi_ulong) - 1);
host_mask_size = (target_mask_size + (sizeof(*host_mask) - 1)) &
~(sizeof(*host_mask) - 1);
--
2.48.1
--
Andreas Schwab, SUSE Labs, schwab@suse.de
GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE 1748 E4D4 88E3 0EEA B9D7
"And now for something completely different."
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH] linux-user: fix handling of cpu mask in riscv_hwprobe syscall
2025-03-04 16:52 [PATCH] linux-user: fix handling of cpu mask in riscv_hwprobe syscall Andreas Schwab
@ 2025-03-10 23:10 ` Alistair Francis
2025-03-11 7:28 ` Robbin Ehn
0 siblings, 1 reply; 3+ messages in thread
From: Alistair Francis @ 2025-03-10 23:10 UTC (permalink / raw)
To: Andreas Schwab
Cc: Laurent Vivier, Palmer Dabbelt, Robbin Ehn, Alistair Francis,
qemu-devel
On Wed, Mar 5, 2025 at 2:54 AM Andreas Schwab <schwab@suse.de> wrote:
>
> The third argument of the riscv_hwprobe syscall contains the size of the
> cpu mask in bytes, not bits.
>
> Signed-off-by: Andreas Schwab <schwab@suse.de>
Richard sent a v2 that I have applied:
https://patchew.org/QEMU/20250308225902.1208237-3-richard.henderson@linaro.org/
Alistair
> ---
> linux-user/syscall.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/linux-user/syscall.c b/linux-user/syscall.c
> index a22a5df8cc..4cc1a31d0d 100644
> --- a/linux-user/syscall.c
> +++ b/linux-user/syscall.c
> @@ -9120,16 +9120,16 @@ static void risc_hwprobe_fill_pairs(CPURISCVState *env,
>
> static int cpu_set_valid(abi_long arg3, abi_long arg4)
> {
> - int ret, i, tmp;
> + int ret, i;
> size_t host_mask_size, target_mask_size;
> unsigned long *host_mask;
>
> /*
> * cpu_set_t represent CPU masks as bit masks of type unsigned long *.
> - * arg3 contains the cpu count.
> + * arg3 contains the size of the cpu mask.
> */
> - tmp = (8 * sizeof(abi_ulong));
> - target_mask_size = ((arg3 + tmp - 1) / tmp) * sizeof(abi_ulong);
> + target_mask_size = (arg3 + (sizeof(abi_ulong) - 1)) &
> + ~(sizeof(abi_ulong) - 1);
> host_mask_size = (target_mask_size + (sizeof(*host_mask) - 1)) &
> ~(sizeof(*host_mask) - 1);
>
> --
> 2.48.1
>
>
> --
> Andreas Schwab, SUSE Labs, schwab@suse.de
> GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE 1748 E4D4 88E3 0EEA B9D7
> "And now for something completely different."
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] linux-user: fix handling of cpu mask in riscv_hwprobe syscall
2025-03-10 23:10 ` Alistair Francis
@ 2025-03-11 7:28 ` Robbin Ehn
0 siblings, 0 replies; 3+ messages in thread
From: Robbin Ehn @ 2025-03-11 7:28 UTC (permalink / raw)
To: Alistair Francis
Cc: Andreas Schwab, Laurent Vivier, Palmer Dabbelt, Alistair Francis,
qemu-devel
On Tue, Mar 11, 2025 at 12:10 AM Alistair Francis <alistair23@gmail.com> wrote:
>
> On Wed, Mar 5, 2025 at 2:54 AM Andreas Schwab <schwab@suse.de> wrote:
> >
> > The third argument of the riscv_hwprobe syscall contains the size of the
> > cpu mask in bytes, not bits.
> >
> > Signed-off-by: Andreas Schwab <schwab@suse.de>
>
> Richard sent a v2 that I have applied:
>
> https://patchew.org/QEMU/20250308225902.1208237-3-richard.henderson@linaro.org/
Hey, thanks for fixing!
Acked-by: Robbin Ehn <rehn@rivosinc.com>
>
> Alistair
>
> > ---
> > linux-user/syscall.c | 8 ++++----
> > 1 file changed, 4 insertions(+), 4 deletions(-)
> >
> > diff --git a/linux-user/syscall.c b/linux-user/syscall.c
> > index a22a5df8cc..4cc1a31d0d 100644
> > --- a/linux-user/syscall.c
> > +++ b/linux-user/syscall.c
> > @@ -9120,16 +9120,16 @@ static void risc_hwprobe_fill_pairs(CPURISCVState *env,
> >
> > static int cpu_set_valid(abi_long arg3, abi_long arg4)
> > {
> > - int ret, i, tmp;
> > + int ret, i;
> > size_t host_mask_size, target_mask_size;
> > unsigned long *host_mask;
> >
> > /*
> > * cpu_set_t represent CPU masks as bit masks of type unsigned long *.
> > - * arg3 contains the cpu count.
> > + * arg3 contains the size of the cpu mask.
> > */
> > - tmp = (8 * sizeof(abi_ulong));
> > - target_mask_size = ((arg3 + tmp - 1) / tmp) * sizeof(abi_ulong);
> > + target_mask_size = (arg3 + (sizeof(abi_ulong) - 1)) &
> > + ~(sizeof(abi_ulong) - 1);
> > host_mask_size = (target_mask_size + (sizeof(*host_mask) - 1)) &
> > ~(sizeof(*host_mask) - 1);
> >
> > --
> > 2.48.1
> >
> >
> > --
> > Andreas Schwab, SUSE Labs, schwab@suse.de
> > GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE 1748 E4D4 88E3 0EEA B9D7
> > "And now for something completely different."
> >
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2025-03-11 7:29 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-03-04 16:52 [PATCH] linux-user: fix handling of cpu mask in riscv_hwprobe syscall Andreas Schwab
2025-03-10 23:10 ` Alistair Francis
2025-03-11 7:28 ` Robbin Ehn
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).