On Tue, 2009-10-27 at 12:38 +, Eric B Munson wrote:
> The scripts added by this patch are used to measure the cost of a tlb
> miss on the running machine. They rely on the administrator obtaining
> the source for calibrator from:
> http://monetdb.cwi.nl/Calibrator/src/calibrator.c compiling it
On Fri, 2009-01-30 at 15:56 +, Eric B Munson wrote:
> From: Adam Litke
>
> Now that libhugetlbfs can work with multiple huge page sizes, testing this
> support has become a priority. The following patch enables automatic testing
> of page sizes that are mounted and have at
ed ??
>
> Since libhugetlbfs is a library, it has it's own text and data
> segments (independent of the program). So the library code runs
> uninhibited since only the program segments (not the library segments)
> are unmapped.
>
>
> thanks,
> Mehul.
>
>
On Mon, 2009-01-05 at 10:25 -0800, mehul vora wrote:
>
> Adam,
>
> I just commented out printfs, but still program gets a seg fault. I
> have a tight loop after "munmap", so ideally program should never come
> out, but for me program gets a SIGSEGV after first munmap
>
> Could you run your
On Mon, 2009-01-05 at 09:07 -0800, Nishanth Aravamudan wrote:
> On 05.01.2009 [06:26:55 -0800], mehul vora wrote:
> > Nishanth,
> >
> > I went through "munmap" code and found that core kernel (arch
> > independent) code removes all "vma" references of the passed vaddr and
> > then it calls a
nr_used);
> pool->pagesize = size;
> pool->minimum = nr_static;
> pool->maximum = nr_static + nr_over;
> - pool->size = nr_used + nr_resv;
> + pool->size = nr_used;
> pool->
> +support enabled, the binary segfaults.
Is the NX bit a feature of x86 chips only? Perhaps we should say "on an
x86 host with support for the NX bit enabled".
> +TRY:Recompiling with the --hugetlbfs-align options and use the new
> relinking
> +method or
I recently added a troubleshooting section for things like this. Would
you consider putting it there?
--
Adam Litke - (agl at us.ibm.com)
IBM Linux Technology Center
-
This SF.Net email is sponsored by the Moblin Your Move De
; all of the linker scripts.
>
> Credit goes to Alan Modra for identifying what change needed to be made
>
> Signed-off-by: Eric B Munson <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
--
Adam Litke - (agl at us.
cripts/elf_x86_64.xBDT
> @@ -31,6 +31,7 @@ SECTIONS
>.hash : { *(.hash) } :htext
>.note.SuSE : { *(.note.SuSE) } :htext :note
>.note.ABI-tag : { *(.note.ABI-tag) } :htext :note
> + .note.gnu.build-id : { *(.note.gnu.build-id) } :text :note
>.dynsym
t; read back instead. This can be useful for debugging failures.
>
> Signed-off-by: David Gibson <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
>
> Index: libhugetlbfs/tests/readback.c
> ===
&
: Adam Litke <[EMAIL PROTECTED]>
---
tests/fork-cow.c | 26 ++
1 files changed, 22 insertions(+), 4 deletions(-)
diff --git a/tests/fork-cow.c b/tests/fork-cow.c
index ef0f3e7..70d3904 100644
--- a/tests/fork-cow.c
+++ b/tests/fork-cow.c
@@ -47,6 +47,18 @@
#
Now that use of the libhugetlbfs linker scripts is deprecated, print a warning
message when ld.hugetlbfs is invoked with the --hugetlbfs-link option to
encourage the user to migrate to --hugetlbfs-align.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
ld.hugetlbfs |2 ++
1 files c
On Mon, 2008-11-17 at 11:50 +1100, David Gibson wrote:
> On Sat, Nov 15, 2008 at 11:56:54AM +1100, David Gibson wrote:
> > On Fri, Nov 14, 2008 at 10:06:57AM -0600, Adam Litke wrote:
> > > On Fri, 2008-11-14 at 17:51 +1100, David Gibson wrote:
> > > > On Thu, Nov 13,
On Sat, 2008-11-15 at 11:56 +1100, David Gibson wrote:
> On Fri, Nov 14, 2008 at 10:06:57AM -0600, Adam Litke wrote:
> > On Fri, 2008-11-14 at 17:51 +1100, David Gibson wrote:
> > > On Thu, Nov 13, 2008 at 08:26:12PM +, Adam Litke wrote:
> [snip]
> > > The one b
strongly recommend we proceed with the conversion.
On Mon, 2008-11-17 at 16:50 +, Mel Gorman wrote:
> On Fri, Nov 14, 2008 at 05:51:29PM +1100, David Gibson wrote:
> > On Thu, Nov 13, 2008 at 08:26:12PM +0000, Adam Litke wrote:
> > > Now that libhugetlbfs can work with multip
Changes since V1:
- In the event that the normal default huge page size cannot be used by
libhugetlbfs (due to no usable mount point), if the user specified the size
print a WARNING (since semantics are altered), otherwise just print an INFO
message.
Signed-off-by: Adam Litke <[EM
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
shm.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/shm.c b/shm.c
index ac3b52a..ed7731e 100644
--- a/shm.c
+++ b/shm.c
@@ -67,7 +67,7 @@ int shmget(key_t key, size_t size, int
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
alloc.c |5 +++--
1 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/alloc.c b/alloc.c
index 750b2cb..ccf550c 100644
--- a/alloc.c
+++ b/alloc.c
@@ -35,7 +35,7 @@ static void *fallback_base_pages(size_t len, ghp_t
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
hugeadm.c | 12 ++--
hugectl.c |8
pagesize.c |4 ++--
3 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/hugeadm.c b/hugeadm.c
index 6e9e939..f190a41 100644
--- a/hugeadm.c
+++ b/hugeadm.c
@@
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
morecore.c | 36 ++--
1 files changed, 18 insertions(+), 18 deletions(-)
diff --git a/morecore.c b/morecore.c
index 95d0470..ccda51f 100644
--- a/morecore.c
+++ b/morecore.c
@@ -74,7 +74,7 @@ stati
Changes since V1:
- When find_or_create_share_path() fails, segment remapping is disabled.
Print a message to make this clear to the user.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
elflink.c | 105 +
1 files chang
, unrecoverable errors
WARNING - Recoverable condition, but may result in altered semantics
INFO- Detailed information about normal library operations
DEBUG - Diagnostic information for debugging, potential runtime overhead
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
d
idea. Unfortunately this variable is part of the
API and must be maintained.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
Acked-by: Mel Gorman <[EMAIL PROTECTED]>
---
HOWTO | 44 +---
debug.c| 11 ++-
man/li
be maintained.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
HOWTO | 45 ++---
debug.c| 11 ++-
man/libhugetlbfs.7 | 16 +---
3 files changed, 45 insertions(+), 27 deletions(-)
diff --git a/H
On Mon, 2008-11-17 at 17:32 +, Mel Gorman wrote:
> On Fri, Nov 14, 2008 at 09:01:09PM +0000, Adam Litke wrote:
> > Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
> > ---
> >
> > hugeutils.c | 12 ++--
> > kernel-features.c | 16 ++
On Mon, 2008-11-17 at 17:30 +, Mel Gorman wrote:
> On Fri, Nov 14, 2008 at 09:01:03PM +0000, Adam Litke wrote:
> >
> > ---
> >
> > shm.c |8
> > 1 files changed, 4 insertions(+), 4 deletions(-)
> >
> > diff --git a/shm.c b/shm.
On Mon, 2008-11-17 at 17:28 +, Mel Gorman wrote:
> On Fri, Nov 14, 2008 at 09:00:50PM +0000, Adam Litke wrote:
> >
> > ---
> >
> > alloc.c |8 +---
> > 1 files changed, 5 insertions(+), 3 deletions(-)
> >
> > diff --git a/alloc.
ll look like errors to me that result in broken sharing.
We seem to disagree on the definition of a warning. Since all of these
cases result in fallback to no sharing, I'd consider them noteworthy
warning messages.
