Hi,
memory-barriers.txt says:
[on "store tearing"]
"In fact, a recent bug (since fixed) caused GCC to incorrectly use
this optimization in a volatile store.".
I was wondering if you could help me retrieve some reference/discussions
about this?
Thanks,
Andrea
Hopefully, with Paul's proper email address this time,
Andrea
On Mon, Oct 29, 2018 at 12:06:27AM +0100, Andrea Parri wrote:
> Hi,
>
> memory-barriers.txt says:
>
> [on "store tearing"]
>
> "In fact, a recent bug (since fixed) caused GCC to incorrectly use
> this optimization in a volatile store.".
>
> I was wondering if you could help me retrieve some reference/discussions
> about this?
>
> Thanks,
> Andrea
On Mon, Oct 29, 2018 at 12:10:03AM +0100, Andrea Parri wrote:
> Hopefully, with Paul's proper email address this time,
>
> Andrea
>
> On Mon, Oct 29, 2018 at 12:06:27AM +0100, Andrea Parri wrote:
> > Hi,
> >
> > memory-barriers.txt says:
> >
> > [on "store tearing"]
> >
> > "In fact, a recent bug (since fixed) caused GCC to incorrectly use
> > this optimization in a volatile store.".
> >
> > I was wondering if you could help me retrieve some reference/discussions
> > about this?
This was quite some time ago, but it involved a 32-bit volatile store
of a constant such as 0x10001. The machine in question had a narrow
store-immediate instruction, so the compiler emitted a pair of 16-bit
store-immediate instructions. This bug was fixed, though only after
significant screaming and shouting.
Thanx, Paul
On Sun, Oct 28, 2018 at 06:20:42PM -0700, Paul E. McKenney wrote:
> On Mon, Oct 29, 2018 at 12:10:03AM +0100, Andrea Parri wrote:
> > Hopefully, with Paul's proper email address this time,
> >
> > Andrea
> >
> > On Mon, Oct 29, 2018 at 12:06:27AM +0100, Andrea Parri wrote:
> > > Hi,
> > >
> > > memory-barriers.txt says:
> > >
> > > [on "store tearing"]
> > >
> > > "In fact, a recent bug (since fixed) caused GCC to incorrectly use
> > > this optimization in a volatile store.".
> > >
> > > I was wondering if you could help me retrieve some reference/discussions
> > > about this?
>
> This was quite some time ago, but it involved a 32-bit volatile store
> of a constant such as 0x10001. The machine in question had a narrow
> store-immediate instruction, so the compiler emitted a pair of 16-bit
> store-immediate instructions. This bug was fixed, though only after
> significant screaming and shouting.
That does sound like an interesting discussion. ;D Thanks for the info,
Andrea
>
> Thanx, Paul
>
On Mon, Oct 29, 2018 at 2:21 AM Paul E. McKenney <[email protected]> wrote:
>
> On Mon, Oct 29, 2018 at 12:10:03AM +0100, Andrea Parri wrote:
> > Hopefully, with Paul's proper email address this time,
> >
> > Andrea
> >
> > On Mon, Oct 29, 2018 at 12:06:27AM +0100, Andrea Parri wrote:
> > > Hi,
> > >
> > > memory-barriers.txt says:
> > >
> > > [on "store tearing"]
> > >
> > > "In fact, a recent bug (since fixed) caused GCC to incorrectly use
> > > this optimization in a volatile store.".
> > >
> > > I was wondering if you could help me retrieve some reference/discussions
> > > about this?
>
> This was quite some time ago, but it involved a 32-bit volatile store
> of a constant such as 0x10001. The machine in question had a narrow
> store-immediate instruction, so the compiler emitted a pair of 16-bit
> store-immediate instructions. This bug was fixed, though only after
> significant screaming and shouting.
A related issue I remember was on ARMv5 (an architecture without
unaligned access) where a function like )not sure if this specific
one triggers it, but something like it did)
struct my_registers {
u32 a;
u32 b;
u32 c;
} __attribute__((packed));
#define __raw_writel(p, v) do { (volatile u32 __iomem *)(p) = (v); } while (0)
void my_write_a(struct my_registers __iomem *r, u32 val)
{
__raw_writel(&r->a, val);
}
The above is undefined behavior because we cast from an unaligned
data type to a 32-bit aligned type, and gcc resolved this by turning the
intended 32-bit store into a set of 8 bit stores. We worked around this
by changing __raw_writel() into a inline assembly that always uses a
32-bit store.
Arnd
On Mon, Oct 29, 2018 at 10:23:07AM +0100, Arnd Bergmann wrote:
> On Mon, Oct 29, 2018 at 2:21 AM Paul E. McKenney <[email protected]> wrote:
> >
> > On Mon, Oct 29, 2018 at 12:10:03AM +0100, Andrea Parri wrote:
> > > Hopefully, with Paul's proper email address this time,
> > >
> > > Andrea
> > >
> > > On Mon, Oct 29, 2018 at 12:06:27AM +0100, Andrea Parri wrote:
> > > > Hi,
> > > >
> > > > memory-barriers.txt says:
> > > >
> > > > [on "store tearing"]
> > > >
> > > > "In fact, a recent bug (since fixed) caused GCC to incorrectly use
> > > > this optimization in a volatile store.".
> > > >
> > > > I was wondering if you could help me retrieve some reference/discussions
> > > > about this?
> >
> > This was quite some time ago, but it involved a 32-bit volatile store
> > of a constant such as 0x10001. The machine in question had a narrow
> > store-immediate instruction, so the compiler emitted a pair of 16-bit
> > store-immediate instructions. This bug was fixed, though only after
> > significant screaming and shouting.
>
> A related issue I remember was on ARMv5 (an architecture without
> unaligned access) where a function like )not sure if this specific
> one triggers it, but something like it did)
>
> struct my_registers {
> u32 a;
> u32 b;
> u32 c;
> } __attribute__((packed));
> #define __raw_writel(p, v) do { (volatile u32 __iomem *)(p) = (v); } while (0)
> void my_write_a(struct my_registers __iomem *r, u32 val)
> {
> __raw_writel(&r->a, val);
> }
>
> The above is undefined behavior because we cast from an unaligned
> data type to a 32-bit aligned type, and gcc resolved this by turning the
> intended 32-bit store into a set of 8 bit stores. We worked around this
> by changing __raw_writel() into a inline assembly that always uses a
> 32-bit store.
I had either missed or forgotten this one, nice example of store tearing!
Thanx, Paul