Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

>> We don't need it; this is ancient history. Nonetheless, feel free to >> be consistent with the old style. I'll approve it. I understand. Thank you, Andrew! On 2020/07/30 21:13, Yasumasa Suenaga wrote: Hi Koichi, Looks good your change. BTW have you got a sponsor? I will sponsor you if you

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

Hi Koichi, Looks good your change. BTW have you got a sponsor? I will sponsor you if you do not have yet. Thanks, Yasumasa On 2020/07/30 18:45, Andrew Haley wrote: On 7/30/20 6:58 AM, Koichi Sakata wrote: I just formatted the patch to be same as the other code. Moreover, I saw the

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

On 7/30/20 6:58 AM, Koichi Sakata wrote: > I just formatted the patch to be same as the other code. > > Moreover, I saw the following code in k_standard.c. > > if (_LIB_VERSION == _SVID_) >exc.retval = zero; > else >exc.retval = zero/zero; > > Do we need to do like that in this patch?

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

Hi Andrew, I just formatted the patch to be same as the other code. Moreover, I saw the following code in k_standard.c. if (_LIB_VERSION == _SVID_) exc.retval = zero; else exc.retval = zero/zero; Do we need to do like that in this patch? It seems to be related to the old matherr(), but

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

Hi Andrew, Thank you for teaching me. I made the patch that returns NaN. Could you please sponsor it? In this case I think it is better to fix k_standard.c itself rather than ignoring the warning. However, I was able to learn there are some good ways to suppress warnings. So I'd like to thank

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

On 17/07/2020 12:26, Koichi Sakata wrote: > > > You'll need to find a reviewer that understands what that > > method is supposed to do in that case, that's not me ;-) > > I understand. This ML is suitable for finding a reviewer, isn't it? > Or, there is another way. We can avoid the error by

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

On 7/17/20 6:48 AM, Yasumasa Suenaga wrote: Hi Koichi, On 2020/07/17 20:26, Koichi Sakata wrote: Hi Daniel,  > The changes to NetworkInterface.c look good to me. Thank you.  > You'll need to find a reviewer that understands what that  > method is supposed to do in that case, that's not

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

If you are going to change makefiles you need to get review on build-...@openjdk.java.net :) David On 17/07/2020 11:48 pm, Yasumasa Suenaga wrote: Hi Koichi, On 2020/07/17 20:26, Koichi Sakata wrote: Hi Daniel,  > The changes to NetworkInterface.c look good to me. Thank you.  > You'll

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

Hi Koichi, On 2020/07/17 20:26, Koichi Sakata wrote: Hi Daniel, > The changes to NetworkInterface.c look good to me. Thank you. > You'll need to find a reviewer that understands what that > method is supposed to do in that case, that's not me ;-) I understand. This ML is suitable for

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

Hi Daniel, > The changes to NetworkInterface.c look good to me. Thank you. > You'll need to find a reviewer that understands what that > method is supposed to do in that case, that's not me ;-) I understand. This ML is suitable for finding a reviewer, isn't it? Or, there is another way. We

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

Hi Koichi, On 13/07/2020 08:03, Koichi Sakata wrote: > I understand that. I respect your idea. > I fixed the patch as follows. The changes to NetworkInterface.c look good to me. > > By the way, k_standard.c still remains. Is there a way to proceed with it? You'll need to find a reviewer that

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

Hi Daniel, On 2020/07/09 18:16, Daniel Fuchs wrote: > Hi Koichi, > > If you don't mind I'd prefer to keep the simple approach > that will fix the warning without changing the behavior > of the function. I understand that. I respect your idea. I fixed the patch as follows. By the way,

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

Hi Koichi, If you don't mind I'd prefer to keep the simple approach that will fix the warning without changing the behavior of the function. memset((char *), 0, sizeof(if2)); - strncpy(if2.ifr_name, name, sizeof(if2.ifr_name) - 1); + strncpy(if2.ifr_name, name,

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

Hi Daniel, Thank you for your response. > > +if (sizeof(if2.ifr_name) < sizeof(name)) { > > +return -1; > > +} > > If I'm not mistaken `sizeof(name)` where name is a const char* > will always be 8 (on 64 bits architecture) - so this is probably > not doing what you want. I'm

Re: Build error with GCC 10 in NetworkInterface.c and k_standard.c

Hi, I will not comment on the changes to libfdlibm/k_standard.c Concerning NetworkInterface.c I believe the proposed changes are incorrect - and I do not see the issue with the current code. > +if (sizeof(if2.ifr_name) < sizeof(name)) { > +return -1; > +} If I'm not mistaken

Build error with GCC 10 in NetworkInterface.c and k_standard.c

Hi all, (I've sent a similar e-mail before to this ML, but I extract the part only related to core-libs-dev ML from the previous one.) I tried to build OpenJDK fastdebug with GCC 10.1 on Ubuntu 18.04, but I saw some compiler warnings as follows: