2009/7/23 Jeff Johnson n3...@mac.com
This isn't the right fix imho.
js/src needs to build standalone, not include config.h.
Otherwise, JS is gonna become MANDATORY INTERNAL to rpm.
I thought it already was..?
--
Regards,
Per Øyvind
I experienced problems with Smart and RPM 5.2 when trying to add URPMI channel
(MDV cooker contribs).
It happens when the channel is signed.
Smart was aborting because of assert in rpmdb/pkgio.c(173):
rpmRC rpmtsFindPubkey(rpmts ts, void * _dig)
{
HE_t he = memset(alloca(sizeof(*he)), 0,
On Jul 24, 2009, at 6:44 AM, Per Øyvind Karlsen wrote:
2009/7/23 Jeff Johnson n3...@mac.com
This isn't the right fix imho.
js/src needs to build standalone, not include config.h.
Otherwise, JS is gonna become MANDATORY INTERNAL to rpm.
I thought it already was..?
MANDATORY (as in reliably
On Jul 24, 2009, at 6:54 AM, David Smid wrote:
As you can see, this assert gives little sense - it tests if
rpmtsDig(ts) ==
dig, but that can be true only if the _dig argument is NULL (or
happens to equal
to rpmtsDig(ts)). Smart is obviously calling the function with
different value.
Jeff Johnson napsal(a):
On Jul 24, 2009, at 6:54 AM, David Smid wrote:
As you can see, this assert gives little sense - it tests if
rpmtsDig(ts) ==
dig, but that can be true only if the _dig argument is NULL (or
happens to equal
to rpmtsDig(ts)). Smart is obviously calling the function
On Jul 24, 2009, at 7:55 AM, David Smid wrote:
A reproducer please, the issue is deeper than removing an assert.
I'm afraid I am not able to provide simpler test case.
This is what I found out using gdb.
How many packages in the reproducer?
73 de Jeff
Jeff Johnson napsal(a):
On Jul 24, 2009, at 7:55 AM, David Smid wrote:
A reproducer please, the issue is deeper than removing an assert.
I'm afraid I am not able to provide simpler test case.
This is what I found out using gdb.
How many packages in the reproducer?
73 de Jeff
On Jul 24, 2009, at 7:55 AM, David Smid wrote:
I don't understand it, what's the point in passing an argument that
is not
allowed to have value different from rpmtsDig(ts) ( or NULL but that
gives the
same result as passing rpmtsDig(ts) ) ?
The first statement allows the dig variable to
On Jul 24, 2009, at 8:27 AM, David Smid wrote:
Jeff Johnson napsal(a):
On Jul 24, 2009, at 7:55 AM, David Smid wrote:
A reproducer please, the issue is deeper than removing an assert.
I'm afraid I am not able to provide simpler test case.
This is what I found out using gdb.
How many
Jeff Johnson napsal(a):
I was asking how many packages are in the transaction that is assert
failing, not how many packages are being used.
If the number of packages is tolerable, then I will next ask
What packages are in the transaction?
and attempt a reproducer.
I can likely debug
On Jul 24, 2009, at 9:34 AM, David Smid wrote:
Jeff Johnson napsal(a):
I was asking how many packages are in the transaction that is assert
failing, not how many packages are being used.
If the number of packages is tolerable, then I will next ask
What packages are in the transaction?
and
On Thu, Jul 23, 2009, Per Øyvind Karlsen wrote:
[...]
fix va_list autofoo stuff breaking libjs build
I'm wondering:
1. Why is HAVE_VA_LIST_AS_ARRAY defined but never used in the code?
2. Why are such changes committed directly to 5.2 and not to HEAD first?
2009/7/24 Ralf S. Engelschall rse+rpm-de...@rpm5.orgrse%2brpm-de...@rpm5.org
On Thu, Jul 23, 2009, Per Ųyvind Karlsen wrote:
[...]
fix va_list autofoo stuff breaking libjs build
I'm wondering:
1. Why is HAVE_VA_LIST_AS_ARRAY defined but never used in the code?
it's actually used
13 matches
Mail list logo