> > @@ -1063,7 +1064,7 @@ static int obtain_prepared_file(struct
On Mon, 2008-11-17 at 17:17 +, Mel Gorman wrote:
> On Fri, Nov 14, 2008 at 09:00:32PM +0000, Adam Litke wrote:
> > The default verbosity level in libhugetlbfs is 1, meaning that only errors
> > are
> > printed. Many of the new library features can have outcomes that ar
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
hugeutils.c | 12 ++--
kernel-features.c | 16
2 files changed, 14 insertions(+), 14 deletions(-)
diff --git a/hugeutils.c b/hugeutils.c
index 1121966..999da74 100644
--- a/hugeutils.c
+++ b/hugeu
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
morecore.c | 36 ++--
1 files changed, 18 insertions(+), 18 deletions(-)
diff --git a/morecore.c b/morecore.c
index 95d0470..ccda51f 100644
--- a/morecore.c
+++ b/morecore.c
@@ -74,7 +74,7 @@ stati
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
shm.c |8
1 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/shm.c b/shm.c
index ac3b52a..5e96400 100644
--- a/shm.c
+++ b/shm.c
@@ -38,7 +38,7 @@ int shmget(key_t key, size_t size, int shmflg)
size_t aligne
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
alloc.c |8 +---
1 files changed, 5 insertions(+), 3 deletions(-)
diff --git a/alloc.c b/alloc.c
index 750b2cb..f231f2c 100644
--- a/alloc.c
+++ b/alloc.c
@@ -35,7 +35,7 @@ static void *fallback_base_pages(size_t len, ghp_t
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
hugeadm.c | 12 ++--
hugectl.c |8
pagesize.c |4 ++--
3 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/hugeadm.c b/hugeadm.c
index 6e9e939..f190a41 100644
--- a/hugeadm.c
+++ b/hugeadm.c
@@
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
elflink.c | 101 -
1 files changed, 53 insertions(+), 48 deletions(-)
diff --git a/elflink.c b/elflink.c
index 63229f6..e38e796 100644
--- a/elflink.c
+++ b/elflink.c
@@ -
INFO- Detailed information about normal library operations
DEBUG - Diagnostic information for debugging, potential runtime overhead
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
debug.c |3 +--
libhugetlbfs_debug.h| 17 +
libhugetlbfs_inte
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
hugeutils.c | 12 ++--
kernel-features.c | 16
2 files changed, 14 insertions(+), 14 deletions(-)
diff --git a/hugeutils.c b/hugeutils.c
index 1121966..999da74 100644
--- a/hugeutils.c
+++ b/hugeu
---
shm.c |8
1 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/shm.c b/shm.c
index ac3b52a..5e96400 100644
--- a/shm.c
+++ b/shm.c
@@ -38,7 +38,7 @@ int shmget(key_t key, size_t size, int shmflg)
size_t aligned_size = size;
int hugetlbshm_enabled = 0;
---
hugeadm.c | 12 ++--
hugectl.c |8
pagesize.c |4 ++--
3 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/hugeadm.c b/hugeadm.c
index 6e9e939..f190a41 100644
--- a/hugeadm.c
+++ b/hugeadm.c
@@ -162,7 +162,7 @@ void pool_adjust(char *cmd, unsigned in
---
alloc.c |8 +---
1 files changed, 5 insertions(+), 3 deletions(-)
diff --git a/alloc.c b/alloc.c
index 750b2cb..f231f2c 100644
--- a/alloc.c
+++ b/alloc.c
@@ -35,7 +35,7 @@ static void *fallback_base_pages(size_t len, ghp_t flags)
{
int fd;
void *buf;
- DEBUG(
---
elflink.c | 101 -
1 files changed, 53 insertions(+), 48 deletions(-)
diff --git a/elflink.c b/elflink.c
index 63229f6..e38e796 100644
--- a/elflink.c
+++ b/elflink.c
@@ -252,14 +252,14 @@ static int find_or_create_share_path(long
---
morecore.c | 36 ++--
1 files changed, 18 insertions(+), 18 deletions(-)
diff --git a/morecore.c b/morecore.c
index 95d0470..ccda51f 100644
--- a/morecore.c
+++ b/morecore.c
@@ -74,7 +74,7 @@ static void *hugetlbfs_morecore(ptrdiff_t increment)
void
INFO- Detailed information about normal library operations
DEBUG - Diagnostic information for debugging, potential runtime overhead
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
debug.c |3 +--
libhugetlbfs_debug.h| 17 +
libhugetlbfs_inte
INFO- Detailed information about normal library operations
DEBUG - Diagnostic information for debugging, potential runtime overhead
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
debug.c |3 +--
libhugetlbfs_debug.h| 17 +
libhugetlbfs_inte
---
elflink.c | 101 -
1 files changed, 53 insertions(+), 48 deletions(-)
diff --git a/elflink.c b/elflink.c
index 63229f6..e38e796 100644
--- a/elflink.c
+++ b/elflink.c
@@ -252,14 +252,14 @@ static int find_or_create_share_path(long
