# [PATCH] Use dev_t for device id (st_dev) from stat in setup_git_directory_gently() 6 messages from 2010-07-13 to 2010-07-14. Participants: Raja R Harinath, Ævar Arnfjörð Bjarmason, Junio C Hamano. Thread: https://gitlist.dev/t/24383 ## Raja R Harinath, 2010-07-13 09:02 Subject: [PATCH] Use dev_t for device id (st_dev) from stat in setup_git_directory_gently() Message-ID: <1279011720-21424-1-git-send-email-harinath@hurrynot.org> URL: https://gitlist.dev/e/1279011720-21424-1-git-send-email-harinath%40hurrynot.org ``` The original declaration was int, which seems to cause trouble on my machine. It causes spurious "filesystem boundary" errors when running the testsuite. The cause seems to be $ stat -c%d . 2147549952 which is too large for a 32-bit int type. Using the correct type, dev_t, solves the issue. (Because I'm paranoid and forgetful, I checked -- yes, Unix v7 had dev_t.) Other uses of st_dev seem to be reasonably safe. fill_stat_cache_info truncates it to an 'unsigned int', but that value seems to be used only to validate the cache, and only if USE_STDEV is defined. --- setup.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/setup.c b/setup.c index 7e04602..87c21f0 100644 --- a/setup.c +++ b/setup.c @@ -323,7 +323,8 @@ const char *setup_git_directory_gently(int *nongit_ok) const char *gitdirenv; const char *gitfile_dir; int len, offset, ceil_offset, root_len; - int current_device = 0, one_filesystem = 1; + dev_t current_device = 0; + int one_filesystem = 1; struct stat buf; /* -- 1.7.2.rc2.11.g03e33 ``` ## Ævar Arnfjörð Bjarmason, 2010-07-13 09:26 Subject: Re: [PATCH] Use dev_t for device id (st_dev) from stat in setup_git_directory_gently() Message-ID: URL: https://gitlist.dev/e/AANLkTiku82MylNMUX6P8be3xSLVg8g8aYsNWodaOM6LQ%40mail.gmail.com In-Reply-To: <1279011720-21424-1-git-send-email-harinath@hurrynot.org> ``` On Tue, Jul 13, 2010 at 09:02, Raja R Harinath wrote: > The original declaration was int, which seems to cause trouble on my > machine. What OS/architecture is your machine running? ``` ## Raja R Harinath, 2010-07-13 12:01 Subject: Re: [PATCH] Use dev_t for device id (st_dev) from stat in setup_git_directory_gently() Message-ID: <87d3ur7fr1.fsf@hariville.hurrynot.org> URL: https://gitlist.dev/e/87d3ur7fr1.fsf%40hariville.hurrynot.org In-Reply-To: ``` Hi, Ævar Arnfjörð Bjarmason writes: > On Tue, Jul 13, 2010 at 09:02, Raja R Harinath wrote: >> The original declaration was int, which seems to cause trouble on my >> machine. > > What OS/architecture is your machine running? Nothing fancy $ uname -a Linux hariville.hurrynot.org 2.6.35-rc3-00397-g123f94f #146 SMP PREEMPT Mon Jul 5 15:50:47 IST 2010 x86_64 GNU/Linux $ df . Filesystem 1K-blocks Used Available Use% Mounted on /dev/sda1 959314384 389150456 521433560 43% / - Hari ``` ## Junio C Hamano, 2010-07-13 20:55 Subject: Re: [PATCH] Use dev_t for device id (st_dev) from stat in setup_git_directory_gently() Message-ID: <7voceb2jaw.fsf@alter.siamese.dyndns.org> URL: https://gitlist.dev/e/7voceb2jaw.fsf%40alter.siamese.dyndns.org In-Reply-To: <1279011720-21424-1-git-send-email-harinath@hurrynot.org> ``` Raja R Harinath writes: > The original declaration was int, which seems to cause trouble on my > machine. It causes spurious "filesystem boundary" errors when running > the testsuite. The cause seems to be > > $ stat -c%d . > 2147549952 > > which is too large for a 32-bit int type. > > Using the correct type, dev_t, solves the issue. (Because I'm > paranoid and forgetful, I checked -- yes, Unix v7 had dev_t.) > > Other uses of st_dev seem to be reasonably safe. fill_stat_cache_info > truncates it to an 'unsigned int', but that value seems to be used only > to validate the cache, and only if USE_STDEV is defined. > --- Makes sense; thanks. Sign-off? > setup.c | 3 ++- > 1 files changed, 2 insertions(+), 1 deletions(-) > > diff --git a/setup.c b/setup.c > index 7e04602..87c21f0 100644 > --- a/setup.c > +++ b/setup.c > @@ -323,7 +323,8 @@ const char *setup_git_directory_gently(int *nongit_ok) > const char *gitdirenv; > const char *gitfile_dir; > int len, offset, ceil_offset, root_len; > - int current_device = 0, one_filesystem = 1; > + dev_t current_device = 0; > + int one_filesystem = 1; > struct stat buf; > > /* > -- > 1.7.2.rc2.11.g03e33 ``` ## Raja R Harinath, 2010-07-14 01:08 Subject: Re: [PATCH] Use dev_t for device id (st_dev) from stat in setup_git_directory_gently() Message-ID: <878w5elvk2.fsf@hariville.hurrynot.org> URL: https://gitlist.dev/e/878w5elvk2.fsf%40hariville.hurrynot.org In-Reply-To: <7voceb2jaw.fsf@alter.siamese.dyndns.org> ``` Hi, Junio C Hamano writes: > Raja R Harinath writes: > >> The original declaration was int, which seems to cause trouble on my >> machine. It causes spurious "filesystem boundary" errors when running >> the testsuite. The cause seems to be >> >> $ stat -c%d . >> 2147549952 >> >> which is too large for a 32-bit int type. >> >> Using the correct type, dev_t, solves the issue. (Because I'm >> paranoid and forgetful, I checked -- yes, Unix v7 had dev_t.) >> >> Other uses of st_dev seem to be reasonably safe. fill_stat_cache_info >> truncates it to an 'unsigned int', but that value seems to be used only >> to validate the cache, and only if USE_STDEV is defined. >> --- > > Makes sense; thanks. > > Sign-off? Signed-off-by: Raja R Harinath Do I need to resend? >> setup.c | 3 ++- >> 1 files changed, 2 insertions(+), 1 deletions(-) >> >> diff --git a/setup.c b/setup.c >> index 7e04602..87c21f0 100644 >> --- a/setup.c >> +++ b/setup.c >> @@ -323,7 +323,8 @@ const char *setup_git_directory_gently(int *nongit_ok) >> const char *gitdirenv; >> const char *gitfile_dir; >> int len, offset, ceil_offset, root_len; >> - int current_device = 0, one_filesystem = 1; >> + dev_t current_device = 0; >> + int one_filesystem = 1; >> struct stat buf; >> >> /* >> -- >> 1.7.2.rc2.11.g03e33 ``` ## Junio C Hamano, 2010-07-14 14:55 Subject: Re: [PATCH] Use dev_t for device id (st_dev) from stat in setup_git_directory_gently() Message-ID: <7voceayuwz.fsf@alter.siamese.dyndns.org> URL: https://gitlist.dev/e/7voceayuwz.fsf%40alter.siamese.dyndns.org In-Reply-To: <878w5elvk2.fsf@hariville.hurrynot.org> ``` Raja R Harinath writes: > Hi, > > Junio C Hamano writes: > >> Raja R Harinath writes: >> >>> The original declaration was int, which seems to cause trouble on my >>> machine. It causes spurious "filesystem boundary" errors when running >>> the testsuite. The cause seems to be >>> >>> $ stat -c%d . >>> 2147549952 >>> >>> which is too large for a 32-bit int type. >>> >>> Using the correct type, dev_t, solves the issue. (Because I'm >>> paranoid and forgetful, I checked -- yes, Unix v7 had dev_t.) >>> >>> Other uses of st_dev seem to be reasonably safe. fill_stat_cache_info >>> truncates it to an 'unsigned int', but that value seems to be used only >>> to validate the cache, and only if USE_STDEV is defined. >>> --- >> >> Makes sense; thanks. >> >> Sign-off? > > Signed-off-by: Raja R Harinath > > Do I need to resend? No. Again thanks for a fix. >>> setup.c | 3 ++- >>> 1 files changed, 2 insertions(+), 1 deletions(-) >>> >>> diff --git a/setup.c b/setup.c >>> index 7e04602..87c21f0 100644 >>> --- a/setup.c >>> +++ b/setup.c >>> @@ -323,7 +323,8 @@ const char *setup_git_directory_gently(int *nongit_ok) >>> const char *gitdirenv; >>> const char *gitfile_dir; >>> int len, offset, ceil_offset, root_len; >>> - int current_device = 0, one_filesystem = 1; >>> + dev_t current_device = 0; >>> + int one_filesystem = 1; >>> struct stat buf; >>> >>> /* >>> -- >>> 1.7.2.rc2.11.g03e33 > > -- > To unsubscribe from this list: send the line "unsubscribe git" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html ```