/
struct rte_eth_txmode {
+ int reserved[]; /* force the size of struct to be 0 */
};
/**
--
1.7.10.4
-- next part --
An HTML attachment was scrubbed...
URL:
<http://dpdk.org/ml/archives/dev/attachments/20130612/cc28c4c3/attachment.html>
All,
Following many requests, 6WIND is releasing the source code of our
librte_pmd_vmxnet3 under a BSD license.
It is available at:
http://dpdk.org/browse/vmxnet3-usermap/snapshot/vmxnet3-usermap-1.0.tar.gz
git://dpdk.org/vmxnet3-usermap
It allows running any DPDK application i
The new release can be downloaded here:
http://dpdk.org/browse/dpdk/tag/?id=v1.2.3r3
Changelog
- fixes for:
* write-back threshold for 82576
* forbidden empty struct with C++
- enhancements:
* option to load external drivers
12/06/2013 17:35, Olivier MATZ :
> On 06/12/2013 05:24 PM, Thomas Monjalon wrote:
> > From: Damien Millescamps
> >
> > Extract rte_mem_virt2phy() from find_physaddr().
>
> Acked-by: Olivier Matz
pushed
--
Thomas
On 06/12/2013 05:24 PM, Thomas Monjalon wrote:
> From: Damien Millescamps
>
> Extract rte_mem_virt2phy() from find_physaddr().
Acked-by: Olivier Matz
From: Damien Millescamps
Extract rte_mem_virt2phy() from find_physaddr().
rte_mem_virt2phy() permits to obtain the physical address of any
virtual address mapped to the current process calling this function.
Note that this function is very slow and shouldn't be called
after initialization to avo
12/06/2013 16:46, Olivier MATZ :
> On 06/12/2013 04:30 PM, Thomas Monjalon wrote:
> > The previous fix was for g++ but is broken with gcc:
> > error: flexible array member in otherwise empty struct
>
> Sorry Thomas for the previous one, I should have tested it.
No problem, I should also bette
On 12.06.2013 16:54, Olivier MATZ wrote:
>> Unless I'm missing something, testpmd still needs
>> an explicit rte_pmd_foo_init() call. It's of course not difficult to
>> sprinkle some __attribute__((constructor))'s into the pmd's
>
> Yes, you need to do it in the pmd's constructor.
Ok, I thought it
12/06/2013 13:55, Olivier MATZ :
> On 06/12/2013 09:51 AM, Thomas Monjalon wrote:
> > From: Damien Millescamps
> >
> > Add an option to specify libraries to be loaded before probing the PCI.
>
> Looks good to me.
pushed
--
Thomas
Hi Antti,
> > For instance, testpmd -d librte_pmd_xxx.so can be used to enable xxx
> driver
> > support on testpmd without any recompilation of testpmd.
>
> What's the use case?
Load any other pmd driver dynamically (ex: mellanox driver)
> Unless I'm missing something, testpmd still needs
> an
On 06/12/2013 04:30 PM, Thomas Monjalon wrote:
> The previous fix was for g++ but is broken with gcc:
> error: flexible array member in otherwise empty struct
> See http://gcc.gnu.org/onlinedocs/gcc-4.2.2/gcc/Zero-Length.html
>
> Let's fix it with #ifdef __cplusplus.
Sorry Thomas for the pre
The previous fix was for g++ but is broken with gcc:
error: flexible array member in otherwise empty struct
See http://gcc.gnu.org/onlinedocs/gcc-4.2.2/gcc/Zero-Length.html
Let's fix it with #ifdef __cplusplus.
Signed-off-by: Thomas Monjalon
---
lib/librte_ether/rte_ethdev.h |2 ++
12/06/2013 14:23, lxu :
> Great! Thanks for the quick feedback.
>
> -- Original --
> From: "Olivier MATZ";
>
> I agree with your solution. Do you approve the following patch ?
>
> From: lxu
> Date: Wed, 12 Jun 2013 09:32:30 +0200
> Subject: [PATCH] ethdev: force
On 12.06.2013 09:51, Thomas Monjalon wrote:
> From: Damien Millescamps
>
> Add an option to specify libraries to be loaded before probing the PCI.
>
> For instance, testpmd -d librte_pmd_xxx.so can be used to enable xxx
driver
> support on testpmd without any recompilation of testpmd.
What
Hello,
On 06/12/2013 09:51 AM, Thomas Monjalon wrote:
> From: Damien Millescamps
>
> Add an option to specify libraries to be loaded before probing the PCI.
Looks good to me.
Hi Thomas,
Please find some comments below.
> +/**
> + * Get physical address of any mapped virtual address in the current
process.
> + * It is found by browsing the /proc/self/pagemap special file.
> + * The page won't be swappable anymore.
> + */
> +phys_addr_t rte_mem_virt2phy(const voi
xmode { int
reserve[]; }; /* sizeof(struct rte_eth_txmode) == 0 in gcc and g++ */
-- next part --
An HTML attachment was scrubbed...
URL:
<http://dpdk.org/ml/archives/dev/attachments/20130612/08df08bd/attachment.html>
05/06/2013 16:36, Vincent JARDIN :
> On 30/05/2013 19:12, Stephen Hemminger wrote:
> > Signed-off-by: Stephen Hemminger
>
> Reviewed-by: Vincent Jardin
applied with modified title:
log: add ability to override syslog parameters
--
Thomas
05/06/2013 16:34, Vincent JARDIN :
> On 30/05/2013 19:12, Stephen Hemminger wrote:
> > Signed-off-by: Stephen Hemminger
>
> Reviewed-by: Vincent Jardin
>
> It does not hurt to move both to a cold section. TODO: some other init
> and setup functions should be cold too.
applied with modified titl
05/06/2013 16:25, Vincent JARDIN :
> On 30/05/2013 19:12, Stephen Hemminger wrote:
> > Signed-off-by: Stephen Hemminger
>
> Reviewed-by: Vincent Jardin
>
> It is a nice to have for performance since HPET is quite slow.
applied
--
Thomas
05/06/2013 16:22, Vincent JARDIN :
> Reviewed-by: Vincent Jardin
>
> It looks good. Thank you.
applied
--
Thomas
-- next part --
An HTML attachment was scrubbed...
URL:
<http://dpdk.org/ml/archives/dev/attachments/20130612/1304bd9f/attachment.html>
Hello,
> Can we change the struct rte_eth_txmode ?
>
> Such as :
>
> struct rte_eth_txmode { int reserve[]; }; /* sizeof(struct rte_eth_txmode)
> == 0 in gcc and g++ */
I agree with your solution. Do you approve the following patch ?
From: lxu
Date: Wed, 12 Jun 2013 09:32:30 +0200
Subject: [
From: Damien Millescamps
Add an option to specify libraries to be loaded before probing the PCI.
For instance, testpmd -d librte_pmd_xxx.so can be used to enable xxx driver
support on testpmd without any recompilation of testpmd.
Signed-off-by: Damien Millescamps
Signed-off-by: Jean-Mickael Gu
From: Damien Millescamps
Extract rte_mem_virt2phy() from find_physaddr().
This function permits to obtain the physical address of any
virtual address mapped to the current process calling this function.
Note that this function is very slow and shouldn't be called
after initialization to avoid a
25 matches
Mail list logo
