So use PRIx64, as you're supposed to do.
Or, in fact, change this to a tracepoint.
yes. I will change all of them. It should be useful for everyone.
C.
On 9/17/21 7:10 AM, Cédric Le Goater wrote:
On 9/17/21 16:00, Luis Fernando Fujita Pires wrote:
From: Cédric Le Goater
+ target_long signed_value;
+ target_long signed_decr;
Since these values will be the results of sextract64, it's probably better to
define them as int64_t.
but then
On 9/17/21 16:00, Luis Fernando Fujita Pires wrote:
From: Cédric Le Goater
+target_long signed_value;
+target_long signed_decr;
Since these values will be the results of sextract64, it's probably better to
define them as int64_t.
but then it breaks the code doing the logging on PPC3
From: Cédric Le Goater
> >> +target_long signed_value;
> >> +target_long signed_decr;
> >
> > Since these values will be the results of sextract64, it's probably better
> > to
> define them as int64_t.
>
> but then it breaks the code doing the logging on PPC32 targets :/
You mean here?
Hello,
On 9/16/21 20:22, Luis Fernando Fujita Pires wrote:
Hi Cédric,
These changes look good and seem to replicate the exact behavior we had before,
while fixing the 64-bit dec from MicroWatt.
Yes. That's the goal. I would like to minimize the possible impact on other
platforms. I checked M
Hi Cédric,
These changes look good and seem to replicate the exact behavior we had before,
while fixing the 64-bit dec from MicroWatt.
A few notes, in case they help others, too:
According to the Power ISA:
When not in Large Decrementer mode:
- the maximum positive value for the decrement
The current way the mask is built can overflow with a 64-bit decrementer.
Use sextract64() to extract the signed values and remove the logic to
handle negative values which has become useless.
Cc: Luis Fernando Fujita Pires
Fixes: a8dafa525181 ("target/ppc: Implement large decrementer support for
