Re: [U-Boot] [PATCH] travis: Use kernel.org pre-built toolchain for riscv
From: Tom Rini [mailto:trini@konsulko.com] Sent: Tuesday, September 04, 2018 3:24 AM To: Bin Meng Cc: U-Boot Mailing List; Rick Jian-Zhi Chen(陳建志) Subject: Re: [PATCH] travis: Use kernel.org pre-built toolchain for riscv
On Mon, Sep 03, 2018 at 05:50:39PM +0800, Bin Meng wrote:
Hi Tom,
On Thu, Aug 30, 2018 at 9:12 PM Bin Meng bmeng.cn@gmail.com wrote:
This updates travis configuration to use kernel.org pre-built toolchain for riscv.
Signed-off-by: Bin Meng bmeng.cn@gmail.com
.travis.yml | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
Since this is travis-ci changes, can we take this for v2018.09? I've verified that it works well for current riscv build on travis-ci.org.
Seeing as things are still working before this change I would like
to see an ack
from the custodian, thanks!
-- Tom
Hi Bin
I think lt is a positive way to use kernel.org pre-built toolchain for code coverage. And very agree with that.
But there is a problem, it maybe encounter in the future. If I or someone try to add a private csr which the public toolchain may not recognize it. How shall it be overcomed ? Do you have any ideas ?
Rick
Hi Rick,
On Tue, Sep 4, 2018 at 1:36 PM Rick Chen rickchen36@gmail.com wrote:
From: Tom Rini [mailto:trini@konsulko.com] Sent: Tuesday, September 04, 2018 3:24 AM To: Bin Meng Cc: U-Boot Mailing List; Rick Jian-Zhi Chen(陳建志) Subject: Re: [PATCH] travis: Use kernel.org pre-built toolchain for riscv
On Mon, Sep 03, 2018 at 05:50:39PM +0800, Bin Meng wrote:
Hi Tom,
On Thu, Aug 30, 2018 at 9:12 PM Bin Meng bmeng.cn@gmail.com wrote:
This updates travis configuration to use kernel.org pre-built toolchain for riscv.
Signed-off-by: Bin Meng bmeng.cn@gmail.com
.travis.yml | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
Since this is travis-ci changes, can we take this for v2018.09? I've verified that it works well for current riscv build on travis-ci.org.
Seeing as things are still working before this change I would like
to see an ack
from the custodian, thanks!
-- Tom
Hi Bin
I think lt is a positive way to use kernel.org pre-built toolchain for code coverage. And very agree with that.
But there is a problem, it maybe encounter in the future. If I or someone try to add a private csr which the public toolchain may not recognize it.
I assume what you said "private" csr means "vendor-specific" csr that is not defined by the risc-v ISA?
How shall it be overcomed ? Do you have any ideas ?
I believe this shall be implemented properly by the toolchain itself. Or we hardcode the instruction for such "private" csr if there is no good solution from the toolchain side. Having a custom toolchain seems not a good idea.
Regards, Bin
Hi Rick,
On Tue, Sep 4, 2018 at 4:56 PM Bin Meng bmeng.cn@gmail.com wrote:
Hi Rick,
On Tue, Sep 4, 2018 at 1:36 PM Rick Chen rickchen36@gmail.com wrote:
From: Tom Rini [mailto:trini@konsulko.com] Sent: Tuesday, September 04, 2018 3:24 AM To: Bin Meng Cc: U-Boot Mailing List; Rick Jian-Zhi Chen(陳建志) Subject: Re: [PATCH] travis: Use kernel.org pre-built toolchain for riscv
On Mon, Sep 03, 2018 at 05:50:39PM +0800, Bin Meng wrote:
Hi Tom,
On Thu, Aug 30, 2018 at 9:12 PM Bin Meng bmeng.cn@gmail.com wrote:
This updates travis configuration to use kernel.org pre-built toolchain for riscv.
Signed-off-by: Bin Meng bmeng.cn@gmail.com
.travis.yml | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
Since this is travis-ci changes, can we take this for v2018.09? I've verified that it works well for current riscv build on travis-ci.org.
Seeing as things are still working before this change I would like
to see an ack
from the custodian, thanks!
-- Tom
Hi Bin
I think lt is a positive way to use kernel.org pre-built toolchain for code coverage. And very agree with that.
But there is a problem, it maybe encounter in the future. If I or someone try to add a private csr which the public toolchain may not recognize it.
I assume what you said "private" csr means "vendor-specific" csr that is not defined by the risc-v ISA?
How shall it be overcomed ? Do you have any ideas ?
I believe this shall be implemented properly by the toolchain itself. Or we hardcode the instruction for such "private" csr if there is no good solution from the toolchain side. Having a custom toolchain seems not a good idea.
I did not see a confirmative response on whether this patch is OK. Can you please comment, and if no issue I hope this can catch up the train of upcoming release :)
Regards, Bin
Bin Meng bmeng.cn@gmail.com 於 2018年9月6日 週四 上午9:43寫道：
Hi Rick,
On Tue, Sep 4, 2018 at 4:56 PM Bin Meng bmeng.cn@gmail.com wrote:
Hi Rick,
On Tue, Sep 4, 2018 at 1:36 PM Rick Chen rickchen36@gmail.com wrote:
From: Tom Rini [mailto:trini@konsulko.com] Sent: Tuesday, September 04, 2018 3:24 AM To: Bin Meng Cc: U-Boot Mailing List; Rick Jian-Zhi Chen(陳建志) Subject: Re: [PATCH] travis: Use kernel.org pre-built toolchain for riscv
On Mon, Sep 03, 2018 at 05:50:39PM +0800, Bin Meng wrote:
Hi Tom,
On Thu, Aug 30, 2018 at 9:12 PM Bin Meng bmeng.cn@gmail.com wrote:
This updates travis configuration to use kernel.org pre-built toolchain for riscv.
Signed-off-by: Bin Meng bmeng.cn@gmail.com
.travis.yml | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
Since this is travis-ci changes, can we take this for v2018.09? I've verified that it works well for current riscv build on travis-ci.org.
Seeing as things are still working before this change I would like
to see an ack
from the custodian, thanks!
-- Tom
Hi Bin
I think lt is a positive way to use kernel.org pre-built toolchain for code coverage. And very agree with that.
But there is a problem, it maybe encounter in the future. If I or someone try to add a private csr which the public toolchain may not recognize it.
I assume what you said "private" csr means "vendor-specific" csr that is not defined by the risc-v ISA?
How shall it be overcomed ? Do you have any ideas ?
I believe this shall be implemented properly by the toolchain itself. Or we hardcode the instruction for such "private" csr if there is no good solution from the toolchain side. Having a custom toolchain seems not a good idea.
Hi Bin
Agree with you, maybe I shall consider to isolate the vendor-specific code with CONFIG_XXX.
I did not see a confirmative response on whether this patch is OK. Can you please comment, and if no issue I hope this can catch up the train of upcoming release :)
Hi Tom
I think it is OK to pull this change into the master. Hope everything goes fine :)
Rick
Regards, Bin
participants (2)
-
Bin Meng -
Rick Chen