<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"> <head> <meta http-equiv="Content-Type" content="text/html; charset=us-ascii"> <meta name="Generator" content="Microsoft Word 15 (filtered medium)"> <style></style> </head> <body lang="ZH-CN" link="#0563C1" vlink="#954F72" style="text-justify-trim:punctuation"> <div class="WordSection1"> There is also a condition that the hugepagesz is 1G.<o:p></o:p> If the hugepagesz is 2M, this problem cannot be repeated.<o:p></o:p> <o:p> </o:p> Thanks,<o:p></o:p> Yunjian<o:p></o:p> <o:p> </o:p> <div style="border:none;border-left:solid blue 1.5pt;padding:0cm 0cm 0cm 4.0pt"> <div> <div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0cm 0cm 0cm"> From: wangyunjian Sent: Monday, February 7, 2022 10:44 AM To: 'Honnappa Nagarahalli' <Honnappa.Nagarahalli@arm.com>; dev@dpdk.org; users@dpdk.org Cc: Feifei Wang <Feifei.Wang2@arm.com>; Ruifeng Wang <Ruifeng.Wang@arm.com>; Huangshaozhang <huangshaozhang@huawei.com>; dingxiaoxiong <dingxiaoxiong@huawei.com>; nd <nd@arm.com> Subject: RE: [dpdk-dev][dpdk-users] A problem about memory may not be all-zero allocated by rte_zmalloc_socket()<o:p></o:p> </div> </div> <o:p> </o:p> Hi, Honnappa<o:p></o:p> <o:p> </o:p> This problem is probability. Test case need to be executed multiple times.<o:p></o:p> <o:p> </o:p> The test steps and code are as follows:<o:p></o:p> <o:p> </o:p> /home/dpdk #./arm64-armv8a-linuxapp-gcc/app/dpdk-testpmd --legacy-mem -c 0xC -m 8192<o:p></o:p> <o:p> </o:p> app/test-pmd/testpmd.c | 14 ++++++++++++++<o:p></o:p> 1 file changed, 14 insertions(+)<o:p></o:p> <o:p> </o:p> diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c<o:p></o:p> index 55eb293cc0..3c127f9623 100644<o:p></o:p> --- a/app/test-pmd/testpmd.c<o:p></o:p> +++ b/app/test-pmd/testpmd.c<o:p></o:p> @@ -4251,6 +4251,20 @@ main(int argc, char** argv)<o:p></o:p> rte_stats_bitrate_reg(bitrate_data);<o:p></o:p> }<o:p></o:p> #endif<o:p></o:p> +<o:p></o:p> + printf("start test rte_zmalloc_socket\n");<o:p></o:p> + char *a;<o:p></o:p> + while((a = rte_zmalloc_socket(NULL, 1024 * 1024, 0, SOCKET_ID_ANY)) != NULL) {<o:p></o:p> + for (int i = 0; i < 1024 * 1024; i++) {<o:p></o:p> + if (a[i] != 0) {<o:p></o:p> + printf("a[%d] = %d\n",i,a[i]);<o:p></o:p> + }<o:p></o:p> + a[i] = 255; // This assignment is important. It can increase the probability.<o:p></o:p> + }<o:p></o:p> + }<o:p></o:p> + printf("end test rte_zmalloc_socket\n");<o:p></o:p> + return EXIT_SUCCESS;<o:p></o:p> +<o:p></o:p> #ifdef RTE_LIB_CMDLINE<o:p></o:p> if (strlen(cmdline_filename) != 0)<o:p></o:p> cmdline_read_from_file(cmdline_filename);<o:p></o:p> <o:p> </o:p> Thanks,<o:p></o:p> Yunjian<o:p></o:p> <o:p> </o:p> <div style="border:none;border-left:solid blue 1.5pt;padding:0cm 0cm 0cm 4.0pt"> <div> <div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0cm 0cm 0cm"> From: Honnappa Nagarahalli [<a href="mailto:Honnappa.Nagarahalli@arm.com">mailto:Honnappa.Nagarahalli@arm.com</a>] Sent: Monday, January 31, 2022 12:22 PM To: wangyunjian <<a href="mailto:wangyunjian@huawei.com">wangyunjian@huawei.com</a>>; <a href="mailto:dev@dpdk.org">dev@dpdk.org</a>; <a href="mailto:users@dpdk.org">users@dpdk.org</a> Cc: Feifei Wang <<a href="mailto:Feifei.Wang2@arm.com">Feifei.Wang2@arm.com</a>>; Ruifeng Wang <<a href="mailto:Ruifeng.Wang@arm.com">Ruifeng.Wang@arm.com</a>>; Huangshaozhang <<a href="mailto:huangshaozhang@huawei.com">huangshaozhang@huawei.com</a>>; dingxiaoxiong <<a href="mailto:dingxiaoxiong@huawei.com">dingxiaoxiong@huawei.com</a>>; Honnappa Nagarahalli <<a href="mailto:Honnappa.Nagarahalli@arm.com">Honnappa.Nagarahalli@arm.com</a>>; nd <<a href="mailto:nd@arm.com">nd@arm.com</a>> Subject: RE: [dpdk-dev][dpdk-users] A problem about memory may not be all-zero allocated by rte_zmalloc_socket()<o:p></o:p> </div> </div> <o:p> </o:p> Hi Yunjian,<o:p></o:p> That’s interesting. Is it possible to elaborate the use case or possibly provide the code snippet?<o:p></o:p> <o:p> </o:p> It is possible that it is a synchronization problem due to relaxed memory model that Arm architecture uses. There could be a barrier missing in the code.<o:p></o:p> <o:p> </o:p> Thanks,<o:p></o:p> Honnappa<o:p></o:p> <o:p> </o:p> <div style="border:none;border-left:solid blue 1.5pt;padding:0cm 0cm 0cm 4.0pt"> <div> <div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0cm 0cm 0cm"> From: wangyunjian <<a href="mailto:wangyunjian@huawei.com">wangyunjian@huawei.com</a>> Sent: Saturday, January 29, 2022 9:21 PM To: <a href="mailto:dev@dpdk.org">dev@dpdk.org</a>; <a href="mailto:users@dpdk.org"> users@dpdk.org</a> Cc: Feifei Wang <<a href="mailto:Feifei.Wang2@arm.com">Feifei.Wang2@arm.com</a>>; Ruifeng Wang <<a href="mailto:Ruifeng.Wang@arm.com">Ruifeng.Wang@arm.com</a>>; Huangshaozhang <<a href="mailto:huangshaozhang@huawei.com">huangshaozhang@huawei.com</a>>; dingxiaoxiong <<a href="mailto:dingxiaoxiong@huawei.com">dingxiaoxiong@huawei.com</a>> Subject: [dpdk-dev][dpdk-users] A problem about memory may not be all-zero allocated by rte_zmalloc_socket()<o:p></o:p> </div> </div> <o:p> </o:p> Hi, all<o:p></o:p> <o:p> </o:p> There's a problem that the memory are allocated by rte_zmalloc_socket()<o:p></o:p> may not be all-zero on the ARM platform.<o:p></o:p> <o:p> </o:p> However, the x86 platform does not have this problem.<o:p></o:p> <o:p> </o:p> Any ideas ?<o:p></o:p> <o:p> </o:p> Thanks,<o:p></o:p> Yunjian<o:p></o:p> <o:p> </o:p> </div> </div> </div> </div> </body> </html>