* Re: sendfile syscall offset bug ??? [not found] <gh-mailinglist-notifications-41a7ca26-5023-4802-975b-f1789d68868e-void-packages-30741@inbox.vuxu.org> @ 2021-05-08 20:21 ` ericonr 2021-05-09 8:37 ` malv-c ` (6 subsequent siblings) 7 siblings, 0 replies; 8+ messages in thread From: ericonr @ 2021-05-08 20:21 UTC (permalink / raw) To: ml [-- Attachment #1: Type: text/plain, Size: 300 bytes --] New comment by ericonr on void-packages repository https://github.com/void-linux/void-packages/issues/30741#issuecomment-835498277 Comment: Is this not reproducible with C? What happens if you use sendfile as the function exposed by libc? Does it behave differently with different kernel versions? ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: sendfile syscall offset bug ??? [not found] <gh-mailinglist-notifications-41a7ca26-5023-4802-975b-f1789d68868e-void-packages-30741@inbox.vuxu.org> 2021-05-08 20:21 ` sendfile syscall offset bug ??? ericonr @ 2021-05-09 8:37 ` malv-c 2021-05-09 9:37 ` malv-c ` (5 subsequent siblings) 7 siblings, 0 replies; 8+ messages in thread From: malv-c @ 2021-05-09 8:37 UTC (permalink / raw) To: ml [-- Attachment #1: Type: text/plain, Size: 4987 bytes --] New comment by malv-c on void-packages repository https://github.com/void-linux/void-packages/issues/30741#issuecomment-835745838 Comment: sorry all you don't know asm me too i restart it but i know enough to understand that sendfile off_t refuse the integer lseek one accept also the exit value and size_t work [sf.txt](https://github.com/void-linux/void-packages/files/6447150/sf.txt) commented source of working example for too busy people to change a value in code ;) Duncaen % strace ./sf3-3 +255 =q sf3.s b* execve("./sf3-3", ["./sf3-3", "+255", "=q", "sf3.s", "bb", "bh%", "bh0"], 0x7ffc6a137880 /* 42 vars */) = 0 open("q", O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE|O_NOATIME, 0777) = 3 open("sf3.s", O_RDONLY) = 4 sendfile(3, 4, NULL, 255) = 255 open("bb", O_RDONLY) = 5 sendfile(3, 5, NULL, 255) = 255 open("bh%", O_RDONLY) = 6 sendfile(3, 6, NULL, 255) = 255 open("bh0", O_RDONLY) = 11 sendfile(3, 11, NULL, 255) = 255 exit_group(0) = ? +++ exited with 0 +++ /tmp % strace ./sf3-3 ^255 =qf sf3.s b* execve("./sf3-3", ["./sf3-3", "^255", "=qf", "sf3.s", "bb", "bh%", "bh0"], 0x7ffde3499810 /* 42 vars */) = 0 open("qf", O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE|O_NOATIME, 0777) = 3 open("sf3.s", O_RDONLY) = 4 sendfile(3, 4, 0xff, 4611686018427387904) = -1 EFAULT (Bad address) open("bb", O_RDONLY) = 5 sendfile(3, 5, 0xff, 4611686018427387904) = -1 EFAULT (Bad address) open("bh%", O_RDONLY) = 6 sendfile(3, 6, 0xff, 4611686018427387904) = -1 EFAULT (Bad address) open("bh0", O_RDONLY) = 11 sendfile(3, 11, 0xff, 4611686018427387904) = -1 EFAULT (Bad address) exit_group(255) = ? +++ exited with 255 +++ % strace ./sf3 +63 :q qq qqa qqs qqw execve("./sf3", ["./sf3", "+63", ":q", "qq", "qqa", "qqs", "qqw"], 0x7ffeaec3d830 /* 42 vars */) = 0 open("q", O_RDONLY) = 3 open("qq", O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE|O_NOATIME, 0777) = 4 sendfile(4, 3, NULL, 63) = 63 open("qqa", O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE|O_NOATIME, 0777) = 5 lseek(3, 0, SEEK_SET) = 0 sendfile(5, 3, NULL, 63) = 63 open("qqs", O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE|O_NOATIME, 0777) = 6 lseek(3, 0, SEEK_SET) = 0 sendfile(6, 3, NULL, 63) = 63 open("qqw", O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE|O_NOATIME, 0777) = 11 lseek(3, 0, SEEK_SET) = 0 sendfile(11, 3, NULL, 63) = 63 exit_group(0) = ? +++ exited with 0 +++ /tmp % strace ./sf3 ^63 :q qq qqa qqs qqw execve("./sf3", ["./sf3", "^63", ":q", "qq", "qqa", "qqs", "qqw"], 