On Fri, 2008-11-14 at 17:51 +1100, David Gibson wrote:
> On Thu, Nov 13, 2008 at 08:26:12PM +0000, Adam Litke wrote:
> > Now that libhugetlbfs can work with multiple huge page sizes, testing this
> > support has become a priority. The following patch enables automatic
> >
If -z common-page-size and -z max-page-size are not recognized by the linker
then the linkhuge_rw binary won't be suitably aligned and failures are
expected.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
tests/Makefile |2 +-
tests/linkhuge_r
Now that libhugetlbfs can work with multiple huge page sizes, testing this
support has become a priority. The following patch enables automatic testing
of page sizes that are mounted and have at least one page allocated. Care is
taken to assure that only valid combinations are tested. For exampl
ns with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> ___
> Libhugetlbfs-devel mailing list
> Libhugetlbfs-
t; #endif
>
> struct libhugeopts_t {
> + int min_copy;
> + int sharing;
> + unsigned long force_elfmap;
> + char*ld_preload;
> + char*elfmap;
> + char*share_path;
> };
>
> /*
--
>
> hpagesize = check_hugepagesize();
>
> + /* Must be root because this test modifies the overcommit pool */
> + check_must_be_root();
> +
> + pool_size = save_overcommit_pages(hpagesize);
> + set_nr_overcommit_hugepages(hpagesize, 0);
> +
> freepages =
y be wasted to
> offset the buffer by some random cacheline. Effort is made to select a
> random cache-line to give good average performance.
>
> Signed-off-by: Mel Gorman <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTEC
/proc/mounts is now used in addition to /proc/meminfo for detecting huge page
sizes. Override the contents of that file as well.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
tests/meminfo_nohuge.c |7 +++
1 files changed, 7 insertions(+), 0 deletions(-)
diff --git a
for example. This patch enforces the meaning implied by the function name by
> removing the ability of the function to fallback to base pages and redefines
> GHP_DEFAULT to mean "Use the default hugepage size to back the region".
>
> Signed-off-by: Mel Gorman <[EMAIL PROT
oes not care for such things, this patch adds get_hugepage_region().
> It allocates regions of memory that are backed by hugepages where possible
> but callers are not required to align their length and can request fallback
> to base pages.
>
> Signed-off-by: Mel Gorman <[EMAIL PRO
e selection */
> + linemod += len % numlines;
> + }
> + DEBUG("Using line offset %d from start\n", line);
> +
> + return bytebuf;
> +}
> +
--
Adam Litke - (agl at us.ibm.com)
IBM Linux Technology Center
--
-type linker
scripts because it leaves the text and data segments untouched and creates a
new segment for a huge page BSS. We need to ignore the boundary between the
text and data segment because it is not important.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
elflink.c |9 +++
behave more like anonymous memory (in that the huge
pages are freed after an unmap and the data will not persist) and that
another process cannot open the same fd to snoop data.
Otherwise these all look good.
> +.SH SEE ALSO
> +.I oprofile(1)
> +,
> +.I libhugetlbfs(7)
> +.SH AUTHORS
On Thu, 2008-11-06 at 10:15 +1100, David Gibson wrote:
> On Wed, Nov 05, 2008 at 09:43:40AM -0600, Adam Litke wrote:
> > On Wed, 2008-11-05 at 14:02 +, Mel Gorman wrote:
> > > @@ -127,6 +140,34 @@ void *get_huge_pages(size_t len, ghp_t flags)
> >
being
> a useless and unnecessary burden, it prevents anything useful being done
> with the wasted bytes. This patch relieves the caller of some responsibility.
>
> Signed-off-by: Mel Gorman <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> alloc.c | 3
gt; +
> + buf = (void *)bytebuf;
> + }
> +
> /* woo, new buffer of shiny */
> return buf;
> }
--
Adam Litke - (agl at us.ibm.com)
IBM Linux Technology Center
-
This SF.Net email is
On Tue, 2008-11-04 at 15:21 +, Mel Gorman wrote:
> Currently an offset is being passed in to mmap() when allocating a
> region. This is pointless so pass in 0.
>
> Signed-off-by: Mel Gorman <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
>
but expected semantics will be altered
INFO - Detailed information about normal operation
DEBUG - Diagnostic information with a potential run-time acquisition
cost.
