Hi!
On Sat, 1 Nov 2014 13:11:29 +0100, Bernd Schmidt ber...@codesourcery.com
wrote:
[nvptx mkoffload]
To support the --enable-offload-targets=nvptx-none=[install directory]
configuration option, I committed the following to trunk in r220782 (and
filed https://gcc.gnu.org/PR65097):
commit
Hi!
On Wed, 11 Feb 2015 16:20:51 +0100, Bernd Schmidt ber...@codesourcery.com
wrote:
On 02/11/2015 03:44 PM, Thomas Schwinge wrote:
Note the global cold/hot labels. This confuses mkoffload, and it runs
into a busy loop due to what I understand to be a bug in skipping of
strange tokens,
Hi!
On Wed, 28 Jan 2015 18:05:25 +0100, I wrote:
On Sat, 1 Nov 2014 13:11:29 +0100, Bernd Schmidt ber...@codesourcery.com
wrote:
I'm sending this for reference more than anything else - this is the
patch that adds the target support for offloading to the nvptx port. It
depends on the
Hi!
On Wed, 28 Jan 2015 20:59:45 +0300, Ilya Verbin iver...@gmail.com wrote:
On 28 Jan 18:05, Thomas Schwinge wrote:
+ fprintf (out, #define PTX_ID 1\n);
+ fprintf (out, static __attribute__((constructor)) void init
(void)\n{\n);
+ fprintf (out, GOMP_offload_register
On 02/11/2015 03:44 PM, Thomas Schwinge wrote:
Note the global cold/hot labels. This confuses mkoffload, and it runs
into a busy loop due to what I understand to be a bug in skipping of
strange tokens, cited above, which such global labels would fall under.
Here is what might be a fix for this
Hi Bernd!
On Sat, 1 Nov 2014 13:11:29 +0100, Bernd Schmidt ber...@codesourcery.com
wrote:
I'm sending this for reference more than anything else - this is the
patch that adds the target support for offloading to the nvptx port.
(I committed this in r220209.)
I actually expect this to
Hi!
On Wed, 28 Jan 2015 18:05:25 +0100, I wrote:
On Sat, 1 Nov 2014 13:11:29 +0100, Bernd Schmidt ber...@codesourcery.com
wrote:
I'm sending this for reference more than anything else - this is the
patch that adds the target support for offloading to the nvptx port. It
depends on the
Hi!
On Wed, 11 Feb 2015 15:44:26 +0100, I wrote:
If -freorder-blocks-and-partition is active, this results in PTX code
such as:
// BEGIN PREAMBLE
.version3.1
.target sm_30
.address_size 64
// END PREAMBLE
$LCOLDB0:
On 28 Jan 18:05, Thomas Schwinge wrote:
+ fprintf (out, #define PTX_ID 1\n);
+ fprintf (out, static __attribute__((constructor)) void init (void)\n{\n);
+ fprintf (out, GOMP_offload_register (__OPENMP_TARGET__, PTX_ID,\n);
The file include/gomp-constants.h already contains:
#define
Hi!
On Sat, 1 Nov 2014 13:11:29 +0100, Bernd Schmidt ber...@codesourcery.com
wrote:
I'm sending this for reference more than anything else - this is the
patch that adds the target support for offloading to the nvptx port. It
depends on the other offloading patches Ilya is currently
I'm sending this for reference more than anything else - this is the
patch that adds the target support for offloading to the nvptx port. It
depends on the other offloading patches Ilya is currently submitting.
I actually expect this to change a little in the near future; the nvptx
mkoffload
11 matches
Mail list logo