0x7ffcf6c22ce0 /* 42 vars */) = 0 open("q", O_RDONLY) = 3 open("qq", O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE|O_NOATIME, 0777) = 4 sendfile(4, 3, 0x3f, 4611686018427387904) = -1 EFAULT (Bad address) open("qqa", O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE|O_NOATIME, 0777) = 5 lseek(3, 63, SEEK_SET) = 63 sendfile(5, 3, 0x3f, 4611686018427387904) = -1 EFAULT (Bad address) open("qqs", O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE|O_NOATIME, 0777) = 6 lseek(3, 63, SEEK_SET) = 63 sendfile(6, 3, 0x3f, 4611686018427387904) = -1 EFAULT (Bad address) open("qqw", O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE|O_NOATIME, 0777) = 11 lseek(3, 63, SEEK_SET) = 63 sendfile(11, 3, 0x3f, 4611686018427387904) = -1 EFAULT (Bad address) exit_group(63) = ? +++ exited with 63 +++ % strace ./sf3 ^63 b* execve("./sf3", ["./sf3", "^63", "bb", "bh%", "bh0"], 0x7ffff06f9d30 /* 42 vars */) = 0 open("bb", O_RDONLY) = 3 fcntl(1, F_SETFL, O_RDONLY) = 0 sendfile(1, 3, 0x3f, 4611686018427387904) = -1 EFAULT (Bad address) open("bh%", O_RDONLY) = 4 sendfile(1, 4, 0x3f, 4611686018427387904) = -1 EFAULT (Bad address) open("bh0", O_RDONLY) = 5 sendfile(1, 5, 0x3f, 4611686018427387904) = -1 EFAULT (Bad address) exit_group(63) = ? +++ exited with 63 +++ /tmp % strace ./sf3 +63 b* execve("./sf3", ["./sf3", "+63", "bb", "bh%", "bh0"], 0x7fff8f955ab0 /* 42 vars */) = 0 open("bb", O_RDONLY) = 3 fcntl(1, F_SETFL, O_RDONLY) = 0 sendfile(1, 3, NULL, 63export HISTCONTROL=ignoredups:erasedups shopt -s histreedit his) = 63 open("bh%", O_RDONLY) = 4 sendfile(1, 4, NULL, 63AE ri M man tr ll /bin/tr gcl https://github.com/coreutils/cor) = 63 open("bh0", O_RDONLY) = 5 sendfile(1, 5, NULL, 63./sbx -z alsactl store alsamixer create_ap --help |more ee /) = 63 exit_group(0) = ? +++ exited with 0 +++ ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: sendfile syscall offset bug ??? [not found] <gh-mailinglist-notifications-41a7ca26-5023-4802-975b-f1789d68868e-void-packages-30741@inbox.vuxu.org> 2021-05-08 20:21 ` sendfile syscall offset bug ??? ericonr 2021-05-09 8:37 ` malv-c @ 2021-05-09 9:37 ` malv-c 2021-05-09 9:37 ` malv-c ` (4 subsequent siblings) 7 siblings, 0 replies; 8+ messages in thread From: malv-c @ 2021-05-09 9:37 UTC (permalink / raw) To: ml [-- Attachment #1: Type: text/plain, Size: 517 bytes --] New comment by malv-c on void-packages repository https://github.com/void-linux/void-packages/issues/30741#issuecomment-835765753 Comment: for small tasks the time c exec is loaded asm exec did the job ... also to return to code in c may aska long time and the c lib can put any values where it want to know it you just have to verify the code reverify it after compilation with the hope nothing happen while installing deal only with kernel problems is less painfull anyway i can test anything you want ... ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: sendfile syscall offset bug ??? [not found] <gh-mailinglist-notifications-41a7ca26-5023-4802-975b-f1789d68868e-void-packages-30741@inbox.vuxu.org> ` (2 preceding siblings ...) 2021-05-09 9:37 ` malv-c @ 2021-05-09 9:37 ` malv-c 2021-05-09 11:59 ` Duncaen ` (3 subsequent siblings) 7 siblings, 0 replies; 8+ messages in thread From: malv-c @ 2021-05-09 9:37 UTC (permalink / raw) To: ml [-- Attachment #1: Type: text/plain, Size: 517 bytes --] New comment by malv-c on void-packages repository https://github.com/void-linux/void-packages/issues/30741#issuecomment-835765757 Comment: for small tasks the time c exec is loaded asm exec did the job ... also to return to code in c may aska long time and the c lib can put any values where it want to know it you just have to verify the code reverify it after compilation with the hope nothing happen while installing deal only with kernel problems is less painfull anyway i can test anything you want ... ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: sendfile syscall offset bug ??? [not found] <gh-mailinglist-notifications-41a7ca26-5023-4802-975b-f1789d68868e-void-packages-30741@inbox.vuxu.org> ` (3 preceding siblings ...) 2021-05-09 9:37 ` malv-c @ 2021-05-09 11:59 ` Duncaen 2021-05-09 11:59 ` Duncaen ` (2 subsequent siblings) 7 siblings, 0 replies; 8+ messages in thread From: Duncaen @ 2021-05-09 11:59 UTC (permalink / raw) To: ml [-- Attachment #1: Type: text/plain, Size: 270 bytes --] New comment by Duncaen on void-packages repository https://github.com/void-linux/void-packages/issues/30741#issuecomment-835791971 Comment: This has nothing to do with void linux, you are using the syscalls wrong. lseek takes off_t and sendfile tasks a off_t pointer. ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: sendfile syscall offset bug ??? [not found] <gh-mailinglist-notifications-41a7ca26-5023-4802-975b-f1789d68868e-void-packages-30741@inbox.vuxu.org> ` (4 preceding siblings ...) 2021-05-09 11:59 ` Duncaen @ 2021-05-09 11:59 ` Duncaen 2021-05-09 14:01 ` malv-c 2021-05-09 14:25 ` Duncaen 7 siblings, 0 replies; 8+ messages in thread From: Duncaen @ 2021-05-09 11:59 UTC (permalink / raw) To: ml [-- Attachment #1: Type: text/plain, Size: 269 bytes --] New comment by Duncaen on void-packages repository https://github.com/void-linux/void-packages/issues/30741#issuecomment-835791971 Comment: This has nothing to do with void linux, you are using the syscall wrong. lseek takes off_t and sendfile tasks a off_t pointer. ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: sendfile syscall offset bug ??? [not found] <gh-mailinglist-notifications-41a7ca26-5023-4802-975b-f1789d68868e-void-packages-30741@inbox.vuxu.org> ` (5 preceding siblings ...) 2021-05-09 11:59 ` Duncaen @ 2021-05-09 14:01 ` malv-c 2021-05-09 14:25 ` Duncaen 7 siblings, 0 replies; 8+ messages in thread From: malv-c @ 2021-05-09 14:01 UTC (permalink / raw) To: ml [-- Attachment #1: Type: text/plain, Size: 188 bytes --] New comment by malv-c on void-packages repository https://github.com/void-linux/void-packages/issues/30741#issuecomment-835813595 Comment: sorry Duncaen i didn't read the man correctly ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: sendfile syscall offset bug ??? [not found] <gh-mailinglist-notifications-41a7ca26-5023-4802-975b-f1789d68868e-void-packages-30741@inbox.vuxu.org> ` (6 preceding siblings ...) 2021-05-09 14:01 ` malv-c @ 2021-05-09 14:25 ` Duncaen 7 siblings, 0 replies; 8+ messages in thread From: Duncaen @ 2021-05-09 14:25 UTC (permalink / raw) To: ml [-- Attachment #1: Type: text/plain, Size: 269 bytes --] New comment by Duncaen on void-packages repository https://github.com/void-linux/void-packages/issues/30741#issuecomment-835791971 Comment: This has nothing to do with void linux, you are using the syscall wrong. lseek takes off_t and sendfile takes a off_t pointer. ^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2021-05-09 14:25 UTC | newest] Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- [not found] <gh-mailinglist-notifications-41a7ca26-5023-4802-975b-f1789d68868e-void-packages-30741@inbox.vuxu.org> 2021-05-08 20:21 ` sendfile syscall offset bug ??? ericonr 2021-05-09 8:37 ` malv-c 2021-05-09 9:37 ` malv-c 2021-05-09 9:37 ` malv-c 2021-05-09 11:59 ` Duncaen 2021-05-09 11:59 ` Duncaen 2021-05-09 14:01 ` malv-c 2021-05-09 14:25 ` Duncaen
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).