We will then switch the default verbosity level to WARNING.
--
Adam Litke - (agl at us.ibm.com)
IBM Linux Technology Center
-
TECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> Makefile |4 ++--
> 1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/Makefile b/Makefile
> index b9e2540..4554154 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -297,12 +29
We want to document the possible need to use --relax with GNU ld to avoid
truncated relocations. Update the documentation.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
HOWTO | 10 ++
1 files changed, 10 insertions(+), 0 deletions(-)
diff --git a/HOWTO b/HOWTO
index 5
alidate this cache to allow us test with various values.
>
> Add and expose kernel_default_hugepage_size_reset() as a test probe,
> and use this within the gethugepagesizes test.
>
> Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMA
44
> --- a/version.lds
> +++ b/version.lds
> @@ -4,6 +4,8 @@ VERS_1.0 {
> hugetlbfs_test_path;
> hugetlbfs_find_path;
> hugetlbfs_unlinked_fd;
> +
> + __tp_*;
Are these 1.0 symbols? I would think they should go in HTLBFS_2.1
> local:
>
> Acked-by: Mel Gorman <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> hugeutils.c | 33 +
> libhugetlbfs_privutils.h |3 +++
> shm.c|2 +-
> 3 files changed, 25 inserti
The HUGETLB_MORECORE and HUGETLB_ELFMAP options can now accept page size
specifiers. Provide the details on how to do this in the HOWTO.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
HOWTO | 61 +++--
1 files changed, 55 inse
X, 1, expected_sizes);
>
> - INIT_LIST(expected_sizes, base_size, meminfo_size);
> + INIT_LIST(expected_sizes, base_size, 2048 * 1024);
Magic numbers :( Since these have some correlation to the values in the
fake memin
instead of the c program directly. The shell script must perform any
required discovery, execute the real test case, and interpret the result.
Comments?
Changes since V1:
- Revamped the build so wrappers properly depend on their test cases
Signed-off-by: Adam Litke <[EMAIL PROTEC
ned-off-by: Adam Litke <[EMAIL PROTECTED]>
---
tests/Makefile |4 +++
tests/quota.sh | 13 +++
tests/run_tests.sh |2 +-
tests/wrapper-utils.sh | 56
4 files changed, 73 insertions(+), 2 deletions(-)
cr
Make use of pre-existing kernel version comparison code and create a helper
that can be used by test wrapper scripts to compare two kernel versions.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
kernel-features.c| 11 +++
libhugetlbfs_privutils.h |3 +++
tests/Ma
Oops. The greater-than operator (>) is not commutative. While trying to test
for a > b, we are accidentally re-testing a < b. Fix.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
kernel-features.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/ke
lure
detection. Hmm. Maybe I can have an unexpected pass return RC_BUG? Or
perhaps we don't care about them and can just return PASS?
--
Adam Litke - (agl at us.ibm.com)
IBM Linux Technology Center
-
This SF.Net email
quota is known to fail on kernels < 2.6.24. Create a wrapper script to check
for an old running kernel and report the failure as expected. Place reusable
shell functions into their own wrapper-utils.sh script.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
tests/Makefile
instead of the c program directly. The shell script must perform any
required discovery, execute the real test case, and interpret the result.
Comments?
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
tests/Makefile | 18 +-
tests/hugetests.h |2 ++
Make use of pre-existing kernel version comparison code and create a helper
that can be used by test wrapper scripts to compare two kernel versions.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
kernel-features.c| 11 +++
libhugetlbfs_privutils.h |3 +++
tests/Ma
Oops. The greater-than operator (>) is not commutative. While trying to test
for a > b, we are accidentally re-testing a < b. Fix.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
kernel-features.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/ke
HOWTO
> index 54519e2..02fb6a8 100644
> --- a/HOWTO
> +++ b/HOWTO
> @@ -2,7 +2,7 @@ libhugetlbfs HOWTO
> ======
>
> Author: David Gibson <[EMAIL PROTECTED]>, Adam Litke <[EMAIL PROTECTED]>,
> and others
> -Last updated: Wed
h 25 hugepages allocated, most tests should succeed. However, with
> +smaller hugepages sizes, (like PowerPC 64kb pages) many more hugepages
> +may be necessary.
>
> To use libhugetlbfs features, as well as to run the testsuite, hugetlbfs
> must be mounted:
--
Adam Litke - (
nor part of the libhugetlfs API.
