On Mon, Dec 09, 2013 at 11:23:14AM -0800, Richard Henderson wrote:
On 12/08/2013 05:48 AM, Aurelien Jarno wrote:
It looks fine to me, though I don't know if it has any impact on the
performances.
I couldn't measure any performance difference at all.
Ok great. Still in the long term we
On 12/08/2013 05:48 AM, Aurelien Jarno wrote:
It looks fine to me, though I don't know if it has any impact on the
performances.
I couldn't measure any performance difference at all.
But for me, the improvement in debugging for the translators is important.
r~
On Thu, Sep 19, 2013 at 12:58:37PM -0700, Richard Henderson wrote:
We previously allocated 32-bits per temp for the next_free_temp entry.
We now allocate 4 bits per temp across the 4 bitmaps.
Using a linked list meant that if a translator is tweeked, resulting in
temps being freed in a
On 19 September 2013 20:58, Richard Henderson r...@twiddle.net wrote:
We previously allocated 32-bits per temp for the next_free_temp entry.
We now allocate 4 bits per temp across the 4 bitmaps.
Using a linked list meant that if a translator is tweeked, resulting in
temps being freed in a
Ping.
r~
On 11/19/2013 07:57 PM, Richard Henderson wrote:
Ping?
r~
On 09/20/2013 05:58 AM, Richard Henderson wrote:
We previously allocated 32-bits per temp for the next_free_temp entry.
We now allocate 4 bits per temp across the 4 bitmaps.
Using a linked list meant that if a
Ping?
r~
On 09/20/2013 05:58 AM, Richard Henderson wrote:
We previously allocated 32-bits per temp for the next_free_temp entry.
We now allocate 4 bits per temp across the 4 bitmaps.
Using a linked list meant that if a translator is tweeked, resulting in
temps being freed in a different
We previously allocated 32-bits per temp for the next_free_temp entry.
We now allocate 4 bits per temp across the 4 bitmaps.
Using a linked list meant that if a translator is tweeked, resulting in
temps being freed in a different order, that would have follow-on effects
throughout the TB. Always