> + *
> + * All functions declared external here must be externalised using a define
> + * of the following form:
> + *
> + * #define foo __pu_foo
> + */
> +
> #ifndef _LIBHUGETLBFS_PRIVUTILS_H
> #define _LIBHUGETLBFS_PRIVUTILS_H
>
--
Ada
Andy Whitcroft had some good suggestions on my check_linkhuge_tests patch.
- Comment the format of LINKHUGE_WORDSIZES
- Use a case statement instead of an if/else
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
tests/run_tests.sh | 14 --
1 files changed, 8 insertions
ust test the real one.
>
> Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> Makefile | 31 ++-
> init_privutils.c | 25 +
> libhugetlbfs_int
OTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> version.lds |2 ++
> 1 files changed, 2 insertions(+), 0 deletions(-)
>
> diff --git a/version.lds b/version.lds
> index 1429230..626ee02 100644
> --- a/version.lds
> +++ b/version.
On Fri, 2008-10-17 at 17:14 +0100, Andy Whitcroft wrote:
> Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> version.lds |6 ++
> 1 files changed, 6 insertions(+), 0 deletions(-)
>
> diff --git a/version.lds
; Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Fair enough reasoning. I agree.
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> version.lds |1 +
> 1 files changed, 1 insertions(+), 0 deletions(-)
>
> diff --git a/version.lds b/version.lds
> index ba06c8
On Fri, 2008-10-17 at 17:14 +0100, Andy Whitcroft wrote:
> Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> elf32ppclinux.c |2 ++
> elf64ppc.c |2 ++
> libhugetlbfs_internal.h |
;
> Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> libhugetlbfs_internal.h |4
> 1 files changed, 4 insertions(+), 0 deletions(-)
>
> diff --git a/libhugetlbfs_internal.h b/libhugetlbfs_internal.h
> inde
On Fri, 2008-10-17 at 17:14 +0100, Andy Whitcroft wrote:
> Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> version.c |2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/version.c b/vers
On Fri, 2008-10-17 at 17:14 +0100, Andy Whitcroft wrote:
> Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> hugeutils.c |2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/hugeutils.c
On Fri, 2008-10-17 at 17:14 +0100, Andy Whitcroft wrote:
> Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> hugetlbfs.h |4
> libhugetlbfs_internal.h |4
> 2 files changed, 4 insertions(+
On Fri, 2008-10-17 at 17:14 +0100, Andy Whitcroft wrote:
> Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> hugetlbfs.h |1 -
> kernel-features.c |4 ++--
> libhugetlbfs_internal.h |
document in each .c file if it is a part of a
library and which one. That will make it easier for someone to figure
out in which file a new function is to be implemented.
On Fri, 2008-10-17 at 17:14 +0100, Andy Whitcroft wrote:
> Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Acked-by: A
On Fri, 2008-10-17 at 17:14 +0100, Andy Whitcroft wrote:
> Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> hugeutils.c |4 ++--
> 1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/hugeutils.c
On Fri, 2008-10-17 at 17:14 +0100, Andy Whitcroft wrote:
> Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Acked-by: Adam Litke <[EMAIL PROTECTED]>
> ---
> Makefile |6 +++---
> hugetlbfs.h | 16
> huge
; +++ b/tests/counters.c
> @@ -181,6 +181,7 @@ void _set_nr_hugepages(unsigned long count, int line)
> out:
> verify_counters(line, et, ef, er, es);
> }
> +#undef set_nr_hugepages
Why is this needed?
> #define set_nr_hugepages(c) _set_nr_hugepages(c, __LINE
> will remain unexported:
>
> #define hpool_sizes __lh_hpool_sizes
> extern int hpool_sizes(struct hpage_pool *, int);
>
> This patch converts all current library local functions to this new idiom.
>
> Signed-off-by: Andy Whitcroft <[EMAIL PROTECTED]>
Acked-
.
Signed-off-by: Adam Litke <[EMAIL PROTECTED]>
---
tests/run_tests.sh | 38 ++
1 files changed, 26 insertions(+), 12 deletions(-)
diff --git a/tests/run_tests.sh b/tests/run_tests.sh
index 4fe6eed..dc2370f 100755
--- a/tests/run_tests.sh
+++ b/tests/run_te
1 - 100 of 530 matches
Mail list logo
